Prevalence of Cervical Human Papillomavirus (HPV) Infection in Vanuatu Free

Oceania is a very diverse continent with respect to the burden of cervical cancer (1, 2) and of its cause, that is infection with high-risk (HR) human papillomavirus (HPV) types (3). Cervical cancer age-standardized incidence rates are low (below 8 in 100,000) in well-screened populations in Hawaii, Australia, and New Zealand, but they are 15.6 in 100,000 in French Polynesia, the only population-based cancer registry in lower resource Oceanic islands (1). Approximate estimates of cervical cancer incidence suggested the existence of substantially more elevated rates in other island nations (e.g., 31.4 in 100,000 in Fiji; ref. 2).

The Republic of Vanuatu is located in the South Pacific Ocean and consists of more than 80 islands including the most populous, Efate, where the capital Port Vila is situated. Vanuatu has a fast growing population estimated to be 221,000 in 2007. Eighty percent of the population lives in rural, isolated villages that have minimal access to basic health care and education services. Christianity is the predominant religion but many indigenous beliefs survive, and Vanuatu culture retains a strong diversity through local regional variations and through foreign influence. Teenage pregnancies and arranged marriages are common.

The age-standardized incidence rate of cervical cancer in Vanuatu was estimated to be approximately 17 per 100,000 in 2008, but cancer ascertainment is incomplete and mortality data not available (2). Pathology, radiotherapy, and palliative care are not available and the use of traditional remedies (mainly plants) is frequent. With respect to HPV prevalence, a study of different screening modalities in 488 women aged between 30 and 50 years based on hybrid capture 2 (HC2) was carried out in Port Vila in 2006. It showed an overall prevalence of HR HPV types of 9.0% (4).

The aims of our present new survey of HPV prevalence in Vanuatu were to fill some knowledge gaps with respect to HPV prevalence in a broader age range, notably young women, and the distribution of individual HPV types in women with and without cervical lesions. This information was not available from the study by McAdam and colleagues (4) and is deemed to be essential to introduce and monitor the impact of HPV vaccine (5). We, therefore, conducted a new survey in Vanuatu using a well-validated PCR assay for HPV testing and HC2.

Between February 2009 and January 2010, a survey was conducted by the Ministry of Health of Vanuatu in collaboration with the University of Queensland (Brisbane, Queensland, Australia), and the International Agency for Research on Cancer (IARC; Lyon, France). Women between ages 18 and 64 were invited to join the present survey by poster and flier advertisements, and nurse “awareness” visits to villages around Santo Urban, Espiritu Santo Island, and Port Vila, Efate Island (the first and third largest islands in Vanuatu, respectively). Support to our present study was asked to community leaders, including religious leaders and chiefs, and radio publicity was also used in Port Vila. Women who were pregnant or had undergone gynecologic surgery were not eligible. A total of 998 women visited 6 study clinics located in hospitals and family planning organizations in Port Vila (887) and Santo Urban (111) and provided written consent. They underwent an interview on lifestyle factors and medical history and a gynecologic examination. Although there were very few refusals among women who were personally invited, participation rates are difficult to estimate due to the uncertainty on the number of women who had been reached by information about the present study. The age distribution of participating women, however, was very similar to the age distribution of Vanuatu population according to the last census survey except for a slight excess of participation among women aged between 25 and 34 years of age (43.9% in our study compared with 31.4% in Vanuatu female population) and a lack among women aged 45 years or older (11.2% vs. 20.9%, respectively).

Women underwent a direct light unmagnified speculum examination of the cervix. Trained nursing staff collected exfoliated cervical cells with Cervex-Brush (Rovers Medical Devices). Cells were placed in a vial containing PreservCyt media (Cytyc-Hologic) and used for liquid-based cytology and HPV testing. Cytology was read at the Victorian Cytology Service according to the Australian Modified Bethesda System 2004 (6). Briefly, in the Australian System, atypical squamous cells of undetermined significance and atypical squamous cells, possible high-grade lesions, are classified, respectively, as low-grade squamous intraepithelial lesions (LSIL) and high-grade squamous intraepithelial lesions (HSIL). Cytology samples from 152 women were inadequate mainly due to insufficient cell yield.

