Sie sind auf Seite 1von 7

Vaccine 32 (2014) 733739

Contents lists available at ScienceDirect

Vaccine
journal homepage: www.elsevier.com/locate/vaccine

Estimation of the potential overall impact of human papillomavirus


vaccination on cervical cancer cases and deaths
Georges Van Kriekinge a, , Xavier Castellsagu b , David Cibula c , Nadia Demarteau a
a
b
c

Health Economics, GlaxoSmithKline Vaccines, Avenue Fleming, 20, Wavre, 1300, Belgium
Cancer Epidemiology Research Program, Institut Catal dOncologia (ICO)-IDIBELL, CIBERESP, LHospitalet de Llobregat, Catalonia, Spain
Gynecologic Oncology Centre, General University Hospital of the First Medical School, Charles University, Prague, Czech Republic

a r t i c l e

i n f o

Article history:
Received 16 May 2013
Received in revised form 30 October 2013
Accepted 17 November 2013
Available online 26 November 2013
Keywords:
Vaccination
Human papillomavirus
Cervical cancer

a b s t r a c t
Background: Human papillomavirus (HPV) vaccination offers potential for primary prevention of HPVrelated pre-cancers and cancers as demonstrated in clinical trials. Mathematical models have estimated
the potential real-life impact of vaccination on the burden of cervical cancer (CC). However, these are
restricted to evaluations in a limited number of countries.
Methods: Potential decline in CC cases and deaths with the AS04-adjuvanted HPV-16/18 vaccine of young
girls nave to HPV, was estimated at steady-state (vaccine coverage: 0100%) based on clinical trial and
country-specic incidence data. Data on vaccine efcacy were taken from the end of study PATRICIA trial
of the AS04-adjuvanted HPV-16/18 vaccine. The numbers of cases and deaths due to HPV-16/18 were
estimated and compared with those due to any HPV type to estimate the additional cases prevented. This
difference estimates CC cases and deaths avoided due to protection against non-vaccine HPV types. Costoffsets due to reductions in CC treatment were estimated for ve countries (Brazil, Canada, Italy, Malaysia
and South African Republic) using country-specic unit cost data. Additionally, cervical intraepithelial
neoplasia grade 2 or 3 (CIN2/3)-related burden (cases and treatment costs) prevented by vaccination
were estimated for two countries (Italy and Malaysia).
Results: HPV vaccination could prevent a substantial number of CC cases and deaths in countries worldwide, with associated cost-offsets due to reduced CC treatment. Cross-protection increased the estimated
potential number of CC cases and deaths prevented by 34 and 18% in Africa and Oceania, respectively.
Moreover, vaccination could result in a substantial reduction in the number of CIN2/3 lesions and associated costs.
Conclusion: HPV vaccination could reduce the burden of CC and precancerous lesions in countries worldwide, part of disease burden reduction being related to protection against non HPV-16/18 related types.
2013 The Authors. Published by Elsevier Ltd. All rights reserved.

1. Introduction
Cervical cancer (CC) is the third most common cancer in women,
with an estimated 530,000 new cases worldwide in 2008 [1].
Despite screening, the burden of CC remains high, with 275,000

Abbreviations: CC, cervical cancer; CIN, cervical intraepithelial neoplasia; HPV,


human papillomavirus; TVC, total vaccinated cohort; HSIL, high-grade squamous
intraepithelial lesion; ICO, Institut Catal dOncologia; VE, vaccine efcacy; WHO,
World Health Organization; LEEP, loop electrosurgical excision procedure; LLETZ,
large loop excision of the transformation zone.
This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-No Derivative Works License, which permits
non-commercial use, distribution, and reproduction in any medium, provided the
original author and source are credited.
Corresponding author. Tel.: +32 10 85 41 98; fax: +32 10 85 3367.
E-mail address: georges.m.van-kriekinge@gsk.com (G. Van Kriekinge).

deaths estimated for 2008 [1]. The burden of CC varies considerably


between countries, with 85% of cases and 88% of deaths occurring
in developing nations [1,2]. Human papillomavirus (HPV) is established as a necessary cause of CC, with HPV identied in 99.7% of
CC cases worldwide [3]. The two HPV types most commonly associated with CC are HPV types 16 and 18. However, approximately a
quarter of CC cases are associated with other oncogenic HPV types
and therefore, impose a substantial health burden [4].
Precancerous lesions also known as cervical intraepithelial
neoplasia (CIN) impose a health burden beyond that of CC itself, particularly in countries with well-established screening programmes
where CIN lesions are more likely to be detected [5]. High-grade
cervical intraepithelial neoplasia (CIN2/3), when diagnosed, results
in conisation or surgical excision to remove the lesion, as per consensus guidelines for management of CIN2/3 [6].
Vaccination against oncogenic HPV infection offers the potential for primary prevention of precancerous lesions and CC. Two

