Newsletteron Human

PapillomavirusWWW.HPVTODAY.COM

1

HPV SCREENING ENDORSED BY SCIENTIFIC SOCIETIES AND REGULATORY AGENCIES IN THE UNITED STATES AND EUROPE

Nº 38

08/2015

2015ATHENA TRIAL

SPECIAL ISSUE

It seems that the time is now mature for a gen-

eralized change in the cervical cancer prevention

paradigm. How do you see the process and the

milestones in the last decade?

I certainly agree that over the last decade or so, the

maturation of the data we have for evaluating choices

in cervical cancer screening demands that each society

evaluate the data compared to historic norms. In the

US, the process began in 2001 with the ASCUS LSIL

Triage Study (ALTS) that clinically validated the con-

cept of HPV testing for triage of equivocally abnormal

cytology. The data from that study stimulated a global

conversation about the relative performance of HPV

versus cytology for screening. In parallel, the proof of

effi cacy in 2006 of HPV vaccines for the prevention

of cervical cancer has led to the concept of potential

cervical cancer eradication strategies using both vac-

cination and more optimized screening. The impact

of vaccination and the improved screening sensitivity

of HPV testing –including the US ATHENA trial and

the large scale European experiences– dominates the

conversation now with the recognition that screening

must evolve to continue to be effective.

What is the importance of the FDA resolution on

HPV primary screening?

If one follows current US news reporting, science or

the lack of appreciation of science, even the repudia-

tion of science, is constantly in the news. Climate

change and vaccine safety versus medical policy are

very hot topics. Medical policy in general continues

to be of great national interest. In this context, the

(cont. page 3) interview with

MARK H. STOLER

2

SCIENTIFIC COMMITTEE

EDITORIAL COMMITTEE

EDITORIAL STAFF

General Coordinator: F. Xavier Bosch (Spain)

International Coordinators: Xavier Castellsagué (Spain) Patti Gravitt (USA)

Coordinators for Spain: Silvia de SanjoséXavier Cortés

Coordinator for Portugal and Brazil: Clara Bicho (Portugal)

Coordinator for Germany: Karl Ulrich Petry

Coordinator for France:Christine Clavel

Coordinator for Italy:Flavia Lillo

Coordinator for Russia and NIS:Svetlana I. Rogovskaya

Coordinators for Latin America:Eduardo Lazcano (Mexico) Silvio Tatti (Argentina)

Coordinator for Asia-Pacific Area:Suzanne Garland (Australia)

Coordinator for China:You-Lin Qiao

Coordinator for Japan:Ryo Konno

Th Agorastos (Greece)L Alexander (USA)Ch Bergeron (France)HV Bernard (USA)JC Boulanger (France)T Broker (USA)Ll Cabero (Spain)S Campo (Scotland)P Coursaget (France)T Cox (USA)J Cuzick (UK)Ph Davies (UK)L Denny (South Africa)S Dexeus (Spain)E Diakomanolis (Greece)A Ferenczy (Canada)S Franceschi (France)E Franco (Canada)I Frazer (Australia)L Gissmann (Germany)S Goldie (USA)F Guijon (Canada)A Guerra (Spain)M Hernández (Mexico)R Herrero (Costa Rica)T Iftner (Germany)I-Wuen Lee (Singapore)D Jenkins (UK)A Jenson (USA)WM Kast (USA)V Késic (Yugoslavia)S Krüger Kjaer (Denmark)R Kurman (USA)Ch Lacey (UK)CJLM Meijer (The Netherlands)J Monsonego (France)L Olmos (Spain)G de Palo (Italy)H Pfister (Germany)L Pirisi-Creek (USA)R Prado (Chile)W Prendiville (Ireland)Ll M Puig-Tintoré (Spain)T Rohan (USA)R Richart (USA)S Robles (USA)P Sasieni (UK)J Schiller (USA)KV Shah (USA)J Sherris (USA)A Singer (UK)P Snijders (The Netherlands)M Stanley (UK)M Steben (Canada)P Stern (UK)S Syrjanen (Finland)R Testa (Argentina)M Tommasino (France)M van Ranst (Belgium)L Villa (Brazil)R Viscidi (USA)G Von Krogh (Sweden).

Website: www.hpvtoday.com

Correspondence and collaborations: E-mail: box@hpvtoday.com

Copyright: All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the written permission of the copyright holder.© Fundación para el Progreso de la Educación y la Salud (FPES). The intellectual responsibility of the signed contributions is primarily of the authors and does not necessarily reflect the views of the editorial or scientific committees.

Alejandro Santos, Jesús Muñoz, Cristina Rajo.

Published by: Fundación para el Progreso de la Educación y la Salud (FPES)C/ Serrano 240, 5, 28016 Madrid. SPAIN.

Legal deposit: M-35437-2001

ISSN: 1885-9291

INDEX

interview with Mark H. Stoler 1

HPV SCREENING ENDORSED BY SCIENTIFIC SOCIETIES AND REGULATORY AGENCIES IN THE UNITED STATES AND EUROPE

REPORTSUNDERSTANDING THE LIMITATIONS OF CYTOLOGY FOR SCREENING – Philip E Castle

5

KEY FINDINGS OF THE ATHENA STUDY – Thomas C Wright and Catherine M Behrens

9

TRIAGING HPV-POSITIVE WOMEN – Nicolas Wentzensen 13

THE ROLE OF EPIDEMIOLOGY IN DISCERNING OPTIMAL CERVICAL CANCER SCREENING STRATEGIES – Julia C Gage

17

THE INTERIM GUIDANCE TO HPV SCREENING – Warner K Huh 20

HPV TESTING IN THE UNITED STATES: PERSPECTIVES FROM SYSTEMS, PROVIDERS AND WOMEN – Vicki Benard and Mona Saraiya

22

THE AUSTRALIAN EXAMPLE: AN INTEGRATED APPROACH TO HPV VACCINATION AND CERVICAL SCREENING – Karen Canfell

24

international agenda 28

FREE ELECTRONIC SUBSCRIPTION AT WWW.HPVTODAY.COM

RECEIVE IT IN YOUR E-MAIL OR RETRIEVE ANY PREVIOUS ISSUES IN ENGLISH, SPANISH, RUSSIAN OR JAPANESE

HPV TODAY is an independent publication created by a voluntary group of scientists and clinicians, and edited and distributed freely by the Foundation FPES (Foundation for the Advancement in Education and Health) thanks to unrestricted educational grants from the HPV related industry. Opportunities for sponsorship, request for printed copies, meeting advertising and correspondence: box@hpvtoday.com

3

interview

(from page 1)

Federal Drug Administration (FDA) approval of an

algorithm for cervical cancer screening that starts

with HPV testing rather than traditional cytology

is a true testimony to the fact that data driven

science can actually produce a potential political or

philosophical outcome that agrees with the data.

But of course, not everyone agree...

What should be the expected gains of the

introduction of HPV testing into primary

screening protocols?

Simply put, the expected gains should be better

health for women who are screened. Secondarily

one might expect suffi cient recovery of resources

due to the effi ciencies of the algorithm to actu-

ally be able to extend screening to populations

who have not been screened. The push back to

these ideas centers mainly on the concern for the

potential harms of HPV screening by identifying

women with precancer that may not progress

over the screening

interval to cancer, risk-

ing over-treatment.

Yet, as should always

be pointed out, HPV

primary screening is

always coupled to a

triage test to focus

treatment on those

that need it most.

Furthermore, screening

for cervical cancer only has utility if one fi nds the

precancer in the population and treats it to prevent

cancer development. The focus of the current

debates are how best to identify that population.

The US is largely adopting a co testing strat-

egy (cytology and HPV tests) for primary

screening whereas Europe tends to favor HPV

screening alone and cytology as one of the

triage options. How do you interpret these

different resolutions?

I don’t think the distinctions are quite so clear.

The US, unlike Europe has a long tradition of

relative lack of cost-constraint and relative over-

utilization of testing. But fundamentally, I think

both European and US physicians as well as

women, want the same thing, namely safety from

cervical cancer. The difference is now we have an

abundance of data to interpret and decide how

best to achieve that goals while balancing the good

with the bad. Furthermore, the conversation in the

US has only just begun. No clinical practice in the

US could even consider primary HPV testing until

the interim guidance was published. In contrast, we

have had “permission” to do co-testing for more

than 10 years in US guidelines.

Non participants in screening activities remain

one of the most vexing population from which

cervical cancer mortality persists. Any devel-

opments on the self-sampling front that may

increase screening coverage in the US?

As noted above, lack of screening is the major

US risk factor for cervical cancer. More than half

of the cancer in the US is found in unscreened or

inadequately screened women. In some states

and counties in the US, the incidence of cervi-

cal cancer is similar to

that in Sub-Saharan

Africa. The barriers to

screening are manifold

in the US especially

in the absence of an

organized screening

program. Given recent

published data, self-

sampling approaches

could well have a major

impact on cervical cancer incidence in under-

screened women. But that conversation has yet to

be started in any meaningful way. The good news

is the recent US health care policy (Obamacare),

mandates access to free cervical cancer screening

for all insured individuals in the US.

The US is increasing HPV vaccination coverage.

Do you anticipate any change in the screening

protocols among vaccinated cohorts?

The mathematics of screening is driven by

prevalence. As vaccine coverage increases

the prevalence of the targets of screening

will decrease, demanding that more sensitive

algorithms be applied. In this regard, the docu-

mentation of the ongoing Australian experience

leads the way, with many European countries not

Mark H. StolerProfessor(Emeritus) of Pathology and Clinical Gynecology. Associate Director of Surgical and Cytopathology. Department of Pathology. University of Virginia Health System. Charlottesville. USA

"The FDA approval of an algorithm for cervical cancer screening that starts

with HPV testing rather than traditional cytology is a true testimony to the fact that data driven science can actually produce

a potential political or philosophical outcome that agrees with the data."

4

far behind. HPV vaccination is absolutely a primary

driver for the necessary adoption of HPV primary

screening. Let us not forget that these arguments will

only be stronger once the just approved nonavalent

HPV vaccine penetrates the market.

Would you recommend integrated protocols of

vaccination and screening for each age strata

along women's lifetime? If so what would they

look like?

Obviously, the HPV community has labored long

and hard to try and bring primary prophylaxis and

improved data driven secondary prevention through

screening to the world’s women. The detail of how

to best spread the fruits of these labors is unfortu-

nately a complex economic and political question,

less so a medical or scientifi c one except in how the

science and political economics collide. That being

said, I believe every population, male and female

deserves protection from HPV-related disease.

Universal and sustained prophylactic vaccination of

our children, now with the nonavalent vaccine would

drive HPV disease related prevalence rates down

to a level where screening would mathematically

be impossible. But for the next few decades, we

need a transition strategy of screening all women

in populations where the prevalence of disease still

allows screening to work. We now have the tech-

nology to potentially bring affordable state of the

art molecular screening at rational and infrequent

intervals to populations where the economics and

infrastructure of traditional cytology based screen-

ing would be impossible to implement and sustain.

The details of any such integrated program really

have to be determined on a regional or local basis.

And of course screening must be coupled to treat-

ment or screening is a waste of time.

Any fi nal comments?

We are at the dawn of a new era in cervical cancer

prevention, where we can now even talk about eradi-

cation, and the elimination of screening. In developed

countries we are perhaps a bit too focused on the

details of optimization or how to implement change.

While the idea that screening requires a balancing of

the benefi ts with the potential harms, in my opinion

the biggest harm is not screening. I will also state

it yet again, if we don’t treat precancer, screening

will have no impact on women’s suffering from

cervical cancer. Hence in my opinion the scales

tip in favor of eradicating true precancer. The

scientifi c community will work in the next decade

to develop better biomarkers to better refi ne the

target population needing treatment. Meanwhile,

the HPV vaccination programs will continue to

drive down the prevalence of precancer and

cancer. We are beginning to realize our dreams…

PROGRESS IN SCREENING TECHNOLOGIES

Conventional cytology

Monolayer liquid-based cytology

Normal cytology with double HPV 52 and 31 infections

interview

Dr. Mark H. Stoler has served as a consultant in clinical trial design and as an expert pathologist for HPV vaccine and/or diagnostic trials for Roche, Ventana Medical Systems, Hologic/Gen-Probe, Becton Dickinson, Cepheid, Qiagen, Inovio and Merck.

5

REPORT

UNDERSTANDING THE LIMITATIONS OF CYTOLOGY FOR SCREENING

Philip E. Castle, PhD, MPH1,2 1 Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY, USA.2 Global Coalition Against Cervical Cancer, Arlington, VA, USA.

