Article Text

Short report
Seroprevalence of 10 human papillomavirus types in the general rural population of Anyang, China: a cross-sectional study
  1. Chanyuan Zhang1,
  2. Fangfang Liu1,
  3. Qiuju Deng1,
  4. Zhonghu He1,
  5. Longfu Xi2,3,
  6. Ying Liu1,
  7. Yaqi Pan1,
  8. Tao Ning1,
  9. Chuanhai Guo1,
  10. Ruiping Xu4,
  11. Lixin Zhang4,
  12. Hong Cai1,
  13. Yang Ke1
  1. 1Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education), Laboratory of Genetics, Peking University Cancer Hospital & Institute, Beijing, People's Republic of China
  2. 2Department of Epidemiology, School of Public Health, University of Washington, Seattle, Washington, USA
  3. 3Department of Pathology, School of Medicine, University of Washington, Seattle, Washington, USA
  4. 4Anyang Cancer Hospital, Anyang, Henan, People's Republic of China
  1. Correspondence to Hong Cai and Yang Ke, Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education), Laboratory of Genetics, Peking University Cancer Hospital & Institute, No. 52 Fucheng Road, Haidian District, Beijing 100142, People's Republic of China;,


Objectives Data on the seroprevalence of human papillomavirus (HPV) in China are limited. The objective of this study was to characterise the serological profiles of HPV infection in a rural Chinese population and help establish effective vaccine policy.

Methods Serum antibodies against the major capsid protein L1 of 10 HPV types (HPV-3, 6, 11, 16, 18, 45, 52, 57, 58 and 75) were evaluated with Luminex-based multiplex serology in a population-based study of 5548 adults (including 1587 couples) aged 25–65 years enrolled from rural Anyang, China, in 2007–2009.

Results The seroprevalence for any HPV type and any of the types HPV-6/11/16/18 was 64.8% and 34.4%, respectively. 30.3% of adults were seropositive for any mucosal high-risk (HR) HPV, and HPV-58 (10.6%), HPV-16 (9.7%) and HPV-18 (9.3%) were the three most common types. 24.8% of seropositive individuals were positive for multiple mucosal HR-HPV serotypes. Seroprevalence for most HPV types was similar among men and women. While mucosal low-risk HPV seropositivity was found to significantly decrease with age, the prevalence of antibodies to mucosal HR antigens showed a general trend of increase with age. The lifetime number of sex partners was independently associated with mucosal HR-HPV seropositivity. Positive correlation of spousal seropositivity was observed for mucosal HPV but not for cutaneous HPV.

Conclusions HPV infection was common in both men and women in rural China. HPV seroprevalence differed significantly with age, sexual behaviour and spousal infection status. These findings will be useful for evaluating and establishing HPV vaccination programmes.

  • HPV

Statistics from

Human papillomavirus (HPV) is the sexually transmitted aetiological agent of cervical and other anogenital cancers.1 HPV-6/11/16/18 vaccine can effectively reduce the risk of HPV-associated diseases.1 The structuring of HPV vaccination programmes requires population-based data on HPV epidemiology.

Epidemiological knowledge of HPV infection relies heavily on prevalence data of genital HPV DNA. DNA detection methods identify only current infections and are limited by small specimens of certain anatomical sites. Although a substantial proportion of people (∼50% of women) do not develop an antibody response after a natural infection, HPV antibodies are a useful tool for measuring cumulative exposure.2 Seroepidemiological studies on HPV have previously been conducted mainly in western countries, and HPV-16 is the most commonly observed type.3 However, most previous investigations had limited representativeness of the general population, involved women only or evaluated only small numbers of HPV types.

Knowledge about HPV seroepidemiology in Asia including China is especially limited. Before the vaccination programme gets started in China, this study will provide evaluation of HPV seroprevalence and associated risk factors in a large Chinese population.


Serum specimens and questionnaire data were collected from all participants (aged 25–65 years and without prior diagnosis of cancer, immunodeficiency disorders, cardiovascular illness or HBV/HCV/HIV) in six of the nine villages which were cluster-sampled in the baseline survey of a population-based oesophageal cancer cohort in rural Anyang, China, in 2007–2009.4 ,5 All participants provided written informed consent and the study protocol was approved by the Institutional Review Board of the School of Oncology, Peking University.

Serum antibodies to 10 HPV types were tested using a bead-based glutathione S-transferase (GST)-L1 multiplex serology assay, using a Bio-Plex 200 system (BIO-RAD, USA).4 ,6 The median fluorescence intensity (MFI) values for each HPV antigen were calculated by subtracting the MFI for beads coated with GST alone.

