Safaeian, Mahboobeh PhD*; Kiddugavu, Mohammed MD§; Gravitt, Patti E. PhD*; Ssekasanvu, Joseph BSc§; Murokora, Dan MD§; Sklar, Marc MD∥; Serwadda, David MD¶; Wawer, Maria J. MD‡; Shah, Keerti V. MD, DrPH†; Gray, Ron MD‡
HUMAN PAPILLOMAVIRUS (HPV) IS THE causal agent for cervical cancer, and international studies have shown carcinogenic HPV types present in greater than 99.7% of cervical cancer cases.1–3 The Papanicolaou (Pap) test, introduced as a screening test in the 1940s, has reduced the incidence and mortality from cervical cancer mainly in countries with established screening program.4 In contrast, 80% of cervical cancers occur in developing countries with minimal or no screening programs.5
The motivation for conducting this study stems from the observation that in some cultures and settings, women will not seek a pelvic examination if they are asymptomatic, which creates a barrier for screening programs and for research, particularly among nonclinic populations in HPV DNA testing has been shown to have higher sensitivity compared with cytology6 for detecting high-grade cervical lesions; specificity is comparable over the age of 30.7 To increase screening coverage and overcome possible reticence among women reluctant to undergo pelvic examination, the use of self-collection of vaginal samples has been widely investigated. Of the studies that compare HPV DNA by self- and physician-collection methods, the majority have been conducted among high-risk populations: women from cervical cancer screening or colposcopy clinics, sexually transmitted disease or teen health clinics, and women from a case–control study of cervical cancer.8–19 There are only a minority of studies from resource-poor countries with limited or no access to screening programs,8,19–22 and only 2 were conducted among HIV-positive women.11,15 Although these studies have been instrumental in determining the prevalence of HPV and establishing the comparability of self- to physician-collection methods in the specific settings, they lack generalizability to general populations.
To study the validity of a self-collection method for HPV detection, we compared self-collected vaginal and physician-collected cervical samples from women enrolled in the Rakai Community Cohort Study (RCCS), a large population-based cohort in Uganda that provides a unique opportunity to assess the use of self-collection for HPV detection. Additionally, if performance of self-collection is as good as physician-collected samples, using self-collected swabs will facilitate prospective natural history studies of HPV infection in resource-poor, rural settings.
Materials and Methods
Participants for this study were part of the ongoing RCCS described in detail elsewhere.23 The primary aims of the RCCS are to assess trends in HIV and sexually transmitted infection prevalence and incidence in 50 rural communities in Southwestern Uganda and to conduct prevention studies. Briefly, the RCCS conducts annual home-based surveys with comprehensive interviewer-administered questionnaires on sociodemographic, behavioral, and health characteristics on a population of approximately 12,000 adults aged 15 to 49 years. Biologic samples are collected after the interview at home and include venous blood for HIV-1 serology.
Two HPV-related substudies were conducted within the RCCS. The schematics for these substudies are presented in Figure 1. The first, started in 1998, is the HPV natural history substudy, and the second, conducted in 2002 to 2003, is the pelvic examination substudy. The samples from the pelvic examination substudy were used in the present analysis. The pelvic examination substudy was conducted to formally evaluate the use of self-administered vaginal swabs for HPV DNA detection as a surrogate for cervical sampling. Two thousand five hundred women who had provided self-collected swabs during the follow-up visits (natural history substudy, 1998–2001) were identified and targeted for this substudy. Women were contacted at home and informed of the details of this substudy and were invited to come to a clinic for further evaluation by pelvic examination, colposcopy, and biopsy/cryotherapy as needed. Women signed a written informed consent. The study was approved by Institutional Review Boards in Uganda and at Johns Hopkins and Columbia Universities. Consenting women were asked to provide a self-collected vaginal sample in the clinic just before having a trained physician conduct a pelvic examination with direct cervical cell sampling. Self-collected vaginal swabs were obtained using the Digene Sampler Kit; however, instead of the usual conical brush provided with the kit, a Dacron swab was used. Women were asked to squat and insert a sterile 20-cm Dacron swab into the vagina up to the vault and to rotate the swab 3 times in the vaginal vault. The swab was then placed in 1 mL of Digene's standard transport media (STM) and kept on ice until it was transported back to the field laboratories and frozen at −80°C. After self-collection, after inserting the speculum, the physician first collected a cervical swab for HPV determination using a Dacron swab, which was placed in 1 mL of Digene's STM and next, performed a colposcopic examination of the cervix. Suspicious lesions were biopsied and a cervigram image was saved on the computer for future evaluation. Suspicious lesions were treated with cryotherapy or referral. Women could refuse study participation at any time during the examination. Transportation was provided but there was no monetary compensation.
