Background Homosexual men have high rates of anal cancer but an understanding of the epidemiology of HSIL, the presumed precursor, is lacking. We aimed to describe the epidemiology of anal HSIL, and association with human papillomavirus (HPV), in a community-recruited cohort of homosexual men.

Methods The Study of the Prevention of Anal Cancer is a three-year prospective study of anal HPV infection and cancer precursors in homosexual men aged ≥ 35. At each visit all men receive an anal swab for cytology and HPV genotyping (Roche Linear Array), and high resolution anoscopy with biopsy of suspected lesions. Anal HSIL was defined as having either intraepithelial neoplasia grade 2/3 on histology and/or HSIL on cytology.

Results 293 men were recruited by December 2012. Median age was 49 and 28.3% were HIV-positive. The baseline prevalence of anal HSIL was 44.6% and 34.8% in the HIV-positive and -negative respectively (p = 0.119). Among those without HSIL at baseline, HSIL incidence was 30.0 and 20.0 per 100 person-years in the HIV-positive and -negative (p = 0.467). Among those with HSIL at baseline, the clearance rate was 37.0 and 44.1 per 100 person-years in the HIV-positive and -negative (p = 0.771). Men who tested HPV16 positive on their anal swab at baseline were more likely to develop incident HSIL (58.1 vs 16.1 per 100 person-years, p = 0.002), and less likely to clear prevalent HSIL (15.9 vs. 65.9 per 100 person-years, p = 0.006).

Conclusion Anal HSIL were highly prevalent. Incidence and clearance were common and closely associated with HPV16 status. The high rate of clearance is consistent with the observation that anal HSIL progress to cancer less commonly than do cervical HSIL. The close association of persistence with HPV16 status suggests that HPV testing should be investigated as a means of identifying those with HSIL who are at highest risk of developing cancer.