Cervical cancer screening has been the most

successful of all cancer screenings and in many

ways remains the exception. The reasons for its

success have been many-fold. Cervical cancer

occurs in a highly localized region of the cervix,

the squamocolumnar junction, and this zone is

relatively accessible (unlike for virtually all other

cancers) making sampling and treatment relatively

easy and accurate. And because invasive cervical

cancer develops slowly from a precancerous lesion,

on average over a time period of 1-2 decades,1

there are many opportunities to intervene (screen,

diagnose, and treat) prior to invasion.

Papanicolaou or Pap testing, invented by Dr.

George Papanicolaou in the mid-20th century,

was the fi rst cervical cancer screening test. Pap

testing, or cervical cytology, is the process of

microscopic assessment of exfoliated cervical cells

for morphologic changes indicative of neoplastic

alterations. Where cytology-based screening

has been effectively implemented, e.g. United

States,2,3 United Kingdom,4 Nordic countries,5 the

Netherlands,2 New Zealand,4 and Australia4 to name

a few, cervical cancer incidence and mortality have

declined signifi cantly because of the timely detec-

tion and treatment of cervical precancerous lesions

and progressive down staging of invasive cervical

cancer. It is not hyperbole to say that Pap testing

has saved millions of lives.

Yet, despite its successes, Pap testing has a

number of well-known limitations. Pap testing has

only moderate one-time sensitivity for cervical

precancer (cervical intraepithelial neoplasia grade

2 [CIN 2], grade 3 [CIN 3], and adenocarcinoma in

situ [AIS]) and cancer. Thus, Pap testing requires

many repeat screenings in a lifetime to achieve pro-

grammatic effectiveness. In particular, cytology has

poor sensitivity for detection of adenocarcinoma

precursors, e.g. AIS, perhaps because of poor

sampling of the lesions higher in the endocervical

canal. As a consequence, in the context of secular

trends of increased exposure to HPV, annual rates

of adenocarcinoma have not declined signifi cantly

in most countries and have even increased in some

over the last few decades.2;6-8

Some of the limitations of Pap testing are due in

part to its subjective and laborious nature. As a

consequence, it has only fair-to-poor reproduc-

ibility (inter-rater agreement).9 And because of its

laborious nature, there is a limited number that any

one reader can review per day so many readers

are needed to meet the demands of a high-volume

"It is not hyperbole to say that Pap testing has saved millions of lives."

Figure. Screening with cervical cytology has reduced incidence and mortality from cervical cancer in countries with signifi cant population coverage. Incidence reduction tends to stagnation primarily due to non participation and secondarily to limita-tions of cytology, notably to detect cervical adenocarcinomas.

1950 1960 1970 1980 1990 2000

40

35

30

25

20

15

10

5

0

INCIDENCE OF INVASIVE CERVICAL CANCER US -FROM SEER

6

clinical laboratory diagnostic, adding to overall

variability in performance. Thus, in order to achieve

high-quality Pap testing, signifi cant investments in

infrastructure and extensive quality assurance and

control measures are required.

High-quality, high-throughput cytology-based

screening is therefore best achieved through an

organized screening program. As reported by

Simonella and Canfell4, greater reductions in the

annual cervical cancer incidence and mortality were

achieved in Australia, New Zealand, and England

when each country switched from opportunistic

screening to organized screening. In the US, where

screening is opportunistic, the cervical cancer

screening program is ineffi cient and costs more

than $6 billion per annum. Comparing the US to the

Netherlands, which have experienced comparable

reductions in cervical cancer incidence and mortal-

ity, women from the US undergo 3- to 4-fold more

Pap tests than women from the Netherlands.3

Therefore, women living in the US potentially

experience a much greater burden of the harms of

cervical cancer screening10 than those living in the

Netherlands.

However, a new paradigm of targeting HPV, the obli-

gate, viral cause of cervical cancer and its immediate

precursor lesions, for cervical cancer prevention is

emerging. Technical advances, including prophy-

lactic vaccination against certain high-risk HPV

(hr HPV) types for primary prevention and hrHPV

testing for cervical cancer screening secondary

prevention, are highly effi cacious and when used in

an age-appropriate manner, highly cost effective.11,12

Hr HPV testing13-18 and reliable19,20 than Pap testing

for the detection of cervical precancer and cancer.

Importantly, a negative hrHPV test provides greater

reassurance against cervical precancer and cancer

than Pap,21-23 safely permitting longer intervals

between screens or increased safety for similar

interval. A single round of hrHPV testing was more

effective than Pap testing in reducing cervical cancer

incidence in 6.5 years23 and in reducing mortality

due to cervical cancer in 8 years.24

High-risk HPV testing is best used only as the

screening test, to rule-out disease, and clinical

decisions should not be made based on a single

hrHPV-positive result alone, since most hrHPV

positive women do not have and will not develop

cervical precancer or cancer. The application of

Pap testing or potentially other markers, including

biomarker-enhanced cytology using p16/Ki-67

immunocytochemistry,25 can be limited to the

at-risk hrHPV positive group to “rule-in” i.e., to

determine what kind of care is needed.

Indeed, the Pap result, because of its inherent

ability to grade cytologic severity, unlike the

hrHPV result, can further stratify risk for clinical

decision making on what kind of care is needed

in HPV-positive women:26

A) those with negative cytology are monitored

closely until there is evidence of short-term

hrHPV persistence, a strong risk factor for

concomitant or future cervical precancer and

cancer;27,28

B) those with mild Pap abnormalities (e.g., atypical

squamous cells of undetermined signifi cance

[ASC-US] or low-grade squamous intraepithelial

lesion [LSIL]) are referred to colposcopy with

the noted exception of young women, who are

less likely to have cervical precancer and more

likely to have a transient hrHPV infection, and

C) severe Pap abnormalities (e.g., high-grade

squamous intraepithelial lesion [HSIL]) are

"As reported by Simonella and Canfell, greater reductions in the annual cervical

cancer incidence and mortality were achieved in Australia, New Zealand, and

England when each country switched from opportunistic screening to

organized screening."

"A single round of hrHPV testing was more effective than Pap testing in

reducing cervical cancer incidence in 6.5 years and in reducing mortality due

to cervical cancer in 8 years."

7

UNDERSTANDING THE LIMITATIONS OF CYTOLOGY FOR SCREENING

Pap Screening Unvaccinated Populations Vaccinated Populations

In all women

CIN 2+ < CIN 2 Total Se= 50.0% CIN 2+ < CIN 2 Total Se= 50.0%Pap+ 500 4,500 5,000 Sp= 95.5% Pap+ 250 3,500 3,750 Sp= 96.5%

Pap- 500 94,500 95,000 PPV= 10.0% Pap- 250 96,000 96,250 PPV= 6.7%Total 1,000 99,00 100,000 NPV= 99.5% Total 500 99,500 100,000 NPV= 99.7%

Triage of hrHPV Positives (cytology blinded)

CIN 2+ < CIN 2 Total Se= 50.0% CIN 2+ < CIN 2 Total Se= 45,0%Pap+ 450 2,050 2,500 Sp= 77.5% Pap+ 225 1,650 1,875 Sp = 76,4%

Pap- 450 7,050 7,500 PPV= 18.0% Pap- 225 5,400 5,625 PPV = 12,0%Total 900 9,100 100,000 NPV= 94.0% Total 450 7,050 7,500 NPV = 95,1%

Triage of hrHPV Positives (cytology informed)

CIN 2+ < CIN 2 Total Se= 80.0% CIN 2+ < CIN 2 Total Se= 80.0%Pap+ 720 1,905 2,625 Sp= 79.1% Pap+ 360 1,609 1,969 Sp= 77.2%

Pap- 180 7,195 7,375 PPV= 27.4% Pap- 90 5,441 5,531 PPV= 18.3%Total 900 9,100 10,000 NPV= 97.6% Total 450 7,050 7,500 NPV= 98.4%

referred to colposcopy but might, with informed

decision-making between patient and provider,

lead to excisional treatment without histologic

confirmation of cervical precancer.

Shifting Pap testing from all women to the

at-risk, hrHPV-positive women improves the

performance of Pap testing in two ways:

• First, enriching for (increasing prevalence of) cer-

vical precancer and cancer algebraically results

in a better positive predictive value.29

• Second, informing cytologists that the Pap is

from hrHPV-positive women, “informed screen-

ing” or “screening with prejudice”, may increase

sensitivity without significantly decreasing the

specificity.30

The latter needs to be reproduced in other settings

to show that it is a generalizeable phenomenon.

This approach, rule-out with hrHPV testing and

rule-in with Pap testing (or another marker) or

evidence of persistent hrHPV infection, is not only

more efficient, but it becomes necessary in the

context of secular trends of decreasing prevalence

of cervical precancer and cancer due to screening

and now HPV vaccination. Of the latter, HPV vac-

cination against HPV16 and HPV18 in HPV-naïve

populations is expected to reduce the prevalence

of CIN 2/3 by ~50% and cervical cancer by 70%.

Even in the US, where HPV vaccination coverage

is embarrassing low,31 there is already evidence of

reduced prevalence of HPV16 and HPV18 infec-

tions32 and HPV16 and HPV18-related CIN 2/3.33

Table 1 presents the theoretical performance

of Pap testing in the general population and in

a vaccinated population, among all women and

hrHPV-positive women, and the latter with or

without screening with prejudice. The positive

predictive value (PPV) of Pap testing at a positive

Pap sensitivity = 50% hrHPV testing sensitivity = 90%Prevalence of CIN2+ in an unvaccinated population = 1%; Prevalence of CIN2+ in an HPV-naïve population vaccinated against HPV16 and HPV18 = 1%Prevalence of hrHPV in an unvaccinated population = 10%; Prevalence of hrHPV in an HPV-naïve population vaccinated against HPV16 and HPV18 = 7.5%Prevalence of ASC-US+ in an unvaccinated population = 5%; Prevalence of ASC-US+ in an HPV-naïve population vaccinated against HPV16 and HPV18 = 3.75%Informed screening or screening with prejudice increases sensitivity by 60% and decreases specificity by 5%.30

Table 1. Hypothetical positive predictive values (PPV) for cytology in HPV unvaccinated and vaccinated populations of women, as a general screen and as triage of high-risk HPV (hrHPV) positive women without and with a priori knowledge that the slides are from hrHPV-positive women.

8

References: 1. Moscicki AB et al. Vaccine 2012;%20;30 Suppl 5:F24-33. doi: 10.1016/j.vaccine.2012.05.089.:F24-F33. 2. Wang SS et al. Cancer 2004;100:1035-1044. 3. Habbema D et al. Milbank Q 2012;90:5-37. 4. Simonella L, Canfell K. Cancer Causes Control 2013;24:1727-1736. 5. Vaccarella S et al. Br J Cancer 2014;111:965-969. 6. Bray F et al. Cancer Epidemiol Biomarkers Prev 2005;14:2191-2199. 7. Liu S et al. CMAJ 2001;164:1151-1152. 8. Liu S et al. Int J Gynecol Cancer 2001;11:24-31. 9. Stoler MH, Schiffman M. JAMA 2001;285:1500-1505. 10. Saslow D et al. CA Cancer J Clin 2012;62:147-172. 11. Kim JJ et al. Ann Intern Med 2009;151:538-545. 12. Goldhaber-Fiebert JD et al. J Natl Cancer Inst 2008;100:308-320. 13. Cuzick J et al. Int J Cancer 2006;119:1095-1101. 14. Mayrand MH et al. N Engl J Med 2007;357:1579-1588. 15. Naucler P et al. N Engl J Med 2007;357:1589-1597. 16. Rijkaart DC et al. . Lancet Oncol 2012;13:78-88. 17. Ronco G, Giorgi-Rossi P, Carozzi F et al. Effi cacy of human papillomavirus testing for the detection of invasive cervical cancers and cervical intraepithelial neoplasia: a randomised controlled trial. Lancet Oncol 2010. 18. Castle PE et al. Lancet Oncol 2011;12:880-890. 19. Castle PE et al. Am J Clin Pathol 2004;122:238-245. 20. Carozzi FM et al. Am J Clin Pathol 2005;124:716-721. 21. Dillner J et al. BMJ 2008;337:a1754. doi: 10.1136/bmj.a1754.:a1754. 22. Castle PE et al. Clinical Human Papillomavirus Detection Forecasts Cervical Cancer Risk in Women Over 18 Years of Follow-Up. J Clin Oncol 2012. 23. Ronco G et al. Lancet 2014;383:524-532. 24. Sankaranarayanan R et al. N Engl J Med 2009;360:1385-1394. 25. Ikenberg H et al. J Natl Cancer Inst 2013;105:1550-1557. 26. Massad LS et al. J Low Genit Tract Dis 2013;17:S1-S27. 27. Castle PE et al. BMJ 2009;339:b2569. doi: 10.1136/bmj.b2569.:b2569. 28. Kjaer SK et al. J Natl Cancer Inst 2010;102:1478-1488. 29. Wentzensen N, Wacholder S. Cancer Discov 2013;3:148-157. 30. Bergeron C et al. J Natl Cancer Inst 2015;107:dju423. 31. Elam-Evans LD et al. MMWR Morb Mortal Wkly Rep 2014;63:625-633. 32. Markowitz LE et al. J Infect Dis 2013;208:385-393. 33. Hariri S et al. Vaccine 2015;10. 34. New guidelines on screening and treatment for cervical cancer. 2013. South Africa, World Health Organization. Ref Type: Pamphlet.