All swab samples were periodically shipped to Johns Hopkins University and stored at −80°C until the time of assay.
Women were tested for HIV at the time of the annual cohort survey. The HIV status of each participant was determined using 2 enzyme immunoassays (EIAs) (Vironostika HIV-1; Organon Teknika, Charlotte, NC, and Cambridge Biotech, Worcester, MA), and discordant EIA results or new seroconversions were confirmed by Western blot (HIV-1 Western Blot; Bio-Merieux-Vitek, St. Louis, MO).
Human Papillomavirus Detection and Genotyping
HPV status was determined using a 2-stage approach. First, all samples were screened for the presence of carcinogenic HPV using the B-probe of the hybrid capture-2 (hc2) assay. For stage 2, all hc2 HPV-positive samples and a 10% random sample of hc2 HPV-negatives were genotyped using the PGMY09/11 consensus-primer polymerase chain reaction (PCR) and reverse hybridization technique (Roche Molecular Systems, Pleasanton, CA).24,25 Previous studies have confirmed that the correlation between hc2 and L1 consensus PCR is excellent.26
Human Papillomavirus DNA Testing by the Hybrid Capture-2
hc2 (Digene Corp., Gaithersburg, MD) is a U.S. Food and Drug Administration-approved, commercially available HPV test, which collectively targets 13 carcinogenic HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68). In fact, because of genetic relatedness, other cancer-associated types such as HPV-66 and HPV-53 are also detected by hc2.27 The hc2 assay was performed according to the manufacturer's instructions on a 100-μL aliquot of each specimen, which was denatured with 50 μL of denaturation solution and 75 μL of denatured product was used for the hc2 assay. The hc2 assay has a detection threshold of 1.0 pg/mL HPV DNA (approximately 5,000 copies).28 The full specimen was not processed for hc2 testing to preserve enough sample volume for future PCR-based analyses.
Human Papillomavirus DNA Testing by Roche Polymerase Chain Reaction Line Blot Assay
Fifty microliters of STM sample was denatured with 10× digestion buffer (4 mg/ml proteinase K and 1% laureth-12) for 1 hour at 65°C followed by heat denaturation of the proteinase K at 95°C for 10 minutes. DNA was precipitated by addition of 200 μL precipitation buffer (absolute ethanol and 0.825M ammonium acetate). Samples were mixed by inversion and precipitated overnight at −20°C. The precipitated DNA was pelleted by centrifugation. The supernatant was removed using fine-tipped transfer pipettes, and DNA pellet was dried and resuspended in 25 μL Tris-EDTA buffer and stored at −20°C.17 For quality control purposes, a negative control sample (containing 1 × 106 HPV-negative cells [K562] per mL of STM) was processed after every 11th sample.
Five microliters of DNA extracted from the STM samples was used for consensus amplification of HPV using a cocktail of biotinylated PGMY 09/11 and β-globin primers in a final volume of 100 μL. The PCR product was denatured in 0.4 N NaOH and HPV genotyping performed using the prototype reverse line blot (LB) system developed by Roche Molecular Systems (RMS), Inc. as described earlier.24,25 Briefly, the PGMY 09/11 primer system targets a 450 bp conserved sequences in the HPV L1 gene. Thirty-seven HPV genotypes, which include 22 common carcinogenic and 15 low-risk HPV types and a human β-globin internal control target for sample quality and PCR integrity24,25 were discriminated by hybridizing the biotinylated PCR products to the Roche prototype probe array (RMS line blot).25,29,30 We included 4 positive controls (known low and high quantities of HPV-16 and HPV-18), the negative control DNA extracted with the original samples at every 12th space and a water negative control as a quality control measure for PCR integrity in each PCR plate.
