ORIGINAL STUDY

Human Papillomavirus Type Distribution in Invasive

Cervical Cancer and High-Grade Cervical Intraepithelial
Neoplasia Across 5 Countries in Asia
Swee Chong Quek, MBBCh,* Boon Kiong Lim, MBBS, MRCOG,Þ Efren Domingo, MD,þ
Ruey Soon, MBBS, MRCOG,§ Jong-Sup Park, MD, PhD,|| Thi Nhung Vu, MD, PhD,¶
Eng Hseon Tay, MBBS, MMED, DGO,Þ Quang Thanh Le, MMed,# Young-Tak Kim, MD, PhD,**
Ba Quyet Vu, MD, PhD,ÞÞ Ngoc Thanh Cao, MD, PhD,þþ Genara Limson, MD,§§
Viet Thanh Pham, MD, PhD,|||| Anco Molijn, MSc,¶¶ Gunasekaran Ramakrishnan, MSc,##
and Jing Chen, MD***

Objective: Independent, prospective, multicenter, hospital-based cross-sectional studies

were conducted across 5 countries in Asia, namely, Malaysia, Vietnam, Singapore, South
Korea, and the Philippines. The objectives of these studies were to evaluate the prevalence of
human papillomavirus (HPV) types (high risk and others including coinfections) in women
with invasive cervical cancer (ICC) and high-grade precancerous lesions.
Methods: Women older than 21 years with a histologic diagnosis of ICC and cervical
intraepithelial neoplasia [CIN 2 or 3 and adenocarcinoma in situ (AIS)] were enrolled.
Cervical specimens were reviewed by histopathologists to confirm the presence of ICC or
CIN 2/3/AIS lesion and tested with short PCR fragment10-DNA enzyme immunoassay-line
probe assay for 14 oncogenic HPV types and 11 non-oncogenic HPV types. The prevalence
of HPV 16, HPV 18, and other high-risk HPV types in ICC [including squamous cell car-
cinoma (SCC) and adenocarcinoma/adenosquamous carcinoma (ADC/ASC)] and CIN 2/3/
AIS was estimated.
Results: In the 5 Asian countries, diagnosis of ICC was confirmed in 500 women [SCC
(n = 392) and ADC/ASC (n = 108)], and CIN 2/3/AIS, in 411 women. Human papillo-
mavirus DNA was detected in 93.8% to 97.0% (84.5% for the Philippines) of confirmed ICC
cases [94.0%Y98.7% of SCC; 87.0%Y94.3% (50.0% for the Philippines) of ADC/ASC] and
in 93.7% to 100.0% of CIN 2/3/AIS. The most common types observed among ICC cases
were HPV 16 (36.8%Y61.3%), HPV 18 (12.9%Y35.4%), HPV 52 (5.4%Y10.3%), and HPV 45
(1.5%Y17.2%), whereas among CIN 2/3/AIS cases, HPV 16 (29.7%Y46.6%) was the most
commonly observed type followed by HPV 52 (17.0%Y66.7%) and HPV 58 (8.6%Y16.0%).
Conclusions: This article presents the data on the HPV prevalence, HPV type distribution,
and their role in cervical carcinogenesis in 5 Asian countries. These data are of relevance
to public health authorities for evaluating the existing and future cervical cancer prevention
strategies including HPV-DNA testingYbased screening and HPV vaccination in these Asian
populations.
Key Words: Asia, Cervical intraepithelial neoplasia, Epidemiology,
Human papillomavirus, Invasive cervical cancer

*Parkway Gynaecology Screening and Treatment Center, Singapore,
Singapore; †Department of Obstetrics and Gynecology, University
Malaya Medical Centre, Kuala Lumpur, Malaysia; ‡Department
of Obstetrics and Gynecology, University of The Philippines,
Philippine General Hospital, Manila, Philippines; §Department of
Obstetrics and Gynecology, Hospital Likas, Kota Kinabalu, Sabah,
Malaysia; ||Department of Obstetrics and Gynecology, The Catholic
University of Korea, Seoul St. Mary’s Hospital, Seoul, Korea;
¶Department of Obstetrics and Gynecology, Hung Vuong Hospital,
Ho Chi Minh City, Vietnam; †Thomson Women Cancer Center,
Singapore; #Department of Obstetrics and Gynecology, Tu Du
Hospital, Ho Chi Minh City, Vietnam; **Department of Obstetrics
and Gynecology, College of Medicine, University of Ulsan, Asan
Medical Center, Seoul, Korea; ††Department of Gynaecological
Oncology, National Hospital for Obstetrics and Gynaecology,
Hanoi, Vietnam; ‡‡Department of Obstetrics and Gynecology, Hue

