The
new england journal
of
medicine
n engl j med 372;8 nejm.org February 19, 2015
711
The authors’ full names, academic de-
grees, and affiliations are listed in the Ap-
pendix. Address reprint requests to Dr.
Joura at the Department of Gynecology
and Obstetrics, Medical University of Vi-
enna, Währinger Gürtel 18-20, Vienna,
Austria, or at elmar . joura@ meduniwien
. ac . at.
*A complete list of the investigators for
the Broad Spectrum HPV Vaccine Study
is provided in the Supplementary Ap-
pendix, available at NEJM.org.
N Engl J Med 2015;372:711-23.
DOI: 10.1056/NEJMoa1405044
Copyright © 2015 Massachusetts Medical Society.
BACKGROUND
The investigational 9-valent viruslike particle vaccine against human papillomavirus
(HPV) includes the HPV types in the quadrivalent HPV (qHPV) vaccine (6, 11, 16,
and 18) and five additional oncogenic types (31, 33, 45, 52, and 58). Here we present
the results of a study of the efficacy and immunogenicity of the 9vHPV vaccine in
women 16 to 26 years of age.
METHODS
We performed a randomized, international, double-blind, phase 2b–3 study of the
9vHPV vaccine in 14,215 women. Participants received the 9vHPV vaccine or the
qHPV vaccine in a series of three intramuscular injections on day 1 and at months
2 and 6. Serum was collected for analysis of antibody responses. Swabs of labial,
vulvar, perineal, perianal, endocervical, and ectocervical tissue were obtained and
used for HPV DNA testing, and liquid-based cytologic testing (Papanicolaou testing)
was performed regularly. Tissue obtained by means of biopsy or as part of definitive
therapy (including a loop electrosurgical excision procedure and conization) was
tested for HPV.
RESULTS
The rate of high-grade cervical, vulvar, or vaginal disease irrespective of HPV type
(i.e., disease caused by HPV types included in the 9vHPV vaccine and those not
included) in the modified intention-to-treat population (which included participants
with and those without prevalent infection or disease) was 14.0 per 1000 person-years
in both vaccine groups. The rate of high-grade cervical, vulvar, or vaginal disease re-
lated to HPV-31, 33, 45, 52, and 58 in a prespecified per-protocol efficacy popula-
tion (susceptible population) was 0.1 per 1000 person-years in the 9vHPV group
and 1.6 per 1000 person-years in the qHPV group (efficacy of the 9vHPV vaccine,
96.7%; 95% confidence interval, 80.9 to 99.8). Antibody responses to HPV-6, 11, 16,
and 18 were noninferior to those generated by the qHPV vaccine. Adverse events re-
lated to injection site were more common in the 9vHPV group than in the qHPV group.
CONCLUSIONS
The 9vHPV vaccine prevented infection and disease related to HPV-31, 33, 45, 52,
and 58 in a susceptible population and generated an antibody response to HPV-6,
11, 16, and 18 that was noninferior to that generated by the qHPV vaccine. The
9vHPV vaccine did not prevent infection and disease related to HPV types beyond
the nine types covered by the vaccine. (Funded by Merck; ClinicalTrials.gov num-
ber, NCT00543543).
ABSTRACT
A 9-Valent HPV Vaccine against Infection
and Intraepithelial Neoplasia in Women
E.A. Joura, A.R. Giuliano, O.-E. Iversen, C. Bouchard, C. Mao, J. Mehlsen,
E.D. Moreira, Jr., Y. Ngan, L.K. Petersen, E. Lazcano-Ponce, P. Pitisuttithum,
J.A. Restrepo, G. Stuart, L. Woelber, Y.C. Yang, J. Cuzick, S.M. Garland, W. Huh,
S.K. Kjaer, O.M. Bautista, I.S.F. Chan, J. Chen, R. Gesser, E. Moeller, M. Ritter,
S. Vuocolo, and A. Luxembourg, for the Broad Spectrum HPV Vaccine Study*
Original Article
The New England Journal of Medicine
Downloaded from nejm.org at FIOCRUZ on March 21, 2016. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
n engl j med 372;8 nejm.org February 19, 2015
712
The
new england journal
of
medicine
T
he human papillomavirus (HPV)
causes premalignant and malignant lesions
of the cervix,
1,2
vagina,
3,4
vulva,
4,5
anus,
4,6
penis,
7
and oropharynx,
8
as well as genital warts.
