Review

Journal of the Royal Society of Medicine; 2020, Vol. 113(2) 64–78

DOI: 10.1177/0141076819899308

Will HPV vaccination prevent cervical cancer?

Claire P Rees1 , Petra Brhlikova2 and Allyson M Pollock2

1
Centre for Global Public Health, Institute of Population Health Sciences, Barts and The London School of Medicine and Dentistry,
Queen Mary University, London E1 2AB, UK
2
Institute of Health and Society, Newcastle University, Newcastle NE2 4AX, UK
Corresponding author: Allyson M Pollock. Email: [email protected]

Summary offered to girls aged 9–13 years before sexual debut

We conducted a critical appraisal of published Phase 2 and and naı̈ve to HPV infection. Box 1 gives an overview
3 efficacy trials in relation to the prevention of cervical of licensing and indications in Europe and the US.
cancer in women. Our analysis shows the trials themselves
Public health agencies promote the position that
generated significant uncertainties undermining claims of
efficacy in these data. There were 12 randomised control
the vaccine has been shown to prevent cervical cancer
trials (RCTs) of Cervarix and Gardasil. The trial popula- (see Supplement 1). Not all routinely emphasise the
tions did not reflect vaccination target groups due to dif- limitations of the evidence or the uncertainties which
ferences in age and restrictive trial inclusion criteria. The we will discuss.
use of composite and distant surrogate outcomes makes it
impossible to determine effects on clinically significant out-
comes. It is still uncertain whether human papillomavirus
Background
(HPV) vaccination prevents cervical cancer as trials were A key issue for the design of trials and studies of
not designed to detect this outcome, which takes decades efficacy is the complexity of the epidemiology of the
to develop. Although there is evidence that vaccination HPV subtypes and the lesions used as surrogate end-
prevents cervical intraepithelial neoplasia grade 1 (CIN1)
points for cervical cancer, each with their own differ-
this is not a clinically important outcome (no treatment is
given). Trials used composite surrogate outcomes which
ent natural histories, prevalence and incidence and
included CIN1. High efficacy against CIN1þ (CIN1, 2, 3 strength of association with cancer. These measures,
and adenocarcinoma in situ (AIS)) does not necessarily especially if combined as composite surrogate end-
mean high efficacy against CIN3þ (CIN3 and AIS), which points in trials, generate new uncertainties.
occurs much less frequently. There are too few data to
clearly conclude that HPV vaccine prevents CIN3þ. CIN i) HPV infection
in general is likely to have been overdiagnosed in the trials
because cervical cytology was conducted at intervals of 6– There are 100þ types of the HPV: 12 of which are
12 months rather than at the normal screening interval of carcinogenic to humans, according to the
36 months. This means that the trials may have overesti- International Association of Cancer Research
mated the efficacy of the vaccine as some of the lesions
(IARC).4 Types vary in prevalence, as does their
would have regressed spontaneously. Many trials diagnosed
persistent infection on the basis of frequent testing at short
association with cervical cancer. HPV vaccines are
intervals, i.e. less than six months. There is uncertainty as licensed for use against oncogenic HPV types 16
to whether detected infections would clear or persist and and 18 and now 31, 33, 45, 52, 58 in Gardasil-9.
lead to cervical changes. Gardasil and Gardasil-9 are also licensed against
non-oncogenic types 6 and 11 linked to genital warts.
Keywords The lifetime risk of an incident of HPV infection is
Vaccination programmes, cervical cancer 79%;5 the majority of HPV infections are transient
and 67% clear within one year.6 Around 10% of
Received: 8th November 2019; accepted: 17th December 2019
women without CIN have HPV infection at any
one time.7 The mechanism of progression from
HPV infection to cervical cancer and its precursors
The human papillomavirus (HPV) vaccination pro- is not well understood.4,8–11
gramme aims to prevent cervical cancer. Globally
around 13.1/100,000 women are diagnosed with cer- ii) Cervical cancer and pre-cancerous lesions as sur-
vical cancer each year.2 Typically, vaccination is rogate endpoints

! The Royal Society of Medicine 2020

Article reuse guidelines: sagepub.com/journals-permissions
Rees et al. 65

Key Messages.

 We do not know how well HPV vaccination will protect against cervical cancer. Trials have not focused on the outcome of
cervical cancer because they had too few participants and did not follow them up for long enough: cervical cancer may take
decades to develop.

 Published numbers from randomised controlled trials may overstate efficacy because: (a) testing occurred too frequently in the
trials when, in real-world settings, lesions may regress spontaneously; (b) trials used composite surrogate outcomes, some of
which, such as HPV-infection and CIN1, occur more frequently than others and are very unlikely to progress to cancer; and (c)
subgroups were over-analysed.

 The trial populations have limited relevance and validity for real world settings: for example, women in the trials were older
than the target population; we do not have enough data on the benefits in women who may have been exposed to HPV before
they were vaccinated and who do not know their HPV status.

 We do not have enough data on the impact of the vaccine on CIN3, which is more likely than CIN1 and 2 to progress to
cervical cancer. We also have less data on the impact on cervical disease due to any HPV type rather than just lesions due to
HPV 16 and 18.

 Women should still attend regular cervical screening because efficacy in preventing cervical precursors is <100% and there are
more oncogenic types than those covered by the vaccines. We have good evidence that cervical screening significantly reduces the
risk of cervical cancer in women regardless of whether they have been vaccinated. The number of new cancers and deaths has
decreased markedly such that cervical cancer now accounts for only 1% of cancer deaths in women in the UK (854 deaths in
2016).1

 Information from the trials can tell us what happens between five and nine years after vaccination, but we do not know if
protection wanes after this time.

 A recent observational study provides some evidence of efficacy against CIN3þ in girls vaccinated before sexual debut.
Ongoing observational studies may tell us about the long-term effect on rates of cervical cancer, but it will take many years
before we have the evidence.

Box 1. Licensing and guidelines.

Licensing

 Gardasil, Gardasil-9 and Cervarix vaccines have been approved for marketing and used in females and males from the age of 9
years throughout the world to prevent cervical cancer.

 The European Medicines Agency (EMA) and US Food and Drug Administration (FDA) granted marketing approval for Gardasil
in 2006, and for Cervarix in 2007 and 2009, respectively.

 Gardasil-9 was approved in 2014 by the FDA and in 2015 by the EMA, but it is not currently used in the UK.

 The EMA has licensed all three vaccines for females and males with no upper age limit. The FDA has licensed Gardasil up to age
26 and Gardasil-9 up to age 45 for females and males, and Cervarix for females only up to age 25.

Guidelines

 The US Centers for Disease Control and Prevention recommends ‘routine vaccination at age 11 or 12 years. (Vaccination can
be started at age 9.) The Advisory Committee on Immunization Practices also recommends vaccination for females aged 13
through 26 years not adequately vaccinated previously’.3

 The UK uses Gardasil. Public Health England advises girls to be vaccinated from age 12–18 years. Immunisation Scotland offer
the vaccine for girls aged 11–13 years. There is a planned roll-out to boys aged 12–13 in England and Scotland.