The primary objective was to determine concordance between self- and physician-collected swabs for the detection of carcinogenic HPV by hc2 and Roche LB assay. All samples were tested by hc2, which is calibrated to maximize clinical sensitivity for detection of cervical lesions. The Roche line blot genotyping was not uniformly applied to all samples (all hc2-positives and 10% randomly selected hc2-negative samples were genotypes); hence, to use all available data, the focus of our analysis is based on the hc2 data. Subsequently, we compared the prevalence of each HPV type detected from self- and physician-collection methods on the subset that were genotyped. Because only 10% of the hc2 HPV-negatives was chosen for genotyping, in this article, we report on our findings of the carcinogenic HPV types present in the hc2 probe (HPV type 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68). The agreement between sampling methods was determined by unweighted κ statistics and 95% confidence intervals (CIs), which calculate percent agreement beyond that expected by chance alone. Number of HPV types detected by each collection method was determined and nonparametric Wilcoxon test for differences between medians was used as assessment of statistical significance. We used the semiquantitative RLU values from the hc2 assay as a proxy measure for HPV viral burden. The RLU values were log-transformed and differences between medians by collection methods were assessed by nonparametric Wilcoxen test of median. We grouped the genotype data to include the 13 hc2-targeted HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68. Because of the well-described crossreactivity of hc2 with HPV66 and HPV53,26,30,31 we also included these 2 high-risk types in this group. All analyses were stratified by participants' HIV status to investigate whether HIV infection affects the performance.
To examine if there was a preference of detection of HPV types in either the vaginal (self-collected) or cervical (physician-collected) tissue, we examined the misclassification by types using a McNemar's test, which is a nonparametric test for matched data. It determines whether the proportion of samples classified as positive by self-collected vaginal swab and negative by physician-collected cervical swabs is equal to the proportion of samples classified as negative by self-collected vaginal swab and positive by physician-collected cervical swabs.
Figure 1 presents the schematics for the HPV substudies conducted within the RCCS and reasons for refusing to participate in the pelvic examination substudy. During RCCS home visits, over 86% of women routinely provide self-administered vaginal swabs. However, only 51% agreed to accept a pelvic examination conducted by a physician. Reasons for nonparticipation are shown on the righthand panel. Of the 2,223 women in active follow up, invited to participate in the pelvic examination substudy, 51% enrolled, whereas 25% declined at the outset and 19% who initially showed interest did not attend their appointment.
Women who agreed to a pelvic exam were significantly older (median age 30; IQR 25–38) compared to nonparticipants (median age 26; IQR 21–35, P <.0001) and more likely to have higher number of lifetime sex partners (P <0.0001). They were not different in terms of HIV status, current pregnancy, and current marital status relative to nonparticipants (Table 1). Using data from the self-collected samples from the routine Rakai follow-up visits showed that participants accepting a physician examination were also more likely to have had a positive prior hc2 test (18%) than nonparticipants (14%) (P = 0.02, data not shown). However, at the time data were collected for this substudy, participants would not have known their prior HPV infection status; hence, this is unlikely to have biased participation in this study.
Paired self- and physician-collected swabs were available from 606 women. hc2 results were available for all of these; however, when performing the genotyping, the internal β-globin target was not amplified in 2 of the self- and 4 of the physician-collected samples and these samples were excluded when reporting genotype-specific analyses.
Table 2 presents the overall and HIV-stratified demographic characteristics of the population. HIV results were unknown for 7 participants. The overall demographic characteristics at the time of dual specimen collection were as follows: median age, 30 years, and median age of sexual debut, 16 years; 94 (15.7%) were HIV-positive. HIV-positive women differed from HIV-negative women in that HIV-positive women were older and more likely to report a higher lifetime number of sexual partners.