0264-410X/$ see front matter 2013 The Authors. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.vaccine.2013.11.049

734

G. Van Kriekinge et al. / Vaccine 32 (2014) 733739

HPV vaccines are currently available: a HPV-6/11/16/18 vaccine


(Gardasil , Merck/Sano-Pasteur) and a HPV-16/18 vaccine with
Adjuvant System 04 (AS04) (Cervarix , GlaxoSmithKline Vaccines).
The PApilloma TRIal against Cancer In young Adults (PATRICIA)
is the largest trial conducted so far with a licenced HPV vaccine. This trial assessing the HPV-16/18 AS04-adjuvanted vaccine
enrolled 18,729 healthy women aged 1525 years irrespective of
their baseline HPV DNA status [7,8]. Data from the end-of-study
analysis of the PATRICIA trial showed that the AS04-adjuvanted
HPV-16/18 vaccine demonstrated 100% efcacy against CIN3+
lesions associated with HPV 16/18 and further had an overall vaccine efcacy (VE) of 93.2% against CIN3+ lesions irrespective of
HPV type in the HPV-nave1 total vaccinated cohort (TVC) after a
follow-up time up to 48 months [9]. These results demonstrated
that protection against non-vaccine HPV types is present, with
or without co-infection with HPV 16/18 [9,10]. These ndings
suggest that this vaccine could offer important health benets in reducing precancerous lesions and CC cases beyond that
expected from the prevention of lesions caused by HPV types-16/18
alone.
The objective of the present study was to estimate the potential
real life impact of the AS04-adjuvanted HPV-16/18 vaccine on CC
cases and deaths at country level in all WHO reported countries differentiating number of cases potentially prevented irrespective of
the causative HPV type as well as cases prevented causally related to
HPV-16/18. These number of cases and deaths were subsequently
grouped by continent. Additionally, potential reduction in the treatment costs of CC in ve countries located in ve different regions
and the potential effect of vaccination on the burden and cost of
precancerous lesions in two other countries (one from Europe and
one from Asia) was evaluated.

2.1.2. Country specic CC cases and deaths prevented causally


related to HPV-16/18
The number of cases prevented for each country that could be
attributed to protection against HPV-16/18 alone was estimated by
multiplying the annual number of CC cases and deaths by vaccine
coverage and the expected vaccine effectiveness against HPV-16/18
related-CC cases and deaths. The HPV-16/18 related effectiveness
was estimated using country-specic data of the proportion of CC
cases attributable to HPV-16/18 multiplied by the reported vaccine
efcacy against HPV-16/18-related CC.
Vaccine efcacy of 100% against HPV-16/18-related CC was
used based on the AS04-adjuvanted HPV-16/18 VE against CIN3+
causally related to HPV-16/18 in the HPV-nave1 TVC from the
end-of-study data from the PATRICIA trial [9]. The distribution of
HPV-16/18 in CC cases specic for each country was taken from the
Institut Catala dOncologia (ICO) Information Centre on HPV and
cancer database [2], using a weighted distribution if the summed
distribution exceeded 100% (all HPV = 100%) or the unadjusted
distribution if the sum of the distribution did not exceed 100%.
Country-specic HPV distributions were used where available or
valid. Data were considered not valid when data for less than 7
HPV types were reported or the sum of the minimum and maximum number of samples for the determination of any of the HPV
type distribution was less than 100. For countries without countryspecic data, regional values when available or continental values
were used.
The annual numbers of CC cases and deaths (irrespective of HPV
type and HPV-16/18-related) potentially prevented by HPV vaccination at steady-state were tabulated for each individual country
for four scenarios of vaccine coverage i.e. 50, 70, 90 and 100%.
The formulae below formally describes the calculations used.
CC preventedirrespective = CC VEirrespective vac coverage

2. Materials and methods


2.1. Potential impact of vaccination on the number of CC cases
and deaths prevented over a one-year period
2.1.1. Country specic CC cases and deaths prevented irrespective
of HPV type
The number of CC cases and deaths, irrespective of HPV type,
potentially prevented annually by HPV vaccination at steady-state
(once the entire susceptible population has been vaccinated prior
to sexual debut and thus the total population is assumed to benet
from vaccination) for each country was estimated by multiplying
the reported annual number of cases by vaccine coverage and VE
as a proxy for vaccine effectiveness as follows:
Cases prevented = annual number of cases coverage
vaccine effectiveness

The annual number of cases and deaths was specic to each


country, using projections for 2008 from the International Agency
for Research in Cancer [1].
The proxy vaccine effectiveness irrespective of HPV type used
against CC cases and deaths was 93% (95% CI:7999%). It is based
on the most recent data on the HPV-16/18 AS04-adjuvanted VE
against CIN3+ irrespective of HPV type in the HPV- nave1 TVC
from the end-of-study results from the PATRICIA trial [9]. The
efcacy observed in this population is thought to be representative of the VE among the primary target population for HPV
vaccination programmes in many countries worldwide, i.e. girls
pre-sexual debut [11,12]. Vaccination was assumed to offer lifetime
protection.