9,10
The recent development of prophylactic vaccines
directed against the most relevant disease-causing
HPV types has helped to prevent diseases related
to HPV.
11
In clinical trials, the bivalent HPV viruslike
particle vaccine against HPV types 16 and 18 was
efficacious against related infection with these
types and against cervical dysplasia,
12
and the
quadrivalent HPV viruslike particle vaccine against
types 6, 11, 16, and 18 was efficacious against
related infection and against cervical, vaginal,
vulvar, and anal dysplasia and against condyloma
related to HPV-6 and 11.
13-17
Post-licensure reports
from countries with established HPV vaccination
programs indicate that HPV vaccination has a ben-
eficial effect at the population level as early as
3 years after the introduction of an HPV vacci-
nation program, including decreases in the in-
cidence of high-grade cervical abnormalities,
18-21
the prevalence of vaccine HPV types,
22-24
and the
incidence of genital warts.
25-29
Analyses of clini-
cal trial and post-licensure safety data have not
identified safety concerns associated with HPV
vaccination.
30-37
Current HPV vaccines address approximately
70% of cervical cancers through protection from
HPV-16 and HPV-18. Partial cross-protection
against nonvaccine HPV types has been reported
for both licensed vaccines, although the clinical
significance of the partial cross-protection re-
mains uncertain.
38
The investigational 9-valent
HPV viruslike particle (9vHPV) vaccine addresses
the four HPV types (6, 11, 16, 18) that are in the
quadrivalent HPV vaccine plus five additional on-
cogenic types (31, 33, 45, 52, and 58). The 9vHPV
vaccine offers the potential to increase overall
prevention of cervical cancer from approximately
70% to approximately 90%.
2,39,40
Here we report
the findings of an efficacy and immunogenicity
study of the 9vHPV vaccine.
Methods
Study Design
We conducted a randomized, international, multi-
center, double-blind study of the immunogenicity,
efficacy, and side-effect profile of the 9vHPV vac-
cine in women 16 to 26 years of age. The study
was based on a phase 2–3 adaptive design (see
the Supplementary Appendix, available with the
full text of this article at NEJM.org). An initial
group of 1242 women were randomly assigned
to receive one of three doses of the 9vHPV vac-
cine or a qHPV vaccine control. A larger group
of 13,598 women were then randomly assigned
to receive either the 9vHPV vaccine at the dose
selected on the basis of results in the initial
group or the qHPV vaccine control. The efficacy
study included these 13,598 women together with
the 307 women in the initial group assigned to
receive the 9vHPV vaccine at the dose selected
and the 310 women in the initial group assigned
to the qHPV vaccine, representing a total of
14,215 women (Fig. S1 and S2 in the Supplemen-
tary Appendix).
Since HPV vaccination is widely recommend-
ed and has been shown to prevent HPV disease
related to oncogenic HPV types, the use of a
placebo was not considered to be acceptable for
ethical reasons. Consequently, the study used the
qHPV vaccine as an active comparator. Participants
were eligible if they had no history of an abnor-
mal result on a Papanicolaou (Pap) test, no more
than four lifetime sexual partners, and no previ-
ous abnormal finding on cervical biopsy.