Estimated rates of regression and progression The IARC has acknowledged that composite end-
for CIN1, 2 and 3 are presented in Figure 1.12 points in intervention studies involving CIN2 are
However, there remain uncertainties due to methodo- sub-optimal13 as CIN2 is often misclassified due to
logical issues in the epidemiological studies from its diagnosis having lower reproducibility and valid-
which these findings originate.12 ity.14 Women with CIN2 are currently offered
66 Journal of the Royal Society of Medicine 113(2)

Figure 1. CIN natural history. itself has raised concerns.11 We chose to focus on
randomised controlled trials as this offers the highest
level of evidence and this is the evidence used for
decisions by regulatory bodies and decisions on initi-
ating vaccination programmes.
We also found 39 meta-analyses and systematic
reviews of HPV vaccine efficacy; of them many are
restricted to post-hoc analyses of subgroups and have
inappropriately combined trials in the same analysis,
e.g. for different vaccines (see Supplement 3). The
2018 Cochrane review18 has been criticised for failing
to include all relevant trials, ignoring evidence of
treatment which complicates research into progres- harms and using composite endpoints with different
sion to CIN3. natural histories.19
CIN3 can develop via progression of CIN1 and We compared the eligibility criteria, testing meth-
CIN2 or directly as a result of HPV infection, so ods for HPV and cervical lesions, outcomes measures,
CIN1 and CIN2 may not be good predictors of pro- length of follow-up, target group and subgroup def-
gression. Rate of progression from CIN3 to invasive initions used in the different trials. We focused on the
cancer is likely to be higher than Ostor’s estimate of evidence for efficacy for CIN3þ and 12-month per-
>12%.12 Lifetime risk may be up to 40% without sistent infection which are deemed the more stringent
cervical screening and treatment.15 outcome measures.

Aim Does HPV vaccine prevent cervical cancer?

To describe the uncertainties generated by the design None of the trials were designed to determine efficacy
of Phase 2 and 3 efficacy trials for prevention of cer- or effectiveness against cervical cancer. There were no
vical cancer and its precursors and how they affect reported cases of cervical cancer in any trials; one
the interpretation of efficacy data. case of vulval carcinoma was reported in the vacci-
nated group of FUTURE I.48
The time between first exposure to HPV and peak
Methods and analysis development of CIN3 is 7–10 years.13 It takes a fur-
We searched Embase and Medline for papers relating ther 10 years or so for cervical cancer to develop
to blinded controlled trials of HPV vaccination in according to natural history studies.13 All trials had
females considering efficacy against cervical cancer a mean length of follow-up of six or fewer years,
and its precursors. See Supplement 2 for search strat- apart from the HPV-023 extension with a mean
egy. No trial has tested Gardasil and Cervarix vac- follow-up of 8.9 years.
cines head to head using clinical outcomes (rather
than immunological outcomes). We reviewed 35 pub- Does HPV vaccine prevent pre-cancerous
lished papers relating to 12 published randomised
blinded non-HPV vaccine-controlled Phase 2 and 3
lesions?
trials of Gardasil and Cervarix conducted from 2001 There were eight powered outcomes relating to cer-
to 2016 assessing efficacy against cervical cancer and vical disease used across the trials, all of which were
its precursors (Table 1 and Figure 2). Throughout surrogates or composite surrogate outcomes
this article, we refer to trials by their protocol name (Table 2). Surrogate outcomes are biomarkers used
as presented in bold in Table 1. But for trials 104798 as substitutes for clinical endpoints and used to pre-
and 107638, we use the name of their first authors, dict an intervention’s benefits. Key pre-marketing
Konno and Zhu, respectively. trials evaluated the effect of HPV vaccines against
We excluded trials of the HPV 16 monovalent vac- pre-cancerous lesions, endpoints that were accepted
cine (as it was not licensed and data suggest it had by authorities as a surrogate for cervical cancer.
different pharmacodynamics to Gardasil and There are limitations to surrogate outcomes in gen-
Cervarix16) and the one efficacy trial of Gardasil-9 eral,54 and for HPV and cervical cancer. Using com-
(as the control was Gardasil, not a placebo).17 All posite surrogate outcomes (combining two or more
the efficacy studies used an active vaccine (Hepatitis surrogate outcomes together) is problematic because
A) as the control, or a control containing an alumi- of differences in epidemiology and their natural his-
nium-adjuvant, rather than a true placebo; this in tory and management – see ‘Background’ section and
Table 1. All Phase 2 and 3 trials by vaccine and control, sponsor, country, start and end dates, number and age of participants and length of follow-up.
Average (mean)
Trial name, NCT Number of Age at length of
Vaccine and control number (and sponsor) Research papers Phase Country Start date End date participants enrolment follow-up

Cervarix versus HPV-001/580299 Harper et al.20 2 USA (incl. Puerto Rico), Jan-01 Apr-03 1113 15–25 Up to 27 months
aluminium-hydroxide NCT00689741 Brazil, Canada (average not given)
containing control (GSK)

Cervarix versus placebo HPV-007; follow-on Harper et al.,21 2 USA, Brazil, Canada Nov-03 Aug-07 776 15–25 5.9 years from first
aluminium-hydroxide study from Romanowski et al.22 vaccination
containing control HPV-001
NCT00120848
(GSK)

Cervarix versus HPV-023/109616; De Carvalho et al.,23 2 Brazil Nov-07 Jul-08 437 15–25 8.9 years from first
aluminium-hydroxide follow-on study Roteli-Martins et al.,24 vaccination
containing control from HPV-001 and Naud et al.25
HPV-007
NCT00518336 (GSK)

Cervarix versus Hepatitis 104798 (Konno) Konno et al.26,27 2 Japan Apr-06 Feb-09 1040 20–25 24 month after first
A vaccine NCT00316693 vaccination
(GSK)

Cervarix versus Hepatitis PATRICIA/HPV-008 Paavonen et al.,28,29 3 USA, Australia, Belgium, May-04 Nov-09 18,644 15–25 Mean 43.7 months
A vaccine NCT00122681 Lehtinen et al.,30 Brazil, Canada, Finland, (median 47.4)
(GSK) Wheeler et al.,31 Germany, Italy, Mexico,
Palmroth et al.,32 Philippines, Spain,
Szarewski et al.,33 Taiwan, Thailand, UK
Apter et al.,34
Struyf et al.35

Cervarix versus Hepatitis Costa Rica Vaccine Herrero et al.,36 3 Costa Rica Jun-04 Dec-10 7465 18–25 53.8 months
A vaccine Trial/CVT/ Kreimer et al.,37 (initiation (final data
HPV-009 Rodriguez et al.,38 into trial) collection
NCT00128661 Hildesheim et al.,39,40 for primary
(NCI & GSK) Beachler et al.41 outcome)

Cervarix versus VIVIANE/HPV-015/ Skinner et al.,42 3 Australia, Canada, Mexico, Feb-06 Jan-14 5747 26þ 5.9 years TVC from
aluminium-hydroxide 104820 Wheeler et al.43 Netherlands, Peru, first vaccination,
containing control NCT00294047 Philippines, Portugal, 5.7 years in ATP-E
(GSK) Russia, Singapore, group from third
Thailand, UK, USA vaccination

Cervarix versus 107638/Zhu Zhu et al.44,45 3 China Oct-08 Oct-14 6051 18–25 Mean 57 months TVC-
aluminium-hydroxide NCT00779766 E from first vaccin-
containing control (GSK) ation, 52 months
ATP-E group from
third vaccination

(continued)
Table 1. Continued.
Average (mean)
Trial name, NCT Number of Age at length of
Vaccine and control number (and sponsor) Research papers Phase Country Start date End date participants enrolment follow-up
46,47
Gardasil versus V501-007 Villa et al. 2 Brazil, Finland, Sweden, May-00 May-04 552 (initial 16–23 Initial study up to 36
aluminium NCT00365716 Norway, USA study up to months (average not
hydroxyphosphate (Merck) three years), given). Extension
sulphate adjuvant 241 (extension study up to five
containing control study up to years (results given
five years) for 226 women who
completed study to
60 months, average
not given)