148 International Journal of Gynecological Cancer & Volume 23, Number 1, January 2013

Copyright © 2012 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer & Volume 23, Number 1, January 2013
College of Medicine and Pharmacy, Hue, Vietnam; §§Department
of Obstetrics & Gynecology, Makati Medical Center, Makati,
Philippines; ||||Department of Mother and Child Health, Ministry
of Health, Vietnam; ¶¶DDL Diagnostic Laboratory, Rijswijk, The
Netherlands; ##GlaxoSmithKline Pharmaceuticals, Bangalore, India;
and ***GlaxoSmithKline Pte Ltd, Singapore, Singapore.
Address correspondence and reprint requests to Jong-Sup Park, MD,
PhD, Department of Obstetrics and Gynaecology, Catholic
University Medical College, 505 Banpodong, Seochogu, Seoul,
137-040, Republic of Korea. E-mail: jspark@catholic.ac.kr.
A core writing team (CWT) was formed to take lead in driving the
development of this publication. The main responsibilities of the
CWT were to review, recommend, and endorse decisions relating
to the manuscript on behalf of all coauthors. Members of the
CWTwere Swee Chong Quek, Boon Kiong Lim, Efren Domingo,
Ruey Soon, Jong-Sup Park, and Jing Chen. Other than the first
author (Swee Chong Quek) and last author (Jing Chen) who were
decided by a voting exercise by the CWT, the other members of
the CWT are listed ahead of other co-authors and in accordance to
the number of subjects they had recruited, with the one with the
highest recruitment number listed first.
GlaxoSmithKline Biologicals SA was the funding source for all of
the 5 studies mentioned in this article and was involved in all
stages of the conduct and analysis of the studies.
GlaxoSmithKline Biologicals SA also took charge of all costs
associated to the development and the publishing of the present
article. [Study eTrack numbers: 110425 (Malaysia); 110994
(Philippines); 106714 (Singapore); 110676 (South Korea);
110428 (Vietnam)].
Received July 16, 2012, and in revised form September 24, 2012.
Accepted for publication September 27, 2012.
(Int J Gynecol Cancer 2013;23: 148Y156)
C ervical cancer is the third most common cancer in women
worldwide and second most common cancer in women
in the developing world.1,2 As per the latest estimation by the
International Agency for Research on Cancer, in 2008, globally
there were 530,000 new cases of cervical cancer and 275,000
cervical cancer deaths; more than 85.0% of cervical cancer
cases and deaths reported worldwide were from the developing
countries.1,2 Cervical cancer was reportedly the most common
cancer in South-Central Asia, with an age-adjusted incidence
of 24.6 per 100,000. Overall, 159,800 cervical cancer deaths
were recorded in Asia in 2008.1
In countries with well-organized cytologic screening
programs, Papanicolaou smear testing has facilitated early
detection and treatment of precancerous cervical lesions,
thereby reducing the number of cases progressing to cervical
cancer.3 However, in Asia, the coverage of cervical cancer
screening program is low, mainly because of limited access
(G5%) to screening in most countries and the absence of
organized screening programs in some countries.4Y6 Hence,
the burden of cervical cancer in most Asian countries continues
to be high.
Human papillomavirus (HPV) has been identified as
the necessary cause of cervical cancer (overall prevalence of
HPV in cervical cancer is È99.7%).7 The 8 most common HPV
* 2013 IGCS and ESGO
Copyright © 2012 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
HPV Type Distribution in ICC and CIN 2/3
Swee Chong Quek, Boon Kiong Lim, Efren Domingo, Ruey Soon,
Jong-Sup Park, Thi Nhung Vu, Eng Hseon Tay, Quang Thanh Le,
Young-Tak Kim, Ba Quyet Vu, Ngoc Thanh Cao, Genara
Limson, and Viet Thanh Pham were principal
investigators involved in the 5 studies sponsored by
GlaxoSmithKline (GSK) Biologicals SA. GSK provided research
funds to their institutions for purposes of conducting the study.
Swee Chong Quek, Boon Kiong Lim, Ruey Soon, Thi Nhung Vu,
and Quang Thanh Le had received honoraria from GSK for
lectures including services on speaker bureaus. GSK provided
travel support to Swee Chong Quek, Boon Kiong Lim, Thi
Nhung Vu, Quang Thanh Le, Ba Quyet Vu, and Anco Molijn for
travel to study meetings or congresses for the presentation of
study results. GSK had also paid Anco Molijn’s institution for
the laboratory tests involved in these studies and consultancy
fees for his research collaborations with GSK. Gunasekaran
Ramakrishnan and Jing Chen are employees of the
GlaxoSmithKline group of companies.
All authors had full access to the study data and have made
substantive intellectual contributions to the publication. The
corresponding author had final responsibility to submit for
publication.
Supplemental digital content is available for this article. Direct URL
citation appears in the printed text and is provided in the HTML
and PDF versions of this article on the journal’s Web site
(www.ijgc.com).
Copyright * 2013 by IGCS and ESGO
ISSN: 1048-891X
DOI: 10.1097/IGC.0b013e31827670fd
types in cervical cancer worldwide are HPV 16, 18, 45, 31, 33,
52, 58, and 35.8,9 Two prophylactic HPV vaccines are now
available in many Asia-Pacific countries. Data on the pre-
dominantly circulating HPV types is essential to estimate the
potential impact of prophylactic HPV vaccination and to assess
the importance of HPV testingYbased cervical screening.
However, limited data are available on HPV prevalence and
type distribution in most Asian countries. Moreover, the
available information had several limitations, including
small sample sizes, HPV detection methods that were not
fully validated, and analyzed samples that were collected in
the last decade or were extremely restrictive in the selection
of the study population.10Y16 These limitations restricted the
capability to draw unambiguous conclusion from the avail-
able data. Even a more recent review17 of HPV prevalence in
Hong Kong, Singapore, and Taiwan had certain inherent lim-
itations (results included unpublished data and pooled data),
which impeded inferential data interpretation.17
To address the gap in the knowledge of the epidemiol-
ogy of HPV in Asia, a series of epidemiological studies with
similar designs were conducted in 5 Asian countries. The
countries were chosen with the aim of collecting data from both
affluent economies with better health care resources (South
Korea and Singapore) as well as the developing economies in
149
