OCTOBER 2016 Volume 2, Issue 5 assets/JGO_2_5.pdf · Compulsory Licenses for Cancer Drugs: Does Circumventing Patent Rights Improve Access to Oncology ... Surendra S. Shastri, MD,

Breaking Cancer Bad News to Patients With CancerJamal Zekri et al

Prioritizing US Cervical Cancer Prevention With Results From a Geospatial ModelJonathan K. Kish et al

Phenotype and Treatment of Breast Cancer in HIV-Positive and -Negative Women in Cape Town, South Africa

Lizanne Langenhoven et al

OCTOBER 2016Volume 2, Issue 5

An official publication of

ContentsEDITORIAL

Strengthening Global Partnership in Breast Cancer ResearchEdward L. Trimble and Kathy J. Helzlsouer (see article on page 302).................................................................... 253

COMMENTARY

Breast Cancer Among Iraqi Women: Preliminary Findings From a Regional Comparative Breast CancerResearch ProjectNada A.S. Alwan ................................................................................................................................................... 255

ORIGINAL REPORTS

Thoracic Oncology

EGFR Mutations in Latinos From the United States and Latin AmericaAriel Lopez-Chavez, Anish Thomas, Moses O. Evbuomwan, et al........................................................................... 259

Palliative and Supportive Care

Breaking Cancer Bad News to Patients With Cancer: A Comprehensive Perspective of Patients, Their Relatives,and the Public—Example From a Middle Eastern CountryJamal Zekri and Syed Mustafa Karim .................................................................................................................... 268

Gynecologic Cancer

Prioritizing US Cervical Cancer Prevention With Results From a Geospatial ModelJonathan K. Kish, Alicia I. Rolin, Zhaohui Zou, et al............................................................................................... 275

Breast Cancer

Phenotype and Treatment of Breast Cancer in HIV-Positive and -Negative Women in Cape Town, South AfricaLizanne Langenhoven, Pieter Barnardt, Alfred I. Neugut, et al............................................................................... 284

REVIEW ARTICLE

Compulsory Licenses for Cancer Drugs: Does Circumventing Patent Rights Improve Access to OncologyMedications?Cinthia Leite Frizzera Borges Bognar, Brittany L. Bychkovsky, and Gilberto de Lima Lopes Jr ............................... 292

Journal of Global Oncology (ISSN 2378-9506) is published online only six times a year, bimonthly,by the American Society of Clinical Oncology, 2318 Mill Road, Suite 800, Alexandria, VA 22314.

Editorial correspondence should be addressed to David Kerr, MD, Journal of Global Oncology.Phone: 703-797-1900; Fax: 703-684-8720. E-mail: [email protected]. Internet: www.jgo.org.

Journal of Global Oncology® is a registered trademark of American Society of Clinical Oncology, Inc.

continued on next page

October 2016Volume 2, Issue 5

mailto:[email protected]

http://www.jco.org/podcasts

SPECIAL ARTICLES

Breast Cancer and African Ancestry: Lessons Learned at the 10-Year Anniversary of the Ghana-MichiganResearch Partnership and International Breast RegistryEvelyn Jiagge, Joseph Kwaku Oppong, Jessica Bensenhaver, et al (See editorial on page 253)............................. 302

Management and Care of Women With Invasive Cervical Cancer: American Society of Clinical OncologyResource-Stratified Clinical Practice GuidelineLinus T. Chuang, Sarah Temin, Rolando Camacho, et al ....................................................................................... 311

CASE REPORT

Gallbladder Adenomyomatosis Mimicking Carcinoma: A Diagnostic DilemmaAbhishek Mahajan and Smiti Sripathi .................................................................................................................... 341

Also in This Issue

Information for AuthorsCurrent Abstracts

@JGO_ASCO

October 2016 Volume 2, Issue 5

jgo Editorial RosterEDITOR-IN-CHIEF

David Kerr, MD, DSc, Oxford, United Kingdom

ASSOCIATE EDITORS

Otis Brawley, MD, Atlanta, GA

Mary Gospodarowicz, MD, Toronto, Canada

Tezer Kutluk, MD, PhD, Ankara, Turkey

Gilberto Lopes, MD, MBA, S~ao Paulo, Brazil, andMiami, FL

Rengaswamy Sankaranarayanan, MD, Lyon,France

D. Cristina Stefan, MD, PhD, Cape Town, SouthAfrica

SOCIAL MEDIA EDITOR

Gilberto Lopes, MD, MBA, S~ao Paulo, Brazil, andMiami, FL

EDITORIAL BOARD

Isaac F. Adewole, MB, BS, Ibadan, Oyo, Nigeria

Nada A. S. Al Alwan, MD, PhD, Baghdad, Iraq

Benjamin O. Anderson, MD, Seattle, WA

Ami S. Bhatt, MD, PhD, Stanford, CA

Gouri Shankar Bhattacharyya, MD, PhD, DNB,MRCP, Kolkata, India

Philip E. Castle, PhD, MPH, Arlington, VA

Franco Cavalli, MD, Ticino, Switzerland

Eduardo L. Cazap, MD, PhD, Buenos Aires,Argentina

James Cleary, MB, BS, FAChPM, Madison, WI

C. Norman Coleman, MD, Chevy Chase, MD

Anil D’Cruz, Mumbai, India

Paul de Souza BSc(Med), MBBS, MPH, PhD,FRACP, Sydney NSW, Australia

Nagi S. El Saghir, MD, Beirut, Lebanon

Ahmed Elzawawy, MD, Port Said, Egypt

Kenneth Fleming, MB ChB, MA(Oxon), DPhil,FRCPath, FRCP, Oxford, United Kingdom, andBethesda, MD

Tamer M. Fouad, MD, PhD, Cairo, Egypt

Ophira Ginsburg, MSc, MD, Toronto, Canada

William J. Gradishar, MD, Chicago, IL

Serigne Gueye, MD, FWACS, Dakar, Senegal

Francisco Gutierrez-Delgado, MD, PhD, FACP,Oaxaca, Mexico

Abdul Rahman Jazieh, MD, MPH, Riyadh, SaudiArabia

Felicia Knaul, PhD, Coral Gables, FL

Patrick J. Loehrer, MD, Indianapolis, IN

Gary H. Lyman, MD, MPH, Seattle, WA

Danny Milner, MD, Boston, MA

Twalib Ngoma, MD, Dar es Salaam, Tanzania

Nir Peled, MD, PhD, FCCP, Petah Tikva, Israel

Surendra S. Shastri, MD, Mumbai, India

Yuankai Shi, MD, PhD, Beijing, China

Lawrence N. Shulman, MD, Philadelphia, PA

George W. Sledge Jr., MD, Stanford, CA

Richard Sullivan, MD, PhD, London, UnitedKingdom

Edward L. Trimble, MD, MPH, Bethesda, MD

Daniel A. Vorobiof, MD, Johannesburg, SouthAfrica

Christopher P. Wild, PhD, Lyon, France

William C. Wood, MD, Atlanta, GA

Annie Young, Coventry, UK

Publisher:David SampsonManaging Editors:Ken G. KornfieldEmilie Gunn

Volume 2, Issue 5, October 2016 jgo.ascopubs.org JGO – Journal of Global Oncology

http://jgo.ascopubs.org

jgo Journal of Global Oncology

JOURNAL OF GLOBAL ONCOLOGY

(ISSN 2378-9506) is published online only sixtimesa year, bimonthly, by theAmericanSociety ofClinical Oncology, 2318 Mill Road, Suite 800,Alexandria, VA 22314.

EDITORIAL CORRESPONDENCE(manuscript-related inquiries):

David Kerr, MD, Editor-in-ChiefJournal of Global OncologyPhone: 703-797-1900; Fax: 703-684-8720E-mail: [email protected]; Internet: jgo.org

AMERICAN SOCIETY OF CLINICAL ONCOLOGY(membership-related inquiries):

ASCO Member ServicesPhone: 703-299-0158Toll-free: 888-282-2552Fax: 703-299-0255E-mail: [email protected]: www.asco.orgHours: Monday-Friday, 8:30 a.m.-5:00 p.m.Eastern Time

CUSTOMER SERVICE

JGO Customer ServicePhone: 703-519-1430; Toll-free: 888-273-3508; Fax: 703-518-8155E-mail: [email protected]

ADVERTISING SALES

The Walchli Tauber Group, Inc.225 Old Emmorton Road, Suite 201Bel Air, MD 21015Phone: 443-512-8899; Fax: 443-512-8909www.wt-group.com

BUSINESS-TO-BUSINESS SALES

Rick WerdannSpringer Healthcare, LLC233 Spring StreetNew York, NY 10013Phone: 212-460-1523Mobile: 646-209-1840E-mail: [email protected]: www.SpringerHealthcare.com

FREE PUBLIC ACCESS

Journal of Global Oncology (JGO) provides freeonline access at jgo.org.

PERMISSIONS REQUESTS

Licensing, Rights, and Permissions DivisionAmerican Society of Clinical OncologyPhone: 571-483-1722Fax: 703-518-5094E-mail: [email protected]

Licensing: All articles are published under a CC BY or CC BY-NC-ND Creative Commons license, found here:https://creativecommons.org/licenses/.Copyright: Copyright © 2016 by American Society of Clinical Oncology unless otherwise indicated. All rights reserved. Nopart of this publication may be reproduced or transmitted in any form or by any means now or hereafter known, electronicor mechanical, including photocopy, recording, or any information storage and retrieval system, without permission inwriting from the Publisher. Printed in the United States of America. Journal of Global Oncology® is a registered trademarkof American Society of Clinical Oncology, Inc.

Support: Journal of Global Oncology has received support from the following funders and thanks them for their generouscontributions: Conquer Cancer Foundation Mission Endowment; Doris Duke Charitable Foundation; Novartis Oncology; NationalCancer Institute; CRDFGlobal; andNational Science Foundation. For information about supporting Journal of Global Oncology orother programs of the American Society of Clinical Oncology (ASCO), please contact the Conquer Cancer Foundation of ASCO at(571) 483-1700 or visit www.conquercancerfoundation.org/corporate-foundation-giving.

Disclaimer:The ideas and opinionsexpressed in JGO donot necessarily reflectthose of the AmericanSociety of ClinicalOncology (ASCO). Themention of any product,service, or therapy inthis publication or inany advertisement inthis publication shouldnot be construed asan endorsement of theproducts mentioned. Itis the responsibility ofthe treating physicianor other health careprovider, relying onindependent experienceand knowledge of thepatient, to determinedrug dosages and thebest treatment for thepatient. Readers areadvised to check theappropriate medicalliterature and theproduct informationcurrently provided by themanufacturer of eachdrug to be administeredto verify approveduses, the dosage,method, and durationof administration, orcontraindications.Readers are alsoencouraged to contactthe manufacturer withquestions about thefeatures or limitationsof any products. ASCOassumes no responsibilityfor any injury or damageto persons or propertyarising out of or relatedto any use of thematerial contained inthis publication or to anyerrors or omissions.



http://jgo.org


http://www.asco.org


http://www.wt-group.com


http://www.SpringerHealthcare.com

http://jgo.org


https://creativecommons.org/licenses/

http://www.conquercancerfoundation.org/corporate-foundation-giving


jgo Current Abstracts Volume 2, Issue 5 October 2016

EGFR Mutations in Latinos From the United States and Latin America

abstract

Purpose Epidermal growth factor receptor (EGFR) mutations confer sensitivity to EGFR tyrosine kinaseinhibitors in patients with advanced non–small-cell lung cancer (NSCLC). There are limited and con-flicting reports on the frequency of EGFR mutations in Latinos.

Patients and Methods Samples from 642 patients with NSCLC from seven institutions in the United Statesand Latin America were assessed for EGFRmutations (exons 18 to 21) at Clinical Laboratory ImprovementAmendments-certified central laboratories.

Results EGFR mutation analysis was successfully performed in 480 (75%) of 642 patients; 90 (19%)were Latinos, 318 (66%) were non-Latino whites, 35 (7%) were non-Latino Asians, 30 (6%) were non-Latino blacks, and seven (2%) were of other races or ethnicities. EGFRmutations were found in 21 (23%)of 90 Latinos with varying frequencies according to the country of origin; Latinos from Peru (37%),followed by the United States (23%), Mexico (18%), Venezuela (10%), and Bolivia (8%). In never-smoker Latinos and Latinos with adenocarcinoma histology, EGFR mutation frequencies were 38% and30%, respectively. There was a significant difference in the frequency of EGFR mutations among thedifferent racial and ethnic subgroups analyzed (P < .001), with non-Latino Asians having the highestfrequency (57%) followed by Latinos (23%), non-Latino whites (19%), and non-Latino blacks (10%).There was no difference between Latinos (23%) and non-Latinos (22%; P = .78) and Latinos and non-Latino whites (P = .37). Patients from Peru had an overall higher frequency of mutations (37%) than allother Latinos (17%), but this difference only exhibited a trend toward significance (P = .058).

Conclusion There was no significant difference between the frequency of EGFR mutations in NSCLC inLatinos and non-Latinos.

J Glob Oncol 2. © 2016 by American Society of Clinical Oncology Licensed under the Creative Commons Attribution 4.0 License

continued

Ariel Lopez-Chavez

Anish Thomas

Moses O. Evbuomwan

et al

pp 259-267




Breaking Cancer Bad News to Patients With Cancer: AComprehensive Perspective of Patients, Their Relatives, and thePublic—Example From a Middle Eastern Country

abstract

Purpose Empowering patients with cancer requires that they be continuously informed about theircondition. In some Eastern cultures, this concept is often opposed by caregivers. We aim to compare theextent of disclosure desired by actual and presumed patients with cancer and their relatives in ourpractice.

Methods Nine questions reflecting possible bad news communication to patients from diagnosis to theend of life were designed to investigate the extent of desired disclosure and were answered by 100patients (cohort I) and 103 accompanying relatives (cohort II) in an outpatient setting. In addition, 103public participants attending a family medicine clinic (cohort III) each answered the questions from theperspective of a presumed patient (cohort IIIA) and the perspective of a relative (cohort IIIB). The primaryend point was affirmative response to six or more questions (AR ‡ 6), reflecting a preference to beinformed of the majority (‡ 67%) of possible bad news.

Results AR ‡ 6 was reported in 85% of cohort I and 52% of cohort II (x2 P < .001). On multivariableanalysis, AR ‡ 6 showed significant association with being a patient (in cohorts I and II) and havingnonmetastatic disease (only in cohort I). In the public cohort, AR ‡ 6 was reported in 91% in cohort IIIAand 63% in cohort IIIB (x2 P < .001). On multivariable analysis, AR ‡ 6 in cohort III was significantlyassociated with being a presumed patient and having at least a college education.

Conclusion Patients desire disclosure of the majority of cancer-related bad news. This is in contrast to theviews and requests of relatives. The public participants would also desire similar disclosure if they wereto be diagnosed with cancer.


continued

Jamal Zekri

Syed Mustafa Karim

pp 268-274




Prioritizing US Cervical Cancer Prevention With Results From aGeospatial Model

abstract

Purpose To determine if differences in screening and vaccination patterns across the population mayaccentuate ethnic and geographic variation in future burden of disease.

Methods Using Cancer in North America data provided by the North American Association of CentralCancer Registries, county cervical cancer incidence trends from 1995 to 2009 were modeled for theentire United States using ecologic covariates. Rates for health service areas were also modeled byethnicity. State-level incidence was mapped together with Papanicolaou (Pap) screening, past 3 years(women ‡ 18 years old), and three-dose human papillomavirus (HPV) vaccine coverage (girls 13 to 17years old) to identify potential priority areas for preventive services.

Results US cervical cancer incidence decreased more during the periods 1995 to 1999 and 2000 to2004 than during the period 2005 to 2009. During these 15 years, the most affected areas becameincreasingly confined to Appalachia, the lower Mississippi Valley, the Deep South, Texas, and Florida.Hispanic and black women experienced a higher incidence of cervical cancer than both white and Asianand Pacific Islander women during each period. Women in 10 of 17 states/districts with a high incidence(‡ 8.14/100,000) reported low Pap testing (< 78.5%), HPV vaccine coverage (< 33.9%), or bothprevention technologies.

Conclusion The decline in cervical cancer incidence has slowed in recent years. Access to HPVvaccination, targeted screening, and treatment in affected populations is needed to reduce cervicalcancer disparities in the future.


continued

Jonathan K. Kish

Alicia I. Rolin

Zhaohui Zou

et al

pp 275-283




Phenotype and Treatment of Breast Cancer in HIV-Positive and-Negative Women in Cape Town, South Africa

abstract

Purpose An estimated 5.9 million people in South Africa are infected with HIV. Because antiretroviraltherapy has made infection with HIV a treatable, chronic condition, HIV-infected individuals are nowsurviving to middle and older age. We investigated the implications of HIV status for breast cancer inSouth Africa.

Methods We compared clinical and demographic characteristics of women newly diagnosed with a firstprimary breast cancer at Tygerberg Hospital, Cape Town, South Africa, from January 2010 to December2011 by HIV status. We then compared HIV-positive patients with HIV-negative controls, matched 2:1 onage and ethnicity, with respect to chemotherapy regimens, toxicities, completion of systemic chemo-therapy, and changes in CD4 cell count.

Results Of 586 women with breast cancer, 31 (5.3%) were HIV positive, 420 (71.7%) were HIV negative,and 135 (23%) were untested for HIV. Women with HIV were younger than other women (P < .001). Thegroups did not differ in regard to stage at presentation, histologic subtype, tumor grade, nodal in-volvement, or hormone receptor positivity. More than 84% of patients who initiated systemic chemo-therapy, regardless of HIV status, completed it without serious toxicity. Among HIV-positive patientsreceiving chemotherapy, the mean baseline CD4 cell count was 477 cells/mL (standard deviation, 160cells/mL), and the mean nadir was 333 cells/mL (standard deviation, 166 cells/mL).

Conclusion HIV-infected women were younger at breast cancer diagnosis than HIV-negative women butotherwise similar in phenotype and completion of chemotherapy. Longer term follow-up is needed toevaluate the effects of HIV, antiretroviral therapy, and chemotherapy on the survival and quality of life ofpatients with breast cancer.


continued

Lizanne Langenhoven

Pieter Barnardt

Alfred I. Neugut

et al

pp 284-291




Compulsory Licenses for Cancer Drugs: Does Circumventing PatentRights Improve Access to Oncology Medications?

abstract

Worldwide, there are enormous inequities in cancer control that cause poor outcomes among patientswith cancer who live in low- and middle-income countries (LMICs). One of the biggest challenges thatoncology faces today is how to increase patient access to expensive, but life-saving, therapies in LMICs.Access to cancer medications in LMICs is a major problem, especially in recent years, as the costs ofthese therapies continue to rise exponentially. One mechanism available to LMICs to improve access tocancer medications allows a country to pursue a compulsory license for a given drug. Here, we willreview how the legal framework in the World Trade Organization’s Trade-Related Aspects of IntellectualProperty Rights Agreement and the Doha Declaration supports countries to circumvent patent laws andacquire compulsory licenses for essential medicines. We will also discuss the current and future role ofcompulsory licenses in oncology and how compulsory licenses may improve access to cancer drugs inLMICs.


continued

Cinthia Leite FrizzeraBorges Bognar

Brittany L. Bychkovsky

Gilberto de LimaLopes Jr

pp 292-301




Breast Cancer and African Ancestry: Lessons Learned at the 10-YearAnniversary of the Ghana-Michigan Research Partnership andInternational Breast Registry

executivesum

mary

Women with African ancestry in western, sub-Saharan Africa and in the United States represent apopulation subset facing an increased risk of being diagnosed with biologically aggressive phenotypes ofbreast cancer that are negative for the estrogen receptor, the progesterone receptor, and the HER2/neumarker. These tumors are commonly referred to as triple-negative breast cancer. Disparities in breastcancer incidence and outcome related to racial or ethnic identity motivated the establishment of theInternational Breast Registry, on the basis of partnerships between the Komfo Anokye Teaching Hospitalin Kumasi, Ghana, the University of Michigan Comprehensive Cancer Center in Ann Arbor, Michigan, andthe Henry Ford Health System in Detroit, Michigan. This research collaborative has featured educationaltraining programs as well as scientific investigations related to the comparative biology of breast cancerin Ghanaian African, African American, and white/European American patients. Currently, the In-ternational Breast Registry has expanded to include African American patients throughout the UnitedStates by partnering with the Sisters Network (a national African American breast cancer survivors’organization) and additional sites in Ghana (representing West Africa) as well as Ethiopia (representingEast Africa). Its activities are now coordinated through the Henry Ford Health System International Centerfor the Study of Breast Cancer Subtypes. Herein, we review the history and results of this internationalprogram at its 10-year anniversary.


continued

Evelyn Jiagge

Joseph Kwaku Oppong

Jessica Bensenhaver

et al

pp 302-310




Management and Care of Women With Invasive Cervical Cancer:American Society of Clinical Oncology Resource-Stratified ClinicalPractice Guideline

executivesum

mary

Purpose To provide evidence-based, resource-stratified global recommendations to clinicians andpolicymakers on the management and palliative care of women diagnosed with invasive cervical cancer.

Methods ASCO convened a multidisciplinary, multinational panel of cancer control, medical and radiationoncology, health economic, obstetric and gynecologic, and palliative care experts to produce recommen-dations reflecting resource-tiered settings. A systematic review of literature from 1966 to 2015 failed to yieldsufficiently strong quality evidence to support basic- and limited-resource setting recommendations; a formalconsensus-based process was used to develop recommendations. A modified ADAPTE process was also usedto adapt recommendations from existing guidelines.

Results Five existing sets of guidelines were identified and reviewed, and adapted recommendations formthe evidence base. Eight systematic reviews, along with cost-effectiveness analyses, provided indirectevidence to inform the consensus process, which resulted in agreement of 75% or greater.

Recommendations Clinicians and planners should strive to provide access to the most effective evidence-based antitumor and palliative care interventions. If a woman cannot access these within her own orneighboring country or region, shemay need to be treated with lower-tier modalities, depending on capacityand resources for surgery, chemotherapy, radiation therapy, and supportive and palliative care. For womenwith early-stage cervical cancer in basic settings, cone biopsy or extrafascial hysterectomy may beperformed. Fertility-sparing procedures or modified radical or radical hysterectomy may be additionaloptions in nonbasic settings. Combinations of surgery, chemotherapy, and radiation therapy (includingbrachytherapy) should be used for women with stage IB to IVA disease, depending on available resources.Pain control is a vital component of palliative care. Additional information is available at www.asco.org/rs-cervical-cancer-treatment-guideline and www.asco.org/guidelineswiki. It is the view of ASCO that healthcare providers and health care system decision makers should be guided by the recommendations for thehighest stratum of resources available. The guideline is intended to complement but not replace localguidelines.


Linus T. Chuang

Sarah Temin

Rolando Camacho

et al

pp 311-340


http://www.asco.org/rs-cervical-cancer-treatment-guideline


http://www.asco.org/guidelineswiki


editorial

Strengthening Global Partnership inBreast Cancer ResearchSee accompanying article on page 302

The accompanying article by Jiagge et al1 raisesnumerous issues critical to global breast cancerresearch and control. It is important to note thatbreast cancer is the most common cancer amongwomen as well as the leading cause of cancerdeaths among women worldwide. Understandingthe different patterns of breast cancer around theglobe can facilitate prevention, early diagnosis,and treatment.

The global diaspora of peoples has major impli-cations for health and disease and can providecritical insights for understanding etiology andimproving treatment. Historically, for example,women in East Asia have experienced lower risksof breast cancer than women of Europeandescent living in Europe and North America.Studies on women of East Asian descent whomigrated to North America, however, have dem-onstrated rates of breast cancer comparable tothose of women of European descent amongsecond-generation immigrants, with intermedi-ate rates among first-generation women. TheMichigan-Ghana study team report a higher pro-portion of triple-negative breast cancer (TNBC)among women in West Africa compared withwomen of European descent living in North Amer-ica, with North American women of mixed Africanand European descent falling in between the twoproportions. Because TNBC is most commonamong women with premenopausal onset ofbreast cancer, the age distribution of the reportedpatients with breast cancer is important to con-sider. Theauthors state that the rankingspersistedamong women younger than age 50 years.More recently, the Michigan-Ghana group hasexpanded their work to include breast can-cers among Ethiopian women. Rates of TNBCare much lower among these women comparedwith those in West Africa, consistent with geneticadmixture with peoples of Arab descent. Theirpreliminary work suggests potential biologic un-derpinnings, which hopefully will be followed by

more expansive investigations of multiple factors.A key remaining question is to what extent theseobserved differences in the proportion of specificsubtypesofbreast cancerarea result of underlyinggenetic factors, constitutional or environmentalfactors, or, most likely, a combination of these riskfactors. Studying differences in exposures acrossvaried populations may help in detecting clues forprevention.

It will be important to link this work with studiesof breast cancer among other populations of Afri-can descent, both in Africa as well as in the Ca-ribbean and Central and South America. It hasbeen estimated that more Africans were forciblybrought through the slave trade to South America,particularly Brazil, than were brought to NorthAmerica. We know relatively little about the mo-lecular characteristics of breast cancer inMozam-bique and Angola, sites from which many of theAfricans taken toBrazil originated.Wealsoneed tostrengthen research on the molecular character-istics of breast cancer occurring among women ofNative American ancestry and those with admix-tures of Native American, European, and Africandescent. Suchworkwill requireclosecollaborationwith colleagues across Africa, the Caribbean, andCentral and South America.

The work by Jiagge et al1 also makes clear theimportance of building the infrastructure for di-agnosing and treating breast cancer in low- andmiddle-income countries. One of the first steps isto build awareness among the general public andhealth care providers, both traditional medicineand allopathic practitioners, of the importance ofprompt evaluation of breast masses. Another keystep is to facilitate efficient and timely evaluationof breast masses with fine-needle aspiration,core needle biopsies, ultrasound, and diagnosticmammography, accompanied by the appropri-ate training in biopsy technique, anatomic pa-thology, and imaging. The presence of a reliablesupply chain for biopsy needles, glass slides, and

Edward L. Trimble

Kathy J. Helzlsouer

National Cancer Institute,National Institutes ofHealth, Rockville, MD

Corresponding author:Edward L. Trimble, MD,9609 Medical Center Dr,Room 3W562, Rockville,MD20850; e-mail: [email protected].

253 Volume 2, Issue 5, October 2016 jgo.ascopubs.org JGO – Journal of Global Oncology

© 2016 by American Society of Clinical Oncology Licensed under the Creative Commons Attribution 4.0 License




pathology reagents is also critical. The develop-ment of the appropriate human resources forhealth is also central. Themultidisciplinaryman-agement and complex treatment of breast can-cer requires training and career support forlay health workers, nurses, doctors, and tech-nicians. Methods to efficiently deliver complextreatment in aneffective andcost-efficientmannerare critically needed for low-resource environ-ments. Building the overall capacity of the healthcare infrastructure in low-resource communitiesalso requires making health care affordable andaccessible.

Finally, the article by Jiagge et al1 highlights theimportance of sustained collaboration and men-torship. The commitment by Dr Newman and hercolleagues at the University of Michigan andthe Henry Ford Health System has been long-standing. This partnership has helped improvethe standard of care as well as broaden our un-derstanding of breast cancer etiology and genet-ics. We need to encourage many more suchmodels of collaboration to reduce the global bur-den of breast cancer.

DOI: 10.1200/JGO.2016.005264Published online on jgo.ascopubs.org on July 27, 2016.

AUTHOR CONTRIBUTIONS

Manuscript writing: All authorsFinal approval of manuscript: All authors

AUTHORS’ DISCLOSURES OFPOTENTIAL CONFLICTS OF INTEREST

The following represents disclosure information provided byauthors of this manuscript. All relationships are consideredcompensated. Relationships are self-held unless noted. I 5

Immediate Family Member, Inst 5 My Institution. Relation-ships may not relate to the subject matter of this manuscript.For more information about ASCO’s conflict of interest policy,please refer to www.asco.org/rwc or jco.ascopubs.org/site/ifc.

Edward L. TrimbleNo relationship to disclose

Kathy J. HelzlsouerNo relationship to disclose

REFERENCES

reference

1. Jiagge E, Oppong JK, Bensenhaver J, et al: Breast cancer and African ancestry: Lessons learned atthe 10-year anniversary of the Ghana-Michigan research partnership and International BreastRegistry. J Glob Oncol 2:302-310, 2016


http://dx.doi.org/10.1200/JGO.2016.005264


http://www.asco.org/rwc

http://jco.ascopubs.org/site/ifc


commentaries

Breast Cancer Among Iraqi Women:Preliminary Findings From a RegionalComparative Breast Cancer ResearchProject

Breast cancer is the most frequent malignancyamong women worldwide, accounting for 25% ofall cancers, with an estimated 1.57 million newcases in 2012.1 It is also the leading cause offemale cancer-related deaths. Although substan-tial improvement in survival from this disease hasbeen reported in high-resource countries, the riskcontinues to increase, yielding highmortality ratesinmiddle- and low-incomecountries.1,2Within theEastern Mediterranean Region (EMR), cancer isthe fourth-ranked cause of death, after cardiovas-cular diseases, infectious diseases, and injuries.3

According to WHO classification, the EMR com-prises 21member states in theMiddle East, NorthAfrica, and Central Asia. The included nations areAfghanistan, Bahrain, Djibouti, Egypt, Iran, Iraq,Jordan, Kuwait, Lebanon, Libya, Morocco, Pales-tinian territory, Oman, Pakistan, Saudi Arabia,Somalia, Sudan, Syria, Tunisia, United Arab Emir-ates, and Yemen. The International Agency forResearch on Cancer (IARC) estimated that 292,677cases of cancer were newly diagnosed among thefemale population in EMR during 2012, and176,139 died of the disease.1 The five most com-monly recorded cancers in women were those ofthe breast, colorectum, cervix, ovary, and non-Hodgkin lymphoma. Overall, 99,000 cases wereregistered as breast cancer in that region.

In addition to being the most important cancer,other features that justify increasing efforts forbreast cancer control in the EMR include theexponential rise in the incidence and the higherprevalenceof affected youngwomenpresentingatadvanced stages of disease. These factors led tolow 5-year survival rates from breast cancer inmany countries within the region as comparedwith high-income settings.1,3-9

Breast cancer has become a major threat to fe-male health in Iraq, where it is the leading causeof death after cardiovascular diseases among

women, with a cancer-related mortality rate of23%.1,4,8,9 It has been the highest-ranked malig-nancy among the Iraqi population in general since1986. The latest Iraqi Cancer Registry8 revealedthat among an estimated population size of32,500,000, a total of 21,101 newcases of cancerwere registered in 2012; 9,268 were in men and11,833 were women.8 The crude incidence of allcancers was 61.69 per 100,000 (53.31 in menand 70.59 in women). During that year, 4,115cases of breast cancer were reported, accountingfor 19.5% of all newly diagnosed malignanciesand 34%of the registered female cancers, with anincidence approximating 22 per 100,000 femalepopulation. The highest frequency was observedinmiddle-agedwomen (45-49 years old), whereasthe peak age-specific incidence was reported inwomen 50-54 years old. It has been documentedthat there is a tendency for the disease to bediagnosed at advanced stages, with a likely prev-alence of poorly differentiated tumor forms il-lustrated in significantly high rates of nuclearaneuoploidy, thus yielding amortality incidenceof approximately 60%.1,4,8-10

At the Main Training Center for Early Detection ofBreast Cancer inBaghdad, it was reported in 2010that breast cancer was diagnosed in 19.8% ofwomen presenting with palpable breast lumps.4

Although 90.6% of those patients detected thelumps by themselves, only 32% sought medicaladvice within the first month. Approximately one-third of those patients were diagnosed in their fifthdecade of life, 47%presented at advanced stagesof disease, and 16% recalled a positive familyhistory.4 Another survey that was conductedto explore the level of knowledge, attitude,and practice toward breast cancer and breastself-examination (BSE) among educated Iraqiwomen demonstrated in 2012 that almost half ofthe participants had low knowledge scores.11

Nada A.S. Alwan

Nada A.S. Alwan, BaghdadUniversity, Baghdad, Iraq

Author’s disclosures ofpotential conflicts ofinterest are found at the endof this article.Corresponding author:Nada A.S. Alwan, MD,National Cancer ResearchCenter, BaghdadUniversity, Baghdad,Iraq; e-mail: [email protected]






Although 90% of the respondents have heardabout BSE, only 43% actually practiced the tech-nique. By multiple logistic regression analysis,researchers found that the level of knowledgeamong a university-affiliated population in north-ern Iraq and participants’ age were significantlyassociated with performing BSE.12

The outcomes of those studies obviously illustratesignificant knowledge gaps about the relative im-portance of breast cancer among the Iraqi com-munity and emphasize the urgent need forpractical policy decisions to promote early detec-tion through elevating the level of awareness. Ingeneral, the poor survival in less-developed coun-tries, including Iraq, ismainly attributed to the lackof strategic,well-designeddiagnostic policies cou-pled with inadequate treatment facilities.13

WHO, in collaboration with IARC, organized aconsultative Regional Meeting on Cancer ControlandResearch Priorities in Doha, Qatar, in October2013. The following recommendations weremade: strengthening cancer registration and sur-veillance, conducting priority research on canceretiology, and strengthening screening and earlydetection of priority cancers. It was concludedthat the most common cancers amenable to earlydetection in EMR are those of the breast, colo-rectum, and cervix, and it was emphasized thatstrengthening needs to be built on the best in-ternational evidence and existing regional experi-ence, taking into consideration the availableresources, challenges, and opportunities.14

In the context of breast cancer control, informationon the putative risk factors for breast cancerand the clinical profile of patients are of utmostimportance.3,15 Evaluating such processes inevi-tably demands ensuring the provision of an ap-propriate sustained database operated by trainedpersonnel. Within hospital records, in the majorityof countries belonging to EMR, there is improperdocumentation of critically important clinical fac-tors such as tumor size, nodal status, stage ofbreast cancer at initial diagnosis, hormonal re-ceptor status, frequency of distant metastasis,prevailing treatment modalities, and survival. Infact, most of the national cancer registries lackdata regarding stage distribution and survivalrates.

In an attempt to address these information needsin the clinical profile of patientswith breast cancer,and to emphasize the role of research as one of thebasic pillars in the adoption of a national cancercontrol strategy, a national breast cancer researchprogram was established in Iraq in 2009. Under

supervision of the IARC Screening Unit, the authordeveloped a comprehensive information systemdatabase for Iraqi patients diagnosed with breastcancer. In 2011, the WHO Eastern MediterraneanRegional Office proposed using that model to com-pare the demographic characteristics, clinicopath-ologic presentations, and management outcomesamong patients in the EMR affected with the dis-ease through implementing a regional comparativebreast cancer research project.10 An online infor-mation system supervised by IARC has been pro-vided to collect data systematically from patientsattending targeted breast cancer facilities belong-ing to eight member nations.

In Iraq, a preliminary analysis of the relevant data-base findings belonging to 855patients, diagnosedand treated for breast cancer, documented thefollowing results. Overall, 24.6% of the patientswere illiterate, 70.8% sought medical consultationwithin the sameyear after detecting abnormal signsor symptoms, and 46% were in their premeno-pausal age, whereas 35% were diagnosed in the45-54 year age group. Interestingly, 86.3% weremarried, merely 7% had their first childbirth afterthe age of 35 years, and only 8.5% were nullipara.History of lactation and hormonal therapy wasreported in 48% and 20.5%, respectively. Overall,35% documented a positive family history of ma-lignancy, and 18.5% confirmed having relativeswith breast cancer. Themainpresenting signswerepalpable breast lumps (94%), skin changes(9.8%), and bloody nipple discharge (4.7%). Bi-lateral breast cancerwas reported in4.6%.Accord-ing to TNM classification, 9.8% presented withstage I disease, and 46%were diagnosed in stagesIII and IV. Infiltrativeductal carcinomawas themostcommon pathology (67%), followed by intraduc-tal carcinoma (13.6%) and lobular carcinoma(18.5%). Less than 7% of malignant tumors werewell differentiated. More than two-thirds of thepatients (65.5%) had positive lymph node involve-ment at the time of initial diagnosis. Immunohisto-chemical assays demonstrated that estrogen,progesterone and Her2 receptors were positive in67%, 69%, and 49.2%of specimens, respectively.The majority of patients (92.3%) were provisionallytreated by modified radical mastectomy, 35.2%received palliative treatment, hormonal therapywas prescribed to 54.2%, and recurrence wasregistered in 9.4%.

Comparing our statistics with those reported inhigh-resource settings (eg, the United States),obvious significant differences are displayed spe-cifically regarding the stage distribution of thedisease, with 61%, 32%, and 6%of breast cancer



cases present at localized, regional, and distantstages, respectively.16 Nevertheless, Iraq, catego-rizedasamiddle-incomecountrybyWHO/EasternMediterranean Regional Office,17 documents farbetterprognostic indicators than those recorded inlow-resource settings such as Eritrea in easternAfrica, where two-thirds of the cases are detectedin advanced stages; the mean duration from theonset of symptoms to the time of seeking medicaladvice approaches 3 years.18

In general, these findings justify increasing effortsto establish comprehensive breast cancer controlprograms in Iraq, focusing provisionally on pro-moting education and early diagnosis as majorapproaches to controlling thedisease. The strikingpatterns of breast cancer among women in ourregion highlight the urgent need to consider earlydetection a priority.

DOI: 10.1200/JGO.2015.003087Published online on jgo.ascopubs.org on March 16, 2016.

AUTHOR’S DISCLOSURES OFPOTENTIAL CONFLICTS OF INTEREST

The following represents disclosure information provided byauthors of this manuscript. All relationships are consideredcompensated. Relationships are self-held unless noted. I 5Immediate Family Member, Inst 5 My Institution. Relation-ships may not relate to the subject matter of this manuscript.For more information about ASCO’s conflict of interest policy,please refer to www.asco.org/rwc or jco.ascopubs.org/site/ifc.

Nada A.S. AlwanNo relationship to disclose

ACKNOWLEDGMENTI thank R. Sankaranaravanan, MD (Head, Screening Unit/IARC), Eric Lucas, and Ibtihal Fadhil, MD (Regional Advisor,NCD, EMRO) for the support and supervision of IARC and theWHO Eastern Mediterranean Regional Office.

REFERENCES1. International Agency for Research on Cancer: Globocan 2012. Lyon, France, World Health Or-

ganization International Agency for Research on Cancer, 2013

2. Anderson BO, Yip CH, Smith RA, et al: Guideline implementation for breast healthcare in low-income and middle-income countries: Overview of the Breast Health Global Initiative GlobalSummit 2007. Cancer 113(8 Suppl):2221-2243, 2008

3. World Health Organization. Towards a Strategy for Cancer Control in the Eastern MediterraneanRegion (ed. 1). Cairo, Egypt, Regional Office for the Eastern Mediterranean.World Health Orga-nization, 2010

4. Alwan NAS: Breast cancer: Demographic characteristics and clinico-pathological presentation ofpatients in Iraq. East Mediterr Health J 16:1159-1164, 2010

5. El-Zaemey S, Nagi N, Fritschi L, et al: Breast cancer among Yemeni women using the NationalOncology Centre Registry 2004-2010. Cancer Epidemiol 36:249-253, 2012

6. Khokher S, Qureshi MU, Riaz M, et al: Clinicopathologic profile of breast cancer patientsin Pakistan: Ten years data of a local cancer hospital. Asian Pac J Cancer Prev 13:693-698,2012

7. Arkoob K, Al-Nsour M, Al-Nemry O, et al: Epidemiology of breast cancer in women in Jordan:Patient characteristics and survival analysis. East Mediterr Health J 16:1032-1038, 2010

8. Iraqi Cancer Board. Results of the Iraqi Cancer Registry 2012. Baghdad, Iraq, Iraqi Cancer RegistryCenter, Ministry of Health, 2015

9. Al Alwan NA: DNA proliferative index as a marker in Iraqi aneuploid mammary carcinoma. EastMediterr Health J 6:1062-1072, 2000

10. AlwanN: Iraqi initiative of a regional comparative breast cancer research project in theMiddle East.J Cancer Biol Res 2:1016-1020, 2014

11. AlwanNA, Al-AttarWM, EliessaRA, et al: Knowledge, attitude and practice regarding breast cancerand breast self-examination among a sample of the educated population in Iraq. East MediterrHealth J 18:337-345, 2012

12. Alwan N, et al: Knowledge, attitudes and practice towards breast cancer and BSE in KirkukUniversity, Iraq. Asian Pac J Reprod 1:308-311, 2012







13. Von Karsa L, Qiao Y, Ramadas K, et al: Prevention/Screening Implementation, in Stewart BW andWild CP (eds): World Cancer Report 2014. Lyon, France, World Health Organization InternationalAgency for Research on Cancer, 2014

14. Miller T, Alwan N: Concept note on screening and early detection of breast cancer in the EasternMediterranean Region. Regional Meeting on Cancer Control and Research Priorities, WHO/EMR.Doha, Qatar, October 22-24, 2013

15. International Agency for Research on Cancer. Breast Screening. Handbooks on Cancer Pre-vention, vol. 15. Lyon, France, International Agency for Research on Cancer (in press)

16. National Cancer Institute. Fact Sheets: Female Breast Cancer. Bethesda, MD, NIH, NationalCancer Institute, Surveillance, Epidemiology and End Result Program, 2012

17. World Health Organization. Health system strengthening in countries of the EasternMediterraneanRegion. Regional Committee for the EasternMediterranean, 59th session. Provisional agenda item3. Cairo, Egypt, WHO Regional Office for the Eastern Mediterranean, 2013

18. Tesfamariam A, Gebremichael A, Mufunda J: Breast cancer clinicopathological presentation,gravity and challenges in Eritrea, East Africa:Management practice in a resource-poor setting. S AfrMed J 103:526-528, 2013



originalreport

EGFR Mutations in Latinos From theUnited States and Latin America

abstract

Purpose Epidermal growth factor receptor (EGFR) mutations confer sensitivity to EGFR tyrosine kinaseinhibitors in patients with advanced non–small-cell lung cancer (NSCLC). There are limited and conflictingreports on the frequency of EGFR mutations in Latinos.

Patients and Methods Samples from 642 patients with NSCLC from seven institutions in the United Statesand Latin America were assessed for EGFRmutations (exons 18 to 21) at Clinical Laboratory ImprovementAmendments-certified central laboratories.

Results EGFRmutation analysis was successfully performed in 480 (75%) of 642 patients; 90 (19%) wereLatinos, 318 (66%) were non-Latino whites, 35 (7%) were non-Latino Asians, 30 (6%) were non-Latinoblacks, and seven (2%) were of other races or ethnicities. EGFR mutations were found in 21 (23%) of 90Latinos with varying frequencies according to the country of origin; Latinos from Peru (37%), followed bythe United States (23%),Mexico (18%), Venezuela (10%), and Bolivia (8%). In never-smoker Latinos andLatinos with adenocarcinoma histology, EGFR mutation frequencies were 38% and 30%, respectively.Therewas a significant difference in the frequency of EGFRmutations among the different racial and ethnicsubgroups analyzed (P < .001), with non-Latino Asians having the highest frequency (57%) followed byLatinos (23%), non-Latino whites (19%), and non-Latino blacks (10%). There was no difference betweenLatinos (23%) and non-Latinos (22%; P = .78) and Latinos and non-Latino whites (P = .37). Patients fromPeru had an overall higher frequency of mutations (37%) than all other Latinos (17%), but this differenceonly exhibited a trend toward significance (P = .058).

Conclusion There was no significant difference between the frequency of EGFR mutations in NSCLC inLatinos and non-Latinos.


INTRODUCTION

Racial and ethnic disparities in the incidence,stage at diagnosis, treatment, and survival of pa-tients with lung cancer have been described;however, the reasons for these disparities arenot completely understood.1 It is estimated thatthe ethnic and racial composition of the UnitedStates will change dramatically in the next fewdecades.2 Given this, it is paramount that racialand ethnic health care disparities be studied andsteps taken to have a positive impact on healthoutcomes. To effectively target and eliminate lungcancer–related health care disparities, a betterunderstanding of the molecular characteristicsof the disease and their relationship with raceand ethnicity is needed.

Activating mutations in the epidermal growth fac-tor receptor (EGFR) gene confer sensitivity toEGFR tyrosine kinase inhibitors in patients withadvanced non–small-cell lung cancer (NSCLC).3-6

Racial, ethnic, and demographic differences in the

distribution of such mutations have been previouslydescribed. For example, activating EGFRmutationsare significantly more common in East Asians, wom-en, and never-smokers.7 The frequency of EGFRmutations varies from approximately 10% of lungadenocarcinomas in North America and Europe toas high as 50% to 60% inAsia.8,9However, there arelimited and conflicting reports of the frequency ofEGFRmutations among Latinos (also referred to asHispanics). Published reports have been limited toretrospective cohorts that are constrained, amongother factors, by patient selection for mutation test-ing based on clinical characteristics.10,11

According to the US Census Bureau, Latinoscurrently comprise 17.4% of the US populationand are projected to grow to 29% of the populationby 2060—more than one quarter of the total pop-ulation.2 Lung cancer is the third most commonlydiagnosedcancer amongLatinomenandwomen; itis the leading cause of cancer death among Latinomen and the second-leading cause among Latino

Ariel Lopez-Chavez

Anish Thomas

Moses O. Evbuomwan

Liqiang Xi

Guinevere Chun

Tatiana Vidaurre

Oscar Arrieta

George Oblitas III

Ana Belen Oton

Alejandro R. Calvo

Arun Rajan

Mark Raffeld

Seth M. Steinberg

Lorena Arze-Aimaretti

Giuseppe Giaccone

Author affiliations appear atthe end of this article.

A.L.-C. and A.T.contributed equally to thiswork.

Supported by the NationalCancer Institute (NCI)Intramural ResearchProgram, the KnightCancer Institute at OregonHealth and ScienceUniversity, and the NCICancer TherapyEvaluationProgram.Presented in part at the51stAnnualMeetingof theAmerican Society ofClinical Oncology,Chicago, IL, May 29-June2, 2015.

Authors’ disclosures ofpotential conflicts ofinterest and contributionsare found at the end ofthis article.Corresponding author:Giuseppe Giaccone, MD,PhD, GeorgetownUniversity, 3970Reservoir Rd, Washington,DC 20007; e-mail:[email protected].





women.12 The incidence, clinical course, and out-comes of lung cancer among Latinos are distinctfromnon-Latinowhites, who currently constitute thelargest racial and ethnic group in the United States.Lung cancer incidence rates are lower among La-tinos13; however, despite the lower socioeconomicstatus,more limitedaccess tocare,anddiagnosesatadvancedstagesofdisease thatwouldpredictother-wise, lung cancer mortality rates are 50% loweramong Latinos than non-Latino whites.13,14 Al-though this so-called Latino paradox is still a matterofdebate, given the limited reports todate,10,11 thereis a clear need for amore comprehensivemolecularcharacterization of NSCLC in this ethnic group tobetter understand outcome disparities in theUnitedStates. In particular, patients with EGFR mutationsare known to have better survival than patients withwild-type EGFR.15

In this prospective study, using central testing, wesought to characterize the types and frequenciesof EGFR-activating mutations among Latino pa-tients with NSCLC living in the United States andLatin America and compare them with US non-Latinos.