Table 3 presents the overall and HIV stratified HPV prevalence (determined by hc2) by selected sociodemographic and behavioral factors. Carcinogenic HPV prevalence was 19.0% (115 of 606) in self-collected and 19.0% (116 of 606) in physician-collected swabs (P = 0.5). In general, HPV prevalence rates were similar between self- and physician-collected specimens in each stratum of characteristics. Using either the self- or physician-collected samples, HPV prevalence was higher among HIV-positive women (40% and 37% in self- and physician-collected samples, respectively) compared with HIV-negative women (15% and 16% in self- and physician-collected samples, respectively). As expected, carcinogenic HPV prevalence declined with age, later age at first intercourse, and fewer sex partners in the prior year. Further stratifying by HIV status, we observed similar HPV prevalence rates by collection method; however, HIV-positive women were more likely to have higher HPV prevalence rates compared with HIV-negative women regardless of collection method.
Table 4 shows the overall and HIV-stratified agreement and κ values between self- and physician-collection methods using data from the hc2 assay. Overall, the κ for HPV detection from self-collected vaginal and physician-collected cervical swabs was 0.75 (95% CI = 0.68–0.82). Overall, there were 47 discordant records; however, they were equally distributed among the self- and physician-collected swabs. There were 24 samples that were positive by the physician collection and negative by self-collection. Conversely, there were 23 self-collected samples that were positive but were negative by physician collection (McNemar's P ≤1.00). The κ between self- and physician-collected samples were similar by HIV strata (κ = 0.71, 95% CI = 0.56–0.86) and 0.75 (95% CI = 0.67–0.83) for HIV-positive and HIV-negative women, respectively.
The type-specific prevalence rates and kappas for carcinogenic HPV are shown in Figure 2. Overall, HPV-16 was the most prevalent type detected in both self- and physician-collected samples (5.3% and 4.7% in self- and physician-collected samples respectively) with a κ of 0.90. Kappa for type-specific agreement ranged from 0.41 (HPV-70) to 1.00 (IS-39). The κ for any carcinogenic HPV type detected by LB was 0.87 (95% CI = 0.82–0.92).
HIV-stratified prevalence and κ of carcinogenic HPV types determined from the LB assay were calculated (data not shown, available on request). Among HIV-positive women, carcinogenic HPV prevalence ranged from 0% (HPV IS-39) to 12% (HPV-52) by either collection method. Among HIV-negative women, prevalence rates of carcinogenic HPV ranged from 0.2 (HPV-69 and HPV-26) to 4.5% (HPV-16) by either collection method. When stratified by HIV status, the κ value for any carcinogenic HPV among HIV-positive women was 0.89 (95% CI = 0.80–0.98), and among HIV-negative women, κ was 0.85 (95% CI = 0.79–0.91) (data not shown).
Overall, the median number of carcinogenic HPV types detected was 2 (IQR 1–3) in self-collected vaginal samples and one (IQR 1–3) among MD-collected (P = 0.14). Stratifying by HIV, the median number of HPV types identified in self- and physician-collected specimens did not differ by collection method (P = 0.2 and 0.5 among HIV-negative and HIV-positive, respectively); however, HIV-positive women had a significantly larger number of concurrent HPV infections (median of 3 [IQR 2–5]) compared with HIV-negative women (median of 2 [IQR 1–3], P = 0.000 for both self-collected and physician-collected samples). Similarly, using the semiquantitative RLU by the hc2 assay as a marker of viral burden, we observed no differences in the cumulative viral burden in either self- or physician-collected samples (P = 0.3 and 0.6 for HIV-negative and HIV-positive, respectively); however, HIV-positive women had significantly higher viral burden as measured by the RLU values (P = 0.000 and 0.006 among self-collected and physician-collected, respectively; data not shown).