Quek et al International Journal of Gynecological Cancer
Asia (Malaysia, Vietnam, and the Philippines) because differ-
ences in socioeconomic conditions, demographic features, and
cervical screening programs might be associated with variations
in the burden of HPV infection (see Supplemental Digital
Content, Table 1, http://links.lww.com/IGC/A135).
The objectives of these studies were to evaluate the pre-
valence of HPV types (high risk and others including co-
infections) in women with invasive cervical cancer (ICC) and
high-grade precancerous lesions and to identify the associ-
ated risk factors. This article presents the findings on HPV
prevalence and type distribution in the study populations.
MATERIALS AND METHODS
Study Design and Population
Five single-country, hospital-based cross-sectional studies
were conducted in the following Asian countries: Singapore (July
2007Y2009), South Korea (September 2007 to August 2009),
Malaysia (December 2007Y2009), Vietnam (January 2009 to
March 2010), and the Philippines (March 2009 to June 2010).
Infection with HPV was investigated in women older
than 21 years diagnosed with ICC [including squamous cell
carcinoma (SCC), adenocarcinoma (ADC), and adenosqua-
mous carcinoma (ASC)] or high-grade precancerous lesions
[including cervical intraepithelial neoplasia (CIN) 2 or 3 and
adenocarcinoma in situ (AIS)]. Women with previous HPV
vaccination and recurrent episodes of ICC or high-grade lesions
and whose cervical specimens were collected after radiotherapy
or chemotherapy were excluded. Cervical specimens fixed in
10% buffered formalin obtained during routine clinical pro-
cedures were analyzed for HPV DNA. A questionnaire was
administered to the women in a face-to-face interview for in-
formation on their education level, smoking habits, and repro-
ductive history. The studies were approved by independent ethics
committees of all participating hospitals. Informed consent was
obtained from all women before study participation.
Laboratory Procedures
All specimens were prepared in a histology unit spe-
cialized in molecular analysis. The 10.0% buffered formalin-
fixed and paraffin-embedded cervical specimens were sectioned
using sandwich technique. A 4.0-Km section was cut for
pathologic confirmation of diagnosis on a hematoxylin-
eosinYstained section. Two sets of 3  8 Km sections were
taken for whole-section HPV DNA analysis, then followed
by a hematoxylin-eosinYstained section. Extensive measures
were implemented to prevent cross contamination. After sec-
tioning a paraffin block, the microtome was cleaned, and a new
knife was used. Negative control sections were used to check for
cross-contamination. The paraffin sections from all sites were
processed and reviewed in 1 laboratory, DDL Diagnostic Lab-
oratory (Voorburg, The Netherlands), by a panel of independent
histopathologists. Once the presence of ICC or CIN 2/3/AIS
lesions was confirmed, they were tested for HPV genotype.
Specimens were tested for HPV DNA by polymerase chain re-
action (PCR) amplification/typing using the HPV short PCR
fragment/line probe assay (SPF10 PCR/LiPA25; version 1.0)
system.18 Each DNA isolation run and PCR run contained
HPV-positive and HPV-negative controls. Ten microliters of
150
Copyright © 2012 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
& Volume 23, Number 1, January 2013
proteinase K-treated DNA was added to 40 KL of PCR mix.
The SPF10 PCR primer set amplifies a small fragment of 65 bp
from the L1 region of mucosal HPV genotypes. Amplification
products were detected using the HPV SPF10 PCR (version
1.0) DNA enzyme immunoassay (DEIA) system. The DEIA-
positive SPF10 amplimers were used to identify the HPV
genotype by reverse hybridization with the HPV LiPA25, con-
taining probes for 25 different HPV genotypes, 14 oncogenic
HPV types (HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59,
66, and 68), and 11 non-oncogenic HPV types (HPV 6, 11, 34,
40, 42, 43, 44, 53, 54, 70, and 74); SPF10 HPV LiPA25 version
1.0 (Labo Biomedical Products, Rijswijk, The Netherlands,
based on licensed Innogenetics technology). Following this, total
DNA was isolated from formalin-fixed and paraffin-embedded
material using a procedure involving treatment with proteinase
K.19 Proteinase K was heat-inactivated by incubation at 96-C for
10 minutes.
Statistical Analyses
Human papillomavirus positivity rate and the preva-
lence of each individual HPV type with 95% confidence
intervals (CIs) were calculated for each country. All statistical
analyses were descriptive; the primary analyses were performed
on subjects with laboratory-confirmed diagnosis of ICC and
high-grade precancerous lesions. Demographic characteristics
of study subjects and overall prevalence of HPV types were
analyzed for all 5 countries. In addition, pooled analyses of the
prevalence of HPV types in the 4 Southeast Asian countries
(Singapore, Malaysia, the Philippines, and Vietnam) were per-
formed to compare with that in Korea. All statistical analyses
were performed using Statistical Analysis System version 9.1.3
(SAS, NC).
In the pooled analyses of the 4 Southeast Asian coun-
tries, the subjects from the same study/country were considered
as clusters, and the estimates (with 95% CI) were computed
using generalized estimating equations, to take the correlation
between subjects from each study/country into account (ie,
within and between cluster correlation). The generalized esti-
mating equation was fitted using the Statistical Analysis
System procedure GENMOD,20,21 assuming an exchange-
able correlation matrix.
RESULTS
Demographic Characteristics
A total of 1012 women were enrolled into the 5 studies.
After histologic review of cervical specimens obtained from
these women, 500 cases of ICC (SCC, 392; ADC, 108) and
411 cases of high-grade cervical lesions (CIN 2/3 and AIS) were
analyzed for HPV types. The reasons for exclusion of women
from analyses are presented in Table 1A. Table 1B summarizes
the cases analyzed by study (country) and histologic diagnosis.
The mean age at sample collection and other characteristics
of the women are summarized in Table 2.
HPV Prevalence
Overall, HPV DNA was detected in 93.8% or more of
ICC samples from all countries except the Philippines where
it was detected in 84.5% of ICC samples. Furthermore, HPV
* 2013 IGCS and ESGO
International Journal of Gynecological Cancer & Volume 23, Number 1, January 2013 HPV Type Distribution in ICC and CIN 2/3