Gardasil versus FUTURE I/ Garland et al.48 3 Australia, Austria, Brazil, Dec-01 Jul-07 5455 16–24 Average three years
aluminium V501-013 Canada, Colombia, from first
hydroxyphosphate NCT00092521 Czech Republic, vaccination
sulphate adjuvant (Merck) Germany, Hong Kong,
containing control Italy, Mexico, New
Zealand, Russia,
Thailand, UK, USA (incl.
Puerto Rico)

Gardasil versus FUTURE II/ Future II Study 3 Brazil, Colombia, Denmark, Jun-02 Jul-07 12,167 15–26 Average three years
aluminium V501-015 Group49 Finland, Iceland, Mexico, from first
hydroxyphosphate NCT00092534 Norway, Peru, Poland, vaccination
sulphate adjuvant (Merck) Singapore, Sweden, UK,
containing control USA (incl. Puerto Rico)

Gardasil versus FUTURE III/V501- Munoz et al.50 3 Colombia, France, Jun-04 May-09 3819 24–45 Median 4 years (mean
aluminium 019 Castellsague et al.51 Germany, Philippines, 3.8 years)
hydroxyphosphate NCT00090220 Thailand, USA, Spain
sulphate adjuvant (Merck)
containing control

Gardasil versus V501-027 Yoshikawa et al.52 2 Japan Jun-06 Sep-09 1021 18–26 30 months (23 months
aluminium NCT00378560 after month 7)
hydroxyphosphate (Merck)
sulphate adjuvant
containing control

Gardasil versus V501-041 Wei et al.53 3 China Jan-09 Sep-16 3006 20–45 Mean 6.07 years after
aluminium NCT00834106 first vaccination
hydroxyphosphate (Merck)
sulphate adjuvant
containing control
Rees et al. 69

Figure 2. Phase 2 and 3 randomised controlled trials on HPV vaccination efficacy.

Trials
Cervarix

Gardasil HPV-16
Costa Rica Gardasil-9 monovalent
Vaccine Trial vaccine

HPV-001 V501-005
V503-001
Konno HPV-007
FUTURE I
PATRICIA HPV-023
FUTURE II
Zhu
V501-007
VIVIANE
FUTURE III

V501-027

V501-041

Legend: Trial sponsor

GSK

NCI & GSK

The study was sponsored and funded by

NCI, but it was done under an FDA
invesgaonal new drug applicaon
(IND), for which the manufacturer has to
funcon as sponsor.

Merck

Figure 1. Five of the 12 trials (VIVIANE, V01-007, recommended for CIN1. Seven trials (FUTURE I,
FUTURE III, V501-027, V501-041) were powered VIVIANE, V01-007, FUTURE III, V501-027, Zhu,
for composite outcomes that combined cervical dis- V501-041) included CIN1 with CIN2, CIN3 and AIS
ease and persistent HPV infection.43,47,51–53 Four in the same primary outcome (making a composite
trials combined cervical disease and genital warts in outcome), potentially inflating vaccine efficacy as
the same primary outcome (FUTURE III, V501-007 there are many more CIN1 cases than
and V501-027, V501-041) thereby inflating efficacy CIN2þ.43,45,47,48,51–53
measures.47,51–53
CIN2þ
CIN1þ The incidence (rate of detection) of CIN2, CIN3 and
The trial outcomes included surrogates CIN1 and AIS in the trials was low so although many trials
CIN2, which are more common than CIN3/AIS showed high efficacy for the vaccine, this was in the
and cervical cancer, but which often regress and are context of very few cases of CIN2þ. For example, the
of limited clinical concern (see ‘Background’ section HPV-023 trial showed high vaccine efficacy (100%
and Figure 1). For example, intervention is not against CIN2 and CIN3 over nine years follow-up)
70 Journal of the Royal Society of Medicine 113(2)

Table 2. Powered endpoints across trials.

Number of trials
Powered endpoint using the endpoint Vaccine: Trials using the endpoint

HPV 16/18 incident infection 1 Cervarix: HPV-001/007/023

HPV 16/18 12-month persistent infection 1 Cervarix: CVT

HPV 16/18 6-month persistent infection 1 Cervarix: Konno

HPV 6/11/16/18 CIN1þ 1 Gardasil: FUTURE I

HPV 16/18 CIN2þ 5 Cervarix: CVT, PATRICIA, Zhu

Gardasil: FUTURE II, V501-041

HPV 16/18 6-month persistent infection or 2 Cervarix: VIVIANE, Zhu

CIN1þ

HPV 16/18 6-month persistent infection or 1 Gardasil: FUTURE III

external genital lesions or CIN1þ

HPV 6/11/16/18 6-month persistent infection 4 Gardasil: V501-007, FUTURE III, V501-027,
or external genital lesions or CIN1þ V501-041

with very low incidence (only three cases, all in the zero).49,56 Vaccine efficacy against CIN3 and AIS
control group, out of 212 participants).25 The trials due to any HPV type varied substantially between
were powered for the minimum number of events the vaccines.30,56
needed to obtain a statistically significant result,
and many trials were designed to stop once this What is the evidence that vaccination
number had been achieved. But the powered out-
prevents clinically meaningful HPV infection?
comes often included CIN1, which means those
trials were not powered to reach a minimum It is possible to diagnose new HPV infections
number of higher-grade CIN cases. Instead of mul- (incident) and ongoing infection (persistent).
tiple short duration trials, this problem of power Studies have shown median length of HPV 16 infection
could have been overcome by having one large trial to be 8.5–19.4 months and HPV 18, 7.8–12 months.13
of longer duration in each country. The HPV001/007/023 trial used incident infection
of HPV 16/18 as the primary outcome.20,22,25 The
results are not relevant to policy decision making as
CIN3þ
the current consensus reported by the WHO is that
CIN3 is generally agreed to be the best marker for incident HPV infection is not an adequate surrogate
risk of cervical cancer, with rates of progression of at outcome because it rarely progresses to cervical
least 12%.12 New evidence suggests that clinical inter- disease.57
vention following detection on screening may be best There is a lack of agreement on what time period
reserved for women with CIN3.55 Only three of the 12 defines persistent infection,13 and the trials may have
trials (FUTURE I, FUTURE II, PATRICIA) overestimated vaccine efficacy by picking time peri-
reported CIN3þ or AIS in subgroups that repre- ods that are shorter than the duration of most self-
sented the target population of women naı̈ve to limiting infections, for example six months. In some
HPV (see Supplement 4).30,49,56 The incidence of trials, the testing interval for diagnosing six-month
AIS in the trials is very low and only three trials persistent infection was four months36,47,52 or five
(FUTURE I, FUTURE II, PATRICIA) published months.44,51,53
results for AIS alone.30,49,56 Only one Gardasil trial, V501-041, used 12-month
In these three trials, vaccine efficacy against CIN3 persistent infection as an outcome; however, the
and AIS due to HPV 16/18 was 100% (see study authors only presented data for combined
Supplement 4) but there were small numbers and HPV 6/11/16/18, not for 16/18 or any oncogenic
wide confidence intervals, sometimes showing non- type.53 In the Cervarix trials, 12-month HPV 16/18
significance (where the confidence interval crosses persistent infection vaccine efficacy varied from
Rees et al. 71