PATIENTS AND METHODS

Participants

This study was conducted at the following institu-tions: National Cancer Institute (NCI; Bethesda,MD; coordinating center), Oregon Health andScience University (OHSU; Portland, OR), Insti-tuto Nacional de Enfermedades Neoplasicas(Lima, Peru), Instituto Nacional de Cancerologia(Mexico City, Mexico), Hospital Oncologico LuisRazetti (Caracas,Venezuela),DenverHealthMed-ical Center (Denver, CO), and Universidad Mayorde San Simon (Cochabamba, Bolivia).

Samples from patients with histologically con-firmed NSCLC were collected from two proto-cols that were registered with clinicaltrials.gov:NCT01306045 (CUSTOM; Molecular Profilingand Targeted Therapies in Advanced Thoracic Ma-lignancies Trial) and NCT01255150 (Frequencyof EGFR Mutations in Latinos With Non-Small CellLung Cancer). Patients eligible for NCT01306045had histologically confirmed advanced NSCLC forwhom surgical resection or multimodality therapywith curative intent was not feasible. Patients hadto have biopsiable disease and be willing to under-go biopsy for molecular profiling or had to haveenough and adequate archival material from a pre-vious biopsy to perform molecular profiling analy-ses. For protocol NCT01255150, all samples werecollected frompatientswith histologically confirmedNSCLC who were dead at the time of tissue

collection. For the purposes of both studies, theterms Hispanic or Latino included individuals whoself-identified as Hispanic or Latino or were born inany Latin American country.

EGFR mutation testing was performed centrallyat Clinical Laboratory Improvement Amendments(CLIA) –certified central laboratories of the NCIor OHSU. EGFR mutation analysis of exons 18through 21 was performed with either pyrose-quencing at the NCI or Sequenom MassArray(San Diego, CA) at OHSU. The institutional reviewboards of the participating centers approved theresearch protocol, and all living participants pro-vided written informed consent. Tumor samplesfrom dead individuals for whom basic clinical in-formation was available were included in thisanalysis.

Pyrosequencing

DNA was extracted from paraffin-embedded tis-sue sections using the QiagenQIAampDNAFFPETissue Kit (Hilden, Germany), according to man-ufacturer’s instructions. For identification of pointmutations, coamplifications at lower denaturationtemperature–polymerase chain reaction (PCR;cold PCR) were performed either individually orin a single 96-well microtiter plate (complete genepanel) in an Applied Biosystems (ABI) 9700 ther-mocycler (Foster City, CA). After PCR, the prod-ucts were subjected to pyrosequencing on aQiagen PyroMark Q24 system. For identificationof deletions and insertions, independent PCR re-actions were performed with fluorescein-labeledprimers in the ABI 9700, and the products wereanalyzed by capillary electrophoresis on an ABI3130xl Genetic Analyzer. Five PCR reactions weredesigned to interrogate the most commonly oc-curringmutations, including deletionmutations inexon 19, point mutations (codons 858, 861, and863) in exon 21, insertions and point mutations inexon 20 (codon790), andmutations at codon 719in exon 18. The products of the exon 20, exon 21,and exon 18 reactions were analyzed by pyrose-quencing, and the products of deletion mutationsin exon19and insertions in exon20wereanalyzedby capillary electrophoresis, as described.16

Sequenom MassArray

Initial PCR reactions were set upwith an EpMotion5075 liquid handler (Eppendorf, Hamburg, Ger-many) and used 10-ng DNA per multiplex in a to-tal volume of 5 ml, with 100 nmol/L of primers,2 mmol/L of magnesium chloride, 500 mmol/Lof deoxynucleotide, and 0.1 units of Taq polymer-ase. Amplification included one cycle at 94°C for



4 minutes, followed by 45 cycles at 94°C for 20seconds, 56°C for 30 seconds, and 72°C for1minute and one final cycle at 72°C for 3 minutes.Unincorporated nucleotides were inactivated byaddition of 0.3 units of shrimp alkaline phospha-tase and incubation at 37°C for 40 minutes, fol-lowed by heat inactivation of shrimp alkalinephosphatase at 85°C for 5 minutes. Single-baseprimer extension reactions were performed with0.625 to1.25mmol/L of extensionprimer and1.35units of TypePLEX thermosequenase DNA poly-merase (Sequenom). Extension cycling includedone cycle at 94°C for 30 seconds, and 40 cyclesat 94°C for 5 seconds, five cycles at 52°C for5 seconds and 80°C for 5 seconds, followed byone cycle at 72°C for 3 minutes. Extension prod-ucts were purified by incubation for 30 minutes withan ion exchange resin (SpectroCLEAN; Sequenom),and approximately 10 nl of purified product wasspotted onto SpectroChip II matrices (Sequenom).A Bruker matrix-assisted laser desorption/ionizationtime-of-flightmassspectrometer (MassARRAYCom-pact; Sequenom) was used to resolve extensionproducts. MassARRAY Typer Analyzer software(Sequenom) was used for automated data anal-ysis, accompanied by visual inspection of extensionproducts.

Statistical Analyses

Analyses were performed using Fisher’s exact testor Mehta’s modification to Fisher’s exact test todetermine relationships among parameters,whether for all patients or by adenocarcinoma ornot, smoking or not, and center and countryeffects. All P values are two tailed and reportedwithout adjustment for multiple comparisons.

RESULTS

Demographic Characteristics

We collected samples from 642 patients withNSCLC fromseven institutions in theUnitedStatesand Latin America. EGFR mutation analysis wassuccessfully performed in 480 (75%) of 642.Among patients who had a successful mutationanalysis, 90 (19%) were Latinos, 318 (66%) werenon-Latino whites, 35 (7%) were non-LatinoAsians, 30 (6%) were non-Latino blacks, andseven (2%) were of other races or ethnicities.Women accounted for 53% (255 of 480) of pa-tientsandmen for46%(220of480).Mostpatientswere either age 40 to 64 years (259 [54%] of 480)or age 65 years or older (203 [42%] of 480;Table 1). Adenocarcinoma was the most frequenthistology (370 [77%] of 480), and never-smokersaccounted for 34% (165 of 480) of patients. The

majority of patients were from the United States(420 [88%] of 480), whereas patients from Peru,Bolivia, Venezuela, andMexico accounted for 6%,3%,2%,and2%, respectively.Most of thepatientswho enrolled were current or former smokers (306[64%] of 480) andhad advanceddisease (stage IIIbto IV; 447 [93%] of 480).

Among Latinos who had a successful EGFR mu-tation analysis, those from the United States (30[33%] of 90) and Peru (27 [30%] of 90) accountedfor themajority. In this population, 42% (38 of 90)were age older than 65 years, 54% (49 of 90) werewomen, 73% (66 of 90) had adenocarcinomahistology, 47% (42 of 90) were current or formersmokers, and 63% (57 of 90) had advanceddisease.

EGFR Mutations in Latinos

EGFR mutations were detected in 22% (105 of480) of patients overall (Table 2). Among Latinos,EGFRmutations were found in 23.3% (21 of 90),and the frequency varied based on the country oforigin. The highest frequency was observed inLatinos from Peru (37%) followed by the UnitedStates (23%), Mexico (18%), Venezuela (10%),and Bolivia (8%). Among non-Latinos, the overallfrequency of EGFR mutations was 21.5% (84 of390), and the frequency variedbasedon race. Thehighest frequency was observed in Asians(57.1%) followed by whites (18.6%) and blacks(10%; 95% CI, 2.11 to 26.53).

EGFR exon 19 deletions were the most frequenttype ofmutation amongLatinos (10 [47.6%] of 21;Table 3) followed by EGFR L858R and exon 20insertions, which were found in 38% (eight of 21)and 9.5% (two of 21) of patients, respectively.EGFR exon 19 deletions were also the most fre-quent subtype ofmutations amongwhites, Asians,and blacks, with frequencies ranging from 10%(three of 30) in blacks to 57% (20 of 35) in Asians.The frequency of EGFR L858R mutations was25% each in Asians (five of 20) and whites (15of 59); additionally, one (5%) of 20 Asians andseven (11.8%) of 59 whites had concurrent EGFRL858R and T790M mutations.

EGFR Mutation Frequency Among Racial or EthnicSubgroups

Therewas a significant difference in the frequencyof EGFRmutations among the different racial andethnic subgroups analyzed in the study (P, .05),which was primarily driven by the high frequencyof EGFRmutations seen in non-Latino Asians andthe low frequency observed in non-Latino blacks.However, there was no significant difference in



EGFRmutation frequency overall between Latinos(23%) andnon-Latinos (22%;P= .78; Table 4). Inaddition, there was no significant difference in thefrequency of EGFR mutations between Latinosand non-Latino whites (overall, 21 of 90 v 59 of318; P = .37 v .54 accounting for location) andbetweenpatients enrolled in theUnitedStates and

those enrolled in Latin America (non-US patients;P = .74). Although the EGFR mutation frequencyseemed to be higher in patients from Peru com-pared with patients of other races or ethnicities,this difference only exhibited a trend toward beingstatistically significant (P= .058), and the associationwas further diminished greatly when smoking and

Table 1 – Patient Demographic and Clinical Characteristics

Characteristic

Latinos

Non-Latinos

TotalWhite Asian Black Other

No. % No. % No. % No. % No. % No. %

Total 90 19 318 66 35 7 30 6 7 1 480 100

Age, years

18-39 3 3 9 3 2 6 0 0 0 0 14 3

40-64 46 51 166 52 27 77 18 60 2 29 259 54

. 65 38 42 142 45 6 17 12 40 5 71 203 42

Unknown 3 3 1 0 0 0 0 0 0 0 4 1

Sex

Male 36 40 146 46 14 46 17 57 7 100 220 46

Female 49 54 172 54 21 54 13 43 0 0 255 53

Unknown 5 6 0 0 0 0 0 0 0 0 5 1

Histologic subtype

Adenocarcinoma 66 73 243 76 33 76 23 77 5 71 370 77

Squamous cell carcinoma 11 12 37 12 1 12 3 10 1 14 53 11

NEC 2 2 11 3 0 0 2 7 0 0 15 3

NOS or other 11 12 27 8 1 12 2 7 1 14 42 9

Country

Bolivia 12 13 0 0 0 0 0 0 0 0 12 3

Mexico 11 12 0 0 0 0 0 0 0 0 11 2

Peru 27 30 0 0 0 0 0 0 0 0 27 6

United States 30 33 318 100 35 100 30 100 7 100 420 88

Venezuela 10 11 0 0 0 0 0 0 0 0 10 2

Smoking history

Never 40 44 90 28 24 69 9 30 2 29 165 34

Current or former 42 47 227 71 11 31 21 70 5 71 306 64

Unknown 8 9 1 0 0 0 0 0 0 0 9 2

Disease stage

I 0 0 0 0 0 0 0 0 0 0 0 0

II 1 1 0 0 0 0 0 0 0 0 1 0

IIa 1 1 0 0 0 0 0 0 0 0 1 0

IIb 6 7 0 0 0 0 0 0 0 0 6 1

III 1 1 0 0 0 0 0 0 0 0 1 0

IIIa 8 9 0 0 0 0 0 0 0 0 8 2

IIIb to IVa 57 63 318 100 35 100 30 100 7 100 447 93

Unknown 16 18 0 0 0 0 0 0 0 0 16 3

Abbreviations: NEC, neuroendocrine cancers; NOS, not otherwise specified.



adenocarcinoma histology subgroups were an-alyzed independently (data not shown).

DISCUSSION

In this study, which is the largest to our knowledgeto prospectively evaluate the frequency of EGFRmutations in CLIA-certified laboratories in Latinopatients with NSCLC from the United States andLatin America, there was no difference in thefrequency of EGFR mutations between Latinosand non-Latinos. Although we observed a signif-icant difference in the frequency of EGFR muta-tions among the different racial and ethnicsubgroups studied, this difference was not signif-icant when non-Latino Asians and blacks wereremoved from the analysis. This is consistent with

previous reports that have observed a significantlyhigher frequency of EGFR mutations in Asians.Furthermore, the samples analyzed were part of amultiethnic and international cohort of patientswith NSCLC, which allowed us to make compar-isons among racial and ethnic groups rather thanrelying on historical controls. Finally, an additionalstrength of this study was that molecular testingwas performed in two CLIA-certified central labo-ratories, avoiding quality control issues and po-tential variability in EGFR mutation frequencyresulting fromdifferences in testingmethods usedin different countries.

However, this study has limitations that under-score important aspects of cancermolecular char-acterization studies in Latinos and other under-represented minority populations in the UnitedStates. First, enrollment of Latino patients in theUnited States was low (30 [7%] of 420), despitethe fact that one of the protocols in this study(NCT01255150) was specifically geared towardenrollment of such patients. Furthermore, despitethe establishment of collaborations with severallarge cancer centers in Latin America to addressthe challenges described in this article, the num-ber of Latinos overall was low. This problem high-lights the significant difficulties of researchcollaborations with developing countries in whichresourceconstraints, logistic, and legal challengesmay significantly affect enrollment. This is in linewith previous experience in other molecular char-acterization studies and clinical trials. Further-more, it partially explains the paucity of tumor-specificmolecular characterization data in Latinosand other minority populations in the UnitedStates. In this particular case, although the first

Table 2 – EGFR Mutation Frequency

Race or Ethnicity Positive Negative No. Frequency (%) 95% CI

Latino

All 21 69 90 23.3 15.06 to 33.43

Peru 10 17 27 37.0 19.40 to 57.63

United States 7 23 30 23.3 9.93 to 42.28

Mexico 2 9 11 18.2 2.28 to 51.78

Venezuela 1 9 10 10.0 0.25 to 44.50

Bolivia 1 12 13 7.7 0.19 to 36.03

Non-Latino

All 84 306 390 21.5 17.56 to 25.96

White 59 259 318 18.6 14.43 to 23.27

Asian 20 15 35 57.1 39.35 to 73.68

Black 3 27 30 10.0 2.11 to 26.53

Other 2 5 7 28.6 3.67 to 70.96

Table 3 – EGFR Mutation Subtypes

Subtype

Latinos

Non-Latinos

TotalWhite Asian Black Other

No. % No. % No. % No. % No. % No. %

Total 90 19 318 66 35 7 30 6 7 1 480 100

No. of positive mutations 21 23 59 19 20 57 3 10 2 29 105 22

Exon 19 deletion 10 47.6 22 37.28 10 50 1 3.33 1 50 44 41.9

Exon 19 deletion plus T790M 0 0 3 5.08 3 15 1 3.33 0 0 7 6.66

L858R 8 38 15 25.42 5 25 0 0 1 50 29 27.6

L858R plus T790M 0 0 7 11.86 1 5 0 0 0 0 8 7.61

T790M alone 0 0 1 1.69 0 0 0 0 0 0 1 0.95

Other exon 20 insertion 2 9.52 6 10.16 0 0 1 3.33 0 0 9 8.57

Other 1 4.76 5 8.47 1* 5 0 0 0 0 7 6.66

Total 21 99.88 59 99.96 20 100 3 9.99 2 100 105 99.95

*Insertion exon 19 Asian other.



reports of the frequency of EGFR mutations inNSCLC in the United States date back to2004,17 the initial report for Latinos in the UnitedStates came almost a decade later,11 and the totalnumber of reports and patients tested by race orethnicity differs significantly betweenmajority andminority populations. Of utmost importance is thatin the event the cancer molecular profile of aparticular minority population would differ signif-icantly from the majority, such a disparity in mo-lecular dataavailability could lead to theworseningof the already existing outcome disparities amongpatients of different races or ethnicities in theUnited States.

Another potential limitation of our study is that wedid not account for the significant racial andgenetic differences that exist within the Latinopopulation. This population is significantly hetero-geneous and in our study was defined based oneither ageographicdefinition (ie, LatinAmerica) orself-identification18 and included individuals ofdifferent races (eg, Latinowhites, Latino AmericanIndians, and Latino blacks). In this study, we didnot collect information on race in the Latino pop-ulation, nor did we perform genetic ancestry an-alyses or germline ancestry informative markersthat could characterize the genetic origin withinadmixed populations.19 It is possible that we mayhave primarily sampled a subset of Latino patientswith NSCLC, such as the Latino white population.Because this population is similar, in terms ofgenetic ancestry, to the non-Latino white popula-tion in theUnitedStates, thismay have obscured apotential difference in EGFR mutation frequencybetween the two groups. However, Latinos withAmerican Indian ancestry are of special interest,given that they represent themajority population inMexico, Central America, and parts of SouthAmerica (eg, Peru), and Latinos from such geo-graphic areas comprise the largest subgroup ofLatinos in the United States. Thus, we may nothave captured a significant portion of the popu-lation of interest (ie, Latino American Indians),resulting in the negative results presented here.Conversely, the observation of a high EGFR mu-tation frequency in patients from Peru, which isknown to have a large percentage of Latinos withhigh American Indian ancestry, could be an in-dication that there may in fact be a higher EGFRmutation frequency in a subset of Latinos de-fined by high American Indian ancestry. Alterna-tively, this could be related to sampling of aPeruvian population of Chinese and Japanesedescent, which is also among the largest in LatinAmerica.20 This underscores theneed to account

Table 4 – EGFR Mutation Frequency Comparisons Among Subgroups

Subgroup

Wild Type MutatedTotal

PNo. % No. % No.

Latinos v non-Latino .78

Latino 69 76.67 21 23.33 90

Non-Latino 306 78.46 84 21.54 390

Total 375 105 480

Latinos v white v Asian v black , .001

Latino 69 76.67 21 23.33 90

Non-Latino

White (NCI) 135 79.41 35 20.59 170

White (OHSU) 124 83.78 24 16.22 148

Asian 15 42.86 20 57.14 35

Black 27 90 3 10 30

Total 370 103 473

Latinos v non-Latino whites .54

Latino

United States 23 76.67 7 23.33 30

Non–United States 46 76.67 14 23.33 60

White

NCI 135 79.41 35 20.59 170

OHSU 124 83.78 24 16.22 148

Total 328 80 408

United States v non–United States , .001


United States 329 78.33 91 21.67 60

Total 375 105 480

All patients by smoking history , .05

Never smoker 101 61.21 64 38.79 165

Smoker 265 86.6 41 13.4 306

Total 366 105 471

Never smokers 1.00

Latino 25 62.5 15 37.5 40

Non-Latino 76 60.8 49 39.2 125

Total 101 64 165

Never smokers .016

Latino 25 62.5 15 37.5 40

Non-Latino

White (NCI) 38 66.67 19 33.33 57

White (OHSU) 22 66.67 11 33.33 33

Asian 7 29.17 17 70.83 24

Black 7 77.78 2 22.22 9

Total 99 64 163

Never smokers .97

Latino

(Continued on following page)



for race and genetic ancestry analysis in futurestudies.

Finally, it is important to note that the EGFRmutation frequency in the non-Latino white pop-ulation of our cohort was considerably higher thanwhat has been described in historical controls.This is likely the result of an unintended patientselection bias related to the fact that the popula-tions of patients seeking care at large researchinstitutions in the United States (eg, NCI andOHSU) tend to have clinical characteristics thathave been associated with a higher frequency ofEGFR mutations (eg, never smokers and adeno-carcinoma histology).21 This is of utmost impor-tance, because our conclusions are based ona comparator arm for which EGFR mutation

frequency may have been overestimated as a re-sult of selection bias. Thus, this comparison maynot be appropriate, and our conclusions may notbe accurate. This may explain the contrastingconclusions reached by this and other studies,particularly that conducted by Arrieta et al,10 inwhich the EGFR mutation frequency of a large(N = 5,738) retrospective cohort of NSCLC sam-ples from different countries in Latin America wasanalyzed. In that study, a frequency of 26% wasfound, which was compared with a historical con-trol of 15% in non-Latino patients from the UnitedStates; thus, the authors concluded that the EGFRmutation frequency in Latinos was higher. In con-trast, Zhang et al11 evaluated EGFR mutations inlung adenocarcinomas from US Latino patientsand found an EGFR mutation frequency of 15%(six of 40), which was similar to that observedin non-Latino white patients (eight [18.6%]of 43).

Our study provides what is to our knowledge thefirst prospective assessment of EGFR mutationfrequency in CLIA-certified laboratories amongLatinos from the United States and Latin Americaand allows comparison with frequencies amongUS non-Latino patients of different races. Al-though we did not observe a difference in thefrequency of EGFR mutations between Latinosand non-Latinos, our results should be interpretedwith caution, given the significant limitations of thestudy. Furthermore, our observations are incon-clusive, and more research is needed to betterunderstand the molecular characteristics ofNSCLC in Latinos, with a particular focus on therole of race and genetic admixture and its molec-ular implications and the determination of its im-pact on the treatment and outcome of all patientswith NSCLC regardless of race or ethnicity.



Conceptionanddesign:Ariel Lopez-Chavez,SethM.Steinberg,Giuseppe Giaccone

Collection and assembly of data: Ariel Lopez-Chavez, AnishThomas, Moses O. Evbuomwan, Liqiang Xi, Guinevere Chun,Tatiana Vidaurre, Oscar Arrieta, George Oblitas III, Ana BelenOton, Arun Rajan, Mark Raffeld, Lorena Arze-Aimaretti,Giuseppe GiacconeData analysis and interpretation: Ariel Lopez-Chavez, AnishThomas, Liqiang Xi, Alejandro R. Calvo, Seth M. Steinberg,Lorena Arze-Aimaretti, Giuseppe Giaccone

Manuscript writing: All authors

Final approval of manuscript: All authors


The following represents disclosure information provided byauthors of this manuscript. All relationships are consideredcompensated. Relationships are self-held unless noted. I =Immediate Family Member, Inst = My Institution. Relation-ships may not relate to the subject matter of this manuscript.For more information about ASCO’s conflict of interest policy,please refer to www.asco.org/rwc or jco.ascopubs.org/site/ifc.

Ariel Lopez-ChavezEmployment: GenentechStock or Other Ownership: Genentech

Table 4 – EGFR Mutation Frequency Comparisons Among Subgroups (Continued)

Subgroup

Wild Type MutatedTotal

PNo. % No. % No.

United States 6 66.67 3 33.33 9


White

NCI 38 66.67 19 33.33 57

OHSU 22 66.67 11 33.33 33

Total 85 45 130

Latinos .06

Peru 17 62.96 10 37.04 27

Non-Peru 52 82.54 11 17.46 63

Total 69 21 90

Latino never smokers .53

Peru 12 57.14 9 42.86 21

Non-Peru 13 68.42 6 31.58 19

Total 25 15 40

Abbreviations: NCI, National Cancer Institute; OHSU, Oregon Health and Science University.







Anish ThomasNo relationship to disclose

Moses O. EvbuomwanNo relationship to disclose

Liqiang XiNo relationship to disclose

Guinevere ChunNo relationship to disclose

Tatiana VidaurreNo relationship to disclose

Oscar ArrietaNo relationship to disclose

George Oblitas IIINo relationship to disclose

Ana Belen OtonEmployment: Eli LillyHonoraria: Pfizer

Consulting or Advisory Role: PfizerTravel, Accommodations, Expenses: Pfizer, Eli Lilly

Alejandro R. CalvoSpeakers’ Bureau: Genentech, Boehringer Ingelheim

Arun RajanNo relationship to disclose

Mark RaffeldStock or Other Ownership: Pfizer, GlaxoSmithKline

Seth M. SteinbergNo relationship to disclose

Lorena Arze-AimarettiNo relationship to disclose

Giuseppe GiacconeConsulting or Advisory Role: Astex Pharmaceuticals, Boeh-ringer Ingelheim, Clovis Oncology, Celgene

Research Funding: Karyopharm Therapeutics, AstraZeneca

Affiliations

Ariel Lopez-Chavez, Knight Cancer Institute, Oregon Health and Science University, Portland, OR; Anish Thomas, Moses O.Evbuomwan, Liqiang Xi, Guinevere Chun, Arun Rajan, Mark Raffeld, Seth M. Steinberg, and Giuseppe Giaccone, Center for CancerResearch, National Cancer Institute, National Institutes of Health, Bethesda, MD; Tatiana Vidaurre, Instituto Nacional deEnfermedadesNeoplasicas, Lima,Peru;Oscar Arrieta, InstitutoNacional deCancerologia,MexicoCity,Mexico;GeorgeOblitas III,Hospital Oncologico Luis Razetti, Caracas, Venezuela; Ana Belen Oton, Denver Health Medical Center, University of Colorado,Denver, CO; Alejandro R. Calvo, Kettering Cancer and Blood Specialists, Kettering, OH; and Lorena Arze-Aimaretti, Facultad deMedicina de la Universidad del Valle, Cochabamba, Bolivia.

REFERENCES1. Clegg LX, Li FP, Hankey BF, et al: Cancer survival among US whites and minorities: A SEER (Surveillance, Epide-

miology, and End Results) program population-based study. Arch Intern Med 162:1985-1993, 2002

2. Colby SL, Ortman JM: Projections of the Size and Composition of the U.S. Population: 2014 to 2060—PopulationEstimates and Projections. https://www.census.gov/content/dam/Census/library/publications/2015/demo/p25-1143.pdf

3. Mok TS, Wu YL, Thongprasert S, et al: Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma. N Engl J Med361:947-957, 2009

4. Zhou C, Wu YL, Chen G, et al: Erlotinib versus chemotherapy as first-line treatment for patients with advanced EGFRmutation-positive non-small-cell lung cancer (OPTIMAL, CTONG-0802): A multicentre, open-label, randomised,phase 3 study. Lancet Oncol 12:735-742, 2011

5. Rosell R, Carcereny E, Gervais R, et al: Erlotinib versus standard chemotherapy as first-line treatment for Europeanpatients with advanced EGFR mutation-positive non-small-cell lung cancer (EURTAC): A multicentre, open-label,randomised phase 3 trial. Lancet Oncol 13:239-246, 2012

6. Wu YL, Zhou C, Hu CP, et al: Afatinib versus cisplatin plus gemcitabine for first-line treatment of Asian patients withadvanced non-small-cell lung cancer harbouring EGFRmutations (LUX-Lung 6): An open-label, randomised phase 3trial. Lancet Oncol 15:213-222, 2014

7. Janne PA, Engelman JA, Johnson BE: Epidermal growth factor receptor mutations in non–small-cell lung cancer:Implications for treatment and tumor biology. J Clin Oncol 23:3227-3234, 2005

8. Rosell R, Moran T, Queralt C, et al: Screening for epidermal growth factor receptor mutations in lung cancer. N Engl JMed 361:958-967, 2009

9. Shi Y, Au JS, Thongprasert S, et al: A prospective, molecular epidemiology study of EGFR mutations in Asian patientswith advanced non-small-cell lung cancer of adenocarcinoma histology (PIONEER). J Thorac Oncol 9:154-162, 2014

10. Arrieta O, Cardona AF, Martın C, et al: Updated frequency of EGFR and KRASmutations in nonsmall-cell lung cancerin Latin America: The Latin-American Consortium for the Investigation of Lung Cancer (CLICaP). J Thorac Oncol 10:838-843, 2015


https://www.census.gov/content/dam/Census/library/publications/2015/demo/p25-1143.pdf

https://www.census.gov/content/dam/Census/library/publications/2015/demo/p25-1143.pdf


11. Zhang W, McQuitty EB, Olsen R, et al: EGFR mutations in US Hispanic versus non-Hispanic white patients with lungadenocarcinoma. Arch Pathol Lab Med 138:543-545, 2014

12. American Cancer Society: Cancer Facts and Figures for Hispanics/Latinos 2012-2014. http://www.cancer.org/acs/groups/content/@epidemiologysurveilance/documents/document/acspc-034778.pdf

13. National Cancer Institute: SEER Stat Fact Sheets: Lung and Bronchus Cancer. http://seer.cancer.gov/statfacts/html/lungb.html

14. Saeed AM, Toonkel R, Glassberg MK, et al: The influence of Hispanic ethnicity on nonsmall cell lung cancer histologyand patient survival: An analysis of the Survival, Epidemiology, and End Results database. Cancer 118:4495-4501,2012

15. Eberhard DA, Johnson BE, Amler LC, et al: Mutations in the epidermal growth factor receptor and in KRAS arepredictive and prognostic indicators in patients with non–small-cell lung cancer treated with chemotherapy alone andin combination with erlotinib. J Clin Oncol 23:5900-5909, 2005

16. Chowdhuri SR, Xi L, Pham TH, et al: EGFR and KRASmutation analysis in cytologic samples of lung adenocarcinomaenabled by laser capture microdissection. Mod Pathol 25:548-555, 2012

17. Lynch TJ, Bell DW, Sordella R, et al: Activating mutations in the epidermal growth factor receptor underlying re-sponsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med 350:2129-2139, 2004

18. Yaeger R, Avila-Bront A, Abdul K, et al: Comparing genetic ancestry and self-described race in African Americans bornin the United States and in Africa. Cancer Epidemiol Biomarkers Prev 17:1329-1338, 2008

19. Araujo LH, Timmers C, Bell EH, et al: Genomic characterization of non–small-cell lung cancer in African Americans bytargeted massively parallel sequencing. J Clin Oncol 33:1966-1973, 2015

20. Central Intelligence Agency: The World Factbook Peru. https://www.cia.gov/library/publications/the-world-factbook/geos/pe.html

21. Brewin CR, Bradley C: Patient preferences and randomised clinical trials. BMJ 299:313-315, 1989


http://www.cancer.org/acs/groups/content/@epidemiologysurveilance/documents/document/acspc-034778.pdf

http://www.cancer.org/acs/groups/content/@epidemiologysurveilance/documents/document/acspc-034778.pdf

http://seer.cancer.gov/statfacts/html/lungb.html

http://seer.cancer.gov/statfacts/html/lungb.html

https://www.cia.gov/library/publications/the-world-factbook/geos/pe.html

https://www.cia.gov/library/publications/the-world-factbook/geos/pe.html


originalreport

Breaking Cancer Bad News to PatientsWith Cancer: A ComprehensivePerspective of Patients, Their Relatives,and the Public—Example From a MiddleEastern Country

abstract

Purpose Empowering patients with cancer requires that they be continuously informed about their con-dition. In some Eastern cultures, this concept is often opposed by caregivers.We aim to compare the extentof disclosure desired by actual and presumed patients with cancer and their relatives in our practice.

Methods Nine questions reflecting possible bad news communication to patients from diagnosis to the endof life were designed to investigate the extent of desired disclosure and were answered by 100 patients(cohort I) and 103 accompanying relatives (cohort II) in an outpatient setting. In addition, 103 publicparticipants attending a family medicine clinic (cohort III) each answered the questions from the per-spective of a presumed patient (cohort IIIA) and the perspective of a relative (cohort IIIB). The primary endpoint was affirmative response to six or more questions (AR ‡ 6), reflecting a preference to be informed ofthe majority (‡ 67%) of possible bad news.

ResultsAR‡6was reported in85%ofcohort I and52%ofcohort II (x2P< .001). Onmultivariable analysis,AR ‡ 6 showed significant association with being a patient (in cohorts I and II) and having nonmetastaticdisease (only in cohort I). In the public cohort, AR‡6was reported in 91% in cohort IIIA and 63% in cohortIIIB (x2 P < .001). On multivariable analysis, AR ‡ 6 in cohort III was significantly associated with being apresumed patient and having at least a college education.

Conclusion Patients desire disclosure of the majority of cancer-related bad news. This is in contrast to theviews and requests of relatives. The public participants would also desire similar disclosure if they were tobe diagnosed with cancer.


INTRODUCTION

Over the last few decades, traditional paternalisticmodels of patient care have given way to moreemphasis on patient empowerment, autonomy,and shared decision making. This shift is wellsupported by currently available recommenda-tions and guidelines.1-3

The United Kingdom General Medical Councilgood medical practice guidance clearly stresseseffective communication with patients. The guid-ance states that physiciansmust listen to patients,take account of their views, and respond honestlyto their questions. Theymust also give patients theinformation they want or need to know in a waythey can understand.4

Nevertheless, concealing diagnostic or prognosticinformation from patients with cancer is still com-mon in clinical practice. Studies show that a sig-nificant proportion of patients with cancer are notprovided with much of their health-related news.5

This is more prevalent in Eastern countries thanin Western countries.6 In Eastern culture, familymembers often extend their support to the patientby stepping in and taking over some or all of thepatient’s responsibilities and rights. This support-ive attitude may evolve into a dominating attitudethat hijacks the patient’s basic right to knowledgeand a share in decision making.7 Conversely,evidence shows that a majority of patients withcancer in Western societies need to know the

Jamal Zekri

Syed Mustafa Karim

All authors, AlFaisalUniversity, Riyadh; andKing Faisal SpecialistHospital and ResearchCenter, Jeddah, SaudiArabia.

Authors’ disclosures ofpotential conflicts ofinterest and contributionsare found at the end of thisarticle.Corresponding author:SyedMustafa Karim, MD, KingFaisal Specialist HospitalandResearchCenter,MBCJ64, PO Box 40047,Jeddah 21499, SaudiArabia; e-mail: [email protected].






nature of their diagnosis, and they want as muchinformation as possible about their condition.8

There is no evidence to suggest that patients fromEastern cultures want less information.

Physicians, patients with cancer, and their rela-tives in Eastern cultures often find themselves inthisdilemma,whichcontinues tobeanunresolvedmatter of public debate. Rational decisions cannever be reached in theabsenceof data specific tothis population. Motivated by these realities, weplanned this study to shed light on the perspec-tivesofpatientswithcancer, their relatives, and thegeneral public in a Middle Eastern country ondisclosing cancer bad news.

METHODS

After approval from our institutional review board,consecutive adult patients with cancer (cohort I)and their accompanying relatives (cohort II) wererecruited from medical oncology clinics. Patientswith cancer whowere diagnosed at least 3monthsbefore recruitment, were aware of their diagnosis,and had the ability to read and write or verballycommunicate were included in the study.

The public participants (cohort III) were recruitedfrom consecutive adult nonpatients with cancerpresenting to one family clinic in the family med-icine outpatient department. All individuals wererecruited within 3 weeks.

Nine closed-ended questions were designed totest the individuals’ (patients, their relatives, andthe general public) perspectives regarding disclo-sure of bad news. The questions reflect possiblebad news communication from diagnosis to theendof life (Tables1and2).Questionspresented topatients addressed their personal preferenceabout having cancer bad news delivered to them-selves. Questions presented to relatives of patientswith cancer addressed their preference regardingdelivering news to that patient (their relative).

Questions to members of the public were pre-sented in two formats. Format A addressed theirpersonal preference if they were diagnosed withcancer (self is a patient), and format B addressedtheir personal preference if a close relative werediagnosed with cancer (relative is a patient). Thequestions were designed so that an affirmative(yes) answer indicated willingness to be informedof the bad news. Patients who answered affirma-tively to six or more of the first nine questions wereidentified as indicating a preference to be in-formed of a significant majority (six of nine or67%) of the possible bad news. Another questionwas included to assess preference regarding who

should receive the bad news first (the patient, thefamily member, or to both at the same time).

Participants were approached by the investigator(treating oncologist, family physician, or the re-search nurse) so the study could be explainedverbally and to obtain verbal consent for partici-pation. They were then given the questionnaire intheir native Arabic language to self-completewithin the clinic setting. For those unable to read,the investigator read the questions, documentedthe patient’s exact response, and was available toexplain any ambiguities.

We estimated that 70% of the participants pre-ferred disclosure of bad news and thus wouldanswer each question affirmatively. The total sam-ple size required was calculated to be 267 if theestimatewaswithin6%of the true value (margin oferror). The total samplewasdividedbyusingequalallocation; to increase the precision of the esti-mate, the sample size was increased to 100 par-ticipants in each cohort.

Collectively, 309 individuals were approached. Sixindividuals (three patients with cancer and threemembers of the public) did not participate in thestudy because of lack of interest. In all, 303 in-dividuals (98%) completed the survey.

Results are presented as a percentage of affirma-tive responses (ARs) to each question. Responsesto eachquestion answeredby cohorts I and II werecompared by using the x2 test. Responses to eachquestion in the two formats (to self and to relative)answered by cohort III were compared by usingMcNemar’s test. Multivariable linear regressionanalysis was conducted to investigate the associ-ation between affirmative response to six or morequestions (AR>6)andpossible relevant factors incohorts I and II combined and in cohort III sepa-rately. SPSS version 12 (SPSS, Chicago, IL) wasused for statistical analysis.

RESULTS

Patients and Their Relatives

Of the patients, 100 completed the questionnaire.Median age was 48 years (range, 18 to 80 years),72% were females, and 28% were males. Eighty-seven percent of patients preferred to be informedof their cancer diagnosis, and 98% wanted toknow the serious news about their health. Re-sponses to other questions are presented inTable 1, and they show that themajority of patientspreferred tobe informedof other aspects of cancertreatment and possible unfavorable outcome.A relatively smaller majority (56%) preferred tobe involved in end-of-life discussions. Eighty-three



percent of patientswanted to be the first to knowofanybadnews (either alone [43%]oraccompaniedby a family member [40%]). Eighty-five percent ofpatients answered six or more questions affirma-tively.Nopatients (0%)answeredyes to fewer thanthree questions.

Of the relatives, 103 completed the questionnaire.Median age was 39 years (range, 18 to 77 years),44%were females, and56%weremales.Sixty-eightpercent of relatives preferred that patients be in-formedof theircancerdiagnosis incontrast to87%ofpatientswhopreferred tobe informedof their cancerdiagnosis (68% v 87%; P = .001). Responses ofrelatives to other questions are provided in Table 1;data show that more patients than relatives were infavor of disclosure to the patient. For example, 85%of patients want to be informed of treatment failurecomparedwithonly38%of relativeswhoagreed thatthepatientshouldbe informedof this fact (P, .001).Comparison between patients’ and relatives’ re-sponses is provided in Table 1.Only 39%of relativeswanted the patient to be the first to know of any badnews (either alone [9%] or accompanied by a familymember [30%]).

The disclosure of themajority of bad news (AR> 6)to thepatient wasdesired by 85%of patients but byonly 52% of relatives (x2 P, .001). On multivari-able linear regression analysis, desire for majordisclosure was significantly associated with beinga patient in cohorts I or II and having nonmetastaticdisease in cohort I (Table 3).

Public Participants

Of the public participants, 100 completed thequestionnaire. Median age was 32 years (range,18 to 75 years), 44%were females, and 56%weremales. Ninety-eight percent preferred to be in-formedof thediagnosis if theywerediagnosedwithcancer (format A), whereas only 75% agreed thatdiagnosis should be disclosed to a relative if thatrelative were diagnosed with cancer (format B;98% v 75%; P = .001). Twenty-four percentadvocated disclosure of diagnosis to self but notto a relative.

Responses of the public to other questions areprovided in Table 2, and all favored more disclo-sure to self than to a relative if both were patientswith cancer. For example, 79% wanted to beinformed of treatment failure (format A) comparedwith only 45%who agreed that the relative shouldbe informed (format B; P, .001). Comparison ofresponses to both formats of the questionnaire ispresented in Table 2.

Only 54% advocated that the patient (relative) tobe the first to know of any bad news (either alone[13%] or accompanied by a family member[41%]). In contrast, 90% advocated that the pa-tient (self) to be the first to know of any bad news(either alone [66%] or accompanied by a familymember [24%]).Ninety-onepercent answeredsixor more questions in format A affirmatively com-pared with 63% in format B (P , .001).

The disclosure of themajority of bad news (AR> 6)to the patient was desired by 91% of the publicparticipants if they were diagnosed with cancerbut by only 63% of the public participants iftheir close relatives were diagnosed with cancer(x2 P ,.001). On multivariable linear regressionanalysis, desire formajor disclosurewas significantlyassociated with being a presumed patient andhaving a college or higher education (Table 3).

DISCUSSION

It is clear from daily practice in Middle Easterncountries that serious health-related information(including cancer bad news) is not disclosed tomany patients. Aljubran7 described that in SaudiArabia, family members’ genuine cultural values

Table 1 – Questionnaire and Frequency of Affirmative Response of Patients and TheirRelatives

Question Patients (%) Relatives (%) P

Should the patient be informed of:

1. The diagnosis of cancer 87 68 .001

2. Any possible poor outcome 90 57 , .001

3. The term chemotherapy if it is part oftreatment

71 67 .548

4. Failure of treatment 85 38 , .001

5. Every significant change about thecondition and outcome

92 70 , .001

6. Serious health news 98 85 .002

7. The lack of specific anticancer treat-ment options

90 61 , .001

Do you:

8. Agree that a physician should notwithhold information about you at therequest of a family member

80 56 .001

9. Agree that the patient should be in-volved in end-of-life discussions

56 30 , .001

Affirmative (yes) answers to six or more ofnine questions

85 52 , .001

Should bad news be disclosed first to , .001

The patient 43 9

A family member 16 61

Both at the same time 40 30



oblige them to support thepatient. Theboundariesof this support are unclear and often lead to takingover some or all of the patient’s responsibilities.Eventually, that support may evolve into a domi-nating attitude that hijacks the patient’s basic rightto knowledge and a part in decision making. BouKhalil9 reports similar situations across theMiddle

East. This attitude is prevalent in many Easterncultures. In Turkey, 66% of relatives of patientswith cancer did not want the diagnosis to bedisclosed to the patient.10 Communities in coun-tries recognized as Western that have geographicproximity to Eastern countries show similar atti-tudes. In Italy, 84% of cancer physicians consider

Table 2 – Questionnaire and Frequency of Affirmative Response of the Public

QuestionTo Self

(format A)To Relative(format B)

Yes to Self ButNo to Relative

No to Self ButYes to Relative P

Should the patient be informed of:

1. The diagnosis of cancer 98 75 24 1 , .001

2. Any possible poor outcome 85 60 28 3 , .001

3. The term chemotherapy if it is part oftreatment

80 76 8 4 .388

4. Failure of treatment 79 45 36 2 , .001

5. Every significant change about thecondition and outcome

93 84 12 3 .035

6. Serious health news 99 90 9 0 .004

7. The lack of specific anticancer treat-ment options

79 56 24 1 , .001

Do you:

8. Agree that a physician should notwithhold information from the patient atthe request of family member

87 82 12 7 .359

9. Agree that the patient should be in-volved in end-of-life discussions

76 59 18 1 , .001

Affirmative (yes) answers to six or more ofnine questions

91 63 , .001

Should bad news be disclosed first to , .001

The patient 66 13

A family member 10 46

Both at the same time 24 41

Table 3 – Multivariable Linear Regression Analysis of Relation Between AR > 6 and Possible Confounding Factors

Confounding FactorHR

(cohorts I and II) 95% CI PHR

(cohort III) 95% CI P

Age, years 0.53 0.26 to 1.08 .08 1.34 0.55 to 3.24 .52

, 50 (cohorts I and II)

, 35 (cohort III)

Male sex 0.98 0.44 to 2.22 .97 0.89 0.39 to 2.01 .78

Less than college education 0.95 0.48 to 1.90 .87 0.24 0.11 to 0.56 .001

Employment status 1.40 0.62 to 3.16 .42 0.83 0.35 to 1.99 .68

Being a patient v being a relative 4.35 2.15 to 8.79 , .001 7.11 3.05 to 16.59 , .001

Metastatic v nonmetastatic cancer* 0.18 0.04 to 0.70 .014

Active v nonactive treatment* 2.66 0.69 to 10.25 .15

Abbreviation: AR > 6, affirmative answers to six or more of nine questions; HR, hazard ratio.*Relevant to cohort I only.



the family to be an obstacle to direct communi-cation with the elderly.11 There is increasing ev-idence in Western countries that most patientswant to know more about their health and theirdisease.8,12-14 Evidence, although limited, simi-larly supports that most patients from Easterncultures want to be informed of their health-related issues.15,16

Our results add to the literature from the MiddleEastern region and confirm that a majority ofpatients with cancer (71% to 98%) prefer to knowmost of the information about their diagnosis,outcome, treatment, treatment results, and prog-ress (Table 1). In addition, 90% prefer to be in-formed about themore serious information if thereare no specific anticancer treatment options avail-able for their condition. End of life is an extremelychallenging stage because it can be associatedwith uncomfortable physical symptoms, depres-sion, dignity issues, and loss of hope.17 Even so,more than half of patients (56%) feel that theyshould be involved in the delicate and challengingdiscussions regarding the end of life.

We also attempted to investigate the amount ofbad news patients want to know. We accom-plished this by calculating the percentage of pa-tients who answered different proportions of thefirst nine questions affirmatively. Answering yes tosix or more questions (AR > 6) indicated a pref-erence for wanting to know > 67% of the badnews. We found that 85% of patients prefer toknow the majority of possible bad news. To thebest of our knowledge, this is the only publishedreport that addresses the amount of possible badnews patients with cancer want to know.

The results also confirm the current perceptionthat a significant proportion of relatives in theMiddle Eastern region prefer not to keep patientsinformed (Table 1). Research from South Koreashowed a similar pattern in which 96% of patientswith cancer compared with 77% of family mem-bers believe that patients should be informed of aterminal illnessdiagnosis.18 InChina,patientswithcancer were more likely than families to believethat patients should be informed of the diagnosis(early stage, 90.8% v 69.9%; terminal stage,60.5% v 34.4%).19

Personal experience and the literature show thatrelatives in Eastern cultures act as bearers to badnews and, in many cases, strongly demand a “donot tell” approach. Physicians in these culturesmay be more likely to follow family wishes.6,7,9,10

To investigate this point, we asked patients andtheir relatives two direct questions: “Do you agree

that the doctor should not withhold informationabout you at the request of family member?” Amajority of patients (80%) agreed compared withonly 56% of the relatives. “To whom should badnews be disclosed first?” More patients (83%)than relatives (39%) wanted disclosure first topatients (either alone or with a family member).

Here we have somedegree of conflict between thepatient’s and the relative’s wishes. Physiciansfrequently face this dilemma during daily practicein the Middle East. From a practical point of view,direct communication is about more than provid-ing information. It establishes good therapeuticrelationships and acknowledges the humanity ofpatients. The information provided by the physi-cian will help the patient understand and consentto the management plan. Without adequateinformation, a patient cannot make autonomousdecisions. All these factors impose a duty on thephysician to provide information to patients in anappropriate way.

Worldwide, few research groups attempted tostudy the attitude and opinion of the public andhealthy individuals toward disclosing bad news.Horn and Waingrow20 reported the Americanpublic’s views and beliefs about many issues re-lated to cancer, including disclosure of diagnosisat three time points: in 1948, 1955, and 1962.Their results showed that at least 63% (1948),64% (1955), and 60% (1962) of participants in-dicate that the patient should be informed of thediagnosis. A survey of 200 individuals age65yearsor older from 31 senior citizen centers in LosAngeles, CA, found that ethnicity was the primaryfactor that influenced attitudes toward truth-tell-ing.21 In Japan, 85.4% of the responding publicwanted full disclosure, and 11.3% wanted partialdisclosure.22 A survey of the general population inNepal showed that more than 80% wanted to beinformed if they were diagnosed with cancer.23

Views in Eastern cultures about disclosing serioushealth-related information to patients are dividedand continue to be a matter of unresolved publicdebate. In addition, there is a scarcity of dataregarding the preferences and attitudes of thepublic in Middle Eastern countries. The currentperception is that the public’s attitude is conser-vative and favors limited or no disclosure topatients.

To better understand the public’s attitude, mem-bers of the public sample were presented with twoquestionnaire formats: format A addressed theirpersonal preference if they were diagnosed withcancer (self is a patient); format B addressed their



personal preference if their close relative werediagnosed with cancer (relative is a patient).