Women with visible abnormalities of the cervix suggestive of malignancy, HPV-positive findings according to HC2 or HSIL cytology, were recalled to have a colposcope-aided examination and, if appropriate, a colposcopically-directed biopsy and treatment (cryotherapy or loop electrosurgical excision of the cervix transition zone) by local gynecologists. Among women younger than 30 years, only HPV-positive HSILs were recalled. Histology examinations were conducted by accredited Australian diagnostic pathology providers (4).

The present study had the approval of the Parliament of Vanuatu and the Human Experimentation Ethics Committee of the University of Queensland and of the IARC Ethics Committee.

HPV testing and genotyping were conducted at the VU University Medical Centre (Amsterdam, The Netherlands) according to a protocol similar to that used in previous IARC HPV Prevalence Surveys (7, 8). DNA was extracted from the PreservCyt sample using magnetic beads (Macherey-Nagel) on a robotic system (Hamilton Robotics) according to the manufacturer's instructions. β-Globin PCR analysis was conducted first to confirm the presence of human DNA in all specimens (9). The presence of HPV DNA was determined by conducting a general primer GP5⁺/6⁺-mediated PCR, which permits the detection of a broad spectrum of genital HPV types at the subpicogram level (10). HPV positivity was assessed by hybridization of PCR products in an enzyme immunoassay with 2 oligoprobe cocktails that, together, detect the following 44 mucosal HPV types: HPV6, 11, 16, 18, 26, 30, 31, 32, 33, 34, 35, 39, 40, 42, 43, 44, 45, 51, 52, 53, 54, 55, 56, 57, 58, 59, 61, 64, 66, 67, 68, 69, 70, 71, 72, 73, 81, 82, 83, 84, 85, 86, 89, and 90. Subsequent HPV genotyping was conducted by reverse-line blot hybridization of GP5⁺/6⁺ PCR products (11). HPV types considered HR types for this analysis comprised HPV16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68; all other HPV types were considered low-risk. Eleven women were excluded because their samples were β-globin negative, leaving 987 (98.9%) women with valid PCR-based HPV findings (of whom 840 had adequate cytology).

HR HPV testing (Digene HC2 High-Risk HPV DNA testing from QIAGEN) was conducted according to manufacturer's instructions in the Hematology Laboratory, Vila Central Hospital, Port Vila, Vanuatu. The performance of HC2 results was not possible for 75 women due to temporary malfunctioning of HC2 equipment. HC2 findings were therefore available for 912 (92.4%) women who also had PCR-based HPV findings.

Age-standardized HPV prevalence was computed to allow comparisons with other IARC surveys (7). Age-specific HPV prevalence for 5-year age groups was applied to the world standard population, as routinely done for cancer incidence and mortality rates (1). Odds ratios (ORs) for HPV positivity and corresponding 95% confidence intervals (CI) were computed by unconditional logistic regression adjusted for (i) age group (≤24; 25–29, 30–34, 35–39, 40–44, and ≥45 years) or (ii) age group and all variables significantly associated to HPV positivity in the age-adjusted model. Risk trends were assessed by considering variable categories as continuous variables. The agreement made between HC2 and the PCR assay with respect to 13 HR HPV types included in HC2 was evaluated with κ statistics and by computing the sensitivity and specificity of HC2 compared with the PCR assay.

The prevalence of all HPV types according to the PCR assay was 28.4% (95% CI, 25.6%–31.3%) corresponding to an age-standardized prevalence of 25.0% (95% CI, 21.9%–28.0%; Table 1). The prevalence of HR HPV types was 21.7% (95% CI, 19.1%–24.4%). On account of the large number of frequently HPV-positive young women in the present study, the corresponding age-standardized prevalence was slightly lower (19.2%; 95% CI, 16.4%–22.0%) than the crude prevalence. Multiple HPV types accounted for 37.1% of HPV-positive women. Among 840 women with adequate cytologic results, cervical abnormalities were diagnosed in 114 (13.6%) women, of whom 71.9% were HPV-positive. Cytologic abnormalities included 80 LSIL and 34 HSIL or worse (HSIL⁺). All women with HSIL⁺ and an additional 39 women who were positive at HC2 but an inadequate cytology or LSIL, were recalled. Among 39 women (53.4%) who had a biopsy, histologic findings were normal in 16; cervical intraepithelial neoplasia (CIN) grade 1 in 7; CIN2 in 3; CIN3 in 8 women, and invasive cervical carcinoma in 5 (including a woman with an adenocarcinoma). In addition, invasive cervical carcinoma was clinically detected in 2 women with inadequate cytology and HC2-negative findings. HPV16 was confirmed as the most common type (9.0% overall; Table 1). Among HR types, HPV18, 51, 52, 56, and 59 were each found in more than 2% of women. Among low-risk types (detected in 13.7% of all women, detailed distribution available in Supplementary Appendix), HPV6 and 11 were rare.

