CANCER OF THE UTERINE cervix is the most common form of cancer in women residing in developing countries, with an annual global incidence of 371,000 cases per year.1 Human immunodeficiency virus (HIV)-infected women are at increased risk for developing invasive cervical cancer, and the United States Centers for Disease Control and Prevention (CDC) included cervical cancer as an acquired immune deficiency syndrome (AIDS)-defining illness in the 1993 AIDS case definition.2,3 Cervical cancer was detected in 2.3% of HIV-infected women in Europe4 and in 3.9% of HIV-infected women in the United States.3 The rate of cervical cancer among HIV-infected women in Thailand as reported by the Ministry of Public Health (MOPH) from 1994 through 1998 was much lower, at 0.2%.5 Similarly, cervical cancer did not seem to be associated with HIV infection in Africa.4 Early mortality among HIV-infected women in resource-constrained settings and limited screening services may explain why cervical cancer has not been recognized as an HIV-associated malignancy in those parts of the world.
The primary cause of cervical cancer is infection with high-risk types of human papillomavirus (HPV).6 Considerable effort is devoted to develop both prophylactic7 and therapeutic vaccines to prevent the development of HPV-associated cervical cancer.8 A prophylactic quadrivalent HPV vaccine targeting HPV types 6, 11, 16, and 18 effectively prevented acquisition of infection and clinical disease during 35 months’ follow-up in a placebo-controlled trial.7 This vaccine would be cost-effective in combination with cervical cancer screening, as shown in an economic model,9 and may also offer an opportunity to reduce the risk of invasive cervical cancer among HIV-infected women. However, more data are needed on the prevalence of HPV types in relation to HIV disease to estimate the potential impact of a prophylactic or therapeutic HPV vaccine on cervical cancer. In this study, we assessed the HPV prevalence and distribution of different HPV types among HIV-infected pregnant women in Thailand. Risk factors for HPV detection were studied, including HIV viral load in the genital tract and plasma.
Our study population consisted of women participating in a trial to assess short-course zidovudine for the prevention of mother-to-child HIV transmission. This study has been described in detail elsewhere.10 In short, from May 1996 to December 1997, 1140 HIV-infected pregnant women were screened for enrollment at 2 Bangkok hospitals serving a population of low socioeconomic status. A total of 397 women were eligible and were randomized between placebo and zidovudine, 300 mg orally twice daily from 36 weeks’ gestation until onset of labor and 300 mg every 3 hours orally from onset of labor until delivery. Study nurses collected information during face-to-face interviews, including sociodemographic characteristics, risk factors, and HIV-related symptoms according to the World Health Organization clinical-staging system.11
Blood was drawn at scheduled visits at 36 weeks of gestation for determination of plasma HIV viral load and lymphocyte phenotyping. Cervicovaginal lavage (CVL) samples were obtained at scheduled visits at 38 weeks of gestation from 319 of the 397 women; most (87.2%) women who did not have a CVL collection delivered before 38 weeks of gestation.
CVL samples were obtained according to procedures of the DAIDS Virology Manual for HIV Laboratories13 and as previously described.12 Separate supernatants and pellets were prepared from CVL samples 6 months after the randomization started; samples were available from 256 (80.3%) of 319 women and were stored at −70 °C. CVL cell pellets were transported to Columbia University for HPV analysis.
Maternal HIV viral load in plasma and CVL was determined by using the Amplicor HIV-1 Monitor Test, version 1.5 (Roche Diagnostic Systems, Branchburg, NJ).12 Newborn’s HIV-infection status was tested by Amplicor HIV-1 polymerase chain reaction (PCR) (Roche Diagnostic Systems), and lymphocyte phenotyping was done with the FACScan flow cytometer (Becton Dickinson Immunocytometry Systems, San Jose, CA) as previously described.10
HPV DNA Detection
CVL cellular material was suspended in phosphate-buffered saline, and DNA was isolated using the QIAamp DNA Blood and Cell Culture Mini Kit according to the manufacturer’s instructions (Qiagen Inc., Chatsworth, CA). DNA was eluted with sterile water, and a 10-μl aliquot of the purified DNA was then analyzed utilizing the PCR–based Roche Line-probe Assay (kindly provided by Dr. Janet Kornegay, Roche Molecular Diagnostics).14 This PCR-based assay utilized the PGMY09/11, β-globin primers, and AmpliTaq gold DNA polymerase as described by Kornegay et al.15 PCR products were run on an agarose gel and classified as HPV positive if a band of the correct size was identified. All samples with a band corresponding to HPV DNA were subsequently genotyped using the reverse line-blot detection system which detects 27 HPV types plus the human β-globin gene.16
Cervical smears for cytology were collected 4 to 6 weeks after delivery. The Bethesda system was used for classification into normal, atypical squamous cells of undetermined significance (ASCUS), low-grade squamous intraepithelial lesion (LSIL), and high-grade squamous intraepithelial lesion (HSIL) categories.17 Appropriate diagnosis and treatment of cervical pathology was performed as part of routine medical care.
