Prevalence and type distribution of human papillomavirus in healthy Japanese women aged 20 to 25 years old enrolled in a clinical study
Clinical trial registry: Clinicaltrials.gov.
Name: Human papillomavirus vaccine (Cervarix) efficacy, immunogenicity and safety trial in adult Japanese women with GSK Biologicals HPV-16/18 vaccine.
To whom correspondence should be addressed.
Efficacy, immunogenicity and tolerability of the human papillomavirus (HPV)-16/18 AS04-adjuvanted vaccine were evaluated in Japanese women aged 20–25 years, for which results have been reported previously. We analyzed the baseline data from that study and report the prevalence rates of HPV infection in young healthy Japanese women. One thousand and the forty Japanese women aged 20–25 years were enrolled in a phase II, double-blind, controlled, randomized, multicenter study. At study entry, cervical specimens were collected from the women and tested by line probe assay for 25 HPV-types and by HPV-16/18-specific polymerase chain reaction. The most frequently detected HPV-type in baseline cervical specimens was HPV-52 (8.1%), followed by HPV-16 (6.5%), HPV-51 (4.5%), HPV-18 (4.0%) and HPV-31 (3.8%). The proportion of HPV DNA-positive women increased with severity of cytological abnormalities: 26.1% (237/908) in normal cytology, 93.3% (70/75) in low-grade squamous intraepithelial lesion and 100% (7/7) in high-grade squamous intraepithelial lesion. The relative contribution of HPV-16 and HPV-18 was 4.1 and 3.0% for normal cytology cases, and 20.0 and 16.0% in low-grade squamous intraepithelial lesion, respectively. HPV-16 was found in four of seven high-grade squamous intraepithelial lesion cases (57.1%) and five of the six cervical intraepithelial neoplasia 2+ cases (83.3%). Multiple and single HPV infections were observed in 13.5% (140/1039) and 20.7% (215/1039) of all women, respectively. The HPV prevalence rates in Japanese women aged 20–25 years underline the importance of HPV vaccination at a young age and this report should be useful for monitoring changes in HPV prevalence after widespread HPV vaccination in Japanese women. (Cancer Sci 2011; 102: 877–882)
Persistent infection with HPV is the necessary cause for developing cervical cancer,(1) the second most common cancer in women worldwide.(2) HPV is a common sexually transmitted infection. Population-based studies suggest that up to as much as 80% of sexually active women are exposed to at least one genital HPV type in their lifetime.(3) While HPV infection occurrence is highest in young sexually active women, a relatively high HPV prevalence has also been detected in post-menopausal women. Infection at this later stage may be due to reactivation of earlier acquired infections or infection from new sexual partners later in life.(4) At least 14 oncogenic HPV types (high-risk HPV) have been causally linked to cervical cancer.(5) HPV-16 and HPV-18 are the most prevalent and account for more than 70% of all invasive cervical cancers worldwide; HPV-31 and HPV-45 are responsible for an additional 10% of cases(6,7) followed by HPV-33, HPV-35, HPV-52 and HPV-58.(6,7) HPV vaccines are now licensed in more than 100 countries, coinciding with large-scale national and regional immunization programs aimed at young adolescent girls.(8) Since infection with HPV may occur throughout the lifetime of a sexually active woman, it is important that vaccination induces a strong, sustained antibody response to ensure long-term protection.(9)
Currently, there are two HPV vaccines available in the world: a bivalent (Cervarix®, GlaxoSmithKline Biologicals) and a quadrivalent (Gardasil®, Merck). The bivalent vaccine is an AS04-adjuvanted vaccine specifically targeting HPV-16 and -18 types, while the quadrivalent vaccine is an aluminium-adjuvanted vaccine specifically targeting HPV-6, -11, -16 and -18 types. Both vaccines are immunogenic, generally well-tolerated, with clinically acceptable safety profiles.(10–13) Prevention by both vaccines of CIN2+ lesions caused by HPV-16 and -18 was >90% in women aged 15–25 years old and negative for these HPV-types at study entry.(11,14,15) Furthermore, excisional therapies for CIN2+ are reduced by approximately 70% for the bivalent vaccine(16) and 40% for the quadrivalent.(17) Sustained efficacy of both vaccines has been documented for 5 years or more.(11,15,16,18,19) However, it may take 5–10 years after the start of a vaccination program before significant reductions in the incidence of cervical cancer will be apparent.(20)