Table 2 shows that the proportion of HPV16 and/or 18 among women positive for HPV by PCR assay substantially increased with the severity of cervical lesion (ORs for HSIL⁺ vs. normal cytology = 2.80; 95% CI, 1.25%–6.28%). All women with invasive cervical carcinomas were HPV-positive by PCR assay: 4 for HPV16, and 1 each for HPV18, 35, and a combination of HPV30 and 52.

Figure 1 shows the age-specific prevalence of HPV, classified hierarchically into (i) HPV16 and/or 18, (ii) other HR types, and (iii) low-risk types only. HPV prevalence decreased from 46.1% among women aged ≤21 years to 15.3% in women aged ≥45. HR HPV type prevalence by HC2 showed a similar age pattern. The proportion of cytologic abnormalities was highest (17.4%) in women aged 25 to 29 years and then remained stable at about 11% from age 30 to 34 years onward (Fig. 1).

Table 3 shows the relationship between HPV positivity and various potential risk factors. HPV prevalence was significantly lower in rural Efate (OR, 0.55; 95% CI, 0.35%–0.88%) and, nonsignificantly, in Espiritu Santo, than in Port-Vila. OR for HPV positivity in women aged ≥45 compared with those ≤24 years was 0.25 (95% CI, 0.14%–0.45%) and the downward trend with age was highly statistically significant. Other significant risk factors for HPV positivity in the age-adjusted analysis included being a current smoker (OR, 1.77; 95% CI, 1.08%–2.91%), or single (OR, 1.46; 95% CI, 1.04%–2.04%) and reporting a high number of lifetime sexual partners (OR for ≥5 vs. 1 = 1.67; 95% CI, 1.10%–2.54%). There were no significant associations between HPV positivity and age at first sexual intercourse, number of births, and use of hormonal contraceptives. Age at first marriage and first birth were unrelated to HPV positivity. Multivariate analysis led to a slight attenuation (<20%) of all the significantly elevated ORs except that for being single (Table 3).

HR HPV positivity by PCR assay and HC2 were 21.2% and 16.2%, respectively, among 912 women for whom both tests were available. Seventy-five percent of women were double-negative and 12.5% were double-positive (Table 4). Seventy-nine (8.7%) women were PCR-positive but HC2-negative whereas 18 (2.0%) were PCR-negative but HC2-positive (κ test = 0.59, i.e., moderate agreement). Sensitivity and specificity of HC2 compared with the PCR assay were 59.1% and 95.3%, respectively (Table 4). Sensitivity and specificity were similar for different HR types. HPV53 (6 women) and 66 (3 women) predominated in HC2-positive women who were only infected by low-risk types according to the PCR assay (data not shown). The relative sensitivity of HC2 compared with the PCR assay improved to 62.5% when women with inadequate cytology were excluded on account of higher proportion of HC2-negative results among women with inadequate cytology.

Table 5 shows the prevalence of HSIL⁺ (corresponding to the positive predictive value) among women testing positive for one or more HR HPV type by PCR assay or HC2. Restricting our analysis to, respectively, 188 and 136 women with PCR-based or HC2-based positivity to HR HPV types, 61.2% and 52.2% of women, respectively, did not have cytologic abnormalities. Infection with HR HPV in women younger than 30 years was significantly more likely to be associated with LSIL than among older women (P < 0.001 for both HPV tests). Conversely, infection with HR HPV in women 30 years or older was significantly more likely to be associated with HSIL⁺ cytology (27.6% for PCR and 35.2% for HC2) than among younger women (P < 0.001 for both HPV tests; data not shown in tables). Although comparisons were based on small numbers, the positive predictive values of HR HPV positivity for HSIL⁺ among women aged 40 or older were 36.8% for PCR and 30.0% for HC2, not inferior to those among women aged 30 to 39 years (Table 5).