In the public cohort, 79% to 98% answered thefirst eight questions affirmatively, indicating thatthe majority would prefer to know most of theinformation about their diagnosis, outcome,treatment, treatment results, and progress ifthey were diagnosed with cancer (Table 1).For example, 98% preferred disclosure of acancer diagnosis. A smaller study from the Mid-dle Eastern region published in 1998 reported asimilar outcome (93%) in 40 members of thepublic.15 Among 2,422 individuals from thegeneral population in Japan, 73% answeredthat they would want to know about their diseaseand prognosis if an incurable disease werediagnosed; 90% desired direct disclosure tothemselves.24

Interestingly, there was a significant difference inthe public’s response to both formats of the ques-tionnaire, indicating preferential desire for moredisclosure to self (91%) if the responder werediagnosed with cancer and less desire for disclo-sure to the relative (63%) if the relative werediagnosed with cancer (Table 2). The responseto the question “To whom should bad news bedisclosed first” shows that amajority of individualswanted to retain their right to be informed about

their own condition. Conversely, many individualswould deny their relatives this right. One can onlyassume that applying different principles to similarsituations in this context arises mostly from asincere and genuine protective motive, althoughintermediate-term and long-term implicationswere not considered. An uninformed patient willnot be able tomake a rational decision and cannotbeexpected tomakea valid consent to treatments.Unaware of the prognosis, this patient may missthe opportunity to plan social, professional, andfinancial matters appropriately.

We recruited patients from one hospital, whichmay be considered a limitation. However, ourhospital is a tertiary referral center serving all re-gions of the country. Therefore, we believe ourresults can be generalized to the whole Saudipopulation.

Interventions and reforms are urgently needed toguarantee that appropriate attention is paid topatients’ wishes and rights to be kept informedof their health-related issues. Physicians have aduty to explain to relatives the importance andpositive consequences of informing the patient. Inaddition, significant action should involve educat-ing the public on a wider scale.

DOI: 10.1200/JGO.2015.001925Published online on jgo.ascopubs.org on April 20, 2016.


Conception and design: Jamal Zekri

Administrative support: Jamal ZekriProvision of study materials or patients: Jamal ZekriCollection and assembly of data: Jamal Zekri

Data analysis and interpretation: All authorsManuscript writing: All authorsFinal approval of manuscript: All authors


The following represents disclosure information provided byauthors of this manuscript. All relationships are considered

compensated. Relationships are self-held unless noted. I =Immediate Family Member, Inst = My Institution. Relation-ships may not relate to the subject matter of this manuscript.For more information about ASCO’s conflict of interest policy,please refer to www.asco.org/rwc or jco.ascopubs.org/site/ifc.

Jamal ZekriNo relationship to disclose

Syed Mustafa KarimNo relationship to disclose

REFERENCES1. Okamura H, Uchitomi Y, Sasako M, et al: Guidelines for telling the truth to cancer patients. Jpn J Clin Oncol 28:1-4,

1998

2. Girgis A, Sanson-Fisher RW: Breaking bad news. 1: Current best advice for clinicians. Behav Med 24:53-59, 1998

3. Scope: Right from the start template: Good practice in sharing the news. http://www.cen.scot.nhs.uk/files/4z-scope_right_from_the_start_template.pdf

4. General Medical Council: Good medical practice. http://www.gmc-uk.org/guidance/good_medical_practice.asp

5. Costantini M, Morasso G, Montella M, et al: Diagnosis and prognosis disclosure among cancer patients: Results froman Italian mortality follow-back survey. Ann Oncol 17:853-859, 2006

6. Baile WF, Lenzi R, Parker PA, et al: Oncologists’ attitudes toward and practices in giving bad news: An exploratorystudy. J Clin Oncol 20:2189-2196, 2002






http://www.cen.scot.nhs.uk/files/4z-scope_right_from_the_start_template.pdf

http://www.cen.scot.nhs.uk/files/4z-scope_right_from_the_start_template.pdf

http://www.gmc-uk.org/guidance/good_medical_practice.asp


7. Aljubran AH: The attitude towards disclosure of bad news to cancer patients in Saudi Arabia. Ann Saudi Med 30:141-144, 2010

8. Meredith C, Symonds P, Webster L, et al: Information needs of cancer patients in west Scotland: Cross sectional surveyof patients’ views. BMJ 313:724-726, 1996

9. Bou Khalil R: Attitudes, beliefs and perceptions regarding truth disclosure of cancer-related information in the MiddleEast: A review. Palliat Support Care 11:69-78, 2013

10. OzdoganM, SamurM, ArtacM, et al: Factors related to truth-telling practice of physicians treating patients with cancerin Turkey. J Palliat Med 9:1114-1119, 2006

11. Locatelli C, Piselli P, CicerchiaM, et al: Physicians’ age and sex influence breaking bad news to elderly cancer patients:Beliefs and practices of 50 Italian oncologists—The G.I.O.Ger study. Psychooncology 22:1112-1119, 2013

12. Cassileth BR, Zupkis RV, Sutton-Smith K, et al: Information and participation preferences among cancer patients. AnnIntern Med 92:832-836, 1980

13. Sutherland HJ, Llewellyn-Thomas HA, Lockwood GA, et al: Cancer patients: Their desire for information and par-ticipation in treatment decisions. J R Soc Med 82:260-263, 1989

14. Jenkins V, Fallowfield L, Saul J: Information needs of patients with cancer: Results from a large study in UK cancercentres. Br J Cancer 84:48-51, 2001

15. Al-Ahwal M: Cancer patients’ awareness of their disease and prognosis. Ann Saudi Med 18:187-189, 1998

16. Al-Amri AM: Cancer patients’ desire for information: A study in a teaching hospital in Saudi Arabia. East MediterrHealth J 15:19-24, 2009

17. Woo JA, Maytal G, Stern TA: Clinical challenges to the delivery of end-of-life care. Prim Care Companion J ClinPsychiatry 8:367-372, 2006

18. Yun YH, Lee CG, Kim SY, et al: The attitudes of cancer patients and their families toward the disclosure of terminalillness. J Clin Oncol 22:307-314, 2004

19. Jiang Y, Liu C, Li JY, et al: Different attitudes of Chinese patients and their families toward truth telling of different stagesof cancer. Psychooncology 16:928-936, 2007

20. Horn D, Waingrow S: What changes are occurring in public opinion toward cancer: National public opinion survey. AmJ Public Health Nations Health 54:431-440, 1964

21. Blackhall LJ, Murphy ST, Frank G, et al: Ethnicity and attitudes toward patient autonomy. JAMA 274:820-825, 1995

22. Miyata H, Tachimori H, Takahashi M, et al: Disclosure of cancer diagnosis and prognosis: A survey of the generalpublic’s attitudes toward doctors and family holding discretionary powers. BMC Med Ethics 5:E7, 2004

23. Gongal R, Vaidya P, Jha R, et al: Informing patients about cancer in Nepal: What do people prefer? Palliat Med 20:471-476, 2006

24. Miyashita M, Hashimoto S, Kawa M, et al: Attitudes toward disease and prognosis disclosure and decision making forterminally ill patients in Japan, based on a nationwide random sampling survey of the general population and medicalpractitioners. Palliat Support Care 4:389-398, 2006



originalreport

Prioritizing US Cervical CancerPrevention With Results From aGeospatial Model

abstract

Purpose To determine if differences in screening and vaccination patterns across the population mayaccentuate ethnic and geographic variation in future burden of disease.

MethodsUsingCancer inNorth America data provided by theNorth AmericanAssociation of Central CancerRegistries, county cervical cancer incidence trends from1995 to 2009weremodeled for the entire UnitedStatesusingecologiccovariates. Rates for health serviceareaswerealsomodeledbyethnicity. State-levelincidence wasmapped together with Papanicolaou (Pap) screening, past 3 years (women‡ 18 years old),and three-dosehumanpapillomavirus (HPV)vaccinecoverage (girls13 to17yearsold) to identify potentialpriority areas for preventive services.

Results US cervical cancer incidence decreasedmore during the periods 1995 to 1999 and 2000 to 2004than during the period 2005 to 2009. During these 15 years, the most affected areas became increasinglyconfined to Appalachia, the lower Mississippi Valley, the Deep South, Texas, and Florida. Hispanic andblack women experienced a higher incidence of cervical cancer than both white and Asian and PacificIslander women during each period. Women in 10 of 17 states/districts with a high incidence (‡ 8.14/100,000) reported low Pap testing (< 78.5%), HPV vaccine coverage (< 33.9%), or both preventiontechnologies.

Conclusion The decline in cervical cancer incidence has slowed in recent years. Access to HPV vac-cination, targeted screening, and treatment in affected populations is needed to reduce cervical cancerdisparities in the future.


INTRODUCTION

Cervical cancer incidence and mortality in theUnited States have declined steadily in the de-cades since the adoption of cytology-basedPapanicolaou (Pap) testing.1 Recently, cervicalcancer death rates seem to have plateaued,and a slight decrease has been reported inthe use of Pap testing.2 Cervical cancer dispar-ities (the uneven distribution of burden to areasof lower socioeconomic status and minority eth-nic groups) are increasing.3-6 Cervical cancerincidence is elevated among older and blackwomen7,8 and in impoverished areas such asAppalachia, the lower Mississippi Valley, andthe United States–Mexico border area.9 Re-duced access to screening services is a drivingfactor in these areas. Nationally, 3-year Papscreening completion is . 80% for women 21to 65 years of age, with lower levels of screeningamong immigrants, minorities, the poor, and

older women.10 Approximately 47% of girls bornin theUnitedStates in 2000had received at leastone dose of human papillomavirus (HPV) vac-cine by age 13 years.11 Safety concerns andviews on sexual activity influence parental de-cisions regarding the vaccination of adolescentgirls.11 Although the safety and efficacy of theHPV vaccination is well established, in 2013,38% of girls 13 to 17 years of age in the UnitedStates had completed the three-dose HPV vac-cine series,11 far less than optimal.12 Becausethe reduction in cervical cancer incidence as aresult of HPV vaccination will take decades to berealized,12 surveillance is needed to identifycommunities with a higher burden of cervicalcancer and to direct cervical cancer screeningand vaccination services where they are neededmost.1

To define high-priority areas for preventive ser-vices, we estimated small-area cervical cancer

Jonathan K. Kish

Alicia I. Rolin

Zhaohui Zou

James E. Cucinelli

Zaria Tatalovich

Mona Saraiya

Sean F. Altekruse

Jonathan K. Kish, Alicia I.Rolin, Zaria Tatalovich, andSean F. Altekruse, NationalCancer Institute,Rockville; Zhaohui Zou andJames E. Cucinelli,Information ManagementServices, Silver Spring,MD; and Mona Saraiya,Centers forDiseaseControland Prevention, Atlanta,GA.The findings andconclusions in this reportare those of the authorsand do not necessarilyrepresent the officialposition of the Centers forDisease Control andPrevention.

Authors’ disclosures ofpotential conflicts ofinterest and contributionsare found at the end of thisarticle.Correspondingauthor:SeanF. Altekruse, DVM, MPH,PhD,9609Medical CenterDr, Room 4E536,Rockville, MD 20850;e-mail: [email protected].






incidence, including by ethnicity, and mappedstate-level incidence with percentage screeningand HPV vaccination coverage.

METHODS

Incidence Data

TheNorthAmericanAssociation ofCentral CancerRegistries (NAACCR) provided a Cancer in NorthAmerica data set with county-level incidence andpopulation denominators for this analysis. A totalof 36 of 50 state registries, all of which metNAACCR silver or gold standards for tumor regis-tration data quality, gave active consent to includetheir incident cases in the analytic data set. For theyears 1995 to 2009, complete or partial county-level cervical cancer incidence datawere reportedbyNAACCR registries in areascovering74%of theUnited States population. This included completedata from 1995 to 2009 for 22 states encompass-ing 56% of the population (Arizona, California,Colorado, Connecticut, Hawaii, Idaho, Illinois,Iowa, Kentucky, Louisiana, Maine, Michigan,Nebraska, New Jersey, New Mexico, New York,Pennsylvania, Rhode Island, Texas, Utah, Wash-ington, and Wyoming) and multiple years of datafor 14 states encompassing 18%of the population(Alaska, Arkansas, Georgia, Massachusetts, Mis-sissippi, Montana, Nevada, North Carolina, NorthDakota, South Carolina, South Dakota, Tennes-see,Virginia, andWestVirginia). In these14states,data were missing for a median of 2 reportingyears, generally the earliest reporting years. Twolower Mississippi Valley states were missing datafor 1995 to 2003. The remaining 14 states and theDistrict of Columbia, which did not provide data,accounted for 26% of the US population.

Spatiotemporal Model

We developed a model similar in composition toone used commonly to predict current-year can-cer incidence in the United States13 to estimatecounty-level cervical cancer incidence in 5-yearintervals from 1995 to 2009.14 The multivariablelogistic regression model included a set of cova-riates selected through a forward and backwardprocess. A generalized linear model was used(PROC GLIMMIX, SAS 9.3, SAS Institute, Cary,NC)with three random terms to account for spatialautocorrelation (longitude and latitude of thecounty), temporal autocorrelation (year of diagno-sis), and residual autocorrelation of covariates.Missing cervical cancer incidence at the countylevel was modeled on the basis of the reportedincidence in counties with comparable attributes.The incidence for countieswith observeddatawas

also updated according to model predictions toslightly adjust reported rates. The time periods ofinterest for modeled all-ethnicity county-level in-cidence were 1995 to 1999, 2000 to 2004, and2005 to 2009. To present incidence among ethnicgroups, datawere aggregated at the health servicearea (HSA) level,15 reducing instability from smallcounts at the county level. HSAs are either a singlecounty or a cluster of contiguous counties that arerelatively self-contained with respect to hospitalcare.Datawere suppressedwhen therewere16orfewer modeled incident cases. The accuracy ofpredicted rates for areas with missing data de-pends on how well covariates in themodel predictthe actual incidence.

Model Inputs

Demographic inputs of cervical incident caseswere non-Hispanic ethnicity (hereafter referredto as white, black, American Indian and AlaskaNative, and Asian and Pacific Islander), Hispanicethnicity (all ethnicities), and age. County of res-idence was geocoded on the basis of latitude andlongitude. County-level population estimates wereobtained from the US Census Bureau SummaryFile for each year from 1995 through 2009. In-cidence data, stratified by age (, 44, 45 to 64,and > 65 years) and year of diagnosis were re-trieved using SEER*Stat 8.1.2 (Information Man-agement Services, Calverton, MD).

County-level covariates included in the modelwere county-level rural-urban density data,16 anAreaHealth Resources covariate enumerating thenumberof hospital-basedphysiciansat thecountylevel,15 and data on the percentages of the countypopulation who were black, Asian and PacificIslander, and American Indian and Alaskan Na-tive. Cervical cancermortality data reported by theNational Center for Health Statistics were alsoincluded for each county in the United States.Socioeconomic covariates incorporated into themodel were the percentage of the county popu-lation with income below the poverty level and thepercentage of the population > 25 years of agewith > 4 years of college education. The modelalso contained a variable indicating whether thecounty was in a National Program of CancerRegistries–funded area. The model provided es-timates of county-level cervical cancer incidencefor the entire United States, including areas andyears with missing data as well as those withreported data. To illustrate geographic distribu-tions in rates, modeled county estimates weremapped using 2000 Census county designations(ArcGIS 10.1; ESRI, Redlands, CA). In the all



ethnicities combined model uncertainty relatedto small numbers was addressed with spatialsmoothing. The population-weighted, nonpara-metric algorithm used universal Kriging afterdetrending.17 Data were approximately normallydistributed and no transformations were applied,although first-order surface trend was removed.Smoothing was not applied to incidence maps ofspecific ethnic groups because of the potential forinstability.

State-Level Incidence Trends

State-level modeled incidence trends on the basisof estimated age-standardized incidence wereanalyzed using joinpoint regression analysis (Join-point 3.5.0; Information Management Services).The technique fit a series of joined straight linesona logarithmic scale for annual age-standardizedrates.18 Using data from the model for each year,state trends were estimated for 5-year fixedintervals, or annual average percent change(AAPC), using weighted annual percent changesfrom joinpoint models.

Identification of Priority Areas

Three state-level variables were visualized: mod-eled cervical cancer incidence during the period2005 to 2009, state-level proportions ofwomen21to 65 years of age who had had a Pap test duringthe past 3 years19 estimated from pooled 2008 to2010 responses to the National Health Interview

Survey and Behavioral Risk Factor SurveillanceSystem, and HPV vaccination coverage by state(obtained from the National Immunization Surveyof Teens performed during 2013)20 defined as thepercentage of girls 13 to 17 years old who hadreceived three doses or more of either the bivalentor the quadrivalent vaccine.

State-level cervical cancer incidencewasmappedin tertiles (low, midlevel, and high incidence). Lowincidence was 5.17 to 6.75, midlevel incidenceranged from 6.76 to 8.13, and high incidence was8.14 to9.76casesper100,000women.Lowstate-level Pap screening coverage was defined as ,78.5%, midlevel as 78.5% to 81.4%, and high as81.5% to 88.5%. Low state-level HPV vaccinationuptake was defined as , 33.9%, midlevel as33.9% to 40.1%, and high as 40.2% to 56.5%.States in the bottom third distribution for three-doseHPVvaccine receipt (,33.9%), Pap screen-ing (,78.5%), or both vaccine andPap screeningwere depicted with horizontal, vertical, and cross-hatched lines, respectively. The status of stateMedicaid expansion as of January 2016 was alsoassessed.21

RESULTS

Modeled cervical cancer incidence and observeddeath rates per 100,000 women from 1995 to2009 are presented in Table 1. Incidence anddeath rates were highest among non-Hispanicblack followed by Hispanic women in each time

Table 1 – Age-Adjusted Cervical Cancer Incidence and Mortality by Selected Characteristics, 1995 to 2009

Characteristic

1995-1999 2000-2004 2005-2009 Overall

ModeledIncidence

ObservedDeath

ModeledIncidence

ObservedDeath

ModeledIncidence

ObservedDeath

ModeledIncidence

ObservedDeath

N Rate N Rate N Rate N Rate N Rate N Rate N Rate N Rate

Overall 75,085 10.8 22,086 3.1 66,406 8.9 20,013 2.6 63,174 7.9 19,838 2.4 204,665 9.1 61,937 2.7

Ethnicity

White 51,100 9.7 13,951 2.6 42,942 7.9 12,323 2.2 38,879 6.9 11,897 2.1 132,922 8.1 38,171 2.3

Black 11,891 15.7 4,272 6.3 10,629 12.6 3,796 5.0 10,243 11.0 3,723 4.4 32,763 12.9 11,791 5.1

Hispanic 9,280 17.1 1,817 3.8 9,872 14.3 2,047 3.4 10,891 12.8 2,325 3.0 30,042 14.4 6,189 3.3

API 2,369 10.5 597 3.0 2,532 8.7 647 2.4 2,745 7.6 648 1.9 7,646 8.7 1,892 2.3

AI/AN 444 9.2 99 3.2 430 7.8 107 2.7 416 6.9 134 2.8 1,291 7.9 340 2.8

Age, y

< 44 30,434 6.7 5,294 1.2 25,104 5.5 4,470 1.0 21,875 4.9 3,977 0.9 77,413 5.7 13,741 1.0

45-64 29,757 20.5 8,357 5.7 28,726 16.9 8,306 4.9 29,363 15.2 9,079 4.6 87,845 17.3 25,742 5.0

> 65 14,895 14.6 8,435 8.2 12,576 11.9 7,237 6.8 11,936 10.7 6,782 6.1 39,407 12.3 22,454 7.0

NOTE. Incidence anddeath rates are per 100,000women. Ethnicmortality statistics exclude deaths in theDistrict of Columbia,Maine,Minnesota,NewHampshire, NorthDakota,Oklahoma, andSouthCarolinabecause of instability inHispanic populationestimates.22Data sources:NorthAmericanAssociation of Central CancerRegistries for cervical cancerincidence, National Center for Health Statistics for deaths, 1995 to 2009. Incident case counts estimated using spatiotemporal model; observed death counts, cervical cancer.Abbreviations: API, non-Hispanic Asian/Pacific Islander; AI/AN, non-Hispanic American Indian/Alaska native; y, years.



period. Incidence and death rates decreased fromthe period 1995 to 1999 to the period 2005 to 2009among all ethnic groups. Age-specific incidence per100,000womenwas highest among 45- to 64-year-old women, averaging 17.3 and decreasing from20.5 during the period 1995 to 1999 to 15.2 duringtheperiod2005 to 2009. Thehighest death ratewasalso seen among women > 65 years of age, aver-aging 7.0 anddecreasing from8.2during the period1995 to1999 to6.1during theperiod2005 to 2009.

Figure1displays the smoothed, county-levelmod-eled estimates of cervical cancer incidence bytime period. From 1995 to 1999, the highest in-cidence was found in counties extending southand west from Appalachia to southeastern Colo-rado, northeastern NewMexico, and Texas, with alow incidence found in southwestern New En-gland, the Northern Plains, and Mountain West.During the period 2000 to 2004, elevated ratesbecamemore localized to central Appalachia, theOhio River Valley, the Lower Mississippi RiverValley, rural Texas including the Mexican borderarea, and the rural Southeast. During the period2005 to 2009, areas with the highest incidencewere even more contained within Appalachia, thelower Mississippi Valley, the Deep South, Texas,and Florida.

Overall, in the United States during 1995 to 1999,2000 to 2004, and 2005 to 2009, the AAPC inmodeled cervical cancer incidence declined sig-nificantly by 23.5, 23.1, and 21.1 percent peryear over successive periods (Data Supplement).This pattern of slowing but still statistically signif-icant decreasing incidence trends in recent yearswas seen in 27 states. In nine states and in theDistrict of Columbia, trends did not change signif-icantly over the three time periods. In three states(Georgia, Virginia, and Louisiana) therewas a non-significant increase in cervical cancer incidencebetween 2005 and 2009. In nine other states(Alaska, Tennessee,Maryland,NewJersey, SouthCarolina, South Dakota, Illinois, Iowa, and Ore-gon), the decrease in cervical cancer incidenceduring theperiod2005 to2009wasnot statisticallysignificant. In Utah and Oklahoma the AAPC dur-ing the period 1995 to 1999 was not statisticallysignificant; however, incidence decreased signif-icantly during more recent time periods.

Figure 2 presents modeled HSA-level incidenceby ethnicity during the period 2005 to 2009. Colorranges vary by map. Higher cervical cancer in-cidencewas seen amongHispanic andblack thanamong white and Asian and Pacific Islanderwomen. Among whites, incidence was highest

1995–1999 2000–2004

2005–2009

5.2–7.6

7.7–8.6

8.7–9.6

9.7–11.0

11.1 +

Fig 1 –

Smoothed, county-levelmodeled US cervicalcancer incidence per100,000 women, 5-yearintervals, 1995 to 2009.



in Appalachia, the Mississippi and Ohio Rivervalleys, Indiana, and rural parts of Illinois, Texas,Louisiana, Arkansas, Oklahoma, Tennessee, Ala-bama, and Georgia. Isolated areas with highincidence were found in central California,northwestern Arizona, central Pennsylvania,southern New Jersey, and northern and centralFlorida. High modeled incidence areas for blackswere found in eastern Texas and adjacent areas ofOklahoma, Louisiana, and Arkansas. The inci-dence was also elevated along the MississippiRiver Valley from Illinois to Tennessee, Arkansas,and Mississippi. Affected rural areas includedadjacent areas of Mississippi and Alabama, andGeorgia and north central Florida. A high inci-dence was also seen in northern and centralIndiana, rural South Carolina, southern New Jer-sey, New York City, and southeastern Florida.

Of areas with large Hispanic populations, the re-gions with the highestmodeled incidence includedwest Texas, central and southern Florida, andmetropolitan Houston, Chicago, and New York.

AmongAsianandPacific Islanders, ahighmodeledincidence was seen in central California and Hous-ton, Texas; central and south Florida; and metro-politan New York. The incidence for AmericanIndians/AlaskaNativeswas not presented becauseof low counts.

Figure 3 presents state-level modeled cervicalcancer incidence during the period 2005 to2009. States are classified into three categories:low, midlevel, and high cervical cancer incidencestates. States in the bottom tertile of Pap screeninguse and HPV vaccine uptake were classified ashaving low adoption of these preventive services.Sixteen states plus theDistrict of Columbia, shownin red, had a high cervical cancer incidence. Intwo high-incidence states, both Pap screeningand HPV vaccine uptake were low (Arkansasand Nevada). There was a low proportion of Papscreening in three high-incidence states (Tennes-see, West Virginia, and Indiana) and of HPVvaccine uptake in five high-incidence juris-dictions (District of Columbia, Mississippi, Illinois,

Hispanic

9.4–10.7

10.8–11.611.7–12.6

12.7–13.5

13.6–15.1

Suppressed

6.5–9.6

9.7–10.710.8–11.2

11.3–12.3

12.4–15.4

Suppressed

Black

White Asian/Pacific Islander

4.5–6.4

6.5–7.17.2–7.6

7.7–8.4

8.5–10.6

Suppressed

5.5–6.3

6.4–7.37.4–8.0

8.1–8.7

8.8–9.4

Suppressed

Ethnic/racial groups with highest cervical cancer incidence rates

Racial groups with lowest cervical cancer incidence rates

Fig 2 –

Modeled health servicesarea–level cervical cancerincidence by ethnicity,2005 to 2009.



Kentucky, and New Jersey). Many of the 16 stateswith a midlevel cervical cancer incidence (shown inorange)were adjacent to high-incidence states. Onemidlevel-incidence state (Missouri) had low uptakeof both Pap screening andHPV vaccination. Amongthe midlevel-incidence states, five (New Mexico,Wyoming, Nebraska, Oklahoma, and Ohio) had alow percentage of Pap screening, and five (Alaska,Kansas,Maryland,NorthCarolina, andGeorgia) hadlow vaccine uptake. States with the lowest cervicalcancer incidence (shown in gold) were located inNew England, the northern tier of states, the fourcorners region, and Virginia. In three low-incidencestates (Montana, Idaho,andUtah),a lowpercentageof women received Pap screening, and a low per-centageof girls received threedosesofHPVvaccine.Therewas also a lowpercentage of Pap screening inArizona, North Dakota, and South Dakota, and thepercentage of girls in Virginia who received HPVvaccination was less than 33.9%.

As of January 2016, six of 17 jurisdictions withthe highest cervical cancer incidence had not

expanded Medicaid coverage (Alabama, Florida,Mississippi, South Carolina, Tennessee, andTexas). Seven of 16 stateswith amidlevel cervicalcancer incidence had not expanded Medicaid(Georgia, Kansas, Missouri, North Carolina,Nebraska, Oklahoma, and Wyoming). Amongthe 18 states with the lowest cervical cancerincidence, three had not expanded Medicaid(Idaho, Montana, and Utah).

DISCUSSION

This study suggests that the long-termdecrease inUS cervical cancer incidence is slowing down.This finding is consistent with the nonsignificantdecreasing trend reported during the period 2005to 2009 in areas in the United States with high-quality incidence data.12 High-incidence geo-graphic areas were confined increasingly to ruralareas within Appalachia, Texas, the lower Missis-sippi Valley, and the southeastern United States.Hispanic and black women had a higher cervicalcancer incidence than did white and Asian and

WA

MT

ID

OR

WY

ND

SD

NE

CO

UTNV

CA

KS

IA

MN

WI

IL

MO

OKNM

AZ

AK

HI

Age-adjusted incidence(per 100,000)

AR

MS AL

TN

KY

IN

MI

OH

PA

NY

VT

NH

ME

MACT

NJ

DE

MDDCVA

WV

SC

NC

GA

FL

LATX

5.17–6.75 (low)

6.76–8.13 (moderate)

8.14–9.75 (high)

Low HPV vaccination and Pap screeningLow HPV vaccination Low Pap screening

Fig 3 –

States with high, midlevel,and low cervical cancerincidence during the period2005 to 2009, furtherdepicting states withlow percentages ofPapanicolaou (PAP)screening or humanpapillomavirus (HPV)vaccination. Less than78.5% of women . 18years old reported PAPscreening in the past 3years during the period2008 to 2010, and ,33.9%ofgirls13 to17yearsof age had received threedoses of HPV vaccine as of2013.



Pacific Islander women. In 10 of 17 states/districtswith a high cervical cancer incidence, there waslow use of Pap screening or HPV vaccination.Locally tailored cervical cancer vaccination,screening, and treatment efforts that target poorwomen living in medically underserved geo-graphic areas are needed to maintain progressin reducing cervical cancer disparities.

As recently as the 1970s, cervical cancer was aleading cause of cancer among US women; how-ever, the incidence has decreased in subsequentdecades.8 Trends in this study differ from sus-tained decreasing cervical cancer incidencetrends over the past half-century, but are consis-tent with recent findings describing a leveling off ofcervical cancer mortality2 and incidence12 in theUnited States. A possible explanation for the slow-ing of the decreasing trend is that women withaccess to health care are benefiting from preven-tive services such as Pap screening and HPVtesting,2 to a greater extent than are women inmedically underserved groups. These under-served women, who experience a higher burdenof cervical cancer,23 include ethnic minorities,women from low socioeconomic backgrounds,5

and women living in impoverished geographicareas.10 Future progress to reduce the burdenof cervical cancer depends on access to vaccina-tion, screening, and treatment of these hard-to-reach groups.24 Of note, some states with thehighest incidence of cervical cancer have lowpercentages of Pap screening and HPV vaccineuptake. Provisions of the Affordable Care Act,which require most health insurance plans tocover cervical cancer screening and HPV vacci-nationwith nocost sharing, could improve cervicalcancer prevention among low-income women.25

In this study spanning the years1995 through2009,there were progressively smaller areas with an ele-vated cervical cancer incidence over time. Regionswith the highest burden of disease during the period2005to2009were largelycontained toeconomicallydeprived counties within Appalachia, Texas, thelower Mississippi Valley, and the southeasternUnited States. The limited progress in reducingthe incidence of this cancer in areas with sloweconomic development or an influx of immigrantpopulations is consistent with findings reported inMexico,26 Brazil,27 and England.28

Lack of awareness, lack of access to health care,and cultural beliefs are barriers to cervical cancerprevention within population subgroups.29-31 Forinstance, in Connecticut,32 heterogeneity in theoccurrenceofcervicalcancerprecursors is reported.

Culturallycompetent locally targetedoutreachneedsto be part of cervical cancer control programs. Onestudy of African-American women living in the highcervical cancer mortality area of Sunflower County,Mississippi, indicated that door-to-door visits to offerhome self-collection HPV test kits increased partic-ipation in cervical cancer screening almost four-foldcomparedwithclinic-basedPap testingalone.33 In anational study, predictors of not being screened forcervical cancer included not having made a physi-cian office visit within thepast 12monthsbecauseofcost,minority ethnicity, lackof ahigh school diplomaamong residents of metropolitan areas, and self-reported fair or poor general health among non-metropolitan area residents.34 The heterogeneity ofunderserved women suggests a need for screeningandHPV vaccination outreach across broad areas.35

Other researchers have reported ethnic and geo-graphic disparities in cervical cancer screening34

and incidence.36 In the United States–Mexico bor-der area,Hispanicwomenwere less likely thanotherwomen to have had a recent Pap test,37 and whitewomen in Appalachia had higher rates of HPVinfection compared with the US population.38

Although provider recommendation improvesacceptance of HPV vaccination, minority andlow-income women are least likely to receivesuch recommendations.39 Cervical cancer preven-tion can be advanced through community-basedinterventions,40 particularly in communities withlimitedaccess toa formalhealthcaresystem.Thesecommunity-based interventions may be more ef-fective than one-size-fits-all approaches.41-43

This study identified populations that would benefitfrom cervical cancer outreach by ethnicity, geogra-phy, and access to screening and HPV vaccination.Ongoing spatial analysis is recommended tomonitorcervical cancer trends in the HPV vaccine era.44

Study limitations include missing data for somestates,whichwaspartially addressedwith geospatialmodeling. Furthermore, HPV vaccination data wereavailable only at the state level. County-level datawould improve prioritization of outreach to areaswith a high cervical cancer burden. Despite limita-tions, the analysis identifies priority areas for inter-ventions to improvescreeningandvaccination rates.Although progress has been made in reducing theincidence of cervical cancer, outreach is needed inlow-socioeconomic areas of the United States. Pro-visions of the Affordable Care Act that eliminate costsharing for cervical cancer screening and HPV vac-cination inmosthealthplansshould reducecost asabarrier to receiving these prevention services.

DOI: 10.1200/JGO.2015.001677Published online on jgo.ascopubs.org on April 11, 2016.






Conception and design: Jonathan K. Kish, Mona Saraiya, SeanF. AltekruseAdministrative support: Mona Saraiya, Sean F. AltekruseCollection and assembly of data: Jonathan K. Kish, Alicia I.Rolin, James E. Cucinelli, Zaria TatalovichData analysis and interpretation: Jonathan K. Kish, Alicia I.Rolin, Zhaohui Zou, Mona SaraiyaManuscript writing: All authorsFinal approval of manuscript: All authors


The following represents disclosure information provided byauthors of this manuscript. All relationships are consideredcompensated. Relationships are self-held unless noted. I =Immediate Family Member, Inst = My Institution. Relation-ships may not relate to the subject matter of this manuscript.For more information about ASCO’s conflict of interest policy,please refer to www.asco.org/rwc or jco.ascopubs.org/site/ifc.

Jonathan K. KishEmployment: Cardinal HealthConsulting or Advisory Role: Cardinal Health (Inst)Travel, Accommodations, Expenses: Cardinal Health

Alicia I. RolinNo relationship to disclose

Zhaohui ZouNo relationship to disclose

James E. CucinelliNo relationship to disclose

Zaria TatalovichNo relationship to disclose

Mona SaraiyaNo relationship to disclose

Sean F. AltekruseNo relationship to disclose

REFERENCES1. Smith RA, Brooks D, Cokkinides V, et al: Cancer screening in the United States, 2013: A review of current American

Cancer Society guidelines, current issues in cancer screening, and new guidance on cervical cancer screening andlung cancer screening. CA Cancer J Clin 63:88-105, 2013

2. Benard VB, Thomas CC, King J, et al: CDC vital signs, Cervical cancer incidence, mortality, and screening—UnitedStates, 2007–2012. MMWR. Morb Mortal Wkly Rep 63:1004-1009, 2014

3. Akers AY, Newmann SJ, Smith JS: Factors underlying disparities in cervical cancer incidence, screening, andtreatment in the United States. Curr Probl Cancer 31:157-181, 2007

4. SaraiyaM, Ahmed F, Krishnan S, et al: Cervical cancer incidence in a prevaccine era in the United States, 1998-2002.Obstet Gynecol 109:360-370, 2007

5. Simard EP, Fedewa S,Ma J, et al: Widening socioeconomic disparities in cervical cancermortality amongwomen in 26states, 1993-2007. Cancer 118:5110-5116, 2012

6. Yabroff KR, Lawrence WF, King JC, et al: Geographic disparities in cervical cancer mortality: What are the roles of riskfactor prevalence, screening, and use of recommended treatment? J Rural Health 21:149-157, 2005

7. DeSantis C, Naishadham D, Jemal A: Cancer statistics for African Americans, 2013. CA Cancer J Clin 63:151-166,2013

8. Howlader N, Noone AM, Krapcho M, et al (eds): SEER cancer statistics review, 1975–2010. Bethesda, MD: NationalCancer Institute. http://seer.cancer.gov/csr/1975_2010/

9. Horner MJ, Altekruse SF, Zou Z, et al: U.S. geographic distribution of prevaccine era cervical cancer screening,incidence, stage, and mortality. Cancer Epidemiol Biomarkers Prev 20:591-599, 2011

10. Centers for Disease Control and Prevention; National Center for Health Statistics: Health, United States, 2010: WithSpecial Feature on Death and Dying. 2011. http://www.cdc.gov/nchs/data/hus/hus10.pdf

11. Stokley S, Jeyarajah J, Yankey D, et al: Human papillomavirus vaccination coverage among adolescents, 2007-2013,and postlicensure vaccine safety monitoring, 2006-2014–United States. MMWRMorb Mortal Wkly Rep 63:620-624,2014

12. Jemal A, Simard EP, Dorell C, et al: Annual Report to the Nation on the Status of Cancer, 1975-2009, featuring theburden and trends in human papillomavirus(HPV)-associated cancers and HPV vaccination coverage levels. J NatlCancer Inst 105:175-201, 2013

13. American Cancer Society; Cancer facts & figures 2015. http://www.cancer.org/research/cancerfactsstatistics/cancerfactsfigures2015/index

14. Pickle LW, Hao Y, Jemal A, et al: A newmethod of estimatingUnited States and state-level cancer incidence counts forthe current calendar year. CA Cancer J Clin 57:30-42, 2007

15. US Department of Health and Human Services; Health Resources and Services Administration (HRSAA). Area HealthResources Files (AHRF). http://ahrf.hrsa.gov/

16. United States Department of Agriculture, Economic Research Service: Rural-urban continuum codes. www.ers.usda.gov/data-products/rural-urban-continuum-codes/.aspx

17. Goovaerts P. Geostatistics for Natural Resources Evaluation. New York, NY, Oxford University Press, 1997



http://www.jco.ascopubs.org/site/ifc

http://seer.cancer.gov/csr/1975_2010/

http://www.cdc.gov/nchs/data/hus/hus10.pdf

http://www.cancer.org/research/cancerfactsstatistics/cancerfactsfigures2015/index

http://www.cancer.org/research/cancerfactsstatistics/cancerfactsfigures2015/index

http://ahrf.hrsa.gov/

http://www.ers.usda.gov/data-products/rural-urban-continuum-codes/.aspx

http://www.ers.usda.gov/data-products/rural-urban-continuum-codes/.aspx


18. Kim HJ, Fay MP, Feuer EJ, et al: Permutation tests for joinpoint regression with applications to cancer rates. Stat Med19:335-351, 2000

19. National Cancer Institute: Small area estimates for states, counties, & health service areas. http://sae.cancer.gov/

20. Centers for Disease Control and Prevention: 2013 NIS-Teen vaccination voverage table data. http://www.cdc.gov/vaccines/imz-managers/coverage/nis/teen/data/tables-2013.html

21. Kaiser Family Foundation: Status of state action on the Medicaid expansion decision. http://kff.org/health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act/

22. National Cancer Institute Surveillance, Epidemiology, and End Results Program: Policy for calculating Hispanicmortality for 1990+ data. http://seer.cancer.gov/seerstat/variables/mort/origin_recode_1990+/

23. Freeman HP, Wingrove BK: Excess Cervical Cancer Mortality: A Marker for Low Access to Health Care in PoorCommunities. Rockville, MD: National Cancer Institute, Center to Reduce Cancer Health Disparities, May 2005. NIHPub. No. 05–5282, 2007 http://www.cancer.gov/about-nci/organization/crchd/about-health-disparities/resources/excess-cervical-cancer-mortality.pdf

24. Scarinci IC, Garcia FA, Kobetz E, et al: Cervical cancer prevention: New tools and old barriers. Cancer 116:2531-2542,2010

25. The Henry J. Kaiser Family Foundation: Women’s Health Policy. Women’s Health Insurance Coverage. http://kff.org/womens-health-policy/fact-sheet/womens-health-insurance-coverage-fact-sheet/

26. Sanchez-Barriga JJ: Mortality trends from cervical cancer in the seven socioeconomic regions and the thirty twofederative entities of Mexico, 2000-2008 [in Spanish]. Gac Med Mex 148:42-51, 2012

27. Gonzaga CM, Freitas-Junior R, Barbaresco AA, et al: Cervical cancermortality trends in Brazil: 1980-2009. Cad SaudePublica 29:599-608, 2013

28. Shack L, Jordan C, Thomson CS, et al: Variation in incidence of breast, lung and cervical cancer and malignantmelanoma of skin by socioeconomic group in England. BMC Cancer 8:271, 2008

29. Lai JY, Tinker AV, Cheung WY: Factors influencing the willingness of US women to vaccinate their daughters againstthe human papillomavirus to prevent cervical cancer. Med Oncol 30:582, 2013

30. Williams WW, Lu PJ, Saraiya M, et al: Factors associated with human papillomavirus vaccination among young adultwomen in the United States. Vaccine 31:2937-2946, 2013

31. Downs LS Jr, Scarinci I, Einstein MH, et al: Overcoming the barriers to HPV vaccination in high-risk populations in theUS. Gynecol Oncol 117:486-490, 2010

32. Niccolai LM, Julian PJ, Bilinski A, et al: Geographic poverty and racial/ethnic disparities in cervical cancer precursorrates in Connecticut, 2008-2009. Am J Public Health 103:156-163, 2013

33. Castle PE, Rausa A, Walls T, et al: Comparative community outreach to increase cervical cancer screening in theMississippi Delta. Prev Med 52:452-455, 2011

34. Stanley SL, Thomas CC, King JB, et al: Predictors of never being screened for cervical cancer by metropolitan area.J Community Health 39:400-408, 2014

35. Bach PB: Gardasil: From bench, to bedside, to blunder. Lancet 375:963-964, 2010

36. Downs LS, Smith JS, Scarinci I, et al: The disparity of cervical cancer in diverse populations. Gynecol Oncol 109:S22-S30, 2008

37. Coughlin SS, Uhler RJ, Richards T, et al: Breast and cervical cancer screening practices among Hispanic and non-Hispanic women residing near the United States-Mexico border, 1999-2000. Fam Community Health 26:130-139,2003

38. Reiter PL, Katz ML, Ruffin MT, et al: HPV prevalence among women from Appalachia: results from the CARE project.PLoS One 8:e74276, 2013 [Erratum: PLoS One 8, 2013]

39. Jeudin P, Liveright E, Del Carmen MG, et al: Race, ethnicity, and income factors impacting human papillomavirusvaccination rates. Clin Ther 36:24-37, 2014

40. Kobetz E, Menard J, Barton B, et al: Patne en Aksyon: Addressing cancer disparities in Little Haiti through researchand social action. Am J Public Health 99:1163-1165, 2009

41. Gwede CK, Ashley AA, McGinnis K, et al: Designing a community-based lay health advisor training curriculum toaddress cancer health disparities. Health Promot Pract 14:415-424, 2013

42. Wells KJ, Luque JS, Miladinovic B, et al: Do community health worker interventions improve rates of screeningmammography in the United States? A systematic review. Cancer Epidemiol Biomarkers Prev 20:1580-1598, 2011

43. Zapka J, Taplin SH, Ganz P, Grunfeld E, Sterba K. Multilevel factors affecting quality: examples from the cancer carecontinuum. J Natl Cancer Inst Monogr 2012:11-19, 2012

44. RichardsonDB, VolkowND, KwanMP, et al: Medicine. Spatial turn in health research. Science 339:1390-1392, 2013


http://sae.cancer.gov/

http://www.cdc.gov/vaccines/imz-managers/coverage/nis/teen/data/tables-2013.html

http://www.cdc.gov/vaccines/imz-managers/coverage/nis/teen/data/tables-2013.html

http://kff.org/health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act/

http://kff.org/health-reform/state-indicator/state-activity-around-expanding-medicaid-under-the-affordable-care-act/

http://seer.cancer.gov/seerstat/variables/mort/origin_recode_1990+/

http://www.cancer.gov/about-nci/organization/crchd/about-health-disparities/resources/excess-cervical-cancer-mortality.pdf

http://www.cancer.gov/about-nci/organization/crchd/about-health-disparities/resources/excess-cervical-cancer-mortality.pdf

http://kff.org/womens-health-policy/fact-sheet/womens-health-insurance-coverage-fact-sheet/

http://kff.org/womens-health-policy/fact-sheet/womens-health-insurance-coverage-fact-sheet/


originalreport

Phenotype and Treatment of BreastCancer in HIV-Positive and -NegativeWomen in Cape Town, South Africa

abstract

Purpose An estimated 5.9 million people in South Africa are infected with HIV. Because antiretroviraltherapy has made infection with HIV a treatable, chronic condition, HIV-infected individuals are nowsurviving tomiddle and older age.We investigated the implications of HIV status for breast cancer in SouthAfrica.

Methods We compared clinical and demographic characteristics of women newly diagnosed with a firstprimary breast cancer at Tygerberg Hospital, Cape Town, South Africa, from January 2010 to December2011 by HIV status. We then compared HIV-positive patients with HIV-negative controls, matched 2:1 onage and ethnicity, with respect to chemotherapy regimens, toxicities, completion of systemic chemo-therapy, and changes in CD4 cell count.

Results Of 586 women with breast cancer, 31 (5.3%) were HIV positive, 420 (71.7%) were HIV negative,and 135 (23%) were untested for HIV. Women with HIV were younger than other women (P < .001). Thegroups did not differ in regard to stage at presentation, histologic subtype, tumor grade, nodal involvement,or hormone receptor positivity. More than 84% of patients who initiated systemic chemotherapy, re-gardless of HIV status, completed it without serious toxicity. Among HIV-positive patients receivingchemotherapy, themeanbaselineCD4cell countwas477cells/mL (standard deviation, 160cells/mL), andthe mean nadir was 333 cells/mL (standard deviation, 166 cells/mL).

Conclusion HIV-infected women were younger at breast cancer diagnosis than HIV-negative women butotherwise similar in phenotype and completion of chemotherapy. Longer term follow-up is needed toevaluate the effects of HIV, antiretroviral therapy, and chemotherapy on the survival and quality of life ofpatients with breast cancer.


INTRODUCTION

Of the 35.3 million people worldwide who wereestimated tobe infectedwithHIV/AIDS in2012,25million (71%) lived in Sub-Saharan Africa, includ-ing 6.1million in South Africa, the country with thelargest number of people living with HIV/AIDS inthe world.1

In the United Kingdom, between 1996 and 2008,the life expectancy of HIV/AIDS-infected peopleincreased by an average of 15 years, presumablybecause of antiretroviral therapy (ART). By theend of the period, the life expectancy of peoplewith HIV/AIDS was only 13 years shorter than thatof the general population of the United Kingdom.2

In South Africa, although no comparable dataare yet available, it is clear that persons with HIVinfection are living longer as a result of ARTthan infected people did before the use of ART.

Therefore, like uninfected people, those with HIVinfections will incur increasing risks for the epi-thelial malignancies associated with aging.

By 2010, 19%of all deaths in theSwissHIVCohortStudywereattributable to non–AIDS-definingcan-cers.3 Persons with HIV had a higher than aver-age risk for anal cancer, lung cancer, certain headand neck cancers, hepatocellular carcinoma, andHodgkin lymphoma, but not for breast cancer,prostate cancer, or colorectal cancer. Little isknownabout thephenotype of thesemalignanciesand about the outcomes of standard oncologictreatment among HIV-positive patients, includingthose receiving ART.4-9

In a series of 1,092 black women consecutivelydiagnosed with breast cancer in Soweto, SouthAfrica, between January 2006 and July 2012,10

19.7% were found to be HIV positive. Nearly a

Lizanne Langenhoven

Pieter Barnardt

Alfred I. Neugut

Judith S. Jacobson

Lizanne Langenhoven andPieter Barnardt,Stellenbosch University,South Africa; Alfred I.Neugut and Judith S.Jacobson, Mailman Schoolof Public Health andHerbert IrvingComprehensive CancerCenter, ColumbiaUniversity; and Alfred I.Neugut, College ofPhysicians and Surgeons,Columbia University, NewYork, NY.

