Assessing Urine Human Papillomavirus Polymerase Chain Reaction Testing As a Tool for Screening Anal HPV Infection in HIV-Positive MSM

Abstract

Multiple types of human papillomavirus (HPV) are responsible for most cervical cancers but also cause anal cancers—especially in HIV-positive patients. Furthermore, men who have sex with men (MSM) are twice as likely to develop anal cancers as non-MSM. A simple screening test for HPV infection would be useful in these patients. The aim of our study was to evaluate the detection of HPV by real-time polymerase chain reaction (PCR) in urine as a marker of anal infection in MSM. The study included 52 HIV-positive MSM treated at Amiens University Hospital (Amiens, France). After obtaining informed consent, we performed an anal swab and gathered 10 mL of first-void urine. Samples were extracted and amplified in a real-time PCR. Genotypes were determined with a PapilloCheck® system (Greiner Bio-One, Frickenhausen, Germany). The anal test was the gold standard for calculating the characteristics of the urine test. The sensitivity of the urine test for diagnosing anal HPV infection was 15%, the specificity was 66%, the positive predictive value was 87.5%, and negative predictive value was 4.5%. The prevalence of anal HPV infection in the study population was 94%. Genotype 42 was the most common. The anal HPV viral load was significantly lower in men in a stable relationship than in single men. However, there was no statistically significant relationship between anal viral load and anal intraepithelial lesions. We conclude that urine-based HPV is a poor predictor of anal HPV infection in HIV-positive MSM.

Introduction

T he multiple types of human papillomavirus (HPV) are potentially pathogenic DNA viruses. Certain HPV genotypes are responsible for the majority of cervical cancers.¹ These observations prompted the production of a vaccine, the administration of which has been eligible for reimbursement by the French public health insurance system since 2008. However, HPV also infect men and can trigger the development of condyloma and anal cancers.² Men who are HIV positive are especially susceptible to HPV infections and, subsequently, to anal cancer.^3

–6 Furthermore, men who have sex with men (MSM) are approximately twice as susceptible to HPV infection as men who do not have sex with men,^3,7

–10 and HPV prevalence is approximately twice as high in HIV-positive MSM than in HIV-negative MSM.^11
–13 Several different HPV genotypes are often found in anal samples, especially in MSM and even more so in HIV-positive patients.^14,15 Of the high-risk genotypes, HPV-16 is the most prevalent, followed by HPV-26, -30, and -58. However, many differences from one population to another have been reported and the genotype results cannot be generalized.^14,16,17

In our experience and that reported by colleagues, the detection of HPV anal lesions and infection is not easy to perform or undergo; we therefore decided to look for an easier way to screen for this virus.^18,19 Given that HPV is a multisite infection and can be found everywhere on the genital organs,^12,20,21 we chose to test urine samples as a potentially simple way of detecting HPV anal infections in a high-prevalence population. Whereas anal cytology and polymerase chain reaction (PCR) are well-studied tools, there is not yet a standard of care. However, Palesky et al.^6,8 showed in 1998 that PCR can be used to detect HPV infection. Previous studies have tested urine with in a PCR assay^22

–25 but few determined the accuracy of urine testing as a screen for anal infection²⁵ and none used with real-time PCR. The prevalence of HPV in urine samples was as high as 30.9% in the study by Costa et al.,²² although Giuliano et al.²⁰ showed that an urethral swab was the worst sample for HPV detection (urine sampling was abandoned, due to the poor levels of sample adequacy and HPV detection). Nevertheless, PCR in urine samples can be successfully performed, as reported by Payan et al.²⁶ and Cuschieri²⁷ (with a sensitivity of between 37% and 56%). These observations prompted us to perform this study.

Methods

Setting

This cross-sectional study was performed at Amiens University Hospital, a 1400-bed teaching hospital in Amiens, France.

Patients

Inclusion criteria was: adult (over 18 years of age) HIV-positive MSM who presented at our infectious diseases outpatient clinic between 2008 and 2009. Noninclusion criterion was failure to give consent or any anal operation that prevented sampling. Exclusion criteria was incomplete data or incomplete sampling.