Medical records from all women in our study were reviewed for postpartum cervical cytology and contact information was abstracted from these records.
Women testing positive for at least 1 high-risk HPV type were considered to be at higher risk for developing cervical dysplasia compared to women without high-risk HPV types.18,19 These women were contacted and referred for cervical cancer screening. Information about HPV and cervical cancer was provided, and a cervical smear was obtained for cytology at the outpatient gynecologic clinic. Expenses, including transportation, cervical cytology, colposcopy, and treatment, were reimbursed.
Data Collection and Analysis
Possible associations between women’s characteristics and HPV detection, high-risk HPV types, and detection of multiple HPV types were investigated. For univariate analysis of the potential risk factors of HPV detection, 2-sided t tests, χ2 tests, and odds ratios (OR) with 95% confidence intervals (CI) were calculated; significant (P <0.05) factors from univariate analyses were subsequently included in multivariate models using logistic regression. The Mantel-Haenszel χ2 test was used for trend analysis.
The original protocol and this protocol amendment for HPV testing were reviewed and approved by a CDC institutional review board and the Ethical Review Committee of the Thai Ministry of Public Health.
All women who participated in the 1996 randomized clinical trial of zidovudine gave written informed consent for storing biologic specimens for future testing.
Samples from 256 HIV-infected pregnant women were available for HPV DNA analysis, with a median age of 25 years (range, 17–39 years). Less than half of the women (45.9%) had completed junior high school, most women (70.0%) were nullipara, reported 1 (38.7%) or 2 (37.9%) lifetime sex partners, and some (9.4%) reported a history of sex work. One hundred twenty-eight (50.0%) women were randomized to receive zidovudine and 127 (50.0%) to receive placebo.
HPV Prevalence by Type
Testing of CVL samples was performed between February 2003 and June 2003. All 256 CVL samples were positive for the β-globin PCR and were considered appropriate for HPV DNA analysis. The overall prevalence of HPV was 35.5% (91/256) (95% CI, 29.9%–41.6%). The prevalence of high-risk HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68)20 was 23.4% (60/256) (95% CI, 18.6%–28.9%). The most prevalent HPV types were HPV types 16 (n = 11), 18 (n = 10), 39 (n = 14), 51 (n = 10), 52 (n = 12), and 53 (n = 11) (Table 1). The prevalence of infection with either HPV type 16 or 18 was 8.2% (21/256). Fifteen of 91 PCR-positive samples showed a band of the correct size on the agarose gel but could not be typed by the line-blot hybridization. Seventy-six women had HPV-positive samples that could be typed, of whom 43 (56.6%) women had 1 HPV type and 33 (43.4%) women had multiple HPV types (Table 1). Multiple infections consisted of 2 HPV types (n = 19), 3 HPV types (n = 8), 4 HPV types (n = 3), or 5 HPV types (n = 3).
No difference was found between the prevalence of high-risk versus low-risk HPV types in samples with single or multiple HPV types (P = 0.8). Of the 43 samples with single HPV types, 29 contained high-risk HPV types. Of the 33 samples with multiple HPV types, 30 contained at least 1 high-risk HPV type and 18 contained at least 2 high-risk HPV types. At least 1 other high-risk HPV type was detected together with HPV-16 in 63.6% (7/11) and with HPV type 18 in 50.0% (5/10) of infections.