Supported by US NationalCancer Institute Grant No.D43-CA153715.Authors’ disclosures ofpotential conflicts ofinterest and contributionsare found at the end of thisarticle.Corresponding author:Judith S. Jacobson, DrPH,MBA, Department ofEpidemiology, MailmanSchool of Public Health,Columbia University, 722West168thSt,R732,NewYork, NY 10032; e-mail:[email protected].




quarter of theHIV-positivewomenwerediagnosedwith a CD4 count less than 200 cells/mL. The HIV-positive women were younger at diagnosis thanthosewhowereHIVnegative ornot tested, but theydid not differ in tumor characteristics or prognosticfactors. This finding was in stark contrast to an-ecdotal reports of a more aggressive phenotype inHIV-positive women.

It was recently shown that HIV-positive patientswere diagnosed with more advanced-stage cervi-cal cancer than their HIV-negative peers, hada decreased likelihood of completing radicalchemoradiotherapy, and fared worse overall inregard to treatment-related toxicity, especiallywhen chemotherapywas used in conjunctionwithradiotherapy.11,12 Whether these findings alsoapply to women diagnosed with breast cancerremains to be determined because the risk factorsassociated with these epithelial malignancies areinherently different.

Given that breast cancer is the most commonnon–AIDS-defining cancer in the global femalepopulation13 and that risk increases with age,the proportion of patients with breast cancer inSouth Africa who are HIV positive is likely to in-crease in the next few years. Yet at present, nospecific guidelines are available to clinicians car-ing for patients with both diagnoses. A case seriesin the erabeforewidespreaduseof ART found thatsuch patients had poor tolerance for systemiccytotoxic therapy and that most experienced pro-gressive immune impairment with progression ofHIV. The authors cautioned that hormonal ther-apy, rather than systemic chemotherapy, shouldbe considered in the adjuvant setting.14 The pri-mary objective of this study is to compare HIV-positive and -negative women with breast cancerin terms of their clinical and demographic char-acteristics and tolerance of standard chemother-apy regimens.

METHODS

After obtaining approval from our Ethical ReviewBoard, we reviewed the charts of all women reg-istered with a diagnosis of breast cancer on theoncology service at the Tygerberg Academic Hos-pital in Cape Town, South Africa, from January2010 through December 2011. Of 816 patientswhose records we reviewed, we excluded 154 forinsufficient data, 47 for a prior diagnosis of breastcancer, 12 for male sex, 16 for ductal carcinomain situ, and five for benign breast disease.

We then compared patients by HIV status withrespect to age at diagnosis, ethnic group, and

menopausal status. We also compared them withrespect to disease phenotype (stage at diagno-sis,15 histologic subtype, grade of differentiation,nodal status, and positivity to estrogen receptor,progesterone receptor [PR], and humanepidermalgrowth factor receptor 2). Chemotherapy toleranceand toxicities were captured for the HIV-positivesubgroup and an HIV-negative comparison groupmatched 2:1 on age and ethnicity. We determinedhow many of the assigned chemotherapy cycleswere completed. Allowance was made for inter-ruptions or postponements as well as dose re-ductions as a result of toxicities. Patients wereconsidered to have completed chemotherapyif they had completed the number of cyclesassigned to them at the beginning of treatment,regardless of dose reduction or treatment inter-ruption. Toxicity was scored according to theNational Cancer Institute Common TerminologyCriteria for Adverse Events (version 4.0).16 Weassigned a toxicity grade to each patient at eachcycle for the following hematologic parameters:hemoglobin, WBC count, platelets, neutrophils,lymphocytes, and CD4 count. Clinical toxicitiesevaluated were nausea and vomiting, alopecia,thrombophlebitis, and weight loss. The CD4 nadirwas defined as the lowest value recorded at anytime after initiation of treatment. It is standardpractice to measure the CD4 cell count at cycle 3and on completion of chemotherapy.

Statistical analysis was performed using Statisticasoftware (Dell Statistica, Tulsa, OK). Where datawere incomplete, an unknown category was cre-ated before data analysis. The statistical signifi-canceof differences amongHIV status groupswasevaluated by x2 test and analysis of variance, withsignificance set at P , .05.

RESULTS

Of the 586 patients whomet the inclusion criteria,451 (77%) were tested for HIV and 135 (23%)were not; 31 patients (5.3%) tested positive and420 (71.7%) tested negative (Table 1). A total of359 patients (61.3%) were of mixed race, 158(26.9%) were of European descent, and 69(11.8%) were of African descent.

Ethnicity was strongly associated with HIV status;27.1%ofAfricanpatients but only 3.6%of thoseofmixed race and none of those of European back-ground were HIV positive (P, .001). Nearly 90%of African patients were tested for HIV;mixed-racepatients and those of European descent weremuch less likely to be tested. The patients’medianage at presentation was 56 years (range, 20 to 97years). The median age of the HIV-positive group



Table 1 – Demographic and Clinical Characteristics of Study Participants by HIV Status

Characteristic

HIV Status

All Patients

P

Positive Negative Unknown

No. % No. % No. % No. %

All patients 31 5.3 420 71.7 135 23.0 586.0 100.0

Age at diagnosis, years , .001

Median 42 54 62 56

Range 29-60 20-91 30-97 20-97

Age by ethnic group, years , .001

African

Median 40 49 63 48

Range 29-54 20-77 44-82 20-82

Mixed race

Median 45 54 59 55

Range 32-60 22-89 30-86 22-89

European

Median — 58 68 61

Range — 30-91 24-97 24-97

Ethnic group , .001

African 18 58.1 43 10.2 8 5.9 69 11.8

Mixed race 13 41.9 272 64.8 74 54.8 359 61.3

European 0 0.0 105 25.0 53 39.3 158 26.9

Menopausal status , .001

Premenopausal 20 64.5 143 34.1 37 27.4 200 34.1

Postmenopausal 7 22.6 271 64.5 91 67.4 369 63.0

Perimenopausal 4 12.9 6 1.4 1 0.7 11 1.9

Unknown 0 0.0 0 0.0 6 4.4 6 1.0

Stage at diagnosis .04

I 2 6.5 21 5.0 9 6.7 32 5.5

II 8 25.8 174 41.4 49 36.3 231 39.4

III 11 35.5 164 39.0 44 32.6 219 37.4

IV 10 32.3 61 14.5 33 24.4 104 17.7

Histologic subtype .34

Infiltrating ductal 22 71.0 355 84.5 110 81.5 487 83.1

Lobular 2 6.5 28 6.7 8 5.9 38 6.5

Other/unspecified 7 22.5 37 8.8 17 12.6 61 10.4

Grade .83

Well differentiated 3 9.7 38 9.0 13 9.6 54 9.2

Moderately differentiated 9 29.0 119 28.3 46 34.1 174 29.7

Poorly differentiated 5 16.1 89 21.2 22 16.3 116 19.8

Ungraded 14 45.2 174 41.4 54 40.0 242 41.3

No. of nodes involved .10

0-3 13 41.9 182 43.3 48 35.6 243 41.5

. 3 2 6.5 83 19.8 13 9.6 98 16.7

Unknown 16 51.6 155 36.9 74 54.8 245 41.8




was 42 years, that of the HIV-negative group was 54years, and that of the groupwith unknownHIV statuswas 62 years (P, .001). Within each ethnic group,HIV-positivepatientswereyounger thanHIV-negativepatients, who were younger than untested patients(P, .001). Within the HIV-positive and -negativegroups, African patients were younger than pa-tients of mixed race and patients of Europeanbackground. In the untested group, the Africanpatients were older than those of mixed race.

Fifty-five percent of all patients were diagnosedwith stage III or IV breast cancer, including 68.8%of HIV-positive patients, 54% of HIV-negative pa-tients, and 57% of untested patients (P, .001 forthe association of HIV status with stage). Overall,488 patients (83.1%) had infiltrating ductal car-cinoma; only 38 patients (6.5%) had infiltratinglobular carcinoma. The 61 remaining patients hadtumors of other rare or unspecified histologic sub-types. Although HIV-positive patients were morelikely than others to have rare epithelial subtypes,the differences were not statistically significant.The grade of tumor differentiation was not asso-ciated with HIV status.

Of the 341 patients offered axillary sampling (be-cause they had surgical treatment andwere youn-ger than 70 years), 243 (71.3%) had fewer thanfour nodes involved. Nodal status was not associ-ated with HIV status.

With regard to hormonal prognostic markers, HIV-negative and untested patients were nearly twiceas likely as HIV-positive patients to have PR-positive disease (P = .02). The groups did notdiffer with respect to estrogen receptor, humanepidermal growth factor receptor 2, or triple-negative status (Table 1).

Of the 31 HIV-positive patients, 22 (71%) wereknown to be HIV positive at breast cancer diag-nosis; the other nine were diagnosed at their firstclinic visit. Twelve patients had been on ART for3 months or longer. All patients newly diagnosedwith HIV were referred for initiation of ART. TwoHIV-positive patients had pulmonary tuberculosisat the time of their cancer diagnosis, one patientwas newly diagnosed with a pulmonary aspergil-loma, and one patient was diagnosed with Staph-ylococcus pneumonia.

Of the 31HIV-positive patients, five failed to returnto the hospital for surgical or other treatment afterstaging examinations were performed. Anotherfive patients did not receive chemotherapy be-cause of comorbidities (eg, ischemic heart dis-ease). One patient presented with stage I breastcancer and was managed with surgery only. Onepatient had just begun chemotherapy and had norecord of toxicity at time of data analysis. There-fore, we report on toxicities among 19HIV-positivepatients who received systemic chemotherapyand 39 matched HIV-negative controls.

In the HIV-positive group, 16 patients received ananthracycline-based regimen in combination withcyclophosphamide and/or fluorouracil. The otherthree patients received a cyclophosphamide/methotrexate/fluorouracil (CMF) regimen be-cause of cardiac compromise.

Three HIV-positive patients did not complete theirassigned chemotherapy. The first patient discon-tinued in themiddle of chemotherapybut returned7months later with local progression of her breastcancer. The second patient completed four cy-cles of palliative chemotherapy for stage IV dis-ease before discontinuing treatment. She was

Table 1 – Demographic and Clinical Characteristics of Study Participants by HIV Status (continued)

Characteristic

HIV Status

All Patients

P

Positive Negative Unknown

No. % No. % No. % No. %

Receptor status

Estrogen receptor positive 16 51.6 268 63.8 91 67.4 375 64.0 .17

Progesterone receptor positive 9 29.0 215 51.2 71 52.6 295 50.3 .02

HER2 receptor positive 10 32.3 149 35.5 41 30.4 200 34.1 .72

Triple negative 5 16.1 65 15.5 19 14.1 89 15.2 .92

Completion of systemic chemotherapy

Chemotherapy administered 19 61.3 39 100.0

Chemotherapy completed 16 84.2 33 84.6

Chemotherapy-related mortality 1 5.3 1 2.6

Abbreviation: HER2, human epidermal growth factor receptor 2.



eventually ascertained to have died, possibly as aresult of chemotherapy-related toxicity. The thirdpatient completed four cycles of chemotherapybut discontinued treatment after her tumor spreadto the contralateral breast.

In the HIV-negative group, 35 patients received ananthracycline-based regimen in combination withcyclophosphamide and/or flourouracil, and fourpatients received CMF. Six patients in the HIV-negative cohort did not complete assigned chemo-therapy. Two patients developed disease progres-sion on treatment; three patients were lost to follow-up or transferred to a different hospital; and onepatient died after three cycles of CMF, possibly as aresult of treatment-related toxicity. Thus, 84% ofboth HIV-positive and HIV-negative patients com-pleted their assigned chemotherapy (Table 1).

Table 2 compares the toxicity parameters of theHIV-positive and -negative subgroups. Alopecia,thrombophlebitis, and nausea and vomiting werethe most frequently observed clinical toxicities inboth groups (data not shown). No statisticallysignificant difference between the HIV-positive

and -negative subgroups could be demonstratedfor hemoglobin, WBC count, platelets, and neu-trophils (Table 2). However, HIV-positive patientshad more lymphopenia than HIV-negative pa-tients; five HIV-positive patients (26.4%) com-pared with no HIV-negative patients developedgrade 3 or 4 toxicity (P = .001).

Among patients in the HIV-positive cohort, themean CD4 cell count at diagnosis was 402cells/mL (standard deviation [SD], 244 cells/mL;range, 66 to 1,075 cells/mL). Among the 15 pa-tients who received chemotherapy and had com-plete serial CD4values, themeanCD4cell count atdiagnosis was 477 cells/mL (SD, 160 cells/mL),and the mean nadir was 333 cells/mL (SD, 166cells/mL;P, .001;Fig1).FivepatientswhoseCD4nadir was not recorded were excluded from thenadir value analysis.

Although it did not affect the completion of theirplanned chemotherapy, two patients required anadjustment to their ART regimen during chemo-therapy, one as a result of a grade 4 anemia andthe other as a result of a deterioration of herglomerular filtration rate.

DISCUSSION

In our cohort, the majority of HIV-positive pa-tients (58.1%) were of African descent, and theHIV-positive group was much younger than theHIV-negative group (42 v 54 years, respec-tively). Two important arguments in the litera-ture are relevant to these observations. First,African American women are known to be youn-ger at breast cancer diagnosis than Americanwomen of European descent. Second, Voutsa-dakis and Silverman17 argued that HIV-infectedwomen with breast cancer are younger thanother women with breast cancer because HIV-infected women are younger than HIV-negativewomen in the general population. Supportingthat argument, in South Africa, the highest prev-alence of HIV has been recorded in Africanwomen of reproductive age.18,19 The associa-tion of menstrual status with HIV status reflectsthe relative youth of the HIV-positive patients,most of whom were well below the mean age atmenopause in the general population, which iscurrently estimated to be 51.4 years.20

The HIV prevalence of 5.3% observed in the totalsample of 586 patients was less than the preva-lenceof16%reported for thegeneral populationofSouth Africa. However, our institution is located intheWesternCapeprovince,whereHIVprevalenceis known tobe lower thanelsewhere in thecountry.

Table 2 – Hematologic Toxicities Among Patients With Breast Cancer by HIV Status

Toxicity

HIV Status

All Patients

P

Positive Negative

No. % No. % No. %

All patients 19 32.8 39 67.2 58 100.0

Hemoglobin .17

Grade 0 1 5.3 10 25.6 11 19.0

Grade 1/2 17 89.5 28 71.8 45 77.6

Grade 3/4 1 5.3 1 2.56 2 3.4

WBC count .67

Grade 0 16 84.2 31 79.5 47 81.0

Grade 1/2 3 15.8 8 20.5 11 19.0

Grade 3/4 0 0.0 0 0.0 0 0.0

Platelets .48

Grade 0 19 100.0 38 97.4 57 98.3

Grade 1/2 0 0.0 1 2.56 1 1.7

Grade 3/4 0 0.0 0 0.0 0 0.0

Neutrophil count .33

Grade 0 12 63.2 17 43.6 29 50.0

Grade 1/2 5 26.3 18 46.2 23 39.7

Grade 3/4 2 10.5 4 10.3 6 10.3

Lymphocytes .001

Grade 0 2 10.5 15 38.5 17 29.3

Grade 1/2 12 63.2 24 61.5 36 62.1

Grade 3/4 5 26.3 0 0.0 5 8.6



Because HIV remains highly stigmatized in SouthAfrica, obtaining accurate prevalence data for thegeneral population is challenging. Current HIVinfection estimates are based on indirect mea-sures, such as antenatal incidence data, shiftsin mortality patterns, and the analysis of annualsurveys.18,19 The prevalence of HIV among pa-tients inour samplewhowere testedwas7.1%,notmuch lower than the reported national prevalenceof 7.7% in women age 55 to 59 years. However,135 of our patients (23%) were not tested. Thedilutional effect of including the untested, pre-dominantly older patients with breast cancer inthe prevalence calculation became evident whenwe examined the data by age group. Among theAfrican patients, 26.1% tested positive for HIV;their median age was 40 years (range, 39 to 45years). The prevalence of HIV among patients ofmixed race and those of European backgroundwas less than the expected national average butstill consistent with the age/race distribution inother samples.

Estrogen receptor sensitivity was not associatedwithHIV status, butHIV-positive patientswere lesslikely than others to have hormone-sensitive dis-ease, perhaps because they were younger thanthe other patients. No statistically significant dif-ference was detected in stage at presentation,histologic subtype, tumor grade, or nodal involve-ment between the HIV-positive group and theother groups.We found little support for the notionthat HIV altered the breast cancer phenotype, inaccordance with the phenotype described byHurley et al14 and Sarhan et al.22

Our HIV-positive patients who received chemo-therapy tolerated it well. Earlier case reports onHIV-positive patients receiving systemicanthracycline-based therapy described adverseoutcomes, including neutropenic sepsis, acuterespiratory distress syndrome, Candida esophagi-tis, and death.14,23 Our patients started ART atinitiation of chemotherapy regardless of CD4count, per national ART guidelines. We believe

that with ART, most HIV-positive patients who areotherwise eligible for chemotherapy can benefitfrom it.

Although the literature implies that grade 3 or 4myelosuppression is commonamongHIV-positivepatients with breast cancer receiving chemother-apy, only 26.4% of our HIV-positive patients onchemotherapy (but none of the HIV-negative pa-tients on chemotherapy) experienced grade 3 or 4lymphopenia. They did not require dose adjust-ments, and they completed treatment. No dose-limiting toxicity was observed in the neutrophilcount, and only two patients developed grade 3neutropenia. A single case of grade 4 toxicity wasobserved in the hemoglobin level of a patient on anevirapine-based ART regimen.

Failure to return to the hospital for treatment is notuncommon in South Africa. Many patients whocome to Tygerberg Hospital are members of theXhosa tribe residing in the rural Eastern Cape.Many of these patients speak only Xhosa, do nothave access to telephones, have never seen theinside of ahospital, and travelmore than800kmtoreceive care at Tygerberg Hospital, which is theclosest public hospital with full oncology services.Few patients have themeans to return even once,let alone every 21 days, to receive chemotherapy.In our cohort, patients with and without HIVseemed to be equally affected by socioeconomicbarriers to care. Further research is needed tohelpus understand and address the barriers faced byour patients.

Our study was not designed to determine whetherHIV either prevented or promoted the develop-ment of breast cancer. All we observed in our HIV-positive patients with breast cancer was the co-existence of two chronic diseases. To date, theonly association between CD4 count and the rel-ative risk of developing a malignancy has beenshown in cancerswith a known viral pathogenesis,the so-called AIDS-defining cancers.

Among our HIV-positive patients who receivedchemotherapy, the CD4 count declined by a thirdof pretreatment values, fromamedianof477cells/mL (range, 234 to 807 cells/mL) to a median nadirof 333 cells/mL (range, 62 to 713 cells/mL). Lym-phoma studies published in the pre-ART eraestablished the perception that HIV-positive pa-tients who received chemotherapy were at highrisk for persistent, severe compromise of immu-nologic function and poor outcomes.24 In AIDSMalignancies Consortium Trial 010, an excessrisk of death was associated with postchemother-apy opportunistic infections in patients with CD4

1,000

800

600

400

CD4

Coun

t (ce

lls/µ

L)

200 Diagnosis

Nadir0CD4 Count (time)

Fig 1 –

Trend of CD4 count duringchemotherapy treatment.



counts less than 50 cells/mL.25 However, inmore recent studies, despite initial low CD4counts, patients receiving ART tolerated che-motherapy without significant toxicity.26-28 Fur-ther studies are needed to confirm thesefindings among patients with non–HIV-relatedcancers. South Africa is an ideal setting in whichto conduct such studies because of its dispro-portionate burden of HIV.

The greatest limitation of this study was that itwas based on chart reviews. The chart data werenot collected for the purpose of evaluating theeffects of HIV on breast cancer phenotypes andoutcomes. In addition, because CD4 valueswere not determined at set intervals after com-pletion of treatment, we have no data on thelonger term effects of systemic chemotherapyon the progression of HIV and on patientsurvival.

Among the patients in our sample, thosewhowereHIV positive were diagnosed at a younger age than

other patients with breast cancer and were pre-dominantly of African ethnicity. The age distribu-tion of HIV-positive patients with breast cancerseems to correspond to the agedistribution ofHIV-positive women in the general population. Al-thoughHIV-positive patients weremore likely thanothers to have PR-negative cancers, that differ-ence in phenotype was likely to have been aresult of age and ethnicity rather than HIV itself.The HIV-positive patients who received systemicchemotherapy tolerated it well, experienced nosignificant hematologic toxicity, and developed noopportunistic infections during treatment. Themean decline in CD4 cell count during chemo-therapy was 30%. Our findings should reassurepatients with HIV and breast cancer, as well astheir providers, that, with ART and careful moni-toring, they can benefit from appropriatechemotherapy.



Conception and design: Lizanne Langenhoven, Pieter Bar-nardt, Judith S. JacobsonAdministrative support: Lizanne Langenhoven

Provision of study materials or patients: Lizanne Langenhoven,Pieter Barnardt

Collection and assembly of data: Lizanne Langenhoven, PieterBarnardt

Data analysis and interpretation: Lizanne Langenhoven, AlfredI. Neugut, Judith S. Jacobson



The following represents disclosure information provided byauthors of this manuscript. All relationships are considered

compensated. Relationships are self-held unless noted. I =Immediate Family Member, Inst = My Institution. Relation-ships may not relate to the subject matter of this manuscript.For more information about ASCO’s conflict of interest policy,please refer to www.asco.org/rwc or jco.ascopubs.org/site/ifc.

Lizanne LangenhovenNo relationship to disclose

Pieter BarnardtNo relationship to disclose

Alfred I. NeugutStock or Other Ownership: Stemline TherapeuticsConsulting or Advisory Role: Pfizer, Teva, Otsuka, UnitedBioSource Corporation, EHE International

Judith S. JacobsonNo relationship to disclose

REFERENCES1. UNAIDS: Global report: UNAIDS report on the global AIDS epidemic. http://www.unaids.org/sites/default/files/

media_asset/UNAIDS_Global_Report_2013_en_1.pdf

2. May M, Gompels M, Delpech V, et al: Impact of late diagnosis and treatment on life expectancy in people with HIV-1:UK Collaborative HIV Cohort (UK CHIC) Study. BMJ 343:d6016, 2011

3. Weber R, Ruppik M, Rickenbach M, et al: Decreasing mortality and changing patterns of causes of death in the SwissHIV Cohort Study. HIV Med 14:195-207, 2013

4. Grulich AE, van Leeuwen MT, Falster MO, et al: Incidence of cancers in people with HIV/AIDS compared withimmunosuppressed transplant recipients: A meta-analysis. Lancet 370:59-67, 2007

5. Engels EA, Biggar RJ, Hall HI, et al: Cancer risk in people infected with human immunodeficiency virus in the UnitedStates. Int J Cancer 123:187-194, 2008

6. Powles T, Macdonald D, Nelson M, et al: Hepatocellular cancer in HIV-infected individuals: Tomorrow’s problem?Expert Rev Anticancer Ther 6:1553-1558, 2006

7. Engels EA, Pfeiffer RM, Goedert JJ, et al: Trends in cancer risk among people with AIDS in the United States 1980-2002. AIDS 20:1645-1654, 2006






http://www.unaids.org/sites/default/files/media_asset/UNAIDS_Global_Report_2013_en_1.pdf

http://www.unaids.org/sites/default/files/media_asset/UNAIDS_Global_Report_2013_en_1.pdf


8. Chaturvedi AK, Madeleine MM, Biggar RJ, et al: Risk of human papillomavirus-associated cancers among personswith AIDS. J Natl Cancer Inst 101:1120-1130, 2009

9. SilverbergMJ, AbramsDI: AIDS-defining and non-AIDS-definingmalignancies: Cancer occurrence in the antiretroviraltherapy era. Curr Opin Oncol 19:446-451, 2007

10. Cubasch H, Joffe M, Hanisch R, et al: Breast cancer characteristics and HIV among 1,092 women in Soweto, SouthAfrica. Breast Cancer Res Treat 140:177-186, 2013

11. Simonds H, Wright J.D, du Toit N, et al: Completion of and early response to chemoradiation among human im-munodeficiency virus (HIV)-positive and HIV-negative patients with locally advanced cervical carcinoma in SouthAfrica. Cancer 118:2971-2979, 2012

12. Simonds H, Neugut A.I, Jacobson JS:HIV status and acute hematologic toxicity among patients with cervix cancerundergoing radical chemoradiation. Int J Gynecol Cancer 25:884-890, 2015

13. World Health Organization: WHO National Cancer Control Programmes 2012. http://www.who.int/cancer/en/

14. Hurley J, Franco S, Gomez-Fernandez C, et al: Breast cancer and human immunodeficiency virus: A report of 20cases. Clin Breast Cancer 2:215-220, 2001

15. Breast, in Edge SB, Byrd DR, Compton CC, et al (eds): AJCC Cancer Staging Manual, 7th edition. New York, NY,Springer, 2010, pp 345-376

16. National Cancer Institute:Common Terminology Criteria for Adverse Events, version 4.0. http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_QuickReference_5x7.pdf

17. Voutsadakis IA, Silverman LR: Breast cancer in HIV-positive women: A report of four cases and review of the literature.Cancer Invest 20:452-457, 2002

18. Shisana O, Rehle T, Simbayi LC, et al: The South African National HIV Prevalence, HIV Incidence, Behavior and Com-municationSurvey, 2008. http://www.health-e.org.za/wp-content/uploads/2013/05/2966e129fc39e07486250fd47fcc266e.pdf

19. National Department of Health, South Africa: The 2010 National Antenatal Sentinel HIV and Syphilis PrevalenceSurvey in South Africa. http://www.hst.org.za/sites/default/files/hiv_aids_survey.pdf

20. Gold EB, Bromberger J, Crawford S, et al: Factors associated with age at natural menopause in amultiethnic sample ofmidlife women. Am J Epidemiol 153:865-874, 2001

21. Bahadini B, Herzog J, Somlo G, et al: Prevalence of BRCA mutations in a cohort of young high-risk breast cancerpatients. J Clin Oncol 22, 2004 (suppl; abstr 9662)

22. Sarhan M, DePaz HA, Oluwole SFD: Breast cancer in women with human immunodeficiency virus infection:Pathological, clinical, and prognostic implications. J Womens Health (Larchmt) 19:2261-2266, 2010

23. El-Rayes BF, Berenji K, Schuman P, et al: Breast cancer in women with human immunodeficiency virus infection:Implications for diagnosis and therapy. Breast Cancer Res Treat 76:111-116, 2002

24. Sparano JA, Hu X, Wiernik PH, et al: Opportunistic infection and immunologic function in patients with humanimmunodeficiency virus-associated non-Hodgkin’s lymphoma treated with chemotherapy. J Natl Cancer Inst 89:301-307, 1997

25. Kaplan LD, Lee JY, Ambinder RF, et al: Rituximab does not improve clinical outcome in a randomized phase 3 trial ofCHOP with or without rituximab in patients with HIV-associated non-Hodgkin lymphoma: AIDS-Malignancies Con-sortium Trial 010. Blood 106:1538-1543, 2005

26. Bower M, Stebbing J, Tuthill M, et al: Immunologic recovery in survivors following chemotherapy for AIDS-related non-Hodgkin lymphoma. Blood 111:3986-3990, 2008

27. Antinori A, Cingolani A, Alba L, et al: Better response to chemotherapy and prolonged survival in AIDS-relatedlymphomas responding to highly active antiretroviral therapy. AIDS 15:1483-1491, 2001

28. Vaccher E, Spina M, di Gennaro G, et al: Concomitant cyclophosphamide, doxorubicin, vincristine, and prednisonechemotherapy plus highly active antiretroviral therapy in patients with human immunodeficiency virus-related, non-Hodgkin lymphoma. Cancer 91:155-163, 2001


http://www.who.int/cancer/en/

http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_QuickReference_5x7.pdf

http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_QuickReference_5x7.pdf

http://www.health-e.org.za/wp-content/uploads/2013/05/2966e129fc39e07486250fd47fcc266e.pdf

http://www.health-e.org.za/wp-content/uploads/2013/05/2966e129fc39e07486250fd47fcc266e.pdf

http://www.hst.org.za/sites/default/files/hiv_aids_survey.pdf


reviewarticle

Compulsory Licenses for Cancer Drugs:Does Circumventing Patent RightsImprove Access to OncologyMedications?

abstract

Worldwide, there are enormous inequities in cancer control that cause poor outcomes among patients withcancer who live in low- andmiddle-income countries (LMICs). One of the biggest challenges that oncologyfaces today is how to increase patient access to expensive, but life-saving, therapies in LMICs. Access tocancermedications in LMICs is amajor problem, especially in recent years, as the costs of these therapiescontinue to riseexponentially. Onemechanismavailable to LMICs to improveaccess tocancermedicationsallows a country to pursue a compulsory license for a given drug. Here, we will review how the legalframework in the World Trade Organization’s Trade-Related Aspects of Intellectual Property RightsAgreement and theDohaDeclaration supports countries to circumvent patent laws and acquire compulsorylicenses for essential medicines.Wewill also discuss the current and future role of compulsory licenses inoncology and how compulsory licenses may improve access to cancer drugs in LMICs.


INTRODUCTION

Cancer is one of the leading global causes ofmorbidity and mortality, with approximately 14million new cases and 8.2 million deaths in2012.1 Worldwide, there are enormous inequitiesin cancer control that result in poor outcomes forpatients in low- and middle-income countries(LMICs).2,3 More than 60% of new cases occurin Africa, Asia, and Latin America, and these re-gions account for 70% of the world’s cancerdeaths.4 The burden of cancer in LMICs alsosignificantly impacts the economy of these re-gions, yet only 6% of global resources for cancerare spent in the developing world.5,6 In the nearfuture, this gap in cancer care between LMICs andhigh-income regions is predicted to grow. By2020, cancer is likely to kill more than twice asmany people worldwide as in 2000 and the deathrate in LMICs will be at least five times greater thanin high-income countries.7

Cancer outcomes are clearly related to where onelives.8 Patients with breast cancer in the UnitedStates have a 5-year overall survival of 84%,whereas, in Gambia, 5-year overall survival is only12%.7One of the biggest challenges that faces theworld of oncology is how LMICs will address the

rising burden of cancer in their regions—this in-cludes difficult decisions of when to offer expen-sive cancer therapies to patients and how tobest organize cancer prevention programs withinhealth systems that have suboptimal infrastruc-ture and support.2,3,5 Access to cancer drugs forLMICs is a growing problem because many newmedications in oncology are exorbitantly expen-sive and prices have risen in recent years: theaverage price of a cancer therapy has doubledfrom US$5,000 per month in 2003 to US$10,000per month in 2013 and continues to rise.9 Highcosts are a major, and often insurmountable,barrier in poor countries for which the price ofstandard cancer medications are simply too highrelative to the national and individual income.10

Countries design and enforce patent laws to pro-tect intellectual property of pharmaceutical andother products and services. Many authors arguethat this is an important way to incentivize drugdevelopment, as it creates a de facto monopoly,which, in theory, allowsmedications tobepriced insuch a way that covers the costs of production,recoups investment in researchanddevelopment,and, ideally, brings in profits to innovative compa-nies and adequate financial returns for their

Cinthia Leite FrizzeraBorges Bognar

Brittany L. Bychkovsky

Gilberto de Lima Lopes Jr

Cinthia Leite FrizzeraBorges Bognar, CentroPaulista deOncologia, andOncoclinicas do Brasil;Gilberto de Lima Lopes Jr,Oncoclinicas Group, SaoPaulo, Brazil; Brittany L.Bychkovsky, Dana-FarberCancer Center HarvardMedical School, Boston,MA; and Gilberto de LimaLopes Jr, Johns HopkinsUniversity School ofMedicine, Baltimore, MD.Authors’ disclosures ofpotential conflicts ofinterest and contributionsare found at the end of thisarticle.Corresponding author:Gilberto de Lima Lopes Jr,MD, MBA, Centro Paulistade Oncologia andOncoclinicas do BrasilGroup, Rua Maranh~ao569/4, S~ao Paulo, Brazil01240-001; e-mail:[email protected].





investors. Newmedications covered under patentlaws are priced for high-income countries, whichalmost always makes them unaffordable to largenumbers of patients and health systems in LMICs.

WTO TRADE-RELATED ASPECTS OF INTELLECTUALPROPERTY RIGHTS AGREEMENT AND THE DOHADECLARATION

Before 1995, there was significant variabilityacross countries in how patented medicationswere regulated. This changed when the WorldTrade Organization (WTO) andmember countriesapproved the Trade-Related Aspects of Intellec-tual Property Rights (TRIPS) Agreement, whichgreatly influenced interpretation of patent laws.11

Before TRIPS, more than 40 countries did not of-fer patent protection for pharmaceutical productsand many developing countries that did only of-fered protection for 5 to 7 years.11 The agreement,which went into effect in 1995, required all WTOcountries to provide patent protection for a min-imum of 20 years, and this included patents forboth pharmaceutical processes and products;however, the TRIPS agreement contains provi-sions that allow individual countries to balancetheir intellectual property and patents with theirown health and development needs.11 For exam-ple, countries can issue compulsory licenses tomake generic medications on the grounds ofpublic interest. This means that a country canproduce generic drugs without the consent ofthe patent holder even when intellectual propertyrights are still in effect.

After the TRIPS Agreement, the Doha Declarationwas adopted in 2001 by the WTO and supportedthe notion that member states could circumventpatent rights by issuing compulsory licenses thatwould allow them to access essential medicationsif these medications were urgently needed to pro-tect the public’s health. TheDohaDeclaration alsocontained provisions that allowed LMICs that werewithout drug manufacturing capabilities to importmedications produced elsewhere under compul-sory licensing. The pharmaceutical company thatowns the original patent still holds the right to itsinvention and is entitled to compensation underTRIPS, which, therefore,means that governmentsmust first negotiate with the pharmaceutical in-dustry directly to ask to purchase the desiredmedication at a reduced price or they mustrequest a voluntary license from the pharmaceu-tical company to manufacture the drug beforethey can issue a compulsory one.5,12,13 Onlyin situations of an emergency or extreme urgency,such as an epidemic, can governments forgo the

process of negotiating with the patent holder.14 Itwas after the Doha Declaration that several coun-triesusedcompulsory licensing to increaseaccessto HIV/AIDs drugs.15

An amendment to the WTO TRIPS agreementallows for least developed countries (LDCs) thatare members of the WTO to import generic med-ications, ignoring both local and international pat-ent laws.16 This pharmaceutical product waiverwas recently extended until 2033 and applies to34 WTO member countries.16,17 These LDCscan also produce their own generics, but, todate, most countries involved do not have apharmaceutical generic manufacturing indus-try. Among all WTO LDCs, only Uganda, Nepal,and Bangladesh have a nascent pharmaceuti-cal industry, and these three countries indeedled the most recent negotiations to get theamendment approved.18

USE OF COMPULSORY LICENSES FOR HIV THERAPY

To date, compulsory licenses have been widelyused to enhance access to medications to treatcommunicable disease, such as HIV, tuberculo-sis, and malaria, and helped to bring life-savingdrugs to patients around the world.15 When anIndian manufacturer (Cipla) began to offer HIV/AIDs triple therapy forUS$350per patient per yearin 2001, it made international headlines becausethe patented equivalent had cost US$10,000 toUS$15,000 per patient.12 As soon as the genericfrom Cipla became available, governments beganto issue compulsory licenses for thismedication sothat they could purchase the generic.19 In May2007, after negotiations failed with the patentholder, the Brazilian government granted its firstcompulsory license for the public noncommercialuse of efavirenz, an essential HIV antiretrovial.20

By using a generic version of efavirenz, the Brazil-ian government saves approximately US$30 mil-lion per year—money that can now be used forother public health needs.19

USE OF COMPULSORY LICENSES FOR HEPATITIS C

In2013after theUSFoodandDrugAdministrationapproved sofosbuvir for hepatitis C, Gilead Sci-ences, the manufacturer of the drug (note that asof this writing, Gilead is involved in an ongoingdispute with Merck for the patent of sofosbuvir),21

was immediately criticized for its high cost as itsprice is so elevated relative to production cost.14

For example, a typical 12-week course of sofos-buvir is US$84,000 per patient, although totalproduction costs are only US$68 to US$136.22,23

To circumvent the high cost, efforts in many LMICs



were quickly taken to gain access to this medicationforpatientswithhepatitis C virus (HCV). Indiawasoneof the first countries to start production of a genericversionof sofosbuvir, anda12-weekcourse isnowonthe market for US$567.24,25

In response to the generic version, Gilead Sciencesannounced in September 2014 that it granted avoluntary license for sofosbuvir and ledipasvir (asister drug to sofosbuvir) to 11 Indian drug manu-facturers.17 The agreement allows these Indiancompanies to produce and sell generic versionsof sofosbuvir and ledipasvir to 91 LMICs25; how-ever, the agreement excluded many LMICs, coun-tries inwhich73millionpeoplewithHCV live.26 Thelicensing agreement left out 46% of people whoneed HCV treatment worldwide, including thosewho live in Brazil (2.6 million with HCV), Thailand(1.5 million with HCV), and Morocco (625,000 withHCV).26 Countries left out of this agreement need toeither negotiate a discounted rate directly with GileadSciences or pursue compulsory licenses for sofosbu-vir and ledipasvir, both of which take time and delaypatient access to these medications.

Brazil, as one of the largest countries in LatinAmerica, does have bargaining power unlike othersmall LMICs and has now negotiated with GileadSciences to purchase sofosbuvir, daclastavir, andsimeprevir. Through this agreement, a course ofsofosbuvir will cost US$9,425 for 12 weeks oftreatment.27 After this agreement, Mercosur, atrade collaborative between Argentina, Brazil,Bolivia, Uruguay, Paraguay, and Venezuela, ne-gotiated the purchase of sofosbuvir, daclastavir,and simeprevir with Gilead Sciences in November2014 at a 90% discount.27

The status of sofosbuvir in Egypt is evolving butimportant to follow, because Egypt has one of thehighest rates of hepatitis C infection in the worldafter nearly 9 million people were infected whena national mass therapy program against schisto-somiasis used contaminated needles.28,29 TheEgyptian government has rejected Gilead Sci-ences’ application for sofobusvir’s patent, whichmade it inevitable that generic versions wouldeventually be sold there. In response to this,GileadSciences offered to sell the drug for US$10 a pill orfor US$900 for a 12-week course. Subsequent tothis, the government dispensed sofosbuvir free topatients, with some restrictions to prevent a blackmarket trade of the bottles. In the first year,125,000 patients were treated with sofosbuvir.Now, generic versions of sofosbuvir are availablefor as little as US$4 a pill, and Egypt is introducingthese generics into the public health system.28

EXAMPLES OF COMPULSORY LICENSES GRANTEDFOR CANCER MEDICATIONS

Todate, compulsory licenseshavenotbeenwidelyused in LMICs to increase access to essentialmedicines for patients with cancer (Table 1); how-ever, there are two important examples in whichcompulsory licenses were used for cancer drugsin Thailand and India.

In 2008, the Thai government issued compulsorylicenses for erlotinib, letrozole, and docetaxel, andwas one of the first countries to grant a compulsorylicense for a noncommunicable disease.19 Intro-duction of a generic version of letrozole was esti-mated to save US$88 to US$102 million per year,docetaxel US$46 to US$53 million per year, and

Table 1 – Compulsory Licenses Requested and Other Patent Decisions in Low- andMiddle-Income Countries for OncologyDrugs Since 2001

Year Market Product Outcome

2006 India Imatinib Patent rejected

2007-2008 Thailand Imatinib Drug discounted, no CL issued

2007-2008 Thailand Docetaxel, letrozole, erlotinib CLs granted

2012 India Sorafenib CL granted

2012 India Sunitinib Patent revocation

2012 India Erlotinib Ruling allows generic

2013 India Trastuzumab Ruling pending clinical trial data fromgeneric

2013 India Dasatinib CL denied

2013 India Ixabepilone Application withdrawn

2014 Ecuador Sorafenib CL granted

NOTE. See IMS Consulting Group,15 Intellectual Property Watch,30 and Ministerio de Salud Publica.31

Abbreviation: CL, compulsory license.



erlotinib US$6 to US$8 million per year.32 A com-pulsory license for imatinib was initially pursued,but then canceled after negotiations with Novartisproceeded—Novartis has now made imatinibavailable to all patients who receive care in thepublic health system as part of their funded In-ternational Patient Assistance Program.6,33 TheThai government decision to issue compulsorylicenses for oncology drugs coincided with othercost-containment measures and efforts to expandpublic health coverage.

After compulsory licenses were pursued for erlo-tinib, letrozole, and doxcetaxel in Thailand, therewere clear benefits in terms of reducing drug costsand improving access to treatments for patientswith cancer.6 For letrozole, thecompulsory licensereduced the cost per pill from US$7.35 to US$0.19to US$0.22 per pill, which represents a 30-folddifference in price.6 Within 5 years of offering letro-zole, docetaxel, imatinib, and erlotinib in the publichealth system, an additional 8,916 patients receivedletrozole, 10,813 were treated with docetaxel, 1,846with imatinib, and 256 with erlotinib.6

The first compulsory license for an oncology drugin India was issued in 2012 for sorafenib. At thattime, Bayer’s sorafenib was used primarily foradvanced liver and renal cancer and improvedoutcomes only by a fewmonths; however, a year oftreatment cost US$96,000.6 By pursuing a com-pulsory license, generic manufacturing of sorafe-nib was started in India, which reduced the cost oftreatment to US$2,124 for 1 year—US$177 permonth from US$8,000 per month.6

Since 2013, the government of India has pursuedcompulsory licenses for trastuzumab, dasatinib,and ixabepilone. As a result, Roche abandoned itspatent claims for trastuzumab, and the Indianhighcourt approved a local drug company, Biocon, toproduce a biosimilar34; however, Roche subse-quently sued the Indian drug regulatory agency forapproving Biocon-Mylan’s trastuzumab as a bio-similar without carrying out clinical trials.35 Cur-rently, Biocon has entered phase III trials withtrastuzumab to demonstrate that their biosimilarversion of trastuzumab has efficacy against hu-man epidermal growth factor receptor 2 (HER2)–positive breast cancer.36 In our opinion, trastu-zumab isagoodchoice foracompulsory licenseasit has excellent efficacy against HER2-positivebreast cancer in both the metastatic and earlydisease settings. However, as the case from Indiashows, the process for obtaining a compulsorylicense and identifying a manufacturer to supportthe drug’s development can take years and delayaccess to important medications.

For dasatinib, the Indian patent office rejected therequest for a compulsory license, saying that thegovernment failed to explore the proper channelsto obtain a voluntary license from the patentholder.37,38 A request for a subsequent compul-sory license was pursued in 2015 and the DelhiHigh Court rejected the request and upheld thepatent held by Bristol-Myers Squibb.37,38 Thecompulsory license request for ixabepilone waswithdrawn as a result of toxicity concerns relatedto the drug.39

From an economic perspective, substitution ofpatented drugs with generic versions is cost sav-ing, and from a public health standpoint, not onlypermitting but also encouraging generic drug pro-duction and use increases access to essential can-cer medications in LMICs. For example, in India, ifgeneric versions of paclitaxel, docetaxel, gemcita-bine, oxaliplatin, and irinotecan—five commonlyused chemotherapeutic agents—were introduced,the potential annual savings for the health caresystem is nearly US$843 million (or V670 million).40

In fact, generic versions of these drugs are alreadyavailable and cost 8.9% to 36% less than the equiv-alent brandeddrug, and there is only aneed topermittheir use in the Indian market.40

IS THERE A ROLE FOR COMPULSORY LICENSES TOIMPROVE ACCESS TO CANCER MEDICATIONS?

Although some critics have suggested that thefailure to uphold intellectual property rights willdecrease incentives for innovation and, therefore,lead to fewer new medications in the future, evi-dence in support of this notion is scant. In fact,.80%of financial gains fromcancerdrugscomesfrom high-income countries in which compulsorylicensing is rarely used or approved.13 One obser-vational study investigated this issue and foundthat pharmaceutical companies affected by com-pulsory licensesdidnothaveadecline in the rateofnew medicines patented or their measured in-ventive and innovation activity.41

Obviously, some pharmaceutical companies viewcompulsory licenses as a threat to their intellectualproperty, research and development, and medi-cation sales. Therehavebeena fewcases inwhichthe pharmaceutical industry has tried to pressurecountries to deter them from issuing compulsorylicenses. For example, Pfizer announced that itwould rethink investments in Egypt after the coun-try issued a compulsory license for sildenafil in200242; however, this is unusual, and pharma-ceutical investment andgrowth continues inmanycountries, such as Brazil and South Africa, wherecompulsory licenses have been issued.13



Inplaceof usingcompulsory licensing, LMICsmaybenefit from negotiating directly with the patentowner, that is, pharmaceutical companies, to offeressential drugs to their populations. In many cases,pharmaceutical companies have offered affordableprices for medications, if ordered in bulk, to serve alarge population, which is an approach that we sup-port. However, there are no data on how long thesenegotiations take before a decision is made and ifpharmaceutical companies purposefully delay thisprocess so that anagreement or a compulsory licenseis not immediately granted.

There is also the argument that the price of pat-ented medications is not the main barrier to med-ication access in LMICs and, in fact, that lack ofmanufacturing capacity or poor health care systemsare larger contributors that impact access.10,15

Counter arguments to this point are simple: if LMICssave money on medication expense, then thesesavings can be invested in improvement of theirown drug manufacturing capacity and health sys-tems. In Thailand, a study found that if relevant HIV/AIDs drugs were not patented, an additional 10,000prescriptions could be made, which would increaseaccess by 50%.10

The high costs of cancer drugs threaten access tocancer treatment even in high-income countries.As a result of its extremely high cost, trastuzumabemtanzine (T-DM1), a drug used to treat meta-static HER2-positive breast cancer, has not beenmade available to patients treated in the nationalhealth system in the United Kingdom, accordingto a recent recommendation by the National In-stitute for Health and Care Excellence. The Na-tional Institute for Health and Care Excellenceestimates that only 1,500 women in the UnitedKingdom would benefit from treatment with T-DM1every year and that a year of treatment costs£102,405, roughly 3.9 times the 2014 per capitaincome of £26,350 in the United Kingdom.43,44

Comparedwith lapatinib plus capecitabine therapyin this setting, T-DM1 costs £166,400 per quality-adjusted life year (QALY) gained,45 which is signif-icantly higher than the cost-effectiveness thresholdin the United Kingdom of £30,000/QALY gained.46

In contrast to lapatinib and capecitabine, T-DM1therapy has a more favorable adverse effect profileand is generally well tolerated, an important con-sideration in patients with advanced cancer wherepreserving quality of life is a major goal; this fact isnot accounted for in the cost and QALY calculation.

Out of concern of the access barrier to T-DM1therapy, the Coalition for Affordable T-DM1, a civilorganization, sent a formal letter to United King-dom secretary of state for health to ask that the

government use provisions in United Kingdompatent laws to authorize the manufacture or im-portation of generic versions of T-DM1without thepermission of Roche.43,44 This case simply exem-plifies the exorbitantly high price of cancer med-ications and the urgent need to find solutions tothis problem, especially in resource-conscious orresource-constrained settings.