Outcomes

The primary outcome was the performance of a urine test for screening for anal HPV infection; secondary outcomes were (1) the ability of the anal HPV real-time PCR to detect abnormal cytologies and (2) the determinants of HPV viral load. Informed consent was obtained from all participants and the study was approved by the Nord-Picardie II independent ethics committee.

Clinical data

On the day of inclusion, a preestablished questionnaire was used to collect separate data on sexual behavior (recorded by the patient) and clinical signs (recorded by the clinician): the relationship status (single or living with a partner), the frequency of anal and nonanal intercourse, the lifetime sexual preference (inserter, receptor, or both), the number of partners per month, the lifetime number of partners, the lifetime use of condom, smoking habit, age at first sexual intercourse, history of HPV diseases for the patient himself and his partner, the history of the HIV disease (CD4 count, viral load and any history of opportunistic diseases), and the history of follow-up by a proctologist. The physician noted the presence of anal or penile condyloma or any other symptoms and then took a blind anal sample (without anoscopy), as described below.

Samples

On the same day, approximately 10 mL of first-void urine was sampled and stored at −80°C prior to virological testing. The anal sample was taken with a Dacron swab (Eurotubo, Rubi, Spain) with three 360° rotations. The sample was suspended in PreservCyt^® cell transport medium (as used in the ThinPrep Pap Test from Cytyc Corporation, Boxborough, MA). Two samples were collected for all patients.

Cytology

Tests were reported according to a simplified version of the 2001 Bethesda system of reporting, as follows: no lesion (including cases with pronounced inflammation), atypical squamous cells of undetermined significance (ASC-US), low- or high-grade intraepithelial squamous lesion, or suspected carcinoma. To rule out false-negatives, a second round of cytological analysis was performed independently by two cytologists blinded to the first set of results. The results were concordant in 75% of cases; a third (consensus) round was made in the remaining cases.

Virology

DNA extraction.

Two milliliters of urine were centrifuged (at 18,000g, for 30 min) and sediment was resuspended in 200 μL of sterile water, prior to extraction. DNA extraction was performed on 200 μL of resuspended urine sediment or 200 μL of PreservCyt with an EZ1 Biorobot (Qiagen, Courtaboeuf, France; elution, 50 μL).

Real-time PCR.

Five microliters of extracted DNA were amplified with SPF₁₀ primers in a Sybr-Green real-time PCR (as described by Payan²⁶) on an AB7300^® thermocycler (Applied Biosystems, Carlsbad, CA). Dilutions of an HPV16 complete genome plasmid (kindly provided by EM de Villiers, Deutsches Krebsforschungszentrum, Heidelberg, Germany) were used as standards to evaluate the samples' viral loads. As an internal control, GAPDH was assayed in each sample.²⁶

Genotyping.

All samples were genotyped with the PapilloCheck® system (Greiner Bio-One, Frickenhausen, Germany), according to the manufacturer's instructions.

Statistical analysis

The gold standard for anal HPV infection was detection of HPV DNA in anal samples with the quantitative assay. We calculated the urine test's sensitivity, specificity and the positive and negative predictive values. In view of our primary outcome, we did not need to calculate the size of the study population. Continuous variables (age, CD4 count, time since HIV onset and anal and urinary viral loads) were expressed as the mean±standard deviation (if normally distributed) or as the median and range. Categorical variables were expressed as the number and percentage. Comparisons of anal viral load were performed with Student's t tests or an analysis of variance for continuous variables. Potential correlations between HPV viral load, CD4 count, and CD4/CD8 ratio were probed with a Pearson correlation test. The threshold for statistical significance was set to 0.05. All statistical tests were two-sided. Database management and all statistical analyses were performed with SPSS software (version 11, SPSS Inc., Chicago, IL).

Results

Ninety patients met the inclusion criteria and 55 were included (giving a study acceptance rate of 61%). Three were subsequently withdrawn (2 for incomplete clinical data and 1 for a missing urine sample) and so the full analysis set comprised 52 patients.