Our present survey disclosed high prevalence (28.4%) of HPV infection and HR HPV infections (21.7%) among the general female population in Vanuatu, notably among women younger than 25 years (41.8% and 31.7%, respectively). Few population-based surveys similar to our present one have been carried out in Oceania and none from the lowest resource islands included either women younger than 30 years or provided HPV type–specific distribution. This information is important to introduce and monitor the impact of HPV vaccine (5).

HPV prevalence had been previously reported in Hawaiian women (25.6%; ref. 12) and among non-Indigenous and Indigenous women in Australia (41.5% and 47.5%, respectively; ref. 13). Similarly to our survey, the highest HPV prevalences were detected in women younger than 25 years, but findings must be compared with caution due to the use of a PCR assay (PGMY09/PGMY11) that has higher analytic sensitivity than GP5⁺/6⁺ (12, 13). Conversely, HPV prevalence in the survey in Vanuatu can be confidently compared with that in the other IARC HPV Surveys in the framework of which the present study has been conducted. The age-standardized HPV prevalence in Vanuatu (25.0%) ranks as the third highest, after that in Guinea (51.5%; ref. 8) and Mongolia (36.7%; ref. 14). It is similar to the prevalence detected in Nigeria and more elevated than that found in other areas at high-risk for cervical cancer in Asia and South and Central America (7). We previously showed that HR HPV prevalence is strongly correlated internationally with cervical cancer incidence rates, especially among women aged 35 years or older (15), pointing to existence of a high burden of the disease in Vanuatu. This possibility is confirmed by the high prevalence of invasive cervical carcinoma (0.8%) in our study population.

Our present findings on the prevalence of individual HPV types among Vanuatu general female population contribute substantially to present knowledge on HPV infection in Oceania, as no information from the region could be included in the most recent meta-analysis of HPV type distribution among women with normal cytology (16). HPV16 was by far the most frequently detected type in Vanuatu (6.5% among women with normal cytology and 9.0% in all women). The next frequent HR types were HPV56, 52, 59, 51, and 18, with each a prevalence above 2.0%. Infections with vaccine types HPV16 and/or 18 accounted for 49.5% of HR HPV positivity in the general female population and they ranged between 48.1% among women younger than 25 years and 53.9% among women aged 45 or older. Among HPV-positive women, the prevalence of HPV16 or 18 was similar in women with normal cytology (32.9%) and LSIL (37.7%) but significantly higher in women with HSIL⁺ (56.3%).

Risk factors associated with cervical HPV infection in the age-adjusted analysis in our present study were consistent with previous IARC HPV Surveys, that is, high number of lifetime sexual partners (17), being single, or current smokers (18). In fact, 71.4% of women in our present study reported 2 lifetime sexual partners or more (mean = 3.1). The corresponding proportion was lower in previous IARC HPV Surveys (8, 14, 17). In agreement with previous studies, high parity and age at first sexual intercourse and first birth were unrelated to HPV positivity (19). In a multivariate analysis that included all risk factors significantly associated to HPV positivity in the age-adjusted analysis, the only association that was not weakened and remained statistically significant was the one with being single. In fact, we discovered a complex relationship between age, marital status, and lifetime sexual partners. Multiple partners were frequently reported among women younger than 30 years regardless of their marital status, whereas among older women, high number of sexual partners was concentrated among noncurrently married women. This interesting finding shows that the presence of high-risk sexual behaviour in Vanuatu is concentrated among the young generations and among older women who do not have a husband.

Of note, McAdam and colleagues (4) reported 9% HR HPV prevalence based on HC2 testing of 488 women aged between 30 and 50 years from Port Vila, Vanuatu. Women had been recruited in a way similar to that used in the present study. HR HPV prevalence in the study by McAdam and colleagues was not significantly lower than HC2-based findings in our present study in the same age groups (12.0%).