Study Oversight
The study was conducted in accordance with
principles of Good Clinical Practice and was ap-
proved by the institutional review board at each
participating institution and by regulatory agen-
cies. Written informed consent was provided by
all adult participants and by a parent or legal
guardian of participants who were minors. A
scientific advisory committee comprising both aca-
demic and sponsor (Merck) investigators developed
the protocol (available at NEJM.org), formulated
the statistical analysis plan, analyzed and inter-
preted the data, and wrote the manuscript. The
external data and safety monitoring committee
whose members were aware of the group assign-
ments assessed safety findings throughout the
study. All the authors vouch for the complete-
ness and accuracy of the data and analyses pre-
sented.
Vaccine Dosing
A 0.5-ml dose of qHPV vaccine contains 20 μg
of HPV-6, 40 μg of HPV-11, 40 μg of HPV-16,
and 20 μg of HPV-18 viruslike particles, and
The New England Journal of Medicine
Downloaded from nejm.org at FIOCRUZ on March 21, 2016. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
n engl j med 372;8 nejm.org February 19, 2015
713
A 9-Valent HPV Vaccine
Characteristic
9vHPV Vaccine
(N = 7106)
qHPV Vaccine
(N = 7109)
Total
(N = 14,215)
Age — yr
Mean 21.9±2.5 21.8±2.5 21.9±2.5
Median 22.0 22.0 22.0
Range 16–26 16–26 16–26
Age at first sexual intercourse — yr 17.4±2.2 17.4±2.2 17.4±2.2
Region — no. (%)
Asia–Pacific 905 (12.7) 909 (12.8) 1,814 (12.8)
Europe 2406 (33.9) 2409 (33.9) 4,815 (33.9)
Latin America 2372 (33.4) 2372 (33.4) 4,744 (33.4)
North America 1423 (20.0) 1419 (20.0) 2,842 (20.0)
Smoking status — no. (%)
Current smoker 1071 (15.1) 1005 (14.1) 2,076 (14.6)
Former smoker 382 (5.4) 358 (5.0) 740 (5.2)
Never smoked 5647 (79.5) 5744 (80.8) 11,391 (80.1)
Unknown 6 (0.1) 2 (0) 8 (0.1)
Lifetime sexual partners — no. (%)†
1 2063 (29.0) 2023 (28.5) 4,086 (28.8)
2 1691 (23.8) 1698 (23.9) 3,389 (23.8)
3 1648 (23.2) 1646 (23.2) 3,294 (23.2)
4 1520 (21.4) 1527 (21.5) 3,047 (21.4)
>4 11 (0.2) 15 (0.2) 26 (0.2)
Non-HPV–related cervicovaginal infections
or sexually transmitted diseases
Any 298 (4.2) 292 (4.1) 590 (4.2)
Chlamydia 284 (4.0) 285 (4.0) 569 (4.0)
Gonorrhea 19 (0.3) 11 (0.2) 30 (0.2)
Contraceptive use‡
Barrier 2318 (32.6) 2303 (32.4) 4,621 (32.5)
Behavior 1014 (14.3) 1035 (14.6) 2,049 (14.4)
Hormonal 4273 (60.2) 4292 (60.4) 8,565 (60.3)
Day 1 composite HPV positivity — no./
total no. (%)§
Serologic test 2771/7082 (39.1) 2647/7078 (37.4) 5418/14,160 (38.3)
PCR assay 1887/6919 (27.3) 1920/6943 (27.7) 3807/13,862 (27.5)
Serologic test or PCR assay 3365/6970 (48.3) 3345/6983 (47.9) 6710/13,953 (48.1)
* Plus–minus values are means ±SD. The baseline characteristics of the two study groups were similar. The quadrivalent
human papillomavirus (HPV) vaccine (qHPV) targets HPV types 6, 11, 16, and 18; the 9-valent viruslike particle vaccine
(9vHPV) targets the HPV types in the qHPV vaccine and five additional oncogenic types (31, 33, 45, 52, and 58). PCR
denotes polymerase chain reaction.