In Japan, cervical cancer ranks approximately 7th in women overall (incidence rate of 13.6 per 100 000) and 2nd in women aged 15–44 years (incidence rate of 12.0 per 100 000).(21,22) It is estimated that 15 000 women are diagnosed with cervical cancer yearly, leading to approximately 3500 deaths.(21–23) These figures include approximately 2000 new cases and 200 deaths that are estimated to occur every year in Japanese women in their twenties and thirties.(23,24) HPV is currently one of the least known sexually transmitted infections in Japan, resulting in a lower level of public knowledge of the risks of HPV and cervical cancer.(25) HPV-16 and HPV-18 are the most frequently identified HPV-types in invasive cervical cancers in Japan (67.1%), and HPV-52 and HPV-58 appear to be the next most common types accounting for 11.5% of cervical cancers.(26) Prevalence of HPV infection in the Japanese population has been mainly reported based on evaluations of women who have been referred to hospitals and clinics to receive cervical cancer screening. Furthermore, the women examined in previous studies have come from a wide range of ages and, thus, information on the rates of HPV infection in young women, particularly those in their twenties, is lacking. For instance, Inoue et al.(27) reported on a large scale study evaluating HPV testing in over 8000 women in the Ishikawa Prefecture. The median age of these women was 36 years but the age span of the participants extended from 14 to 83 years of age. A further concern is that these studies were usually conducted in small geographic areas relating to a particular hospital, clinic or prefecture. Very little information regarding nationwide HPV prevalence is available.
We have recently reported the results of a clinical study to evaluate the bivalent HPV-16/18 AS04-adjuvanted vaccine (Cervarix®, GlaxoSmithKline Biologicals) in healthy Japanese women aged 20–25 years.(28–30) While the results of this study showed that the HPV-16/18 AS04-adjuvanted vaccine was effective and immunogenic, with a clinically acceptable safety profile in the population studied, an accurate knowledge of the actual HPV infection rates in young healthy Japanese women could further confirm the importance of vaccination against HPV types 16 and 18. We therefore determined the HPV infection status of each woman at study entry as an indicator of the national infection rates. In particular, we analyzed DNA of 25 HPV types (14 oncogenic: 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66 and 68; 11 non-oncogenic types: 6, 11, 34, 40, 42, 43, 44, 53, 54, 70 and 74) using PCR and report here the baseline prevalence of both the oncogenic and non-oncogenic HPV types in the cervical cytology samples of young healthy Japanese women.
Materials and Methods
Healthy Japanese women aged 20–25 years were recruited in this phase II double-blind (observer-blind) controlled, randomized, multicenter study (104798, NCT00316693) between April and October 2006. The 13 centers were located in Aomori, Tokyo, Fukui, Osaka, Hiroshima, Miyazaki and Kagoshima in Japan. Study participants were not screened before enrolment with respect to baseline serological, cytological or HPV DNA status. Inclusion and exclusion criteria were as previously described.(30) The study was conducted following the Declaration of Helsinki (version 1996) and all participants provided written informed consent. All recruitment materials, informed consent, protocol, and amendments were approved by independent institutional review boards.
