Presence of multiple human papillomavirus types in cervical samples from HIV-infected women

Gynecologic Oncology 92 (2004) 225 – 231 www.elsevier.com/locate/ygyno

Presence of multiple human papillomavirus types in cervical samples $ from HIV-infected women Jose´ Eduardo Levi, a,* Silvana Fernandes, a Adriana Fumie Tateno, a Eduardo Motta, b Liliam Pereira Lima, c Jose´ Eluf-Neto, c and Cla´udio Se´rgio Pannuti a a

Laborato´rio de Virologia do Instituto de Medicina Tropical de Sa˜o Paulo, LIMHC, Departamento de Doenc¸as Infecciosas e Parasita´rias, Faculdade de Medicina da Universidade de Sa˜o Paulo, Sa˜o Paulo, SP, Brazil b Departamento de Ginecologia e Obstetrı´cia da Faculdade de Medicina da Universidade de Sa˜o Paulo, Sa˜o Paulo, SP, Brazil c Departamento de Medicina Preventiva da Faculdade de Medicina da Universidade de Sa˜o Paulo, Sa˜o Paulo, SP, Brazil Received 30 May 2003

Abstract Objectives. The aim of this study was to detect and identify human papillomavirus (HPV) genotypes on a population of women infected by the human immunodeficiency virus (HIV) and to investigate the role of multiple infections on cervical dysplasia. Methods. Two hundred and fifty-five HIV-infected women were enrolled on a study to evaluate the prevalence of HPV and cervical intraepithelial neoplasia (CIN). A group of HIV-negative women with confirmed CIN diagnosis was included for comparison. A polymerase chain reaction (PCR)-reverse hybridization method was applied to detect and precisely identify HPV types, specifically multiple infections. Results. On HIV patients, an altered Pap smear confirmed by biopsy was observed on 45 (18%); HPV-DNA prevalence was 87% (223/ 255), with 45% (116/255) infected by more than two types. In contrast, HPV-DNA was detected in all 36 women of the control group but only 3 were infected by more than two types. Cervical dysplasia was associated with low CD4 counts and elevated high-risk HPV viral load. However, the presence of multiple HPV types did not correlate with the degree of immune suppression or the presence of cervical lesions. Conclusions. Infection with multiple HPV types is a rather frequent finding on Brazilian HIV-infected women. On this population, concomitant infection with three or more HPV types does not seem to confer an additional risk of cervical dysplasia in comparison to single/ double infections, nor to be related to more severe immunesuppresion. D 2003 Elsevier Inc. All rights reserved. Keywords: HPV; PCR; Genotyping; Viral load; Cervical intraepithelial neoplasia; Multiple infections

Introduction It is well documented that women infected with the human immunodeficiency virus (HIV) have a high prevalence of human papillomavirus (HPV) infection [1,2] and cervical squamous intra-epithelial lesions (SIL) [3]. In this group of immunocompromised hosts, these lesions have a

$ This work was supported by the Brazilian Ministry of Health, STD/ AIDS Program and FAPESP (Grant 97/00541-7). * Corresponding author. Laborato´rio de Virologia, Instituto de Medicina Tropical da Universidade de Sa˜o Paulo, Rua Dr. Ene´as de Carvalho Aguiar 470, 2o andar, CEP 05403-000, Sa˜o Paulo, SP, Brazil. Fax: +55-11-3063-2659. E-mail address: [email protected] (J.E. Levi).

0090-8258/$ - see front matter D 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2003.10.004

worse outcome, progressing faster than in immunocompetent patients [4]. Lesions are difficult to treat, with a high rate of recurrence [5], frequently containing the same pretreatment HPV type [6]. Several epidemiological studies identified HPV persistence and viral load as independent factors conferring risk for further development of cervical abnormalities [7,8]. In fact, both variables are interrelated since a high-viral load infection may be more difficult to clear, resulting in persistence, while a persisting infection may allow more cycles of replication, leading to an increase of HPV-DNA. Persistence of HPV-DNA also fully explains the high association between HIV-mediated immunesupression and SIL [9,10]. In the near past, investigators working with molecular methods for HPV typing on cervical samples from immunocompromised patients have described that multiple HPV

226

J.E. Levi et al. / Gynecologic Oncology 92 (2004) 225–231

types was a frequent finding. However, the extent of multiple infections and a precise characterization of HPV types involved on that was not feasible with the typing techniques then available. Recently, the development of polymerase chain reaction (PCR)-based methods that are more sensitive and able to identify more than 20 specific HPV types simultaneously [11,12] has allowed multiple infections to be systematically investigated. In the current study, 255 HIV-infected women and a control group of 36 women at risk for HPV infection but HIV seronegative were analyzed for the presence of HPV-DNA and subsequent typing, by a PCR-reverse hybridization method to fully characterize the spectrum of HPV types and assess the frequency of multiple HPV infections. Moreover, we attempted to correlate multiple HPV infections with other virological and clinical parameters.