Dr. Castle has received commercial HPV tests for research at a reduced or no cost from Roche, Qiagen, Norchip, Arbor Vita Corporation, BD, and mtm. He has been compensated fi nancially as a member of a Merck Data and Safety Monitoring Board for HPV vaccines. Dr. Castle has been a paid as consultant for BD, Gen-Probe/Hologic, Roche, Cepheid, ClearPath, Guided Therapeutics, Teva Pharmaceutics, Genticel, Inovio, and GE Healthcare. Dr. Castle has received honoraria as a speaker for Roche and Cepheid.

Ris

k of

CIN

2+

UnvaccinatedVaccinated

Pap

Triage

Triage*

30

25

20

15

10

5

0

%

Figure 1. A graphic representation of the positive predictive value (risks) for CIN2 or more severe diagnoses (CIN2+) as shown in Table 1. The risks are shown for Pap test-ing among the general population and as a triage test among HPV-positive women without and with informed screening.*30

cutpoint of ASC-US in a vaccinated population is

6.7% i.e. only 1 of 16 Pap-positive women will have

CIN 2+ (and 1 of 32 will have CIN 3+). The PPV of

Pap doubles when restricted to hrHPV-positive

women and triples when slides from hrHPV-positive

women are screened with prejudice.

The introduction of Pap testing in the mid-20th

century was a major public health intervention

and the fi rst and most successful cancer screen.

Pap testing led to the discovery of new and more

effective prevention tools, HPV vaccination and

hrHPV testing, These tools should and need to be

deployed globally if we are to avert the unnecessary

burden of cervical cancer, especially in populations

living in low- and middle-income countries, where

effective Pap programs were never established.34

Pap testing had its day and should be “celebrated”

for all that it accomplished in cancer prevention. But

“based on the weight of the current evidence”, the

Pap should no longer be the standard of care for

cervical cancer prevention and its use should be

limited to deciding what kind of care hrHPV-positive

women need: increased surveillance, colposcopy, or

possibly treatment.

UNDERSTANDING THE LIMITATIONS OF CYTOLOGY FOR SCREENING

9

REPORT

The ATHENA (Addressing the Need for Advanced

HPV Diagnostics) study is the US registrational

study that has been used to obtain Federal Drug

Administration (FDA) approval of the Roche cobas

HPV Test for use in the US1 The ATHENA study was

begun in 2008 and con-

sisted of two distinct

phases. The Baseline

Phase was cross-

sectional and enrolled

46,887 eligible non-

pregnant women > 21

years undergoing rou-

tine cervical cancer screening at 61 clinical centers

in 23 states across the United States. At enrollment

all women had a ThinPrep liquid-based cytology

(LBC) sample and HPV testing with multiple high-risk

HPV DNA assays (AMPLICOR HPV Test, LINEAR

ARRAY High Risk HPV Genotyping Test, cobas

HPV Test –all manufactured by Roche Molecular

Systems, Pleasanton, CA). All women with either an

abnormal cytology result or who were HPV positive

were referred to colposcopy. In addition, a random

subsample of cytology and HPV negative women

were referred to colposcopy (n=886) to allow for

verifi cation bias adjustment.

Data obtained from the Baseline Phase of ATHENA

were used to obtain FDA-approval for use of cobas

HPV Test in women > 21 years with atypical squa-

mous cells of undetermined signifi cance (ASC-US)

and for cotesting using both cytology and HPV

testing in women > 30 years. The cobas HPV Test is

a real-time polymerase chain reaction (PCR) assay

that detects 12 pooled high-risk HPV genotypes

(31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66 and

68) and HPV 16 and HPV 18 individually and

includes a separate B-globin control. Data from the

Baseline Phase were

also used to obtain

FDA-approval for HPV

16/18 genotyping in

both ASC-US patients

and when cotesting.2-4

KEY FINDINGS OF THE ATHENA STUDY

Thomas C Wright, Jr. MD1 and Catherine M Behrens, MD, PhD2

1 Department of Pathology and Cell Biology Columbia University (Emeritus), New York, NY, USA.2 Roche Molecular Systems, Pleasanton, CA, USA.

"Data from the Baseline Phase were also used to obtain FDA-approval for

HPV 16/18 genotyping in both ASC-US patients and when cotesting."

Athena was the Greek virgin god-dess of reason, intelligent activity, ar ts and literature. She became the patron goddess o f A thens after winning a contest against Poseidon by offering the olive tree to the Athenians.

"Data obtained from the Baseline Phase of ATHENA were used to obtain

FDA-approval for use of cobas HPV Test in women >21 years with ASC-US and for cotesting using both cytology and HPV testing in women >30 years."

10

All women who underwent colposcopy during

the Baseline Phase of ATHENA and found not to

have cervical intraepithelial neoplasia of grade 2

(CIN 2+), were eligible for enrollment into a 3 year

Follow-up Phase. During the Follow-up Phase,

women had yearly gynecological exams and both

liquid based cytology (LBC) and HPV testing and

were referred to colposcopy if found to have

> ASC-US. At the 3 year visit, women were offered

an exit colposcopy. Data from both the Baseline

and the Follow-up Phases were subsequently used

to obtain FDA-approval in April of 2014 of the

cobas HPV Test with HPV 16/18 genotyping for

primary screening in women > 25 years.5

WHAT WE LEARNED FROM ATHENA REGARDING THE ROLE OF HPV GENOTYPING IN COTESTING.The prevalence of HPV in women with negative for

intraepithelial lesion or malignancy (NILM) results

decreased with increasing age, however most of

the decrease occurred between the 30-39 and the

40-49 age groups, Table 1. Only small decreases

were observed for HPV (14 pooled types), HPV 16,

HPV 18, and 12 “other” high-risk HPV genotypes

between the other age groups. Although the overall

prevalence of HPV in women > 30 yrs with NILM

was 6.7%, the majority of this was due to positiv-

ity for the 12 “other” high-risk HPV genotypes.

Positivity for HPV 16/18 was relatively uncommon,

1.5%. Verifi cation bias-adjusted estimates for rela-

tive risk of CIN 3+ in women with NILM cytology

by HPV genotypes status clearly demonstrates the

importance of a woman’s HPV 16/18 status. Women

with NILM cytology who were HPV negative had

an estimated absolute risk of CIN 3+ of only 0.3,

which is low enough to allow interval screening at 3

to 5 years. Women with NILM cytology who were

high-risk HPV positive (for any of the 14 high-risk

Age Group (yrs)

30 - 39 40 - 49 50 - 59 60 - 69 Overall

HPV (+)

prevalence 9.0 5.7 5.3 4.9 6.7

risk of CIN 3+ 5.7 3.5 1.1 2.7 4.1

HPV 16 (+) / 18 (+)

prevalence 2.3 1.1 1.0 0.9 1.5

risk of CIN 3+ 13.5 6.6 4.4 0.0 9.8

HPV 16 (+)

prevalence 1.6 0.7 0.6 0.7 1.0

risk of CIN 3+ 16.4 7.5 2.4 0.0 11.7

HPV 18 (+)

prevalence 0.7 0.4 0.4 0.2 0.5

risk of CIN 3+ 7.1 2.4 3.7 0.0 5.7

HPV 12 other

prevalence 6.7 4.6 4.4 4.0 5.2

risk of CIN 3+ 3.0 2.8 0.3 3.3 2.4

HPV (-)

risk of CIN 3+ 0.0 0.4 0.0 1.6 0.3

Total

risk of CIN 3+ 0.5

Table 1. Impact of age on prevalence of HPV and estimated absolute risk of CIN 3+ (in % over the 3-year follow up period) by HPV status in women with NILM cytology.*

* modifi ed from Wright TC et al. Am. J. Clin. Pathol. (2011)

"Data from both the Baseline and the Follow-up Phases were subsequently

used to obtain FDA-approval in April of 2014 of the cobas HPV Test with HPV

16/18 genotyping for primary screening in women >25 years."

HPV genotypes combined) have an absolute risk of

4.1 and those with one of the 12 "other" high-risk

HPV genotypes have a risk of 2.4. Since their risk of

CIN 3+ is < 5.0, it is generally accepted that these

women can be followed up at 12 months with repeat

co-testing and only need colposcopy if found to be

persistently HPV positive or have a repeat cytol-

ogy of equal or greater than low-grade squamous

intraepithelial lesions (LSIL) with a negative HPV

result. In contrast, women positive for either HPV 16

or HPV 18 have an estimated absolute risk of CIN 3+

of 9.8 and would clearly benefi t from a referral to

immediate colposcopy. Risk for all combinations

of genotypes was highest in women 30-39 years

of age and lowest in women > 50 years. Thus the

results from ATHENA clearly demonstrate the use-

fulness of being able to genotype for HPV 16 and

HPV 18 when cotesting.

11

KEY FINDINGS OF THE ATHENA STUDY

WHAT WE LEARNED FROM ATHENA REGARDING THE SAFETY OF HPV PRIMARY SCREENING.At Baseline, 10.5% of the 40,901 women > 25

years were HPV positive and 6.4% had an abnor-

mal cervical cytology. During the course of the

3-year study, a total of 319 CIN 3 lesions, 20

adenocarcinoma in-situ, and 8 invasive cervical

cancer cases were detected. Figure 1 shows the

3-year cumulative incidence rate (CIR) of CIN 3+

adjusted for sampling fraction and follow-up. A

negative cytology result at Baseline conferred a

lower level of protection against the subsequent

identification of CIN 3+ than does a negative

HPV test result. Of the CIN 3+ identified after

3 years 47.3% had a NILM cytology result at

Baseline whereas only 9.8% were HPV negative.

A negative cotest at Baseline offered minimal

additional protection compared to a negative

HPV test alone. HPV genotype status was also

highly predictive of CIN 3+ over the course of

the study. The highest 3-year CIR of CIN 3+ was

25.2% (95% CI, 21.7-28.7%) in HPV 16 positive

women and the lowest was 0.3% (95% CI, 0.1-

0.6%) in HPV negative women, Figure 2. Women

positive for HPV 18 had a 3-year CIR for CIN 3+

that was intermediate between HPV 16 positive

women and those positive for the 12 “other”

high-risk HPV genotypes. The high prevalence

of HPV positivity (10.5%) in women >25 years

means that some form of triage is required

to identify those women who should undergo

immediate colposcopy and those who can be

followed without immediate colposcopy. The HPV

primary screening algorithm that was approved

by the FDA for use in the U.S is a combination

of genotyping for HPV 16/18 and reflex cytology

for women positive for the 12 other high-risk HPV

genotypes, Figure 3. This is also the strategy that

the recent Society of Gynecologic Oncology (SGO)

/ American Society for Colposcopy and Cervical

Pathology (ASCCP) Interim Guidance adopted for

HPV primary screening.6

We compared the FDA approved strategy to

a strategy of either cytology alone (with reflex

HPV testing for women with ASC-US) for women

Figure 2. 3-year adjusted cumulative incidence of CIN 3+ in women >25 years stratified by baseline HPV test result.*

0 1 2 3

HPV 16 positive HPV 18 positive HPV 12 other positive HPV negative

Cum

ulat

ive

Inci

denc

e R

ate

(%)

- VB

A

Years of Follow-up

35

30

25

20

15

10

5

0

Figure 1. 3-year adjusted cumulative incidence of CIN 3+ in women >25 years stratified by baseline screening test result.*

Baseline Year 1 Year 2 Year 3

0.78

0.34

0.30*

PAP negative HPV negative PAP and HPV negative

CIR

CIN

3+

(%)

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0

"Of the CIN 3+ identified after 3 years 47.3% had a NILM cytology result at Baseline

whereas only 9.8% were HPV negative."

* modified from Wright TC et al. Gynecol. Oncol. (2015)

12

KEY FINDINGS OF THE ATHENA STUDY

strategy requiring only 10.8 colposcopy exams

per CIN 3+. The hybrid strategy and HPV primary

screening had similar effi ciencies, 12.8 and 12.9

colposcopy exams per CIN 3+, respectively.

These data clearly demonstrate that HPV primary

screening using the FDA-approved screening strat-

egy is as safe as cervical cytology over a 3-year

screening interval in women > 25 years and that

it is as effi cient as using cytology in women 25-29

years and cotesting in women > 30 years.

> 25 years or a hybrid strategy that uses cytol-

ogy in women 25-29 years and cotesting with

cytology and HPV testing in women > 30 years.