BARRIERS TO COMPULSORY LICENSES

Pursuingacompulsorylicensedoesseemtobeapos-sible solution to improve access to medications inLMICs.Nonetheless, therearesignificantbarriers tothis route for many LMICs and, consequently, thisroute has not been frequently pursued for cancerdrugs since the introduction of the TRIPS Agree-ment and Doha Declaration in 2001 for essentialcancer medications. A 2005 report by the WHOfound that many LMICs did not implementmany TRIPS flexibilities into their legislation—compulsory licensing, parallel importation, limitson data protection, use of broad research, andother exceptions to patentability—and this wasprimarily attributed to a lack of legal and techni-cal expertise needed to draft such legistlation.47

Furthermore, even when compulsory licenses areapproved, therecanstill bedelays in the introductionof these generic medications into the market. Forexample, in Thailand, introduction of generic med-ications was delayed for 1 year from the timecompulsory licenses were approved.32 The reasonfor this was that some patent-holding companiesalleged that importation and production of genericdrugs under Thailand’s government-use licens-ingpolicy, a formof compulsory licensing,wasnotprotected by TRIPS/DOHA. As a result, this causedconfusionandreluctanceamonggenericproducersand, consequently, a delay in importing the medi-cations.32 In recognition of these issues, efforts areneeded to resolve these barriers in obtaining com-pulsory licenses for essential medications.

ALTERNATIVE STRATEGIES TO IMPROVE ACCESSTO CANCER MEDICATIONS

New strategies are being considered to ensurethat cheaper medicines flow to countries most inneed. These include tiered price schemes, public–private partnerships, patent pools, and taxincentives.48,49

Tiered pricing, also known as price discriminationanddifferential or equity pricing, consists of charg-ing different prices for the same product or servicein different markets or segments of a market. Themethodusually isbasedonconsumerability topayand not necessarily on the market demand. For



Table 2 – 2015 WHO Essential Medicines List for Cancer Medications

Essential Medicine On-Patent in United States? (patent holder) On-Patent in Europe? (patent holder)

Antineoplastic and adjuvant

All-trans-retinoid acid No No

Allopurinol No No

Asparaginase No No

Bendamustine Yes (Teva Pharmaceuticals) Yes (Mundi Pharma)

Bleomycin No No

Calcium folinate No No

Capecitabine No No

Carboplatin No No

Chlorambucil No No

Cisplatin No No

Cyclophosphamide No No

Cytarabine No No

Dacarbazine No No

Dactinomycin No No

Daunorubicin No No

Docetaxel No No

Doxorubicin No No

Etoposide No No

Fludarabine No No

Fluorouracil No No

Filgrastim No No

Gemcitabine No No

Hydroxycarbamide No No

Ifosfamide No No

Imatinib No Yes (Novartis)

Irinotecan No No

Mercaptopurine No No

Mesna No No

Methotrexate No No

Procarbazine No No

Rituximab Yes (Roche) No

Tiguanine No No

Trastuzumab Yes (Roche) No

Vinblastine No No

Vincristine No No

Vinorelbine No No

Hormones and antihormones

Anastrozole No No

Bicalutamide No No

Dexamethasone No

Leuprorelin Yes, suspension for subcutaneous injection(Tolmar Pharmaceuticals)

Yes, suspension for subcutaneous injection(Tolmar Pharmaceuticals)




example, medications will be more expensive inhigh-income countries and more affordable inLMICs.

In the last two decades, tiered pricing has beenfrequently used to lower the cost of vaccines andAIDSmedications in LMICs. Until recently, thoughthe situation has started to change, tiered pricinghad not been applied frequently to oncology med-ications. For example, in 2011, there was littlevariation in price between widely used cancermedications—oxaliplatin, bevacizumab, cetu-zimab, trastuzumab, sorafenib, erlotinib, andgefitinib—in Southeast Asia despite gross na-tional income per capita varying between coun-tries by three-fold or more.5 A few pharmaceuticalcompanies have introduced tiered pricing fortheir cancer products in the last few years, in-cluding GlaxoSmithKline, Eli Lilly, Sanofi, andRoche.5

One effort underway to promote tiered pricing hasbeen organized by the Global Fund and GAVIAlliance. This effort is known as the EquitableAccess Initiative and aims to improve global ac-cess to essential medicines.50 Stratifying pricesinto tiers can be done in several ways. The group’sfirst proposal recommended prices be set by aninternational public health body, such as theWHO; however, a more recent proposal considersvoluntary discounts by pharmaceutical compa-nies, and this could lower prices even further.50

Despite these efforts, it is unclear whether strati-fying prices would significantly reduce the cost ofcancer drugs. A study from the President’s Emer-gency Plan for AIDSRelief in 2005 concluded thattiered pricing had failed to lower the price of on-patent HIV/AIDS medications, and that genericformulas were primarily responsible for reducingthe price of AIDS drugs and improving access.50

From 2004 to 2008, the average price of HIVantiretrovirals decreasedby48%,and it isdebatedwhether this is primarily a result of tiered pricing,generic manufacturing, targeted negotiations, ora combination of the three.7 On one hand, be-causeof thesecontroversies, theEquitableAccess

Initiative and other organizations, such asMedecins Sans Frontieres, do not endorse tieredpricing as they worry that this may result inmiddle-income countries paying higher prices formedicines.25,50,51 On the other hand, some believetiered pricing could work well in small-volumemarkets or in regions with uncertain productioncapacities on which an occasional short-term so-lution is needed to temporarily secure access to amedication.51

Whatever solution is pursued, there is growingrecognition that the price of oncology medicationsis a problem. A recent joint symposium that in-volved the WHO, WTO, and the World IntellectualProperty Organization agreed that prices for es-sential medicines continue to be an issue in LMICs,and concerns over how to increase access to med-icines were discussed.30 In 2015, theWHO updatedtheir list of essential medicines and included 16 newoncology drugs, several of which are under patentand are priced high (Table 2).52-54

In conclusion, price negotiationswith pharmaceu-tical companies and use of compulsory licenses,albeit controversial, has led to an important ad-vance in access to drugs for communicable dis-eases, such as HIV and hepatitis C and, morerecently, for cancer, in LMICs. Even the threat ofcompulsory licenses can work as a bargainingchip when discussing drug prices with indus-try. Indeed, the pharmaceutical industry hasresponded with a series of price-discriminationand market-access strategies to increase salesand access to medications in LMICs, as seen inHIV treatment access in the last decade. In thiscontext, it is imperative that we review existingapproaches to drug pricing and identify solutionsto improve access to cancer medications inemerging markets and LMICs. We will have themost success solving this problem through collab-oration between the pharmaceutical industry,governments, private funds, civil organizations,andhealthprofessionals; only thenwill webeableto best control the growing burden of cancerworldwide.55

Table 2 – 2015 WHO Essential Medicines List for Cancer Medications (Continued)

Essential Medicine On-Patent in United States? (patent holder) On-Patent in Europe? (patent holder)

Hydrocortisone No No

Methylprednisolone No No

Prednisolone No No

Tamoxifen No No

NOTE. See WHO,52 Union for International Cancer Control,53 and European Patent Office.54



DOI: 10.1200/JGO.2016.005363Published online on jgo.ascopubs.org on June 29, 2016.


Conception and design: Cinthia Leite Frizzera Borges Bognar,Gilberto de Lima Lopes Jr

Collection and assembly of data: All authorsData analysis and interpretation: All authorsManuscript writing: All authorsFinal approval of manuscript: All authorsAccountable for all aspects of the work: All authors


The following represents disclosure information provided byauthors of this manuscript. All relationships are consideredcompensated. Relationships are self-held unless noted.I = Immediate Family Member, Inst = My Institution. Rela-tionshipsmaynot relate tothesubjectmatterofthismanuscript.For more information about ASCO’s conflict of interest policy,please refer to www.asco.org/rwc or jco.ascopubs.org/site/ifc.

Cinthia Leite Frizzera Borges BognarNo relationship to disclose

Brittany L. BychkovskyNo relationship to disclose

Gilberto de Lima Lopes JrHonoraria: AstraZeneca, Genentech, Merck Serono, MerckSharp & Dohme, Fresenius Kabi, Novartis, Bristol-MyersSquibb, Janssen-CilagConsulting or Advisory Role: Pfizer, Bristol-Myers Squibb, EliLilly, ImCloneResearch Funding: Eli Lilly, ImClone, Pfizer, AstraZeneca,Merck Sharp & Dohme, EisaiExpert Testimony: Sanofi

ACKNOWLEDGMENTB.L.B. thanks theDana-Farber Cancer Institute for supportingher research work.

REFERENCES1. International Agency for Research on Cancer: GLOBOCAN 2012: Estimated cancer incidence, mortality and prev-

alence worldwide in 2012. http://globocan.iarc.fr

2. Goss PE, Lee BL, Badovinac-Crnjevic T, et al: Planning cancer control in Latin America and the Caribbean. LancetOncol 14:391-436, 2013

3. Goss PE, Strasser-Weippl K, Lee-Bychkovsky BL, et al: Challenges to effective cancer control in China, India, andRussia. Lancet Oncol 15:489-538, 2014

4. Stewart B, Wild CP (eds): World Cancer Report. Lyon, France, IARC, 2014

5. Lopes G de L Jr, de Souza JA, Barrios C: Access to cancer medications in low- and middle-income countries. Nat RevClin Oncol 10:314-322, 2013

6. ’t Hoen E: Access to cancer treatment: A study of medicine pricing issues with recommendations for improving accessto cancer medication. A report prepared for OXFAM. http://apps.who.int/medicinedocs/documents/s21758en/s21758en.pdf

7. CanTreat International: Access to cancer treatment in low- and middle-income countries. An essential part of globalcancer control. A CanTreat position paper. http://axios-group.com/assets/Uploads/CanTreat-UICC-IssuePaper-2010.pdf

8. de Souza JA, Hunt B, Asirwa FC, et al: Global health equity: Cancer care outcomedisparities in high-,middle-, and low-income countries. J Clin Oncol 34:6-13, 2016

9. IMS Institute for Healthcare Informatics: Innovation in cancer care and implications for health systems: Global on-cology trend report. http://obroncology.com/imshealth/content/IMSH_Oncology_Trend_Report_020514F4_screen.pdf

10. Yamabhai I, Smith RD: A review of the health and economic implications of patent protection, with a specific focus onThailand. Health Res Policy Syst 10:24, 2012

11. World Health Organization: WTO and the TRIPS agreement. http://www.who.int/medicines/areas/policy/wto_trips/en/

12. Ooms G, Forman L, Williams OD, et al: Could international compulsory licensing reconcile tiered pricing of phar-maceuticals with the right to health? BMC Int Health Hum Rights 14:37, 2014

13. Lopes G: Compulsory licensing: A double-edged sword in the fight for access to cancer medications in low- andmiddle-income countries. https://connection.asco.org/blogs/compulsory-licensing-double-edged-sword-fight-access-cancer-medications-low-and-middle-income

14. Beall R, Kuhn R: Trends in compulsory licensing of pharmaceuticals since the Doha Declaration: A database analysis.PLoS Med 9:e1001154, 2012

15. IMS Consulting Group: Securing IP and access to medicine: Is oncology the next HIV? An IMS Consulting GroupWhitePaper on compulsory licensing and other risks to intellectual property rights in oncology. https://www.imshealth.com/files/web/Global/Services/Services%20TL/IMSCG_Compulsory_Licensing.pdf

16. World Trade Organization: WTO members agree to extend drug patent exemption for poorest members. https://www.wto.org/english/news_e/news15_e/trip_06nov15_e.htm






http://globocan.iarc.fr

http://apps.who.int/medicinedocs/documents/s21758en/s21758en.pdf

http://apps.who.int/medicinedocs/documents/s21758en/s21758en.pdf

http://axios-group.com/assets/Uploads/CanTreat-UICC-IssuePaper-2010.pdf

http://axios-group.com/assets/Uploads/CanTreat-UICC-IssuePaper-2010.pdf

http://obroncology.com/imshealth/content/IMSH_Oncology_Trend_Report_020514F4_screen.pdf

http://www.who.int/medicines/areas/policy/wto_trips/en/

https://connection.asco.org/blogs/compulsory-licensing-double-edged-sword-fight-access-cancer-medications-low-and-middle-income

https://connection.asco.org/blogs/compulsory-licensing-double-edged-sword-fight-access-cancer-medications-low-and-middle-income

https://www.imshealth.com/files/web/Global/Services/Services%20TL/IMSCG_Compulsory_Licensing.pdf

https://www.imshealth.com/files/web/Global/Services/Services%20TL/IMSCG_Compulsory_Licensing.pdf

https://www.wto.org/english/news_e/news15_e/trip_06nov15_e.htm

https://www.wto.org/english/news_e/news15_e/trip_06nov15_e.htm


17. World Trade Organization: Understanding the WTO: The organization. Least-developed countries. https://www.wto.org/english/thewto_e/whatis_e/tif_e/org7_e.htm

18. Barigaba J: Africa: Uganda leads push for permanent waiver on drug. The East African, November 9, 2015. http://allafrica.com/stories/201511091872.html

19. Stirner B, Thangaraj H: Learning from practice: Compulsory licensing cases and access to medicines. Pharm Pat Anal2:195-213, 2013

20. Luo J, Oliveira MA, Ramos MBC, et al: Antiretroviral drug expenditure, pricing and judicial demand: An analysis offederal procurement data in Brazil from 2004-2011. BMC Public Health 14:367, 2014

21. Reuters: Merck wins hepatitis C drug patent claim against Gilead. The New York Times, March 22, 2016. http://www.nytimes.com/2016/03/23/business/merck-wins-hepatitis-c-drug-patent-claim-against-gilead.html

22. Harris G: Maker of costly hepatitis C drug sovaldi strikes deal on generics for poor countries. The New York Times,September 15, 2014. http://www.nytimes.com/2014/09/16/business/international/maker-of-hepatitis-c-drug-strikes-deal-on-generics-for-poor-countries.html

23. Hill A, Khoo S, Fortunak J, et al: Minimum costs for producing hepatitis C direct-acting antivirals for use in large-scaletreatment access programs in developing countries. Clin Infect Dis 58:928-936, 2014

24. Wei L, Lok ASF: Impact of new hepatitis C treatments in different regions of the world. Gastroenterology 146:1145-1150.e1-4

25. Medecins Sans Frontieres: Strategies to secure access to generic hepatitis C medicines. http://www.msfaccess.org/sites/default/files/MSF_assets/HepC/Docs/HepC_brief_OvercomingbarriersToAccess_ENG_2015.pdf

26. HepCoalition: Home. http://www.hepcoalition.org

27. Brazil Ministry of Health: Compra conjunta de medicamentos gera economia de ate 83%. http://www.brasil.gov.br/saude/2015/11/compra-conjunta-de-medicamentos-gera-economia-de-ate-83

28. McNeil DG Jr: Curing hepatitis C, in an experiment the size of Egypt. The New York Times, December 15, 2015. http://www.nytimes.com/2015/12/16/health/hepatitis-c-treatment-egypt.html

29. Struthers A: Fromschistosomiasis to hepatitis C: The spread ofHCV in Egypt. http://mjota.org/images/SpreadofHCVEgypt.pdf

30. Intellectual Property Watch: Home. http://www.ip-watch.org

31. Ministerio de Salud Publica: Ecuador concedio nueve licencias obligatorias para medicamentos estrategicos. http://www.salud.gob.ec/ecuador-concedio-nueve-licencias-obligatorias-para-medicamentos-estrategicos/

32. Mohara A, Yamabhai I, Chaisiri K, et al: Impact of the introduction of government use licenses on the drug expenditureon seven medicines in Thailand. Value Health 15:S95-S99, 2012 (suppl 1)

33. Novartis Oncology: Patient assistance. https://www.novartisoncology.com/patients-and-caregivers/patient-assistance

34. Machado K: Generic herceptin approved in India: India’s Biocon to market generic version of breast-cancer drug. The WallStreet Journal, November 27, 2013. http://www.wsj.com/articles/SB10001424052702304281004579221971613720720

35. Generics and Biosimilars Initiative: Biocon and Mylan challenge Indian ban on trastuzumab ‘similar biologics.’ http://www.gabionline.net/Biosimilars/News/Biocon-and-Mylan-challenge-Indian-ban-on-trastuzumab-similar-biologics

36. Pilla V: Two biocon biosimilar drugs enter global clinical trials. Live Mint, May 8, 2015. http://www.livemint.com/Companies/n8dgr8Fbg5iVA3d8PmuYaP/Two-Biocon-biosimilar-drugs-enter-global-clinical-trials.html

37. Jayaraj S: DHC upholds validity of dasatinib patent. Spicyip, July 16, 2015. http://spicyip.com/2015/07/dhc-upholds-validity-of-dasatinib-patent.html

38. LaneEJ: Indian court upholds BMSSprycel patent, but list of challengers grows. FiercePharma http://www.fiercepharma.com/regulatory/indian-court-upholds-bms-sprycel-patent-but-list-of-challengers-grows

39. Mukherjee R: Cancer drug may lose patent shield. The Times of India, February 7, 2014. http://timesofindia.indiatimes.com/business/india-business/Cancer-drug-may-lose-patent-shield/articleshow/29965162.cms

40. Lopes G de L: Cost comparison and economic implications of commonly used originator and generic chemotherapydrugs in India. Ann Oncol 24:v13-v16, 2013 (suppl 5)

41. ColleenC: Cheap drugs at what price to innovation: Does the compulsory licensing of pharmaceuticals hurt innovation?Berkeley Technol Law J 18:853-908, 2003

42. Allam A: Seeking investment, Egypt tries patent laws. TheNew York Times, October 4, 2002. http://www.nytimes.com/2002/10/04/business/seeking-investment-egypt-tries-patent-laws.html

43. Union for Affordable Cancer Treatment: Home. http://cancerunion.org/tdm1_uk.html

44. Knowledge Ecology International: Home. http://keionline.org/node/2328

45. National Institute for Health and Care Excellence: Trastuzumab emtansine for treating HER2-positive, unresectablelocally advanced or metastatic breast cancer after treatment with trastuzumab and a taxane. https://www.nice.org.uk/guidance/ta371


https://www.wto.org/english/thewto_e/whatis_e/tif_e/org7_e.htm

https://www.wto.org/english/thewto_e/whatis_e/tif_e/org7_e.htm

http://allafrica.com/stories/201511091872.html

http://allafrica.com/stories/201511091872.html

http://www.nytimes.com/2016/03/23/business/merck-wins-hepatitis-c-drug-patent-claim-against-gilead.html

http://www.nytimes.com/2016/03/23/business/merck-wins-hepatitis-c-drug-patent-claim-against-gilead.html

http://www.nytimes.com/2014/09/16/business/international/maker-of-hepatitis-c-drug-strikes-deal-on-generics-for-poor-countries.html

http://www.nytimes.com/2014/09/16/business/international/maker-of-hepatitis-c-drug-strikes-deal-on-generics-for-poor-countries.html

http://www.msfaccess.org/sites/default/files/MSF_assets/HepC/Docs/HepC_brief_OvercomingbarriersToAccess_ENG_2015.pdf

http://www.msfaccess.org/sites/default/files/MSF_assets/HepC/Docs/HepC_brief_OvercomingbarriersToAccess_ENG_2015.pdf

http://www.hepcoalition.org

http://www.brasil.gov.br/saude/2015/11/compra-conjunta-de-medicamentos-gera-economia-de-ate-83

http://www.brasil.gov.br/saude/2015/11/compra-conjunta-de-medicamentos-gera-economia-de-ate-83

http://www.nytimes.com/2015/12/16/health/hepatitis-c-treatment-egypt.html

http://www.nytimes.com/2015/12/16/health/hepatitis-c-treatment-egypt.html

http://mjota.org/images/SpreadofHCVEgypt.pdf

http://mjota.org/images/SpreadofHCVEgypt.pdf

http://www.ip-watch.org

http://www.salud.gob.ec/ecuador-concedio-nueve-licencias-obligatorias-para-medicamentos-estrategicos/

http://www.salud.gob.ec/ecuador-concedio-nueve-licencias-obligatorias-para-medicamentos-estrategicos/

https://www.novartisoncology.com/patients-and-caregivers/patient-assistance

http://www.wsj.com/articles/SB10001424052702304281004579221971613720720

http://www.gabionline.net/Biosimilars/News/Biocon-and-Mylan-challenge-Indian-ban-on-trastuzumab-similar-biologics

http://www.gabionline.net/Biosimilars/News/Biocon-and-Mylan-challenge-Indian-ban-on-trastuzumab-similar-biologics

http://www.livemint.com/Companies/n8dgr8Fbg5iVA3d8PmuYaP/Two-Biocon-biosimilar-drugs-enter-global-clinical-trials.html

http://www.livemint.com/Companies/n8dgr8Fbg5iVA3d8PmuYaP/Two-Biocon-biosimilar-drugs-enter-global-clinical-trials.html

http://spicyip.com/2015/07/dhc-upholds-validity-of-dasatinib-patent.html

http://spicyip.com/2015/07/dhc-upholds-validity-of-dasatinib-patent.html

http://www.fiercepharma.com/regulatory/indian-court-upholds-bms-sprycel-patent-but-list-of-challengers-grows

http://www.fiercepharma.com/regulatory/indian-court-upholds-bms-sprycel-patent-but-list-of-challengers-grows

http://timesofindia.indiatimes.com/business/india-business/Cancer-drug-may-lose-patent-shield/articleshow/29965162.cms

http://timesofindia.indiatimes.com/business/india-business/Cancer-drug-may-lose-patent-shield/articleshow/29965162.cms

http://www.nytimes.com/2002/10/04/business/seeking-investment-egypt-tries-patent-laws.html

http://www.nytimes.com/2002/10/04/business/seeking-investment-egypt-tries-patent-laws.html

http://cancerunion.org/tdm1_uk.html

http://keionline.org/node/2328

https://www.nice.org.uk/guidance/ta371

https://www.nice.org.uk/guidance/ta371


46. Towse A, Pritchard C, Devlin N: Cost-Effectiveness Thresholds: Economic and Ethical Issues. London, UnitedKingdom, Office of Health Economics, The King’s Fund, 2002

47. Musungu SF, Oh C: The use of flexibilities in TRIPS by developing countries: can they promote access to medicines?http://www.who.int/intellectualproperty/studies/TRIPSFLEXI.pdf

48. Nicol D, Owoeye O: Using TRIPS flexibilities to facilitate access to medicines. Bull World Health Organ 91:533-539,2013

49. World Health Organization: Global action plan for the prevention and control of noncommunicable diseases: 2013-2020. http://apps.who.int/iris/bitstream/10665/94384/1/9789241506236_eng.pdf?ua=1

50. Holmes CB, Coggin W, Jamieson D, et al: Use of generic antiretroviral agents and cost savings in PEPFAR treatmentprograms. JAMA 304:313-320, 2010

51. Moon S, Jambert E, Childs M, et al: A win-win solution?: A critical analysis of tiered pricing to improve access tomedicines in developing countries. Global Health 7:39, 2011

52. World Health Organization: WHO moves to improve access to lifesaving medicines for hepatitis C, drug-resistant TBand cancers. http://www.who.int/mediacentre/news/releases/2015/new-essential-medicines-list/en/

53. Shulman LN, Wagner CM, Barr R, et al: Proposing essential medicines to treat cancer: Methodologies, processes, andoutcomes. http://www.uicc.org/sites/main/files/atoms/files/JCO-2015-Shulman-JCO.2015.61.8736.pdf

54. European Patent Office: Espacenet patent search. www.epo.org/searching/free/espacenet.html

55. Global Task Force on Expanded Access to Cancer Care and Control: An initiative directed by HGEI, HMS, HSPH,FHCRC, UW andUWMed. http://gtfccc.harvard.edu/icb/icb.do?keyword=k69586&tabgroupid=icb.tabgroup138264


http://www.who.int/intellectualproperty/studies/TRIPSFLEXI.pdf

http://apps.who.int/iris/bitstream/10665/94384/1/9789241506236_eng.pdf?ua=1

http://www.who.int/mediacentre/news/releases/2015/new-essential-medicines-list/en/

http://www.uicc.org/sites/main/files/atoms/files/JCO-2015-Shulman-JCO.2015.61.8736.pdf

http://www.epo.org/searching/free/espacenet.html

http://gtfccc.harvard.edu/icb/icb.do?keyword=k69586&tabgroupid=icb.tabgroup138264


specialarticle

Breast Cancer and African Ancestry:Lessons Learned at the 10-YearAnniversary of the Ghana-MichiganResearch Partnership and InternationalBreast RegistrySee accompanying editorial on page 253

executivesum

mary

Women with African ancestry in western, sub-Saharan Africa and in the United States represent a pop-ulation subset facing an increased risk of being diagnosed with biologically aggressive phenotypes ofbreast cancer that are negative for the estrogen receptor, the progesterone receptor, and the HER2/neumarker. These tumors are commonly referred to as triple-negative breast cancer. Disparities in breastcancer incidence and outcome related to racial or ethnic identity motivated the establishment of theInternational Breast Registry, on the basis of partnerships between the Komfo Anokye Teaching Hospital inKumasi, Ghana, the University of Michigan Comprehensive Cancer Center in Ann Arbor, Michigan, and theHenry Ford Health System in Detroit, Michigan. This research collaborative has featured educationaltraining programs aswell as scientific investigations related to the comparative biology of breast cancer inGhanaian African, African American, and white/European American patients. Currently, the InternationalBreast Registry has expanded to include African American patients throughout the United States bypartneringwith theSistersNetwork (anationalAfricanAmericanbreast cancer survivors’organization) andadditional sites in Ghana (representing West Africa) as well as Ethiopia (representing East Africa). Itsactivities are now coordinated through the Henry Ford Health System International Center for the Studyof Breast Cancer Subtypes. Herein, we review the history and results of this international program at its10-year anniversary.

INTRODUCTION

Financial constraints are a constant reality in low-andmiddle-income countries and pose enormousbarriers to both quantifying and addressing thecancer burden in sub-Saharan Africa. The limiteddata available have been generated by the Globo-can 2008 database of the International Agency forResearch on Cancer and from recent attempts toreport population-based cancer incidence ratesfrom selected countries such as Uganda andGhana.1-6 Taken together, these resources sug-gest that breast cancer is an increasing problem.Likely explanations for breast cancer as an in-creasing health threat include more prolongedlongevity in many African communities (becausebreast cancer risk increases as women age); ac-ceptance of Westernized, higher fat dietary pat-terns (whichcan increasebreast cancer risk inbothpre- and postmenopausal women); and increasedadoption of reproductive patterns that are more

prevalent inWesternpopulations (suchasdelayedchildbearing and reduced overall parity, which arerisk factors that increase incidence of hormonereceptor–positive breast cancer). Increasing ratesof breast cancer cases are particularly alarmingwithin most sub-Saharan African countries, wherethe already overburdened health care system isunequipped to afford early detection and multi-disciplinary treatment programs. The disturbingand excessively high mortality-to-incidence ra-tios of breast cancer in sub-Saharan Africa com-pared with other parts of the world are depictedin Figure 1.

IntheUnitedStates, race-and/orethnicity-associateddisparities in breast cancer incidence and out-come have been documented for many decades.Breast cancer mortality rates are disproportion-ately higher for African American compared withwhite/European American women, and AfricanAmerican women tend to be diagnosedwith breast

Evelyn Jiagge

Joseph Kwaku Oppong

Jessica Bensenhaver

Francis Aitpillah

Kofi Gyan

Ishmael Kyei

Ernest Osei-Bonsu

Ernest Adjei

Michael Ohene-Yeboah

Kathy Toy

Karen Eubanks Jackson

Marian Akpaloo

Dorcas Acheampong

Beatrice Antwi

FaustinaObengAgyeman

Zainab Alhassan

Linda Ahenkorah Fondjo

Osei Owusu-Afriyie


B.A. and L.N. share seniorauthorship for this article.Authors’ disclosures ofpotential conflicts ofinterest and contributionsare found at the end of thisarticle.Corresponding author: LisaNewman, MD, MPH,Henry Ford Health SystemInternational Center forthe Study of Breast CancerSubtypes, 2799 WestGrand Blvd, Detroit,MI 48202; e-mail:[email protected].


© 2016 by the American Society of Clinical Oncology Licensed under the Creative Commons Attribution 4.0 License



cancer at younger ages. Furthermore, although thepopulation-based breast cancer incidence ratesare increasing among African American women,the survival disparity is widening, with current datarevealing a 42% higher mortality rate among Af-rican Americans.11 Themortality differences havehistorically been ascribed to socioeconomic dif-ferences, because poverty rates and barriers toaccessing adequate health care are more preva-lent among African Americans. However, theyounger age distribution for breast cancer, studiesdocumenting higher mortality risk after adjustingfor socioeconomic status, and the two-fold higherpopulation-based incidence rates of the biologi-cally aggressive triple-negative breast cancer(TNBC; tumors that are negative for estrogen re-ceptor, progesterone receptor, and HER2/neu)phenotype within the African American commu-nity have fueled speculation that African ancestryitself might be associated with hereditary suscep-tibility for specific patterns of breast cancer.12-14

These issues provided the rationale for estab-lishing the International Breast Registry (IBR),involving a breast cancer research partnershipbetween the Komfo Anokye Teaching Hospital(KATH), the University of Michigan (UM), andthe Henry Ford Health System (HFHS). Most re-cently, this program has evolved into the HFHSInternational Center for the Study of Breast CancerSubtypes. Exploratory conversations and biconti-nental introductory visits occurred in 2004 to2005, and the institutional review boards of each

institution provided their initial human researchethics approvals in 2006. The early goals of thiscollaborative were therefore related to studyingthe biology of breast cancer in womenwith Africanancestry, and indeed, the first joint publicationfrom this team was an article providing how-toguidelines for other investigators regarding theconduct of cancer research in developing, low-and middle-income countries.15

The IBR has grown enormously since its inception.Although this collaborative continues to feature arobust breast tumor repository that has providedprovocative, hypothesis-generating data regardingbreast cancer in Ghanaian African aswell as AfricanAmericanwomen, ithasalsoexpandeditsportfolioofeducational and training exchangeprograms. In thisreview, we summarize the various outcomes of thisinternational effort over the past 10 years, aspresented in peer-reviewed publications and aca-demicmeeting abstracts. We also review the non–research-related productivity of this partnership,featuring investment in the oncology services infra-structure of Ghana.

BREAST CANCER AND AFRICAN ANCESTRY:PATTERNS IDENTIFIED THROUGH THE IBR

The concept of subtyping breast cancer has as-sumed increasing importance as our knowledgeof targeted therapy has advanced. Invasive breastcancers that are positive for the estrogen receptorand/or theprogesterone receptor canbemanagedsystemically with a variety of endocrine agents,

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0

10

20

30

40

50

60

70

80

90

100

Rate

(per

100

,000

)

Incidence Rate Mortality Rate Ratio

Mor

talit

y-to

-Inci

denc

e Ra

tio

More

Dev

eloped

Reg

ions

Less

Dev

eloped

Reg

ions

Wes

tern

Euro

pe

Souther

n Euro

pe

Norther

n Euro

pe

North A

mer

ica

Norther

n Afri

ca

Easte

rn A

frica

Mid

dle Afri

ca

Wes

tern

Afri

ca

Wes

tern

Asia

Easte

rn A

sia

Southea

st Asia

South A

mer

ica

Souther

n Afri

ca

Robert Newman Brewer

Amma Gyamfuah

Barbara Salem

Timothy Johnson

Max Wicha

Sofia Merajver

Celina Kleer

Judy Pang

Emmanuel Amankwaa-Frempong

Azadeh Stark

Francis Abantanga

Lisa Newman

Baffour Awuah

Fig 1 –

International variation inbreast cancer incidence,mortality, and mortality-to-incidence ratios. Data fromJemalet al,7Newmanetal,9

and Jiagge et al.10



such as tamoxifen for premenopausal patientsand tamoxifen or one of the aromatase inhib-itors for postmenopausal patients. Tumors thatoverexpress HER2/neu benefit greatly from tar-geted, anti-HER2 therapy such as trastuzumaband/or pertuzumab. Equally important and relevantto treatment planning is the fact that these targetedagents are contraindicated in patients whose tu-morsarenegative for thesemarkers, andusing themin patients with TNBC results in exposing patientsunnecessarily to the toxicity of an ineffective regi-men. Studies of breast cancer in the United Stateshave revealed that frequency and population-basedincidence rates of TNBC are significantly higher inAfrican American women compared with women ofother racial or ethnic identities.16,17 The IBR hasinvestigated the hypothesis of inherited TNBC sus-ceptibility associated with African ancestry by com-paringbreast tumorphenotypes inAfricanAmericanandwestern, sub-SaharanAfricanwomen, becausethese two population subsets have shared ancestryresulting from the colonial-era slave trade.14

Therefore, one of the initial publications of theKATH-UM IBR collaborative focused on compari-sons of TNBC prevalence in white Americanwomen comparedwith African American andGha-naian women. The HFHS, which provides care tothe robustly diverse metropolitan Detroit, MI, com-munity, served as the source for the comparisonpatient population. This study revealed that TNBCaccounted for the majority of KATH patients withbreast cancer.18 Subsequent studies from the IBRtissue repository, onthebasisof largersamplesizes,have confirmed these results. The most recentanalyses, on the basis of immunohistochemistrystudies of 234 Ghanaian patients with breast can-cer (withcompletemarkerprofilingon173 invasivetumors), revealed that 92 (53.2%) of 173 tumorswere triple negative.19 In contrast, the lowest fre-quency of TNBC was seen in white Americanpatients (15.5%), and TNBC frequency for AfricanAmericanpatientswas intermediate between thesetwo extremes at 30%. This pattern persisted insubset analyses of patients younger than age50 years. These findings are consistent with thetheory that extent of African ancestry correlateswith likelihood of being diagnosed with TNBC, be-cause African American patients represent a genet-ically admixedpopulation. Table 1 lists the results ofstudies characterizing the breast cancer burden ofGhanaian women on the basis of data from theIBR research collaborative.

Although studies from the IBR have been enlight-ening with regard to understanding the breastcancer burden of western sub-Saharan Africa as

exemplifiedbyGhanaianwomen, expansionof theregistry has provided opportunities to study breastcancer in East Africa as well. Contributions frompartners at the St Paul’s Millennium Hospital inAddis Ababa, Ethiopia, have yielded interestingpreliminary findings regarding breast tumor phe-notypes from Ethiopian women. Immunohisto-chemistry performed at UM on 95 invasive breastcancers from Addis Ababa revealed a low fre-quency of TNBC (9%).24 The history of populationmigrations through the African diaspora andshared ancestry may explain these disparate re-sults regarding the prevalence of TNBC in AfricanAmerican and Ghanaian women compared withEthiopians. The colonial-era slave trade that wascontrolled by Europeans for the purpose ofgenerating a labor supply in North America fo-cused on capture and enforced trans-Atlantictransport of West Africans.25 In contrast, enslave-ment of East Africans from the region of Ethiopiaoften involved Arabic slave traders and enforcedmigration to northern Africa and Asia.26,27 There-fore, African American women are more likely tohave shared ancestry with women from westernAfrica, and this has been confirmed by genotypingstudies of markers associated with geographicancestry, commonly referred to as ancestry in-formative markers.28-30

Studies of breast tumor biology have generated thestem-cell theory, which hypothesizes that mammarytumor virulence and metastatic risk are driven by asmall subset of cells within the cancer known as thestem cells. Thus, efforts to identify and characterizethe mammary stem cells represent an exciting bodyof research. UM researchers have pioneered studiesof the mammary stem-cell hypothesis and havereported on aldehyde dehydrogenase-1 (ALDH1)expression as a reliable, immunohistochemically de-tectable marker of themammary stem cell and havealso reported its association with more virulenttumors.31StudiesofKATHtumorsatUMdemonstratedelevated expression of this marker in both benign andmalignant breast tissue from Ghanaian women.20

The polycomb group protein EZH2 is another mol-ecule that has been implicated in mammary stemcells and TNBC progression, as demonstrated byUM researchers involved with the IBR.32,33 There-fore, we incorporated studies of EZH2 into the IBR.This marker was found to have unique patterns ofexpression in Ghanaian breast specimens studiedthrough the KATH-UM partnership.22

Lastly, TNBC subtyping is a promising avenue forgaining insights regarding more refined, person-alized treatment of TNBC.34-37 Existing research



suggests that there are at least six different TNBCsubtypes, and some of the distinguishing featuresare related to luminal-like characteristics seen inthe androgen receptor subtype versus the stemcell–like characteristics seen in the mesenchymal sub-types. Despite the disproportionately high frequencyof TNBC among women with African ancestry, theTNBC subtyping research has been based almostexclusively on data sets representing white Ameri-can, European, and Asian populations. Thus far, nodata sets of gene expression profiles representingAfrican patient populations have been available forinclusion in this body of research. These issuesmotivated exploratory analyses of tissue from theIBR looking at expression of both androgen receptorandALDH1as immunohistochemistrysurrogates forTNBC subtypes. Among the Ghanaian tumors, anintriguing finding appeared suggesting a novelTNBC subtype, featuring coexpression of bothandrogen receptor and ALDH1.38

As noted earlier, the partnerships within the IBRoncology teams have generated valuable insightsregarding breast cancer phenotypes that are moreprevalent among women with West African ances-try, such as triple-negative tumors. Patient-derivedxenografts (PDXs) represent an exciting researchstrategy for studyingbreast tumor biologyandnoveltherapies. Implantation of breast cancer fragmentsinto mice mammary fat pads yields a renewablesupply of human tumors that can be used for a

variety of in vivo rodent model experiments.Through the KATH-UM research partnership, aseries of PDXs have been created on the basis ofthe tumors from Ghanaian, African American, andwhite American patients with breast cancer.

BREAST CANCER IN GHANA: DIAGNOSTIC ANDTREATMENT ADVANCES PROMOTED BY THEGHANA-MICHIGAN PARTNERSHIP

The IBR collaborative has also served as an in-vestment in improving the clinical services avail-able to Ghanaian women with breast problems.Historically, the KATH breast clinics relied pre-dominantly onopensurgical diagnostic biopsies toconfirm or rule out the presence of cancer in anywoman presenting with a breast abnormality ormass. The scheduling and implementation of asurgical diagnostic procedure are time consum-ing and use valuable, costly operating room re-sources. Furthermore, this sequence increasesthe risk that the affected patient (who has likelyalready traveled a distance and expended per-sonal finances to seek medical attention) may belost to follow-up. The early years of this partnershipfeatured a training program in using percutaneouscore needle biopsies to diagnose breast canceron-site during the outpatient clinic visit. The suc-cess of this program in terms of accurately andefficiently establishing a diagnosis of breast can-cer has been reviewed and documented.39

Table 1 – Features of Breast Cancer in Ghana on the Basis of Studies From the Ghana-Michigan Breast Cancer Research Partnership

Study

Sample Size (No.) andSource of Breast Cancers

in Africa% of Triple-Negative

Tumors Others Features Identified

Stark et al,18 2010 75 82 Frequency of TNBC among AfricanAmericans, 33%; frequency of TNBCamong white Americans, 10%; P , .01

Schwartz et al,20 2013 104 56 Frequency of mammary stem-cell markerALDH1 expression in patients with vwithout TNBC 53% v 33%; P , .05

Proctor et al,21 2015 147 61 Frequency of mammary stem-cell markerALDH1 expression in patients with TNBC,45%; frequency of androgen receptorexpression in patients with TNBC, 24%

Pang et al,22 2012 100 60 Oncogene EZH2 more prevalent in bothnucleus and cytoplasm of TNBC

Der et al,23 2015 219 58 NA

Newman et al,19 2016 234 53 NA

Newman et al,24 2016 234 (KATH, Kumasi,Ghana), 95 (St Paul’sMillennium Hospital,Addis Ababa, Ethiopia)

Ghana: 53Ethiopia: 9

Increased frequency (36%) of HER2/neu-overexpressing breast cancers in Ethiopia

Abbreviations: KATH, Komfo Anokye Teaching Hospital; NA, not applicable; TNBC, triple-negative breast cancer.



Multidisciplinary or multimodality treatment ofbreast cancer requires immunohistochemistry re-sources that can efficiently and accurately assessfor expression of the estrogen receptor, the proges-terone receptor, and HER2/neu. These markers arecritical in being able to appropriately determine apatient’s response to endocrine therapy and/or tar-geted anti-HER2/neu therapy. Lacking this molec-ular marker information, patients with hormonereceptor–negative breast cancer may face thetoxicity of ineffective endocrine therapy, and pa-tients with hormone receptor–positive breast cancermay miss the opportunity to receive life-saving en-docrine therapy. Similarly, HER2/neu expressioncan identify tumors that are quite sensitive to che-motherapyandanti-HER2/neuagents.TheGhana-Michigan partnership featured a training programin immunohistochemistry for the KATH pathologyteam, as well as the development of a resourcesupply and allocation system that has enabled theKATH Breast Oncology Program to routinely gen-erate their own molecular marker reports for eachpatient diagnosed with invasive breast cancer.

This partnership has also generated opportunitiesto use telemedicine technology for real-time in-ternational collaboration andmultidisciplinary dis-cussion. The UM partners invested in dedicatedInternet and teleconference equipment on-siteat KATH. This teleconference unit allows theMichigan-based and KATH teams to discuss pa-tient care on a weekly basis and to share confer-ence proceedings through live interactions.

Our Ghanaian partners have also been able topromote general breast health awareness pro-grams as a consequence of the expanded breastcancer attention generated by this research col-laborative. KATH principal investigator BaffourAwuah has worked with regional herbalists, pri-mary care physicians, and nurses to conducteducational seminars promoting breast cancerearly detection strategies. Lacking adequate finan-cial resources aswell as accurate population-basedstatistics on breast cancer incidence, it is not fea-sible to conduct community-wide mammographicscreening programs. Therefore, these early detec-tion programs largely focus on dissemination ofinformation regarding clinical signs and symptomsof breast cancer (eg, dominant lump, bloody nippledischarge) and the importance of prompt biopsywith initiation of treatment. Although we cannotquantify the effectiveness of our efforts withregard to breast cancer stage distribution inGhana, our KATH colleagues have established atumor registry office (B. Awuah, personal com-munication, December 2014) and have seen

increasing volumes of breast cancer annually aswell as anecdotal observations of more womenpresenting with operable, earlier-stage disease.Strengthening this tumor registry (which hasstruggledwithmaintaining consistent personneland completeness of data collection) remains ahigh priority for the Michigan-Ghana Collabora-tive, and we have also invested in advancedtraining of tumor registry personnel. Althoughlumpectomy and breast radiation are optionsthat are available to patients with breast cancerpatients in Ghana, few women present withtumors that are amenable to the breast conser-vation approach despite anecdotal observationsof an earlier-stage distribution for KATH pa-tients. Furthermore, concerns regarding inade-quate pre- and postoperative mammographicimaging availability have generated suspicionthat breast-conserving surgery cannot be plannedwithoptimal information regardingextent of diseaseand adequacy of resection.

BREAST CANCER AND AFRICAN ANCESTRY: THEGHANA-MICHIGAN PARTNERSHIP AS A MODEL FOREXPANSION

The multifaceted success of the Michigan partner-ship with KATH has established the foundationfor expansion and collaboration with other healthcare facilities in Africa. As a consequence, thisprogram has grown, with exchange programs thathave included St Paul’s Millennium Hospital inEthiopia as well as the following three additionalsites in Ghana: the Korle Bu Teaching Hospital inAccra, the Tamale Teaching Hospital in Tamale,and the Sunyani Teaching Hospital in Sunyani.Table 2 lists the studies that have been publishedand presented through these partnerships. Inter-national expansioneffortsbeyondAfricaarealso indevelopment, as partnerships are being pursuedwith the All India Institute of Medical Sciences inNew Delhi, India, and the Third Xiangya Hospital,Central South University in Hunan, China.

The KATH-UM relationship has served as a platformfor trainingthephysician-scientists fromGhana.Morethan a dozen Ghanaians have spent time at UM forobservorships and research programs. E.J. is asurgeon fromKATH, currently completingherPhDwork at UM in cancer biology, after which she willreturn to Ghana to assume leadership of a Ghana-based translational research program.

Within theUnited States, expansion of this registryhas involved partnership with the Sisters Network(Houston, TX). This is a national organization ofAfrican American breast cancer survivors43 that



currentlyhasamembershipofapproximately3,000women in more than 30 chapters across 22 states.TheSistersNetworkmembership contributes to theIBR by providing recruitment opportunities at theirnational as well as local meetings. Recruitmentinvolves participants agreeing to provide accesstomedical records and saliva specimens suitablefor DNA extraction and genotyping studies. Long-term goals of this registry include studies of germ-line breast cancer risk in women with diverse racialand ethnic backgrounds in the United States com-pared with international populations.

We enthusiastically look forward to this expandedresearch as a valuable contribution to precisionmedicine initiatives. We are also committed toongoing investment in the breast oncology

servicesavailable topatients in all of ourpartneringsites as well as throughout other low- and middle-income countries. Core biopsy programs, immu-nohistochemistry programs, and multidisciplinarytumor board conferences are examples of theservices that are promoted through our re-search collaborative.

Other investigators with an interest in this type ofinternational cancer research should be mindful ofevolving policies that regulate transport of humantissue via commercial airlines. Updates on regula-tory requirements and restrictions can be obtainedat the Centers for Disease Control and PreventionWeb site.44 Our group typically transported breasttissues for immunohistochemistry studies asformalin-fixed specimens embedded in paraffinblocks. The fresh specimens for PDX work weretransported in dry ice. Commercial carriers haveregulations for labeling and packaging of dry ice,includingmaximum limits forquantity ofdry ice thatcan be transported. Updates on these regulationscan be accessed through the particular carrier’sdangerous goods office andWeb site. Investigatorsshould alsoworkwith their institution’s liaison to theOccupational Safety and Health Administration fortraining in handling dry ice, and further informationcan be obtained from the Occupational Safety andHealth Administration directly.

International partnerships represent a powerfuland unique opportunity to advance insights re-garding the etiology of domestic disparities inbreast cancer burden related to racial and ethnicidentity. These efforts also provide valuable cul-tural, academic, and educational exchange pro-grams as well as opportunities to strengthen theoncology services in under-resourced countries.Our program is certainly not unique; medicalmissionary–type work has existed for many de-cades. Hopefully this summary of our 10-yearexperience with international breast cancer out-reach and research will provide motivation forothers to add to this growing field.

DOI: 10.1200/JGO.2015.002881Published online on jgo.ascopubs.org on July 27, 2016.