85.3 to 100% (see Supplement 5).25,27,29,33,45 Vaccine How similar were the females in the trials to
efficacy for 12-month persistent infection by any
oncogenic HPV type ranged from 10.4% to 50.1%
the target vaccination groups?
across trials with wide confidence intervals for most Females in the trials are typically older than those in
trials.25,27,29,33,45 The results were not statistically sig- real-life vaccine programmes, and it is unclear
nificant for the HPV-023 trial and Zhu (see whether their outcomes are similar. We do not
Supplement 5).25,45 know efficacy rates in girls aged between 9 and 13
Not all trials analysed HPV types 16 and 18 sep- years.
arately. The incidence of HPV infection varies by The youngest trial participants were aged 15 years
HPV type.30 HPV 18 was much less common than and trials did not restrict recruitment to girls before
HPV 16. Combining their results makes the efficacy sexual debut. Therefore, previous exposure to HPV is
against HPV 18 appear more solid. In some trials the likely for some girls. Per-protocol subgroups with
results for HPV 18 on its own were not statistically much fewer participants were used to analyse those
significant and were only significant when combined with no evidence of previous HPV exposure but as
with results for HPV 16. For example, in the per- shown earlier, most trials did not present data for
protocol population subgroup of V501-027, the six- CIN3þ outcomes in these subgroups.
month persistent infection or genital disease (the trial Efficacy in girls aged 9–13 years has been estimated
primary outcome) vaccine efficacy was 100% using immunobridging trials (where immune response
(59.7,100) for HPV 16, 86.0% (8.9, 99.7) for HPV levels are measured) rather than using clinical out-
18 and 94.5% (65.2, 99.9) for HPV 16/18.52 In the comes.60 We do not know what level of antibody
according-to-protocol cohort for efficacy (ATP-E) titres define a surrogate level of protection against
subgroup of the PATRICIA trial, CIN3þ vaccine cervical cancer or its precursors and how long pro-
efficacy for HPV 16 was 90.2% (59.7, 98.9), HPV tection will last (Gardasil anti-HPV 18 titres are not
18: 100% (8.2, 100) and HPV 16/18: 91.7% (66.6, different from natural infection as early as 24 months
99.1), respectively.30 This means the vaccine may not after vaccination).60 Therefore, it is possible that pro-
protect as well against cervical cancer related to HPV tection will wane by time of peak exposure when
18. The proportion of cervical cancers related to HPV vaccinated at an earlier age.
18 ranges from 13% in South/Central America to Three trials recruited older women (FUTURE III
22% in North America.58 (aged 24–45), VIVIANE (aged 26þ) and V501-041
(aged 20–45)).42,50,53 In VIVIANE a subset of up to
How much information is there on 15% of women with a history of HPV-associated
long-term outcomes and how long infection or disease were included (defined as two or
more abnormal smears in sequence, abnormal colpos-
does protection last? copy, or biopsy or treatment of the cervix after
All trials were six or fewer years in length, apart abnormal smear or colposcopy findings) but this
from the extension study HPV-023 with mean fol- means that the Total Vaccine Cohort may not reflect
low-up of 8.9 years, (which maintained blinding and the proportion of women with a history of HPV-
kept a control group) it only included 437 of the origi- associated disease in the wider population.42 There
nal 1113 participants in HPV-001.25 The longest study were then restrictions based on HPV DNA and ser-
of Gardasil was V501-041, which was extended from 30 ostatus for inclusion in the according to protocol for
to 78 months with 2601 out of the initial 3006 efficacy and total vaccine cohort for efficacy sub-
participants.53 groups. In FUTURE III, women with a history of
Features of the trials may bias the findings in over- past or present genital warts or cervical disease were
estimating long-term efficacy. For example, HPV 16/ excluded; the primary tests of efficacy were in the
18 related CIN3 presents earlier than non-vaccine HPV type-specific per-protocol efficacy analyses
type CIN3 so shorter efficacy trials will be biased in (PPE), which required women to be seronegative to
favour of finding HPV 16/18 related CIN3.59 relevant type on day 1 and PCR negative to that type
Although incidence and progression of disease in cervicovaginal swabs or biopsy samples, or both,
differ over time and by age, V501-007 combined the from day 1 until month 7.50 V501-041 excluded
results of participants from the original trial with women with more than four previous sexual partners
those who completed an extra two-year extension.47 and those with a history of genital warts or ‘signifi-
In HPV-007 and HPV-023, results for participants cant cervical disease’ – the study authors did not spe-
from the preceding trials were considered together.22,25 cify what this meant.53
72 Journal of the Royal Society of Medicine 113(2)

Seven trials excluded women with more than four definitions varied across trials, so that results cannot
to six previous sexual partners.20,28,46,48,52,53,61 Five be compared across trials. Results were not given for
excluded women with previous abnormal cervical all subgroups, and were not broken down by country,
smears20,46,48,52,61 and an additional three excluded by study site or for each outcome. This is important
women with a history of previous colposcopy.26,28,44 given different epidemiology of HPV in different
CVT was the only trial with no restrictions based on areas of the world. It may have been reported this
genital warts or cervical or sexual history.36 These way because incidence was low. We have included a
restrictions may make the vaccine appear more effi- table giving the different subgroup definitions in
cacious in the intention-to-treat (ITT) population Supplement 7.
than in the general population of women of the
same age. In CVT, the efficacy in the ITT group for
Problems with reporting of trial results
12-month persistent infection with any oncogenic
HPV type was 11% (95% CI 2.2, 19.5); they did The trials report vaccine efficacy as the primary out-
not give efficacy against CIN3þ in the ITT group.36 come, which shows relative risk reduction. This can
There is also global variation in the epidemiology over-emphasise efficacy compared with absolute risk
of HPV which means that the trial findings may be reduction such as numbers needed to vaccinate,
poorly generalisable to some settings, including which is more useful for clinicians, patients and
Africa, so it is important to know if results differ by policy makers. None of the trials gave numbers
study region. None of the trials considering efficacy needed to vaccinate. CVT is the only trial that pre-
outcomes were conducted in Africa (we are only sented results in terms of absolute risk reduc-
aware of a safety and immunogenicity trial in tion.36,38,39 The absolute risk reduction for the
Africa62) despite this being the world region with by PATRICIA trial for CIN3þ due to any oncogenic
far the highest incidence of cervical cancer.2 The HPV type (see Supplement 4) (our calculation) is
Cochrane review acknowledges that ‘differences in 0.75%, giving a number needed to vaccinate of 133.
the population HPV prevalence in the trial sites, or
differences in study protocols and assays used, may
explain the contrast in efficacy’ between Cervarix and
Frequency of cervical screening
Gardasil.18 All the trials did Pap cytology at 6–12 monthly inter-
vals. Cervical cancer screening is recommended in
What is the risk of oncogenic HPV-type England every three years, between the age of 25
and 49 years.65 Increased frequency of screening can
substitution? lead to over-diagnosis and overtreatment of cyto-
Vaccines may protect against HPV types, which are logical abnormalities that would normally resolve
not included in the vaccine. There was some evi- and not be detected.66 Increasing the frequency of
dence of cross-protection against three high-risk testing suits early trial completion but may overesti-
HPV types (31, 33 and 45) for Cervarix (see mate vaccine efficacy.
Supplement 6).31,36,43,45 There was cross-protection
against one non-vaccine HPV type by Gardasil31
(see Supplement 6).63 But there was evidence of a
Testing methods for HPV
statistically significant increased risk of HPV type The tests for DNA positivity to a particular HPV
51 and 58 in the Cervarix trials, compared with the type (indicating ongoing infection) and seropositivity
control vaccine.31,36 It is unknown whether vaccine (indicating previous infection) have limited specificity
targeting will lead to substitution by other oncogenic and sensitivity.67 This is another reason HPV infec-
types, as with pneumococcal vaccination.64 tion has limitations as a surrogate of cervical cancer.
Only 50–70% of HPV infections result in detectable
anti-HPV responses,51 and initial seropositive status
Methodological factors from the trials which may revert to negative.68 So subgroups of women
may affect interpretation of the results considered naı̈ve to HPV may have had previous
exposure. Also, latent infection may be undetectable
Multiple underpowered analyses on current tests. The IARC has noted that ‘it is not
All trials undertook multiple subgroup analyses, known how frequently this [latent infection] occurs in
which increase the likelihood of positive statistical immunocompetent individuals, how long it lasts,
findings in the absence of true effect. The subgroup what causes re-emergence into a detectable state or
Rees et al. 73