Population characteristics

The population's mean age was 44.9±9.4 and the mean age at first anal intercourse was 21.2±2.1 for inserters and 20.8±4.6 for receivers. Sixty percent of the patients (n=33) had sexual intercourse between 1 and 4 times a month. Twenty-eight patients (54%) were single. Thirty-three patients (66%) had had more than 20 lifetime sexual partners. Nineteen patients (37%) stated that they had never use a condom in their life, and 14 (27%) were being monitored by a proctologist.

At the time of the study, the patients had been infected with HIV for an average of 11.6±6.6 years. The median CD4 count was 576/mm³ (range, 182–1267); 3 patients had a CD4 count below 300/mm³, 7 had a history of opportunistic infection, and 15 had a detectable HIV viral load (i.e., >50 copies per milliliter). Thirty-three patients were being treated with highly active antiretroviral therapies (details are shown in Table 1).

Table 1.

Patients Characteristics

	All, n=52 (%)
Age (years)	44.9±9.4
Sexual course: Inserter	8 (15.4)
Receptor	10 (19.2)
Both	32 (61.5)
Frequency (>1/week)	9 (21)
Lifetime number of sexual partner more than 20	33 (63.4)
Always use condom	32 (61.5)
Personal condyloma history	46 (88.5)
Partner condyloma history	6 (17.1)
Clinical condyloma	4 (7.6)
Circumcision	7 (13.5)
Urethritis history	14 (27)
Smoking	25 (48)
Patients undergoing HAART	33 (63.5)
Median CD4 count (/mm³)	576
Detectable HIV viral load	15 (28.8)
Length of HIV history (years)	11.6±6.6
Opportunistic infection history	7 (13.5)
History of proctologic follow up	14 (27)

HAART, highly active antiretroviral therapy.

Analytical performance of the urine PCR test

The urine-based PCR was rarely positive (n=8 of 52), whereas the prevalence of HPV infection was high in our population (n=49 out of 52) (for details, see Table 2). The sensitivity was 15%, the specificity was 66%, the positive predictive value was 87.5%, and the negative predictive value was 4.5%.

Table 2.

Evaluation of Human Papilloma Virus Viral Load Urine Test

	Positive anal HPV viral load	Negative anal HPV viral load	Total
Positive urine HPV viral load	7	1	8 (15.4)
Negative urine HPV viral load	42	2	44 (84.6)
Total	49 (94.2)	3 (5.8)	52 (100)

Anal HPV viral load was the gold standard for the analysis of urine test.

Percent are in brackets.

HPV, human papilloma virus.

Cytology

No anal swabs led to the suspicion of carcinoma, 36 samples were normal (with severe inflammation in 1), 13 samples were classified as low-grade intraepithelial lesions, 1 was classified as ASC-US, 2 were classified as high-grade intraepithelial lesions, and 1 anal swab did not yield enough cells to perform the analysis. No patients with intraepithelial lesions had a positive urinary PCR test.

Determinants of high viral load in anal samples

The viral load was significantly higher in single men than in men living with a partner (p=0.008). However, there were no other statistically significant correlations—neither with the cytology results, condom use, the frequency and type of anal intercourse, the age at first anal intercourse, the number of sexual partners nor with the HIV/HPV history (opportunistic infections, condyloma, etc.). We did not find any correlation between HPV viral load and the CD4 count (r=0.38).

Genotypes in isolates from anal samples

Approximately two-thirds of the patients were infected with two or more HPV genotypes: 13 patients had only one genotype, 13 patients had two, 5 patients had three, 9 patients had four, 3 patients had five, and 1 patient had seven. The number of genotypes found per patient was not associated with either the frequency of sexual intercourse or the number of partners. No genotype could be detected for 7 patients and genotype 42 was detected in 1 patient with a negative viral load in the PCR. The most frequently found genotypes were 42 (12 times), 16 and 56 (10 times each), and then 44, 6, and 70 (7 times each). Details are shown in Table 3. Patients with three or more genotypes had significant lower HPV viral loads than those with one or two genotypes (p=0.018).

Table 3.