Seropositivity threshold MFI levels for mucosal high-risk (HR)-HPV (HPV-16, 18, 45, 52, and 58) and mucosal low-risk (LR)-HPV (HPV-6 and 11) were defined using data from 123 women (18–24 years of age) who reported no history of sexual intercourse. The cut-off was set at three SDs above the mean level of MFI reactivity among these women, excluding positive outliers. A cut-off of 200 MFI was set to define seropositivity for cutaneous HPV (HPV-3, 57, and 75).

Logistic regression models were employed to assess risk factors for HPV seropositivity and p<0.05 (two-sided) was considered to be significant.


Of 7362 eligible candidates, 5548 (median age: 42 years; male–female ratio: 0.9) were included in the analysis (other 1814 candidates, more likely to be younger men, were not included because of nonparticipation or inadequate serum) (see online supplementary table S1). Seropositivity for any HPV type was 64.8% and 34.4% for HPV-6, 11, 16 or 18. Overall seroprevalence of mucosal HR-HPV, mucosal LR-HPV and cutaneous HPV was 30.3%, 22.5%, and 41.5% respectively. The most common HR-HPV type was HPV-58 (10.6%), followed by HPV-16 (9.7%) and HPV-18 (9.3%). Of the seropositive individuals, 24.8% were positive for multiple (≥2) HR-HPV serotypes.

HPV seroprevalence among men was slightly higher than that among women (66.1% vs 63.6%, p=0.044) (see online supplementary figure S1). For mucosal HR-HPV, overall seropositivity increased with age (ptrend=0.001), characterised by the highest peak at 61–65 years (37.4%). Seropositivity of mucosal LR-HPV declined with age (ptrend<0.001).

Subjects reporting more sexual partners had a higher risk of seropositivity for mucosal HR-HPV (adjusted OR=1.39, 95% CI 1.03 to 1.88; ≥3 partners vs 1 partner) (table 1). Working outside of the local area increased the risk of seropositivity for mucosal LR-HPV as compared with working locally (adjusted OR=1.39, 95% CI 1.14 to 1.69). After stratification by gender, all these associations remained in men but disappeared in women (see online supplementary table S2 and S3).

Table 1

HPV seroprevalence and associated risk factors in rural Anyang, China, 2007–2009*

Of 1587 couples where both partners provided sera, overall mucosal HPV sero-status between spouses was correlated (adjusted OR=1.25, 95% CI 1.02 to 1.53). In contrast, no such association was found for cutaneous HPV (table 2, see online supplementary table S4).

Table 2

Association of HPV seropositivity in 1587 heterosexual couples in rural Anyang, China, 2007–2009


This is the first large-scale investigation of population-based HPV seroprevalence in both Chinese men and women. Using high-throughput multiplex serology, antibodies against L1 for the 10 most locally prevalent HPV DNA types were measured simultaneously.

The study demonstrated that HPV infection was common in rural China. Seropositivity in this study was comparable with that in studies detecting overall antibodies in similar populations,3 but was much higher than in one recent study of Chinese women using a competitive Luminex immunoassay (cLIA).7 As cLIA measures only protective neutralising antibodies, cLIA-based estimates of seroprevalence which indicate past exposure should be much lower.

HPV-16 and HPV-18, together with HPV-58 which is particularly prevalent in Asia, were the three most common HR serotypes observed. HPV-16 DNA predominates over other HR types in cancer-free women from China and elsewhere.8 However, this predominance may not necessarily be reflected in seroprevalence, for which HPV-16, 18, and 58 showed equivalent values in this study. A possible interpretation for this difference in type distribution at the serological and DNA levels may be due to type-specific variations in persistence of infection or ability to induce a serological response. Since HPV-58 contributes markedly to cervical carcinogenesis in China,8 our findings argue that inclusion of HPV-58 in future vaccination programmes for the Chinese population is particularly important.

In contrast with the well-known lower seroprevalence for mucosal HPV found in men as compared with that in women,9 higher male seroprevalence was observed here. This gender-related difference in HPV antibodies is of uncertain causes. However, considering much higher mucosal DNA prevalence in penile versus cervical specimens from the same population (10.8% vs 6.0%, data not shown), together with the lower propensity of men to induce antibody response,9 it is possible that more exposure or susceptibility to HPV among men may exist in this population. This assumption is also supported by our observation that men reported more sexual partners than women. As heterosexual activity is an important route of HPV transmission, inclusion of men in HPV vaccination programmes may be effective in preventing infections in both men and women.