CC preventedHPV16/18 = CC VE16/18 %HPV16/18 vac coverage


CC preventednon-HPV-16/18
= CC preventedirrespective CC preventedHPV-16/18
with:
CC preventedirrespective = annual number of cc cases/deaths prevented by vaccination irrespective of HPV type
CC preventedHPV-16/18 = annual number of CC cases/deaths prevented by vaccination causally related to HPV-16/18 infection
CC preventednon-HPV-16/18 = annual number of CC cases/deaths prevented by vaccination not related to HPV-16/18 infection
CC = country-specic annual incident number of CC cases/deaths
Vac coverage = HPV vaccination coverage
VEirrespective = 93% vaccine efcacy irrespective of HPV type [9]
VE16/18 = 100% vaccine efcacy against CC cases/deaths causally
related to HPV-16/18 infection [9]
2.1.3. Regional and worldwide CC cases and deaths prevented
The CC cases and deaths prevented are graphically presented by
WHO continent. These were estimated by summing up all the CC
cases and deaths prevented of the countries constituting each of
the WHO continents (i.e. Africa, America, Asia, Europe, Oceania). A
worldwide estimate was made by summing up the results for all
countries for vaccination coverage levels ranging from 0 to 100%.
The number of CC cases and deaths averted not causally related
to HPV-16/18 infection were estimated at three possible scenarios
of vaccination coverage (50, 70 and 90%) for each WHO continent
and worldwide by taking the difference between the CC cases and
deaths prevented that are causally related to HPV-16/18 and the
CC cases prevented by vaccination irrespective of HPV type for all
countries in the analysis.

G. Van Kriekinge et al. / Vaccine 32 (2014) 733739

735

Table 1
Input data for cost-offset analysis.
Parameter
Estimated survival time [1]
Annual number of incident
CC cases
Expected 5-year cumulative
incidence
5-year CC prevalence
Proportion alive after 5 years
Average survival time (years)
Unit cost of CC treatment
Reported unit costs and yeara
Unit costs updated to 2011
value [20]a
lifetime cost

Mexico

Canada

Germany

Thailand

South Africa

10,186

1419

4440

9997

5743

50,930

7095

22,200

49,985

28,715

34,294
0.673
3.4

5149
0.726
3.6

14,745
0.664
3.3

35,338
0.707
3.5

15,922
0.554
2.8

MXP 81,478 (2006) [14]


MXP 100,994

CAD 15,302 (2006) [15]


CAD 16,809

EUR 12,342 (2006) [16,17]


EUR 13,448

THB 170,275 (2009) [18]


THB 182,614

ZAR 35,360(2007) [19]


ZAR 46,254

MXP 343,380

CAD 60,512

EUR 44,378

THB 639,149

ZAR 128,237

Annual average treatment cost reported for all countries.


Note: MXP - Mexican Pesos; CAD - Canadian Dollar; EUR - Euro; THB - Thai Baht; ZAR - South African Rand; CC - Cervical cancer.

2.1.4. CC-related cost-offset


In ve countries (Mexico, Canada, Germany, Thailand and South
African Republic) the expected reduction in CC treatment costs
resulting from the cases potentially prevented by HPV vaccination
(cost-offset) were estimated. One country among countries with
available data was randomly selected from each of the following
continents: Asia, Africa, Europe, South America and North America.
The total estimated cost-offset, from the healthcare payer perspective was calculated by multiplying the number of incident CC cases
prevented by the country-specic estimated lifetime cost per case:

Table 2
Input data for precancerous lesion analysis.
Parameter

Italy

Annual number of CIN2/3


% of HPV-16/18 in HSIL
Vaccine efcacy CIN2+
Cost of CIN2/3 treatment
Cost of CIN2/3 treatment at
2011 values [20]

592 [5]
11,976 [21]
48.7 (Italy) [2]
42.4 (Asia continent) [2]
64.9% [9]
EUR 992 [21]
MYR 1558 [5]
EUR 1019
MYR 1644

Malaysia

Note: CIN2/3 - Cervical intraepithelial neoplasia grade 2/3; HSIL - High-grade squamous intraepithelial lesion; EUR - Euro; MYR - Malaysian Ringgit.