what fraction of cancers arises after a period of screening at age 20 years (now changed to age
latency’.13 This also raises the question of whether 25 years in line with England) with underrepresenta-
subgroups naı̈ve to HPV can reliably represent girls tion of the unvaccinated group (23% screening
before sexual debut, and whether HPV infection is a attendance versus 51% in the vaccinated group at
valid surrogate outcome. aged 20 or 21); and shorter follow-up time for
women born in 1995 and 1996 necessarily affects
Meta-analysis of limited value due to trial the robustness of the estimate of vaccine effectiveness
for younger women. In addition, the basis for the
heterogeneity claim of herd protection is not well explained for
Differences in trial endpoints and subgroups limit the the unvaccinated women in the 1995–1996 cohort,
ability to compare and aggregate data from trials. compared with unvaccinated women in 1988–1990.
This is compounded by lack of standardisation Nor do the authors consider how changes in sexual
across studies for a range of measures: tests of previ- activity may have contributed to the observed
ous HPV exposure, serological assays to detect HPV decrease in CIN prevalence independent of the vac-
infection and sampling methods including frequency cine: between 2002 and 2014 (the latest period for
of testing.67 which there are data) the proportion of 15-year-olds
There are no agreed criteria for defining the causal in Scotland who have ever had sex reduced, although
HPV type for clinical lesions, and different trials used socio-economic inequalities persist for sexual initi-
different criteria (see Supplement 8). There was no ation and condom use.71 Screening uptake also
standard approach to assess efficacy against disease varies by socio-economic status.72
and infection due to HPV types not found in the vac-
cine, for example whether they considered non-
vaccine oncogenic types or all non-vaccine HPV types What should we do in the light of the
(see Supplement 9). Given that Merck and GSK were uncertainty?
involved in all the trials it is unclear why there was no
consistency in methods and tests across trials.
Policy
We ask policy makers to:
Is ongoing research likely to resolve the
1. Establish national baseline epidemiological data
uncertainties?
on cancer incidence, mortality and HPV subtype
The focus of this paper is randomised controlled prevalence to support evidence based decisions
trials, but we have also looked at whether observa- about whether the currently available vaccines
tional studies can answer some of the uncertainties, are likely to be cost-effective and should be a
acknowledging that this is a lower level of evidence priority.
but practically the most likely source of future infor- 2. Ensure that cancer surveillance and registries are
mation in the absence of long-term randomised con- in place before any vaccination programme is
trolled trials. We identified 19 Phase 4, observational implemented so that changes in incidence of cer-
and non-blinded follow-up studies (including a meta- vical cancer and its precursors can be studied.
analysis of ecological studies) that are potentially 3. Initiate national long-term efficacy and effective-
relevant to the uncertainties discussed in this paper ness studies that are free of industry funding,
(see Supplement 10). None of these studies is ideal. focus on clinically meaningful outcomes, and
Many are small, of short duration or not looking at enrol and analyse the vaccine target populations.
CIN3þ. One observational study (#5) showed a
reduction in relative risk of CIN3 amongst those vac-
cinated of 0.45.69 The PATRICIA trial follow-up
Research
(#3), the only trial planned for 20 years post-vaccina-
tion, and the Mexican FASTER trial (#18) are likely In the UK, cervical screening is estimated to prevent
to provide more long-term efficacy data on the more more than 80% of cervical cancers.15 A cost-effective-
clinically relevant efficacy endpoints. ness analysis in Australia suggested that immunisa-
The recently published observational study con- tion is not cost-effective in settings with established
ducted in Scotland by Palmer et al.70 provides new cervical screening.73 We still do not know how many
evidence on reduction of CIN3þ regardless of HPV cases of cervical disease prevented by vaccination
type. The authors note the following limitations would have been detected by cervical screening.
which may have inflated measures of efficacy: the Further research is needed on whether adding vac-
study gathered data only on the first round of cervical cination where screening exists will be cost-effective.
74 Journal of the Royal Society of Medicine 113(2)

Box 2. Recommendation for future trials to address the against CIN3þ (CIN3 and AIS), which occurred
uncertainties. much less frequently. There are too few data to
clearly conclude that HPV vaccine prevents
 Vaccinate prior to onset of sexual activity and begin CIN3þ. CIN in general is likely to have been over-
assessment of endpoints at age of usual cervical diagnosed in the trials because most carried out
screening once sexually active
cervical cytology at intervals of 6–12 months
 Make all clinical study reports including anonymised rather than at the normal screening interval of
individual patient data publicly available 36 months. This means that the trials may have
overestimated the efficacy of the vaccine as some
 Separate trials to assess the benefit in women already of the lesions would have regressed spontaneously.
exposed to HPV, without restrictions based on risk
3. Persistent HPV infection – The outcomes for HPV
factors
infection are difficult to interpret. Many trials
 Analyse data by country and study site diagnosed persistent infection on the basis of fre-
quent testing at short intervals, i.e. less than six
 Ensure the testing interval is in line with usual cervical months. This leaves uncertainty as to whether
screening protocols
detected infections would clear or persist and
 Continue follow-up for minimum 20 years from the lead to cervical changes. In the current Public
time of sexual debut Health England cervical screening programme,
patients who are HPV positive but cytology nega-
 Power trials for primary composite outcome CIN3/ tive are not retested for 12 months.74
AIS/cervical cancer due to oncogenic HPV types
4. Differences between trial and real world populations
 Define secondary outcome of persistent infection with – Most of the people in the trials were older than
HPV 16/18 at a minimum of 12 months the 9- to 13-year-olds who are typically offered
vaccination. Efficacy in girls aged 9–13 years has
 Use standardised testing methods for HPV detection. been estimated using immunobridging trials (where
 Undertake a saline placebo-controlled efficacy trial of immune response levels are measured) rather than
Gardasil-9 in previously unvaccinated participants, as it using clinical efficacy outcomes.60 We do not know
is difficult to draw conclusions on efficacy and risk of what level of antibody titres protect against cervical
harms based on the trial comparing Gardasil-9 against cancer and its precursors, or how long protection
Gardasil. will last.60 Similarly data on the outcomes for
women older than 24 years are limited, and all
trials apart from the Costa Rica Vaccine Trial
Box 2 shows our recommendations for further (CVT) had exclusions on eligibility related to
research to address uncertainties. We also call for sexual history or history of genital warts or cervical
more research on HPV to be free from industry disease, limiting the generalisability to catch-up vac-
funding. cination populations.36 HPV epidemiology varies
globally. No efficacy studies were done in Africa.
5. Cross-protection and HPV-type substitution –
Conclusion There is uncertainty about whether the vaccine
This review has revealed many methodological prob- will provide cross-protection against oncogenic
lems with the Phase 2 and 3 efficacy trials of HPV HPV types not targeted by the vaccines. There is
vaccination leading to uncertainty regarding under- also a risk of substitution where a non-vaccine
standing its efficacy. oncogenic HPV type fills the void left by the reduc-
tion of an HPV type targeted by the vaccines.
1. Cervical cancer – It is uncertain whether HPV vac- 6. Methodological considerations – Many trials
cination prevents cervical cancer. The trials were included multiple underpowered subgroup ana-
not designed to detect this outcome, which takes lyses, which increase the chance of false-positive
decades to develop. For most outcomes, follow-up findings. All trials except CVT reported relative
data exist for an average of only four or five years. rather than absolute effects, which tend to over-
2. CIN – There is evidence that vaccination prevents state efficacy, and none provided numbers needed
CIN1; however, this is not a clinically important to vaccinate.
outcome (no treatment is given). Trials used com- Declarations
posite surrogate outcomes which included CIN1,
Competing Interests: None declared.
but high efficacy against CIN1þ (CIN1, 2, 3 and
AIS) does not necessarily mean high efficacy Funding: None declared.
Rees et al. 75

Ethics approval: Not applicable.