TABLE 1A. Reasons for exclusion of women from analyses

The
Singapore Korea Malaysia Vietnam Philippines
Enrolled 179 200 220 281 122
(total cohort)
Exclusions 8 3 46 21 13
Reasons for Diagnosis of Diagnosis of Diagnosis of Diagnosis of Diagnosis of
exclusions ICC/CIN 2/3 ICC/CIN 2/3 ICC/CIN 2/3 ICC/CIN 2/3 ICC/CIN 2/3
not confirmed, not confirmed, not confirmed, not confirmed, not confirmed,
n = 1; samples n = 1; protocol n = 45; sample n = 19; protocol n = 6; protocol
not processed, violations (inclusion/ not sent to violations (inclusion/ violations (inclusion/
n=7 exclusion criteria), laboratory, exclusion criteria), exclusion criteria),
n = 1; invalid n=1 n=2 n=7
HPV DNA results,
n=1
Included in 171 197 174 260 109
analyses

DNA was detected in 93.7% or more of high-grade lesions
(Table 3). Most HPV infections were present as single infections
in ICC samples (Q88.2%) especially ADC/ASC, with a rela-
tively lower proportion of single infection detected in samples
from Singapore (85.2%). Infections with undetermined HPV
types were rarely detected in other countries except for
Malaysia and Singapore. The prevalence of HPV DNA was
consistently lower in samples from ADC/ASC when compared
with SCC in all countries. The proportion of multiple HPV
type infections was substantially higher among CIN 2/3/AIS
cases (15.3%Y20.0%) than ICC cases (3.1%Y8.2%).
The 8 most common HPV types detected in ICC and
CIN 2/3/AIS samples from different countries are shown
in Figure 1. Human papillomavirus types 16 and 18 pre-
dominated ICC samples in all countries. Although differences
were noted in the prevalence of HPV types across countries,
HPV 52, 45, 58, 33, 31, and 68 always featured among the
8 highest HPV types (next to HPV 16 and 18). The prevalence
of HPV 45 was notably higher in the Philippines and Malaysia
than in the other 3 countries. These 8 HPV types together
accounted for 95% to almost all of the HPV-positive ICC
samples from the respective countries. The proportion of single
HPV infections was high for both ICC cases (Q80.0%;
80.0%Y91.5% across countries) and CIN 2/3/AIS cases
(Q66.7%; 66.7%Y83.9% across countries). The HPV type
distribution in single infections across countries has been
presented in Web appendix Table 1.
Human papillomavirus type distribution in ICC sam-
ples from Korea was distinct when compared with the other
Southeast Asian countries; in Korea, the prevalence of HPV
16 [61.3% (95% CI, 50.6Y71.2)] was significantly higher,
and the prevalence of HPV 18 [12.9% (95% CI, 6.8Y21.5)]
was significantly lower than in the other countries [41.7%
(95% CI, 36.0Y47.7) and 29.6% (95% CI, 25.2Y34.4), re-
spectively]. In addition, HPV 45 was less prevalent [1.1%
(95% CI, 0.0Y5.8)], and HPV 33 was more prevalent [6.5%
(95% CI, 2.4Y13.5)] in ICC samples from Korea compared
with the other Southeast Asian countries [6.9% (95% CI,
2.7Y16.4); 2.8% (95% CI, 1.1Y7.1)] (Fig. 2).
The HPV type distribution in SCC samples was similar
to that in ICC cases. With the exception of Korea, HPV 18 was
overrepresented in ADC/ASC compared with SCC, whereas

TABLE 1B. Samples analyzed for HPV types in the 5 studies

ICC
All SCC*† ADC/ASC† CIN 2/3/AIS Total No. Cases
Korea 97 80 (82.5) 17 (17.5) 100 197
Malaysia 101 78 (77.2) 23 (22.8) 73 174
Singapore 65 50 (76.9) 15 (23.1) 106 171
The Philippines 103 85 (82.5) 18 (17.5) 6 109
Vietnam 134 99 (73.9) 35 (26.1) 126 260
Overall 500 392 (78.4) 108 (21.6) 411 911
Data are presented as number or n (%).
*Including undifferentiated carcinoma.
†Denominators are number of all ICC cases.