Subjects and methods

This study was approved by the Ethical Committee of the participating institutions. Written informed consent was obtained from all subjects. Cytology and histology Pap smears were examined and categorized according to the Pap classification comprising Pap I (normal), Pap II (inflammation), Pap III (dysplasia), Pap IV (carcinoma in situ) and Pap V (invasive carcinoma). All patients showing abnormal cytological smears (higher than class II) were referred for colposcopy-directed biopsy, and biopsies were examined and classified according to the CIN system [13]. HIV viral load Blood was drawn by venipuncture and centrifuged within 6 h. Plasma was separated and frozen until processed for viral load. The AMPLICOR HIV Monitor v 1.0 assay (Roche, Sa˜o Paulo, Brazil) was employed for viral load determination on all samples, the linear range of the assay was from 400 to 750,000 copies/ml.

HIV patients Hybrid capture II HIV-infected women attending two specialized centers (Instituto de Infectologia Emı´lio Ribas and Casa da AIDS) in Sa˜o Paulo, Brazil, between December 1997 and September 1999, were invited to participate in the study during a routine visit to the gynecologist. Exclusion criteria were the inability to provide an informed consent and a CD4 cell count of less than 50/Al. Mean age was 32 years (range 18 – 67 years, median 32 years). Patients enrolled (n = 255) underwent a gynecological evaluation and a cervical scrape was collected for cytological analysis and HPV-DNA PCR. A blood sample was obtained to determine the HIV viral load (Roche AMPLICOR HIV Monitor). A CD4+ cell count was accepted for the study if it was performed within a 3 months’ period from the gynecologic visit. Otherwise, a blood sample for CD4+ cells determination was also obtained at that visit. Two hundred and three patients (80%) were receiving antiretroviral agents, mainly a combination of two nucleoside analogue reverse transcriptase inhibitors and one protease inhibitor.

Cervical samples were collected with the brush provided in Digene’s Hybrid Capture II specimen collection kits (Digene do Brasil, Sa˜o Paulo, Brazil) and transported to the Virology Laboratory within 48 h, in the sample transport medium provided in the kit. Results are presented as a ratio of relative light units (RLU) of sample/calibrator. For Positive Calibrator B (PCB, high-risk cocktail), calibrator is a solution containing 1 pg/ml of HPV 16 DNA. DNA isolation Four hundred and fifty microlitres of the Hybrid Capture denatured specimen was removed and DNA was precipitated using 0.3 M sodium acetate and 70% ethanol. Precipitated DNA was washed, dried and resuspended in 100 Al of 0.1 M Tris – EDTA pH = 8.3.

Control group This group was composed of 36 women attending the ambulatory clinic of Cond/Neo (Condyloma and Intraepithelial Neoplasias) in the Gynecology Department of the Hospital das Clı´nicas da Universidade de Sa˜o Paulo. These women were referred to this specialized center after an initial diagnosis or suspicion of condyloma or cervical intraepithelial neoplasia (CIN) and were all HIV negative by serology. Mean age was 32 years (range 18 – 54 years, median 32 years).

Fig. 1. Number of HPV types in cervical samples from HIV+ and HIV patients.

J.E. Levi et al. / Gynecologic Oncology 92 (2004) 225–231

227

HPV genotyping (line blot method)

DNA sequencing

Five microliters of DNA (containing 50– 1000 ng) was added to a PCR mix containing biotinylated PGMY09/ PGMY11 [500 nM] and PCO4/GH20 [25 nM] primers, 4 mM MgCl2, dNTPs (dATP, dCTP, dGTP [200 AM] and dUTP [600 AM]), and 7.5 Units of AmpliTaq Gold (Applied Biosystems, Foster City, CA, USA) in a final volume of 100 Al. PCR products were analyzed by ethidium bromidestained 2% agarose gel electrophoresis, then denatured and hybridized to a strip containing immobilized probes for 27 HPV types plus 2 concentrations of the h-globin probe, as described [12]. All reagents and strips were assembled in a prototype kit that was provided by Roche Molecular Systems. Typing is based on the position of HPV type-specific oligonucleotides on the strip. Samples that were positive after amplification by visual inspection of the gel but did not hybridize to any of the strip probes were considered to harbor HPV of indeterminate type(s).