Subjects were randomized in a 1:1 fashion to receive either the HPV-16/18 AS04-adjuvanted vaccine (containing 20 μg of HPV-16 L1 virus-like particle (VLP) and 20 μg of HPV-18 L1 VLP adjuvanted with 50 μg 3-O-desacyl-4′-monophosphoryl lipid A and 0.5 mg aluminium hydroxide) or a hepatitis A vaccine licensed in Japan (Aimmugen®, Chem-Sero-Therapeutic Research Institute, Kumamoto, Japan; containing 0.5 μg of inactivated hepatitis A antigen) as the control vaccine. Both vaccines were administered intramuscularly according to a 0-, 1-, and 6-month schedule. Investigators obtained cervical specimens with a cervical brush for cytology and HPV DNA as previously described.(18,28–31)
HPV DNA isolated from the cytology specimen was amplified from an aliquot of purified total DNA with the SPF10 broad-spectrum primers. These primers amplify a 65 bp region of the L1 gene and the generic amplification products were detected by hybridization on a microtiter plate (DEIA). HPV-positive specimens were typed by reverse hybridization LiPA. The broad-spectrum PCR SPF10 HPV LiPA25 version 1 and SPF10 HPV DEIA (manufactured by Labo Biomedical Products, Rijswijk, the Netherlands based on licensed INNOGENETICS SPF10 technology) detected 25 HPV types: 14 oncogenic (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66 and 68) and 11 non-oncogenic HPV types (6, 11, 34, 40, 42, 43, 44, 53, 54, 70 and 74). If a sample was negative for HPV-16 or HPV-18 by the SPF10-LiPA25 system, type-specific PCR was performed to confirm the absence of these types using HPV-16 primers that amplified a 92 bp segment of the E6/E7 gene and HPV-18 primers that amplified a 126 bp segment of the L1 gene.(18,28–31)
Cytology was assessed by liquid-based cytology (ThinPrep, Cytyc Corporation, Marlborough, MA, USA) using a central laboratory (Quest Diagnostics, Teterboro, NJ, USA). Cytology results were reported using the 2001 Bethesda system and cytological abnormalities included: (i) ASC-US; (ii) LSIL; (iii) ASC-H; (iv) HSIL; and (v) AGC. Protocol guidelines recommended colposcopy after one report of HSIL or ASC-H. Biopsy was required for any suspected lesions on colposcopy. The central laboratory (Quest Diagnostics) processed and interpreted results from histology samples. All CIN endpoints were confirmed by an expert histopathology review panel that was blinded to vaccine status, HPV DNA status before biopsy, and cytology reports.
The enrolment target of 1000 unscreened women was estimated to provide 800 women who were DNA negative for HPV-16 or HPV-18 at month 0 and 6, and who would be evaluable for assessment of the primary endpoint in the according to protocol group for efficacy analysis. All analyses were based on the TVC; however, some values were absent due to missing or non-evaluable samples. Data is presented as percentage of subjects per group along with the actual number of subjects. Statistical analysis is descriptive in nature.
One thousand and forty healthy Japanese women aged 20–25 years old (mean age: 22.5) were enrolled in 13 study sites in Japan. All women were vaccinated and included in the TVC: 519 women in the HPV group (HPV-16/18 AS04-adjuvanted vaccine) and 521 women in the control group (HAV group, hepatitis A vaccine licensed in Japan). The study sites were located in seven areas with the following recruitment numbers: Tokyo 457 (44.0%), Aomori 160 (15.4%), Osaka 136 (13.1%), Fukui 128 (12.3%), Kagoshima 91 (8.8%), Hiroshima 39 (3.8%) and Miyazaki 29 (2.8%).
The distributions of the tested HPV types in the TVC (irrespective of cytology) and for each cytological status are shown in Table 1. Data was not available for one participant due to a missing sample. Three hundred and fifty-five women (34.2%) of the TVC (irrespective of cytology) tested positive for HPV DNA at study entry. In particular, oncogenic and non-oncogenic HPV types were detected in 304 (29.3%) and 112 (10.8%) women. HPV-52 (8.1%, 84/1039) was the most frequently detected HPV type, followed by HPV-16 (6.5%, 68/1039), HPV-51 (4.5%, 47/1039), HPV-18 (4.0%, 42/1039) and HPV-31 (3.8%, 39/1039).