Our present survey also confirmed a substantial burden of precancerous and cancerous cervical lesions in the general female population of Vanuatu (4). Not only the prevalence of HSIL⁺ in our present study was much higher than in well-screened populations in Western countries (20) but also it tended to increase steadily with a woman's age. Among women aged 30 years or older, 27.6% of HR HPV–positive women by PCR had a concurrent HSIL⁺ and the positive predictive value of the PCR assay or HC2 was very high also among women aged 40 years or older. High positive predictive value in middle-age women has been also reported in sub-Saharan Africa (21) and represents a strong incentive not to exclude women older than 40 or even 50 years from at least the first screening round in high-risk and previously unscreened populations. The choice of HPV-based primary screening is especially valuable in middle-aged women in low-resource countries on account of the loss on sensitivity of visual inspection techniques as the squamous–columnar junction withdraws in the endocervix (22, 23).

Of note, we were able to compare the performances of our PCR assay done in The Netherlands and HC2 done in Vanuatu. We found a fairly good agreement. HC2 was, as expected, less sensitive than PCR (24, 25). Shortcomings of locally conducted HC2 in our present study included a temporary malfunctioning of HC2 equipment and the relatively high proportion of HC2-negative findings in PCR-positive women in the presence of inadequate cytology.

Our present survey has strengths and weaknesses. Strengths include the relatively large number of women, especially those younger than 30 years. Our findings allow to establish the baseline prevalence on HPV types in the anticipation of a possible introduction of HPV vaccination in Vanuatu. The use of a high-quality, clinically validated PCR assay conducted in a central laboratory adds to our study value in terms of comparability with other studies, notably previous IARC HPV Surveys. The extent to which our study women are representative of the general female population in Vanuatu is difficult to evaluate. However, HPV infection is asymptomatic and the knowledge on the topic was very limited in our study population. Self-selection of women by level of HPV risk, notably by sexual behavior, is therefore unlikely to have occurred.

A severe limitation of our study is that, despite repeated efforts, it has not yet been possible to reexamine nearly half of the women with positive HC2 findings or HSIL and provide them with appropriate histologic confirmation and treatment. Compliance to recall was poor among Vanuatu women and addresses and mobile numbers were often unreliable. Difficulties to follow-up screening-positive women, together with the clear urgency of cervical prevention measures in such a high-risk population, point to the strong need for access to a rapid HPV testing method that allows see-and-treat approaches at the same visit.

P.J.F. Snijders has Honoraria from Speakers Bureau from Roche and is a consultant/advisory board member for Roche and Gen-Probe. C.J.L.M. Meijer has Honoraria of Speakers Bureau from GlaxoSmithKline and is a consultant/advisory board member for Qiagen. I.H. Frazer has received royalties from the sale of HPV vaccines from CSL, Merck, and GSK. No potential conflicts of interest were disclosed by the other authors.

Conception and design: L. Tarivonda, G. Clifford, M. McAdam, I.H. Frazer, S. Franceschi

Development of methodology: L. Tarivonda, M. McAdam, I.H. Frazer

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): B. Aruhuri, L. Tarivonda, P.J.F. Snijders, J. Pang, C.J.L.M. Meijer

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): V. Tenet, R. Sinha, J. Pang, I.H. Frazer, S. Franceschi

Writing, review, and/or revision of the manuscript: L. Tarivonda, M. McAdam, C.J.L.M. Meijer, I.H. Frazer, S. Franceschi

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): L. Tarivonda, J. Pang, M. McAdam

Study supervision: R. Sinha

The authors thank the women who participated to the present study in Vanuatu and the staff at Victorian Cytology Service for reading liquid-based cytology, Qiagen for the donation of the Digene HC2 High-Risk HPV DNA testing kits, and Mark Van Asten for the donation of the HC2 testing machine and technical support.

This work was supported by grants from the National Health and Medical Research Council of Australia, the Balzan Foundation, Australians Helping Abroad, and the Bill & Melinda Gates Foundation (grant number 35537). I.H. Frazer was funded from a Queensland Government Premier's Fellowship and is an NIH grant recipient.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.