The HPV primary screening strategy detected the

most cases of CIN 3+ over the 3 years (n=294)

and had the highest adjusted sensitivity (76.1%)

for CIN 3+, Table 2. For comparison, cytology

detected only 179 cases of CIN 3+ and had an

adjusted sensitivity of 47.8% whereas the hybrid

strategy detected 240 cases of CIN 3+ and had

an adjusted sensitivity of 61.7%. Specifi city for

CIN 3+ was highest for the cytology strategy

(97.1%) and lowest for the HPV primary screening

strategy (93.5%). The effi ciency of a given strategy

can be estimated by determining the number of

colposcopies that are required to detect a single

case of CIN 3+. Cytology was the most effi cient

Strategy Sensitivity# Specifi city# Nº. CIN 3+ detected§

Nº. Colposcopy Exams§

Nº. Colpo per case of CIN 3+§

Cytology 47.8 97.1 179 1,934 10.8

Hybrid strategy¥ 61.7 94.6 240 3,097 12.9

HPV primary 76.1 93.5 294 3,769 12.8

Table 2. Per formance of different screening strate-gies for detection of CIN 3+ in women >25 years.*

# verifi cation-bias adjusted. § crude numbers of detected cases. ¥ cytology in women 25-29 and co-testing in women 30+.

References: 1. Wright TC et al. The ATHENA human papillomavirus study: design, methods, and baseline results. Am J Obstet Gynecol 2011. 2. Wright TC et al. Evaluation of HPV-16 and HPV-18 Genotyping for the Triage of Women With High-Risk HPV+ Cytology-Negative Results. Am J Clin Pathol 2011;136:578-86. 3. Stoler MH et al. High-risk human papillomavirus testing in women with ASC-US cytology: results from the ATHENA HPV study. Am J Clin Pathol 2011;135:468-75. 4. Stoler MH et al. The interplay of age stratifi cation and HPV testing on the predictive value of ASC-US cytology. Results from the ATHENA HPV study. Am J Clin Pathol 2012;137:295-303. 5. Wright TC et al. Primary cervical cancer screening with human papillomavirus: End of study results from the ATHENA study using HPV as the fi rst-line screening test. Gynecologic oncology 2015. 6. Huh WK et al. Use of primary high-risk human papillomavirus testing for cervical cancer screening: interim clinical guidance. Gynecologic oncology 2015;136:178-82.

31 33 35

39 45 51

52 56 58

59 66 68

16 18

hrHPV

Follow up in12 months

COLPOSCOPY

HPV-

12 other hrHPV+

HPV16/18+

Routine screening

Cytology

COLPOSCOPY

NILM

>ASC-US

Figure 3. FDA-approved HPV Primary Screening Strategy Incorporating Genotyping for HPV 16/18 and "refl ex" cytology for 12 other high-risk HPV genotypes.*

"The HPV primary screening algorithm that was approved by the FDA for use in the U.S is a combination of genotyping for HPV 16/18

and refl ex cytology for women positive for the 12 other high-risk HPV genotypes."

* modifi ed from Wright TC et al. Gynecol. Oncol. (2015)

Dr. Thomas C Wright is a consultant and speaker for Roche, BD Diagnostics and Cepheid. Dr. Catherine M Behrens was previously employed by Roche Molecular Systems and now serves as a consultant for Roche and other women’s health care entities.

13

REPORT

TRIAGING HPV-POSITIVE WOMEN

Nicolas Wentzensen M.D., Ph.D., M.S.Division of Cancer Epidemiology and Genetics. National Cancer Institute. Bethesda. USA

HPV testing for cervical cancer screening pro-

vides high sensitivity for detection of cervical

precancers, allowing extended screening intervals

compared to cytology in women who test nega-

tive.1 However, most HPV infections are cleared

and do not progress to cervical cancer. HPV

testing can double the number of screen-positive

women compared

to cytology and it is

neither feasible, nor

effective to send

all HPV-posit ive

women to colpos-

copy. Therefore,

additional tests are

needed to identify

women at high-

est risk of cervical

cancer that require

immediate colposcopy and biopsy. An ideal test

should identify the small group of women at high-

est risk for treatable cervical precancers needing

colposcopy referral, while reassuring the majority

of women that their risk is low. In other words, an

optimal strategy should identify as much disease

as possible, while leading to

as little colposcopy referral

as possible.2 Several strate-

gies for triage of HPV-positive

women have been evaluated,

and some have already been

introduced in screening prac-

tice. In addition to the different

characteristics of currently

available triage tests, their

use may vary substantially

between different regions.

Importantly, the perception

of risk may differ in differ-

ent situations and societies.

Therefore, risk thresholds are not absolute, but

they are tied to a certain societal perception of

risk and are often refl ective of established clinical

practice. For example, different thresholds for col-

poscopy referral, and different intervals for repeat

testing may be used for the same tests. Societal

differences aside, recent guidelines efforts

have attempted to

standardize clinical

management, and

have emphasized

the importance of

absolute risk for

public health and

clinical decisions.3,4

B a s i ng c l i n i c a l

management on

absolute risk levels

rather than indi-

vidual test results allows formulating more

general guidelines that can be easily updated

when new tests become available by conducting

risk equivalence studies. Figure 1 shows the

principle of risk-based management in cervical

cancer screening. The risk of cervical cancer and

(d) Colposcopy referral threshold

(a). Population risk

HPV screening Triage marker

Risk of precancer

(b) PPV: Risk in HPV-positives

(e) PPV: Risk in HPV-positives and triage positives

100%

40%

10%

2%

~0%

(f) cNPV: Risk in HPV-positives and triage negatives

(c) cNPV: Risk in HPV-negatives

Figure 1. Risk based management.

An ideal test should identify the small group of women at highest risk for

treatable cervical precancers needing colposcopy referral, while reassuring the majority of women that their risk is low. In other words, an optimal strategy should identify as much disease as possible, while leading to as little colposcopy

referral as possible.

14

CIN3 in the general population is low (a) HPV

primary screening changes the prior, baseline risk

estimate to a higher risk in test-positive women

(b) and to an even lower risk than baseline in

those who test negative (c). The absolute risk of

disease in test-positives is equal to the positive

predictive value, while the absolute risk of disease

in test-negatives is equal to the complement of

the negative predictive value (cNPV or 1-NPV).5

However, in the example given from US man-

agement guidelines, the risk in HPV-positives

does not cross the threshold for referral to

colposcopy (d). A useful triage marker applied to

the HPV-positive population should change the

risk estimate to a risk in triage-positive women

above the colposcopy referral threshold (e) and

to a lower risk in those who test-negative for the

triage marker (f).

The operational characteristics of a triage test

are important for designing a screening program.

Some triage tests are based on detection of

nucleic acids from lysed specimens (e.g. HPV

genotyping, methylation markers) while others

require intact cells (e.g. cytology, p16/Ki-67

dual stain). Depending on the primary test and

the organization of the screening program, some

triage tests may require an additional specimen

type for the triage test or even an additional visit

for collection following a positive screening test

result. It is obviously preferable to run the triage

test from the specimen collected for primary HPV

testing (refl ex triage).

Proposed markers for triage of HPV-positive

women are summarized in table 1. Cytology has

been evaluated as a triage test for primary HPV

testing in several studies and is the triage strategy

in the proposed organized Dutch cervical cancer

screening program. Women who are HPV-positive

in primary screening can have refl ex cytology test-

ing and are referred to immediate colposcopy when

cytology is positive (the thresholds for cytology

Assay Examples Validation Characteristics

HPV genotypingCobas 4800 (Roche);Cervista (Hologic);Onclarity (BD)

Large screening and triage studies

Type-restriction

Oncogene mRNAAptima HPV (Genprobe);Proofer (NorChip);HPV Oncotect (InCell Dx)

Some large observational studies

Type-dependent

E6 detection OncoE6 (Arbor Vita) Limited data available Type-dependent

p16/Ki-67 cytology CINtec PLUS(Roche Ventana)

Large observational studiesNot type-dependent; slide-based

MCM2/ Top2A cytology ProExC (BD) Limited data availableNot type-dependent; slide-based

3q Oncofi sh (Ikonisys);FHACT (Cancer Genetics)

Limited data availableNot type-dependent; slide-based

Host and viral methylationHost markers: CADM1, MAL, mir-124, EPB41L3, LMX1Viral markers: L2, L1

Some large observational studies and trials

Host: not type dependentViral: type-dependent

Table 1. Triage marker candidates

"The risk of cervical precancer and cancer differs considerably by HPV

genotype, with HPV16 being by far the most carcinogenic type, followed by

HPV18. Because of this remarkable risk stratifi cation, HPV genotyping for HPV16 and HPV18 has been evaluated for triage

of HPV-positive women."

George Papanicolau described cervical cytology as the fi rst cancer screeningmethod and launched the concept of cancer prevention by early treatment of cancer precursor lesions

15

TRIAGING HPV-POSITIVE WOMEN

positivity differ between studies and healthcare

settings). HPV-positive, cytology-negative women

undergo retesting after at least 6-12 months, to

allow a proportion of the transient HPV infections

to clear. While there is concern that the limited

sensitivity of cytology cancels out some of the

advantages of primary HPV screening, there is

evidence suggesting that the sensitivity of cervical

cytology is increased when it is conducted with

knowledge of HPV status, or when it is restricted

to HPV-positive women only.6

The risk of cervical precancer and cancer differs

considerably by HPV genotype, with HPV16 being

by far the most carcinogenic type, followed by

HPV18.7 Because of this remarkable risk strati-

fi cation, HPV genotyping for HPV16 and HPV18

has been evaluated for triage of HPV-positive

women. Several HPV assays now provide separate

results for HPV16 and HPV18, or for more HPV

genotypes, either as a second test or integrating

this triage approach into the primary screen-

ing test. Women who test positive for HPV16

or HPV18 have high enough risk for referral to

colposcopy, while women testing positive for the

other carcinogenic types typically have a repeat

test after 12 months. There is an ongoing debate

about whether other carcinogenic types should

be included for triage to immediate colposcopy.

While adding more types increases the sensitivity

of the approach somewhat, the absolute risk is

reduced since all other types are associated with

a much lower risk compared to HPV16.

In the US, primary HPV testing with the cobas

assay was approved with a combination of HPV

genotyping and cytology for triage of HPV-

positive women. According to the approved

algorithm, all women testing positive for HPV16

or HPV18 should be referred to immediate col-

poscopy. Women testing positive for the other

12 high-risk types have cytology testing and

should be referred to colposcopy when cytol-

ogy shows ASC-US or greater. Women positive

for HPV types other than HPV16/18 and with

normal cytology undergo repeat testing after

12 months.8

Several other possible markers for triage of HPV-

positive women have been evaluated. The p16/

Ki-67 dual stain is a cytology-based assay that

indicates overexpression of two cellular proteins

altered in cervical precancers.9 The detection of

cells stained for both p16 and Ki-67 is considered

a positive result. A previous version of the assay

was based on p16 staining alone followed by

morphological evaluation of p16-stained cells. In a

study using the p16 assay nested in the Italian cer-

vical cancer screening trial, p16 showed good risk

stratifi cation for triage of HPV-positive women.10

p16-positive women had high enough risk for

referral to colposcopy, while the repeat testing

interval could be extended to at least two years

among p16-negative women according to these

data. Similar results were observed in smaller

studies using the dual stain, and several large

studies are now underway to confi rm these

results.11 It has been demonstrated that the p16/

Ki-67 dual stain assay can be implemented with

limited training while achieving high reproducibil-

ity and accuracy and that automated evaluation

of dual stained slides is feasible.12

Recently, there has been a lot of focus on devel-

oping DNA methylation markers for triage of

HPV-positive women. Marker panels including the

cellular genes CADM1, MAL, and miR-124-2 in

various combinations have been evaluated in Dutch

Figure 1. Dual staining with p16 and Ki67 in cytology is under evaluation in triaging HPV positive women for colposcopy. In the image one single cell showing double staining (brown and red) is considered suffi cient to stratify for high risk of underlying neoplasia.