Table 1. HPV DNA genotype status distribution in cervical samples, and cytological and histological status at study entry (total vaccinated cohort)
The majority of women (87.4%, 908/1039 women) had no cytological abnormalities; 12.6% (131/1039) had cytological abnormalities including ASC-US (4.5%, 47/1039), LSIL (7.2%, 75/1039), HSIL (0.7%, 7/1039) and ASC-H (0.2%, 2/1039). Overall, 26.1% of women with normal cytology (237/908) were positive for HPV DNA (Table 1). In contrast, of the 131 women who had abnormal cytology, 90.1% (118/131) were positive for HPV DNA. The HPV positivity rate was also high for women diagnosed with ASC-US (83.0%, 39/47), LSIL (93.3%, 70/75) and 100% (9/9) for women diagnosed to have HSIL or ASC-H. HPV-16 (23.7%, 31/131) was the most frequently detected HPV type, followed by HPV-52 (19.8%, 26/131), HPV-31 and HPV-56 equally detected in 13.0% (17/131) and HPV-51 (12.2%, 16/131). Fifty-five percent (5/9) of the women diagnosed as HSIL or ASC-H had HPV-16 detected in their cytological specimens. The nine women with HSIL or ASC-H had a colposcopy after the cytology testing. CIN3 was diagnosed in five women, CIN2 in one woman, CIN1 in one woman and two women had no lesions. Four of the five women who were diagnosed with CIN3 and the woman diagnosed with CIN2 were positive for HPV-16. The positivity rate of HPV-16 in the women diagnosed as CIN2 or CIN3 reached 83.3% (5/6).
The number of multiple HPV infections is illustrated by Table 2. Multiple infections were observed in 140 women (13.5%, 140/1039) of the TVC (irrespective of cytology), including one woman (0.1%) testing positive for six HPV types (45/51/52/53/56/58) and six women (0.6%) positive for five HPV types. Fourteen women (1.3%) were positive for four HPV-types, 40 women (3.8%) positive for three types and 79 of the women (7.6%) had two HPV types. When only the HPV positive women (n = 355) are examined, these percentages increase to 0.3% testing positive for six HPV types, 1.7% for five types, 3.9% for four types, 11.3% for three types and 22.2% of the HPV-positive women had two HPV types (Table 2).
Table 2. Number of human papillomavirus (HPV) infections at study entry in the total vaccinated cohort (TVC) and in HPV positive women (HPV+)
This study is the first to evaluate the HPV prevalence for healthy young Japanese women nationwide, based in seven different regions of Japan. Study sites were well-distributed with the most northern site (Aomori) located in the farthest north prefecture of the main island and the most southern site (Kagoshima) located in the farthest south prefecture of the third island of Japan. Healthy Japanese women aged 20–25 years of age were recruited for a clinical study to assess the efficacy, immunogenicity and safety of the HPV-16/18 AS04-adjuvanted vaccine against HPV. We assessed the baseline HPV status using a sensitive PCR and cytology of the women enrolled in this clinical study and these are the data we present.