Demographic and Sexual Behavior
The mean age was 25.4 years and 24.6 years for pregnant women with HPV-negative samples and HPV-positive samples, respectively (P = 0.2). The HPV prevalence was similar for different age groups (Table 2) and for education level. The mean age at first sexual intercourse was 19.6 years for women with HPV-negative samples and 19.3 years for women with HPV-positive samples (P = 0.5). History of sex work was associated with detection of any HPV type (OR, 2.3; 95% CI, 1.0–5.5) (Table 2) and of high-risk HPV types (OR, 3.5; 95% CI, 1.5–8.6) but not of multiple HPV types (OR, 0.6; 95% CI, 0.2–2.1). Among women without a history of sex work, a median of more than 3 lifetime sexual partners was a risk factor for detection of any HPV type (OR, 2.1; 95% CI, 1.0–4.7) (Table 2) and high-risk HPV types (OR, 3.6; 95% CI, 1.5–8.9).
Medical records were reviewed from November 2003 through January 2004; cervical cytology results were available from 237 (92.6%) of 256 women. Of these 237 women, 222 (93.7%) had normal results and 15 (6.3%) had abnormal cytology results (3 with ASCUS, 7 with LSIL, and 5 with HSIL). Abnormal postpartum cervical cytology was associated with HPV detection (P <0.001) and high-risk HPV types (P <0.001). The HPV types 16, 18, 39, 53, 55, 58, 66, 68, mm7, mm821 were found in samples of women with abnormal cervical cytology (Table 3).
HIV Clinical and Laboratory Parameters
The presence of HIV-related symptoms was associated with detection of any HPV type (OR, 2.5; 95% CI, 1.1–5.4) (Table 4) and high-risk HPV types (OR, 2.4; 95% CI, 1.0–5.7) but not with multiple HPV types (OR, 0.7; 95% CI, 0.2–2.0). The association between HIV-related symptoms and HPV detection remained after controlling for HIV viral load (OR, 2.4; 95% CI, 1.1–5.3). Mother-to-child HIV transmission and 2 weeks of zidovudine treatment (zidovudine started after 36 weeks’ gestation and CVLs were collected at 38 weeks’ gestation) were not associated with HPV detection (Table 4).
The mean and median CD4-cell count was 445 cells/mm3 and 441 cells/mm3, respectively, for women with HPV-negative samples and 422 cells/mm3 and 371 cells/mm3, respectively, for women with HPV-positive samples (P = 0.3). Mean HIV viral load in plasma was 4.58 log10 copies/ml among women with HPV-positive samples and 4.34 log10 copies/ml among women with HPV-negative samples (P = 0.007). After adjusting for CD4-cell count, the association between HIV viral load and HPV detection remained statistically significant (P = 0.009). Plasma HIV viral load of more than 10,000 copies/ml was associated with detection of any HPV type (OR, 2.0; 95% CI, 1.0–3.8) (Table 4), high-risk HPV types (OR, 3.4; 95% CI, 1.4–8.6), and multiple HPV types (OR, 10.2; 95% CI, 1.3–218.0). Mean HIV viral load in CVL samples was 1.60 log10 copies/ml among women with HPV-positive samples and 1.50 log10 copies/ml among women with HPV-negative samples (P = 0.6). Mean HIV viral load of more than 10,000 copies/ml in CVL samples was not associated with detection of any HPV type (OR, 1.0; 95% CI, 0.4–2.4) (Table 4), high-risk HPV types (OR, 1.2; 95% CI, 0.5–3.2), or multiple HPV types (OR, 2.4; 95% CI, 0.5–11.8). No association was found between mean HIV viral load in CVL samples and HPV detection among women receiving placebo.
The 60 women with high-risk HPV types detected in their CVL samples were contacted for a follow-up visit. Of these 60 women, 15 had died, 24 could not be contacted, 4 women refused to have a pelvic examination, and 17 came for a visit to the gynecologic outpatient department for counseling and cervical cytology. Fifteen women had normal cytology results, and 2 women had HSIL. High-grade dysplasia was histologically confirmed for the 2 women with cytologic HSIL, and these women underwent a loop-excision procedure of the cervix. All women returning for a follow-up visit were subsequently referred for HIV care and treatment.