Financial support: Max Wicha, Lisa Newman

Administrative support: Kofi Gyan, Kathy Toy, RobertNewman Brewer, Barbara Salem, Azadeh Stark, LisaNewmanProvision of study materials or patients: Evelyn Jiagge,Joseph Kwaku Oppong, Jessica Bensenhaver, FrancisAitpillah, Ishmael Kyei, Ernest Osei-Bonsu, ErnestAdjei, Linda Ahenkorah Fondjo, Osei Owusu-Afriyie,Amma Gyamfuah, Max Wicha, Celina Kleer, Judy Pang,

Azadeh Stark, Francis Abantanga, Lisa Newman, BaffourAwuah



The following represents disclosure information provided byauthors of this manuscript. All relationships are consideredcompensated. Relationships are self-held unless noted. I 5

Table 2 – Published Studies and Reports Related to the Ghana-Michigan Breast CancerCollaborative

Report Year

“Guide for Investigators Conducting International Research InvolvingDeveloping Nations”15

2010

“African Ancestry and Higher Prevalence of Triple-Negative BreastCancer: Findings From an International Study”18

2010

“ImplementationofaPercutaneousCoreNeedleBiopsyTrainingProgram:Results from the University of Michigan-Komfo Anokye TeachingHospital Breast Cancer Research Partnership”39

2011

“Invasive Breast Carcinomas in Ghana: High Frequency of High Grade,Basal-Like Histology and High EZH2 Expression”22

2012

“Expression of AldehydeDehydrogenase 1 as aMarker of Mammary StemCells in Benign and Malignant Breast Lesions of Ghanaian Women”40

2013

“Androgen Receptor Expression in Ghanaian Breast Cancer Cases: NovelCorrelation With ALDH1 in Triple-Negative Tumors”21,38

2014

“Distinct Pathways Differentiate the CD441Mesenchymal-Like From theALDH1 Epithelial-Like Phenotype of Triple Negative Breast CancerStem Cells”41

2014

“Creating a Comprehensive Patient-Derived Xenograft (PDX) Bank toRepresent Racial Disparities in Triple Negative Breast Cancer”42

2014

“Global Surgical Oncology Disease Burden: Addressing Disparities viaGlobal Surgery Initiatives—The University of Michigan InternationalBreast Cancer Registry”10

2015

“Comparative Analysis of Breast Cancer Phenotypes in African American,White American, and West Versus East African Patients: CorrelationBetween African Ancestry and Triple Negative Breast Cancer”24

2016





Immediate Family Member, Inst 5 My Institution. Relation-ships may not relate to the subject matter of this manuscript.For more information about ASCO’s conflict of interest policy,please refer to www.asco.org/rwc or jco.ascopubs.org/site/ifc.

Evelyn JiaggeNo relationship to disclose

Joseph Kwaku OppongNo relationship to disclose

Jessica BensenhaverNo relationship to disclose

Francis AitpillahNo relationship to disclose

Kofi GyanNo relationship to disclose

Ishmael KyeiNo relationship to disclose

Ernest Osei-BonsuNo relationship to disclose

Ernest AdjeiNo relationship to disclose

Michael Ohene-YeboahNo relationship to disclose

Kathy ToyNo relationship to disclose

Karen Eubanks JacksonNo relationship to disclose

Marian AkpalooNo relationship to disclose

Dorcas AcheampongNo relationship to disclose

Beatrice AntwiNo relationship to disclose

Faustina Obeng AgyemanNo relationship to disclose

Zainab AlassanNo relationship to disclose

Linda Ahenkorah FondjoNo relationship to disclose

Osei Owusu-AfriyieNo relationship to disclose

Robert Newman BrewerNo relationship to disclose

Amma GyamfuahNo relationship to disclose

Barbara SalemNo relationship to disclose

Timothy JohnsonNo relationship to disclose

Max WichaConsulting or Advisory Role: AstraZeneca/MedImmune,Verastem, Paganini Biopharma, Cormorant, Roche

Research Funding: Medmmune, Dompe Farmaceutici,Paganini Biopharma, Verastem

Sofia MerajverNo relationship to disclose

Celina KleerNo relationship to disclose

Judy PangNo relationship to disclose

Emmanuel Amankwaa-FrempongNo relationship to disclose

Azadeh StarkNo relationship to disclose

Francis AbantangaNo relationship to disclose

Lisa NewmanNo relationship to disclose

Baffour AwuahNo relationship to disclose

Affiliations

Evelyn Jiagge, Joseph Kwaku Oppong, Francis Aitpillah, Ishmael Kyei, Ernest Osei-Bonsu, Ernest Adjei, Marian Akpaloo, DorcasAcheampong, Beatrice Antwi, Faustina Obeng Agyeman, Zainab Alhassan, Osei Owusu-Afriyie, Amma Gyamfuah, EmmanuelAmankwaa-Frempong, Francis Abantanga, and Baffour Awuah, Komfo Anokye Teaching Hospital; Linda Ahenkorah Fondjo, KwameNkrumahUniversity of Science and Technology, Kumasi;Michael Ohene-Yeboah, University of GhanaMedical School andKorle-Bu, Accra, Ghana; Evelyn Jiagge, Jessica Bensenhaver, Kathy Toy, Robert Newman Brewer, Timothy Johnson, Max Wicha, SofiaMerajver, Celina Kleer, Judy Pang, and Lisa Newman, University ofMichiganMedical School, Ann Arbor; Jessica Bensenhaver, KofiGyan, Barbara Salem, Azadeh Stark, and Lisa Newman, Henry Ford Health System International Center for the Study of BreastCancer Subtypes, Detroit, MI; and Karen Eubanks Jackson, Sisters Network, Houston, TX.SupportedbySusanG.Komen for theCure, theRichardandSusanBayerBreastCancer Fund,ThinkPinkRocks, theUniversity ofMichigan Comprehensive Cancer Center, QVC Fashion Footwear New York, and National Institutes of Health National CancerCenter Grant No. U01 CA 154224.

REFERENCES

references

1. Forouzanfar MH, Foreman KJ, Delossantos AM, et al: Breast and cervical cancer in 187 countries between 1980 and2010: A systematic analysis. Lancet 378:1461-1484, 2011

2. Ferlay J, Shin HR, Bray F, et al: Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 127:2893-2917, 2008





3. International Agency for Research on Cancer: Globocan Fact Sheet. http://globocan.iarc.fr/Pages/fact_sheets_cancer.aspx

4. Jemal A, Bray F, Forman D, et al: Cancer burden in Africa and opportunities for prevention. Cancer 118:4372-4384,2012

5. Parkin DM, Nambooze S, Wabwire-Mangen F, et al: Changing cancer incidence in Kampala, Uganda, 1991-2006. IntJ Cancer 126:1187-1195, 2010

6. Laryea DO, Awuah B, Amoako YA, et al: Cancer incidence in Ghana, 2012: Evidence from a population-based cancerregistry. BMC Cancer 14:362, 2014

7. Jemal A, Bray F, Center MM, et al: Global cancer statistics. CA Cancer J Clin 61:69-90, 2011

8. Reference deleted.

9. Newman LA: Breast cancer disparities: High-risk breast cancer and African ancestry. Surg Oncol Clin N Am 23:579-592, 2014

10. Jiagge E, Bensenhaver JM, Oppong JK, et al: Global surgical oncology disease burden: Addressing disparities viaglobal surgery initiatives—The University of Michigan International Breast Cancer Registry. Ann Surg Oncol 22:734-740, 2015

11. DeSantis CE, Fedewa SA, Goding Sauer A, et al: Breast cancer statistics, 2015: Convergence of incidence ratesbetween black and white women. CA Cancer J Clin 66:31-42, 2016

12. Jatoi I, Becher H, Leake CR: Widening disparity in survival between white and African-American patients with breastcarcinoma treated in the U.S. Department of Defense Healthcare system. Cancer 98:894-899, 2003

13. Newman LA, Griffith KA, Jatoi I, et al: Meta-analysis of survival in African American and white American patients withbreast cancer: Ethnicity compared with socioeconomic status. J Clin Oncol 24:1342-1349, 2006

14. Fregene A, Newman LA: Breast cancer in sub-Saharan Africa: How does it relate to breast cancer in African-Americanwomen? Cancer 103:1540-1550, 2005

15. Martin IK, Awuah B, Newman LA: Guide for investigators conducting international cancer research involving de-veloping nations. Cancer 116:1396-1399, 2010

16. Kohler BA, Sherman RL, Howlader N, et al: Annual report to the nation on the status of cancer, 1975-2011, featuringincidence of breast cancer subtypes by race/ethnicity, poverty, and state. J Natl Cancer Inst 107:djv048, 2015

17. Newman LA, Reis-Filho JS, Morrow M, et al: The 2014 Society of Surgical Oncology Susan G. Komen for the curesymposium: Triple-negative breast cancer. Ann Surg Oncol 22:874-882, 2015

18. Stark A, Kleer CG,Martin I, et al: African ancestry and higher prevalence of triple-negative breast cancer: Findings froman international study. Cancer 116:4926-4932, 2010

19. Newman LA, Jiagge E, Bensenhaver JM, et al: Comparative analysis of breast cancer phenotypes in African American,white American, and African patients: Correlation between African ancestry and triple negative breast cancer. CancerRes 76, 2016 (4 suppl 4; abstr P6-12-14)

20. Schwartz T, Stark A, Pang J, et al: Expression of ALDH1 as a marker of mammary stem cells in benign and malignantbreast lesions of Ghanaian women. Cancer 119:488-494, 2013

21. Proctor E, Kidwell KM, Jiagge E, et al: Characterizing breast cancer in a population with increased prevalence of triple-negative breast cancer: Androgen receptor and ALDH1 expression in Ghanaian women. Ann Surg Oncol 22:3831-3835, 2015

22. Pang J, Toy KA, Griffith KA, et al: Invasive breast carcinomas in Ghana: High frequency of high grade, basal-likehistology and high EZH2 expression. Breast Cancer Res Treat 135:59-66, 2012

23. Der EM, Gyasi RK, Tettey Y, et al: Triple-negative breast cancer in Ghanaian women: The Korle Bu Teaching Hospitalexperience. Breast J 21:627-633, 2015

24. Newman L, Jiagge E, Bensenhaver JM, et al: Comparative analysis of breast cancer phenotypes in African American,white American, andWest versus East African patients: Correlation betweenAfrican ancestry and triple negative breastcancer. Ann Surg Oncol 23:S8, 2016 (suppl 1; abstr 8)

25. CampbellMC, Hirbo JB, Townsend JP, et al: The peopling of the African continent and the diaspora into the newworld.Curr Opin Genet Dev 29:120-132, 2014

26. Harich N, Costa MD, Fernandes V, et al: The trans-Saharan slave trade: Clues from interpolation analyses and high-resolution characterization of mitochondrial DNA lineages. BMC Evol Biol 10:138, 2010

27. Richards M, Rengo C, Cruciani F, et al: Extensive female-mediated gene flow from sub-Saharan Africa into neareastern Arab populations. Am J Hum Genet 72:1058-1064, 2003

28. Stefflova K, Dulik MC, Barnholtz-Sloan JS, et al: Dissecting the within-Africa ancestry of populations of African descentin the Americas. PLoS One 6:e14495, 2011

29. Salas A, Carracedo A, Richards M, et al: Charting the ancestry of African Americans. Am J Hum Genet 77:676-680,2005


http://globocan.iarc.fr/Pages/fact_sheets_cancer.aspx


30. Bryc K, Auton A, NelsonMR, et al: Genome-wide patterns of population structure and admixture in West Africans andAfrican Americans. Proc Natl Acad Sci USA 107:786-791, 2010

31. Ginestier C, Hur MH, Charafe-Jauffret E, et al: ALDH1 is a marker of normal and malignant human mammary stemcells and a predictor of poor clinical outcome. Cell Stem Cell 1:555-567, 2007

32. Kunju LP, Cookingham C, Toy KA, et al: EZH2 and ALDH-1 mark breast epithelium at risk for breast cancer de-velopment. Mod Pathol 24:786-793, 2011

33. Moore HM, Gonzalez ME, Toy KA, et al: EZH2 inhibition decreases p38 signaling and suppresses breast cancermotility and metastasis. Breast Cancer Res Treat 138:741-752, 2013

34. Christgen M, Geffers R, Kreipe H, et al: IPH-926 lobular breast cancer cells are triple-negative but their microarrayprofile uncovers a luminal subtype. Cancer Sci 104:1726-1730, 2013

35. Lehmann BD, Bauer JA, Chen X, et al: Identification of human triple-negative breast cancer subtypes and preclinicalmodels for selection of targeted therapies. J Clin Invest 121:2750-2767, 2011

36. Lehmann BD, Pietenpol JA: Identification and use of biomarkers in treatment strategies for triple-negative breastcancer subtypes. J Pathol 232:142-150, 2014

37. Masuda H, Baggerly KA, Wang Y, et al: Differential response to neoadjuvant chemotherapy among 7 triple-negativebreast cancer molecular subtypes. Clin Cancer Res 19:5533-5540, 2013

38. Proctor E, Jiagge E, Kleer C, et al: Androgen receptor expression in Ghanaian breast cancer cases: Novel correlationwith ALDH1 in triple-negative tumors. Presented at the Society of Surgical Oncology Annual Cancer Symposium,Phoenix, AZ, March 12-15, 2014

39. Awuah B, Martin IK, Takyi V, et al: Implementation of a percutaneous core needle biopsy training program: Resultsfrom theUniversity ofMichigan-Komfo Anokye TeachingHospital breast cancer research partnership. Ann SurgOncol18:957-960, 2011

40. Schwartz T, Stark A, Pang J, et al: Expression of aldehyde dehydrogenase 1 as a marker of mammary stem cells inbenign and malignant breast lesions of Ghanaian women. Cancer 119:488-494, 2013

41. Jiagge EM, Wong S, Luther T, et al: Distinct pathways differentiate the CD441 mesenchymal-like from the ALDH1epithelial-like phenotype of triple negative breast cancer stem cells. Cancer Res 74, 2014 (suppl 19; abstr LB-59)

42. Luther T, Jiagge E, Lewis MT, et al: Creating a comprehensive patient-derived xenograft (PDX) to represent racialdisparities in triple negative breast cancer. Cancer Epidemiol Biomarkers Prev 24, 2015 (suppl 10; abstr PR09)

43. Newman LA, Jackson KE: Advocacy Corner: Sisters Network, Inc. J Oncol Pract 2009:313-314, 2009

44. Centers for Disease Control and Prevention: Importation. http://www.cdc.gov/importation/index.html


http://www.cdc.gov/importation/index.html


specialarticle

Management and Care of WomenWith Invasive Cervical Cancer:American Society of Clinical OncologyResource-Stratified Clinical PracticeGuideline

executivesum

mary

Purpose To provide evidence-based, resource-stratified global recommendations to clinicians and policy-makers on the management and palliative care of women diagnosed with invasive cervical cancer.

Methods ASCO convened a multidisciplinary, multinational panel of cancer control, medical and radiationoncology,healtheconomic,obstetricandgynecologic,andpalliativecareexperts toproducerecommendationsreflectingresource-tieredsettings.Asystematicreviewof literature from1966to2015failed toyieldsufficientlystrong quality evidence to support basic- and limited-resource setting recommendations; a formal consensus-based process was used to develop recommendations. A modified ADAPTE process was also used to adaptrecommendations from existing guidelines.

Results Five existing sets of guidelines were identified and reviewed, and adapted recommendations formthe evidence base. Eight systematic reviews, along with cost-effectiveness analyses, provided indirectevidence to inform the consensus process, which resulted in agreement of 75% or greater.

Recommendations Clinicians and planners should strive to provide access to the most effective evidence-based antitumor and palliative care interventions. If a woman cannot access these within her own orneighboring country or region, she may need to be treatedwith lower-tier modalities, depending on capacityand resources for surgery, chemotherapy, radiation therapy, and supportive and palliative care. For womenwith early-stage cervical cancer in basic settings, cone biopsy or extrafascial hysterectomy may be per-formed. Fertility-sparing procedures or modified radical or radical hysterectomy may be additional optionsin nonbasic settings. Combinations of surgery, chemotherapy, and radiation therapy (including brachy-therapy) should be used for women with stage IB to IVA disease, depending on available resources. Paincontrol is a vital component of palliativecare. Additional information isavailableatwww.asco.org/rs-cervical-cancer-treatment-guidelineandwww.asco.org/guidelineswiki. It is the viewofASCO that healthcareprovidersand health care system decision makers should be guided by the recommendations for the highest stratum ofresources available. The guideline is intended to complement but not replace local guidelines.

INTRODUCTION

The purpose of this guideline is to provide expertguidance to clinicians and policymakers in allresource settings on work-up, treatment, and pal-liative care for women diagnosed with invasivecervical cancer. The target population is womenwho have been diagnosed with cervical cancer.

There are large disparities regionally and glob-ally in incidence of and mortality resulting fromcervical cancer, in part because of disparitiesin the provision of mass screening. (A separateASCO resource-stratified guideline provides

guidance on screening.) Treatment of cervicalcancer is dependent on the stage of disease.Treatment may include surgical treatments suchas conization, hysterectomy or radical hysterec-tomy, radiation therapy, and/or chemotherapy.Different regions of the world, both among andwithin countries, differ with respect to access tothese treatments. In particular, regions with fewerresources tend to have poorer screening pro-grams, and patients present with more advanceddisease that requires either radical surgery orchemoradiotherapy, neither of which is readily

Linus T. Chuang

Sarah Temin

Rolando Camacho

Alfonso Due~nas-Gonzalez

Sarah Feldman

Murat Gultekin

Vandana Gupta

Susan Horton

Graciela Jacob

Elizabeth A. Kidd

Kennedy Lishimpi

Carolyn Nakisige

Joo-Hyun Nam

Hextan Yuen SheungNgan

William Small

Gillian Thomas

Jonathan S. Berek


This guideline has beenendorsed by the Society ofGynecologic Oncology.

Authors’ disclosures ofpotential conflicts ofinterest and authorcontributions are found atthe end of this article.Corresponding author:American Society ofClinical Oncology, 2318Mill Rd, Suite 800,Alexandria, VA 22314;e-mail: [email protected].









available in these areas. For this reason, standardguidelines that assume ideal availability of surgeryand radiotherapy may not be applicable. The goalof our guideline is to recommend options in set-tings inwhich ideal treatment regimensmaynotbeavailable.

Approximately 85% of incident cervical cancersoccur in less developed regions (also knownas low- and middle-income countries [LMICs])around the world, representing 12% of women’scancers in those regions. Eighty-seven percent ofdeaths resulting from cervical cancer occur inthese less developed regions.1 Some of the re-gions in the world with the highest mortalityrates include the WHO South-East Asia andWestern Pacific regions, followed by India andAfrica.1 As a result of these disparities, the ASCOResource-Stratified Guideline Advisory Groupchose cervical cancer as a priority topic for guide-line development.

Disparities exist not only among countries andglobal regions but also within countries. For ex-ample, within the United States, some regionshave higher incidences of cervical cancer (eg,Texas, 9.2; Arkansas, 9.8; and Mississippi, 9.7per 100,000 women)2 as well as rates of povertygreater than 17%.3 In the United States, blackwomenexperiencehigher incidence andmortalitythan women of other races or ethnicities.2 Aware-ness of these disparities in access to care shouldbe considered in the context of this clinical prac-tice guideline, and health care providers shouldstrive to deliver the highest level of cancer care toall women.

ASCO has established a process for resource-stratified guidelines, which includes mixedmethods of guideline development, adaptationof the clinical practice guidelines of other organi-zations, and formal expert consensus. This articlesummarizes the results of that process and pres-ents resource-stratified recommendations, whichare based, in part, on expert consensus and adap-tation from the Canadian Cancer Care Ontario(CCO),4,5 European Society of Medical Oncology(ESMO),6 Japan Society of Gynecologic Oncology(JSGO),7 USNational Comprehensive Cancer Net-work (NCCN),8 andmultinational WHO9 guidelinesfor the care and palliation of women with invasivecervical cancer. Appendix Table A1 lists links tothese guidelines.

This ASCO guideline reinforces selected recom-mendations offered in the CCO, ESMO, JSGO,NCCN, and WHO guidelines and acknowl-edges the effort put forth by the authors and

aforementioned societies to produce evidence-based and/or consensus-based guidelines in-forming practitioners and institutions caringfor patients with invasive cervical cancer. Indeveloping resource-stratified guidelines, ASCOhas adopted its framework from the four-tierapproach (basic, limited, enhanced, and maxi-mal, summarized inTable1andAppendix TableA2)developed by WHO and applied by the BreastHealth Global Initiative and made modificationsto that framework based on Disease Control Pri-orities 3 and uses an evidence-based approach toinform guideline recommendations.10-12

GUIDELINE QUESTIONS

This clinical practice guideline addresses fouroverarching clinical questions: In the basic-, lim-ited-, enhanced-, and maximal-resource settings,what are the appropriate care options for womenwith invasive cervical cancer in work-up, treat-ment, follow-up and post-treatment surveillance,and palliative care?

METHODS

Guideline Development Process

These recommendations were developed by anASCO Expert Panel with multinational and multi-disciplinary representation. The Expert Panel metvia teleconference and in person and corre-sponded through e-mail. On the basis of theconsideration of the evidence, the authors wereasked to contribute to the development of theguideline, provide critical review, and finalizethe guideline recommendations. Members ofthe Expert Panel were responsible for reviewingand approving the penultimate version of theguideline, which was then circulated for externalreview and submitted to a peer-reviewed journalfor editorial review and consideration for publica-tion. This guideline was partially informed by theASCO modified Delphi formal expert consensusmethodology, during which the Expert Panel wassupplemented by additional experts recruited torate their agreement with the drafted recommen-dations. The entire membership of experts is re-ferred to as the ASCO Consensus Panel (DataSupplement provides a list of members). All ASCOguidelines are ultimately reviewed and approvedby theExpert Panel and theASCOClinical PracticeGuideline Committee (CPGC) before publication.

The guideline development process was also in-formed by the ADAPTE methodology13 and con-sensus processes used together as an alternative tode novo recommendation development for this



guideline. However, a special section on radiationtherapy in resource-constrained settings does in-cludedenovoexpert consensus recommendations.Adaptation of guidelines is considered by ASCOin selected circumstances, when one or moreguidelines from other organizations already existon the same topic. The objective of the ADAPTEprocess is to take advantage of existing guide-lines to enhance efficient production, reduce du-plication, and promote the uptake of guidelinerecommendations.

The ASCO adaptation and formal expert consen-sus processes begin with a literature search toidentify literature including candidate guidelinesfor adaptation. The panel used literature searches(from 1966 to 2015, with additional searches forliterature published in specific areas), existingguidelines and expert consensus publications,some observational studies, and clinical experi-ence as guides.

Adapted guideline manuscripts are reviewed andapproved by the ASCOCPGC. The review includes

Table 1 – Treatment Capacity

Treatment

Setting

Basic Limited Enhanced Maximal

Surgery Simple (extrafascial)hysterectomyormore extensive hysterectomycan be performed*

Modified radical or radicalhysterectomy

Capableofperformingmostmajorsurgeries, including radicalhysterectomy, radical trache-lectomy,† pelvic and para-aortic LN sampling, and pelvicexenteration†

Radical hysterectomy, radicaltrachelectomy, pelvic andpara-aortic LN sampling,sentinel node biopsy, andpelvic exenteration; RT,chemotherapy, interventionalradiology, palliative careservice, and bevacizumab areall available

Following are not available: PETscan, interventional radiology,sentinel node biopsy/IORT, orbevacizumab

Chemotherapy Availability of chemotherapydrugs is unpredictable

Chemotherapy may be available Chemotherapy available;bevacizumab not available

Chemotherapy available;bevacizumab is available

RT No RT available Limited external RT with nobrachytherapy available; insomeareaswhere there is onlybrachytherapy and no externalRT, this will be considered asbasic level

RT including external beam andbrachytherapy available;interventional radiology notavailable

RT including external beam andbrachytherapy available;interventional radiologyavailable

Pathology Pathology services are notavailable; if there is a way tosend pathology for reviewwhen needed, that shouldoccur

Pathology services indevelopment

Pathology services in develop-ment or not always available

Pathology available

(There are basic pathology andfrozen section services;consultations are not readilyavailable)

(Pathology services includingfrozen sections are available;tumor registry and regularmultidisciplinary conferencesare not consistently availablein the region)

(Full pathology services includingdiagnosis, consultation, tumorregistry, and multidisciplinaryconferences are available)(Basic pathology may be avail-

able, but diagnosis is oftendelayed for more than 1month; there are no frozensections or pathologyconsultations in the region)

Palliative care Palliative care service is indevelopment; basic palliativecare, including pain andsymptom management,should be provided‡

Pain and symptom managementavailable; palliative careservice is in development

Palliative care service not alwaysavailable

Palliative care service available

NOTE. It is the view of the American Society of Clinical Oncology that health care providers and health care system decisionmakers should be guided by the recommendations forthe highest stratum of resources available. This guideline is intended to complement but not replace local guidelines. Bold font indicates addition of a recommended action overa previous resource level (eg, in limited setting, a bold action is one that was not recommended in basic).Abbreviations: IORT, intraoperative radiation therapy; LN, lymph node; PET, positron emission tomography; RT, radiotherapy.*Where medical facilities exist to take care of women who are at high risk for postoperative complications.†Can be performed at some enhanced levels.‡Palliative care is multifaceted and in some contexts can be provided concurrently with tumor-directed therapy. Pain management and best supportive care are necessary butinsufficient parts of palliative care in all settings. Womenwith advanced cervical cancer with or without access to tumor-directed therapymay have specific late-stage symptomsthat require clinicians to perform or offer urogenital-specific interventions. See Special Commentary.



THE BOTTOM LINE

Management and Care of Women With Invasive Cervical Cancer: American Society of Clinical OncologyResource-Stratified Clinical Practice Guideline

Guideline Question

In basic-, limited-, enhanced-, andmaximal-resource settings, what are the appropriate care options forwomen diagnosed with invasive cervical cancer?

Target Population

Women at all levels of resource settings diagnosed with invasive cervical cancer.

Target Audience

This clinical practice guideline globally targets health care providers (including gynecologic oncologists,medical oncologists, radiation oncologists, obstetricians and gynecologists, surgeons, nurses, andpalliative care clinicians), policymakers, patients, and caregivers.

Methods

A multinational, multidisciplinary Expert Panel was convened to develop clinical practice guidelinerecommendations based on a systematic review of the medical literature and/or an expert consensusprocess.

Author’s note. It is the view of the American Society of Clinical Oncology (ASCO) that health careproviders and health care system decision makers should be guided by the recommendations for thehighest stratum of resources available. The guidelines are intended to complement but not replacelocal guidelines.

Key Points*

·If follow-up is available, theExpertPanel recommendsconebiopsy forwomenwith stage IA2disease inbasic settings and cone biopsy plus pelvic lymphadenectomy in limited settings. In enhanced andmaximal settings, radical trachelectomy is recommended for patients with stage IB1 cervical cancerwith tumor size up to 2 cm who desire fertility-sparing surgery.

·In basic settings where patients cannot be treated with radiation therapy, extrafascial hysterectomyeither alone or after chemotherapymay be an option for women with stage IA1 to IVA cervical cancer.

·In basic settings, for women with larger tumors or advanced-stage cervical cancer, neoadjuvantchemotherapy is recommended, whenever chemotherapy is available, for the purpose of shrinkingthe tumor before performing hysterectomy.

·Concurrent radiotherapy andchemotherapy is standard in enhancedandmaximal settings forwomenwith stage IB to IVA disease.

·The panel stresses the addition of low-dose chemotherapy during radiotherapy but not at the cost ofdelaying radiation therapy if chemotherapy is not available.

·In limited-resource settings where there is no brachytherapy, the ASCO Expert Panel recommendsextrafascial hysterectomy or its modification for women who have residual tumor 2 to 3 months afterconcurrent chemoradiotherapy and additional boost.

·For patients with stage IV or recurrent cervical cancer, single-agent chemotherapy (carboplatin orcisplatin) is recommended in basic settings.

·If the resources are available and the patient cannot receive treatment with curative intent, palliativeradiotherapy should be used to relieve symptoms of pain and bleeding.

·Where resources are constrained, single- or short-course radiotherapy schemes can be used withretreatments if feasible for persistent or recurrent symptoms.

(continued on following page)



two parts: methodologic review and content re-view. The formerwas completedby twoASCOstaffmembers and the latter by members of the ExpertPanel (Appendix Table A3) convened by ASCO,which included multinational and multidisciplin-ary representation.

The guideline recommendations were crafted,in part, using the Guidelines Into Decision Sup-port (GLIDES) methodology and accompanyingBRIDGE-Wiz software.14 Detailed informationabout the methods used to develop this guide-line is available in the Methodology Supplementand Data Supplement at www.asco.org/rs-cervical-cancer-treatment-guideline.

The ASCO Expert Panel and guideline staff willwork with co-chairs to keep abreast of any sub-stantive updates to the guideline. On the basis of aformal review of the emerging literature, ASCOwilldetermine the need to update this guideline. Thisis themost recent information as of the publicationdate. Visit the ASCO Guidelines Wiki at www.asco.org/guidelineswiki to submit new evidence.

Guideline Disclaimer

The clinical practice guideline and other guidancepublished herein are provided by ASCO to assistproviders in clinical decision making. The infor-mation herein should not be relied upon as beingcomplete or accurate, nor should it be consideredas inclusive of all proper treatments or methods ofcare or as a statement of the standard of care.Withthe rapid development of scientific knowledge,new evidence may emerge between the time in-formation is developed andwhen it is published or

read. The information is not continually updatedandmay not reflect themost recent evidence. Theinformation addresses only the topics specificallyidentified herein and is not applicable to otherinterventions, diseases, or stages of diseases. Thisinformation does not mandate any particularcourse of medical care. Furthermore, the infor-mation is not intended to substitute for the in-dependent professional judgment of the treatingprovider, because the information does not accountfor individual variation among patients. Recommen-dations reflect high, moderate, or low confidencethat the recommendation reflects the net effect of agiven course of action. The use of words like “must,”“must not,” “should,” and “should not” indicatesthat a course of action is recommended or not rec-ommended for either most or many patients, butthere is latitude for the treating physician to selectother courses of action in individual cases. In allcases, the selected course of action should beconsidered by the treating provider in the contextof treating the individual patient. Use of the in-formation is voluntary. ASCO provides this infor-mation on an as-is basis and makes no warranty,express or implied, regarding the information.ASCO specifically disclaims any warranties ofmerchantability or fitness for a particular use orpurpose. ASCO assumes no responsibility for anyinjury or damage to persons or property arisingout of or related to anyuseof this informationor forany errors or omissions.

Guideline and Conflicts of Interest

The Expert Panel was assembled in accor-dance with the ASCO Conflicts of Interest Policy

THE BOTTOM LINE (CONTINUED)

Qualifying Statements

ASCO believes that cancer clinical trials are vital to inform medical decisions and improve cancer careand that all patients should have the opportunity to participate.

Palliative care and pain management are part of the treatment of cancers, including cervical cancer, toavoid unnecessary suffering during the final stages of disease. Pain control is a vital component ofpalliative care, a basic human right often neglected in cancer control programs.

Additional Resources

More information, including a Data Supplement with additional evidence tables, a MethodologySupplement with information about evidence quality and strength of recommendations, slide sets,and clinical tools and resources, is available at www.asco.org/rs-cervical-cancer-treatment-guideline.Patient information is available at www.cancer.net.

*Not all recommendations for all settings are listed. Please see Tables 3-7.







http://www.cancer.net


Implementation for Clinical Practice Guidelines(found at www.asco.org/rwc). All members ofthe panel completed the ASCO disclosure form,which requires disclosure of financial and otherinterests, including relationships with commercialentities that are reasonably likely to experiencedirect regulatory or commercial impact as a resultof promulgation of the guideline. Categories fordisclosure include employment; leadership; stockor other ownership; honoraria; consulting or advi-sory role; speaker’s bureau; research funding; pat-ents, royalties, other intellectual property; experttestimony; travel, accommodations, expenses; andother relationships. In accordance with the ASCOpolicy, themajority of themembers of the panel didnot disclose any relationships constituting a conflictunder the policy.

RESULTS

As part of the systematic literature review, Pub-Med, SAGE, Cochrane Systematic Review, andNational Guideline Clearinghouse databasesweresearched for guidelines, systematic reviews, andmeta-analyses published between1966 and Jan-uary2015.Additional searcheswereconducted inPubMed inFebruary,May, July, andAugust 2015.Inclusioncriteria includedpublicationswithapop-ulation of women with invasive cervical cancer

that addressed treatment, follow-up, and/or palli-ative care andweredevelopedby experts aspart ofa recognized organizational effort. To inform rec-ommendations for basic- or limited-resource set-tings, preference was given to studies conductedin these settings.

Searches for cost-effectiveness analyses werealso conducted. Articles were excluded fromthe systematic review if they were meeting ab-stracts; books, editorials, commentaries, let-ters, news articles, case reports, or narrativereviews; or publications in a language otherthan English.

A total of 13 guidelines and seven systematic re-views were found in the literature search.4-7,15-31

On thebasis of content andmethodology reviews,the Expert Panel chose five sets of guidelines—one from Japan, one from Western Europe,two fromCanada, andone from theUnitedStates—to form the evidentiary basis for the guidelinerecommendations.4-7,15-18 In addition, theWHOPalliative Care Chapter (Chapter 7) in the “WHOGuideline on Comprehensive Cervical Cancer Con-trol: A Guide to Essential Practice” was identifiedand deemed eligible for adaptation.9 An addi-tional eight systematic reviews26-33 and economicanalyses34-36 were selected to provide indirect

Table 2 – Work-Up

Setting


History and physical examination,CBC, cervical biopsy, conebiopsy,and LFT/renal function studies

History and physical examination,CBC, cervical biopsy, pathologicreview, cone biopsy, and LFT/renal function studies



Imaging (optional in < stage IB1disease): chest x-ray

Imaging (optional in < stage IB1):chest x-ray, CT (specifically CTof abdomen and pelvis for womenwith advanced-stage disease fortreatment planning purposes)

Imaging (optional in < stage IB1):chest x-ray, CT, or MRI

Imaging (optional in < stage IB1):chest x-ray, CT, or MRI or PET-CT

Smoking cessation and counseling;may offer HIV testing




Optional: EUA cystoscopy/proctoscopy only if suspicionof bladder or rectum invasion byCT or MRI

Optional: EUA cystoscopy/proctoscopy only if suspicion ofbladder or rectum invasion by CTor MRI

Type of recommendation: evidencebased




Overall evidence quality:intermediate

Overall evidence quality: high Overall evidence quality: high Overall evidence quality: high

Strength of recommendation:moderate

Strength of recommendation:moderate

Strength of recommendation: strong Strength of recommendation: strong

NOTE. Bold indicates addition of a recommended action over a previous resource level (eg, in limited setting, a bold action is one that was not recommended in basic).Abbreviations: CBC, complete blood count; CT, computed tomography; EUA, examination under anesthesia; LFT, liver function test; MRI, magnetic resonance imaging; PET,positron emission tomography.




Table 3 – Recommendations for Stage IA, IB, and IIA Disease

Type of Disease

Setting


IA1, LVSI negative,FS (seeDiscussion)

1A1 (negativemargins): conebiopsy* (with scalpel)

1A1 (negativemargins): conebiopsy



Repeat cone biopsy orextrafascial hysterectomyfor positive margins




Type of recommendation:evidence based




Evidence: high Evidence: low Evidence: high Evidence: highRecommendation: strong Recommendation: weak Recommendation: strong Recommendation: strong

IA1, LVSI positive,FS

Cone biopsy in selectedcases, if follow-up possible

Cone biopsy Cone biopsy plus PLND (seeDiscussion regarding currentevidence on FS for womendesiring fertility preservation)

Cone biopsy plus PLND

Type of recommendation:consensus based


Type of recommendation:evidence and consensusbased


Evidence: intermediate Evidence: intermediate Evidence: high Evidence: highRecommendation: weak Recommendation: weak Recommendation: strong Recommendation: strong

OR radical trachelectomyplus PLND

OR radical trachelectomy plusPLND (may offer6 SLN)



Evidence: intermediate Evidence: intermediateRecommendation: moderate Recommendation: moderate

IA1, non-FS (noLVSI)

Cone biopsy (if follow-uppossible) OR extrafascialhysterectomy,† thenobserve after initial conebiopsy, repeat cone, orextrafascial hysterectomy ifmargins are positive

Cone biopsy (if follow-uppossible); observe(after cone biopsy)‡ ORextrafascial hysterectomy†(extrafascial hysterectomyOR modified radicalhysterectomyplusPLNDORif positive marginsrepeat conization§)

Cone biopsy‡ OR extrafascialhysterectomy† (extrafascialhysterectomy OR modifiedradical hysterectomy plusPLND OR if positivemargins repeatconization§)

Cone biopsy‡ OR extrafascialhysterectomy† (extrafascialhysterectomy OR modifiedradical hysterectomy pluspelvic LN sampling ifpositive margins [mayoffer 6 SLN] OR repeatconization§)





Evidence: high Evidence: high Evidence: high Evidence: highRecommendation: strong Recommendation: strong Recommendation: strong Recommendation: strong

IA1, non-FS (withLVSI)

As above Stage IA1 (with LVSI) andstage IA2: modified radicalhysterectomy

Stage IA1 (with LVSI) andstage IA2: modified radicalhysterectomy (whenpositive margins on repeatcone) plus PLND 6 PANB(pelvic irradiation plusbrachytherapy [with LVSI] ifpatient is not eligible forsurgery)

Stage IA1 (with LVSI) andstage IA2: modified radicalhysterectomyplusPLND6PANB (may offer6SLNORpelvic irradiation plusbrachytherapy [if patient isnot eligible for surgery])





Evidence: low Evidence: low Evidence: intermediate Evidence: intermediateRecommendation: weak Recommendation: weak Recommendation: moderate Recommendation: moderate

IA2, FS Cone biopsy (if follow-uppossible)

Cone biopsy (if follow-uppossible)

Cone biopsy plus PLND 6

para-aortic LN sampling‡Cone biopsy plus PLND 6para-aortic LN sampling‡





Evidence: low Evidence: low Evidence: low Evidence: lowRecommendation: weak Recommendation: weak Recommendation: weak Recommendation: weak




Table 3 – Recommendations for Stage IA, IB, and IIA Disease (Continued)

Type of Disease

Setting


Radical trachelectomy plusPLND

Radical trachelectomy plusPLND




IA2, non-FS Cone biopsy (if follow-uppossible) or extrafascialhysterectomy (non-FS)

Cone biopsy plus PLND 6para-aortic LN sampling‡

Cone biopsy plus PLND 6para-aortic LN sampling‡

See above




Evidence: low Evidence: low Evidence: lowRecommendation: weak Recommendation: weak Recommendation: weak

Extrafascial hysterectomy Modified radical hysterec-tomy plus PLND 6 para-aortic LN sampling§

Modified radical hysterec-tomy plus PLND 6 para-aortic LN sampling§

Modified radical hysterec-tomy plus PLND 6 para-aortic LN sampling§





Evidence: low Evidence: intermediate Evidence: intermediate Evidence: intermediateRecommendation: weak Recommendation: moderate Recommendation: moderate Recommendation: moderate

OR pelvic RT andbrachytherapy

OR pelvic RT andbrachytherapy




IB1, FS No recommendation No recommendation Radical trachelectomy plusPLND (if adding trachelec-tomy . 2 cm)

Radical trachelectomy pluspelvic LN sampling; mayoffer SLN

Adjuvant therapy may beneeded for patients withtumors . 2 cm with riskfactors (see AppendixTable A4)37




IB1, non-FS Extrafascial hysterectomy Radical hysterectomyplus PLND or radical hys-terectomy (see Note) withadjuvant RT or RTwith concurrent low-dosechemotherapy (concurrentchemoRT) if needed

Radical hysterectomy plusPLND

Radical hysterectomy plusPLND; may offer SLN





Evidence: insufficient Evidence: high Evidence: high Evidence: high (SLN option,low)

Recommendation: weak Recommendation: moderateto strong

Recommendation: strong Recommendation: strong(weak)

NACT if available, thenextrafascial hysterectomy

ChemoRT or RT followed byextrafascial or radicalhysterectomy (see Note) 6PLND 6 PANB¶

Pelvic RT plus brachytherapyplus concurrent low-doseplatinum-basedchemotherapy

Pelvic RT plus brachytherapyplus concurrent low-doseplatinum-basedchemotherapy





Type of Disease

Setting



If no RT is available butchemotherapy is available,NACTmay be used to shrinkthe tumor to make itremovable by surgery(extrafascial or modifiedradical hysterectomy [seeNote] 6 PLND 6 PANB¶)



Evidence: insufficient If the patient’s tumor does notshrinkand isnot resectablewith negative margins,palliative measures,including best supportivecare, 6 chemotherapyshould be offered

Evidence: high Evidence: high

Recommendation: weak Type of recommendation:evidence and consensusbased

Recommendation: strong Recommendation: strong

Note Evidence: lowRecommendation: weak

Wherever radical hysterectomywith concurrent chemoRTis listed as a surgicaloption above, extrafascialhysterectomy is recommendedif there is residual diseaseafter RT or chemoRT witha boost of 68 Gy or initialtumor > 6 cm

Radical hysterectomy may beused after RT or chemoRT toa dose of 50 Gy

IB2 and IIA2 If chemotherapy is available,use NACT followed byextrafascial hysterectomy;if chemotherapy isnot available, extrafascialhysterectomy (modificationas deemed necessary) maybe performed if the surgicalcapacity is present

If chemotherapy is available,NACT followed by radicalhysterectomy (see Note)plus PLND6 para-aortic LNsampling may be anoption§k

PelvicRTplus concurrent low-dose platinum-based che-motherapy plusbrachytherapy

Pelvic RT plus concurrentlow-dose platinum-basedchemotherapy plusbrachytherapy


Type of recommendation:evidence



Evidence: low Evidence: intermediate Evidence: high Evidence: highRecommendation: weak Recommendation: moderate Recommendation: strong Recommendation: strong

If EBRT is available, but notbrachytherapy, thenchemoRTfollowed by extrafascialhysterectomy or RT (ifchemotherapy not available)followed by extrafascialhysterectomy (see Note)

Pelvic RT plus concurrentlow-dose platinum-basedchemotherapy plusbrachytherapy plus adjuvanthysterectomy; adjuvanthysterectomy is notrecommended except ifevidence of residualdisease

Pelvic RT plus concurrentlow-dose platinum-basedchemotherapy plusbrachytherapy plusadjuvant hysterectomy;adjuvant hysterectomy isnot recommended exceptif evidence of residualdisease








Type of Disease

Setting


Evidence: low Evidence: intermediate Evidence: intermediateRecommendation: weak Recommendation: weak Recommendation: weakOR if no EBRT is available, thenbrachytherapyandconcurrentlow-dose platinum-basedchemotherapy followed byradical hysterectomy (seeNote)k

When brachytherapy is notavailable, extrafascial orradical hysterectomy isrecommended only whenthere is persistent centralpelvic disease and selectivelymphadenectomy or LNbiopsy for suspicious lesions


Evidence: low to intermediateRecommendation: weak tomoderate

Radical hysterectomy plusPLND 6 para-aortic LNsampling

Radical hysterectomy plusPLND 6 para-aortic LNsampling‡ and adjuvant RTor chemoRT if needed

Radical hysterectomy plusPLND6 para-aortic LNsampling and adjuvant RTor chemoRT if needed (plusRT6 concurrent low-doseplatinum-based chemotherapyafter hysterectomy if riskfactors)‡




Evidence: low Evidence: low Evidence: lowRecommendation: weak Recommendation: weak Recommendation: weak

Note With risk factors on pathologyspecimen: adjuvantchemotherapy afterhysterectomy (Sedlis et al37

criteriausedinUnitedStates)

With risk factors on pathologyspecimen: adjuvant RT 6chemotherapy afterhysterectomy

With risk factors on pathologyspecimen: adjuvant RT 6concurrent low-dose plati-num-based chemotherapyafter hysterectomy

With risk factors on pathologyspecimen: adjuvant RT 6concurrent low-doseplatinum-based chemo-therapy after hysterectomy


Adjuvant RT (intermediaterisk) or with concurrentlow-dose platinum-basedchemotherapy (high risk) ina referral center



Evidence: insufficient Wherever radical hysterectomywith concurrentchemoRT listedasasurgicaloption above, extrafascialhysterectomy is recom-mended if there is residualdisease after RT or chemoRTwith a boost of 68Gyor initialtumor > 6 cm

Evidence: intermediate Evidence: intermediateRecommendation: weak Recommendation: moderate Recommendation: moderate

Radical hysterectomy may beused after RT or chemoRT toa dose of 50 Gy




but supporting information to supplement formalconsensus.

The identified guidelines were published be-tween 2012 and 2015; however, only one setconsidered resources (from NCCN). The CCOguidelines were systematic review–based guide-lines; the WHO guideline had the largest globalconstituency, andall the selectedguidelinesweredeveloped in maximal resource–level settings.Appendix Table A1 lists links to these guidelines,and theData Supplement provides an overview ofthese guidelines, including information on theirclinical questions, target populations, developmentmethodology, and key evidence.

ASCO METHODOLOGIC REVIEW

The methodologic review of the guidelines wascompleted by two ASCO guideline staff membersusing the Rigour of Development subscale of theAGREE II instrument. The score for the Rigour ofDevelopment domain is calculated by summingthe scores across individual items in the domainand standardizing the total score as a proportionof the maximum possible score. Detailed resultsof the scoring and the AGREE II assessmentprocess for this guideline are available in theMethodology Supplement.

Searches for evidence on specific aspects (eg,radiation therapy use, treatment in developingcountries, conization for small tumors, andlymph node biopsy at the time of hysterec-tomy performed after chemoradiotherapy) wereconducted by ASCO guideline staff to identifyrelevant randomized clinical trials, systematic

reviews, meta-analyses, and guidelines to supple-ment the original search. The English-languageandother inclusion criteria listed under LiteratureSearch were applied to the search results.

FINAL RECOMMENDATIONS

The recommendations were developed by a mul-tinational, multidisciplinary group of experts usingevidence from existing guidelines, supplemen-tary literature, and clinical experience as a guide.All recommendations gained formal consensus.The ASCO Expert Panel underscores that healthcare practitioners who implement the recommen-dations presented in this guideline should firstidentify the available resources in their localand referral facilities and endeavor to providethe highest level of care possible with thoseresources. All providers should provide mini-mum supportive and palliative care to all pa-tients. Complete recommendations with ratingsof evidence and strength of recommendationsare listed in Tables 2-6. Recommendations la-beled as evidence based are supported by exist-ing guidelines.

Work-Up

The purpose of work-up is to assess a patient’soverall health status and gather data to informtreatment. Modalities include history and physi-cal examination, biopsies, blood tests, and imag-ing. Tests available in maximal settings, such asmagnetic resonance imaging or positron emissiontomography (PET) –computed tomography (CT),are optional.


Type of Disease

Setting



Evidence: lowRecommendation: weak

IIA1 See IB1 See IB1 See IB1 See IB1

IIA2 See IB2 See IB2 See IB2 See IB2

NOTE. Bold indicates addition of a recommended action over a previous resource level (eg, in limited setting, a bold action is one that was not recommended in basic).Abbreviations: chemoRT, chemotherapy plus radiotherapy; EBRT, external-beam radiation therapy; FS, fertility sparing; LN, lymph node; LND, lymph node dissection; LVSI,lymphovascular space invasion; NACT, neoadjuvant chemotherapy; PANB, para-aortic node biopsy; PLND, pelvic lymph node dissection; RT, radiotherapy.*This option in basic level only if follow-up is available.†For negative margins or operable tumor or positive margins for dysplasia or carcinoma.‡For negative margins or inoperable tumor.§Margins for dysplasia or carcinoma.¶Selective lymphadenectomy or LN biopsy for suspicious lesions.kRecommended in setting where chemotherapy is not consistently available.



