Article Text

Short report
Prolonged antiretroviral therapy is associated with fewer anal high-grade squamous intraepithelial lesions in HIV-positive MSM in a cross-sectional study
  1. Agnès Libois1,
  2. Francesco Feoli2,
  3. Marcel Nkuize3,
  4. Marc Delforge1,
  5. Deborah Konopnicki1,
  6. Nathan Clumeck1,
  7. Stéphane De Wit1
  1. 1Department of Infectious Diseases, University Saint-Pierre Hospital, Université Libre de Bruxelles, Brussels, Belgium
  2. 2Department of Pathology, Jules Bordet Institute, Université Libre de Bruxelles, Brussels, Belgium
  3. 3Department of Gastroenterology, University Saint-Pierre Hospital, Université Libre de Bruxelles, Brussels, Belgium
  1. Correspondence to Dr Agnès Libois, Department of Infectious Diseases, University Saint-Pierre Hospital, 322 rue haute, Bruxelles 1000, Belgium; agnes_libois{at}


Objective HIV-positive men who have sex with men (MSM) are at increased risk of anal cancer. We evaluate the risk factors for anal high-grade squamous intraepithelial lesion (HSIL) (the precursor of anal cancer) in HIV-positive MSM.

Methods In this cross-sectional study within a cohort, 320 HIV-positive MSM were screened by anal cytology followed by high-resolution anoscopy (HRA) in case of abnormal cytology. Risk factors for anal HSIL were analysed.

Results Men were mostly middle-aged Caucasians with median CD4+ T lymphocytes of 638 cells/µL, 87% on combined antiretroviral therapy (cART) for a median of 5 years. 198 anal cytology samples were normal. In the 122 patients with abnormal cytology, HRA with biopsies were performed: 12% (n=15) normal, 36% (n=44) anal low-grade squamous intraepithelial lesion (LSIL) and 51% (n=63) anal HSIL. Comparing patients with or without anal HSIL (normal cytology or normal biopsy or LSIL), we found in multivariate analysis significantly fewer anal HSIL in patients with cART ≥24 months (OR 0.32 CI 95% 0.162 to 0.631, p=0.001).

Conclusions Prolonged cART (≥24 months) is associated with fewer anal HSIL.

  • HIV
  • HPV

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Over the last 30 years, incidence of anal cancer has increased. It is still an uncommon cancer in the general population but in HIV-positive men who have sex with men (MSM), incidence is particularly high, with rates of 131/100 000 persons-years (80 times higher than in HIV-negative individuals).1

Although controversial, combined antiretroviral therapy (cART) does not seem to be associated with a decreased risk of anal cancer.2 ,3 This fact combined with the longer life expectancy and ageing of patients who are HIV positive, raises concerns for a continuous increase in the incidence of anal cancer.

As with cervical cancer, most cases of anal cancer are caused by persistent infection with high-risk human papillomavirus (HPV), and anal high-grade squamous intraepithelial neoplasia (HSIL) is the precursor of anal cancer.4

Screening and treatment of cervical dysplasia have decreased cervical cancer mortality in women. Although a similar strategy to prevent anal cancer has not demonstrated its efficacy in a randomised study, screening for anal HSIL in a high-risk population as HIV-positive MSM is increasingly proposed in clinical practice, as in our centre.

Our aims are to evaluate the risk factors for anal HSIL in HIV-positive MSM and to evaluate the potential impact of cART.


Since June 2011, a systematic anal cancer screening programme is offered to all HIV-positive MSM followed at the AIDS reference centre of Saint-Pierre University Hospital in Brussels, where 2950 patients who are HIV positive (33% MSM) are currently in care and included in the Saint-Pierre HIV Cohort. Anal swab for cytology was followed by high-resolution anoscopy (HRA) in case of any significant abnormalities at cytology. Between June 2011 and February 2013, patients who performed this screening and had appropriate follow-up (normal swab or HRA with biopsies in case of abnormal cytology) were included in this cross-sectional study.

The sample (Copan Flocked swabs, FLOQSwabs) was collected by the clinician or self-collected by the patient and placed in a liquid cytology media (BD-SUREPATH). Anal cytology was categorised according to the revised 2001 Bethesda System: HSIL (high-grade squamous intraepithelial lesions), LSIL (low-grade squamous intraepithelial lesions) and ASC-US (atypical squamous cells of undetermined significance) and ASC-H (atypical squamous cells, cannot rule out a high-grade lesion). In case of abnormal cytology (including ASC-US), HRA with biopsies was performed by a single gastroenterologist (MN), specifically trained in HRA. Results of biopsies (histology) were classified as normal, anal LSIL or anal HSIL. Anal LSIL correspond to anal intraepithelial neoplasia (AIN) 1 and anal HSIL to AIN 2 and AIN 3. Patients with no anal HSIL were defined as having normal cytology or normal biopsy or LSIL at biopsy.

Immunostaining for Ki67 and P16 was routinely carried out on all biopsies as help for histological grading of lesions. The same expert pathologist (FF) assessed all the samples. The first 250 anal cytology samples were blindly reviewed by another pathologist with international experience in quality control (G Negri, Italy).

Clinical and pathological data were collected prospectively.

Ethics statement

Data collection and analyses were approved by the local ethics committee of Saint-Pierre University Hospital, Brussels (No. O.M. 007). All data were treated and analysed as anonymised.

Statistical methods

Factors associated with HSIL at biopsy were identified by univariate analysis using the Mann–Whitney U-test for quantitative data and the χ2 test or Fisher's exact test for qualitative data. Variables with a p≤0.05 in univariate analysis were included in a multivariable model. Multivariate analyses were performed by logistic regression. Analyses were performed using SAS statistical software (V.9·2; SAS Institute, Cary, North Carolina, USA).