16

TRIAGING HPV-POSITIVE WOMEN

References: 1. Schiffman M et al. J Natl Cancer Inst 2011;103(5):368-83. 2. Wentzensen N. Lancet Oncol 2013;14(2):107-9. 3. Massad LS et al. J Low Genit Tract Dis 2013;17(5 Suppl 1):S1-S27. 4. Saslow D et al. CA Cancer J Clin 2012;62(3):147-72. 5. Wentzensen N, Wacholder S. Cancer Discov 2013;3(2):148-57. 6. Moriarty AT et al. Arch Pathol Lab Med 2014;138(9):1182-5. 7. Schiffman M et al. Cancer Epidemiol Biomarkers Prev 2011;20(7):1398-409. 8. Huh WK et al. Gynecol Oncol 2015. 9. Wentzensen N et al. Clin Cancer Res 2012;18(15):4154-62. 10. Carozzi F et al. Lancet Oncol 2013;14(2):168-76. 11. Petry KU et al. Gynecol Oncol 2011;121(3):505-9. 12. Wentzensen N et al. Cancer Cytopathol 2014. 13. Verhoef VM et al. Lancet Oncol 2014;15(3):315-22. 14. Mirabello L et al. J Natl Cancer Inst 2012;104(7):556-65. 15. Wentzensen N et al. J Natl Cancer Inst 2012;104(22):1738-49. 16. Schiffman M, Wentzensen N. Gynecol Oncol 2015;136(2):175-7.

cervical cancer screening studies and have shown

comparable performance to cytology-based

triage.13 This approach is particularly appealing for

primary screening based on self-sampling, since

the test can be run from the screening sample

and does not require an additional collection for

a cytology specimen. Beyond host methylation,

it has also been demonstrated that methylation

of the HPV genome, especially in the L2 and

L1 regions, is associated with prevalent and

future precancer and cancer. Initial studies have

shown that measuring HPV methylation could

provide suffi cient risk stratifi cation for triage of

HPV-positive women and development of HPV

methylation assays is currently underway.(14,15)

Other approaches that have been proposed to

triage HPV-positive women include measuring

HPV oncogene mRNA or protein expression,

detection of recurrent chromosomal changes at

3q and other sites, and staining for other cellular

proteins that are associated with HPV-related

transformation, such as mcm2 and Top2a.

The number of triage assay candidates is stead-

ily increasing and choosing the optimal triage

strategy is becoming increasingly a challenge.(16)

The evaluation of screening and triage strategies

needs to be conducted on a program level, since

different assays have very different implications

on management procedures.

For example, a certain triage assay could have a

low immediate colposcopy referral rate, but still

send the majority of women to colposcopy after

the repeat test, e.g., a year later. It is already

now apparent that some triage strategies have

similar performance, and other factors may be

important to decide which test to use, such

as the sampling buffer, assay throughput, and

ultimately cost. A theoretical optimal triage

strategy may combine assays from different

manufacturers that are not compatible or cannot

be combined because of commercial concerns.

Furthermore, different population characteris-

tics, screening and management procedures can

infl uence performance estimates for individual

assays, making comparisons across studies

diffi cult for the small differences expected for

some of the strategies. It is not possible to

conduct randomized controlled trials for every

strategy comparison. Therefore, it is necessary

to evaluate triage options in large head-to-head

observational comparisons that allow evaluating

disease detection and management prospec-

tively over multiple years, by multiple testing of

aliquots of the same clinical specimens. Defi ning

an acceptable, management set of triage strat-

egies for women that screen HPV-positive is

one of the major ongoing tasks we face in the

introduction of HPV-based screening.

"It has been demonstrated that the p16/Ki-67 dual stain assay can be implemented with limited training while achieving high reproducibility and accuracy and that

automated evaluation of dual stained slides is feasible."

Dr. Nicolas Wentzensen is employed at the National Cancer Institute (NCI). The NCI has received commercial HPV tests for research at a reduced or no cost from BD, Cepheid, Hologic, and Roche.

17

REPORT

THE ROLE OF EPIDEMIOLOGY IN DISCERNING OPTIMAL CERVICAL CANCER SCREENING STRATEGIESJulia C. Gage, PhD, MPHClinical Genetics Branch. Division of Cancer Epidemiology and Genetics. National Cancer Institute. Bethesda. USA

Numerous large randomized studies1,2 and

screening trials3,4 have established that hig-risk

HPV (hrHPV) testing can improve cervical cancer

screening. Currently, the optimal screening

strategy is uncertain with ongoing debates over

primary hrHPV screening versus cotesting (hrHPV

testing concurrently with Pap testing), the man-

agement of women screening positive, and the

correct screening interval for women screening

negative.

The most scientifically rigorous approach to

resolve these debates would be randomized

screening trials with head-to-head comparisons of

candidate screening strategies. Such trials require

many thousands of women followed for years to

discern meaningful differences in risk of cervical

cancer, which occurs very rarely in undeniably

efficacious screening strategies such as primary

hrHPV and cotesting at 3 or 5-year intervals.

With so many options for screening and triage

tests, and screening intervals, it is not possible

to conduct randomized trials for each comparison

of interest.

In the absence of randomized clinical trials, we

turn to observational data produced as a matter

of course from very large screening programs

and registries, to glean the risks after one or

multiple screening rounds using different testing

strategies.5-7 In clinical programs, each woman is

managed using a strategy; the statistical challenge

is to estimate what would have happened had she

been managed using

alternative, “counter-

factual” strategies.

Key outcomes consid-

ered in the analysis

of observational data

u s e d t o d e c i d e

whether to extend a

screening interval or

use a different screening test, are the extent and

duration of protection against cervical cancer

afforded by the negative screen. Cervical cancer

is the optimal epidemiologic outcome because

screening programs are intended to prevent

cervical cancer mainly by identifying women at

greatest risk of cervical cancer, namely those

with precancer (best equated with histologically-

diagnosed cervical intraepithelial neoplasia of

grade 3 [CIN3]). Few cohorts are large enough

with sufficient follow-up to estimate precisely

the risk of prevalent and incident cancer among

women screening negative. Thus, we are forced

sometimes to estimate cancer risk by the use

of surrogate endpoints, most commonly the

detection of CIN3 (or rare cancers, CIN3+) at

the first screen and risk of CIN3+ in subsequent

screens. The use of CIN3 for screening studies

is conceptually difficult, because finding CIN3 in

time to avert cancer is a success of screening,

and the concern regarding overdiagnosis (finding

CIN3 cases that would not have become invasive

cancer) is always present.

Fortunately, data are now available that directly

provide estimates of cancer risk after a negative

screen from Kaiser Permanente Northern California

(KPNC). KPNC is a large integrated health delivery

system; since 2003 >1 million women age 30+

have been screened with cotesting, at approximate

3-year intervals. Using logistic-Weibull modeling,

we have estimated the risks of cancer up to 5

years after a nega-

tive Pap, hrHPV and

cotest (Figure 1). How

to interpret and com-

pare these risks is not

straightforward, espe-

cially in the context

of deciding screening

guidelines.

"The risks following an HPV-negative and cotest negative test are

definitely lower than those following a negative Pap test. Adding a Pap test to hrHPV testing (cotesting)

confers only a very slight marginal gain in reassurance against cancer."

18

THE ROLE OF EPIDEMIOLOGY IN DISCERNING OPTIMAL CERVICAL CANCER SCREENING STRATEGIES

We use counterfactual reasoning to estimate risks

for screening contexts that differ from actual

clinical practice in an attempt to understand

risks for screening strategies that did not actually

occur (e.g., cotesting every 5 years or primary

hrHPV testing every 3 years). This counterfactual

reasoning is evident in several assumptions. We

assume the estimated risk among women with a

negative Pap in a cotesting screening program

is similar to risk among women with a negative

Pap in a primary Pap program. Because the

vast majority of women at KPNC had cotesting,

women testing Pap-negative are managed based

upon their hrHPV result. Thus, in our recent

screening analyses from KPNC, 3.7% of women

testing Pap-negative were concurrently hrHPV-

positive and had a repeat screen in 1 year with

colposcopy referral if their repeat screen was

positive.9 Cancer risk estimates for Pap-negative

results are thereby likely overestimated slightly

because of the higher cancer risk among women

screening Pap-negative/hrHPV-positive versus

Pap-negative.5

This challenge of counterfactual reasoning is also

present when considering whether to extend

screening intervals beyond standard practice,

e.g., 5 instead of 3 years. Because women at

KPNC typically return for screening 3 years after

a negative cotest, some precancers destined

to progress to cancer between 3 and 5 years

are detected at the 3-year return and treated,

thereby preventing cancer. The 5-year cancer

risks are therefore slightly underestimated.

Conversely, the 5-year risk of precancer is slightly

overestimated because the precancer destined

to regress between 3 and 5 years is detected

and treated at the 3-year screen. This dilemma

also exists for estimating risks that occur before

returning for screening or during time points

between screening visits, e.g., 1-year risks after

a negative screen at KPNC. We are left to make

careful assumptions regarding the natural history

of cervical precancer and cancer between screen-

ing visits to permit risk modeling.

In spite of these challenges, the analytic approach

of risk comparison from large clinical datasets has

provided valuable evidence for decision-making

with statistical power to quantify and distinguish

risks that are extremely low and close to one

another. Importantly, the same general trends

observed with a surrogate endpoint (CIN3+) in

other cohorts and in KPNC have been shown

to hold true when we analyze invasive cancer

outcomes in KPNC. Namely, the risks following

an HPV-negative and cotest negative test are

defi nitely lower than those following a negative

Pap test.4,10,11 Adding a Pap test to hrHPV testing

Figure 1. Cumulative risks of cancer among women aged 30-64 at Kaiser Permanente Northern California by enrollment Pap and HPV test result, 2003-2012.

Adapted from Gage, JC et al. JNCI 2014 8

Cum

ulat

ive

Ris

k of

Cer

vica

l Can

cer

(%)

1 2 3 4 5

0.05

0.04

0.03

0.02

0.01

0.00

Years since baseline

PAP-negative HPV-negative Cotest-negative

0.031

0.020

0.009

0.005

0.011

0.007

0.017

0.014

0.003

19

(cotesting) confers only a very slight marginal gain

in reassurance against cancer.

The choice of cotesting rather than primary HPV

testing necessarily implies that that the value

of a very small reduction in cervical cancer is

very high. A related and deeper question (which

is societal, not statistical) is how much safety

should screening provide. Assuming the risk

models accurately measure risks that would be

observed in their respective screening strategies,

what risk threshold is appropriate for women to

follow routine screening? In the US, the standard

of care for many years was the annual Pap smear

and therefore, it is argued that the threshold

risk for evaluating other strategies should be

the risk of cervical cancer within 1 year after a

negative Pap.12

Public health policymakers in other settings have

defi ned less stringent acceptable cancer risks

for population-based screening programs.13,14

Fortunately, the majority of cervical cancer can

be prevented through existing well-run screening

programs. The debates now focus on relative

minimal reductions in cervical cancer and how

to fi ne-tune cervical cancer screening.

Importantly, the reassurance against cancer

provided by a negative screening test is just the

tip of the iceberg for evaluating a screening inter-

vention. Risks are cumulative and extend over a

lifetime of screening and they must be balanced

with harms and fi nancial costs. Mathematical

decision modeling is currently the only approach

that can incorporate the many factors infl uencing

References: 1. Sankaranarayanan R et al. N Engl J Med. 2009; 360(14): 1385-94. 2. Ronco G et al. Lancet. 2014; 383(9916): 524-32. 3. Wright TC et al. Gynecol Oncol. 2015; 136(2): 189-97. 4. Dillner J et al. BMJ. 2008; 337: a1754. 5. Katki HA et al. Lancet Oncol. 2011; 12(7): 663-72. 6. Katki HA et al. J Low Genit Tract Dis. 2013; 17(5 Suppl 1): S56-63. 7. Wheeler CM et al. Int J Cancer. 2014; 135(3): 624-34. 8. Gage JC et al. JNCI 2014; 136(7):1665-71. 9. Gage JC et al. Cancer cytopathology. 2014; 122(11): 842-50. 10. Wright TC et al. Gynecol Oncol 2015;136:189-97. 11. Ronco G et al. Lancet Oncol. 2010; 11(3): 249-57. 12. Kinney W et al. Obstet Gynecol. 2015. 13. National Institute for Public Health and the Environment (RIVM). Feasibility study for improvements to the population screening for cervical cancer [in Dutch]. 2013. 14. Vink MA et al. Int J Cancer. 2014. 15. Kulasingam SL et al. J Low Genit Tract Dis. 2013; 17(2): 193-202. 16. Burger EA et al. Br J Cancer. 2012; 106(9): 1571-8.

cervical cancer prevention in the context of

repeated screening, changing screening intervals,

treatment and vaccination. Such modeling pro-

vides projections of long-term population-based

outcomes and cost-effectiveness. These analyses

can be powerful tools for comparing screening

strategies over time.15,16 Yet, they have their own

inherent challenges and collaborative efforts to

compare and standardize models are important

(CISNET) to provide robustness and foster trust

in their conclusions.

As epidemiologists, statisticians, or decision ana-

lysts, we are called upon because of our expertise

in risk estimation. However, no discipline including

clinical medicine can claim superiority in the act

of balancing the harms and benefi ts of alternative

cervical cancer screening programs. By nature,

the debate is complicated for all but the simplest

of questions (whether or not to screen). Many

of the opinions are value-laden and philosophi-

cal, not necessarily scientifi c. In addition to the

debates over screening test and interval, we can

add the questions of age of fi rst screen, age of last

screen and screening among vaccinated women.

It is useful to realize that formulating screening

guidelines contains one part in which we are

expert (risk estimation), and one part in which we

provide just one voice (values regarding safety).