Because HPV is a necessary cause of cervical cancer and given the public health benefits of a screening test for cervical neoplasia, increasing emphasis has been placed on HPV testing, which has higher or comparable sensitivity and specificity for detecting high-grade cervical lesions when compared with cytologic screening.7,32 In developed countries with established screening protocols, a dramatic decrease in incidence of cervical precancerous lesions, cancer, and mortality from cervical cancer has been observed over the past 50 years.4 However, as a result of lack of infrastructure and expense, screening programs are unavailable in most developing countries, and as a result, declines in cervical neoplasia have not been observed in resource-poor countries. Additionally, a constraint to screening in many developing countries is the cultural reticence to seek routine pelvic examination; even in some countries with established national screening protocols, women do not seek pelvic examinations and therefore often present with advanced disease.33,34 Hence, there is an urgent need for acceptable and cost-effective screening methods, which can minimize the need for a physician-performed pelvic examination.
Earlier studies of comparability of self- and physician-collection methods have mainly been conducted among high-risk populations.8–19 Our study was conducted in a well-characterized rural African, population-based cohort, and results are likely to be generalizable to other rural areas of sub-Saharan Africa, which constitute 80% or more of the total African population. We demonstrated that self-collection can be a reliable method for sample collection, not only for screening, but potentially for the purpose of natural history research and possibly for the conduct of future HPV vaccine trials.
Of note was our finding of higher acceptability of self-administered samples collected during annual home visits (≥86%) compared with collection by pelvic examination (50%), suggesting that self-collection is likely to provide better population coverage. Qualitative research conducted with eligible women suggested that a reason for noncompliance with a pelvic examination was travel distance from place of residence to the clinic. Subsequent to this finding, women were provided with free transportation to the clinic, but this only marginally improved participation rates and acceptance of pelvic examination, suggesting that cultural factors were still a barrier to screening by pelvic examination. There is a need for higher service coverage to achieve effective screening; thus, self-collected samples for HPV detection might be of use to identify women in need of a pelvic examination and treatment. However, the ability to follow up an HPV-positive result from a self-collected swab has yet to be sufficiently determined. This will require practical effectiveness trials.
Using the hc2 assay, self-collection classified 24 (4%) of the samples as negative and physician-collection classified 23 (4%) of the samples as negative (Table 4). This is lower but consistent with findings by other studies.16,17,21 The discordance, as suggested by Gravitt et al, may be explained by the fact that the HPV DNA quantity may be below the detection level of the assays used.17 However, the clinical relevance of detecting very low-level DNA loads is questionable.35 The observed discordance could also be the result of differences in collection by the participants or the physician. Further examination of the data revealed that, either by hc2 or line blot, the number of discordant pairs was not statistically different between the self-collected vaginal and physician-collected cervical samples. When we analyzed the discordant samples by age, HIV status, and HPV type distribution, we did not observe any patterns; hence, our conclusion that from an analytic view, discordants are randomly distributed. Unfortunately, at this point, we do not have the data to address whether more clinically relevant lesions could be missed if we relied on self-collection, but because the discordants are randomly distributed, one can conclude that some clinically relevant lesions will be missed even with physician-administered swabs.
An important finding of this study was that women from the general population were capable of collecting reliable samples that can be used in research studies and for screening. Sample integrity as measured by the β-globin gene in the self-collected sample was inadequate for only 2 of the 606 self-collected samples; however, 4 of the 606 samples collected by the physician had inadequate β-globin amplification.
The predominant HR-HPV types in the overall cohort were HPV-16 followed by HPV-52. The finding of higher HPV prevalence in HIV-positive compared with HIV-negative women is consistent with other studies.36 Agreement between self- and physician collection was not affected by HIV status. Also, the number of concurrent HPV infections detected and cumulative viral burden was not affected by collection method.
There are limitations to this study. RCCS is a well-established cohort with over 10 years of annual surveillance. Participants trust and have good rapport with the study personnel, which might increase the acceptability of self-collection and limit generalizability to other populations. Women who accepted a pelvic examination were significantly older with more lifetime sex partners than those who did not. We compared κ values by age and lifetime number of sex partners and observed similar kappas within those subgroups. Moreover, because we compared paired self-collected vaginal and physician-collected cervical samples, our study had strong internal validity and hence, these differences are unlikely to affect our findings of similar HPV detection rates in self and physician samples.