Total cost offset = incident CC prevented lifetime cost

2.2. Potential impact of HPV vaccination on high grade cervical


lesions

For each of the ve countries, the total cost-offset for CC cases


prevented irrespective of the causative HPV type, CC prevented
causally related to HPV-16/18 infection, and the difference between
them, i.e. the additional cost-offset from protection against HPV
types other than HPV-16/18 was estimated. For comparison purposes the cost-offset related to CC cases other than HPV-16/18
were converted to international dollars using the 2011 purchase
power parity conversion factor for gross domestic product based on
data from the World Bank for each country [13]. For each analysis,
vaccination coverage was assumed to be 80%.

The potential annual effect of HPV vaccination on the burden


related to CIN 2/3 at vaccination steady state was estimated in
two countries: Italy and Malaysia, randomly selected based on data
availability. The method used is identical to the one used to estimate the vaccine impact on CC cases and deaths. The number of
CIN2/3 cases prevented with vaccination irrespective of HPV type,
the expected number of HPV-16/18 related CIN2/3 avoided by vaccination cases as well as the difference between the two were
estimated. Vaccination coverage was assumed to be 80% in both
countries.

2.1.4.1. Unit costs. Lifetime CC treatment costs, from a healthcare payer perspective, were obtained from published literature
[1419]. For each country in this cost analysis, the publications
reported average annual costs rather than lifetime costs. An approximation of lifetime cost was obtained by multiplying the average
annual cost by the estimated average survival time for patients with
incident CC in each country over the 5 years post diagnosis. It was
assumed that a cancer patient alive for 5 years post diagnosis is
cured and hence without any treatment and costs associated. The
average survival time was estimated for each country using data
on the number of annual incident cases and estimates of 5 year
prevalence reported by Globocan 2008 [1] as follows:
(5 years prevalent cases/incident cases 5) 5
= average survival over the 5 years post diagnosis

Costs for CC treatment were expressed in local currency and


updated to 2011 values using the country-specic Consumer Price
Index reported by the World Bank for each country [20]. Estimated
survival times and lifetime costs are shown in Table 1.

2.2.1. CIN2/3 cases prevented irrespective of HPV type


The prevalent annual numbers of CIN2/3 lesions prior to the
introduction of vaccination for Italy and Malaysia were retrieved
from literature (Table 2) [5,21]. The vaccine effectiveness against
CIN2/3 lesions irrespective of HPV type was assumed at 64.9% based
on the VE reported against CIN2+ lesions, irrespective of HPV type in
the HPV-nave1 TVC from the end-of-study results from the PATRICIA trial [9].
2.2.2. CIN2/3 cases prevented causally related to HVP-16/18
Vaccine effectiveness against CIN2/3 lesions related to HPV16/18 was estimated based on the effectiveness against HPV-16/18
CIN2/3 lesion and the proportion of CIN2/3 related to HPV-16/18.
The vaccine effectiveness against HPV-16/18-related CIN2/3 lesions
was assumed to be 100%, based on VE against CIN2+ causally related
to HPV-16/18 reported from the end-of-study from the PATRICIA
trial among the HPV-nave1 TVC [9].

1
TVC nave: includes all vaccinated subjects who received at least one dose of
vaccine, who had normal cytology, were HPV DNA negative for 14 oncogenic HPV
types (including HPV-16 and HPV-18), and seronegative for HPV-16 and HPV-18 at
baseline.

736

G. Van Kriekinge et al. / Vaccine 32 (2014) 733739

Fig. 1. Cervical cancer cases avoided by HPV vaccination at varying levels of coverage in (a) Africa, (b) America, (c) Asia, (d) Europe, (e) Oceania (f) World.

The proportion of CIN2/3 related to HPV type-16/18 was calculated based on the HPV-16/18 distribution reported for high-grade
cervical lesions in the ICO HPV Information Centre database for each
country [2] (Table 2).

Consumer Price Index reported by the World Bank for each country
[20].

2.2.3. CIN2/3-related cost-offset


The expected CIN2/3-related treatment costs potentially offset
by HPV vaccination was estimated assuming that 100% of CIN2/3
lesions prevented by vaccination would be treated. The offset on
treatment costs was estimated by multiplying the number of cases
potentially prevented by the CIN2/3 treatment unit cost. For comparison purposes the cost-offset related CIN2/3 treatment other
than HPV-16/18 were converted to international dollars using 2011
purchase power parity conversion factor for gross domestic product, based on data from the World Bank for each country [13].