* 2013 IGCS and ESGO 151

Copyright © 2012 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
Quek et al International Journal of Gynecological Cancer & Volume 23, Number 1, January 2013

TABLE 2. Demographic characteristics of study subjects and population characteristics by country

Korea Malaysia Singapore Philippines Vietnam
No. women 197 174 171 109 260
Age at sample collection for ICC cases,* 47.68 (11.62) 52.86 (11.36) 53.72 (12.87) 50.53 (10.06) 47.46 (10.30)
mean (SD), y
Age at sample collection for CIN2/3 cases,* 40.14 (11.29) 49.30 (10.60) 40.96 (10.94) 46.50 (11.83) 43.13 (9.72)
mean (SD), y
Age at sexual debut, mean (SD), y 22.8 (3.1)† 20.2 (3.9) 20.7 (4.0) 19.7 (3.8)† 21.8 (3.6)
Smoking, n (%) 36 (18.3) 50 (28.7) 51 (29.8) 51 (46.8) 10 (3.8)
Education level, n (%)
No formal 2 (1.0) 20 (11.5) 18 (10.5) 0 (0) 2 (0.8)
Primary 11 (5.6) 68 (39.1) 47 (27.5) 47 (43.1) 122 (46.9)
Secondary 81 (41.1) 74 (42.5) 58 (33.9) 45 (41.3) 122 (46.9)
Postsecondary/university 103 (52.3) 12 (6.9) 48 (28.1) 17 (15.6) 14 (5.4)
Marital status, n (%)
Single 24 (12.2) 6 (3.4) 12 (7.0) 10 (9.2) 1 (0.4)
Married 142 (72.1) 129 (74.1) 127 (74.3) 51 (46.8) 223 (85.8)
Widowed 19 (9.6) 18 (10.3) 13 (7.6) 15 (13.8) 20 (7.7)
Divorced or separated 12 (6.1) 19 (10.9) 17 (9.9) 17 (15.6) 14 (5.4)
Living with a partner 0 (0) 2 (1.1) 2 (1.2) 16 (14.7) 2 (0.8)
Parity, n (%)
0 35 (17.8) 8 (4.6) 29 (17.0) 4 (3.7) 8 (3.1)
1Y2 132 (67.0) 32 (18.4) 76 (44.4) 31 (28.4) 131 (50.4)
3Y5 28 (14.2) 90 (51.7) 56 (32.7) 49 (45.0) 105 (40.4)
Q6 2 (1.0) 44 (25.3) 10 (5.8) 25 (22.9) 16 (6.2)
Population characteristics
Gross domestic product per capita current prices 21,529.252 8238.508 44,968.376 2219.003 1272.180
(US dollar) 2011‡
Total population (millions) 2011‡ 49.062 28.713 5.185 95.834 89.316
Population of women aged between 15Y64 y 17197 9117 1847 28379 31219
(thousands) 2010*
Average life expectancy of women 2009§ 83 y 76 y 84 y 73 y 74 y
Established cervical screening program Yes No Yes No No
*From United Nations: Dept. of economic and Social Affairs; Available at: http://esa.un.org/unpd/wpp/Excel-Data/population.htm; Accessed
on April 04, 2012.
†Based on available data.
‡From International Monetary Fund; Available at: http://www.imf.org/external/pubs/ft/weo/2010/02/weodata/weorept.aspx?sy=2011&ey=2011
&scsm=1&ssd=1&sort=country&ds=.&br=1&c=548%2C566%2C576%2C542%2C582&s=NGDPD%2CNGDPRPC%2CNGDPDPC%2CPPPPC
%2CLP&grp=0&a=&pr.x=85&pr.y=5; Accessed on April 04, 2012.
§From World Health Organization: http://www.who.int/countries/en/; Accessed on April 04, 2012.

HPV 16 was significantly underrepresented in ADC/ASC
samples of all countries. A substantial proportion of ADC/
ASC cases were infected with HPV 45 in Malaysia and the
Philippines, but this was not observed in the other countries.
In CIN 2/3/AIS samples, HPV 16, 52, 58, 51, 33, 31, 18, and
35 were the most common HPV types.
DISCUSSION
A recent meta-analysis reported an overall prevalence
of HPV DNA in 89.7% of ICC cases from Eastern Asia; HPV
16, 18, 58, 52, 33, 31, 45, and 59 were the most common HPV
types detected.22 The latest data from a worldwide cross-
sectional study by the Catalan Institute of Oncology showed a
similar prevalence of HPV DNA and overall HPV type dis-
tribution in ICC cases from Asia, the only difference being
that the prevalence of HPV 45 was the third highest among
all types.9 In our study, although HPV DNA was detected in a
higher percentage of ICC cases (Q95.9%) and high-grade
lesions (Q93.7%), other findings confirmed that the HPV
genotypes in these 5 Asian countries were consistent with the
existing information; however, the relative importance of each

152 * 2013 IGCS and ESGO

Copyright © 2012 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer & Volume 23, Number 1, January 2013 HPV Type Distribution in ICC and CIN 2/3