Table 4 – Recommendations for Stage IIB, III, IVA, and IVB and Recurrent Disease

Type ofDisease

Setting


IIBand IIIA NACT followedby extrafascialhysterectomy (modifica-tion as deemed necessary)

ChemoRT or RT* followed byextrafascial or modifiedhysterectomy 6 PLND† 6

PANB



NACT followed by extrafascialormodifiedhysterectomy6PLND† 6 PANB*

Adjuvant hysterectomy is anoption only if residual dis-ease after chemoRT

Adjuvant hysterectomy is anoption only if residual dis-ease after chemoRT





Evidence: insufficient Evidence: low to intermediate Evidence: high Evidence: highRecommendation: weak Recommendation: weak to

moderateRecommendation: strong Recommendation: strong

Extrafascial hysterectomywhen chemotherapy is notconsistently available

Extrafascial or modified hys-terectomy plus pelvic LND6para-aortic LN sampling‡plus adjuvant therapy



Evidence: insufficient Evidence: insufficientRecommendation: weak Recommendation: weakPalliative careType of recommendation:

consensus basedEvidence: intermediateRecommendation: strong

IIIB to IVA Palliative care ChemoRT or RT* followed byextrafascial or radical hys-terectomy (see Note) 6PLND† 6 PANB

Pelvic RT plus brachytherapyplus concurrent low-doseplatinum-based chemo-therapy (in some casesextended-field RT)

Pelvic RT plus brachytherapyplus concurrent low-doseplatinum-based chemo-therapy (in some casesextended-field RT)

NACT (followed by radicalhysterectomyplusPLND†6PANB may be an option)and/or palliative care

AND/OR palliative care AND/OR palliative care (op-tions before palliative carealone include: RT boost,salvage surgery, orchemotherapy)




Type of recommendation: evi-dence and consensus based

Evidence: intermediate Evidence: low to intermediate Evidence: high Evidence: highRecommendation: strong Recommendation: weak to

moderateRecommendation: strong Recommendation: strong

NACT followedby extrafascialhysterectomy

RT 6 concurrent low-doseplatinum-based chemo-therapy (may offer systemicadjuvant chemotherapy)

RT 1 brachytherapy 6 con-current low-doseplatinum-based chemotherapy (mayoffer systemic adjuvantchemotherapy)

RT 1 brachytherapy 6 con-current low-doseplatinum-based chemotherapy (mayoffer systemic adjuvantchemotherapy)





Evidence: insufficient Evidence: intermediate Evidence: intermediate Evidence: intermediateRecommendation: weak Recommendation: moderate Recommendation: weak Recommendation: weak

Note Wherever radical hysterec-tomy with concurrent che-moRT listed as a surgicaloption above, extrafascialhysterectomy is preferred ifthere is residual disease orinitial tumor > 6 cm


Evidence: intermediateRecommendation: weak




Table 4 – Recommendations for Stage IIB, III, IVA, and IVB and Recurrent Disease (Continued)

Type ofDisease

Setting


IVB Palliative care and chemo-therapy (if available)

Palliative care and/or chemo-therapy 6 individualizedRT (palliative care may in-clude palliative RT)

Chemotherapy 6 individual-ized RT AND/OR palliativecare

Chemotherapy 6 bevacizu-mab 6 individualized RTAND/OR palliative care





Evidence: high Evidence: high Evidence: high Evidence: highRecommendation: strong Recommendation: strong Recommendation: strong Recommendation: strong

Recurrent Palliative care Depending on previous RT andeither “no prior RT or failureoutside of previouslytreated field”16(CERV-11)

then may offer tumor-di-rected RT plus platinum-based chemotherapy

Depending on previous RT andcentral v noncentraldisease:

Depending on previous RT andcentral v noncentraldisease:



Central disease: chemoRT orRT 6 brachytherapy if noprior RT

Central disease: chemoRT orRT 6 brachytherapy if noprior RT

Evidence: high Evidence: high If central and prior RT:exenteration

If central and prior RT:exenteration

Recommendation: strong Recommendation: strong Noncentral: chemotherapy,tumor-directed RT, andpalliative care

Noncentral: chemotherapy,tumor-directed RT, andpalliative care



Evidence: high Evidence: highRecommendation: strong Recommendation: strong

AND/OR central disease:chemotherapy

Prior RT plus central disease:pelvic exenteration ORradical hysterectomy ORbrachytherapy [latter two“in carefully selected pa-tients with small (< 2 cm)lesions”15 (CERV-11)]

Prior RT plus central disease:pelvic exenteration 6intraoperative RT OR radi-cal hysterectomy OR bra-chytherapy [latter two “incarefully selected patientswith small (, 2 cm) le-sions”15 (CERV-11)]




Evidence: insufficient Evidence: high Evidence: highRecommendation: weak Recommendation: strong Recommendation: strong

Note This is best managed withexenteration (type of sur-gery that is not feasible toperform in low-resourcesetting)

Prior RT plus noncentral dis-ease: chemotherapy orbest palliative care

Prior RT plus noncentral dis-ease: tumor-directed RT 6

chemotherapy or best pal-liative care

Prior RT plus noncentral dis-ease: tumor-directed RT6chemotherapy OR resec-tion with intraoperative RTfor close or positive mar-gins OR clinical trial ORchemotherapy plus bevaci-zumab AND/OR palliativecare




Evidence: high Evidence: high Evidence: highRecommendation: strong Recommendation: strong Recommendation: strong




Treatment

The treatment of women with invasive cervicalcancer consists of surgery, chemotherapy, andradiation therapy, sometimes in combination. Therecommendations for women with stage IA1 to IVdisease and women with recurrence are listed inTables 3-5, and selected treatment recommenda-tions are discussed in the following sections.

DISCUSSION OF SELECTED TREATMENT ISSUES

Early-Stage and Locally Advanced Disease

Three sets of treatment guidelines from maximalsettings recommended radical hysterectomy, pel-vic lymphadenectomy,andpara-aortic lymphnodesampling8 or concurrent chemoradiotherapy6,7

for women with early-stage or locally advanceddisease (Table3). Therewasno literature to informpractice in the basic setting. Therefore, the panelchose to make consensus recommendations andrelied on clinical experience, training, and judg-ment to formulate these recommendations, giventhat there were no conclusive data regarding thisquestion.

In contrast to other guidelines, this ASCOresource-stratified guideline addresses settingswhere there are no or limited personnel withsufficient surgical expertise to perform radicalhysterectomies for patients with cervical cancer.For a patient who has early-stage disease (stageIA2, IB1, or IIA1), if the surgeon can remove the

tumor safely, with negative margins, the ExpertPanel recommends performing extrafascial hys-terectomy. In basic settings where surgery ca-pacity is absent or lacking, and patients cannotbe treated with radiation therapy, extrafascialhysterectomy may be offered to women with stageIBcervical cancer. Forwomenwith stage IA2or IB1disease with tumors smaller than 2 cm in size and1 cm in depth in the non–fertility-sparing setting,studies are ongoing exploring whether cone biopsyor extrafascial hysterectomy with pelvic lymphade-nectomyisadequate.Forwomenwith larger tumorsor advanced-stage cervical cancer, the ExpertPanel recommends neoadjuvant chemotherapy(NACT) whenever chemotherapy is available, forthe purpose of shrinking the tumor before per-forming hysterectomy. The specific chemotherapyagents may be carboplatin, cisplatin, or paclitaxelplus carboplatin. There are two randomized phaseIII trials (EORTC [European Organisation for Re-search and Treatment of Cancer] 55994 andClinicalTrials.gov identifier NCT00193739) com-paringNACT followed by surgery with chemoradio-therapy in these patients. The EORTC 55994 trialwas recently closed for accrual, whereas theNCT00193739 trial is ongoing (more informationis provided in Future Directions). Extrafascialhysterectomymaybeused for patientswith stageIB2 or IIA2 to IIIA disease after NACT.

For women with locally advanced cervical cancer,the role of surgery has been debated for many

Table 4 – Recommendations for Stage IIB, III, IVA, and IVB and Recurrent Disease (Continued)

Type ofDisease

Setting


Note Before palliative care alone,try options such as RTboost, salvage surgery, orchemotherapy

If recurrence after any of theabove, then clinical trial ORchemotherapy OR bestsupportive care


Evidence: highRecommendation: strong

NOTE. Bold indicates addition of a recommended action over a previous resource level (eg, in limited setting, a bold action is one that was not recommended in basic). Additionalrecommendations regarding settings with limited radiotherapy resources are provided in the main article.Abbreviations: chemoRT, chemotherapy plus radiotherapy; LN, lymph node; LND, lymph node dissection; NACT, neoadjuvant chemotherapy; PANB, para-aortic node biopsy;PLND, pelvic lymph node dissection; RT, radiotherapy.*Recommended in setting where chemotherapy is not consistently available.†Whenbrachytherapy is not available, extrafascial or radical hysterectomy is recommended only when there is persistent central pelvic disease and selective lymphadenectomy orLN biopsy for suspicious lesions.

‡Margins for dysplasia or carcinoma.


http://ClinicalTrials.gov


years. Studies with other comparisons involvinghysterectomy after chemoradiotherapy or radio-therapy (with or without brachytherapy) were an-alyzed in a 2015 Cochrane systematic review,which found little benefit for most women.26

Concurrent radiotherapyandchemotherapy isstan-dard in enhanced andmaximal settings for womenwith stage IB to IVA disease (Tables 3 and 4), asstatedby theNational Cancer Institute; this is basedon sevenclinical trials.38However, the Expert Panelis aware of some negative trials on concurrentchemoradiotherapy regimens. For example, tworandomized controlled trials were conducted, nei-ther of which found an overall survival (OS) differ-ence. One of these found a statistically significantlyimproved disease-free interval.39,40 Therefore,the panel stresses the addition of low-dose

chemotherapy during radiotherapy, but not at thecostofdelayingradiation therapy if chemotherapy isnot available.

In limited-resource settings, where brachytherapyis not available, the panelists considered a phaseIII randomized trial involving participants withstage IB2 to IIB cervical cancer, conducted inMexico, to evaluate if radical hysterectomy andpelvic lymphadenectomy performed after concur-rent chemoradiotherapy was equivalent to treat-ment with brachytherapy.41 The Cochrane reviewincluded this trial, which demonstrated equivalentsurvival and progression-free survival outcomesfor women who underwent radical hysterectomycompared with women who were treated withbrachytherapy. In that study, 72% (62 of 86) ofthe patients receiving surgery alone had pathologic

Table 5 – Chemotherapy Regimens for Stage IV or Recurrent Disease

Setting


Single-agent platinum-based ther-apy (cisplatin or carboplatin)

Cisplatin or carboplatin, cisplatinplus paclitaxel, or carboplatin pluspaclitaxel

Cisplatin plus paclitaxel or carbo-platinpluspaclitaxel (highest-levelevidence for cisplatin: CCO4)

Cisplatin plus paclitaxel plus beva-cizumab or carboplatin plus pac-litaxel plus bevacizumab





Evidence: intermediate Evidence: high Evidence: high Evidence: high

Recommendation: moderate Recommendation: moderate tostrong

Recommendation: strong Recommendation: strong

Abbreviation: CCO, Cancer Care Ontario.

Table 6 – Follow-Up

Options for Follow-Up for All Settings

Follow-up should be based on each individual’s risk of cervical cancer recurrence; high-quality evidence is lacking on the bestmethods of post-treatmentsurveillance; some guidance is offered in other guidelines5 and is provided here as guidance rather than as recommendations:

After 1 to 2 years, every 3 to 6 months

After 3 to 5 years, every 6 to 12 months

After > 5 years, every year based on risk of recurrence

Type of recommendation: consensus based

Evidence: insufficient

Recommendation: weak

Pelvic and physical examination

Imaging and laboratory tests based on symptoms or suspicion

Patient education

Cytologymay be offered, if available, every 3 years after cone biopsy, radical hysterectomy, or trachelectomy; cytology should not be performed after RT

In patients at high risk for locoregional failure, PET-CT 3 months after therapy is optional

Type of recommendation: consensus based

Evidence: insufficient

Recommendation: weak

Abbreviations: CT, computed tomography; PET, positron emission tomography; RT, radiotherapy.



Table7–Overviewof

Selected

Stud

ies:RTSc

hemean

dOutco

meAsPropo

rtionof

Wom

enWith

theCom

plaint

FirstAutho

ran

dYe

arof

Pub

lication

Design

RT

No.

ofPatients

Propo

rtionof

Wom

enWith

Com

plaintsTh

atSho

wed

Partia

l(>50%)or

Com

plete

Improvem

ent

Toxicity

Dosepe

rFrac

tion

(Gy)

No.

ofFrac

tions

Bleed

ing

Pain

Disch

arge

No.

%No.

%No.

%

Bou

lware,

1979

Observationa

lretrospective

101

86a

39of

8645

14of 31

428of

86ac

ute

102b

5547

of55

8513

of 2259

15of

86late

103b

2020

of20

100

5of

863

Hod

son,

1983

Observationa

lretrospective

103c

149of

910

03of

310

01of

110

02of

14late

Halle,1

986

Observationa

lretrospective

101-3d

4227

of30

904of

92of

30ac

ute;

5of

42severe

late

Onsrud,

2001

Observationa

lretrospective

101

2819

of24

790of

30

4of

1136

GI:grad

e1-2,

27of

64;

grad

e2-4,

5of

6410

2c24

103c

1

Mishra,

2005

eObservationa

lretrospective

101

100

50of

6774

23of 48

4746

of 6966

10pa

tientsha

dgrad

e3-4toxicity

102c

6180

5947

103c

3310

050

34

Alternativefrac

tiona

tion

Patric

io,1

987f

Observationa

l6,

52g

,h56

33of

3594

5of

1145

16%

serio

usco

mplications

Span

os,1

996f

Subg

roup

analysisof

aprospe

ctive

trial

3,7

4i61

37of

4976

11of 36

313%

acutetoxicity;

7%lateco

mplications

Grig

sby,20

02f

Observationa

l5

2j15

14of

1593

Nosevere

orac

ute

toxicity

NOTE

.Dataad

aptedwith

perm

ission

from

vanLonk

huijzen

etal.33

Abb

reviation:

RT,

radiationtherap

y.a 77cervix,sixvagina

,three

cxstum

p.b3-

to4-weekinterval.

c 4-weekinterval.

dAt4

weeks’interval

orrecu

rren

ce.

e .50

%respon

seisrepo

rted

.f Alternativefractiona

tion.

g 48ho

urs.

hFlashdo

sefollowed

bysecond

flash

dose

orconven

tiona

lfractiona

tion;

results

areafterfirstdo

se.

i 48ho

urs,repe

ated

after2-4weeks

forup

tothreecycles.

j 1week,

brachytherap

y.



complete response, with no residual tumor in theradical hysterectomy specimen.26 Because of theconcern of the risks associated with performingradical hysterectomy in limited settings, the ASCOExpert Panel recommends extrafascial hysterec-tomy or its modification for women who have re-sidual tumor 2 to 3 months after concurrentchemoradiotherapy and additional boost in placeof brachytherapy. For some women without theserisks factors, the potential harms of surgery mayoutweigh incremental potential risk factors (infor-mal consensus based on subset analyses of pa-tients with bulky stage IB disease).42 If women donot receive surgery, they should be closely moni-tored; however, thepanel acknowledges the seriousissue of loss to follow-up, especially when womenlack access to transportation and/or face other highpotential opportunity costs to attend follow-up.

The Expert Panel also discussed adjuvant chemo-therapy after chemoradiotherapy. The phase III,open-label, randomized study comparing concur-rent gemcitabine plus cisplatin and radiotherapyfollowed by adjuvant gemcitabine and cisplatinversus concurrent cisplatin and radiotherapy inpatients with stage IIB to IVA carcinoma of thecervix by Due~nas-Gonzalez et al43 was the firstadjuvant chemotherapy study. In this study, therewas improved efficacy in the adjuvant therapy armbut also more adverse events (both statisticallysignificant). The primary outcome was 3-yearprogression-free survival, which was 74.4%(95% CI, 68% to 79.8%) in the interventionarm versus 65% (95% CI, 58.5% to 70.7%;P5 .029). There were more grade 3 to 4 adverseevents in intervention arm (eg, neutropenia), andthere were two potentially study-related deaths.There is one ongoing trial examining differ-ent adjuvant chemotherapy agents: OUTBACK,a phase III trial of adjuvant chemotherapy afterchemoradiotherapy as primary treatment of lo-cally advanced cervical cancer compared withchemoradiotherapy alone (ClinicalTrials.govidentifier NCT01414608). This trial is investi-gating the addition of adjuvant carboplatinand paclitaxel after chemoradiotherapy versusconcurrent cisplatin and radiotherapy in pa-tients with stage IB to IVA carcinoma of thecervix. An additional phase III trial of concur-rent chemotherapy and pelvic irradiation withor without adjuvant carboplatin and pacli-taxel is open for high-risk patients with stageIA2, IB, or IIA cervical carcinoma after ra-dical hysterectomy (RTOG [Radiation Ther-apy Oncology Group] 0724/GOG [Gynecologic

Oncology Group] 0724; ClinicalTrials.gov identifierNCT00980954).

Management of women who have early cervicalcancer and desire fertility preservation requirescareful counseling, because radical trachelec-tomy requires surgical expertisenot usually availablein basic or limited settings. Although prospectiveclinical trials to evaluate the outcomes of conebiopsy and pelvic lymphadenectomy for womenwith stage IB1 cervical cancer with tumor sizesmaller than 2 cm are ongoing, retrospective stud-ies have suggested equivalent outcomes can beachieved with the less aggressive approach. Iffollow-up is available, the Expert Panel recom-mends cone biopsy for women with stage IA2disease in basic settings and cone biopsy pluspelvic lymphadenectomy for those in limited set-tings. In enhanced and maximal settings, radicaltrachelectomy is recommended for patients withstage IB1 cervical cancer with tumor size up to2 cm. Because the risk of metastasis increaseswith tumor size between 2 and 4 cm,44,45 radicaltrachelectomy and pelvic lymphadenectomy havebeen performed after NACT in the research setting,but there is no conclusive evidence to support theirroutineuse.Completehysterectomymaybeusedinpatients with stage IB diseasewith high risk factors,including positive margins and lymph nodemetas-tasis.46 (The presence of lymphovascular spaceinvasion [LVSI] is one of the three highest riskfactors for disease recurrence. Radical trachelec-tomy may be considered if only one of these riskfactors is present. See definitions of Sedlis et al37

regarding risk factors under Definition.)

The Expert Panel recommends pelvic lymphade-nectomy in the enhancedandmaximal settings forthe purpose of fertility preservation; this is basedon the current evidence for patients with stage IA1to IA2 disease and positive LVSI. Fertility sparing isrecommended as an option clinicians should offerto women in their reproductive years (up to approx-imately 45 years of age) andwith a desire for fertilitypreservation. The ASCO guideline on fertility pres-ervation discusses these issues in depth.47 Thatguideline states for conservative gynecologicsurgery, “it has been suggested that radicaltrachelectomy (surgical removal of the uterinecervix) should be restricted to stage IA2 to IBcervical cancer with diameter , 2 cm and in-vasion , 10 mm.”47(p2503) For the purposes ofthe current guideline, the authors suggest thatfor women with stage IA1 to IA2 disease whoare approximately 45 years of age and have adesire for fertility preservation, radical trache-lectomy and pelvic lymphadenectomy may be





an option in enhanced and maximal settings.Three ongoing clinical trials will help determineif cone biopsy or extrafascial hysterectomy andpelvic lymphadenectomy may have equiva-lent outcomes for these patients (see FutureDirections).

Late-Stage or Advanced Disease

For patients with stage IV or recurrent cervicalcancer, single-agent chemotherapy (carboplatinor cisplatin) is recommended in basic settings.The purpose is to minimize the increased toxic-ity associated with the doublet chemotherapyregimen or to offer treatment when doublet che-motherapy is not readily available and/or thereis limited capacity for managing adverse effectswomen experience from chemotherapy (Table 4).On the basis of the recent report by the JSGO, inwhich carboplatin plus paclitaxel was not found tobe inferior to a cisplatin plus paclitaxel combination,the panel recommends carboplatin as the first-linechemotherapeutic agent; it is associated with afavorable therapeutic index.48 In the CochraneDatabase Systematic Review in 2012, single-agent cisplatin resulted in a lower response rate(relative risk [RR], 0.60; 95%CI, 0.44 to0.81)butwas associated with less toxicity when comparedwith a platinum-based combination, including lessserious (grade 3 to 4) neutropenia (RR, 0.04; 95%CI, 0.02 to 0.12) and thrombocytopenia (RR, 0.16;95%CI, 0.05 to 0.48). Overall survival was longer inthe combination versus single-agent group (13 to 15v 7 to 9 months). In settings where resources arelimited, single-agent cisplatin or carboplatin is areasonable alternative.32 The GOG 240 studyreported a 3.7-month survival benefit with theaddition of bevacizumab to the combinationchemotherapy regimen in patients with ad-vanced or recurrent cervical cancer.49 The Ex-pert Panel recommends the inclusion of thisagent with carboplatin or cisplatin and paclitaxelcombination in maximal settings, with carefulattention to potential toxicities of this regimen. Inaddition, this guideline contains suggestions onmodifications to radiation therapy protocols inlimited-resource settings; please see RadiationTherapy in Resource-Constrained Settings.

Lymphadenectomy

Throughout the treatment recommendations, theExpert Panel primarily recommends para-aorticlymph node sampling rather than complete lym-phadenectomy; however, if the clinician findsenlarged nodes, he or shemay remove those. This

is performed for prognostic purposes anddoes nothave therapeutic implications.

RADIATION THERAPY IN RESOURCE-CONSTRAINEDSETTINGS

Radiation therapy is a mainstay of cervical can-cer treatment. For women with advanced cervi-cal cancer in maximal resource–level settings(>bulky stage IB), treatment involves radiother-apywithconcurrentplatinum-basedchemotherapy.This radiotherapy includes both external-beamradiation therapy (EBRT) andbrachytherapy.How-ever, radiation therapy is limited or nonexistentin some settings, and there is little evidence toguide clinicians in these situations. In settingswith more limited resources, the most appropri-ate management must depend on the availabletreatments, including surgical and radiationoncology expertise, available chemotherapyagents, EBRT and brachytherapy equipment,and trained physics and dosimetry personnel.De novo expert consensus–based recommenda-tions on alternative management of patients withcervical cancer in settings with no or limitedradiation oncology resources are provided asfollows.

Clinical Question R1

Which alternative treatment options can be offeredwhen there is a lack of radiation therapymachinesor physics supports?

Recommendation R1A: Basic setting. As listed inthe recommendations table (Table 3), for patientswith stage IA cervical cancer, qualified surgeonsmay perform extrafascial hysterectomy (withoutLVSI). For larger tumors (stage IB1 to IIA1), if thereis surgical expertise, extrafascial hysterectomy oritsmodificationmaybe performed if the tumor canbe removed completely without cutting throughgrossdisease. If chemotherapy is availableand thepatient’s tumor is too large to remove completely,NACT may be used to shrink the tumor to make itsurgically removable. If the patient’s tumor doesnot shrink and is not resectable with negativemargins, palliative measures, including best sup-portive care, should be used. (This recommenda-tion concurs with the other recommendations[Table 3].)

Recommendation R1B: Limited-resource set-tings, with limited EBRT and no brachytherapyavailable. If chemotherapy is available and thepatient’s tumor is too large to remove completely(> stage IB2), NACT may be used to shrink thetumor to make it removable by surgery. (Evidence



quality: intermediate. Strength of recommenda-tion: moderate. Qualifying statement: There areongoing trials regarding NACT, and on publication,the Expert Panel may reconsider this recommen-dation.) Additional adjuvant therapies may poten-tiallybeused if risk factorsexist on thehysterectomyspecimen. If EBRT is available, up to50Gy externalradiation followedbyanadditional radiationboost of18 to 20 Gy (portal size, 10 3 10 or 12 3 12 cm)may be used. (Evidence quality: intermediate.Strength of recommendation: moderate.) Extrafas-cial hysterectomy or its modification may be per-formed if there is limited residual tumor on thecervix. (Evidence quality: low. Strength of recom-mendation:weak.) If the tumordoesnot shrinkafterNACT or radiotherapy, palliative measures, includ-ingbest supportivecare, shouldbeused. (Evidencequality: intermediate. Strengthof recommendation:strong. Qualifying statement: If a patient has a largeinoperable tumor, chemotherapy is available, and itis possible to perform surgery if the tumor shrinks,NACT may be used. If NACT does not shrink thetumor, radical radiation therapy may be used. Ifafter this the tumor does not shrink and/or there ismore than limited residual disease and the tumor isinoperable,palliativemeasures, includingbest sup-portive care, should be used.)

Discussion of R1.With no radiotherapy available insome settings, treatment must rely on other avail-able resources. The Expert Panel is in favor ofNACT followed by surgery compared with surgeryfollowed by chemotherapy. In settings where re-sources are limited, risks associated with surgeryin locally advanced cervical cancer may be min-imizedwith the administration ofNACT.NACTcanfrequently shrink tumors, but hysterectomy shouldbe performed only if negative surgery marginscan be achieved. The evidence for NACT comesprimarily from retrospective studies. There aretwo ongoing randomized controlled trials comparingNACT plus surgery with chemoradiotherapy, whichmay inform recommendations regarding NACT inthe future.


Which alternative treatment options should beoffered where there is limited availability of radio-therapy resources? What is the reasonable mini-mal fractionation?

Recommendation R2A: Limited setting. Where thenumber of patients requiring radiotherapy over-whelms the available EBRT resources, providersmay have to triage patients. Thiswould include the

whole population andmay involve using resourcesfor those with higher likelihood of potential cure.Radiotherapy could possibly be used for palliativetreatment in those least likely to be cured. Imple-mentation of these suggestions is dependent onthe setting and local conditions. Shorter radiationfractionation schemes with curative intent may beused where resources are constrained. (Evidencequality: intermediate. Strength of recommenda-tion: moderate.)

If surgical expertise is available and there is limitedradiotherapy for those with resectable disease,clinicians may potentially perform surgery afterNACT.Typeof surgery is listedby stageandsettingin Table 3 (eg, extrafascial hysterectomy or mod-ified radical hysterectomy plus pelvic lymph noderesection if there are positive margins). (Evidencequality: weak. Strength of recommendation: low.)

Recommendation R2B: When brachytherapy isnot available or there is residual tumor afterradiotherapy. Ideally, if EBRT and sufficient re-sources are available and the decision has beenmade to treat a patient with curative intent, radicalirradiation should include an initial whole-pelvicdose of 45 to 50 Gy of EBRT, using 1.8- or 2.0-Gyfractions. If brachytherapy is not available, clini-cians should use an external-beam boost up to adose of 68 to 70 Gy. (Evidence quality: intermedi-ate. Strength of recommendation: moderate.) Thefield volume for the boost should be reduced so thatthe superior border is at the lower sacroiliac joint andthe volume encompassing residual disease.

In a situation of constrained radiotherapy resourcesbut an intent to provide cure, clinicians may usefewer fractions of radiation in a higher dose perfraction. Where this may result in higher complica-tion rates, the initial pelvic volume to be treated canbereduced (eg, thesuperior levelof the fieldmaybeat the inferior S-I joint, because failure ismost likelyto occur in the low pelvis). (Evidence quality: low.Strength of recommendation: moderate.)

If residual central disease persists in the pelvis at2 months after treatment completion and there isno overt distant disease, surgery to remove theresiduum should be an option. (Evidence quality:low. Strength of recommendation: weak. Qualify-ing statement: If the surgeon feels that there is areasonable chanceof resecting residual persistenttumor after an incomplete response to radiationtherapy, an extrafascial hysterectomy could beattempted. Expert opinion is that in these circum-stances, resectability ismore likelywhen the tumorsize is , 4 cm in its greatest dimension.)



Recommendation R2C. If patients have diseasewith a low likelihood of cure, a palliative ap-proach may be used (ie, limited number oflarger-fraction treatments delivered to a smallerfield of irradiation [eg, 3.7 Gy twice per day for2days, repeatedup to3 timesevery2 to4weeks]or a single fraction of 8 to 10 Gy, possibly re-peated once permonthup to 3 times). (Evidencequality: intermediate. Strength of recommenda-tion: moderate.)

Discussion of R2. In a situation with high volumesof cervical cancer and limited numbers of ra-diotherapy treatment machines, the goal shouldbe optimizing the use of resources. For patientswho cannot adequately be treated with surgeryor the combination of NACT and surgery, radio-therapy is the optimal treatment approach. Cu-rative radiotherapy generally consists of 45 to50 Gy of EBRT, using 1.8- or 2.0-Gy fractions.Brachytherapy, which may not be available inlimited-resource settings, is introduced in thethird to fifth week of EBRTor after its completion.Once initiated, every effort should be made tocomplete all treatment in approximately 55days.Treatment breaks should be avoided, becauseprolongation of treatment time decreases curerates. With more patients with cervical cancerto be treated than available radiotherapy ma-chine resources, triaging of patients can beimportant for improving disease control out-comes for patients with potentially curative dis-ease and maximizing symptom control forpatients with advanced disease.

If the resources are available and the patientcannot receive treatment with curative intent,irradiation should be used to relieve symptoms ofpain and bleeding (Table 7).33 A single largefraction or palliative approach may be used, asmay a limited number of large-fraction treatmentsdelivered to a smaller irradiation field. No phase IIIdatahaveestablishedanoptimal fractionation, butavailable data as examples of possible fraction-ation schemes are listed in Table 7.33 The ap-proach used in RTOG 8502 of 3.7 Gy twice per dayfor 2 days and repeated up to 3 times every 2 to4 weeks showed 76% and 31% of patients withcervical cancer experiencing significant improve-ment in bleeding and pain, respectively.50 Althoughit is commonly taught that higher doses of protractedradiation provide better palliation in terms of symp-tom relief and duration of such, randomized datain patients with bladder or lung cancer compar-ing shorter versus longer schedules with highertotal doses do not suggest an advantage.33 There-fore, where resources are constrained, single- or

short-courseschemescanbeusedwithretreatmentsif feasible for persistent or recurrent symptoms.

Clinical Question R3A

What shouldproviders do in situationswhere thereis limited brachytherapy?

Recommendation R3A: Limited setting with limitedbrachytherapy. No optimal dose per fractionationscheme has been defined. Various schemes areused, most commonly three- to five-dose cervicalhigh-dose rate brachytherapy. The dose per frac-tion varies with the number used to achieve bi-ologic equivalence (the combined biologicallyequivalent dose at point A achieved by a combi-nation of EBRT and brachytherapy should ideallyapproximate 75 to 85 Gy).51 (Evidence quality:high. Strength of recommendation: strong.)

Discussion of R3. Ideally, definitive cervical irradi-ation should include brachytherapy; its use showsimproved survival and decreased toxicity, but ifbrachytherapy resources are not available, treat-ment with an external-beam boost as describedin Recommendation R2B should be used. Forresidual disease 2 months after radiotherapy,surgery may be an option as described in Rec-ommendation R2A.20

In data reported from the United States, althoughdefinitive cervical treatment without brachyther-apy shows lower survival rates, it does result inlong-termsurvival in somepatients.On thebasis ofUS SEER data collected between 1988 and 2009,cervical cancer 4-year cause-specific survival isapproximately 13% lower without brachyther-apy (64.3% v 51.5%; P , .001), and 4-yearoverall survival is 12% lower (58.2% v 46.2%;P, .001).52 Among 907 patients with stage IIIBcervical cancer treated with curative intent be-tween 1960 and 1993 at MD Anderson CancerCenter (Houston, TX), 641 were treated withEBRT and brachytherapy, and 266 were treatedwith EBRT only; 5-year disease-specific survivalwas approximately 21% lower without brachy-therapy (45% v 24%).53 Details on the external-beamboost volume to decrease complications sug-gest it would be reasonable to use a smaller pelvicfield tocoverapalpable tumor.Asuggestedsuperiorborder would be mid bony pelvis (this level isapproximately 2 cm distal to the inferior sacroiliacjoints) or at the inferior sacroiliac joint. For lateralizedstage IIBor IIIBdisease, the lateralborderswouldbe2 cm beyond the bony pelvis on plain x-ray. Theinferiorbordershouldbe2cmbelowpalpable tumorin the vagina. The field arrangement may be a four-fieldboxoraparallel pair asasecondchoice, if there



are no resources available for delivering a four-fieldbeam arrangement.


Where there are no simulators, what can be usedwith bony anatomy to target correctly?

RecommendationR4: Settingwithout simulators. Inbasic and limited settings where women do nothave access to facilities with simulators at eitherlocal or referral institutions, plain diagnostic x-raysvisualizing the bony pelvis may be used to guidefield placement. Clinicians in enhanced and maxi-mal settings should not apply this recommendation.(Evidence quality: high. Strength of recommenda-tion: strong.)

Discussion of R4. Where there are no simulatorsfor planning pelvic irradiation, plain diagnosticx-rays visualizing the bony pelvis may be usedto guide field placement. The field arrangement ispreferably a four-field box, particularly for patientswho are obese, but parallel anterior and posteriorfields may be used if necessary. The superior andinferior borders of the field are the same for bothfield arrangements; the superior border may bebetween L4-5 and L5-S1, and the inferior border isplaced 2 cmbelow palpable disease in the vagina.The lateral borders are placed 2 cm beyond thebony pelvis. Where lateral fields are used for thefour-field box, the anterior border should be infront of the pubic arch, and the posterior bordervaries with the extent of disease and may ofteninclude the hollow of the sacrum inferiorly to S4-5.Where there is no means for shielding the supe-rior lateral corners around the potential position oflymph nodes in the common iliac region and thefraction size is higher (. 2 Gy), a lower superiorborder, such as the mid S-I joint, may be used tospare normal small bowel. Clinical experience hasindicated a smaller volume may decrease com-plications, and the additional curative value ofhigher superior borders is doubtful.

Alternatively, where there is no simulator and thepatient is to be treated with a parallel pair ofanterior and posterior fields, with the patient ly-ing supine on the linear accelerator (LINAC) orcobalt-60 (60Co) bed, anterior–posterior separationshould be measured. If both a LINAC and 60Coare available, a patient with a larger separation isbetter treated on the LINAC. To locate the skinmark for the isocenter, 2 cm should be measuredsuperior to the pubic tubercle, and a 16 3 16 or18 3 18 cm field should be placed at the axis.Before treatment, it should be verified on the

treatment unit that the field covers the bony land-marks as described.

POST-TREATMENT FOLLOW-UP

The ASCO Expert Panel concludes that there isinsufficient evidence guiding follow-up in all re-source settings, especially those that are resourceconstrained, including supporting theuseof visualinspectionwith acetic acid, humanpapillomavirusDNA testing, cytology (where tissue samplesand/or laboratories are not available), or imagingfor follow-up in those settings. In addition, theevidence for using squamous cell carcinoma an-tigen and/or high-sensitivity C-reactive protein isinsufficient; prospective studies are needed.Therefore, this guideline does not recommendtheir use. ASCO informally endorses the CCOguideline on follow-up for cervical cancer5 andprovides similar intervals as a guide (Table 6).Pelvic and physical examinations should be per-formed at follow-up. Suspicious lesions should bebiopsied. Most evidence by expert opinion is thatvaginal smear with cytology has been of low value.On the basis of the opinion of the ASCO ExpertPanel, cervicovaginal cytology is not recom-mended for women after radiotherapy. Cytologymay assist the diagnosis of early recurrences incases ofwomen treatedwith radical hysterectomy,with the caveat that the evidence is insufficient.PET-CT scans are often used in surveillance 3 to6 months after treatment of patients at risk fortreatment failure in high-resource settings. It is notwithin the scope of this guideline to reviewthe evidence on post-treatment imaging, andthe CCO guideline does not recommend PET-CT, because of the lack of prospective studies.However, there are indications that post-treatmentPET scans may provide information for prognos-tic purposes in maximal-resource settings. Forexample, a prospective cohort study in which92 women received an [18F]fluorodeoxyglucose-PET scan 3 months after treatment found thatabnormal [18F]fluorodeoxyglucose-PET resultswere statistically significantly associated withlower cause-specific and progression-free sur-vival compared with those in women whosePET scans showed complete response totreatment.54

PALLIATIVE CARE

Palliative care and pain management are part ofthe treatment of cancers, including cervical can-cer. The goal of palliative care is to avoid unnec-essary suffering during the final stages ofdisease. Palliative care is multifaceted and in



some contexts can be provided concurrentlywith tumor-directed therapy. Pain control is avital (but not the sole) component of palliativecare and is a basic human right. This guidelineadapts guidance9 on palliative care provided inthe WHO guideline on women with cervical can-cer, plus other guidelines and additional sourcesreferenced in the Special Commentary underPalliative Care. In its guideline, WHO discussesthe teamapproach.Our guideline encourages allpolicymakers to follow the WHO declaration onpalliative care.55 Women with advanced cervicalcancer with or without access to tumor-directedtherapy may have specific late-stage symptomsthat require clinicians to perform urogenital-specific interventions. The management of pri-mary symptoms, which is discussed in theWHO guideline, includes but is not limited topain, fistula, loss of appetite, wasting, bleed-ing, and GI and genitourinary symptoms. WHOprovides practice sheets for caregivers tohelp women with vaginal discharge, fistula, andbleeding. Further discussion occurs in SpecialCommentary.

SPECIAL COMMENTARY

Radiation Therapy Shortages

There are enormous inequities in the supply ofradiotherapy equipment and personnel be-tween maximal and enhanced settings and lim-ited and basic settings.56 There are manypublications calling attention to this critical sit-uation; therefore, this section will only brieflytouch on it.

In somesettings, only 10%ofpatients haveaccessto radiation therapy; in others, such as in sub-Saharan Africa and Latin America, 0% to 4% ofpatients do, whereas in upper- and middle-income countries in Europe and Central Asia,59% to 79% have access.57 Depending on thesource, 39 to 55 LMICs have no radiation therapyresources.56-58 One publication states that “it hasbeen estimated that Africa is functioning at 25%ofits potential treatable capacity for cervical canceralone.”59,60(p459) Several organizations are workingto increase awareness of this shortage and addresssupply issues, including the International AtomicEnergy Agency, Union for International Cancer Con-trol, Global Task Force on Radiotherapy for CancerControl, and American Society for Radiation Oncol-ogy Association of Residents in Radiation Oncology.This guideline strives to support these effortsand provides guidance to clinicians practicing inresource-constrained settings.

PalliativeCare forWomenWithAdvancedCervicalCancer

This commentary supplements this guideline andis based on additional literature, some of whichwas not included in the formal adaptation reviews.Palliative care and pain management are part ofthe treatment of cancers, including cervical can-cer, to avoid unnecessary suffering during the finalstages of disease. Pain control is a vital componentof palliative care; it is a basic human right oftenneglected in cancer control programs. Pain canbe effectively controlled using a combination ofmedical and nonmedical interventions. Patientsand caregivers need training for effective palliativecare, including end-of-life counseling. Patientswith advanced or recurrent cervical cancer mayhaveanyof the following symptoms: vaginal bleed-ing or discharge, pelvic or back pain, urinaryor bowel fistulas, lower-extremity edema, deep-venous thrombosis, dyspnea resulting fromanemiaor pulmonary involvement, and uremia resultingfrom ureteral obstruction.

In limited-resource settings where radiation ther-apy is limited, providers may have to prioritize itsuse to treat selected patients with advanced-stagedisease and palliate symptoms in other patientswho normally receive antitumor treatment inmaximal-level settings. Interventions to controlvaginal bleeding include radiation therapy or bra-chytherapy, embolization of the uterine arteries,surgical resection, and arterial ligation. Vaginalpacking is usually a temporary measure. The typeand length of treatment should depend on thepatient’s Karnofsky performance status.

Pain is often a disabling symptom of advanced orrecurrent cervical cancer. Regional nerve, mus-cle, and bone infiltration can cause severe pain.Narcotic analgesics are a fundamental com-ponent of cancer pain treatment and may beprepared for oral, rectal, vaginal, sublingual, in-travenous, intramuscular, epidural, or topicaladministration.

When pain is directly attributable to specific fociof disease, such as bone metastasis or para-aortic lymph node recurrence, a brief course ofpalliative radiation therapy yields substantial painreduction in a high percentage of patients. How-ever, pain relief may not be maximally achieveduntil weeks after the palliative radiation therapyends.

Advanced cervical cancer may cause urinary fis-tulas, with vesicovaginal occurring more com-monly than ureterovaginal fistulas. Although notnecessarily painful, fistulous drainage can have



an extremely negative impact on quality of lifebecause of the constant odor.

Palliation of fistulas may be surgically accom-plished by creation of a ureterointestinal conduitor placement of bilateral percutaneous nephros-tomies to decompress the ureters. Rectovaginalfistulasmay occur fromprimary tumor invasion ofthe adjacent rectum. These more often result fromradiation injury or tumor recurrence. A divertingcolostomy is the surgical procedure of choice. Insome lower-resource settings, because the avail-ability of radiotherapy is sometimes difficult or non-existent, some fistulas and bleeding will be treatedwith other interventions.

Progressive or recurrent cervical cancer maycause uremia secondary to ureteral obstruction,whichmay inducenausea, vomiting, somnolence,confusion, and seizures. Ureteral obstruction isrelieved by percutaneous nephrostomy or ureteralstents. Treatment of other symptoms is standard inpatients needing palliative care, and the reader isreferred to Chapter 7 of the WHO comprehensivecervical cancer book.5,8,61

COST IMPLICATIONS

There have been few studies of the cost effective-ness of treatment in LMICs. We can make someinferences as to how cost effective treatment is,using studies from threemiddle-income countries(Thailand,which is anuppermiddle–incomecoun-try, and China andMorocco; both of the latter werelower middle–income countries at the time thestudies were conducted).

Katanyoo et al34 compared two different treatmentoptions for stage IIB disease (radical hysterectomywith pelvic lymph node dissection with or withoutpostoperative adjuvant therapy v concurrent che-moradiotherapy). The first option (compared withno intervention) cost $322 per quality-adjustedlife-year (QALY), and the second cost $327.4 perQALY in 2012 US dollars. This is considerablybelow the Thai per capita gross national income($5,180 in the same year).62 Any interventioncosting less than one times the per capita nationalincome per disability-adjusted life-year is consid-ered highly cost effective.63 Although QALYs anddisability-adjusted life-years are not identical,they are roughly comparable. Cost of treatmentof stage IIB disease in Thailand was $5,218 to$5,281 (ie, slightly greater than per capita in-come). However, the number of QALYs saved islarge (15.9 to 16.4).

Cost data areavailable for treatment of similar-stagedisease inMoroccoandChina.Thecost inMorocco

in 2008 was $6,883 (total colpohysterectomy, pel-vic lymphadenectomy, and postoperative chemo-therapy and radiotherapy), approximately 2.7 timesthe per capita national income of Morocco in thatyear.35IfQALYssavedweresimilar tothoseinThailand(approximately 16), the cost per QALY saved of $430would also fall into the cost-effective range.

For China, costs of the treatment componentswere estimated for a second-level (prefecture)hospital for 2008.36 Cost ranged from $404.73to $544.39 for hysterectomy (simple to radical),was $281.65 for radiation therapy, and was$125.46 for adjuvant chemotherapy, such thattreatment costs for stage IIB disease would be alittle less than $1,000, compared with the percapita national income in that year of $2,975.Again, if the QALYs saved compared with notreatment were similar to those of Thailand, thecost per QALY would fall into the cost-effectiverange (, $100 per QALY).

Thus, treatment of patients with stage IIB diseaseis likely cost effective inmiddle-incomecountries,although the cost per patient is significant (rangingfrom one third of the per capita income in China toapproximately one times the per capita income inThailand to 2.7 times the per capita income inMorocco). Key needs are therefore to expand sur-gical capacity andensurewomenaccess treatmentsufficiently early in disease progression.63

The risk of metastasis to pelvic lymph nodes inpatientswithearlydisease is approximately15%to20%and 25% to 40% for patients with stage I andII disease, respectively, so there is a potential forcure in many of these patients whose disease hasnot yet spread. This is the rationale for extrafascialhysterectomy or its modification in the basic set-ting, but this recommendation depends on theability to safely perform these procedures in theselow-resource settings. This would depend as wellon the ability to perform simple hysterectomiessafely in low-resourcesettings.Concentratingsurgicalvolumeinhigh-riskcentersandbyhigh-risksurgeonshas been shown in many clinical settings to improveoutcomes. Thus, even in countries without trainedgynecologic oncologists or access to ideal radiationtherapy facilities, surgical outcomes could be im-proved by concentrating resources and designatingexperts.These typesofchangesmaybecosteffectivebothby improving clinical outcomesandbyoptimallyusing existing resources.

EXTERNAL REVIEW

The draft was submitted to one external reviewerwith content expertise. It was rated as high quality,and it was agreed it would be useful in practice.



Review comments were reviewed by the co-chairsof the Expert Panel and integrated into the finalmanuscript before approval by the CPGC.

GUIDELINE IMPLEMENTATION

ASCO guidelines are developed for implementa-tion across health settings. Barriers to implemen-tation include the need to increase awareness ofthe guideline recommendations among front-linepractitioners, survivors of cancer, and caregiversand the need to provide adequate services in theface of limited resources. The guideline BottomLineBoxwasdesigned to facilitate implementationof recommendations. This guideline will be dis-tributed widely, including through the ASCO Prac-tice Guideline Implementation Network and theASCO Web site.

LIMITATIONS OF RESEARCH

There were several areas where evidence waslacking to make strong recommendations. Theseareas included the following: optimal post-treatmentsurveillance for women with cervical cancer atrisk for recurrence, including the role of PET scansin maximal-resource settings; using squamouscell carcinoma antigen and/or high-sensitivityC-reactive protein; optimal dose fractionation of bra-chytherapy; surgery for womenwith stage IA2 or IB1disease with tumor smaller than 2 cm in size and1 cm in depth in the non–fertility-sparing setting;optimal treatment of patients with stage IB1 cervicalcancerwith tumor size between2and4cm; optimalfertility-sparingprocedures forwomenwithstage IA1or IA2disease;and treatmentofwomenwith invasivecervical cancer inbasic settings, including regardingchemotherapy and radiation therapy.

ASCO believes that cancer clinical trials are vital toinform medical decisions and improve cancercare. All patients should have the opportunity toparticipate.