Samples containing indeterminate types were re-amplified with PGMY09/11 primers and the amplicons purified by column chromatography. These PCR products were submitted to DNA sequencing with Cy5-labelled GP5+ primer [14] and a Thermosequenase Primer Cycle Sequencing 7-deaza-dGTP kit (Pharmacia, Sa˜o Paulo-Brazil). Sequencing products were run on an Alf Express (Pharmacia) automated sequencing apparatus and the sequence was analyzed by a BLAST search on GenBank and aligned to several L1 450 bp fragments of distinct HPV types. Statistical analyses ANOVA was used for comparison of mean number of HPV types, CD4 cell counts, HIV and HPV viral loads among cytological classes and mean number of HPV types among CD4 categories.

Table 1 Type-specific human papillomavirus (HPV) prevalence among 255 HIV+ women and 36 HIV method

women, as determined by the line-blot reverse-hybridization

Dark gray = high-risk types, Light gray = low-risk types, Blank = risk undefined. Classification follows the scheme of Mun˜oz et al. [32]. *—Four cases of HPV 70, 1 HPV 44, 1 HPV 62, 1 HPV 81 and 2 samples for which we could not obtain high-quality sequencing data. f—HPV 73 is epidemiologically a high-risk type but phylogenetically closer to low-risk types.

228

J.E. Levi et al. / Gynecologic Oncology 92 (2004) 225–231 Table 3 Number of HPV types, HIV viral load, HPV viral load and CD4 cell counts in HIV-infected patients stratified by cytological results

Results Cytology and histology

Cytology Pap class

Number of HPV types, mean (95% CI)

CD4/Al, mean (95% CI)

HIV viral load CPS/ml, mean (95% CI)

HPV viral load RLU/PCB (high risk), mean (95% CI)

I (N = 25)

2.4 (1.6; 3.2) 2.5 (2.2; 2.7)

III (N = 45)

3.0 (2.3; 3.6)

HPV types

ANOVA P value

0.288

28,753 (13,078; 70,584) 64,589 (36,798; 89,381) 87,667 (44,877; 130,458) 0.346

63.8 (0.7; 127.0)

II (N = 175)

336.2 (264.2; 408.1) 304.0 (278.7; 329.2) 233.2 (183.4; 283.1) 0.019

All samples were positive on amplification of the hglobin gene confirming their adequacy. Two hundred and twenty-three (87%) HIV+ patients were shown by PCR to harbor HPV-DNA. Sixty (24%) patients were infected by a unique HPV type, 47 (18%) by two types, 41 (16%) by three types (16%), 36 (14%) by four types and 39 (15%) by more than four types (range 1 – 11 types, mean = 2,9 HPV types). HPV-DNA was detected in all 36 women of the control group: 21 (58%) contained only one type, 13 (36%) were infected by two types, 1 (3%) by three types and 1 (3%) by six types (Fig. 1). The distribution of individual HPV types in these two groups is depicted in Table 1. HPV 16 was the most common type in HIV+ patients, while in the control group, HPV 51 appears as the most common followed by HPV 16. Nine (4%) patients were found to bear indeterminate HPV types. Sequencing revealed the presence of HPV types for which corresponding oligonucleotide probes were not included in the strip (HPV 70 in four patients, HPV 44, HPV 62 and HPV 81 in one patient each), while for two of them, we could not obtain reliable sequence data.

Note. ANOVA, analysis of variance; CI, confidence interval. For analysis of HPV viral load, log values were applied since variance was different among groups.

Ten cytological smears could not be analyzed due to absence of cervical cells (2), hemorrhage (2), broken slide (2) and lost slide (4). Of the remaining 245, 200 (82%) were classified as normal or inflammatory and 45 (18%) were shown to harbor dysplastic smears (Pap III). These patients were submitted to colposcopy and biopsy when appropriate; 40 (89%) of them had a diagnosis of SIL (7 high-grade and 33 low-grade SIL) confirmed by a pathologist on inspection of biopsy slides.