In this study, overall seropositivity of mucosal HR-HPV increased with age. This trend, which has previously been reported,7 may reflect the facts that cumulative exposure measured by HPV seropositivity is higher among older individuals and that antibodies to HR-HPV types tend to be persistent. Chinese women demonstrated peaks of HR-HPV DNA prevalence in both early 20s and early 40s.8 Peak seroprevalence of HR-HPV in this population occurred after the secondary HPV DNA prevalence peak, similar to previous reports in China.7 In contrast with HR-HPV, the seroprevalence of LR-HPV decreased with age. According to Carter, antibodies to LR-HPV types, which tend to decay within 2 years after clearance of infection, seem to be less stable over time compared with antibodies to HR-HPV types.2

Similar to previous studies of HPV DNA and antibodies, greater numbers of lifetime sexual partners increased the risk of mucosal HR-HPV seropositivity.2 ,7 However, for mucosal LR-HPV, its association with multiple sexual partners did not approach statistical significance. This may be due to the insufficient power resulting from underreporting of sexual partners, although efforts were made to minimise the under-reporting bias (eg, conducting interviews in private settings). Nevertheless, we found that working outside the home region, an indicator of more risky sexual behaviours,10 increased the risk of seropositivity for mucosal LR-HPV as compared with working locally. When stratified by gender, all the above associations held true in men but not in women. Given the relatively conservative sexual behaviour among women in rural China and the greater possibility for underreporting of sexual partners, a lack of association between HPV seroprevalence and sexual behaviour indicators in women can be explained. These findings taken together confirm that mucosal HPV seropositivity is associated with a higher number of sexual partners.

Notably, this study is the first one to evaluate the spousal concordance for HPV seropositivity. Our results showed that mucosal HPV seropositivity in one partner increased the risk of mucosal HPV seropositivity in the other partner. In contrast, no such spousal correlation was found for cutaneous HPV. This result, together with the absence of association between cutaneous HPV seropositivity and sexual behaviours, argues that sexual routes are not likely to play a major role in cutaneous HPV transmission.

This study has several limitations. First, the biases potentially imposed by the nonparticipation and exclusion of younger men must be noted, which might have diluted the age-related/gender-related association. Second, as not all infected persons seroconvert, it is not fully known what viruses have been harboured in seropositive individuals, and thus serological data are inherently limited for interpreting cumulative exposure. Furthermore, cross-reactivity of serotypes cannot be ruled out.

In conclusion, HPV infection was common in both men and women in rural China. Mucosal HR-HPV seropositivity increased with age and showed significant association with the number of sexual partners. Positive spousal correlation of seropositivity was observed for mucosal HPV but not for cutaneous HPV in couples. Despite its limitations, this is the largest study characterising serological profiles of HPV in the general Chinese population to date. These findings will be useful for planning and monitoring HPV vaccination programmes.


The authors thank Professor Denise A. Galloway and Dr Joseph J. Carter (Fred Hutchinson Cancer Research Center) for technical help and providing the DNA constructs of pGEX.HPV-16 L1, pGEX.HPV-52 L1, and pGEX.HPV-58 L1. The authors thank Dr Ethel-Michelle de Villiers (Reference Center for Papillomaviruses) for providing the plasmids of HPV-3, 6, 11, 18, 45, 57 and 75; Dr Longfu Xi (University of Washington) for providing the sera from HPV vaccine recipients; and Yong Li, Yuqin Song, Xiuyun Tian, Huirong Ding, Chunfeng Zhang, Yue Zhou, Wenjun Yang, Li Zheng, Fang Lu, Xueqian Wang, Yanyan Zhang, Yiqiang Zhao, Fangcen Guo, Ke Chen, Lei Gao, Min Sun and Luyan Shen who contributed to the field work. The authors also thank Michael A. McNutt for editing and correcting this manuscript.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

  • Abstract in Chinese

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • CZ and FL contributed equally.

  • Handling editor Jackie A Cassell

  • Contributors CZ and FL collaborated in the writing of the manuscript. YK and HC were involved in the design and conducting of the survey. ZH, FL, YL, YP, TN, CG, QD, RX and LZ were involved in conducting the field work and collecting the samples. CZ and QD operated the serological experiments. CZ and FL performed the statistical analyses. ZH, LX and HC revised the manuscript before submission.

  • Funding This work was supported by Natural Science Foundation of China [30872937, 30930102]; ‘973’ Project of National Ministry of Science and Technology Grant [2011CB504300, 2012CB910800]; Charity Project of National Ministry of Health [201202014, 200902002]; and Natural Science Foundation of Beijing [7100001] to Y.K.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval Institutional Review Board of the School of Oncology, Peking University, China.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.