Duration of HIV infection was calculated as the time between HIV diagnosis and the date of anal swab. Duration of sustained HIV undetectability was calculated as the interval from the midpoint between the date of the last detectable HIV load and the first date of undetectable HIV load (<50 copies/mL) to the midpoint between the date of the last undetectable HIV load and either the date of the first subsequent detectable HIV load or the date of anal swab.


Patients characteristics

Anal smears and HRA with biopsies in cases of abnormal cytology were performed in 320 HIV-positive MSM during the study period. The majority of patients were middle-aged Caucasians on cART for several years (table 1).

Table 1

Univariate and multivariate analyses comparing characteristics of HIV-positive MSM with or without anal HSIL

Results of anal swabs and biopsies

Among the 320 anal swabs, 122 were abnormal (39%): 28% (n=35) showed ASC-US, 53% (n=65) LSIL, 10% (n=13) HSIL and 7% (n=9) ASC-H.

Twelve per cent (15/122) of biopsies were normal, 36% (n=44) showed LSIL and 51% (n=63) showed anal HSIL. Among the 15 normal biopsies, 9 had ASC-US and none had HSIL at cytology.

Risk factors for HSIL

In univariate analysis, comparing patients with anal HSIL (n=63) and patients with no HSIL (normal swab (n=198) or normal biopsies (n=15) or LSIL (n=44)), being on cART for 24 months or more was significantly associated with no anal HSIL (80% of patients without HSIL had been treated for more than 24 months vs 53% of patients with HSIL, p<0.0001) (table 1). As expected, having a viral load (VL) <50 copies/mL for 24 months or more was also significantly associated with less anal HSIL. Older age was associated with less HSIL, and other variables were not.

cART and VL <50 copies/mL for ≥24 months are two variables linked and to avoid colinearity in the multivariate analysis, we selected cART ≥24 months in the model. In multivariate analysis with age and cART ≥24 months, only cART ≥24 months was significantly associated with less HSIL (OR 0.32, 95% CI 0.162 to 0.631, p=0.001).

Similar results were found if patients with normal cytology or biopsy samples were compared with patients with anal HSIL (ie, excluding patients with LSIL from the analysis).


In this cohort of 320 HIV-positive MSM with efficient cART for several years, we found a significant association between cART for 24 months or more and absence of HSIL. This finding suggests a protective effect of prolonged cART on anal cancer precursors.

There are strengths and limitations in our study. Among strengths, data were collected prospectively among a large sample of HIV-positive MSM. All HRA were performed by a single proctologist specifically trained in HRA. The same reference pathologist examined all cytology and biopsy, with the first 250 samples reviewed by an external expert for quality control. Among limitations, HRA was performed only in patients with abnormal cytology, which could underestimate anal HSIL. HPV infection and HPV typing were not performed. The prevalence of anal HPV in HIV-positive MSM is known to be ≥90%; therefore, HPV screening is highly sensitive but with a poor positive predictive value to detect HSIL.1 Data on anal receptive intercourse and number of sex partners known to be risk factors for anal HSIL were not available.5

Conflicting results have been published on the potential impact of cART on the development of HSIL: most studies in HIV-positive MSM not showing a protective effect of cART on anal HSIL were done in the early cART era (with ‘old regimens’ having a lower ability to suppress viral load due to poorer adherence related to higher pill burden and side effects).6 In a recent study of Piketty et al,3 cART duration was equal or less than 2 years and this may have been too short to provide a protective effect. In contrast, in a Canadian longitudinal study of 247 HIV-positive MSM, de Pokomandy et al5 have shown a significant association between less anal HSIL and cART >4 years. In a recent retrospective study among the large male US veterans cohort, the risk of anal cancer decreased when the viral load was undetectable during more than 60% of the time.7 In women also, recent studies argue in favour of a protective effect of prolonged cART on HPV infection, cervical dysplasia and cancer. Most recent studies have shown a protective impact of cART on cervical dysplasia, especially in patients with sustained HIV suppression.8 It has been shown in women that it could take up to 36 months to decrease cervical HPV infection under efficient cART.9 In Taiwanese HIV-infected women, adherence above 85% and use of cART for at least 3 years reduced the risk of cervical neoplasia.10

It is plausible that sustained HIV control by cART over a period of several years is needed to provide a protective effect on HPV-induced dysplasia. The protective effect of prolonged cART against anal HSIL, found in our study and for the first time in Europe, is consistent with the above-mentioned studies. Recent studies show a stabilisation of incidence of anal cancer in HIV-positive patients.1 ,2 Whether this current plateau could be linked to better and longer HIV suppression with modern cART is likely.

Early start of cART regardless of CD4 cell count is now recommended by numerous international guidelines (WHO, EACS, DHHS, BHIVA, IAS-USA). Early initiation of cART will result in prolonged cART and could protect HIV-positive MSM from anal cancer precursor. Future research is needed to prove it and to assess the potential protective effect of early cART on anal cancer.


I thank Dr S Cattell for the grammar corrections and Dr G Negri for reviewing the first 250 anal cytology samples.


Supplementary materials

  • Abstract in French

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Part of this study was presented at the 14th European AIDS Conference in Brussels, 16–19 October 2013, oral presentation (Abstract PS6/3).

  • Handling editor Jackie A Cassell

  • Contributors AL conceived the study, participated in the data collection and wrote the article, with the input of SDW. FF analysed anal cytology and histology, MN performed HRA, MD conducted data collection, data management and performed statistical analysis. All authors contributed to the final draft and approved it.

  • Competing interests None declared.

  • Ethics approval Ethics Committee of Saint-Pierre University Hospital, Brussels. No. O.M. 007.

  • Provenance and peer review Not commissioned; externally peer reviewed.