Summary A central role in anal carcinogenesis of high-risk types of human papillomaviruses (hrHPV) was recently established, but the possible role of benign anal lesions has not been addressed in hrHPV-positive and -negative anal cancers. As part of a population-based case-control study in Denmark and Sweden, we interviewed 417 case patients (93 men and 324 women) diagnosed during the period 1991-94 with invasive or in situ anal cancer, 534 patients with adenocarcinoma of the rectum and 554 population controls. Anal cancer specimens (n = 388) were tested for HPV by the polymerase chain reaction. Excluding the 5 years immediately before diagnosis, men, but not women, with anal cancer reported a history of haemorrhoids [multivariate odds ratio (OR) 1.8; 95% confidence interval (Cl) 1.04-3.2] and unspecific anal irritation (OR 4.5; Cl 2.3-8.7) significantly more often than controls. Women with anal cancer did not report a history of benign anal lesions other than anal abscess to any greater extent than controls, but they had used anal suppositories more often (OR 1.5; Cl 1.1-2.0). Patients with hrHPV in anal cancer tissue (84%) and those without (16%) reported similar histories of most benign anal lesions, but anal fissure or fistula was more common among hrHPV-positive cases. Ulcerative colitis and Crohn's disease, reported by <11% of study participants, were not associated with anal cancer risk. The higher proportion of hrHPV-positive anal cancers among case patients with anal fissure or fistula suggests that such mucosal lesions may provide direct viral access to basal epithelial layers. Since risk associations with benign anal lesions in men may be confounded by unreported sexual behaviour, and since risk associations in women were generally negative, it seems unlikely that benign anal lesions act as promoters in hrHPV-associated anal carcinogenesis. Moreover, benign anal lesions appear not to be linked to an alternative, hrHPV-unassociated causal pathway to anal cancer. Ulcerative colitis and Crohn's disease were not supported as causal factors for anal cancer.Keywords: anus neoplasms; risk factors; haemorrhoids; anal fistula; anal fissure; inflammatory bowel diseases; ulcerative colitis; Crohn's disease The incidence of epidermoid anal cancer, a rare neoplasm of the anal canal and perianal skin, has increased considerably during the past decades (Goldman et al, 1989;Frisch et al, 1993;. It has been shown that anal cancer has a sexually transmitted aetiology (Daling et al, 1987;Holly et al, 1989;Frisch et al, 1997). Substantial evidence now points to the causal involvement of certain high-risk types of human papillomaviruses (hrHPV), notably HPV type 16, in the majority of anal cancers (Holm et al, 1994;Frisch et al, 1997). One case-control study (Holly et al, 1989) provided data that were interpreted as supportive of the old belief that anal inflammation predisposes to anal cancer (Brofeldt, 1927;Buckwalter and Jurayj, 1957). However, another case-control study (Holmes et al, 1988) and two subsequent cohort s...