FUTURE DIRECTIONS

Furtherprospectivecomparativeresearchisneeded,conducted especially in basic- and limited-resourcesettings. There are ongoing studies of treatment forpatients with stage IA2 versus 1B1 disease. Radicalversussimplehysterectomy(the feasibility andsafetyof performingconebiopsy or simplehysterectomy) isan active area of investigation:

·Nonrandomized GOG 278 (Studying the Physi-cal Function and Quality of Life Before and AfterSurgery in PatientsWith Stage I Cervical Cancer)compares cone and pelvic lymphadenec-tomy versus simple hysterectomy and pelvic

lymphadenectomy for patients with stage IA2versus 1B1IA1 to IB1 disease with tumorssmaller than 2 cm (ClinicalTrials.gov identifierNCT01649089)

·Similar nonrandomized ConCerv (ConservativeSurgery for Women With Low-Risk Early StageCervical Cancer) study (ClinicalTrials.gov identifierNCT01048853)

·Randomized SHAPE (Radical Versus SimpleHysterectomy and Pelvic Node Dissection inPatients With Low-Risk Early Stage CervicalCancer) trial of patients with stage IA2 or IB1diseasewith tumors smaller than2cm, invasionsmaller than 10 mm, and invasion less than50% on magnetic resonance imaging (Clin-icalTrials.gov identifier NCT01658930)

Postchemoradiotherapy with adjuvant chemo-therapy trials:

·OUTBACK (Phase III Trial of Adjuvant Chemo-therapyAsPrimaryTreatmentofLocallyAdvancedCervical Cancer Compared With ChemoradiationAlone) trial (ANZGOG [Australia New ZealandGynaecological Oncology Group] 0902/GOG0274/RTOG 1174; ClinicalTrials.gov identi-fier NCT01414608)

·TACO (Triweekly Cisplatin Based Chemo-radiation in Locally Advanced Cervical Cancer)trial (ClinicalTrials.gov identifier NCT01561586)

·RTOG 0724 (Phase III Randomized Study ofConcurrent Chemotherapy and Pelvic RadiationTherapy With or Without Adjuvant Chemo-therapy in High-Risk Patients With Early-StageCervical Carcinoma Following Radical Hys-terectomy) trial (ClinicalTrials.gov identifierNCT00980954)

Selected ongoing trials of NACT:

·EORTC 55994 (Randomized Phase III Study ofNeoadjuvant Chemotherapy Followed by Sur-gery vs. Concomitant Radiotherapy and Che-motherapy in FIGO Stage Ib2, IIa . 4 cm, IIbCervical Cancer) trial (ClinicalTrials.gov identi-fier NCT00039338)

·NACTcervix (Prospective Randomized Trialof Neoadjuvant Chemotherapy and SurgeryVersus Concurrent Chemoradiation Therapy inPatients With Stage IB2-IIB Squamous Carci-noma of the Uterine Cervix) trial (ClinicalTrials.gov identifier NCT00193739)

·SYSGO002 (Neoadjuvant Chemotherapy andRadical Surgery Versus Concurrent Chemo-irradiation inFIGOStage IIBCervicalCancer) trial(ClinicalTrials.gov identifier NCT02595554)














DEFINITION

The panel advocates para-aortic lymph node sam-pling insteadof lymphadenectomy.This isperformedfor the purpose of prognostic and not therapeuticimplication.Sedlisetal37criteria forpara-aortic lymphnode sampling are as follows (Appendix Table A4):presenceofLVSIplusdeepone-thirdcervical stromalinvasion and tumor of any size, presenceof LVSI plusmiddle one-third stromal invasion and tumor size of2 cmor larger, presence of LVSI plus superficial one-third stromal invasion and tumor size of 5 cm orlarger, or no LVSI but deep or middle one-thirdstromal invasion and tumor size of 4 cm or larger.

ADDITIONAL RESOURCES

More information, including a Data Supplementwith additional evidence tables, a MethodologySupplement with information about evidence qual-ity and strength of recommendations, slide sets,and clinical tools and resources, is available at www.asco.org/rs-cervical-cancer-treatment-guideline. Pa-tient information is available at www.cancer.net. Visitwww.asco.org/guidelineswiki to provide commentson the guideline or submit new evidence.

DOI: 10.1200/JGO.2016.003954Published online on jgo.ascopubs.org on May 25, 2016.


Administrative support: Sarah TeminManuscript writing: All authorsFinal approval of manuscript: All authors


The following represents disclosure information provided byauthors of this manuscript. All relationships are consideredcompensated. Relationships are self-held unless noted.I 5 Immediate Family Member, Inst 5 My Institution. Re-lationships may not relate to the subject matter of thismanuscript. For more information about ASCO’s conflict ofinterest policy, please refer to www.asco.org/rwc orjco.ascopubs.org/site/ifc.

Linus T. ChuangNo relationship to disclose

Sarah TeminNo relationship to disclose

Rolando CamachoNo relationship to disclose

Alfonso Due~nas-GonzalezNo relationship to disclose

Sarah FeldmanNo relationship to disclose

Murat GultekinNo relationship to disclose

Vandana GuptaNo relationship to disclose

Susan HortonNo relationship to disclose

Graciela JacobNo relationship to disclose

Elizabeth A. KiddResearch Funding: Siemens

Kennedy LishimpiNo relationship to disclose

Carolyn NakisigeNo relationship to disclose

Joo-Hyun NamResearch Funding: AstraZeneca, AbbVie

Hextan Yuen Sheung NganHonoraria:Merck Sharp & Dohme (Asia), Amgen (Asia), IpsosHealthcare, Chindex Hong Kong, PfizerConsulting or Advisory Role: Amgen (Asia)Speakers’Bureau:MerckSharp&Dohme (Asia), ChindexHongKong, PfizerResearch Funding: Amgen (Asia; Inst)

Travel, Accommodations, Expenses: Roche Hong Kong, Chin-dex Hong Kong, Merck Sharp & Dohme (Asia), Pfizer,AstraZeneca

William SmallHonoraria: Carl Zeiss MeditecSpeakers’ Bureau: Carl Zeiss MeditecResearch Funding: Carl Zeiss Meditec

Travel, Accommodations, Expenses: Carl Zeiss Meditec

Gillian ThomasNo relationship to disclose

Jonathan S. BerekConsulting or Advisory Role: Atara BiotherapeuticsResearch Funding: Tesaro (Inst), AstraZeneca (Inst)

ACKNOWLEDGMENTWe thank Muhieddine A.F. Seoud, MD, Natasha B. Leighl,MD, Alexi A. Wright, MD, MPH, and the Clinical PracticeGuideline Committee for their thoughtful reviews and in-sightful comments on this guideline document and LynetteDenny,MD,PhD, Allen S. Lichter,MD,Neville F.Hacker,MD,and Shannon McKernin for their assistance.

AFFILIATIONS

Linus T. Chuang, Icahn School of Medicine at Mt Sinai, New York, NY; Sarah Temin, American Society of Clinical Oncology,Alexandria, VA;Rolando Camacho, retired,Mallorca, Spain;Alfonso Due~nas-Gonzalez, InstitutoNacional deCancerologia,Mexico











City, Mexico; Sarah Feldman, Dana-Farber Cancer Institute and Brigham and Women’s Hospital, Boston, MA; Murat Gultekin,Turkish Ministry of Health, Ankara, Turkey; Vandana Gupta, patient representative, V Care Foundation, Mumbai, India; SusanHorton, University of Waterloo, Waterloo; Gillian Thomas, Sunnybrook Odette Cancer Centre and University of Toronto, Toronto,Ontario, Canada; Graciela Jacob, Instituto Nacional de Cancerologia, Buenos Aires, Argentina; Elizabeth A. Kidd and Jonathan S.Berek, Stanford Comprehensive Cancer Institute, Stanford, CA; Kennedy Lishimpi, Cancer Diseases Hospital, Lusaka, Zambia;Carolyn Nakisige, Mulago Hospital, Kampala, Uganda; Joo-Hyun Nam, Asan Medical Center, Seoul, South Korea; Hextan YuenSheung Ngan, University of Hong Kong, Hong Kong, Special Administrative Region, People’s Republic of China; and WilliamSmall, Cardinal Bernardin Cancer Center, Stritch School of Medicine, Loyola University, Chicago, IL.Clinical Practice Guideline Committee approved: February 21, 2016.

Editor’s note: This American Society of Clinical Oncology clinical practice guideline provides recommendations, with com-prehensive review and analysis of the relevant literature for each recommendation. Additional information, including a DataSupplement with additional evidence tables, a Methodology Supplement, slide sets, clinical tools and resources, and links topatient information at www.cancer.net, is available at www.asco.org/rs-cervical-cancer-treatment-guideline.Reprint requests: 2318 Mill Rd, Suite 800, Alexandria, VA 22314; e-mail: [email protected].

REFERENCES

references

1. International Agency for Research on Cancer: GLOBOCAN 2012 Cervical Cancer: Estimated Incidence, Mortality andPrevalence Worldwide in 2012. http://globocan.iarc.fr/old/FactSheets/cancers/cervix-new.asp

2. Howlader N, Noone AM, Krapcho M, et al: SEER Cancer Statistics Review, 1975-2012. http://seer.cancer.gov/csr/1975_2012/

3. Catholic Campaign for Human Development: Interrelationships of three-year average state poverty rates: 2011-2013.http://www.povertyusa.org/the-state-of-poverty/poverty-map-state/#

4. Hirte H, Kennedy EB, Elit L, et al: Systemic therapy for recurrent, persistent, or metastatic cervical cancer: A clinicalpractice guideline. Curr Oncol 22:211-219, 2015

5. Elit L, Kennedy EB, Fyles A, et al: Follow-Up for Cervical Cancer. https://www.cancercare.on.ca/common/pages/UserFile.aspx?fileId5340742

6. Colombo N, Carinelli S, Colombo A, et al: Cervical cancer: ESMO clinical practice guidelines for diagnosis, treatmentand follow-up. Ann Oncol 23:vii27-vii32, 2012 (suppl 7)

7. Ebina Y, Yaegashi N, Katabuchi H, et al: Japan Society of Gynecologic Oncology guidelines 2011 for the treatment ofuterine cervical cancer. Int J Clin Oncol 20:240-248, 2015

8. Koh WJ, Greer BE, Abu-Rustum NR, et al: Cervical cancer, version 2.2015. J Natl Compr Canc Netw 13:395-404,2015, quiz 404

9. World Health Organization: Comprehensive Cervical Cancer Control: A Guide to Essential Practice (ed 2). http://apps.who.int/iris/bitstream/10665/144785/1/9789241548953_eng.pdf

10. World Health Organization: WHO National Cancer Control Programmes: Policies and Managerial Guidelines—2002.http://www.who.int/cancer/publications/nccp2002/en/

11. Anderson BO, Yip CH, Smith RA, et al: Guideline implementation for breast healthcare in low-income and middle-income countries: Overview of the Breast Health Global Initiative Global Summit 2007. Cancer 113:2221-2243, 2008(suppl 8)

12. Horton S, Gauvreau CL: Cancer in low- and middle-income countries: An economic overview, in Gelband H, Jha P,Sankaranarayanan R, et al (eds): Cancer Disease Control Priorities (ed 3). Washington, DC, International Bank forReconstruction and Development/World Bank, 2015

13. ADAPTE Collaboration: Guideline Adaptation: A Resource Toolkit (version 2.0). http://www.g-i-n.net/document-store/working-groups-documents/adaptation/adapte-resource-toolkit-guideline-adaptation-2-0.pdf

14. Shiffman RN, Michel G, Rosenfeld RM, et al: Building better guidelines with BRIDGE-Wiz: Development andevaluation of a software assistant to promote clarity, transparency, and implementability. J Am Med Inform Assoc 19:94-101, 2012

15. Koh WJ, Greer BE, Abu-Rustum NR, et al: NCCN Guidelines Version 2.2015: Cervical Cancer PreliminaryResource Stratification—Maximal Level. Fort Washington, PA, National Comprehensive Cancer Network,2015

16. Koh WJ, Greer BE, Abu-Rustum NR, et al: NCCN Guidelines Version 2.2015: Cervical Cancer Preliminary ResourceStratification—Limited Level. Fort Washington, PA, National Comprehensive Cancer Network, 2015

ADDENDUMThis addendum acknowledges the endorsement by Gynecologic Cancer InterGroup, in addition to the endorsement of theSociety of Gynecologic Oncology as previously acknowledged in the footnotes.





http://globocan.iarc.fr/old/FactSheets/cancers/cervix-new.asp



http://www.povertyusa.org/the-state-of-poverty/poverty-map-state/#

https://www.cancercare.on.ca/common/pages/UserFile.aspx?fileId=340742



http://apps.who.int/iris/bitstream/10665/144785/1/9789241548953_eng.pdf


http://www.who.int/cancer/publications/nccp2002/en/

http://www.g-i-n.net/document-store/working-groups-documents/adaptation/adapte-resource-toolkit-guideline-adaptation-2-0.pdf

http://www.g-i-n.net/document-store/working-groups-documents/adaptation/adapte-resource-toolkit-guideline-adaptation-2-0.pdf


17. Koh WJ, Greer BE, Abu-Rustum NR, et al: NCCN Guidelines Version 2.2015: Cervical Cancer PreliminaryResource Stratification—Enhanced Level. Fort Washington, PA, National Comprehensive Cancer Network,2015

18. Koh WJ, Greer BE, Abu-Rustum NR, et al: NCCN Guidelines Version 2.2015: Cervical Cancer Preliminary ResourceStratification—Basic Level. Fort Washington, PA, National Comprehensive Cancer Network, 2015

19. Lee LJ, Das IJ, Higgins SA, et al: American Brachytherapy Society consensus guidelines for locally advanced car-cinoma of the cervix: Part III—Low-dose-rate and pulsed-dose-rate brachytherapy. Brachytherapy 11:53-57, 2012

20. Nagase S, Inoue Y, Umesaki N, et al: Evidence-based guidelines for treatment of cervical cancer in Japan: JapanSociety of Gynecologic Oncology (JSGO) 2007 edition. Int J Clin Oncol 15:117-124, 2010

21. Viswanathan AN, Beriwal S, De Los Santos JF, et al: American Brachytherapy Society consensus guidelines for locallyadvanced carcinoma of the cervix: Part II—High-dose-rate brachytherapy. Brachytherapy 11:47-52, 2012

22. Oaknin A, Dıaz de Corcuera I, Rodrıguez-Freixinos V, et al: SEOM guidelines for cervical cancer. Clin Transl Oncol 14:516-519, 2012

23. Small W Jr, Beriwal S, Demanes DJ, et al: American Brachytherapy Society consensus guidelines for adjuvant vaginalcuff brachytherapy after hysterectomy. Brachytherapy 11:58-67, 2012

24. Haie-Meder C, Fervers B, Fondrinier E, et al: SOR guidelines for concomitant chemoradiotherapy for patients withuterine cervical cancers: Evidence update bulletin 2004. Ann Oncol 16:1100-1108, 2005

25. Ye Q, Yuan HX, Chen HL: Responsiveness of neoadjuvant chemotherapy before surgery predicts favorable prognosisfor cervical cancer patients: A meta-analysis. J Cancer Res Clin Oncol 139:1887-1898, 2013

26. Kokka F, Bryant A, Brockbank E, et al: Hysterectomywith radiotherapy or chemotherapy or both for womenwith locallyadvanced cervical cancer. Cochrane Database Syst Rev 4:CD010260, 2015

27. Neoadjuvant Chemotherapy for Cervical Cancer Meta-Analysis Collaboration (NACCCMA) Collaboration: Neoadjuvantchemotherapy for locally advanced cervix cancer. Cochrane Database Syst Rev 2:CD001774, 2004

28. Rydzewska L, Tierney J, Vale CL, et al: Neoadjuvant chemotherapy plus surgery versus surgery for cervical cancer.Cochrane Database Syst Rev 12:CD007406, 2012

29. Rosa DD, Medeiros LR, Edelweiss MI, et al: Adjuvant platinum-based chemotherapy for early stage cervical cancer.Cochrane Database Syst Rev 6:CD005342, 2012

30. Kucukmetin A, Biliatis I, Naik R, et al: Laparoscopically assisted radical vaginal hysterectomy versus radical abdominalhysterectomy for the treatment of early cervical cancer. Cochrane Database Syst Rev 10:CD006651, 2013

31. Liu R,Wang X, Tian JH, et al: High dose rate versus low dose rate intracavity brachytherapy for locally advanceduterinecervix cancer. Cochrane Database Syst Rev 10:CD007563, 2014

32. Scatchard K, Forrest JL, Flubacher M, et al: Chemotherapy for metastatic and recurrent cervical cancer. CochraneDatabase Syst Rev 10:CD006469, 2012

33. van Lonkhuijzen L, ThomasG: Palliative radiotherapy for cervical carcinoma: A systematic review. Radiother Oncol 98:287-291, 2011

34. Katanyoo K, Praditsitthikorn N, Tangjitgamol S, et al: Cost-utility analysis of treatments for stage IB cervical cancer.J Gynecol Oncol 25:97-104, 2014

35. Berraho M, Najdi A, Mathoulin-Pelissier S, et al: Direct costs of cervical cancer management in Morocco. Asian Pac JCancer Prev 13:3159-3163, 2012

36. Shi JF, Chen JF, Canfell K, et al: Estimation of the costs of cervical cancer screening, diagnosis and treatment in ruralShanxi Province, China: A micro-costing study. BMC Health Serv Res 12:123, 2012

37. Sedlis A, Bundy BN, Rotman MZ, et al: A randomized trial of pelvic radiation therapy versus no further therapy inselected patients with stage IB carcinoma of the cervix after radical hysterectomy and pelvic lymphadenectomy: AGynecologic Oncology Group study. Gynecol Oncol 73:177-183, 1999

38. National Cancer Institute: Cervical cancer treatment: Health professional version. http://www.cancer.gov/types/cervical/hp/cervical-treatment-pdq

39. Zuliani AC, Esteves SC, Teixeira LC, et al: Concomitant cisplatin plus radiotherapy and high-dose-rate brachytherapyversus radiotherapy alone for stage IIIB epidermoid cervical cancer: A randomized controlled trial. J Clin Oncol 32:542-547, 2014

40. Srivastava K, Paul S, Chufal KS, et al: Concurrent chemoradiation versus radiotherapy alone in cervical carcinoma: Arandomized phase III trial. Asia Pac J Clin Oncol 9:349-356, 2013

41. Cetina L, Gonzalez-Enciso A, Cantu D, et al: Brachytherapy versus radical hysterectomy after external beam che-moradiation with gemcitabine plus cisplatin: A randomized, phase III study in IB2-IIB cervical cancer patients. AnnOncol 24:2043-2047, 2013

42. Eifel PJ: Adjuvant surgery after radiotherapy. J Natl Cancer Inst Monogr 21:85-88, 1996


http://www.cancer.gov/types/cervical/hp/cervical-treatment-pdq

http://www.cancer.gov/types/cervical/hp/cervical-treatment-pdq


43. Due~nas-Gonzalez A, Zarba JJ, Patel F, et al: Phase III, open-label, randomized study comparing concurrent gem-citabine plus cisplatin and radiation followed by adjuvant gemcitabine and cisplatin versus concurrent cisplatin andradiation in patients with stage IIB to IVA carcinoma of the cervix. J Clin Oncol 29:1678-1685, 2011

44. Wethington SL, Sonoda Y, Park KJ, et al: Expanding the indications for radical trachelectomy: A report on 29 patientswith stage IB1 tumors measuring 2 to 4 centimeters. Int J Gynecol Cancer 23:1092-1098, 2013

45. Li J, Wu X, Li X, et al: Abdominal radical trachelectomy: Is it safe for IB1 cervical cancer with tumors> 2 cm? GynecolOncol 131:87-92, 2013

46. Delgado G, Bundy B, Zaino R, et al: Prospective surgical-pathological study of disease-free interval in patients withstage IB squamous cell carcinoma of the cervix: A Gynecologic Oncology Group study. Gynecol Oncol 38:352-357,1990

47. Loren AW, Mangu PB, Beck LN, et al: Fertility preservation for patients with cancer: American Society of ClinicalOncology clinical practice guideline update. J Clin Oncol 31:2500-2510, 2013

48. Kitagawa R, Katsumata N, Shibata T, et al: Paclitaxel plus carboplatin versus paclitaxel plus cisplatin in metastatic orrecurrent cervical cancer: The open-label randomized phase III trial JCOG0505. J Clin Oncol 33:2129-2135, 2015

49. Tewari KS, Sill MW, Long HJ III, et al: Improved survival with bevacizumab in advanced cervical cancer. N Engl J Med370:734-743, 2014

50. SpanosWJ Jr, Pajak TJ, Emami B, et al: Radiation palliation of cervical cancer. J Natl Cancer Inst Monogr 21:127-130,1996

51. Einck JP, Hudson A, Shulman AC, et al: Implementation of a high-dose-rate brachytherapy program for carcinoma ofthe cervix in Senegal: A pragmatic model for the developing world. Int J Radiat Oncol Biol Phys 89:462-467, 2014

52. HanK,MilosevicM, Fyles A, et al: Trends in the utilization of brachytherapy in cervical cancer in theUnited States. Int JRadiat Oncol Biol Phys 87:111-119, 2013

53. Logsdon MD, Eifel PJ: Figo IIIB squamous cell carcinoma of the cervix: An analysis of prognostic factors emphasizingthe balance between external beam and intracavitary radiation therapy. Int J Radiat Oncol Biol Phys 43:763-775,1999

54. Schwarz JK, Siegel BA, Dehdashti F, et al: Association of posttherapy positron emission tomography with tumorresponse and survival in cervical carcinoma. JAMA 298:2289-2295, 2007

55. World Health Organization: Strengthening of palliative care as a component of comprehensive care throughout the lifecourse. http://apps.who.int/gb/ebwha/pdf_files/WHA67/A67_R19-en.pdf

56. Jaffray DA, Gospodarowicz M: Bringing global access to radiation therapy: Time for a change in approach. Int J RadiatOncol Biol Phys 89:446-447, 2014

57. Zubizarreta EH, Fidarova E, Healy B, et al: Need for radiotherapy in low andmiddle income countries: The silent crisiscontinues. Clin Oncol (R Coll Radiol) 27:107-114, 2015

58. Datta NR, Samiei M, Bodis S: Radiation therapy infrastructure and human resources in low- and middle-incomecountries: Present status and projections for 2020. Int J Radiat Oncol Biol Phys 89:448-457, 2014

59. Abdel-Wahab M, Bourque JM, Pynda Y, et al: Status of radiotherapy resources in Africa: An International AtomicEnergy Agency analysis. Lancet Oncol 14:e168-e175, 2013

60. Fisher BJ, Daugherty LC, Einck JP, et al: Radiation oncology in Africa: improving access to cancer care on the Africancontinent. Int J Radiat Oncol Biol Phys 89:458-461, 2014

61. Friedlander M, Grogan M; US Preventative Services Task Force: Guidelines for the treatment of recurrent andmetastatic cervical cancer. Oncologist 7:342-347, 2002

62. TheWorld Bank: GNI per capita, Atlasmethod (current US$). http://data.worldbank.org/indicator/NY.GNP.PCAP.CD?page51

63. Sachs JD: Macroeconomics and Health: Investing in Health for Economic Development. Geneva, Switzerland, WorldHealth Organization, 2001


http://apps.who.int/gb/ebwha/pdf_files/WHA67/A67_R19-en.pdf

http://data.worldbank.org/indicator/NY.GNP.PCAP.CD?page=1




APPENDIX

Table A2 – Framework of Resource Stratification: Treatment

Setting


Core resources or fundamentalservices absolutely necessary forany gynecologic health caresystem to function; basic-levelservices are typically applied ina single clinical interaction

Second-tier resources or servicesthat producemajor improvementsin outcome, such as increasedsurvival, but that are attainablewith limited financial means andmodest infrastructure; limited-level servicesmay involve single ormultiple clinical interactions

Third-tier resources or services thatare optional but important;enhanced-level resources mayproduce minor improvements inoutcome but increase the numberand quality of therapeutic optionsand patient choices

Mayuseguidelines for high-resourcesettings; high-level resources orservices that may be used in somehigh-resource countries; thisshould be considered lowerpriority than those in the othersettings based on cost orimpracticality for limited-resourceenvironment

NOTE. Data adapted (http://apps.who.int/iris/bitstream/10665/94830/1/9789241548694_eng.pdf).11,12

Table A1 – Guidelines Considered in ADAPTE Process

Guideline Link

Colombo6 (ESMO) http://annonc.oxfordjournals.org/content/23/suppl_7/vii27.full

Koh18 (NCCN) http://www.nccn.org/professionals/physician_gls/pdf/cervical_basic.pdf

Koh16 (NCCN) http://www.nccn.org/professionals/physician_gls/pdf/cervical_limited.pdf

Koh17 (NCCN) http://www.nccn.org/professionals/physician_gls/pdf/cervical_enhanced.pdf

Koh8 (NCCN) http://www.nccn.org/professionals/physician_gls/pdf/cervical.pdf

Hirte4 (CCO) https://www.cancercare.on.ca/common/pages/UserFile.aspx?fileId534224

Elit5 (CCO) https://www.cancercare.on.ca/common/pages/UserFile.aspx?fileId514090

Ebina7 (JSGO) http://link.springer.com/article/10.1007/s10147-015-0806-7

WHO9 http://www.who.int/reproductivehealth/publications/cancers/cervical-cancer-guide/en/

Abbreviations: CCO, Cancer Care Ontario; ESMO, European Society of Medical Oncology; JSGO, Japan Society of Gynecologic Oncology;NCCN, National Comprehensive Cancer Network.



http://annonc.oxfordjournals.org/content/23/suppl_7/vii27.full

http://www.nccn.org/professionals/physician_gls/pdf/cervical_basic.pdf

http://www.nccn.org/professionals/physician_gls/pdf/cervical_limited.pdf

http://www.nccn.org/professionals/physician_gls/pdf/cervical_enhanced.pdf

http://www.nccn.org/professionals/physician_gls/pdf/cervical.pdf





http://link.springer.com/article/10.1007/s10147-015-0806-7

http://www.who.int/reproductivehealth/publications/cancers/cervical-cancer-guide/en/


Table A3 – Expert Panel Membership

Member Affiliation Role or Area of Expertise

Jonathan S. Berek, MD,co-chair

Comprehensive Cancer Institute, Stanford,CA

Gynecologic oncology

Linus T. Chuang, MD,co-Chair

Icahn School of Medicine at Mt Sinai, NewYork, NY


Rolando Camacho, MD retired, Mallorca, Spain Cancer control

Alfonso Due~nas-Gonzalez,MD

Instituto Nacional de Cancerologia, MexicoCity, Mexico

Medical oncology

Sarah Feldman, MD Dana-Farber Cancer Institute and Brighamand Women’s Hospital, Boston, MA


Murat Gultekin, MD Turkish Ministry of Health, Ankara, Turkey Cancer control and gynecologic oncology

Susan Horton, PhD University of Waterloo, Waterloo, Ontario,Canada

Health economics

Graciela Jacob, MD Instituto Nacional de Cancerologia, BuenosAires, Argentina

Palliative care

Elizabeth A. Kidd, MD Stanford University, Stanford, CA Radiation oncology

Kennedy Lishimpi, MD Cancer Diseases Hospital, Lusaka, Zambia Medical oncology

Carolyn Nakisige, MD Mulago Hospital, Kampala, Uganda Medical oncology

Joo-Hyun Nam, MD, PhD Asan Medical Center, Seoul, South Korea Obstetrics and gynecology

Hextan Yuen Sheung Ngan,MD

University of Hong Kong, Hong Kong,Special Administrative Region, People’sRepublic of China

Obstetrics and gynecology

William Small, MD Cardinal Bernardin Cancer Center, StritchSchool of Medicine, Loyola University,Chicago, IL

Radiation oncology

Gillian Thomas, MD Sunnybrook Odette Cancer Centre andUniversity of Toronto, Toronto, Ontario,Canada

Radiation oncology

Vandana Gupta V Care, Mumbai, India Patient representative

NOTE. American Society of Clinical Oncology staff: Sarah Temin, MSPH.

Table A4 – Combination of Three Risk Factors: CLS, Stromal Invasion, and Tumor Size by Treatment Regimen

CLS Stromal Invasion Tumor size (cm)

No. (%)

RT No Additional Therapy

Positive Deep one-third Any 60 (43.0) 68 (48.6)

Positive Middle one-third > 2 28 (20.4) 37 (26.4)

Negative Deep or middle one-third > 4 48 (35.0) 34 (29.3)

Positive Superficial one-third > 5 1 (0.7) 1 (0.7)

Total 137 (100.0) 140 (100.0)

NOTE. Data adapted with permission.37

Abbreviation: CLS, capillary lymphatic space.



casereport

Gallbladder AdenomyomatosisMimicking Carcinoma: A DiagnosticDilemma

CASE REPORT

A69-year-oldmanpresentedwith right hypochon-driac pain accompanied by postprandial nauseaand vomiting for 2 months. Additional symptomsincluded fever and fatigue for 4 days. There wasassociated history of weight loss of approximately6 kg over the past few months. Vital signs werenormal, and he was febrile (38.3°C). His skin andsclerae weremildly icteric. He had no signs of liverfailure. His abdomen was soft, with tendernessnoted in the right hypochondriac region (Murphysign positive); no tenderness was noted else-where, and there was no evidence of ascites.The tip of the gallbladder was palpable in theright 9th costal margin. Laboratory evaluationrevealed mild conjugated hyperbilirubinemia(total bilirubin level, 2.1 mg/dL; direct bilirubinlevel, 1.1 mg/dL), with normal levels of trans-aminases and alkaline phosphatase. Mild hypo-albuminemia was present. Other investigationswere unremarkable. On the basis of the clinicaland laboratory findings, the patient was evalu-ated for obstructive jaundice, and ultrasonogra-phy (US) was performed.

IMAGING FINDINGS

US revealed an abnormally thick gallbladder wallmeasuring 3 cm in thickness. The wall thickeninggave the appearance of an endophyticmass caus-ing near-total obliteration of the lumen. Multipleintraluminal and intramural calculi were seen(Fig 1A). Few echogenic intramural foci werealso seen. No color uptake was noted on powerDoppler evaluation (Fig 1B). Sonographic Mur-phy sign was absent, and there was no pericho-lecystic fluid collection. There was no extraor intrahepatic biliary system dilatation. Gall-bladder malignancy/adenomyomatosis withcholelithiasis was considered as the probablediagnosis; in view of absence of acute inflam-matory features, cholecystitis was consideredunlikely. No other sonographic abnormalitywas identified.

PATHOLOGIC EVALUATION

Interval cholecystectomy was performed 4 weeksafter initial presentation. At laparotomy, diffuselyenlarged gallbladder was seen, which was firmand showed multiple adhesions.

A gross cholecystectomy specimen showed dif-fusely enlarged gallbladder weighing 63 g andmeasuring 7.5 3 5.5 3 2.5 cm. On cut sections,the gallbladder was firm, with grayish thickenedwall measuring up to 2.5 cm and velvety granularmucosa containing numerous multifaceted intra-luminal calculi. The gallbladderwall showed areasof hemorrhages and contained multiple pockets,many of which also contained calculi. Microsec-tions showedulcerations andedematous gallblad-der mucosa and myoepithelial hyperplasia withabundant granulation tissue composed of myofi-broblasts. Inflammatory cells, suchasneutrophils,plasma cells, and macrophages, were also pres-ent, with congested vessels (Fig 2). The finaldiagnosis of adenomyomatosis of the gallbladderwas confirmed by the characteristic histopatho-logic appearance of muscular and epithelialhyperplasia contributing to mural thickening,with epithelial invaginations forming the pathogno-monic intramural diverticula known as Rokitansky-Aschoff sinuses (Fig 2).No evidence ofmalignancywas found.

DISCUSSION

Adenomyomatosis is not an uncommonmimic ofthe gallbladder carcinoma, identified in 1% to8.5% of cholecystectomy specimens.1 Table 1summaries the clinicoradiopathological featuresof gallbladder adenomyomatosis. It is most oftenan incidental finding, has no intrinsic malig-nant potential, and usually requires no specifictreatment.2,3 Varied names have been applied tothis lesion in the literature, including adenomyo-matosis, adenomyoma, diverticular disease, intra-mural diverticulosis, cholecystitis cystica, andcholecystitis glandularis proliferans. It ismore com-monly seen in women than in men, the majority

Abhishek Mahajan

Smiti Sripathi

Abhishek Mahajan, TataMemorial Centre,Mumbai; and SmitiSripathi, Kasturba MedicalCollege, Manipal, India.Authors’ disclosures ofpotential conflicts ofinterest and contributionsare found at the end of thisarticle.Corresponding author:Abhishek Mahajan, MD,MRes, Room 120,Department ofRadiodiagnosis, TataMemorial Hospital, TataMemorial Centre, Dr EBorges Road, Parel,Mumbai, India, 400012;e-mail: [email protected].






presenting with complaints of chronic right upperquadrant pain.2 Frequent coexistence with chole-lithiasis is seen; however, no causative relationshiphas been proved.4 Adenomyomatosis occasionallyproduces abdominal pain, and in some cases,cholecystectomy may be indicated for relief ofsymptoms.4

Dysplasia, carcinoma in situ, and invasive carcino-masmay arise from the epitheliumof adenomyom-atous hyperplasia. However, most authors believethat the cause for development of carcinoma inadenomyomatous hyperplasia is the presenceof stones, chronic inflammation, and metaplasticchanges rather thanadenomyomatoushyperplasiaitself. Thus, adenomyomatous hyperplasia is notconsidered a premalignant lesion.2-4

US is considered as the baseline imaging mo-dality to investigate the hepatobiliary system,

and adenomyomatosis of the gallbladder isa not-infrequent pathology, seen in approx-imately 2% of patients.1,5 Evaluation of gall-bladder wall thickness plays an important rolein the sonographic study of the biliary system.The gallbladder wall is no more than 2 mm thick in97% of healthy subjects, if the short axis of thegallbladder is at least 2 cm in diameter.5 Diffusegallbladder wall thickening has a differential diagno-sis that includesthemorecommoninflammatoryandnoninflammatory causes of wall thickening. Thevarious causes include hepatitis, artifacts inducedby the presence of pericholecystic fluid, hypoalbu-minemia, portal hypertension, right-sided heart fail-ure, incomplete gallbladder distention, cirrhosis,hepatitis, and cholecystitis. In addition, focalthickening occurs in gallbladder carcinoma andadenomyomatosis.5,6 However, gallbladder malig-nancyshouldbesuspectedwhen thereare featuresofa focalmasswith locoregional lymphadenopathy,metastases, and features of biliary obstruction atthe level of the porta hepatis.5

The pathognomonic pathologic features of theadenomyomatosis are thickened gallbladder walland intramural diverticula containing bile, withcholesterol crystals, sludge, or calculi that show adistinct correlation with the modality imaging find-ings.7 Longstanding Rokitansky-Aschoff sinuses ofadenomyomatoushyperplasiaresult incalcificationof the intramural sludge, cholesterol, or stones.Rarely, abdominal radiograph may show them asnondependent calcific opacities in the right upperquadrant.7

On ultrasonography, gallbladder involvement byadenomyomatous hyperplasia is variable in loca-tion and extent, which varies from focal to diffusegallbladder wall thickening.7 Luminal narrowing

A B

Fig 1 –

Ultrasonographic images ofgallbladder in a 69-year-oldman with righthypochondriac pain. (A)Longitudinalultrasonographic imageshows diffusely thickenedgallbladder wall (arrow)with associated multipleintramural calculi(arrowhead)andechogenicintramural foci with comet-tail reverberation artifacts,indicative of cholesterolcrystals within Rokitansky-Aschoff sinuses (star). (B)No obvious color uptake onpower Doppler.

Fig 2 –

Photomicrograph (originalmagnification, 320;hematoxylin and eosinstain) showed ulcerationsand edematous gallbladdermucosa and myoepithelialhyperplasia with abundantgranulation tissuecomposed ofmyofibroblasts.Inflammatory cells, such asneutrophils, plasma cells,and macrophages, werealso presentwith congestedvessels.



may be seen in the diffuse and segmental variantsthat may produce a characteristic hourglassconfiguration.

The diagnostic sonographic hallmark on ultraso-nography is the presence of the anechoic or echo-genic intramural diverticula. Anechoic diverticulacontain bile, whereas those that contain sludge,cholesterol, or intramural calculi are seen as echo-genic foci.4,7 Comet-tail/V-shaped reverberationartifact is due to echogenic sound reverberationemanating from the small echogenic foci in thegallbladder wall into the central anechoic lumen.Similar appearance may be caused by intraluminalor intramural air, as seen in emphysematous cho-lecystitis; however, the mobile nature of the airwith dirty echogenic shadowing on ultrasonographyhelps in differentiating them. In equivocal cases,computed tomography (CT) plays an important role.

CT findings of adenomyomatous hyperplasia usu-ally correlate to the sonographic findings. All threepathologic forms (localized, segmental, and dif-fuse) are apparent on CT scans.4,8 Intramuralcystic spaces and calculi demonstration on CTscan is not possible, but they are easily picked upsonographically because of reverberation artifact.This makes ultrasonography a primary diagnosticimagingmodality for thediagnosis of adenomyom-atous hyperplasia. The Rokitansky-Aschoff sinusesthat are large enough may be seen on CT scan,giving a characteristic appearance, the rosary sign,which is due to the enhancing epithelium withinintramural diverticula surrounded by the relativelyunenhanced hypertrophied gallbladder muscularis.9

Magnetic resonance imaging (MRI) plays an im-portant role in distinguishing adenomyomatoushyperplasia from malignant lesions; hyperintenseintraluminal cavities visualized on T2-weightedMRI are suggestive of Rokitansky-Aschoff sinusesand have been reported to be useful.10 Whenarranged in a curvilinear pattern, they give anappearance of pearl necklace sign.11 The intra-mural calculi are seenassignal voidbecauseof themineral content. The pearl necklace sign is spe-cific for adenomyomatosis of the gallbladderand isnot seen in gallbladder carcinoma.11 Differentia-tion of echogenic intramural foci from abnormalenhancement requires a multiphasic MRI proto-col with intravenous contrast material. Diffusionweighted imaging has been found useful in differ-entiating gallbladder adenomyomatosis from gall-bladder carcinoma and significantly improves thediagnostic accuracy.11a

In our case, US revealed an abnormally thickgallbladder wall measuring 3 cm in thickness.

Table 1 – Summary Table for Clinicoradiopathological Features of GallbladderAdenomyomatosis

Etiopathogenesis Adenomyomatosis is a rare disease of thegallbladder characterized by epithelialproliferation and formation of mucosalpouches through the thickenedmuscularlayerof thegallbladderwall—theso-calledRokitansky-Aschoff sinus. Despite thename, this condition does not involve anyadenomatous changes in the gallbladderepithelium and is not a premalignantcondition.

Incidence 1%-8.5%

Sex ratio Male:female = 1:3

Age predilection Predominantly fifth decade

Risk factor None known; mostly an incidental finding.

Clinical presentation The clinical picture can vary fromasymptomatic (60%) to symptomaticpatients. These symptoms can bedyspepsia, pain in the righthypochondrium, fever of unknown origin,acute or chronic cholecystitis.

Treatment The treatment of choice of these tumors issurgical resection, both in symptomaticand asymptomatic cases, accompaniedor not by cholelithiasis, because of theuncertain evolution of this disease and thedifficulty of differential diagnosis withcarcinoma. Cholecystectomy is suitablewhen the process is located in thegallbladder. When the tree biliary isaffected, surgical resection with freemargins and enterobiliary derivation isindicated

Prognosis Favorable

Ultrasound Mural thickening: can be diffuse, focal orannular

Comet-tail artifact: echogenic intramural focifrom which emanate V-shaped comet-tailreverberation artifacts.

CT CT rosary sign as a result of enhancingepithelium within intramural diverticulasurrounded by the relatively unenhancedhypertrophied gallbladder muscularis

MRI Hyperintense on T2-weighted images andhypointense on T1-weighted images;pearl necklace sign as a result of thecharacteristically curvilinear arrangementof multiple rounded hyperintenseintraluminal cavities visualized at T2-weighted MRI and magnetic resonancecholangiopancreatography ofadenomyomatosis. On diffusion weightedimaging, the cystic areas of theadenomyomatosis show high signalintensity.

Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging.



The wall thickening gave an appearance of endo-phytic mass causing near-total luminal oblitera-tion. Multiple intramural and intraluminal calculiwere seen with few echogenic intramural foci,which did not show the classic comet-tail rever-beration artifacts. The cause for the diagnosticdilemma was the absence of the comet-tail/V-shaped reverberation artifact, which is the diag-nostic sonographic hallmark of adenomyomato-sis. Its absence can be explained on the basis ofthe absence of central anechoic gallbladder lu-men because of gross dense fibrous wall thicken-ing causing near-total luminal obliteration.12

Although the imaging features of adenomyomatosiscan be distinctive enough to allow confident diag-nosis, findings such as gallbladder wall thickeningand enhancement are somewhat nonspecific. Thedifferential diagnosis includes adenomatous, hyper-plastic, and cholesterol polyposis, papillomatosis,adenoma, portal hypertension, total parenteral nu-trition, hypoalbuminemia, congestive heart failure,and cystadenoma, as well as malignancies such asgallbladder adenocarcinoma, carcinoid tumor, andmetastatic melanoma of the gallbladder (Table 2).

Metabolic characterization with [18F]fluorodeoxy-glucose (FDG) positron emission tomographymay be a useful adjunct in problematic cases.13

Adenomyomatosis does not show any FDG uptakeon positron emission tomography study, whichmakes it a problem-solving tool in differentiatingit frommalignantgallbladderwall thickening,whichshows focal or diffuse FDG uptake.13 Though theFDG uptake within these lesions represents themetabolic activity of the underlying pathologicaltissue (cancerous cells), false positive resultsmay also be seen in cases with associated activeinflammatory process such as cholecystitis.14

In conclusion, great difficulty arises in differenti-ating the diffuse variant of adenomyomatous hy-perplasia that simulates a gallbladder malignancy,especially in the absence of the diagnostic hallmarkcomet-tail/V-shapedreverberationartifact,whichmaybe absent in cases of total luminal obliteration by thepathology. Surgical resection may be indicated insymptomatic cases and when nonspecific find-ings present a diagnostic dilemma.

DOI: 10.1200/JGO.2016.005249Published online on jgo.ascopubs.org on June 15, 2016.

Table 2 – Differential Diagnosis for Adenomyomatosis of the Gallbladder and Their Imaging Characteristics

Diagnosis Ultrasound CT MRI

Gallbladderadenomyomatosis

Echogenic intramural foci from whichemanate V-shaped comet-tailreverberation artifacts

CT rosary sign as a result of enhancingepithelium within intramuraldiverticula surrounded by therelatively unenhanced hypertrophiedgallbladder muscularis

Rokitansky-Aschoff sinuses asintramural lesions that arehyperintense on T2-weightedimages, hypointense on T1-weightedimages, and nonenhancing. Ondiffusionweighted imaging, thecysticareas of the adenomyomatosis showhigh signal intensity, whichrepresents the Rokitansky-Aschoffsinuses.

Pearl necklace sign

Gallbladder carcinoma May completely fill the gallbladderlumen

Hypoattenuatingor isoattenuatingmassin the gallbladder fossa

The tumor is usually bright onT2-weighted images and is poorlymarginated. On T1-weighted images,relative to the liver, the gallbladdercarcinoma ranges from isointense tohypointense. Ill-defined earlyenhancement is a typical appearanceof these tumors at dynamicgadolinium-enhanced MRI. Ondiffusion weighted imaging, thetumor shows lower signal intensity onthe ADC maps.

Irregular margins and heterogeneousechotexture

Echogenic foci and acoustic shadowingas a result of coexisting gallstones,gallbladder wall calcification, ortumoral calcification

Pronounced wall thickening . 1.0 cmwith associated mural irregularity ormarked asymmetry

Direct extension into adjoining liverparenchyma

Cholelithiasis The characteristic findings of gallstonesare a highly reflective echo from theanterior surface of the gallstone,mobility of the gallstone onrepositioning the patient (typically ina decubitus position), and markedposterior acoustic shadowing.

Gallstones appear as single or multiplefilling defects within the gallbladderand are densely calcified, rimcalcified, or laminated or havea central nidus of calcification.

Most gallstones have no signal on MRIand present as signal-void fillingdefects within the gallbladder. Theseare most apparent on T2-weightedsequences where signal-void stonesare contrasted against high-signalbile.

Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging.








The following represents disclosure information provided byauthors of this manuscript. All relationships are consideredcompensated. Relationships are self-held unless noted. I =

Immediate Family Member, Inst = My Institution. Relation-ships may not relate to the subject matter of this manuscript.For more information about ASCO’s conflict of interest policy,please refer to www.asco.org/rwc or jco.ascopubs.org/site/ifc.

Abhishek MahajanNo relationship to disclose

Smiti SripathiNo relationship to disclose

REFERENCES1. Ootani T, Shirai Y, Tsukada K, et al: Relationship between gallbladder carcinoma and the segmental type of ade-

nomyomatosis of the gallbladder. Cancer 69:2647-2652, 1992

2. Albores-Saavedra J, Hensen DE, Klimsta DS: Tumors of the gallbladder, extrahepatic bile ducts, and ampulla of Vater:Atlas of tumor pathology, fasc 27, ser 3. Washington, DC, Armed Forces Institute of Pathology, 2000

3. Albores-Saavedra J, Henson DE: Tumors of the gallbladder and extrahepatic bile ducts: Atlas of tumor pathology, fasc22, ser 2. Washington, DC, Armed Forces Institute of Pathology, 1986

4. Hwang JI, Chou YH, Tsay SH, et al: Radiologic and pathologic correlation of adenomyomatosis of the gallbladder.Abdom Imaging 23:73-77, 1998

5. Finberg HJ, Birnholz JC: Ultrasound evaluation of the gallbladder wall. Radiology 133:693-698, 1979

6. Levy AD, Murakata LA, Abbott RM, et al: From the archives of the AFIP. Benign tumors and tumorlike lesions of thegallbladder and extrahepatic bile ducts: Radiologic-pathologic correlation. Radiographics 22:387-413, 2002

7. Raghavendra BN, Subramanyam BR, Balthazar EJ, et al: Sonography of adenomyomatosis of the gallbladder:Radiologic-pathologic correlation. Radiology 146:747-752, 1983

8. Gerard PS, Berman D, Zafaranloo S: CT and ultrasound of gallbladder adenomyomatosis mimicking carcinoma.J Comput Assist Tomogr 14:490-491, 1990

9. Chao C, Hsiao HC,Wu CS, et al: Computed tomographic finding in adenomyomatosis of the gallbladder. J FormosMedAssoc 91:467-469, 1992

10. Yoshimitsu K, Honda H, Jimi M, et al: MR diagnosis of adenomyomatosis of the gallbladder and differentiation fromgallbladder carcinoma: Importance of showing Rokitansky-Aschoff sinuses. AJR Am J Roentgenol 172:1535-1540,1999

11. Haradome H, Ichikawa T, Sou H, et al: The pearl necklace sign: An imaging sign of adenomyomatosis of thegallbladder at MR cholangiopancreatography. Radiology 227:80-88, 2003

11a. Tomizawa M, Shinozaki F, Fugo K, et al: Negative signals for adenomyomatosis of the gallbladder upon diffusion-weighted whole body imaging with background body signal suppression/T2-weighted image fusion analysis. ExpTher Med 11:1777-1780, 2016

12. LafortuneM, Gariepy G, Dumont A, et al: The V-shaped artifact of the gallbladder wall. AJR Am J Roentgenol 147:505-508, 1986

13. HammadAY,Miura JT, Turaga KK, et al: A literature review of radiological findings to guide the diagnosis of gallbladderadenomyomatosis. HPB 18:129-135, 2016

14. Mahajan A, Goh V, Basu S, et al: Bench to bedside molecular functional imaging in translational cancer medicine: Toimage or to imagine? Clin Radiol 70:1060-1082, 2015





Documents

OCTOBER 2016 Volume 2, Issue 5 assets/JGO_2_5.pdf · Compulsory Licenses for Cancer Drugs: Does Circumventing Patent Rights Improve Access to Oncology ... Surendra S. Shastri, MD,