10. Haug CJ. Human papillomavirus vaccination – reasons
Guarantor: AMP. for caution. N Engl J Med 2008; 359: 861–862.
11. Tomljenovic L, Spinosa JP and Shaw CA. Human
Contributorship: AMP had the idea for the article; CPR did the papillomavirus (HPV) vaccines as an option for pre-
literature review and wrote the first drafts of the article; CPR, venting cervical malignancies: (how) effective and
AMP and PB contributed to the analysis and redrafting the article.
safe? Curr Pharm Des 2013; 19: 1466–1487.
Acknowledgements: We are grateful to Peter Roderick for 12. Ostor AG. Natural history of cervical intraepithelial
commenting on drafts and help with editing and to the reviewers neoplasia: a critical review. Int J Gynecol Pathol 1993;
and the editorial team at the BMJ. We would also like to thank the 12: 186–192.
Newcastle University PPI team for their helpful comments. 13. IARC. IARC Monographs on the Evaluation of
Carcinogenic Risks to Humans. Volume 90. Human
Provenance: Not commissioned; editorial review of peer review
Papillomaviruses. Lyon: IARC, 2007.
comments from another journal and subsequent revision.
14. Carreon JD, Sherman ME, Guillen D, Solomon D,
ORCID iD Herrero R, Jeronimo J, et al. CIN2 is a much less repro-
Claire P Rees https://orcid.org/0000-0001-5649-7835 ducible and less valid diagnosis than CIN3: results
from a histological review of population-based cervical
Supplemental Material: Supplemental material for this article is samples. Int J Gynecol Pathol 2007; 26: 441–446.
available online. 15. Peto J, Gilham C, Fletcher O and Matthews FE. The
cervical cancer epidemic that screening has prevented
in the UK. Lancet 2004; 364: 249–256.
References 16. EMA. Gardasil: EPAR – Scientific Discussion. See
1. Cancer Research UK. Cervical Cancer Mortality http://www.ema.europa.eu/docs/en_GB/document_
Statistics. See https://www.cancerresearchuk.org/health- library/EPAR_-_Scientific_Discussion/human/000703/
professional/cancer-statistics/statistics-by-cancer-type/ WC500021140.pdf (2006, last checked 28 Oct 2017).
cervical-cancer/mortality (2019, last checked 17 Apr 17. Huh WK, Joura EA, Giuliano AR, Iversen OE, de
2019). Andrade RP, Ault KA, et al. Final efficacy, immuno-
2. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre genicity, and safety analyses of a nine-valent human
LA and Jemal A. Global cancer statistics 2018: papillomavirus vaccine in women aged 16–26 years: a
GLOBOCAN estimates of incidence and mortality randomised, double-blind trial. Lancet 2017;
worldwide for 36 cancers in 185 countries. CA Cancer 390(10108): 2143–2159.
J Clin 2018; 68: 394–424. 18. Arbyn M, Xu L, Simoens C and Martin-Hirsch PP.
3. Centers for Disease Control. HPV Vaccine Prophylactic vaccination against human papilloma-
Recommendations. See https://www.cdc.gov/vaccines/ viruses to prevent cervical cancer and its precursors.
vpd/hpv/hcp/recommendations.html (2016, last Cochrane Database Syst Rev 2018; 5: CD009069.
checked 7 Oct 2017). 19. Jorgensen L, Gotzsche PC and Jefferson T. The
4. IARC. Biological Agents – A Review of Human Cochrane HPV vaccine review was incomplete and
Carcinogens – Human Papillomaviruses. Lyon: IARC, ignored important evidence of bias. BMJ Evid Based
2012. Med 2018; 23: 165–168.
5. Syrjanen K, Hakama M, Saarikoski S, Vayrynen M, 20. Harper DM, Franco EL, Wheeler C, Ferris DG, Jenkins
Yliskoski M, Syrjanen S, et al. Prevalence, incidence, D, Schuind A, et al. Efficacy of a bivalent L1 virus-like
and estimated life-time risk of cervical human papillo- particle vaccine in prevention of infection with human
mavirus infections in a nonselected finnish female papillomavirus types 16 and 18 in young women: a ran-
population. Sex Transm Dis 1990; 17: 15–19. domised controlled trial. Lancet 2004; 364: 1757–1765.
6. Rodriguez AC, Schiffman M, Herrero R, Wacholder S, 21. Harper DM, Franco EL, Wheeler CM, Moscicki AB,
Hildesheim A, Castle PE, et al. Rapid clearance of Romanowski B, Roteli-Martins CM, et al. Sustained
human papillomavirus and implications for clinical efficacy up to 4.5 years of a bivalent L1 virus-like par-
focus on persistent infections. J Natl Cancer Inst ticle vaccine against human papillomavirus types 16
2008; 100: 513–517. and 18: follow-up from a randomised control trial.
7. WHO/ICO Information Centre on HPV and Cervical Lancet 2006; 367: 1247–1255.
Cancer. HPV and cervical cancer in the 2007 report. 22. Romanowski B, de Borba PC, Naud PS, Roteli-
Vaccine 2007; 25: C1–C230. Martins CM, De Carvalho NS, Teixeira JC, et al.
8. Couto E, Saeterdal I, Juvet LK and Klemp M. HPV Sustained efficacy and immunogenicity of the human
catch-up vaccination of young women: a systematic papillomavirus (HPV)-16/18 AS04-adjuvanted vaccine:
review and meta-analysis. BMC Public Health 2014; analysis of a randomised placebo-controlled trial up to
14: 867. 6.4 years. Lancet 2009; 374: 1975–1985.
9. Di Mario S, Basevi V, Lopalco PL, Balduzzi S, 23. De Carvalho N, Teixeira J, Roteli-Martins CM, Naud
D’Amico R and Magrini N. Are the two human papil- P, De Borba P, Zahaf T, et al. Sustained efficacy and
lomavirus vaccines really similar? A systematic review immunogenicity of the HPV-16/18 AS04-adjuvanted
of available evidence: efficacy of the two vaccines vaccine up to 7.3 years in young adult women.
against HPV. J Immunol Res 2015; 2015: 435141. Vaccine 2010; 28: 6247–6255.
76 Journal of the Royal Society of Medicine 113(2)