TABLE 3. Human papillomavirus infection type

All ICC SCC* ADC/ASC CIN 2/3 /AIS
Country
n % 95% CI n % 95% CI n % 95% CI n % 95% CI
Korea Total 97 V V 80 V V 17 V V 100 V V
HPV-positive* 93 95.9 89.8Y98.9 78 97.5 91.3Y99.7 15 88.2 63.6Y98.5 100 100.0 96.4Y100
Single 82 88.2 79.8Y93.9 68 87.2 77.7Y93.7 14 93.3 68.1Y99.8 77 77.0 67.5Y84.8
infection†
Multiple 7 7.5 3.1Y14.9 6 7.7 2.9Y16.0 1 6.7 0.2Y31.9 20 20.0 12.7Y29.2
infection†
Unknown HPV 4 4.3 1.2Y10.6 4 5.1 1.4Y12.6 V V V 3 3.0 0.6Y8.5
infection†
Malaysia Total 101 V V 78 V V 23 V V 73 V V
HPV positive* 97 96.0 90.2Y98.9 77 98.7 93.1Y100 20 87.0 66.4Y97.2 70 95.9 88.5Y99.1
Single 86 88.7 80.6Y94.2 67 87.0 77.4Y93.6 19 95.0 75.1Y99.9 50 71.4 59.4Y81.6
infection†
Multiple 3 3.1 0.6Y8.8 2 2.6 0.3Y9.1 1 5.0 0.1Y24.9 13 18.6 10.3Y29.7
infection†
Unknown HPV 8 8.2 3.6Y15.6 8 10.4 4.6Y19.4 V V V 7 10 4.1Y19.5
infection†
Singapore Total 65 V V 50 V V 15 106 V V
HPV positive* 61 93.8 85.0Y98.3 47 94 83.5Y98.7 14 93.3 68.1Y99.8 103 97.2 92.0Y99.4
Single 52 85.2 73.8Y93.0 38 80.9 66.7Y90.9 14 100.0 76.8Y100.0 80 77.7 68.4Y85.3
infection†
Multiple 5 8.2 2.7Y18.1 5 10.6 3.5Y23.1 0 0 V 17 16.5 9.9Y25.1
infection†
Unknown HPV 4 6.6 1.8Y15.9 4 8.5 2.4Y20.4 0 0 V 6 5.8 2.2Y12.2
infection†
Philippines Total 103 V V 85 V V 18 V V 6 V V
HPV positive* 87 84.5 76.0Y90.9 78 91.8 83.8Y96.6 9 50.0 26.0Y74.0 6 100 54.1Y100.0
Single 79 90.8 82.7Y95.9 70 89.7 80.8Y95.5 9 100.0 66.4Y100.0 4 66.7 22.3Y95.7
infection†
Multiple 7 8.0 3.3Y15.9 7 9.0 3.7Y17.6 V V V V V V
infection†
Unknown HPV 1 1.1 0.0Y6.2) 1 1.3 0.0Y6.9 V V V 2 33.3 4.3Y77.7
infection†
Vietnam Total 134 V V 99 V V 35 V V 126 V V
HPV positive* 130 97.0 92.5Y99.2 97 98.0 92.9Y99.8 33 94.3 80.8Y99.3 118 93.7 87.9Y97.2
Single 119 91.5 85.4Y95.7 88 90.7 83.1Y95.7 31 93.9 79.8Y99.3 99 83.9 76.0Y90.0
infection†
Multiple 8 6.2 (2.7Y11.8) 7 7.2 (3.0Y14.3) 1 3.0 (0.1Y15.8) 18 15.3 9.3Y23.0
infection†
Unknown HPV 3 2.3 (0.5Y6.6) 2 2.1 (0.3Y7.3) 1 3.0 (0.1Y15.8) 1 0.8 0.0Y4.6
infection†
Results of the subgroup analyses of questions 3 to 6: Small centers (SC) was defined as centers with a number of beds less than the median.
The subgroup of large centers (LC) consisted of the other participating hospitals. LSA laparoscopic approach; HE hysterectomy; BSO bilateral
salpingo-oophorectomy; LAN lymphadenectomy.
*Including undifferentiated carcinoma.
†Denominators are number of HPV-positive cases.

* 2013 IGCS and ESGO 153

Copyright © 2012 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
Quek et al International Journal of Gynecological Cancer & Volume 23, Number 1, January 2013

FIGURE 1. Human papillomavirus types among HPV-positive women by histologic diagnosis. Women with single,
multiple, and unknown infections; KR, Korea; MAL, Malaysia; PH, the Philippines; SG, Singapore; VN, Vietnam.

HPV type varied according to the population origin of the

samples. In addition, the data from our study also emphasized
the significant role of HPV 18 in causing ADC/ASC of
the cervix.
Compared with the 4 countries in Southeast Asia, the
HPV type distribution in Korea showed a distinct pattern.
Human papillomavirus 16 and 18 were the predominant HPV
types in ICC cases across the countries in our studies and
accounted for 36.8% to 61.3% and 12.9% to 27.9% of all
HPV-positive ICC cases. The prevalence of HPV 16 in Korea
was noticeably higher (61.3%), and the prevalence of HPV 18
was lower (12.9%) than in the other countries. Moreover,
HPV 45, which was less prevalent in ICC samples from Korea
(1.1%), was more prevalent in ICC samples from the
4 Southeast Asian countries (6.7%). A previous meta-analysis FIGURE 2. Human papillomavirus types among
(1995Y2007)23 of HPV epidemiology in Korean women HPV-positive ICC cases in Korea versus the 4 Southeast
reported that HPV 16 was the predominant type detected Asian countries combined. KR, Korea; SEA, Southeast Asia.