Comparison of Anal and Urine Specimens for Human Papilloma Virus Viral Load and Genotyping

	Urine specimens		Anal specimens
Patient	VL (log/mL)	Genotyping	VL (log/mL)	Genotyping
1		neg	7,02	56/6
2		neg	8,64	16/11
3		neg	8,40	31/33/42
4	4,45	neg	7,28	6
5		neg	7,66	11/44
6	3,23	neg	2,34	neg
7		neg	8,62	51/40
8		neg	8,38	56/68/6/43
9		neg	6,37	6
10		neg	5,10	51/70/44
11		neg	6,07	31
12		neg	5,31	68/70
13	1,96	neg	3,31	42
14		neg	3,05	neg
15		neg	7,54	11
16		neg	4,13	16/42
17		neg	5,25	40
18	1,95	42	8,57	18/45/52/51/66/40/42
19		neg	5,41	16/39
20		neg	6,20	56/42/44
21	3,39	43	7,61	18/70/6
22		neg	7,13	neg
23		neg	5,00	neg
24	6,83	6	7,49	6/44
25		neg	6,77	16/35/59/56/70
26		neg	7,11	16/31/56/39
27		neg	8,81	16/18/11/42
28		neg	7,15	42
29		neg	6,34	16/18/56/43
30		neg	6,44	16/33/58/56
31		neg	7,67	18/56/68/70/43
32		neg	6,35	16/18/56/43
33		neg	6,94	16/11
34		neg	3,90	70
35		neg	8,09	31/33/51/39/42
36		neg	3,75	39
37		neg	5,89	neg
38		neg	1,87	58/42
39		neg	3,64	58/42
40		neg	4,87	59/39
41		neg	7,21	45/82/6/42
42		neg	0,00	neg
43		neg	6,47	neg
44		neg	3,22	68/40
45		neg	6,64	44
46		neg	4,01	70
47		neg	0,00	42
48		neg	5,40	31/33/43/44
49		neg	9,24	56/11/44
50	4,33	neg	3,75	51
51		neg	5,64	45/33/68/44
52	3,05	neg	0,00	neg

VL, viral load; HPV, human papilloma virus.

Genotypes in isolates from urine samples

Only three genotypes were found in urine: 6, 42, and 43. There was no concordance between anal and urine genotyping: the patient with 7 genotypes in the anal sample had only 1 genotype in the urine (genotype 42 was found in both samples), the patient with genotype 43 in the urine had anal genotypes 18, 70, and 6 and the patient with urine genotype 6 had anal genotypes 6 and 44.

Discussion

This is the first published study to have assessed urine HPV real-time PCR as a tool for anal HPV infection screening. Although urine testing is conceptually attractive in view of its speed and ease of performance, it is not a reliable test for detection of anal HPV infection, especially for HPV-induced anal intraepithelial lesions in HIV-positive MSM.

Anal surveys in these patients are important; in France, HIV-positive MSM are advised to undergo annual screening for anal lesions with a proctologist.²⁸ However, many authors have reported the difficulties associated with successfully implementing this type of survey, as illustrated by (for example) the 10% anal screening rate in an HIV clinic in Colorado and the finding that only 36% of HIV patients discuss anal health with their HIV care providers.^18,19 This surveying can be improved by up to a factor of 4 with an anal health program.¹⁸ Although another potential way of improving anal surveys would be urine testing, it seems that a urine test cannot meet this requirement.

The measurement of HPV in the urine raises several issues: the pathogenic colonization route, technical difficulties with the assay, and the diversity of the population concerned. Urinary tract HPV infections seem to have their own pathogenic characteristics, since the genotypes isolated in the urine differ from those found in an anal swab. However, some studies have shown that penile swabs provide representative samples for HPV detection.^25,29,30 Guiliano and colleague's study was undertaken on heterosexual men (whose HIV status was not reported) and the prevalence of HPV was about 50% on the penile shaft, 35% on the coronal sulcus and glans and 34% on the scrotum.²⁵ Guiliano et al.²⁵ also showed that urethral samples were the worst and thus abandoned urine sampling because of the lack of HPV detection. Hernandez et al.²⁹ found the same prevalence results in a mixed population (MSM: 22%; HIV-positives: 4%), but did not test the urine. Weaver et al. ³⁰ tested 30 heterosexual, HIV-negative men and found the penile shaft sample to be more sensitive for HPV infection (24%) than other samples; again, urine was the least sensitive sample (6%).