Although the overall HPV prevalence was slightly higher in the present study, type-specific prevalence data in Japanese women with normal cervical cytology was comparable to two large scale meta-analyses of Japanese data.(32,33) HPV-52 was also the most prevalent HPV type with HPV-16 second in these findings. While HPV-51 was the third most prevalent HPV type based on the results of the current study and of Miura et al.,(33) HPV-51 was fourth in the study of Konno et al.(32) with HPV-58 third. Previous findings from other Japanese studies were strongly supported by the large dataset used for the meta-analysis, in particular for age groupings.(32,33) HPV prevalence in women in the Asia Pacific region who were in their twenties with normal cytology was highest in Australia (30.1%), followed by Japan (23.1%), India (13.2%), Korea (12.7%) and Taiwan (9.9%).(34) Our data for women aged 20–25 years with normal cytology is relatively high (26.1%) compared to other women from the Asia Pacific region, with the exception of Australian women, but similar to that previously reported for Japanese women.(34) Interestingly, the four HPV genotypes with the highest prevalence in this study matched four of the five most common high-risk HPV genotypes in Asian women with normal cytology from China, Singapore, Taiwan and Indonesia.(34)
While HPV-52 is the most prevalent in the Japanese general population, HPV-16 is more closely associated with high-grade precursor lesions and invasive cancer.(32,33) The meta-analysis of Japanese data indicated that in normal cytological samples the prevalence of HPV-16 was 0.84% and of HPV-18 was <0.50%.(32) However, these data were compiled based on screenings taken of women of all ages. While HPV prevalence in the USA was common among women between the ages of 14 and 59 years old, prevalence was highest in women between the ages of 20 and 24 years old.(35) This current nationwide study specifically examined young women aged 20–25 years and shows a higher prevalence of the HPV-16 and HPV-18 types, particularly HPV-18 in normal cytology. Furthermore, HPV prevalence in Japanese women with normal/negative cytology appears to decrease with aging; for example from 26.1% at 22.5 years (average age) in the present study to 22.5% at 35.0 years(26) and 10.2% at 52.4 years.(36) Although these studies were not based in the same geographic locations and used different primer systems for their PCR analysis, it has been previously observed that HPV-16 and HPV-18 DNA positivity appears to be very high in Japanese women aged 20–29 years with CIN2-3 or invasive cervical cancer.(26) Therefore, it is possible that, due to high prevalence of HPV-16 and HPV-18 in Japanese women with normal cytology, cervical cancer caused by these HPV types will increase in the future.
As described above, HPV type distribution in invasive cervical cancers is very similar among countries of eastern Asia (including Japan, China/Hong Kong/Taiwan and Korea). HPV-16 and -18 are the most common types accounting for approximately 70% with the next three most common types being HPV-58, HPV-52 and HPV-33 accounting for an additional 10%.(26,37) Overall HPV and oncogenic HPV prevalence were 34.1% and 29.2% in all study participants. In the pivotal clinical study (PATRICIA) to evaluate the HPV-16/18 AS04-adjuvanted vaccine with approximately 18 000 women aged 15–25 years conducted in Asia Pacific, Europe, and North and South America, oncogenic HPV types were detected in 20.1% of all study participants.(14) Particularly, the prevalence of HPV-16 and HPV-18 was 5.4% and 2.3% in the participants of the clinical study. The broader age range of the women in that study precludes direct comparison of the Japanese data presented here; however, the prevalence of oncogenic HPV types, especially HPV-16 (6.5%) and HPV-18 (4.0%), in Japanese women is higher than that in other populations.
Cervical cancer is preventable if precancerous lesions are detected early enough. Cytological analysis to detect abnormalities such as LSIL and HSIL is used in combination with histology to define the early stages (low-grade cervical lesions) and advanced stages (high-grade cervical lesions) of the disease.(38) The strong association between oncogenic HPV prevalence, particularly of HPV-16, and cytological abnormalities such as LSIL and HSIL observed in this study follows the high incidences of HPV in invasive cervical cancer, HSIL and LSIL throughout Asia.(37) Indeed, an increasing prevalence of oncogenic HPV with increasing cervical lesion severity has been previously reported in Japanese women,(26,32,33) particularly of the HPV-18 genotype.(26) In agreement with this observation, there was also a good correlation between the prevalence of oncogenic HPV types and HSIL across the different parts of Europe.(39)
While the majority of HPV-positive women in this study were infected with a single HPV-type (60.6%), multiple HPV infections were observed in the remaining 39.4%. This is much higher than the 12.0% previously reported for Japanese women with normal cytology and HPV positive that were also positive for multiple HPV infection types.(26) The higher rate observed in our study is most likely due to the differences in the ages of the groups examined as we specifically examined women aged 20–25 years of age. Women aged 20–24 years of age had the highest prevalence of HPV infection compared to younger (14–24 years) and older (25–29, 30–39, 40–49, or 50–59 years) women, especially when restricting the analysis to sexually active females, in a US study.(35) Furthermore, the rate of multiple infection has been demonstrated to be higher in younger than older women in a Danish study and was also linked to sexual behavior.(40) Therefore, HPV vaccines that provide protection against multiple HPV types may be more effective in preventing cervical cancer.