† The percentages for the number of lifetime sexual partners were calculated on the basis of the number of participants
for whom there were data on sexual history at enrollment (7102 in the 9vHPV group and 7108 in the qHPV group).
‡ Participants may have used more than one contraceptive method. A participant is counted once within a category and
may be counted in more than one category. The percentages for the numbers of participants who used contraceptives
were based on the number for whom this information was available (7102 in the 9vHPV group and 7104 in the qHPV
group).
§ Positivity was defined as an anti-HPV titer on immunoassay of at least 30, 16, 20, 24, 10, 8, 8, 8, and 8 for HPV types 6,
11, 16, 18, 31, 33, 45, 52, and 58, respectively. The numerator in this category represents the number of HPV-positive
participants, and the denominator the total number of participants with assay results that could be evaluated.
Table 1. Baseline Characteristics of the Participants.*
The New England Journal of Medicine
Downloaded from nejm.org at FIOCRUZ on March 21, 2016. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
n engl j med 372;8 nejm.org February 19, 2015
714
The
new england journal
of
medicine
End Point
9vHPV Vaccine
(N = 7099)
qHPV Vaccine
(N = 7105)
Risk Reduction
(95% CI)
no./total no.
cases/1000
person-yr no./total no.
cases/1000
person-yr
Modified intention-to-treat population
High-grade cervical, vulvar, and vaginal disease†
All participants 340/7027 14.0 344/7027 14.0 0.7 (−15.7 to 14.8)
HPV-uninfected on day 1 26/3032 2.4 46/3077 4.2 42.5 (7.9 to 65.9)
Not related to 9 vaccine HPV types‡ 26/3032 2.4 33/3077 3.0 19.7 (−34.5 to 52.5)
Related to 9 vaccine HPV types‡ 0/3032 0.0 13/3076 1.2 100 (70.4 to 100)
HPV-infected on day 1 314/3995 23.1 298/3950 22.1 −4.8 (−23.3 to 10.8)
Not related to 9 vaccine HPV types‡ 141/3995 10.0 137/3950 9.8 −2.0 (−30.0 to 19.9)
Related to 9 vaccine HPV types‡ 173/3992 12.4 161/3946 11.6 −6.8 (−33.2 to 14.3)
Average risk reduction§ — — — — 19.0 (−1.6 to 35.3)
High-grade cervical epithelial neoplasia, adenocarcino-
ma in situ, and cervical cancer
All participants 325/6882 14.1 326/6871 14.1 −0.3 (−17.3 to 14.3)
HPV-uninfected on day 1 26/2976 2.5 44/3009 4.2 39.7 (1.8 to 64.3)
Not related to 9 vaccine HPV types‡ 26/2976 2.5 31/3009 3.0 14.3 (−49.1 to 49.1)
Related to 9 vaccine HPV types‡ 0/2976 0.0 13/3009 1.2 100 (70.3 to 100)
HPV-infected on day 1 299/3906 23.3 282/3862 22.2 −5.3 (−24.1 to 10.8)
Not related to 9 vaccine HPV types‡ 131/3906 10.1 132/3862 10.3 1.8 (−26.0 to 23.5)
Related to 9 vaccine HPV types‡ 168/3906 13.0 150/3862 11.7 −11.3 (−39.6 to 11.0)
Average risk reduction§ — — — — 17.1 (−4.2 to 34.0)
Per-protocol efficacy population
High-grade cervical, vulvar, and vaginal disease†
Related to HPV-31, 33, 45, 52, or 58 1/6016 0.1 30/6,017 1.6 96.7 (80.9 to 99.8)
Related to HPV-6, 11, 16, or 18 1/5883 0.1 3/5898 0.2 66.6 (−203.0 to 98.7)
High-grade cervical epithelial neoplasia, adenocarcino-
ma in situ, and cervical cancer
Related to HPV-31, 33, 45, 52, or 58 1/5948 0.1 27/5943 1.5 96.3 (79.5 to 99.8)
Related to HPV-6, 11, 16, or 18 1/5823 0.1 1/5832 0.1 −0.4 (≤ −999 to 97.4)
Persistent infection ≥6 months’ duration¶
Related to HPV-31, 33, 45, 52, or 58 35/5939 2.1 810/5953 52.4 96.0 (94.4 to 97.2)
Related to HPV-6, 11, 16, or 18 59/5812 3.6 80/5830 5.0 26.4 (−4.3 to 47.5)
* The total number of participants (N) includes those who received at least one dose of a study vaccine; the no./total no. refers to the num-
ber of participants with an end point among the participants who received at least one dose of a study vaccine and had at least one follow-
up visit. The modified intention-to-treat population consisted of participants who received at least one dose of vaccine and for whom there
was at least one measurement of efficacy for the end point being analyzed. The per-protocol efficacy population consisted of participants
who received all three doses of vaccine within 1 year, were HPV-uninfected (i.e., were seronegative at day 1 and had negative results on PCR
assays for all HPV types tested from day 1 through month 7) to the vaccine HPV type being analyzed, and had no protocol violations. CI de-
notes confidence interval.