We found an overall HPV prevalence of 35.5% among the 256 HIV-infected pregnant Thai women in our study. Studies in other parts of the world showed higher HPV prevalence among HIV-infected women; a study in Brazil reported 98.1%22 and studies in the United States showed 54.2%18 and 67.8%23 HPV prevalence. Our antenatal-care cohort of Thai women may be different from women identified as HIV infected outside antenatal clinics or residing in other parts of the world. Many HIV-infected Thai women may acquire HIV infection from their husbands,24 and their relatively low-risk sexual behaviour probably results in a lower HPV prevalence. Low sensitivity of our assay seems an unlikely explanation for the low prevalence because the PCR assay that was utilized to detect and type HPV in this study is well established. Although samples had been stored for several years before testing, amplification of human DNA by β-globin PCR showed samples to be adequate for PCR analysis. However, selective inhibition of HPV DNA amplification cannot be ruled out.25 Pregnancy might have affected the HPV prevalence, although no difference was observed in a case-control study.26 Furthermore, the duration of HIV infection has to be considered as incident and persistent HPV infections result in a higher HPV prevalence.27 The duration of HIV infection for women who learned their HIV status during antenatal screening is likely to be shorter than for the women in the Brazilian cohort,22 of whom 79% were receiving antiretroviral treatment, allowing less time for incident and persistent HPV infections to occur.27
Other studies among Thai nonpregnant women showed a 19.2% HPV prevalence by the Hybrid Capture ViraType Plus assay among HIV-infected women28 and a 6.3% population-based HPV prevalence by PCR,29 supporting the hypothesis that the prevalence of HPV is lower in Thailand than in other parts of the world. A recently completed project among HIV-infected women seeking care for HIV or sexually transmitted infections showed a 38.6% prevalence of high-risk HPV types by the Hybrid Capture 2 HPV DNA assay (Digene Cooperation, Gaithersburg, MD).30 These women were probably infected with HIV for a longer time (mean duration of infection was 4 years) compared to women in our study, which may partially explain the difference in HPV prevalence.
Our results confirm the already-established association between HPV and abnormal cytology, although samples for HPV detection and cervical cytology were taken several weeks apart. We detected HPV type 52 only in samples of women with normal cytology and we were unable to confirm findings by others reporting HPV type 52 more often among women with abnormal cytology in Thailand28 and cervical cancer in Asia.20
In our study, no differences were found in mean age between women with HPV-positive and HPV-negative samples which is in accordance with earlier studies showing no decline in HPV prevalence with increasing age among HIV-infected women compared to women without HIV infection.31 As has been previously shown,32,33 history of sex work was a risk factor for HPV detection, reflecting sexual transmission as a transmission route for genital HPV types.
Plasma HIV viral load was strongly associated with HPV detection, which is consistent with earlier reports.22,34 Most likely, the duration of HIV infection was longer for women with higher viral loads, resulting in a greater likelihood of HPV detection. This supports the hypothesis of HPV reactivation during the course of HIV disease. In addition, we found no differences in HIV viral load in CVL samples among women with or without detectable HPV. A study among 47 HIV-infected women reported an association between higher HIV viral load in genital tract specimens and HPV detection, independent of immune suppression.35 Although we did not observe this finding in our study, future research might elucidate whether HPV is associated with HIV detection in the genital tract.
Women with HPV-positive samples were more likely to have had HIV-related symptoms, regardless of HIV viral load, compared to women with HPV-negative samples, supporting the hypothesis of an association between HPV detection and duration of HIV infection. The Thai MOPH is currently expanding its access to antiretroviral treatment for eligible HIV-infected persons (CD4 cell count less than 200 cells/mm3 or HIV-related symptoms),36 thereby prolonging the lifespan of HIV-infected women, with a likely increase in HPV and cervical cancer incidence.