3.1. CC cases, deaths prevented and treatment costs-offset

2.2.3.1. Unit cost. Unit costs for the treatment of CIN2/3 in each
country are shown in Table 2. Costs were expressed in local
currency and updated to 2011 value using the country-specic

3. Results

Fig. 1 presents country level results grouped by WHO continent


and worldwide of the estimated annual numbers of CC cases potentially avoided by HPV vaccination at steady-state at varying levels
of vaccination coverage. Individual country estimates at four levels
of vaccination coverage (50, 70, 90 and 100%) are shown in Supplementary File 1. In all ve WHO continents, numbers of cases
potentially prevented by vaccination was at least 18% greater in
the analyses including cases causally related to HPV irrespective of
type, compared with the cases causally related to HPV-16/18 infection only. The relative difference (i.e. the percentage increase of
cases avoided causally related to all HPV types vs. HPV-16/18 only)
was most pronounced in Africa (34%). Relative increase of number

G. Van Kriekinge et al. / Vaccine 32 (2014) 733739

737

Fig. 2. Cervical cancer deaths avoided by HPV vaccination at varying levels of coverage in (a) Africa, (b) America, (c) Asia, (d) Europe, (e) Oceania.

of cases avoided for other WHO continents was 27% for America,
26% for Asia, 21% for Europe, 18% for Oceania and 27% worldwide.
A similar pattern was observed for the estimated annual
numbers of CC deaths potentially prevented by HPV vaccination
(Fig. 2). Similarly to CC cases prevented, the inclusion of CC deaths
prevented irrespective of HPV type in the analysis increased by
at least 18% the estimated number of deaths potentially avoided,
with the relative difference having the same values as for CC cases

analysis. Individual country estimates for the CC deaths potentially


prevented at four levels of vaccination coverage (50, 70, 90 and
100%) are shown in Supplementary File 2.
Table 3 shows the estimated annual cost-offset associated
with CC prevention at steady-state in Mexico, Canada, Germany,
Thailand and South African Republic. Including VE irrespective of
HPV type in the analysis increased the estimated cost-offset in all
ve countries by at least 10 million Int$.

Table 3
Cost-offset for cervical cancer treatment over one year at vaccine steady-state.
Parameter

Mexico

Canada

Germany

Thailand

South African Republic

CC cases avoided HPV-16/18 alone


Cost avoided HPV-16/18 alone
CC cases avoided irrespective of HPV type
Cost avoided irrespective of HPV type
Difference in cost avoided (local currency)
Difference in cost (2011, Int$)

5500
1,870,281,833 MXN
7578
2,576,832,748 MXN
706,550,915 MXN
Int$ 86,289,462

843
51,443,667 CAD
1056
64,391,131 CAD
12,947,464 CAD
Int$ 10,505,160

2754
122,994,317 EUR
3303
147,522,853 EUR
24,528,536 EUR
Int$ 30,575,717

5903
3,810,735,886 THB
7439
4,802,146,849 THB
991,410,962 THB
Int$ 56,912,151

2885
369,998,774 ZAR
4273
547,928,120 ZAR
177,929,347 ZAR
Itl$ 33,506,831

Note: MXP - Mexican Pesos; CAD - Canadian Dollar; EUR - Euro; THB - Thai Baht; ZAR - South African Rand; Int$ - International Dollar; CC - Cervical cancer; HPV - Human
papillomavirus.

738

G. Van Kriekinge et al. / Vaccine 32 (2014) 733739

Table 4
CIN2/3 cases avoided, CIN2/3 treatment cost avoided and cost-offset over one year
at vaccine steady-state and 80% vaccination coverage rate.
Parameter
CIN 2/3 cases avoided
HPV 16/18 only
Irrespective of HPV type
Difference (irrespective minus HPV
16/18 only)
Cost-offset
HPV 16/18 only
Irrespective of HPV type
Difference (irrespective minus HPV
16/18 only)
Cost difference in Int$

Italy

Malaysia

4665
6218
1553

201
307
106

4,753,635 EUR
6,336,142 EUR
1,582,507 EUR

330,444 MYR
504,708 MYR
174,264 MYR

Int$ 2,003,903

Int$ 91,936

Note: Int$ - International Dollar; HPV - Human papillomavirus; CIN 2/3 - Cervical
intraepithelial lesions grade 2/3.