154 * 2013 IGCS and ESGO

Copyright © 2012 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer & Volume 23, Number 1, January 2013
regardless of the progression of cervical disease. In addition,
HPV 58 accounted for a significant proportion of ICC cases
and high-grade lesions; HPV 58, 33, and 52 together accounted
for almost 20.0% of ICC cases and high-grade lesions. How-
ever, in our study, HPV 58 and 52 were not among the dominant
types in Korea, whereas HPV 33 was more prevalent in Korea
than in the other Southeast Asian countries. As observed in
previous data from Asia,8 HPV 18 and 45 appear to be more
important etiologic agents for cervical carcinogenesis in
Southeast Asian women than in Korean women, whereas the
prevalence of HPV 33 in ICC cases in Korea would require
further evaluation.
Compared with SCC, a much smaller number of ADC/
ASC cases in each country were available for the analysis of
HPV types. In contrast to the abundance of HPV types in
SCC, only a limited number of HPV types were detected in
these ADC/ASC samples. In previous studies,9 HPV 18, 16,
and 45 accounted for most of the HPV-positive ADC/ASC
cases in Asia. The relevant importance of HPV 45 was strik-
ingly higher in Malaysia and the Philippines than in the other
countries. Again, the prevalence of HPV 16 and 18 in Korea
is contrary to that in the 4 Southeast Asian countries. Given
the limited number of ADC/ASC samples tested in all study
countries, further investigation on the HPV genotypes in these
populations is warranted.
Except for the study in the Philippines where only
6 cases of CIN 2/3/AIS were available for analysis, the major
HPV types detected in high-grade lesions was similar to that
observed for the ICC samples. In terms of HPV type distri-
bution, HPV 18, 45, and 16 (to a less extent) were underrep-
resented in high-grade lesions compared with ICC, whereas
almost all the other high-risk types, particularly HPV 52, 58, 33,
and 31, were overrepresented in high-grade lesions. These
findings support the hypothesis that the risk of progression
from high-grade cervical lesions to ICC is dependent on the
HPV type causing the infection.24 The high-risk types es-
pecially HPV 16 and 18 have been reported to be the major
etiologic agents for ICC worldwide.24 Also of note is that
multiple HPV type infection was far more prevalent in high-
grade lesions than in ICC, adding to the difficulty of asso-
ciating the lesion to specific HPV type(s). It is possible to
investigate this further using laser capture25; however, this
was not pursued in our studies.
Although there have been single and multicountry
studies presenting HPV type distribution, there were several
strengths in our studies. First, the use of cervical samples
collected closer to study start, standardized protocols and
study procedures, particularly the use of a central laboratory
for the histologic review of cervical samples and HPV DNA
analysis, ensured scientific exactitude through consistent his-
tologic diagnosis across all studies and comparability of find-
ings across countries. Second, because it is known that the HPV
detection rate is inversely related to the duration for which the
samples are stored,9 the prospective enrolment in our studies
was beneficial, as recently collected cervical samples were
analyzed for HPV DNA as opposed to most of the previous
studies that analyzed samples archived for different dura-
tions. This, together with the use of highly sensitive SPF10
DEIA-LiPA assays, ensured a high HPV detection rate in
* 2013 IGCS and ESGO
Copyright © 2012 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
HPV Type Distribution in ICC and CIN 2/3
our studies. The studies in Singapore, Korea, and Malaysia
(2007Y2009) were conducted a year earlier than those in
Vietnam and the Philippines (2009Y2010). However, this is
unlikely to have an impact on the results presented in this
article. Finally, very few multicountry studies in published lit-
erature have focused primarily on Asian countries. Although
Bosch et al26 and de Sanjose et al9 have reported the prevalence
of HPV in cervical cancer across the world, only a small
proportion of the countries studied were from Asia in the
former-3 of 22 countries, and both the studies included
minimal or no data from several of the 5 countries included
in our study.
There were certain limitations in our studies. In some
countries, the number of samples analyzed for HPV types
were small; this is especially true for ADC/ASC cases from all
countries and high-grade lesions in the Philippines. Findings in
these histologic groups would require further investigations
of more samples. Furthermore, we did not pursue the analyses
of samples infected with multiple HPV types or with HPV
types untypeable by the assay used in this study. The variation
in the HPV type distribution in Korea when compared with that
in other Southeast Asian countries as observed in our study
and in published literature warrants further investigation.
Despite these limitations, findings from these studies
contribute significantly to bridge the gaps in published lit-
erature regarding the HPV type distribution in both ICC and
high-grade lesions in Asian women, especially those from the
Southeast Asian countries. The epidemiology of HPV type
distribution reported in this article may help public health