With this realization, it becomes clear that varia-

tion in screening practices is inevitable and that

international harmonization is highly unlikely. HPV

is the universal cause of cervical cancer; what to

do with that fact is not universal.

Screening is about identifying in a crowd the ones at high risk of cancer. Subsequent diagnostic

techniques will guide treatment decisions

Dr. Julia C. Cage received HPV testing of NCI specimens, not materials, for research at no cost from Roche and BD.

20

THE INTERIM GUIDANCE TO HPV SCREENING

Warner K. Huh, MD, FACOG, FACSProfessor and Division Director. Division of Gynecologic Oncology. University of Alabama at Birmingham. Birmingham, Alabama USA.

On April 24th 2014, the United States Food and

Drug Administration (FDA) approved high risk

HPV (hrHPV) testing for primary cervical cancer

screening in the United States (US). This decision

also reflected unanimous support (13-0) from the

March 2014 FDA Medical Devices Advisory

Committee Microbiology Panel Meeting,

which included numerous US experts in the area

of cervical cancer screening and prevention. This

approval was based and supported from data

derived from the ATHENA (Addressing the Need

for Advanced HPV Diagnostics) trial. ATHENA, the

largest cervical cancer screening study conducted

in the US, was a registration study sponsored

by Roche Molecular Systems that utilized the

cobas® 4800 system. Data from ATHENA was

previously used for approval of hrHPV testing

for ASC-US cytology and concurrent cytology

and hrHPV screening (i.e., cotesting) in women

30 years and older. These two uses are widely

recommended by numerous stakeholder socie-

ties and organizations, as well as the United

States Preventive Services Task Force (USPSTF).

Furthermore, triage through identification of

specific high-risk types of HPV, specifically types

16 and 18, is also an FDA approved use of hrHPV

testing in selected settings. The FDA approved

a specific primary HPV screening algorithm that

utilized genotyping as well as cytology as triage

tests in this new setting.

In 2011, the American Cancer Society,

American Society for Colposcopy and

Cervical Pathology, and the American Society

for Clinical Pathology updated screening guide-

lines for the early detection of cervical cancer

and its precursors. These guidelines stated that

there was insufficient evidence to use HPV test-

ing alone as a screening mechanism. Specific

reasons included lack of information regarding the

specificity, potential harms including increased

rates of colposcopy and treatment, appropriate

screening intervals, and cost-effectiveness. Since

2011, several additional randomized trials have

been published in addition to ATHENA that have

further informed us about the utility and benefit

of primary HPV screening.

The public announcement of an FDA application

by Roche for a primary HPV screening claim trig-

gered the creation of an interim guidance panel

to review recent evidence and address specific

questions and concerns regarding using a hrHPV

test for primary screening, including ATHENA

and data relevant to the primary HPV screening

labeling. This panel was co-sponsored by the

Society of Gynecologic Oncology and the

American Society of Colposcopy and Cervical

Pathology. The primary objective of the panel

was to provide clinicians with a balanced overview

of primary HPV screening including its benefits

and potential harms. This process included an in

depth literature review as well as a scientific sum-

mary presentation provided by Roche Molecular

Systems of ATHENA including data and findings

related to the primary HPV screening compo-

nents of this trial. Panel members were allowed

to submit questions both before and after the

discussion.

Panel members were asked to address two primary questions:

1 Is HPV testing for primary screening as safe and effective as cytology-based screening?

2 Can primary HPV screening be consid-ered as an alternative to current US cervical cancer screening methods?

21

REPORT

There was considerable debate and discussion about

initiating primary HPV screening at 25 years of age.

Despite almost a one third increased detection of

CIN3+ in women 25-29 years of age and fi ndings

indicating that >50% of CIN3+ cases had preceding

normal cytology, this screening algorithm would double

the number of colposcopies performed in this age

bracket. Although it’s unclear whether detection of

CIN3+ in women 25-29 years of age would translate

into a reduction in invasive cervical cancer, the panel

did feel that this increased detection was clinically

meaningful despite the increased rate of colposcopy.

There was also considerable discussion comparing

primary HPV to cotesting. Based on a recent paper by

Gage et al in the Journal of the National Cancer Institute,

which analyzed data from over 1 million women

screened at Kaiser Permanente Northern California, it

was evident that the reassurance of a negative cotest

results was driven by the negative HPV test component

and based on a 3 year screening interval, primary

testing was as effective as 5 year cotesting.

There continue to be multiple areas of future research

and other considerations in the area of primary HPV

THE RECOMMENDATIONS OF THIS PANEL IS AS FOLLOWS:

screening. Some of these include the concern of false

negative results, specimen adequacy, appropriate

internal controls (since cytology might be viewed as

a surrogate for this), and comparative effectiveness

studies that address topics including cost and impact

on lifetime screening. The panel also highlighted this

algorithm is restricted to one assay at present and

assumptions of comparability should not be made,

and most importantly, expressed considerable con-

cern about the confusion that a third recommendation

might create for clinicians and the critical need for

adequate provider education.

In the end, the panel felt that primary HPV testing

was a highly important advance in cervical cancer

screening (perhaps, one of the most important) based

on the overwhelming supporting scientifi c data from

numerous large clinical trials including ATHENA.

But, these advances are meaningless if women are not screened and as such, it continues to be critically important for us to identify women who are unscreened or underscreened.

The panel also made the following additional recommendations:- Based on limited data, triage of hrHPV-positive

women using a combination of genotyping for HPV 16 and 18 and refl ex cytology for women positive for the 12 other hrHPV genotypes appears to be a reasonable approach to managing hrHPV-positive women. (Figure 1)

- Re-screening after a negative primary HPV screen should occur no sooner than every 3 years.

- Primary HPV screening should not be initiated prior to 25 years of age.

1 A negative HPV test provides greater reas-surance of low cervical intraepithelial neoplasia of grade 3 or higher (CIN3+) risk than a negative cytology result.

2 Because of equivalent or superior effec-tiveness, primary HPV screening can be considered as an alternative to current US cytology-based cervical cancer screening methods. Cytology alone and cotesting remain the screening options specifi cally recommended in major guidelines.

References: 1. Huh WK et al. Gynecol Oncol. 2015 Feb;136(2): p.178-82. 2. Wright TC et al. Gynecol Oncol. 2015 Feb;136(2):189-97. 3. Saslow, D et al. CA Cancer J Clin, 2012. 62(3): p. 147-72. 4. Gage J et al. J Natl Cancer Inst, 2014. 106(8): p.1-4. 5. 2014 Meeting Materials of the Food and Drug Administration (FDA) Microbiology Devices Panel, March 12, 2014. http://www.fda.gov/AdvisoryCommittees/CommitteesMeetingMaterials/MedicalDevices/MedicalDevicesAdvisoryCommittee/MicrobiologyDevicesPan-el/ucm388531.htm (Accessed July 2014).

Follow up in12 months

COLPOSCOPY

HPV-

12 other hrHPV+

HPV16/18+

Routine screening

Cytology

COLPOSCOPY

NILM

>ASC-US

31 33 35

39 45 51

52 56 58

59 66 68

16 18

hrHPV

cobas®

HPV Test

Figure 1. Candidate Screening Algorithm. HPV with 16/18 Genotyping and Refl ex Cytology.

Dr. Warner K. Huh has been paid as a consultant for Merck and THEVAX. Dr. Huh also serves on the scientifi c advisory board for IncellDx but does not personally receive any fees (fees are paid to his institution).

22

HPV TESTING IN THE UNITED STATES: PERSPECTIVES FROM SYSTEMS, PROVIDERS AND WOMENVicki Benard and Mona SaraiyaDivision of Cancer Prevention and Control. Centers for Disease Control and Prevention. Atlanta Georgia, USA.

Currently in the United States, two cervical cancer

screening modalities are endorsed by all three

major guideline organizations (American Cancer

Society, American College of Obstetricians and

Gynecologists, and the US Preventive Services

Task Force).1-3

The recommendations include:

1) screening with a Pap test every 3 years for

women aged 21-65 or

2) screening with a Pap test combined with a

test for high-risk types of human papilloma-

virus (hrHPV) every 5 years for women aged

30-65 (also known as co-testing).

In 2014, the Federal Drug Administration (FDA)

approved high-risk HPV (hrHPV) testing for pri-

mary cervical cancer screening in women aged

25 and older. Shortly thereafter, the Society for

Gynecologic Oncology and the American Society

for Colposcopy and Cervical Pathology jointly pub-

lished guidance recommending at least a 3-year

interval after a negative test, with a proprietary

algorithm for management of abnormal results.4

At this time, the three national screening organiza-

tions with most influence on clinical practice and

reimbursement have not updated their guidelines

to include primary HPV testing as an option for

screening.

Until the first half of the 20th century, the United

States had a high burden of cervical cancer that

was similar to the current burden seen in many low-

and middle-income countries. Doctors began to

use the Papanicolaou (Pap) test in the 1950s, when

annual testing was heavily promoted.1 Although

clinical trials to assess its potential effectiveness

were never performed before implementation,

Pap testing in both opportunistic and organized

systems has accompanied large decreases in cervi-

cal cancer mortality and incidence.5,6

The United States does not a have a universal,

nationally organized cervical cancer screening

program with a mechanism that can system-

atically collect information on cervical cancer

screening or generate reminders or recalls.

Therefore, we must rely on special studies that

provide details about provider and patient

practices. Through the years, expert groups have

made changes to the US guidelines, often caus-

ing confusion among health care providers about

screening methods, intervals, and target age

groups. Guidelines have shifted over a decade

to include longer intervals.7 However, provid-

ers have been very slow to adopt these longer

intervals, and surveys and anecdotal reports

show that some/many providers are conduct-

ing co-testing annually.8 Provider surveys have

shown reluctance to extend the screening interval

beyond annual screening until very recently.8-11

Clinicians in managed care organizations have

been the most successful at increasing intervals.12

With all national organizations reporting consistent

recommendations since 2012, the linking of reim-

bursement for clinical preventive services to one

particular guideline (the US Preventive Services

Task Force), and over a decade of co-testing use,

more providers have reported moving towards

longer intervals.13 However, not many providers are

willing to move to the new recommended interval

of 5 years for co-testing, which may be expected

given the recency of these guidelines.13

Barriers reported by providers include lack of

"However, not many providers are willing to move to the new recommended interval of 5 years for co-testing, which may be expected given the recency of these guidelines.

Barriers reported by providers include lack of knowledge of guidelines, patient demand or expectations of annual Pap tests, fear that patients will not come in for other preventive

services, and concern about missing early cancers."

23

REPORT

knowledge of guidelines, patient demand or

expectations of annual Pap tests, fear that patients

will not come in for other preventive services, and

concern about missing early cancers.9,10,13

National surveys of women have shown a decrease

in overall self-reported Pap tests from 2008 to

201014 and increases in Pap tests with longer

intervals.15 When women were surveyed about

their willingness to extend the screening interval

to 3 years if their doctor recommended it, almost

70% agreed, but only 25% said they would extend

screening to 5 years.16 When patients were surveyed

about adherence to guidelines, they cited barriers

such as lack of knowledge about cervical cancer

screening, a desire for more frequent care, and a

higher degree of perceived risk of cervical cancer.16,17

Most of the data collected come from self-reported

surveys of providers and women; however, some of

these practices are validated by other studies. New

Mexico performs the only statewide systematic

collection of cervical cancer screening records in

the United States.

This registry reported that screening intervals

lengthened from annual screening to less frequent

screening (but still not to the desired interval) from

2008 to 2011. Screening use decreased for all ages.18

In 2011, over 12,000 women developed and over

4,000 women died of this highly preventable dis-

ease.19 While 80% of women in the United States are

screened according to guidelines, in 2012, approxi-

mately 8,000,000 women had not been screened for

cervical cancer in the previous 5 years.19 The lack of

an organized monitoring system for cervical cancer

screening leads to unnecessary screening for some

women and lack of screening for others. Prioritizing

women who are rarely or never screened is essential

in reducing the burden of cervical cancer because

more than half of the cases were in women who had

not been adequately screened.(Figure 1)19

"In 2011in the USA, over 12,000 women developed and over 4,000 women died

of this highly preventable disease."