3.2. CIN2/3 cases avoided and treatment cost-offset


Table 4 presents the estimated annual numbers of CIN2/3 cases
avoided by HPV vaccination at steady-state in Italy and Malaysia.
The estimated vaccine impact on CIN2/3 cases, and treatment costs
averted were 33 and 53% higher in Italy and Malaysia respectively,
for the analysis irrespective of HPV type, compared with the estimates for HPV-16/18 only.
4. Discussion
The results presented here suggest that HPV vaccination of
young girls nave to HPV with the AS04-adjuvanted HPV-16/18 vaccine could reduce the number of CC cases and deaths in countries
worldwide, with the absolute number of CC cases and deaths and
hence, lives saved depending on the vaccination coverage achieved.
Additionally, we have attempted to quantify the impact of
protection against HPV types other than HPV-16/18. Although
the vaccine was designed to target HPV-16/18, end-of-study data
from the 4-year PATRICIA trial demonstrated efcacy against nonvaccine HPV types [10], and an overall VE against CIN3+ lesions of
93% irrespective of HPV type in the lesion in the HPV-nave1 TVC
cohort [9]. We have applied these data to the currently observed
disease burden in all WHO reported countries to estimate the additional health benets of vaccination that accrue from protection
against HPV types other than HPV-16/18. When protection irrespective of HPV types was considered, the number of CC cases
and deaths potentially prevented by vaccination was at least 18%
larger than when only HPV-16/18 were considered. The increased
potential benet was seen across all ve WHO continents, and
was particularly pronounced in Africa, where non-HPV-16/18 cases
account for 17,125 cases prevented at 70% vaccination coverage representing an additional 34% cases potentially prevented,
compared with 272 cases additionally prevented in Oceania, representing an additional 18% cases potentially prevented.
HPV vaccination also has the potential to reduce the morbidity
associated with precancerous lesions. Management of precancerous lesions detected by screening may require surgical procedures,
such as conisation. In countries with absent or poorly developed
cervical screening programmes, few precancerous lesions will be
detected and the health impact will mainly be observed when
undetected lesions progress to symptomatic cancer. Conversely,
in countries with well-developed and effective screening programmes, many precancerous lesions are detected at an early stage
and are usually treated before they progress to cancer, so the health
burden will tend to shift away from CC treatment towards precancerous lesion treatment. In some industrialised countries, the
economic burden of precancerous lesions may exceed or approach
the economic burden of CC. For example, a study of patients in a

health maintenance plan in the USA found that treatment of CC


accounted for 10% of healthcare expenditure on HPV-related cervical disease and treatment of precancerous lesions accounted for
17% [22]. In Belgium, the total annual cost of CC treatment to the
healthcare payer was estimated at Euro 6.5 million and the annual
cost of precancerous lesions at Euro 1.97 million in a retrospective
study [23]. Other morbidities associated with treatment of precancerous lesions of the cervix avoidable by vaccination such as
increased risk of perinatal death and pre-term births should also
be considered [24,25].
To our knowledge, this is the rst estimate of the potential
impact of HPV vaccination on CC cases and deaths to apply the
recent data on VE irrespective of HPV type causing the lesion
reported from the end-of-study data in the PATRICIA trial [9,10].
Thus, it builds on previous estimates of the potential health impact
of HPV vaccination by taking into account the most recent data
on protection against non-vaccine HPV types. In this analysis, we
extrapolated VE data from PATRICIA to Africa, thereby implicitly
assuming that VE would not differ between Africa and the regions
included in the trial. Recent study results in African girls and women
showed that immune responses were similar to those observed in
European populations thus strengthening our assumption [26].
Our study has limitations. Although, we have used countryspecic data from WHO databases to ensure consistency by the
use of the same data source, these estimates may differ from local
epidemiological data of the countries. Second, our estimates are
derived at vaccine steady-state, which in a real-life setting will
need many years to be achieved. Consequently, the full potential of reduction in CC cases and deaths estimated here will need
time to be realised. However, the estimated potential reductions in
high-grade CIN could be observed earlier. For example, in Australia,
where a large catch up for the HPV vaccination programme was
put in place, a signicant reduction in the incidence of high-grade
lesions was observed within three years of introduction of the HPV
vaccination programme [27]. We have also assumed that the crossprotective effect of vaccination will have the same duration as
vaccine-type HPV. Recent data from an independently conducted
clinical trial reported persistence of cross-neutralizing antibody
titres 3 years after vaccination, suggesting that cross-reactive antibody responses are likely to persist long-term [29]. This was further
corroborated by data from the follow-up of the phase II trial of
the AS04-adjuvanted HPV-16/18 vaccine have demonstrated crossreactive immune response that is sustained up to at least 7 years
post vaccination. This strengthens our assumption that the crossprotective effect demonstrated in the PATRICIA trial may be of long
duration [28]. The estimated benets of vaccination could however
be less than projected, should the cross-protection be demonstrated to wane over time. Lastly, our estimates did not take account
herd immunity effects, and thus we may have underestimated the
potential effect of HPV vaccination.
5. Conclusion
Our evaluation estimates that vaccination of young girls nave
to HPV with the AS04-adjuvanted HPV-16/18 vaccine could result
in reductions in the number of CC cases and deaths in countries
worldwide resulting in lives saved and CC-related cost-offsets. A
proportion of the estimated potential reduction relates to protection against non-HPV-16/18 related HPV types. Additionally,
prevention of precancerous lesions could reduce the morbidity
associated with these lesions and result in further cost-savings.
Acknowledgements
The authors are grateful to Carole Nadin (Fleetwith Ltd. c/o
GlaxoSmithKline Vaccines) for medical writing assistance and