Multiple types and degree of immunodeficiency in HIV+ women When stratified by CD4 count categories, it was observed that 31% of the patients with less than 200 cells/Al had a diagnosis of Pap III, in contrast to 13% on those with counts

Table 2 Prevalence of Pap III and number of HPV types stratified by CD4 cell counts in HIV-infected patients CD4/Al

Number of HPV types, mean (95% CI)

Prevalence of Pap III (%)

< 200 (N = 84) 200 – 400 (N = 113) >400 (N = 56) ANOVA P value

2.8 (2.4; 3.3) 2.3 (1.9; 2.6) 2.6 (2.1; 3.1) 0.108

31.0 12.3 13.8 nc

Note. ANOVA, analysis of variance; CI, confidence interval; nc, not calculated.

257.5 (157.2; 357.8) 919.6 (603.0; 1236.2) < 0.001

>200 cells/Al (Table 2). However, in the present series, the mean number of HPV types did not vary significantly according to immune status (Table 2, P = 0.108) or cytological class (Table 3, P = 0.288). Cytological class and multiple HPV types Table 3 shows the number of types and other laboratorial data according to the Pap classification. The mean number of HPV types was similar through all categories while a correlation between worse cytological diagnosis and decreasing CD4 counts was clearly observed. HPV viral load Hybrid capture II was performed on all 255 samples. Two were found inadequate for HCII since run failed and there was not enough liquid left for repetition. Among 253 adequate samples, 88 were negative (including 63 PCR positive), 19 were positive only for the low-risk probe (including 2 PCR negative), 86 were positive for the highrisk probe (including 4 PCR negative) and 60 were positive for both high- and low-risk probes (including 1 PCR Table 4 Number of HPV types infecting HIV+ patients and other clinicallaboratorial parameters Number of HPV types

Prevalence of Pap III (%)

Mean HPV viral load (high risk, RLU/PCB)

Mean CD4/Al

Mean HIV viral load (cps/ml)

0 (N = 32) 1 (N = 60) 2 (N = 47) 3 (N = 41) 4 (N = 36) z 5 (N = 39)

9 21 12 20 22 21

12.2 196.6 275.9 229.0 361.7 697.8

307 312 280 281 296 271

27,577 71,922 21,027 93,150 51,535 106,152

J.E. Levi et al. / Gynecologic Oncology 92 (2004) 225–231

negative). Three cases harboring Pap III smears were negative for the HCII test (including one PCR negative and two PCR positive) while two Pap III cases were positive for the low-risk probe only. Table 4 depicts the overall Pap III prevalence, which is around 20% on patients infected with z 1 HPV type. As expected, the HPV viral load increases when more types are present.

Discussion Although infection with multiple HPV types has been described in the past [15], it does not seem to be commonly observed in cervical samples obtained from immunocompetent women from many countries. Multiple infection was found in only 2.2% of the participants at entry, in the Ludwig-McGill Cohort Study consisting of women attending a maternal and child health program for low-income families held in Sa˜o Paulo city, Brazil [16]. Similarly, multiple infection was observed in 2.6% of Mexican women from Morelos state with normal cytology [17]. In our preliminary experience with material derived from HIV-infected women, it became evident that PCR-RFLP was not a suitable technique to type HPV from these samples, since infection with more than two types was very common, precluding a correct assessment. The use of a methodology appropriate to detect even minor HPV populations and its application to samples from HIV-infected women has demonstrated the simultaneous presence of HPV-DNA from several types, in some instances as many as 10 HPV types in single individuals [1,18]. Using the same methodology, the presence of high-risk HPV types (including HPV 16, 18, 45 and others) was also verified in condylomata acuminata lesions from HIV immunocompromised patients [19], an intriguing finding upon the wellestablished concept that genital warts are caused by low-risk HPV types. This concept has been challenged in the past, when Bradshaw et al. [20] demonstrated by in situ hybridization the occurrence of an HPV 16 condyloma and perianal carcinoma on an HIV-infected man. In the current study, infection with more than two HPV types was observed in 45% of the HIV-infected cohort. This is exactly the same proportion reported in another study performed in a population of HIV-infected women from Santos, a city near Sa˜o Paulo [21], confirming the high frequency of multiple infections in Brazilian HIV seropositive women. Palefsky et al. [2] observed that 403 (23%) among 1778 HIV+ American women harbored two or more HPV types, with 3% presenting six or more HPV types. In other studies, the extent of multiple infection may have been underestimated due to use of PCR-RFLP techniques. Sun et al. [9] found a cumulative HPV prevalence of 56% in a longitudinal study but reported difficulties in RFLP typing L1 PCR products in only 3% of the samples, which they attributed to infection with multiple types and Ellerbrock et al. [4] employing the MY09/11 + RFLP strategy [22] observed