The predominant HPV types in Uganda are unknown, and this information is needed to plan future HPV vaccine trials for the most common HPV virus types in Uganda. In addition, alternative methods of follow up of vaccinated individuals for vaccine effectiveness evaluation using virologic end points are required given the low compliance for speculum-assisted cervical sampling. We demonstrated that self-collection provides a reliable method of obtaining samples and suggests that this procedure might be adopted in trials to increase compliance in rural and low-resource settings.
1. Bosch FX, Lorincz A, Munoz N, Meijer CJ, Shah KV. The causal relation between human papillomavirus and cervical cancer. J Clin Pathol 2002; 55:244–265.
2. Munoz N, Bosch FX, de Sanjose S, et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med 2003; 348:518–527.
3. Walboomers JM, Jacobs MV, Manos MM, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999; 189:12–19.
4. Vizcaino AP, Moreno V, Bosch FX, et al. International trends in incidence of cervical cancer: II. Squamous-cell carcinoma. Int J Cancer 2000; 86:429–435.
5. Parkin DM, Bray FI, Devesa SS. Cancer burden in the year 2000. The global picture. Eur J Cancer 2001; 37(suppl 8):S4–66.
6. Cuzick J, Clavel C, Petry KU, et al. Overview of the European and North American studies on HPV testing in primary cervical cancer screening. Int J Cancer 2006.
7. Franco EL. Primary screening of cervical cancer with human papillomavirus tests. J Natl Cancer Inst Monogr 2003:89–96.
8. Salmeron J, Lazcano-Ponce E, Lorincz A, et al. Comparison of HPV-based assays with Papanicolaou smears for cervical cancer screening in Morelos State, Mexico. Cancer Causes Control 2003; 14:505–512.
9. Garcia F, Barker B, Santos C, et al. Cross-sectional study of patient- and physician-collected cervical cytology and human papillomavirus. Obstet Gynecol 2003; 102:266–272.
10. Hillemanns P, Kimmig R, Huttemann U, Dannecker C, Thaler CJ. Screening for cervical neoplasia by self-assessment for human papillomavirus DNA. Lancet 1999; 354:1970.
11. Petignat P, Hankins C, Walmsley S, et al. Self-sampling is associated with increased detection of human papillomavirus DNA in the genital tract of HIV-seropositive women. Clin Infect Dis 2005; 41:527–534.
12. Dannecker C, Siebert U, Thaler CJ, Kiermeir D, Hepp H, Hillemanns P. Primary cervical cancer screening by self-sampling of human papillomavirus DNA in internal medicine outpatient clinics. Ann Oncol 2004; 15:863–869.
13. Sellors JW, Lorincz AT, Mahony JB, et al. Comparison of self-collected vaginal, vulvar and urine samples with physician-collected cervical samples for human papillomavirus testing to detect high-grade squamous intraepithelial lesions. CMAJ 2000; 163:513–518.
14. Baldwin S, Santos C, Mendez BE, et al. Comparison of type-specific human papillomavirus data from self and clinician directed sampling. Gynecol Oncol 2005; 97:612–617.
15. Palmisano ME, Gaffga AM, Daigle J, et al. Detection of human papillomavirus DNA in self-administered vaginal swabs as compared to cervical swabs. Int J STD AIDS 2003; 14:560–567.
16. Kahn JA, Slap GB, Huang B, et al. Comparison of adolescent and young adult self-collected and clinician-collected samples for human papillomavirus. Obstet Gynecol 2004; 103:952–959.
17. Gravitt PE, Lacey JV Jr, Brinton LA, et al. Evaluation of self-collected cervicovaginal cell samples for human papillomavirus testing by polymerase chain reaction. Cancer Epidemiol Biomarkers Prev 2001; 10:95–100.
18. Harper DM, Noll WW, Belloni DR, Cole BF. Randomized clinical trial of PCR-determined human papillomavirus detection methods: Self-sampling versus clinician-directed—Biologic concordance and women's preferences. Am J Obstet Gynecol 2002; 186:365–373.