G. Van Kriekinge et al. / Vaccine 32 (2014) 733739

Maud Boyer and Sarah Fico (both Business and Decision Life Sciences c/o GlaxoSmithKline Vaccines) for editorial assistance and
publication co-ordination. The authors sincerely acknowledge the
input from Eduardo Franco (McGill University, Canada) on the
design of this health economic evaluation.
Trademarks: Gardasil is a registered trade mark of Merck Sharp
& Dohme Corp., Cervarix is a registered trade mark of the GlaxoSmithKline group of companies.
Conict of interest: ND and GVK are employees of the GlaxoSmithKline group of companies and ND owns stock in the
GlaxoSmithKline group of companies. DC has no conict of interest
related to this manuscript. XC has performed consultancy work for
the GlaxoSmithKline group of companies. He received funding for
board membership and lectures from the GlaxoSmithKline group
of companies. None of these activities was directly related to the
current study
Author contributions: GVK, XC, DC and ND conceived and
designed the study; GVK and ND developed the model; GVK and XC
acquired the data; GVK analysed the data; all authors have made
substantial intellectual contributions to the manuscript, reviewed
and commented on drafts and approved the nal manuscript.
Role of the funding source: GlaxoSmithKline Biologicals SA
was the funding source and was involved in all stages of the
study conduct and analysis. GlaxoSmithKline Biologicals SA also
funded all costs associated with the development and the publishing of the present manuscript. All authors had full access to
the data and agreed with the submission of the manuscript for
publication.
Appendix A. Supplementary data
Supplementary material related to this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.vaccine.
2013.11.049.
References
[1] Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. GLOBOCAN 2008
v2.0. Cancer incidence and mortality worldwide in 2008. IARC CancerBase No.
10; 2010, http://globocan.iarc.fr/ cited 2012 Mar 19.
[2] ICO. ICO (Institut Catal dOncologia) Information Centre on HPV and Cancer (HPV Information Centre). Data Query tool; 2013. http://www.hpvcentre.
net/dataquery.php/ cited 2013 Jul 1.
[3] Walboomers JM, Jacobs MV, Manos MM, et al. Human papillomavirus is
a necessary cause of invasive cervical cancer worldwide. J Pathol 1999
Sep;189(1):129.
[4] Clifford GM, Smith JS, Plummer M, Munoz N, Franceschi S. Human papillomavirus types in invasive cervical cancer worldwide: a meta-analysis. Br J
Cancer 2003 Jan 13;88(1):6373.
[5] Aljunid S, Zafar A, Saperi S, Amrizal M. Burden of disease associated with cervical cancer in Malaysia and potential costs and consequences of HPV vaccination.
Asian Pac J Cancer Prev 2010;11(6):15519.
[6] Updated consensus guidelines for managing abnormal cervical cancer
screening tests and cancer precursors. Frederick, MD, USA: American Society
for Colposcopy and Cervical Pathology (ASCCP); 2012.
[7] Paavonen J, Jenkins D, Bosch FX, et al. Efcacy of a prophylactic adjuvanted
bivalent L1 virus-like-particle vaccine against infection with human papillomavirus types 16 and 18 in young women: an interim analysis of a phase
III double-blind, randomised controlled trial. Lancet 2007 30 June;369(9580):
702161.