Literature reports disagree on whether or not a urine sample is a good tool for HPV detection.^25,30,31 However, site sampling differences may be compounded by additional technical difficulties, such as the urine sample's cell content and the presence of PCR inhibitors.

All the samples in our study contained cells. It is known that a high cell content is essential for HPV detection in urine samples, even though Payan et al.^26,31 failed to observe a correlation between the HPV viral load and the cell content. The presence of PCR inhibitors was ruled out by a urine centrifugation/dilution protocol and the use of GAPDH as an internal PCR control. Furthermore, our positive and negative controls ruled out technical problems, since all urine and anal specimens were positive for GAPDH in the PCR assay. However, the low HPV detection rate in urine can also be explained by the assay used. The observed difference may be due to population variations, although the high prevalence of anal infection in the present study (94%) is similar to that reported by other authors.^13,32

The prevalence of HPV in the urine in the present study (13.5%) is lower than the value found in HIV-positive males in the Netherlands (27.2%)²⁴ and in Brazil (30.9%).²² However, the Brazilian study is not comparable with ours because only 75% of its patients had more than 200 CD4 cells/mm³ and 36% had fewer than 1000 copies of HIV RNA per milliliter.

The principal limitation of our study was the small study population (in terms of the secondary outcome), although this aspect does not affect our findings on the poor validity of urine testing in our population. However, our results should not be generalized. We included over half of our active case load; the patients' demographic characteristics made them representative of the HIV-positive MSM population in our hospital and similar to that of our region.

The clinically relevant point in a patient's prognosis is the detection of cytological abnormalities. Only 17 of the 52 included patients (32.7%) had cytological abnormalities. This value is lower than those found in other studies: a Canadian study found cytological abnormalities in 67% of patients,³² with 56% in an American study³³ and 74% in a German one.³⁴ This could mean that either our sampling technique or our cytological analysis (or both) were inefficient. Since anoscopy was not performed, the sampling technique may have been inadequate and may explain the lower rate of patients with abnormal cytology results. As explained in the Methods section, care was taken to enhance the sensitivity and specificity of the cytological screening. The disparities between these various results could also be explained by differences in the populations' habits and the assay methods.

This study was not designed to detect an association between HPV viral load and the presence of cytologically confirmed anal dysplasia and was underpowered in terms of the number of patients included. However, the apparent lack of a significant correlation does not mean that an association does not exist. Some studies have shown that an association between HPV viral load and cytology does exist, especially for grades 2 and 3 intraepithelial neoplasia in HIV-positive MSM.^35,36 Other studies have not yielded the same results.³⁷ However, the clinical significance of the HPV viral load has not been clearly established and further studies are needed. Nevertheless, some authors have shown that anal HPV screening has good sensitivity but poor specificity for the detection of anal, intraepithelial carcinoma.³² It is noteworthy that in the present study, the viral load was correlated with the stability of the patients' sex life. This probably also means that it was associated with the patients' sexual history and habits.

There is probably competition between different viruses because the viral load is lower when there are three or more HPV genotypes. In the present study, 70.4% of the patients had more than one genotype (median, 3; range, 2–7), and this has been often described.^{13,17,32,38,39} The most frequently found genotypes in our population were 42, 16, and 50 but this parameter varies strongly from one study population to another.¹⁷ We were unable to detect a specific genotype for 7 patients but the use of SPF₁₀ primers could have detected HPV types that are not measured in the PapilloCheck genotyping assay.

In conclusion, testing urine for HPV is not a reliable way to screen for anal HPV infections or intraepithelial lesions. There is a need to find other simple ways of surveying high-risk patients. Hence, more studies are required to establish whether or not the anal HPV viral load is a predictive factor for intraepithelial carcinoma.

Footnotes

Acknowledgments

The authors would like to thank all people involved in this study and especially Dr Jean-Fortuné Ikoli for his help with the cytological diagnoses. The study was funded with independent research resources provided by Amiens University Hospital. The study was approved by the Nord-Picardie II independent ethics committee.

Author Disclosure Statement

No competing financial interests exist.

References

Munoz

, Bosch

, de Sanjose

et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med, 2003; 348:518–527.

