To examine the association between breast cancer risk and a T‐to‐C substitution polymorphism at the 5′promoter region of CYP17, a case‐control study was conducted at Aichi Cancer Center Hospital in Japan. Subjects were 144 histologically confirmed breast cancer patients diagnosed in the past 4 years and 166 hospital controls without cancer. Allele frequency among controls was 44.9% (95% confidence interval; 39.5–50.2) for C allele. Odds ratio (OR) of the polymorphism relative to TT‐genotype was 0.97 (0.58‐1.64) for TC‐genotype and 0.81 (0.39–1.68) for CC‐genotype. Subgroup analyses revealed that the OR was not statistically significant for the subgroups stratified by interval after diagnosis, age at menarche, age at first birth, menopausal status, body mass index, and mother/sisters’ history of breast cancer. Consistent with previous studies conducted in other countries, the 5′promoter region polymorphism of CYP17 affected breast cancer risk of Japanese women to a limited extent. Although this is not a large‐scale case‐control study with population controls, these findings provide enough information to discourage further studies on the association between this polymorphism and breast cancer risk in Japan at large, and suggest that this polymorphism is useless for breast cancer risk estimation.