Male breast cancer is a rare tumour in all parts of the world. About 1% of all breast cancers occur in men, but the male/female ratio is higher among black than among white populations. This effect can be seen in US cancer registries and even more markedly in African data. A positive correlation exists on a population scale between male breast cancer and prostate cancer. Seven case-control studies of male breast cancer are available, and a pooled analysis was conducted of the most commonly suspected risk factors. Male breast cancer appears to be associated with marital status: Mantel-Haenszel exposure odds ratio (EOR) for never married = 1.6; 95% confidence limits (CL) = 1.1, 2.3, religion (EOR for being Jewish = 2.1; 95% CL = 1.4, 3.2), previous breast pathology (EOR for positive history of benign breast disease = 2.7; 95% CL = 1.7, 4.2), gynaecomastia (EOR for positive history = 6.2, 95% CL = 3.4, 11.4), previous testicular pathology (EOR for positive history = 2.2; 95% CL = 1.5, 3.3), previous liver diseases (EOR for positive history = 1.6; 95% CL = 1.0, 2.4) and family history of breast cancer (EOR for first-degree relative with breast cancer = 2.5; 95% CL = 1.7, 3.7). No association is found with smoking history. Other potential risk factors such as reproductive history, education, occupation, anthropometric variables, association with various diseases, and specific exposures such as drug use, were not systematically evaluated in all studies and provide sometimes contradictory results, possibly due to small numbers of exposed subjects. Overall, the analytical epidemiology of male breast cancer presents similarities with the epidemiology of female breast cancer, with a potential role of factors related to hormonal status, relative hyperoestrogeny in men being potentially linked to increased risk of disease. Genetics may also play a role, with high risk linked to a familial history of breast cancer, and with a major risk in patients with Klinefelter's syndrome.