The introduction of systematic cytological screening has resulted in a fall of the death rate from cervical cancer in many countries.(41) Vaccination against HPV-16 and HPV-18 would theoretically decrease by approximately 40% the number of oncogenic HPV-positive findings in screening programs.(39) As HPV type distribution in invasive cervical cancers is very similar among countries of eastern Asia (including Japan, China/Hong Kong/Taiwan and Korea), approximately 70% of invasive cervical cancer cases could be potentially prevented by an HPV-16/18 prophylactic vaccine.(18) Moreover, this could increase to approximately 80% if the vaccine targeted additional HPV types, such as HPV-31 and HPV-45, as well as HPV-16 and HPV-18.(37) Both HPV-16 and HPV-18 are phylogenetically related to other HPV subtypes (HPV-16: HPV-31, -33, -35, -52, -58, -67; HPV-18: HPV-39, -45, -59, -68, -70),(42) suggesting that vaccines targeting HPV-16/18 could cross-react with these subtypes. Cross-protection data have been reported for both the bivalent and the quadrivalent vaccines.(11,16,43,44) The bivalent vaccine demonstrated cross-protection against persistent infection with HPV-31, -33 and -45 of 78.7%, 45.7% and 75.7%(16) while the quadrivalent showed cross protection of approximately 46% against persistent infection with HPV-31.(44) Vaccination against HPV may, thus, provide extra protection against cervical cancer caused by non-vaccine HPV types, especially those vulnerable to infection by multiple HPV types.
Indeed, the increasing prevalence of HPV-16 and HPV-18 observed in young healthy Japanese women in this study may indicate that the incidence of cervical cancer in Japan caused by HPV-16/18 will increase in the future. Therefore, a greater emphasis needs to be placed on providing information on HPV and screening for oncogenic types of HPV, in particular HPV-16 and HPV-18. Safe and effective vaccination against HPV-16 and HPV-18 will help prevent the increase of HPV infection and subsequent development of cervical cancer. Furthermore, the HPV prevalence data reported here in healthy young Japanese women would be very useful for monitoring the changes in HPV prevalence after widespread HPV vaccination in Japanese women in the future.
The authors acknowledge all of the investigators, staff and women enrolled in the study. This study (104798, NCT00316693) was funded and coordinated by GlaxoSmithKline Biologicals, Rixensart, Belgium and GlaxoSmithKline K.K., Tokyo, Japan. We thank Fabian Tibaldi (GlaxoSmithKline Biologicals, Global Clinical Research and Development Department, Rixensart, Belgium) for the statistical analyses, Nobuhiro Noro, Atsushi Maruyama and Marie Okutani (all from GlaxoSmithKline K.K, Vaccine Clinical Development, Tokyo Japan) contributed to data interpretation and critical reading, Claire Marie Seymour for writing assistance and Dirk Saerens and Denis Sohy for editorial assistance and manuscript coordination on behalf of GlaxoSmithKline Biologicals. Cervarix is a registered trademark of the GlaxoSmithKline group of companies. Gardasil is a registered trademark of Merck & Co. Aimmugen is a registered trademark of The Chem-Sero-Therapeutic Research Institute.
Ryo Konno and Hiroyuki Yoshikawa have served on advisory boards for GlaxoSmithKline K.K., and received lecture fees from GlaxoSmithKline K.K.. Shinobu Tamura is employee of GlaxoSmithKline K.K.. Kurt Dobbelaere is employee of GlaxoSmithKline Biologicals and has stock ownership of GlaxoSmithKline.
atypical glandular cells
atypical squamous cells, cannot exclude HSIL
atypical squamous cell of undetermined significance
cervical intraepithelial neoplasia
DNA enzyme immunoassay
high-grade squamous intraepithelial lesion
line probe assay
low-grade squamous intraepithelial lesion
polymerase chain reaction
total vaccinated cohort