Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
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: [email protected]
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: [email protected]
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: [email protected]: www.congressmed.com/cogi
Lisbon, Portugal
17th - 21st September 2015
HPV 2015. 30th International Papillomavirus Conference Venue: Lisboa Congress CentreE-mail: [email protected] Web:www.hpv2015.org
Milan, Italy
20th - 23rd September 2015
39th European Congress of CytologyVenue: Milano CongressiE-mail: [email protected]: www.cytology2015.com
Vancouver, Canada
4th - 9th October 2015
XXI FIGO World Congress of Gynecology and ObstetricsVenue: Vancouver Convention CentreE-mail: [email protected]: 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: [email protected]: 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: [email protected]: http://esgo2015.esgo.org/
Madrid, Spain
9th - 11th November 2015
World Vaccine Congress Europe Venue: Silken Puerta AméricaE-mail: [email protected]: www.terrapinn.com/conference/world-vaccine-congress-europe/
Chicago, USA
13th - 16th November 2015
63rd Annual Scientifi c Meeting ASC Venue: Hyatt RegencyE-mail: [email protected]: 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: [email protected] 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: [email protected]: 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: [email protected]: http://congressmed.com/cogi23/
Yokohama, Japan
28th May - 1st June 2016
19th International Congress of Cytology Venue: PACIFICO YokohamaE-mail: [email protected]: www.cytologyjapan2016.com
Melbourne, Australia
21st - 26th August 2016
16th International Congress of Immunology Venue: Melbourne Convention and Exhibition CentreE-mail: [email protected]: www.ici2016.org
Hamburg, Germany
19th - 22nd October 2016
6th European Congress of Virology Venue: Congress Center HamburgE-mail: [email protected]: 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: [email protected]: www.igcs2016.com
Amsterdam, Netherlands
10th - 13th November 2016
24th World Congress on Controversies in Obstetrics, Gynecology & Infertility (COGI)Venue: to be determinedE-mail: [email protected]: 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