Palefsky

, Holly

, Gonzales

, Berline

, Ahn

, Greenspan

. Detection of human papillomavirus DNA in anal intraepithelial neoplasia and anal cancer. Cancer Res, 1991; 51:1014–1019.

Smits

, Bakker

, Jong

et al. High prevalence of human papillomavirus infections in urine samples from human immunodeficiency virus-infected men. J Clin Microbiol, 2005; 43:5936–5939.

Chin-Hong

, Husnik

, Cranston

et al. Anal human papillomavirus infection is associated with HIV acquisition in men who have sex with men. AIDS, 2009; 23:1135–1142.

Breese

, Judson

, Penley

, Douglas

Jr . Anal human papillomavirus infection among homosexual and bisexual men: Prevalence of type-specific infection and association with human immunodeficiency virus. Sex Transm Dis, 1995; 22:7–14.

Palefsky

, Holly

, Ralston

, Jay

. Prevalence and risk factors for human papillomavirus infection of the anal canal in human immunodeficiency virus (HIV)-positive and HIV-negative homosexual men. J Infect Dis, 1998; 177:361–367.

Surawicz

, Critchlow

, Sayer

et al. High grade anal dysplasia in visually normal mucosa in homosexual men: Seven cases. Am J Gastroenterol, 1995; 90:1776–1778.

Palefsky

, Holly

, Hogeboom

et al. Virologic, immunologic, and clinical parameters in the incidence and progression of anal squamous intraepithelial lesions in HIV-positive and HIV-negative homosexual men. J Acquir Immune Defic Syndr Hum Retrovirol, 1998; 17:314–319.

Abramowitz

, Benabderrahmane

, Ravaud

et al. Anal squamous intraepithelial lesions and condyloma in HIV-infected heterosexual men, homosexual men and women: Prevalence and associated factors. AIDS, 2007; 21:1457–1465.

10.

Daling

, Weiss

, Hislop

et al. Sexual practices, sexually transmitted diseases, and the incidence of anal cancer. N Engl J Med, 1987; 317:973–977.

11.

Palefsky

, Holly

, Efirdc

et al. Anal intraepithelial neoplasia in the highly active antiretroviral therapy era among HIV-positive men who have sex with men. AIDS, 2005; 19:1407–1414.

12.

Goldstone

, Palefsky

, Giuliano

et al. Prevalence of and risk factors for human papillomavirus (HPV) infection among HIV-seronegative men who have sex with men. J Infect Dis, 2011; 203:66–74.

13.

de Pokomandy

, Rouleau

, Ghattas

et al. Prevalence, clearance, and incidence of anal human papillomavirus infection in HIV-infected men: The HIPVIRG cohort study. J Infect Dis, 2009; 199:965–973.

14.

Sirera

, Videla

, Pinol

et al. High prevalence of human papillomavirus infection in the anus, penis and mouth in HIV-positive men. AIDS, 2006; 20:1201–1204.

15.

Vajdic

, van Leeuwen

, Jin

et al. Anal human papillomavirus genotype diversity and co-infection in a community-based sample of homosexual men. Sex Transm Infect, 2009; 85:330–335.

16.

van der Snoek

, Niesters

, Mulder

, van Doornum

, Osterhaus

, van der Meijden

. Human papillomavirus infection in men who have sex with men participating in a Dutch gay-cohort study. Sex Transm Dis, 2003; 30:639–644.

17.

Orlando

, Beretta

, Fasolo

et al. Anal HPV genotypes and related displasic lesions in Italian and foreign born high-risk males. Vaccine, 2009; 27,Suppl 1:A24–29.

18.

Kwong

, Cook

, Bradley-Springer

. Improving anal cancer screening in an ambulatory HIV clinic: Experience from a quality improvement initiative. AIDS Patient Care STDs, 2011; 25:73–78.

19.

Rosa-Cunha

, Cardenas

, Dickinson

, Metsch

. Addressing anal health in the HIV primary care setting: a disappointing reality. AIDS Patient Care STDs, 2010; 24:533–538.

20.

Giuliano

, Lee

, Fulp

, Villa

et al. Incidence and clearance of genital human papillomavirus infection in men (HIM): A cohort study. Lancet, 2011; 377:932–940.