19. Lorenzato FR, Singer A, Ho L, et al. Human papillomavirus detection for cervical cancer prevention with polymerase chain reaction in self-collected samples. Am J Obstet Gynecol 2002; 186:962–968.
20. Lack N, West B, Jeffries D, et al. Comparison of non-invasive sampling methods for detection of HPV in rural African women. Sex Transm Infect 2005; 81:239–241.
21. Wright TC Jr, Denny L, Kuhn L, Pollack A, Lorincz A. HPV DNA testing of self-collected vaginal samples compared with cytologic screening to detect cervical cancer. JAMA 2000; 283:81–86.
22. Belinson JL, Qiao YL, Pretorius RG, et al. Shanxi Province cervical cancer screening study II: Self-sampling for high-risk human papillomavirus compared to direct sampling for human papillomavirus and liquid based cervical cytology. Int J Gynecol Cancer 2003; 13:819–826.
23. Wawer MJ, Gray RH, Sewankambo NK, et al. A randomized, community trial of intensive sexually transmitted disease control for AIDS prevention, Rakai, Uganda. AIDS 1998; 12:1211–1225.
24. Gravitt PE, Peyton CL, Apple RJ, Wheeler CM. Genotyping of 27 human papillomavirus types by using L1 consensus PCR products by a single-hybridization, reverse line blot detection method. J Clin Microbiol 1998; 36:3020–3027.
25. Gravitt PE, Peyton CL, Alessi TQ, et al. Improved amplification of genital human papillomaviruses. J Clin Microbiol 2000;38:357–361.
26. Peyton CL, Schiffman M, Lorincz AT, et al. Comparison of PCR- and hybrid capture-based human papillomavirus detection systems using multiple cervical specimen collection strategies. J Clin Microbiol 1998; 36:3248–3254.
27. Schiffman M, Wheeler CM, Dasgupta A, Solomon D, Castle PE. A comparison of a prototype PCR assay and hybrid capture 2 for detection of carcinogenic human papillomavirus DNA in women with equivocal or mildly abnormal Papanicolaou smears. Am J Clin Pathol 2005; 124:722–732.
28. Lorincz AT, Anthony J. Advances in HPV detection by hybrid capture. Papillomavirus Report 2001; 12:145–154.
29. Peyton CL, Gravitt PE, Hunt WC, et al. Determinants of genital human papillomavirus detection in a US population. J Infect Dis 2001; 183:1554–1564.
30. Poljak M, Marin IJ, Seme K, Vince A. Hybrid Capture II HPV test detects at least 15 human papillomavirus genotypes not included in its current high-risk probe cocktail. J Clin Virol 2002; 25(suppl 3):S89–S97.
31. Castle PE, Schiffman M, Burk RD, et al. Restricted cross-reactivity of hybrid capture 2 with nononcogenic human papillomavirus types. Cancer Epidemiol Biomarkers Prev 2002; 11:1394–1399.
32. Sankaranarayanan R, Gaffikin L, Jacob M, Sellors J, Robles S. A critical assessment of screening methods for cervical neoplasia. Int J Gynaecol Obstet 2005; 89(suppl 2):S4–S12.
33. Sankaranarayanan R, Black RJ, Swaminathan R, Parkin DM. An overview of cancer survival in developing countries. IARC Sci Publ 1998:135–173.
34. Hernandez-Avila M, Lazcano-Ponce EC, de Ruiz PA, Romieu I. Evaluation of the cervical cancer screening programme in Mexico: A population-based case–control study. Int J Epidemiol 1998; 27:370–376.
35. Gravitt PE, Burk RD, Lorincz A, et al. A comparison between real-time polymerase chain reaction and hybrid capture 2 for human papillomavirus DNA quantitation. Cancer Epidemiol Biomarkers Prev 2003; 12:477–484.
36. Palefsky JM, Holly EA. Immunosuppression and co-infection with HIV. J Natl Cancer Inst Monogr 2003:41–46.