Our HPV prevalence data are important for developing HPV vaccines and for planning HPV vaccine studies and immunization programs for HIV-infected women. The frequent occurrence of detection of multiple HPV types in our study has to be considered for vaccine development. Therapeutic HPV vaccines currently under development might be considered for women testing HPV positive. Moreover, HPV prevention would be useful for this population as the HPV prevalence is expected to increase during HIV disease.27 After confirming the relatively low HPV prevalence with direct cervical samples, prophylactic vaccines targeting HPV types 16 and 18 should be studied among women identified as HIV-infected during pregnancy without previous exposure to and testing negative for HPV types 16 and 18. As more than 90% of Thai women in antenatal care are screened for HIV infection as part of the national program to prevent mother-to-child HIV transmission,37 future HPV-immunization programs could be integrated in postpartum care to reduce the development of HPV-associated cervical cancer among HIV-infected women.
1. Parkin D, Pisani P, Ferlay J. Estimates of the worldwide incidence of 25 major cancers in 1990. Int J Cancer 1999; 80:827–841.
2. Centers for Disease Control and Prevention. 1993 Revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR Morb Mortal Wkly Rep 1993; 41:1–19.
3. Maiman M, Frucher R, Guy L, et al. Cervical cancer as an AIDS-defining illness. Obstet Gynecol 1997; 89:76–80.
4. Dal Maso L, Serraino D, Franceschi S. Epidemiology of AIDS-related tumours in developed and developing countries. Eur J Cancer 2001; 37:1188–1201.
5. Chariyalertsak S, Sirisanthana T, Saengwonloey O, et al. Clinical presentation and risk behaviors of patients with acquired immunodeficiency syndrome in Thailand, 1994–1998: regional variation and temporal trends. Clin Infect Dis 2001; 32:955–962.
6. Walboomers J, Jacobs M, Manos M, et al. Human papillomavirus is a necessary cause of invasive cervical cancer world-wide. J Pathol 1999; 189:12–19.
7. Villa LL, Costa RL, Petta CA, et al. Prophylactic quadrivalent human papillomavirus (types 6, 11, 16, and 18) L1 virus-like particle vaccine in young women: a randomised double-blind placebo-controlled multicentre phase II efficacy trial. Lancet Oncol 2005 6:271–278.
8. Galloway DA. Papillomavirus vaccines in clinical trials. Lancet Infect Dis 2003; 8:469–475.
9. Kulasingam SL, Myers ER. Potential health and economic impact of adding a human papillomavirus vaccine to screening programs. JAMA 2003; 290:781–789.
10. Shaffer N, Chuachoowong R, Mock PA, et al. Short-course zidovudine for perinatal HIV-1 transmission in Bangkok, Thailand: a randomised controlled trial. Lancet 1999; 353:773–780.
11. World Health Organization. Acquired immune deficiency syndrome (AIDS): interim proposal for WHO staging system for HIV infection disease. Wkly Epidemiol Rec 1990; 65:221–228.
12. Chuachoowong R, Shaffer N, Siriwasin W, et al. Short-course antenatal zidovudine reduces both cervicovaginal human immunodeficiency virus type 1 RNA levels and risk of perinatal transmission. J Infect Dis 2000; 181:99–106.
13. National Institutes of Health, National Institute of Allergy and Infectious Diseases, Division of AIDS. DAIDS Virology Manual for HIV Laboratories. Washington, DC: US Department of Health and Human Services; 1997. Publication NIH-97-3828.
14. Gravitt PE, Peyton CL, Alessi TQ, et al. Improved amplification of genital human papillomaviruses. J Clin Microbiol 2000; 38:357–361.
15. Kornegay JR, Roger M, Davies PO, et al. International proficiency study of a consensus L1 PCR assay for the detection and typing of human papillomavirus DNA: evaluation of accuracy and intralaboratory and interlaboratory agreement. J Clin Microbiol 2003; 41:1080–1086.
16. Gravitt P, Peyton CL, Apple RJ, et al. Genotyping of 27 human papillomavirus types by using L1 consensus PCR products by single-hybridization, reverse line blot detection method. J Clin Microbiol 1998; 36:3020–3027.
17. Solomon D, Davey D, Kurman R, et al. The 2001 Bethesda System: terminology for reporting results of cervical cytology. JAMA 2002; 287:2114–2119.
18. Ellerbrock TV, Chiasson M, Bush T, et al. Incidence of cervical squamous intraepithelial lesions in HIV-infected women. JAMA 2001; 283:1031–1037.