229

multiple types in only 31 (12%) of their New York-Newark HIV-infected women cohort. These results suggest that the prevalence of infection with multiple HPV types in HIV+ women is higher than among HIV women, but the true prevalence of multiple infections on any group seems to have been underestimated by previous studies applying methods such as PCR-RFLP and Southern blot. For example, Peyton et al. [23] reported a prevalence of HPV multiple infections of 17.5% (677/2348) on a casuistic of young women attending routine gynecologic care clinics in New Mexico, USA. This high rate can be partially attributed to characteristics of the study population but also to the increase on sensitivity and broader HPV types range of the PGMY reverse line blot method. Results presented in this work, using samples from HIVinfected women and the line blot HPV detection and typing system, resemble our previous data obtained with part of the same material but employing the Lipa HPV detection and typing system [1]. Both methods apply the same biochemical principles but Lipa amplifies a fragment of only 62 bp while the line blot spans a 450-bp amplicon, both from the HPV L1 (capsid) gene. The question of why HIV-immunocompromised women are infected by multiple HPV types deserves further investigation. This may be the consequence of a greater frequency of unprotected sexual contact. This study has not collected information on sexual activity so we were unable to address this issue. Another explanation would be based on a failure of HPV type-specific immunity. Conceivably, immunocompetent hosts are able to control and eradicate infection with most HPV types, while types more commonly seen such as HPV 16, 53, 6 and 11 are able to evade host response by mechanisms that are as yet undefined. Thus, an immunocompromised host would provide an environment where those ‘‘less-adapted’’ HPV types could replicate. A relationship between HPV replication, as reflected by viral load, and decreased host immunity, measured by CD4 counts, was identified in this, as well as in another study [24]. Patients with altered smears had a mean CD4 count of 233 cells/Al and an HPV high-risk viral load of 919.6 which is considered very high (Table 3). Even more, the prevalence of Pap III was as high as 31% in patients with CD4 < 200 cells/Al (Table 2). Persistence and oncogenic potential of distinct HPV types are well-established factors conferring risk for the development of CIN and more advanced lesions [25], a fact also evidenced in HIV-infected women [26]. On this crosssectional study, a single collection of cervical cells was performed so the role of persistent infections in the development of cervical lesions could not be investigated. Moreover, the presence of multiple types makes impossible to attribute the prevalent lesions to one particular HPV type (Fig. 2). In contrast to the findings of Fife et al. [27], our data do not support an association between multiple types and cervical dysplasia. This is well illustrated in Table 4 where

230

J.E. Levi et al. / Gynecologic Oncology 92 (2004) 225–231

Fig. 2. HPV types detected on HIV+ patients with a Class III Pap smear (N = 45, box lists HPV types detected on each subject). *In two samples that were negative for the PCR-reverse hybridization method, HPV-DNA was detected by alternative methods and DNA sequencing has shown HPV 74 in one case and HPV 70 on the other. §DNA sequencing has demonstrated the presence of HPV 44.

despite higher HPV and HIV viral loads, patients infected by z 3 HPV types are not statistically different, on Pap III prevalence and CD4 levels, from those infected by a unique HPV type . Discrepancy with the data from Fife et al. [27] could be attributed to the distinct populations studied. HIV+ women probably have a higher basal prevalence of multiple HPV types that may obscure the association to cervical dysplasia, while they have taken as reference HIV women with normal smears, a population that apparently has only a few multiple infections. Their control group, composed of patients with normal cytology, had a mean of 1.04 HPV types while our Pap I and II presented means of 2.4 and 2.5 HPV types respectively. Similarly to our results, Chang et al. [28] found a decreasing number of HPV types with worsening of cytological grade. In agreement, we observed that the mean number of HPV types does not increase significantly with worsening cytological class or diminished CD4 counts. Recently, a new HPV typing and detection microarray system was applied to 1983 samples from Korean women and multiple HPV infection was associated less frequently with cervical carcinoma and precancerous lesions compared with normal cytology [29]. In addition, Rousseau et al. [30] on their prospective study, using PCR/reverse hybridization, reported that persistence of HPV infection on Brazilian HIV women,