739

[8] Paavonen J, Naud P, Salmeron J, et al. Efcacy of human papillomavirus (HPV)16/18 AS04-adjuvanted vaccine against cervical infection and precancer caused
by oncogenic HPV types (PATRICIA): nal analysis of a double-blind, randomised study in young women. Lancet 2009 6 July;374(9686):30114.
[9] Lehtinen M, Paavonen J, Wheeler CM, et al. Overall efcacy of HPV-16/18 AS04adjuvanted vaccine against grade 3 or greater cervical intraepithelial neoplasia:
4-year end-of-study analysis of the randomised, double-blind PATRICIA trial.
Lancet Oncol 13 Jan 2012;1:8999.
[10] Wheeler CM, Castellsague X, Garland SM, et al. Cross-protective efcacy of HPV16/18 AS04-adjuvanted vaccine against cervical infection and precancer caused
by non-vaccine oncogenic HPV types: 4-year end-of-study analysis of the randomised, double-blind PATRICIA trial. Lancet Oncol 2012 Jan 13;1:10010.
[11] Introduction of HPV vaccines in European Union countriesan update. Stockholm: European Centre for Disease Prevention and Control; 2012, in press.
[12] FDA licensure of bivalent human papillomavirus vaccine (HPV2 Cervarix)
for use in females and updated HPV vaccination recommendations from the
Advisory Committee of Immunization Practices (ACIP). MMWR 2010 May
28;59(20):6269.
[13] World Bank. PPP conversion factor, GDP (LCU per international $); 2012.
http://data.worldbank.org/indicator/PA.NUS.PPP?display=default/ cited 2012
Dec 4.
[14] Colantonio L, Gomez JA, Demarteau N, Standaert B, Pichon-Riviere A, Augustovski F. Cost-effectiveness analysis of a cervical cancer vaccine in ve Latin
American countries. Vaccine 2009 Sep 4;27(40):295519.
[15] Anonychuk AM, Bauch CT, Merid MF, Van Kriekinge G, Demarteau N. A
cost-utility analysis of cervical cancer vaccination in preadolescent Canadian
females. BMC Public Health 2009;9:401.
[16] Hillemanns P, Petry K, Largeron N, McAllister R, Tolley K, Bsch K. Costeffectiveness of a tetravalent human papillomavirus vaccine in Germany. J of
Public Health 2009 1 April;17(2):7786.
[17] Siebert U, Sroczynski G, Hillemanns P, et al. The German cervical cancer screening model: development and validation of a decision-analytic
model for cervical cancer screening in Germany. Eur J Public Health 2006
April;(2):18592.
[18] Van Enckevort PJ, Kongngamkam K, Van Kriekinge G. Costs and consequences
of HPV vaccination in Thailand: results of a prevalence based model. (Abstract
PCN62 presented at the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) 14th Annual European Congress, 58 Nov, Madrid,
Spain). Value Health 2011;14(7):A445.
[19] Sinanovic E, Moodley J, Barone MA, Mall S, Cleary S, Harries J. The potential
cost-effectiveness of adding a human papillomavirus vaccine to the cervical
cancer screening programme in South Africa. Vaccine 2009;27(44):2026196.
[20] World Bank. Consumer price index (2005 = 100); 2012. http://data.worldbank.
org/indicator/FP.CPI.TOTL/countries/1W?display=default [cited 2012 Sep 11].
[21] Capri S, Gasparini R, Panatto D, Demarteau N. Cost-consequences evaluation between bivalent and quadrivalent HPV vaccines in Italy: the potential
impact of different cross-protection proles. Gynecol Oncol 2011 June
1;121(3):51421.
[22] Insinga RP, Glass AG, Rush BB. The health care costs of cervical human papillomavirus related disease. Am J Obstet Gynecol 2004 July;191(1):11420.
[23] Annemans L, Remy V, Lamure E, et al. Economic burden associated with the
management of cervical cancer, cervical dysplasia and genital warts in Belgium.
J Med Econ 2008;11(1):13550.
[24] Arbyn M, Kyrgiou M, Simoens C, et al. Perinatal mortality and other severe
adverse pregnancy outcomes associated with treatment of cervical intraepithelial neoplasia: meta-analysis. BMJ 2008;337:a1284.
[25] Bruinsma FJ, Quinn MA. The risk of preterm birth following treatment for precancerous changes in the cervix: a systematic review and meta-analysis. BJOG
2011 Aug;118(9):411031.
[26] Sow PS, Watson-Jones D, Kiviat N, et al. Safety and immunogenicity of human
papillomavirus-16/18 AS04-adjuvanted vaccine: a randomized trial in 1025year-old HIV-Seronegative African girls and young women. J Infect Dis 2012
Dec 13.
[27] Brotherton JM, Fridman M, May CL, Chappell G, Saville AM, Gertig DM. Early
effect of the HPV vaccination programme on cervical abnormalities in Victoria,
Australia: an ecological study. Lancet 2011 June 18;377(9783):922085.
[28] Moscicki AB, Wheeler CM, Romanowski B, et al. Immune responses elicited by
a fourth dose of the HPV-16/18 AS04-adjuvanted vaccine in previously vaccinated adult women. Vaccine 2012 Dec 17;31(1):23441.
[29] Kemp TJ, Safaeian M, Hildesheim A, et al. Kinetic and HPV infection
effects on cross-type neutralizing antibody and avidity responses induced by
Cervarix((R)). Vaccine 2012;31(1):16570.