24. Roteli-Martins CM, Naud P, De Borba P, Teixeira JC, evidence of previous exposure to HPV-16/18. Int J
De Carvalho NS, Zahaf T, et al. Sustained Cancer 2012; 131: 106–116.
immunogenicity and efficacy of the HPV-16/18 34. Apter D, Wheeler CM, Paavonen J, Castellsague X,
AS04-adjuvanted vaccine: up to 8.4 years of follow- Garland SM, Skinner SR, et al. Efficacy of human
up. Hum Vaccin Immunother 2012; 8: 390–397. papillomavirus 16 and 18 (HPV-16/18) AS04-
25. Naud PS, Roteli-Martins CM, De Carvalho NS, adjuvanted vaccine against cervical infection and pre-
Teixeira JC, de Borba PC, Sanchez N, et al. cancer in young women: final event-driven analysis of
Sustained efficacy, immunogenicity, and safety of the the randomized, double-blind PATRICIA trial. Clin
HPV-16/18 AS04-adjuvanted vaccine: final analysis of Vaccine Immunol 2015; 22: 361–373.
a long-term follow-up study up to 9.4 years post-vacci- 35. Struyf F, Colau B, Wheeler CM, Naud P, Garland S,
nation. Hum Vaccin Immunother 2014; 10: 2147–2162. Quint W, et al. Post hoc analysis of the PATRICIA
26. Konno R, Tamura S, Dobbelaere K and Yoshikawa H. randomized trial of the efficacy of human papilloma-
Efficacy of human papillomavirus 16/18 AS04- virus type 16 (HPV-16)/HPV-18 AS04-adjuvanted
adjuvanted vaccine in Japanese women aged 20 to vaccine against incident and persistent infection with
25 years: interim analysis of a phase 2 double-blind, nonvaccine oncogenic HPV types using an alternative
randomized, controlled trial. Int J Gynecol Cancer multiplex type-specific PCR assay for HPV DNA. Clin
2010; 20: 404–410. Vaccine Immunol 2015; 22: 235–244.
27. Konno R, Tamura S, Dobbelaere K and Yoshikawa H. 36. Herrero R, Wacholder S, Rodriguez AC, Solomon D,
Efficacy of human papillomavirus type 16/18 Gonzalez P, Kreimer AR, et al. Prevention of persistent
AS04-adjuvanted vaccine in Japanese women aged 20 human papillomavirus infection by an HPV16/18 vac-
to 25 years: final analysis of a phase 2 double-blind, cine: a community-based randomized clinical trial in
randomized controlled trial. Int J Gynecol Cancer 2010; Guanacaste, Costa Rica. Cancer Discov 2011; 1:
20: 847–855. 408–419.
28. Paavonen J, Jenkins D, Bosch FX, Naud P, Salmeron 37. Kreimer AR, Rodriguez AC, Hildesheim A, Herrero
R, Porras C, Schiffman M, et al. Proof-of-principle
J, Wheeler CM, et al. Efficacy of a prophylactic adju-
evaluation of the efficacy of fewer than three doses of
vanted bivalent L1 virus-like-particle vaccine against
a bivalent HPV16/18 vaccine. J Natl Cancer Inst 2011;
infection with human papillomavirus types 16 and 18
103: 1444–1451.
in young women: an interim analysis of a phase III
38. Rodriguez AC, Solomon D, Herrero R, Hildesheim A,
double-blind, randomised controlled trial. Lancet
Gonzalez P, Wacholder S, et al. Impact of human
2007; 369: 2161–2170.
papillomavirus vaccination on cervical cytology screen-
29. Paavonen J, Naud P, Salmeron J, Wheeler CM, Chow
ing, colposcopy, and treatment. Am J Epidemiol 2013;
SN, Apter D, et al. Efficacy of human papillomavirus
178: 752–760.
(HPV)-16/18 AS04-adjuvanted vaccine against cervical
39. Hildesheim A, Wacholder S, Catteau G, Struyf F,
infection and precancer caused by oncogenic HPV
Dubin G, Herrero R, et al. Efficacy of the HPV-16/
types (PATRICIA): final analysis of a double-blind, 18 vaccine: final according to protocol results from
randomised study in young women. Lancet 2009; 374: the blinded phase of the randomized Costa Rica
301–314. HPV-16/18 vaccine trial. Vaccine 2014; 32: 5087–5097.
30. Lehtinen M, Paavonen J, Wheeler CM, Jaisamrarn U, 40. Hildesheim A, Gonzalez P, Kreimer AR, Wacholder S,
Garland SM, Castellsague X, et al. Overall efficacy of Schussler J, Rodriguez AC, et al. Impact of human
HPV-16/18 AS04-adjuvanted vaccine against grade 3 papillomavirus (HPV) 16 and 18 vaccination on preva-
or greater cervical intraepithelial neoplasia: 4-year lent infections and rates of cervical lesions after exci-
end-of-study analysis of the randomised, double-blind sional treatment. Am J Obstet Gynecol 2016; 215:
PATRICIA trial. Lancet Oncol 2012; 13: 89–99. 212.e1–15.
31. Wheeler CM, Castellsague X, Garland SM, Szarewski 41. Beachler DC, Kreimer AR, Schiffman M, Herrero R,
A, Paavonen J, Naud P, et al. Cross-protective efficacy Wacholder S, Rodriguez AC, et al. Multisite HPV16/18
of HPV-16/18 AS04-adjuvanted vaccine against cer- vaccine efficacy against cervical, anal, and oral HPV
vical infection and precancer caused by non-vaccine infection. J Natl Cancer Inst 2016; 108: 1: djv302.
oncogenic HPV types: 4-year end-of-study analysis of 42. Skinner SR, Szarewski A, Romanowski B, Garland
the randomised, double-blind PATRICIA trial. Lancet SM, Lazcano-Ponce E, Salmeron J, et al. Efficacy,
Oncol 2012; 13: 100–110. safety, and immunogenicity of the human papilloma-
32. Palmroth J, Merikukka M, Paavonen J, Apter D, virus 16/18 AS04-adjuvanted vaccine in women older
Eriksson T, Natunen K, et al. Occurrence of vaccine than 25 years: 4-year interim follow-up of the phase 3,
and non-vaccine human papillomavirus types in ado- double-blind, randomised controlled VIVIANE study.
lescent Finnish females 4 years post-vaccination. Int J Lancet 2014; 384: 2213–2227.
Cancer 2012; 131: 2832–2838. 43. Wheeler CM, Skinner SR, Del Rosario-Raymundo
33. Szarewski A, Poppe WAJ, Skinner SR, Wheeler CM, MR, Garland SM, Chatterjee A, Lazcano-Ponce E,
Paavonen J, Naud P, et al. Efficacy of the human papil- et al. Efficacy, safety, and immunogenicity of the
lomavirus (HPV)-16/18 AS04-adjuvanted vaccine in human papillomavirus 16/18 AS04-adjuvanted vaccine
women aged 15–25 years with and without serological in women older than 25 years: 7-year follow-up of the
Rees et al. 77