In 2012, 8 million women were not screened in the last 5 years

7 out of 10 women who were not screened had a regular doctor and health insurance

More than 50% all new cervical cancers are in women who have never been screened, or have not been screened in the last fi ve years

MISSED OPPORTUNITIES FOR CERVICAL CANCER SCREENING

Figure 1. Screening participation in the US in 2012. Source: Behavioral Risk Factor Surveillance System, 2012

24

References: 1. Saslow D et al. Am J Clin Pathol. 2012;137(4):516-542. 2. ACOG. ACOG Practice Bulletin Number 131: Screening for cervical cancer. Obstet Gynecol. 2012;120(5):1222-1238. 3. Moyer VA. Ann Intern Med. 2012;156(12):880-891, W312. 4. Huh WK et al. Gynecol Oncol. 2015;136(2):178-182. 5. Habbema D, De Kok IM, Brown ML. The Milbank quarterly. 2012;90(1):5-37. 6. Vaccarella S et al. British journal of cancer. 2014;111(5):965-969. 7. Saslow D et al. CA Cancer J Clin. 2002;52(6):342-362. 8. Berkowitz Z, Saraiya M, Sawaya GF. JAMA Intern Med. 2013;173(10):922-924. 9. Roland KB et al. Prev Med. 2013;57(5):419-425. 10. Perkins RB et al. Am J Prev Med. 2013;45(2):175-181. 11. Saraiya M et al. Archives of internal medicine. 2010;170(11):977-985. 12. Dinkelspiel H et al. J Low Genit Tract Dis. 2014;18(1):57-60. 13. Teoh DG et al. Am J Obstet Gynecol. 2015;212(1):62 e61-69. 14. Brown ML et al. Preventing chronic disease. 2014;11:E29. 15. Cervical cancer screening among women aged 18-30 years - United States, 2000-2010. MMWR Morb Mortal Wkly Rep. 2013;61(51-52):1038-1042. 16. Silver MI et al. Obstetrics and gynecology. 2015;125(2):317-329. 17. Hawkins NA et al. Prev Med. 2013. 18. Cuzick J et al. Cancer Epidemiol Biomarkers Prev. 2013. 19. Benard VB et al. MMWR Morb Mortal Wkly Rep. 2014;63(44):1004-1009. 20. Giorgi Rossi P, Baldacchini F, Ronco G. Frontiers in oncology. 2014;4:20. 21. Elfstrom KM et al. Cervical cancer screening in Europe: Quality assurance and organisation of programmes. European journal of cancer (Oxford, England : 1990). 2015. 22. Austrailian Government Department of Health. National Cervical Cancer Screening. 23. Kuppermann M, Sawaya GF. JAMA internal medicine. 2015;175(2):167-168. 24. Sawaya GF, Kuppermann M. Obstetrics and gynecology. 2015;125(2):308-310.

As of today, few countries have introduced

the HPV test as a primary screen into organ-

ized screening programs despite large-scale

pilot studies.20,21 Australia recently adopted

one national guideline for primary HPV testing

every 5 years starting at age 25, which is a

change after almost 20 years of recommending

cytology-based screening every 2 years.22 This

has occurred in the context of Austrialia’s HPV

vaccination campaign, which has resulted in

high vaccination coverage among women now

in their early 20s and promises to improve HPV

screening effi ciency by lowering the HPV burden

among screened women. While primary HPV

testing has a lot of promise for the United States,

it is unclear how it compares to other screening

strategies in terms of net benefi t, acceptability

and cost-effectiveness. Indeed, more information

is needed to identify which strategies constitute

“high-value” care. Guidelines for the use of HPV

tests for primary screening among women under

age 30 are now confl icting, with some authors

expressing concern about overtreatment in

young women in whom the prevalence of HPV

is relatively high and where treatment may be

of lesions that would never progress to cancer.24

In addition, some women will be unable to end

screening at age 65 due to persistently positive

HPV tests. In a review by Giorgi-Rossi et al,

there was caution and concern about increased

disparities related to communications about HPV

positivity in that women who were disadvan-

taged and lower educated had higher anxiety

that higher educated women.

Finally, in all of the excitement about new

technologies, we must remember to focus on

improving coverage to women who are not

getting screened at all or not getting screened

regularly, to ensure followup for abnormal

results, and to be clear about how these new

technologies translate into actionable steps for

systems and providers, and improved outcomes

for women.

"Australia recently adopted one national guideline for primary HPV testing every 5

years starting at age 25, which is a change after almost 20 years of recommending

cytology-based screening every 2 years."

HPV TESTING IN THE UNITED STATES: PERSPECTIVES FROM SYSTEMS, PROVIDERS AND WOMEN

The authors declare no confl ict of interest

25

REPORT

THE AUSTRALIAN EXAMPLE:AN INTEGRATED APPROACH TO HPV VACCINATION AND CERVICAL SCREENINGKaren CanfellDirector, Cancer Research Division, Cancer Council NSW, Woolloomooloo, Australia.

In 2014, Australia became the first country to

announce large scale changes to cervical screen-

ing as a direct response to the successful and

widespread national implementation of HPV vac-

cination. The rollout of the Australian National

HPV Vaccination Program from 2007, with routine

vaccination in 12-13 year old girls and an initial two-

year catch-up to age 26 years, has already reduced

confirmed high grade cervical abnormalities in

young women. This, together with an accumulation

of international evidence on primary HPV testing,

has led to the development of new recommenda-

tions for cervical screening in Australia. These

recommendations, which have emerged out of a

structured and evidence-based process of review

(the 'Renewal'), propose that primary HPV screening

be conducted every 5 years in women 25 years

and older, and that women are discharged from

screening in their early seventies.

Australia is thus now transitioning from cytology

screening to an HPV-based cervical screening

program which will be specifically tailored to

interface with HPV vaccination by incorporat-

ing partial genotyping for the vaccine-included

oncogenic types, HPV 16 and 18. From 2017, all

women, whether unvaccinated or vaccinated, will

be offered HPV screening and those found to have

HPV16/18 infections will be classified as higher risk

for development of cervical intraepithelial neopla-

sia grade 3 or invasive cervical cancer (CIN 3+) and

referred directly to colposcopy for further evalu-

ation. Those with other oncogenic type infections

will be classified as intermediate risk and undergo

triage testing, and HPV negative women will be

returned to routine recall at five years. Because

all women in Australia aged 35 years or younger

have now been offered vaccination and coverage

rates have been relatively high, infection rates with

HPV16/18 in the population are expected to be

relatively low and thus colposcopy referral rates

are expected to be lower than, or comparable to,

current rates in the program. Australia will thus be

the first country to implement a population risk

assessment approach to screening in the context

of vaccination, using the screen-positive partially

genotyped HPV results to determine a woman’s

longitudinal risk of developing CIN 3+ and thus

to determine her optimal management, without

needing to know her individual vaccination status

at the time of screening.

As part of the transitional process, a major

randomised controlled trial of 121,000 women

randomised to HPV versus cytology screening is

currently ongoing. This trial, known as Compass,

is stratifying recruitment by whether women are

in birth cohorts offered vaccination, and will thus

provide key information on cervical screening in

a vaccinated population. Compass has already

facilitated the development of systems for pri-

mary HPV screening with partial genotyping, and

the trial will continue to act as a sentinel experi-

ence for the transition of the National Cervical

Screening Program in Australia.

BACKGROUND: CERVICAL SCREENING IN AUSTRALIAAustralia's organised National Cervical Screening

Program, which was established in 1991, rec-

ommends 2-yearly conventional cytology (Pap

smear) screening in sexually active women aged

18-20 to 69 years. Participation rates over 2

years are 58%, with 83% of eligible women being

"These recommendations, which have emerged out of a structured and evidence-based process of review (the 'Renewal'), propose that primary HPV screening be conducted every 5 years in women 25 years and older, and that

women are discharged from screening in their early seventies."

26

screened every 5 years.1 The program has been

very successful in reducing the incidence and

mortality from cervical cancer, which fell by ~50%

in the first decade, although it is notable that

similar falls were also achieved in countries with

longer screening intervals for cervical cytology.2

In the second decade of the cytology screening

program, rates of cervical cancer incidence and

mortality appear to have plateaued, and at the

present time, although Australia is one of the

countries with the lowest incidence of cervical

cancer, it is likely that the cytology screening

program has ‘reached its limits’ due to the

continuing difficulties of reaching some groups

of women (including those in remote and rural

communities) for regular 2-yearly screening and

also due to the limitations of cervical cytology

in detection of adenocarcinoma. It is notable

that the relative proportion of adenocarcinomas

compared to squamous cancers diagnosed has

grown from 11% in 1982 to 26% in 2008 as the

incidence of invasive squamous cervical cancer

has reduced due to the effect of screening.1

Whilst successful, the National Cervical Screening

Program is associated with substantial ongoing

costs. The direct program costs have been esti-

mated as A$195M in 2010 (projected to grow to

$215M in 2015); this is equivalent to a cost of A$23

per adult woman in 2010, whether actually screened

or not.3 Approximately half of these costs are being

spent on frequent ‘front-end’ cytology tests.

Although liquid-based cytology (LBC) was evalu-

ated for use in the program (most recently in

2009) it has not been adopted, in part because

the cost-effectiveness of LBC was adversely

impacted by the frequent screening interval and

the consequent need to apply an incremental cost

for LBC for a large number of screening tests (26

recommended screens per lifetime).

IMPACT OF HPV VACCINATION IN AUSTRALIAAustralia was the first country to initiate a national

publically-funded HPV vaccination program in 2007.

Female vaccination uptake is approximately 71-72%

for 3 dose coverage in 12-13 year old females;

catch-up in 18-26 year old females achieved cover-

age rates of the order of 30-50%.4,5 From 2013,

males aged 12-13 have also been vaccinated at

school with a two-year catch-up to Year 9 (~15

years). Via herd immunity, male vaccination will

also provide incremental benefits to females,

and is expected to lead to further reductions in

vaccine-included types infections and high grade

cervical abnormalities in females.6

Several factors have come together to lead to

a more rapid impact of vaccination on cervical

screening in Australia compared to many other

countries – these include the early introduction

of HPV vaccination, the extended catch-up to

age 26 years, the early age of starting screen-

ing at 18-20 years, and the consequent overlap

of vaccinated and screened populations from

the inception of the vaccination program, and

relatively high vaccination and cervical screen-

ing coverage rates. After the introduction of

vaccination, Australia experienced rapid falls in

"From 2004-6 to 2012, for women aged < 20 years, rates of CIN 2/3 decreased by 53%; for women aged 20-24 years, rates of confirmed CIN 2/3 were stable

until 2010, then decreased by 21% in the following year."

"It is notable that the relative proportion of adenocarcinomas compared to squamous cancers diagnosed has grown from 11%

in 1982 to 26% in 2008 as the incidence of invasive squamous cervical cancer has reduced due to the effect of screening."

27

REPORT

Primary screening test Age range Interval• Evaluation both unvaccinated

and cohorts offered vaccination.

• Total of 132 detailed screen-ing algorithms in main evaluation.

• Supplementary analysis: screening end age 65 or 70 years.

CURRENT PRACTICE: Conventional cytology 18-20 to 69 years 2

1 Conventional cytology

25-65 years

IARC intervals (3-yearly < 50;

5-yrly 50+ years)2 Manually-read LBC+/-HPV triage of LSIL3 Image-read LBC +/-HPV triage of LSIL

4 HPV with LBC triage of pooled oncogenic types

5-yearly5 HPV with partial genotyping for HPV 16/18 & direct referral to colposcopy

6 Co-testing with both HPV and LBC

Table 1. Options considered in the Renewal (review) of the Australian National Cervical Screening Program.

vaccine-included HPV type infections, in anogeni-

tal warts and in histologically confirmed cervical

high grade precancerous abnormalities (CIN 2/3).

These have now been documented extensively

in young females and also in heterosexual males

due to herd immunity effects. From 2004-6 to

2012, for women aged < 20 years, rates of CIN 2/3

decreased by 53%; for women aged 20-24 years,

rates of confirmed CIN 2/3 were stable until 2010,

then decreased by 21% in the following year.1 It is

expected that rates of high grade abnormalities

will continue to decline in these age groups and

that the declines will extend to older age groups

as the cohorts offered vaccination continue to age.

THE RENEWAL OF AUSTRALIA’S NATIONAL CERVICAL SCREENING PROGRAMA major review, known as Renewal, of Australia’s

cervical screening program, was announced in

November 2011. Its aim is “to ensure that all

Australian women, HPV vaccinated and unvacci-

nated, have access to a cervical screening program

that is acceptable, effective, efficient and based on

current evidence.” In the first phase of Renewal,

the Australian government's Medical Services

Advisory Committee (MSAC) commissioned a

systematic review of the international evidence

and modelled evaluation of health outcomes and

costs i.e. an explicitly linked evidence approach

to guide decision making was taken. The process

was guided by an expert reference group, the

Renewal Steering Committee.

A large number of options were considered for

the future screening program, based on six main

primary screening approaches – conventional

cytology, manually-read LBC, image-read LBC,

HPV screening for a pool of oncogenic types

with cytology triage of HPV positive women, HPV

screening with partial genotyping for HPV 16/18,

and adjunctive co-testing using both cytology and

HPV testing (Table 1).