increases the likelihood of acquisition of another HPV type, while co-infection with multiple HPV types did not influence persistence, the most important factor leading to cervical abnormalities. Herrero et al. [31] also described a similar prevalence of multiple infections among all grades of cervical neoplasia, in a cross-sectional study of 9175 women from Costa Rica. They obtained a higher odds ratio for HSIL and cancer for patients infected by HPV 16 alone in comparison to those infected by HPV 16 plus at least one additional type. The same observation was corroborated by a pool of 11 casecontrol studies on HPV types detected in cervical squamous cell carcinomas, where the odds ratio for a single HPV infection was higher than for multiple infections [32]. The concept of clonality proposes that dysplastic lesions arise from an HPV-infected cell stimulated to divide. Our data may be interpreted as if most cervical intra-epithelial lesions are caused by one HPV type while other co-detected types are likely minor players, and possibly do not confer a significant additional risk of neoplasia. In corroboration, it has been previously noticed that on lesions harboring coinfections, one HPV type is proliferating and the other latent, the former being likely responsible for pathogenesis, and the latter a bystander [15]. However, the role played by multiple infections on the establishment and development of SIL will be better addressed by prospective studies.

J.E. Levi et al. / Gynecologic Oncology 92 (2004) 225–231

Acknowledgments We thank Dr. Janet Kornegay (Roche Molecular Systems, Alameda, CA) for kindly providing the PCRreverse hybridization kits and Dr. Michael M. Lederman (Division of Infectious Diseases, Case Western University, Cleveland, OH) for reviewing the manuscript.

[16]

[17]

[18]

References [19] [1] Levi JE, Kleter B, Quint WG, Fink MC, Canto CL, Matsubara R, 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 – 5. [2] Palefsky JM, Minkoff H, Kalish LA, Levine A, Sacks HS, Garcia P, 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 – 36. [3] Klein RS, Ho GY, Vermund SH, Fleming I, Burk RD. Risk factors for squamous intraepithelial lesions on Pap smear in women at risk for human immunodeficiency virus infection. J Infect Dis 1994;170: 1404 – 9. [4] Ellerbrock TV, Chiasson MA, Bush TJ, Sun XW, Sawo D, Brudney K, et al. Incidence of cervical squamous intraepithelial lesions in HIVinfected women. JAMA 2000;283:1031 – 7. [5] Spitzer M. Lower genital tract intraepithelial neoplasia in HIV-infected women: guidelines for evaluation and management. Obstet Gynecol Surv 1999;54:131 – 7. [6] Nuovo GJ, Babury R, Calayag PT. Human papillomavirus types and recurrent cervical warts in immunocompromised women. Mod Pathol 1991;4:632 – 6. [7] Ho GY, Bierman R, Beardsley L, Chang CJ, Burk RD. Natural history of cervicovaginal papillomavirus infection in young women. N Engl J Med 1998;338:423 – 8. [8] Wallin KL, Wiklund F, Angstrom T, Bergman F, Stendahl U, Wadell G, et al. Type-specific persistence of human papillomavirus DNA before the development of invasive cervical cancer. N Engl J Med 1999;341:1633 – 8. [9] Sun XW, Kuhn L, Ellerbrock TV, Chiasson MA, Bush TJ, Wright Jr TC. Human papillomavirus infection in women infected with the human immunodeficiency virus. N Engl J Med 1997;337:1343 – 9. [10] Ahdieh L, Munoz A, Vlahov D, Trimble CL, Timpson LA, Shah K. Cervical neoplasia and repeated positivity of human papillomavirus infection in human immunodeficiency virus-seropositive and -seronegative women. Am J Epidemiol 2000;151:1148 – 57. [11] Kleter B, van Doorn LJ, Schrauwen L, Molijn A, Sastrowijoto S, ter Schegget J, et al. Development and clinical evaluation of a highly sensitive PCR-reverse hybridization line probe assay for detection and identification of anogenital human papillomavirus. J Clin Microbiol 1999;37:2508 – 17. [12] 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 – 7. [13] Richart RM. A modified terminology for cervical intraepithelial neoplasia. Obstet Gynecol 1990;75:131 – 3. [14] Roda Husman AM, Walboomers JM, Van Den Brule AJ, Meijer CJ, Snijders PJ. The use of general primers GP5 and GP6 elongated at their 3V ends with adjacent highly conserved sequences improves human papillomavirus detection by PCR. J Gen Virol 1995;76: 1057 – 62. [15] Nuovo GJ, Darfler MM, Impraim CC, Bromley SE. Occurrence of