† This category includes high-grade cervical epithelial neoplasia, adenocarcinoma in situ, cervical cancer, high-grade vulvar intraepithelial neo-
plasia, high-grade vaginal intraepithelial neoplasia, vulvar cancer, and vaginal cancer.
‡ The nine vaccine HPV types are 6, 11, 16, 18, 31, 33, 45, 52, and 58. Participants with end-point conditions related to the nine vaccine HPV
types were those who at any time during the study received a diagnosis of the indicated disease related to any of the nine HPV types includ-
ed in the vaccine and, in addition, did not receive a diagnosis of the indicated disease at any time during the study that was not related to
one of the nine HPV types included in the vaccine. Participants with end-point conditions were counted only once and in only one of these
two categories. The sum of case counts in these two categories is equal to the number of end-point cases irrespective of HPV type.
§ The risk reduction shown represents the sample-size–weighted average of the risk reduction in the HPV-uninfected and the HPV-infected
subgroups. The HPV-uninfected subgroup consisted of participants who had the following characteristics on day 1: a negative finding for
squamous intraepithelial lesions, a seronegative finding for the nine vaccine-HPV types, and negative results on PCR assays for all HPV
types tested during the study (types 6, 11, 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59). The HPV-infected subgroup consisted of all par-
ticipants who were not in the HPV-infected subgroup.
¶ Persistent infection was defined as detection of the same HPV type in a genital swab or tissue specimen collected on two or more consecu-
tive visits, with an interval of at least 6 months (±1 month) between the visits.
Table 2. Effect of 9vHPV Vaccine on the Incidence of Cervical, Vulvar, and Vaginal Disease and of Persistent HPV-Related Infection.*
The New England Journal of Medicine
Downloaded from nejm.org at FIOCRUZ on March 21, 2016. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
n engl j med 372;8 nejm.org February 19, 2015
715
A 9-Valent HPV Vaccine
225 μg of the adjuvant amorphous aluminum
hydroxyphosphate sulfate (AAHS).
14
A 0.5-ml
dose of 9vHPV vaccine contains 30 μg of HPV-6,
40 μg of HPV-11, 60 μg of HPV-16, 40 μg of
HPV-18, 20 μg of HPV-31, 20 μg of HPV-33, 20
μg of HPV-45, 20 μg of HPV-52, and 20 μg of
HPV-58 viruslike particles, and 500 μg of AAHS.
Vaccines were administered as a 0.5-ml intramus-
cular injection in three doses, on day 1 and at
month 2 and month 6. Information on random-
ization to a vaccine group is available in the
Supplementary Appendix. At study vaccination
visits, all participants received a vaccination re-
port card on which they recorded oral tempera-
tures on each of the 5 days after vaccination and
adverse events related to the injection site as well
as systemic adverse events on each of the 15 days
after vaccination.