19. Cohn J, Gagnon S, Spence M, et al. The role of human papillomavirus deoxyribonucleic acid assay and repeated cervical cytologic examination in the detection of cervical intraepithelial neoplasia among human immunodeficiency virus-infected women. Am J Obstet Gynecol 2001; 3:322–330.
20. Clifford GM, Smith JS, Plummer M, et al. Comparison of HPV type distribution in high-grade cervical lesions and cervical cancer: a meta-analysis. Br J Cancer 2003; 89:101–105.
21. Manos M, Waldman J, Zhang T, et al. Epidemiology and partial nucleotide sequence of four novel genital human papillomaviruses. J Infect Dis 1994; 170:1096–1099.
22. Levi JE, Kleter B, Quint WGV, et al. High prevalence of human papillomavirus (HPV) infections and high frequency of multiple HPV genotypes in human immunodeficiency virus-infected women in Brazil. J Clin Microbiol 2002; 40:3341–3345.
23. Palefsky JM, Minkoff H, Kalish LA, et al. Cervicovaginal human papillomavirus infection in human immunodeficiency virus-1 (HIV)-positive and high-risk HIV-negative women. J Natl Cancer Inst 1999; 91:226–236.
24. Xu F, Kilmarx PH, Supawitkul S, et al. HIV-1 seroprevalence, risk factors, and preventive behaviors among women in Northern Thailand. JAIDS 2000; 25:353–359.
25. Lefevre J, Hankins C, Pourreaux K, et al. Prevalence of selective inhibition of HPV-16 DNA amplification in cervicovaginal lavages. J Med Virol 2004; 72:132–137.
26. Chan PKS, Chang AR, Tam WH, et al. Prevalence and genotype distribution of cervical human papillomavirus infection: comparison between pregnant women and non-pregnant controls. J Med Virol 2002; 67:583–588.
27. Ahdieh L, Klein R, Burk R, et al. Prevalence, incidence and type-specific persistence of human papillomavirus in human immunodeficiency virus (HIV)-positive and HIV-negative women. J Infect Dis 2001; 184:682–690.
28. Rugpao S, Nagachinta T, Wanapirak C, et al. Gynaecological conditions associated with HIV infection in women who are partners of HIV-positive Thai blood donors. Int J STD AIDS 1998; 9:677–682.
29. Sukvirach S, Smith JS, Tunsakul S, et al. Population-based human papillomavirus prevalence in Lampang and Songkla, Thailand. J Infect Dis 2003; 187:1246–1256.
30. Sirivongrangson P, Bollen LJM, Virapat P, et al. Should women with HIV infection be screened for sexually transmitted infections? initial findings of a demonstration project in Bangkok, Thailand. XV International AIDS Conference, 11–16 July 2004, Bangkok, Thailand [abstract ThPeC7390].
31. Thomas D, Ray NB, Kuypers J, et al. Human papillomaviruses and cervical cancer in Bangkok: the role of husbands and commercial sex workers. Am J Epidemiol 2001; 153:740–748.
32. Burk R, Ho G, Beardsley L, et al. Sexual behavior and partner characteristics are the predominant risk factors for genital human papillomavirus infection in young women. J Infect Dis 1996; 174:679–689.
33. Sellors JW, Mahony JB, Kaczorowski J, et al. Prevalence and predictors of human papillomavirus infection in women in Ontario, Canada: survey of HPV in Ontario Women (SHOW) Group. CMAJ 2000; 163:503–508.
34. Hawes SE, Critchlow CW, Faye Niang MA, et al. Increased risk of high-grade cervical squamous intraepithelial lesions and invasive cervical cancer among African women with human immunodeficiency virus types 1 and 2 infections. J Infect Dis 2003; 188:555–563.
35. Spinillo A, Debiaggi M, Zara F, et al. Human immunodeficiency virus type 1-related nucleic acids and papillomavirus DNA in cervicovaginal secretions of immunodeficiency virus-infected women. Obstet Gynecol 1997; 6:999–1004.
37. Amornwichet P, Teeraratkul A, Simonds RJ, et al. Preventing mother-to-child HIV transmission: the first year of Thailand’s national program. JAMA 2002; 288:245–248.