A modelling approach was used to combine the

international evidence on vaccine efficacy and

screening and diagnostic test accuracy with

local information on vaccination and screening

"The Renewal modelling predicted that 5-yearly HPV screening with partial genotyping from age 25 would be

both life year and (potentially) cost saving, and that this would be the most favourable screening approach overall for both unvaccinated and for cohorts

offered vaccination."

THE AUSTRALIAN EXAMPLE: AN INTEGRATED APPROACH TO HPV

VACCINATION AND CERVICAL SCREENING

28

COMPASS TRIAL: A SENTINEL EXPERIENCECompass [Clinicaltrials.gov NCT02328872] is a

large scale randomised controlled trial of 5-yearly

HPV versus 2.5 yearly image read LBC cytology

screening in women aged 25-69 years. Compass is

one of the fi rst large scale cervical screening expe-

riences in a population of women who have been

offered HPV vaccination, and it is being conducted

in the state of Victoria by the Victorian Cytology

Service. Women presenting for screening are

consented by the primary practitioner and an LBC

sample taken with randomisation applied in the

laboratory. HPV screening in the trial incorporates

Figure 1.

1. Australian women should start having HPV tests at 25 years.

2. HPV tests should be undertaken every 5 years until 74 years.

3. Women with positive HPV tests results should be followed up in accordance with cervical screening pathway*.

4. Women 70 to 74 years of age, with a negative HPV test result may exit the cervical screening program.

5. Women 74 years of age and older who have never had, or who request a HPV test at least 5 years after their last cervical screening test, should be screened.

6. HPV an cytology co-testing is not recommended.

*Management Guidance to be updated following development clinical practice guidelines.

Source: National Screening Program Australia, Partner Reference Group E-newsletter. September 2014

THE DRAFT RENEWED NATIONAL POLICY FOR CERVICAL SCREENING IN AUSTRALIA

behaviour and to simulate future outcomes for

the screening program.

The simulation incorporated a dynamic model of

sexual behaviour and HPV transmission, natural

history and screening which was extensively

validated including against post-vaccination out-

comes for infections and CIN 2/3. The Renewal

modelling predicted that 5-yearly HPV screening

with partial genotyping from age 25 would be

both life year and (potentially) cost saving, and

that this would be the most favourable screening

approach overall for both unvaccinated and for

cohorts offered vaccination. It was predicted

that the use of partial genotyping would result

in further improvements in cervical cancer inci-

dence and mortality compared to the current

screening program of at least 13-15%, and up

to 22%, if retaining a screening end-age of 70

years.7 Although partial genotyping strategies

were predicted to increase colposcopies in an

unvaccinated population, in Australia a large

increase in colposcopies was not predicted

because by 2016, women aged 35 years will

have been offered vaccination.7

The MSAC evidence review report was released

on April 28th 2014, with the recommendations

based on the above modelled fi ndings (Figure 1).

A ‘preferred pathway’ or management algorithm

for HPV-positive women was also identifi ed

(Figure 2), but this is yet to be supported by

the development of professional clinical prac-

tice guidelines, which will be occurring as part

of the implementation phase. The Australian

Health Ministers’ Advisory Council has now

endorsed an Interim Implementation Plan and

the transition from evidence to practice will

be guided by a Steering Committee for the

Renewal Implementation Project, and a Quality

and Safety Monitoring Committee has also been

confi gured. The target implementation date

for the Renewed National Cervical Screening

program is May 2017.

29

THE AUSTRALIAN EXAMPLE: AN INTEGRATED APPROACH TO HPV

VACCINATION AND CERVICAL SCREENING

the use of partial genotyping. Recruitment is

stratified according to whether women are in age

cohorts that were offered vaccination (i.e. whether

aged ~35 years or less in 2015).

Compass is a pragmatic trial which has allowed

the development of new systems for HPV

screening, including the implementation of ‘call-

and-recall’, whereby women are proactively

issued an invitation to attend screening when

their test is due. Compass is being performed

in two phases - Phase I (the pilot) has involved

recruiting 5,000 women, and Phase 2 (the main

trial) involves the ongoing recruitment of 121,000

Figure 2. Preferred pathway for cervical screening in Australian: primary HPV screening with partial genotyping.(This preferred pathway was identified as a result of the Medical Services Commitee Evidence Review and modelled analysis; clinical guidelines develop-

ment to underpin this management algorithm is ongoing as part of the implementation phase of Renewal).

Reflex LBC

Repeat HPV test in 12 months

Repeat HPV test in 12 months

Risk of cervical cancer precursors:

Lower

Intermediate

Higher

Reflex LBCto assist management

at colposcopy

Negative HPV

Test was negative (normal)

Test was negative (normal)

Test was negative (normal)

Indicates HPV infection still

present

HPV infection still present

May require treatment

High-risk HPV infection

present

Unsatisfactory test for technical

reasons

Recall for screening in

5 years

Recall for screening in

5 years

Recall for screening in

5 years

Refer to colposcopy

Refer to colposcopy

Refer to colposcopy

Refer to colposcopy

Retest within6 weeks

NegativeHPV

NegativeHPV

Any positive HPV

Any positive HPV

HPV other onco types

Negative Cytology

ASCUS/LSIL Cytology

ASC-H/HSIL cytology

HPV types 16, 18 Unsatisfactory test

HPV TEST WITH PARTIAL GENOTYPING

women. The trial is designed not only to assess

comparative performance of HPV and LBC screen-

ing in both unvaccinated and vaccinated women,

but also to assess optimal triage strategies for

HPV-positive women in both groups.

In HPV-screened women, a secondary randomi-

sation process for intermediate risk women with

other oncogenic HPV infections (i.e. not HPV16/18)

is implemented, these women are randomised to

be triaged either with LBC or with dual-stained

p16/Ki67 cytology (CINtec PLUS, Roche/Ventana).

The sample size of 121,000 incorporates 114,000

women presenting for screening or follow-up and

30

Karen Canfell is a co-PI of an investigator-initiated trial of cytology and primary HPV screening in Australia (`Compass'), which is conducted and funded by the Victorian Cytology Service (VCS), a government-funded health promotion charity. The VCS have received equipment and a funding contribution for the Compass trial from Roche Molecular Systems and Ventana Inc USA. However neither Karen Canfell nor her institution on her behalf (Cancer Council NSW) receives direct funding from industry for this trial or any other project.

References: 1. Australian Institute of Health and Welfare. Cervical screening in Australia 2011-2012. Canberra: AIHW, 2014. 2. Simonella L, Canfell K. Cancer Causes Control. 2013 Sep;24(9):1727-36. 3. Lew JB et al. BMC Health Services Research 2012; 12(1): 446. 4. Gertig DM, Brotherton JML, Saville AM. Sexual Health 2011; 8: 171-8. 5. Brotherton JM et al. Vaccine. 2014 Jan 23;32(5):592-7. 6. 6. Smith MA et al. International Journal of Cancer 2008; 123(8): 1854-63. 7. Lew JB*, Simms K* et al.(*Joint fi rst authors). National Cervical Screening Program Renewal: Effectiveness modelling and economic evaluation in the Australian setting (Assessment Report). Medical Services Advisory Committee Australia. MSAC Application No. 1276. November 2013.

an additional 7,300 recruited to allow for 10%

of HPV negative women to be assigned to early

recall for safety monitoring. The primary outcome

will be cumulative CIN 3+ at 5 years, assessed on

an intention-to-treat basis, following 5 year HPV

exit testing round in both arms. Key secondary

outcomes will include cross-sectional baseline

confi rmed CIN 2+ and CIN 3+ detection rates,

and the rates of cumulative CIN 3+ in baseline

screen-negative women at 5 years.

THE AUSTRALIAN EXAMPLE: AN INTEGRATED APPROACH TO HPV

VACCINATION AND CERVICAL SCREENING

CONCLUSIONAustralia was the fi rst country to implement a free

public HPV vaccination program in young females.

The successful rollout of the vaccination program

and its rapid impact to reduce high grade cervical

abnormalities in young women has prompted a major

review of cervical screening. The Renewed cervical

screening program will be directly tailored to work

with vaccination via specifi c detection and manage-

ment of women with vaccine-included type infections.

HPV16/18 positive women may, for example, have

been in older age cohorts not offered vaccination, or

they may have been infected prior to vaccination, or

they may not have completed the vaccination course

and were subsequently HPV-infected - in any case

they will be managed as higher risk women. Women

positive for other oncogenic type infections will be

managed as intermediate risk via triage testing, and

HPV-negative women will be referred to 5-yearly

screening, which will refl ect a low level of risk even

for unvaccinated women. In the Renewed screening

program, therefore, whether offered vaccination or

not, women will be managed according to their HPV

status and thus their level of risk. Australia is thus

the fi rst country to move to a truly population-based

risk assessment approach to cervical screening in the

context of HPV vaccination.

VACCINATION +SCREENING

A new scientifi c Journal devoted to HPV and other small DNA tumor viruses and the offi cial journal of the International Papillomavirus Society

This special issue has been supported by an unrestricted educational grant from

www.journals.elsevier.com/papillomavirus-research

32

INTERNATIONAL AGENDA

Budapest, Hungary

17th - 20th September 2015

22nd World Congress on Controversies in Obstetrics, Gynecology & Infertility (COGI)Venue: Syma Event and Congress CentreE-mail: cogi@congressmed.comWeb: www.congressmed.com/cogi

Lisbon, Portugal

17th - 21st September 2015

HPV 2015. 30th International Papillomavirus Conference Venue: Lisboa Congress CentreE-mail: ipvsmembership@kenes.com Web:www.hpv2015.org

Milan, Italy

20th - 23rd September 2015

39th European Congress of CytologyVenue: Milano CongressiE-mail: info@cytology2015.comWeb: www.cytology2015.com

Vancouver, Canada

4th - 9th October 2015

XXI FIGO World Congress of Gynecology and ObstetricsVenue: Vancouver Convention CentreE-mail: FIGO2015@icsevents.comWeb: www.fi go2015.org

Prague, Czech Republic

21st - 24th October 2015

11th World Congress of the European Society of Gynecology (ESG)Venue: Prague Congress Centre E-mail: seg2015@btcongress.comWeb: www.seg2015.com

Nice, France

24th - 27th October 2015

19th International Meeting of the European Society of Gynaecological Oncology (ESGO)Venue: Acropolis Congress CenterE-mail: reg_esgo15@kenes.comWeb: http://esgo2015.esgo.org/

Madrid, Spain

9th - 11th November 2015

World Vaccine Congress Europe Venue: Silken Puerta AméricaE-mail: enquiry.uk@terrapinn.comWeb: www.terrapinn.com/conference/world-vaccine-congress-europe/

Chicago, USA

13th - 16th November 2015

63rd Annual Scientifi c Meeting ASC Venue: Hyatt RegencyE-mail: jsmith@cytopathology.orgWeb: www.cytopathologymeeting.org/2015

Rio De Janeiro, Brazil

18th - 21st November 2015

9t h World Congress on Pediatric Infectious Diseases - WSPID 2015 Congress Venue: Sul A merica, Centro de ConvencoesE-mail: reg_wspid2015@kenes.com Web: www.wspid.kenes.com/

Morocco, Marrakech

18th - 22nd November 2015

AORTIC 2015. 10th AORTIC International Cancer Conference on Cancer in Africa Venue: Palais des Congrès de MarrakechE-mail: info@aorticconference.orgWeb: www.aorticconference.org

Melbourne, Australia

21st - 23rd March 2016

23rd World Congress on Controversies in Obstetrics, Gynecology & Infertility (COGI) Venue: Melbourne Convention and Exhibition CentreE-mail: cogi@congressmed.comWeb: http://congressmed.com/cogi23/

Yokohama, Japan

28th May - 1st June 2016

19th International Congress of Cytology Venue: PACIFICO YokohamaE-mail: icc2016@congre.co.jpWeb: www.cytologyjapan2016.com

Melbourne, Australia

21st - 26th August 2016

16th International Congress of Immunology Venue: Melbourne Convention and Exhibition CentreE-mail: ici2016@arinex.com.auWeb: www.ici2016.org

Hamburg, Germany

19th - 22nd October 2016

6th European Congress of Virology Venue: Congress Center HamburgE-mail: ecv2016@interplan.deWeb: www.eurovirology2016.eu/

Lisbon, Portugal

29th - 31st October 2016

16th Biennial Meeting of the International Gynecologic Cancer Society (IGCS 2016) Venue: Lisboa Congress CentreE-mail: igcs2016@kenes.comWeb: www.igcs2016.com

Amsterdam, Netherlands

10th - 13th November 2016

24th World Congress on Controversies in Obstetrics, Gynecology & Infertility (COGI)Venue: to be determinedE-mail: cogi@congressmed.comWeb: www.congressmed.com/cogi/

Cape Town, South Africa

28th February- 4th March 2017

31st International Papillomavirus Conference & Clinical and Public Workshop

Venue: to be determinedE-mail: to be determinedWeb: to be determined

Last accessed 24/08/2015