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

231

multiple types of human papillomavirus in genital tract lesions. Am J Pathol 1991;138:53 – 8. Franco EL, Villa LL, Sobrinho JP, Prado JM, Rousseau MC, Desy M, et al. Epidemiology of acquisition and clearance of cervical human papillomavirus infection in women from a high-risk area for cervical cancer. J Infect Dis 1999;180:1415 – 23. Lazcano-Ponce E, Herrero R, Munoz N, Cruz A, Shah KV, Alonso P, et al. Epidemiology of HPV infection among Mexican women with normal cervical cytology. Int J Cancer 2001;91:412 – 20. Palefsky JM. Anal squamous intraepithelial lesions: relation to HIV and human papillomavirus infection. J Acquir Immune Defic Syndr 1999;21(Suppl. 1):S42 – 8. Brown DR, Schroeder JM, Bryan JT, Stoler MH, Fife KH. Detection of multiple human papillomavirus types in Condylomata acuminata lesions from otherwise healthy and immunosuppressed patients. J Clin Microbiol 1999;37:3316 – 22. Bradshaw BR, Nuovo GJ, DiConstanzo D, Cohen SR. Human papillomavirus type 16 in a homosexual man. Association with perianal carcinoma in situ and condyloma acuminatum. Arch Dermatol 1992; 128:949 – 52. Goncalves MA, Massad E, Burattini MN, Villa LL. Relationship between human papillomavirus (HPV) genotyping and genital neoplasia in HIV-positive patients of Santos City, Sao Paulo, Brazil. Int J STD AIDS 1999;10:803 – 7. Bernard HU, Chan SY, Manos MM, Ong CK, Villa LL, Delius H, et al. Identification and assessment of known and novel human papillomaviruses by polymerase chain reaction amplification, restriction fragment length polymorphisms, nucleotide sequence, and phylogenetic algorithms. J Infect Dis 1994;170:1077 – 85. Peyton CL, Gravitt PE, Hunt WC, Hundley RS, Zhao M, Apple RJ, et al. Determinants of genital human papillomavirus detection in a US population. J Infect Dis 2001;183:1554 – 64. Palefsky JM, Holly EA, Ralston ML, Da Costa M, Greenblatt RM. Prevalence and risk factors for anal human papillomavirus infection in human immunodeficiency virus (HIV)-positive and high-risk HIVnegative women. J Infect Dis 2001;183:383 – 91. Ho GY, Burk RD, Fleming I, Klein RS. Risk of genital human papillomavirus infection in women with human immunodeficiency virusinduced immunosuppression. Int J Cancer 1994;56:788 – 92. Ahdieh L, Klein RS, Burk R, Cu-Uvin S, Schuman P, Duerr A, et al. Prevalence, incidence, and type-specific persistence of human papillomavirus in human immunodeficiency virus (HIV)-positive and HIVnegative women. J Infect Dis 2001;184:682 – 90. Fife KH, Cramer HM, Schroeder JM, Brown DR. Detection of multiple human papillomavirus types in the lower genital tract correlates with cervical dysplasia. J Med Virol 2001;64:550 – 9. Chang DY, Chen RJ, Lee SC, Huang SC. Prevalence of single and multiple infection with human papillomaviruses in various grades of cervical neoplasia. J Med Microbiol 1997;46:54 – 60. An HJ, Cho NH, Lee SY, Kim IH, Lee C, Kim SJ, et al. Correlation of cervical carcinoma and precancerous lesions with human papillomavirus (HPV) genotypes detected with the HPV DNA chip microarray method. Cancer 2003;97:1672 – 80. Rousseau MC, Pereira JS, Prado JC, Villa LL, Rohan TE, Franco EL. Cervical coinfection with human papillomavirus (HPV) types as a predictor of acquisition and persistence of HPV infection. J Infect Dis 2001;184:1508 – 17. Herrero R, Hildesheim A, Bratti C, Sherman ME, Hutchinson M, Morales J, et al. Population-based study of human papillomavirus infection and cervical neoplasia in rural Costa Rica. J Natl Cancer Inst 2000;92:464 – 74. Mun˜oz N, Bosch FX, de Sanjose S, Herrero R, Castellsague X, Shah KV, et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med 2003; 348:518 – 27.