Follow-up
Swabs of labial, vulvar, perineal, perianal, endo-
cervical, and ectocervical tissue and Pap test
(ThinPrep; Hologic) samples were collected on
day 1 and at months 7, 12, 18, 24, 30, 36, 42, 48,
and 54. Swabs were tested for HPV types 6, 11,
16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59
by means of a polymerase-chain-reaction (PCR)
assay to identify participants who had an active
HPV infection at enrollment and to determine
end points for HPV infection. Participants with
an abnormal result on a Pap test were referred for
colposcopy according to a protocol-mandated tri-
age algorithm (Fig. S3 in the Supplementary Ap-
pendix). Histologic sections were first reviewed
for clinical management by pathologists at a cen-
tral laboratory (Diagnostic Cytology Laboratories)
who were unaware of the patient’s treatment-
group assignment and HPV status and were then
reviewed for the determination of end points by
an adjudication panel of four pathologists who
were unaware of the vaccination assignment.
Primary Hypotheses and End Points
The primary efficacy hypothesis was that, as
compared with the qHPV vaccine, the 9vHPV vac-
cine would reduce the combined incidence of
several conditions related to HPV-31, 33, 45, 52,
and 58 in women 16 to 26 years of age who were
seronegative on day 1 and for whom there were
negative results on PCR assays for the relevant
HPV type from day 1 through month 7. These
conditions (the study end points) were high-grade
cervical intraepithelial neoplasia, adenocarcinoma
in situ, invasive cervical carcinoma, high-grade
vulvar intraepithelial neoplasia, high-grade vaginal
intraepithelial neoplasia, vulvar cancer, and vag-
inal cancer. Determination of the end points was
based on a consensus diagnosis, by a panel of at
least two pathologists, of high-grade cervical,
vulvar, or vaginal disease and the detection of
HPV-31, 33, 45, 52, or 58 DNA in an adjacent his-
tologic section of the same biopsy specimen.
14
It was anticipated that the 9vHPV vaccine and
the qHPV vaccine would be similarly efficacious
in preventing disease related to HPV-6, 11, 16,
and 18. Thus, it was determined that a direct
comparison of the two vaccines on the basis of
disease end points related to HPV-6, 11, 16, and
18 would be prohibitive in terms of study size.
The primary analysis of 9vHPV vaccine efficacy
with respect to end points related to HPV-6, 11,
16, and 18 was designed to determine whether
9vHPV was noninferior to qHPV with respect to
immunogenicity. The primary immunogenicity
hypothesis was that geometric mean titers for
anti-HPV-6, 11, 16, and 18 in the 9vHPV vaccine
group would be noninferior to the geometric mean
titers in the qHPV vaccine group (as quantified
with the use of a competitive Luminex immuno-
assay
41
).
Statistical Analysis
The primary efficacy hypothesis was evaluated
in the per-protocol efficacy population, which
consisted of participants who received all three
doses of vaccine within 1 year, did not have the
HPV type being analyzed (i.e., were seronegative
on day 1 and PCR-negative from day 1 through
month 7), and had no protocol violations. Success-
ful demonstration of the primary efficacy hypoth-
esis required the lower boundary of the two-
sided 95% confidence interval of vaccine efficacy
to be greater than 25%, where vaccine efficacy,
or percent risk reduction, was calculated as
100 × (1 − incidence rate of types in 9vHPV ÷ inci-
dence rate of types in qHPV). The 95% confi-
dence interval of vaccine efficacy was calculated
with the use of a binomial, distribution-based
exact method.
42
With further assumptions relat-
ing to end-point accrual rate, losses to follow-up,
and exclusion rate from the per-protocol efficacy
population, we estimated that enrollment of ap-
proximately 7000 participants per group would be
needed for the phase 3 efficacy study.
The New England Journal of Medicine
Downloaded from nejm.org at FIOCRUZ on March 21, 2016. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