authorities in assessing the need of HPV-DNA testing based
cervical screening program and to evaluate the potential im-
pact of HPV vaccination on the population.
ACKNOWLEDGMENTS
We thank all participants of these 5 studies and all staff
at the study sites for their many contributions.
We are also grateful to the following from the
GlaxoSmithKline group of companies for their support of the
studies and manuscript development: the clinical operations
teams and medical affairs teams of Korea, Malaysia, the
Philippines, Singapore, and Vietnam. In addition, we ac-
knowledge Amrita Ostawal and Avishek Pal for providing
writing support to this article and Ming Tung Lim for publi-
cation coordination.
REFERENCES
1. World Health Organization (WHO).International Agency on
Cancer research (IARC). GLOBOCAN 2008. Available at:
http://globocan.iarc.fr/factsheet.asp. Accessed on March 6, 2012.
2. Ferlay J, Shin HR, Bray F, et al. Estimates of worldwide
burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer.
2010;127:2893Y2917.
3. Pagliusi SR, Teresa AM. Efficacy and other milestones for
human papillomavirus vaccine introduction. Vaccine. 2004;23:
569Y578.
4. Wong LP, Wong YL, Low WY, et al. Knowledge and awareness
of cervical cancer and screening among Malaysian women
155
Quek et al International Journal of Gynecological Cancer
who have never had a Pap smear: a qualitative study. Singapore
Med J. 2009;50:49Y53.
5. Singapore Health Promotion Board. Available:
www.hpb.gov.sg. Accessed: 25 Jul 2010.
6. Kwak MS, Choi KS, Spring BJ, et al. Predicting the stages
of adoption of cervical cancer screening among Korean women.
Prev Med. 2009;49:48Y53.
7. Walboomers J M, Jacobs MV, Manos MM, et al.
Human Papillomavirus is a necessary cause of invasive
cervical cancer worldwide. J Pathol. 1999;189:12Y19.
8. Guan P, Howell-Jones R, Li N, et al. Human papillomavirus
(HPV) types in 115,789 HPV-positive women: a meta-analysis from
cervical infection to cancer. Int J Cancer. 2012;131:2349Y2359.
9. De Sanjose S, Quint WGV, Alemany L, et al. Human
papillomavirus genotype attribution in invasive cervical cancer:
a retrospective cross-sectional worldwide study. Lancet Oncol.
2010;11:1048Y1056.
10. Yadav M, Nurhayati ZA, Padmanathan A, et al. Polymerase
chain reaction detection and restriction enzyme typing of human
papillomavirus in cervical carcinoma. Med J Malaysia.
1995;50:64Y71.
11. Chan R, Khoo L, Ho TH, et al. A comparative study of cervical
cytology, colposcopy and PCR for HPV in female sex workers
in Singapore. Int J STD AIDS. 2001;12:159Y163.
12. Chow VT, Loh E, Yeo WM, et al. Identification of multiple
genital HPV types and sequence variants by consensus and
nested type-specific PCR coupled with cycle sequencing.
Pathology. 2000;32:204Y208.
13. Nhung VT. A survey on HPV types infection among female
in Ho Chi Minh City (HCMC) by PCR. Ho Chi Minh City Med J.
2006;10:402Y407.
14. Anh PT, Hieu NT, Herrero R, et al. Human papillomavirus
infection among women in South and North Vietnam.
Int J Cancer. 2003;104:213Y220.
15. Cheah PL, Looi LM, Sivanesaratnam V. Human papillomavirus
in cervical cancers of Malaysians. J Obstet Gynaecol Res.
2011;37:489Y495.
16. Chen HC, Schiffman M, Lin CY, et al. Persistence of
type-specific human papillomavirus infection and increased
156
Copyright © 2012 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
& Volume 23, Number 1, January 2013
long-term risk of cervical cancer. J Natl Cancer Inst.
2011;103:1387Y1396.
17. Tay SK, Ngan HYS, Chu TY, et al. Epidemiology of human
papillomavirus infection and cervical cancer and future
perspectives in Hong Kong, Singapore and Taiwan. Vaccine.
2008;26S:M60YM70.
18. Kleter B, van Doorn LJ, Schrauwen L, et al. Development
and clinical evaluation of a highly sensitive PCR-reverse
hybridization line probe assay for detection and identification
of anogenital human papillomavirus. J Clin Microbiol.
1999;37:2508Y2517.
19. Quint WGV, Scholte G, van Doorn LJ, et al. Comparative
analysis of human papillomavirus infections in cervical scrapes
and biopsy specimens by general SPF10 PCR and HPV
genotyping. J Pathol. 2001;194:51Y58.
20. Boos D. On generalized score tests. The American Statistician.
1992;46:327Y333.
21. Rotnitzky A, Jewell NP. Hypothesis testing of regression
parameters in semiparametric generalized linear models for
cluster correlated data. Biometrika. 1990;77:485Y497.
22. Li N, Franceschi S, Howell-Jones R, et al. Human
papillomavirus type distribution in 30,848 invasive cervical
cancers worldwide: variation by geographical region,
histological type and year of publication. Int J Cancer.
2010;128:927Y935.
23. Chan PKS, Luk ACS, Parl JS, et al. Identification of human
papillomavirus type 58 lineages and the distribution worldwide.
J Infect Dis. 2011;203:1565Y1573.
24. Smith JS, Lindsay L, Hoots B, et al. Human papillomavirus
type distribution in invasive cervical cancer and high-grade
cervical lesions: a meta-analysis update. Int J Cancer. 2007;121:
621Y632.
25. Quint W, Jenkins D, Molijn A, et al. One virus, one
lesion-individual components of CIN lesions contain a
specific HPV type. J Pathol. 2012;227:62Y71.
26. Bosch FX, Manos MM, Munoz N, et al. Prevalence of human
papillomavirus in cervical cancer: a worldwide perspective.
International biological study on cervical cancer (IBSCC)
Study Group. J Natl Cancer Inst. 1995;87:796Y802.
* 2013 IGCS and ESGO