phase 3, double-blind, randomised controlled 55. Tainio K, Athanasiou A, Tikkinen KAO, Aaltonen R,
VIVIANE study. Lancet Infect Dis 2016; 16: Cardenas J, Hernandes, et al. Clinical course of
1154–1168. untreated cervical intraepithelial neoplasia grade 2
44. Zhu F-C, Chen W, Hu Y-M, Hong Y, Li J, Zhang X, under active surveillance: systematic review and meta-
et al. Efficacy, immunogenicity and safety of the analysis. BMJ 2018; 360: k499.
HPV-16/18 AS04-adjuvanted vaccine in healthy 56. Munoz N, Kjaer SK, Sigurdsson K, Iversen OE,
Chinese women aged 18–25 years: results from a ran- Hernandez-Avila M, Wheeler CM, et al. Impact of
domized controlled trial. Int J Cancer 2014; 135: human papillomavirus (HPV)-6/11/16/18 vaccine on
2612–2622. all HPV-associated genital diseases in young women.
45. Zhu FC, Hu SY, Hong Y, Hu YM, Zhang X, Zhang J Natl Cancer Inst 2010; 102: 325–339.
YJ, et al. Efficacy, immunogenicity, and safety of the 57. Pagliusi SR and Aguado MT. Efficacy and other mile-
HPV-16/18 AS04-adjuvanted vaccine in Chinese stones for human papillomavirus vaccine introduction.
women aged 18–25 years: event-triggered analysis of Vaccine 2004; 23: 569–578.
a randomized controlled trial. Cancer Med 2017; 6: 58. Smith JS, Lindsay L, Hoots B, Keys J, Franceschi S,
12–25. Winer R, et al. Human papillomavirus type distribu-
46. Villa LL, Costa RLR, Petta CA, Andrade RP, Ault tion in invasive cervical cancer and high-grade cervical
KA, Giuliano AR, et al. Prophylactic quadrivalent lesions: a meta-analysis update. Int J Cancer 2007; 121:
human papillomavirus (types 6, 11, 16, and 18) L1 621–632.
virus-like particle vaccine in young women: a rando- 59. Kjaer SK, Frederiksen K, Munk C and Iftner T. Long-
mised double-blind placebo-controlled multicentre term absolute risk of cervical intraepithelial neoplasia
phase II efficacy trial. Lancet Oncol 2005; 6: 271–278. grade 3 or worse following human papillomavirus
47. Villa LL, Costa RL, Petta CA, Andrade RP, Paavonen infection: role of persistence. J Natl Cancer Inst 2010;
J, Iversen OE, et al. High sustained efficacy of a 102: 1478–1488.
prophylactic quadrivalent human papillomavirus 60. Harper DM and DeMars LR. HPV vaccines –
types 6/11/16/18 L1 virus-like particle vaccine through a review of the first decade. Gynecol Oncol 2017; 146:
5 years of follow-up. Br J Cancer 2006; 95: 1459–1466. 196–204.
48. Garland SM, Hernandez-Avila M, Wheeler CM, Perez 61. Future II Study Group. Quadrivalent vaccine against
G, Harper DM, Leodolter S, et al. Quadrivalent vac- human papillomavirus to prevent high-grade cervical
cine against human papillomavirus to prevent anogeni- lesions. N Engl J Med 2007; 356: 1915–1927.
tal diseases. N Engl J Med 2007; 356: 1928–1943. 62. Sow PS, Watson-Jones D, Kiviat N, Changalucha J,
49. Future II Study Group. Prophylactic efficacy of a Mbaye KD, Brown J, et al. Safety and immunogenicity
quadrivalent human papillomavirus (HPV) vaccine in of human papillomavirus-16/18 AS04-adjuvanted
women with virological evidence of HPV infection. J vaccine: a randomized trial in 10–25-year-old HIV-
Infect Dis 2007; 196: 1438–1446. seronegative African girls and young women. J Infect
50. Munoz N, Manalastas R Jr, Pitisuttithum P, Dis 2013; 207: 1753–1763.
Tresukosol D, Monsonego J, Ault K, et al. Safety, 63. Brown DR, Kjaer SK, Sigurdsson K, Iversen OE,
immunogenicity, and efficacy of quadrivalent human Hernandez-Avila M, Wheeler CM, et al. The impact
papillomavirus (types 6, 11, 16, 18) recombinant vac- of quadrivalent human papillomavirus (HPV; types 6,
cine in women aged 24–45 years: a randomised, double- 11, 16, and 18) L1 virus-like particle vaccine on infec-
blind trial. Lancet 2009; 373: 1949–1957. tion and disease due to oncogenic nonvaccine HPV
51. Castellsague X, Munoz N, Pitisuttithum P, Ferris D, types in generally HPV-naive women aged 16–26
Monsonego J, Ault K, et al. End-of-study safety, years. J Infect Dis 2009; 199: 926–935.
immunogenicity, and efficacy of quadrivalent HPV 64. Weinberger DM, Malley R and Lipsitch M. Serotype
(types 6, 11, 16, 18) recombinant vaccine in adult replacement in disease after pneumococcal vaccination.
women 24–45 years of age. Br J Cancer 2011; 105: Lancet 2011; 378: 1962–1973.
28–37. 65. Public Health England. Cervical Screening: Programme
52. Yoshikawa H, Ebihara K, Tanaka Y and Noda K. Overview. See https://www.gov.uk/guidance/cervical-
Efficacy of quadrivalent human papillomavirus (types screening-programme-overview (2015, last checked 28
6, 11, 16 and 18) vaccine (GARDASIL) in Japanese Oct 2017).
women aged 18–26 years. Cancer Sci 2013; 104: 66. Force USPST. U.S. Preventive Services Task
465–472. Force Cervical Cancer Screening. Archived Final
53. Wei L, Xie X, Liu J, Zhao Y, Chen W, Zhao C, et al. Recommendation. See https://www.uspreventiveservi-
Efficacy of quadrivalent human papillomavirus vaccine cestaskforce.org/Page/Document/Recommendation
against persistent infection and genital disease in StatementFinal/cervical-cancer-screening#consider
Chinese women: a randomized, placebo-controlled (2013, last checked 26 May 2019).
trial with 78-month follow-up. Vaccine 2019; 37(27): 67. Eklund C, Forslund O, Wallin KL, Zhou T and Dillner
3617–3624. J Network WHOHPL. The 2010 global proficiency
54. Yudkin JS, Lipska KJ and Montori VM. The idolatry study of human papillomavirus genotyping in vaccinol-
of the surrogate. BMJ 2011; 343: d7995. ogy. J Clin Microbiol 2012; 50: 2289–2298.
78 Journal of the Royal Society of Medicine 113(2)

68. Dillner J. The serological response to papillomaviruses. 72. National Statistics. Scottish Cervical Screening
Semin Cancer Biol 1999; 9: 423–430. Programme Statistics 2017/18. See https://www.isds-
69. Pollock KG, Kavanagh K, Potts A, Love J, Cuschieri cotland.org/Health-Topics/Cancer/Publications/2018-
K, Cubie H, et al. Reduction of low- and high-grade 09-04/2018-09-04-Cervical-Screening-Report.pdf
cervical abnormalities associated with high uptake of (2018, last checked 22 Apr 2019).
the HPV bivalent vaccine in Scotland. Br J Cancer 73. Wilyman J. HPV vaccination programs have not been
2014; 111: 1824–1830. shown to be cost-effective in countries with comprehen-
70. Palmer T, Wallace L, Pollock KG, Cuschieri K, sive Pap screening and surgery. Infect Agent Cancer
Robertson C, Kavanagh K, et al. Prevalence of cervical 2013; 8: 21.
disease at age 20 after immunisation with bivalent HPV 74. Public Health England. Cervical Screening:
vaccine at age 12–13 in Scotland: retrospective popula- Implementation Guide for Primary HPV screening. See
tion study. BMJ 2019; 365: l1161. https://www.gov.uk/government/publications/cervical-
71. Neville FG, McEachran J, Aleman-Diaz A, Whitehead screening-primary-hpv-screening-implementation/cer-
R, Cosma A, Currie D, et al. Trends in the sexual vical-screening-implementation-guide-for-primary-
behaviour of 15-year olds in Scotland: 2002–14. Eur J hpv-screening (2019, last checked 29 Mar 2019).
Public Health 2017; 27: 835–839.