21.

Scholefield

, Hickson

, Smith

, Rogers

, Sharp

. Anal intraepithelial neoplasia: Part of a multifocal disease process. Lancet, 1992; 340:1271–1273.

22.

Costa

, da Silva

, Arruda

, Duarte

, Casseb

. Prevalence of human papillomaviruses in urine samples of male patients infected with HIV-1 in Sao Paulo, Brazil. J Med Virol, 2009; 81:2007–2011.

23.

D'Hauwers

, Tjalma

. Screening for human papillomavirus: Is urine useful? Indian J Cancer, 2009; 46:190–193.

24.

Jong

, van Gorp

, Mulder

, Tol

, Smits

. Effect of HIV viral load, CD4 cell count and antiretroviral therapy on human papillomavirus prevalence in urine samples of HIV-infected men. Int J STD AIDS, 2009; 20:262–264.

25.

Giuliano

, Nielson

, Flores

et al. The optimal anatomic sites for sampling heterosexual men for human papillomavirus (HPV) detection: The HPV detection in men study. J Infect Dis, 2007; 196:1146–1152.

26.

Payan

, Ducancelle

, Aboubaker

et al. Human papillomavirus quantification in urine and cervical samples by using the Mx4000 and LightCycler general real-time PCR systems. J Clin Microbiol, 2007; 45:897–901.

27.

Cuschieri

, Nandwani

, McGough

et al. Urine testing as a surveillance tool to monitor the impact of HPV immunization programs. J Med Virol, 2011; 83:1983–1987.

28.

Yeni

. Prise en charge médicale des personnes infectées par le VIH. Rapport 2010. Paris: La documentation française, 2010; 331–349.

29.

Hernandez

, McDuffie

, Goodman

et al. Comparison of physician- and self-collected genital specimens for detection of human papillomavirus in men. J Clin Microbiol, 2006; 44:513–517.

30.

Weaver

, Feng

, Holmes

et al. Evaluation of genital sites and sampling techniques for detection of human papillomavirus DNA in men. J Infect Dis, 2004; 189:677–685.

31.

D'Hauwers

, Depuydt

, Bogers

et al. Urine versus brushed samples in human papillomavirus screening: study in both genders. Asian J Androl, 2007; 9:705–710.

32.

Salit

, Lytwyn

, Raboud

et al. The role of cytology (Pap tests) and human papillomavirus testing in anal cancer screening. AIDS, 2010; 24:1307–1313.

33.

Bean

, Chhieng

, Roberson

et al. Anal-rectal cytology: Correlation with human papillomavirus status and biopsy diagnoses in a population of HIV-positive patients. J Low Genit Tract Dis, 2010; 14:90–96.

34.

Kreuter

, Potthoff

, Brockmeyer

et al. Anal carcinoma in human immunodeficiency virus-positive men: Results of a prospective study from Germany. Br J Dermatol, 2010; 162:1269–1277.

35.

Alvarez

, de Pokomandy

, Rouleau

et al. Episomal and integrated human papillomavirus type 16 loads and anal intraepithelial neoplasia in HIV-seropositive men. AIDS, 2010; 24:2355–2363.

36.

Salit

, Tinmouth

, Chong

et al. Screening for HIV-associated anal cancer: Correlation of HPV genotypes, p16, and E6 transcripts with anal pathology. Cancer Epidemiol Biomarkers Prev, 2009; 18:1986–1992.

37.

Poizot-Martin

, Henry

, Benhaim

, Obry-Roguet

, Figarella

, Tamalet

. High level of HPV 16 and 18 DNA load in anal swabs from male and female HIV-1 infected patients. J Clin Virol, 2009; 44:314–317.

38.

Damay

, Fabre

, Costes

et al. Human papillomavirus (HPV) prevalence and type distribution, and HPV-associated cytological abnormalities in anal specimens from men infected with HIV who have sex with men. J Med Virol, 2010; 82:592–596.

39.

Dunne

, Nielson

, Stone

, Markowitz

, Giuliano

. Prevalence of HPV infection among men: A systematic review of the literature. J Infect Dis, 2006; 194:1044–1057.