We examined risk factors for breast cancer after subdividing cases based on the presence of HER-2/neu oncogene amplification in their tumors. Data were from the Carolina Breast Cancer Study, a population-based, case-control study of 577 invasive breast cancer patients, diagnosed during 1993-1996 and ages 20-74 years, and 790 controls frequency-matched on race and age. Information on breast cancer risk factors was obtained from structured personal interviews. About 20% of paraffin-embedded tissues from the breast cancers of cases were identified as positive for HER-2/neu amplification (HER-2/neu+) by differential PCR. Early age at menarche, higher waist:hip ratio, and family history of breast or ovarian cancer were associated with elevated odds ratios (ORs) for both HER-2/neu+ and HER-2/neu- breast cancers. Breastfeeding for at least 1 year was inversely associated with HER-2/neu+ breast cancer [OR, 0.3; 95% confidence interval (CI), 0.1-0.7] more so than HER-2/neu- breast cancer (OR, 0.8; 95% CI, 0.5-1.2). Most of the remaining risk factors had ORs around 1.0 for both HER-2/neu+ and HER-2/neu- breast cancers, although a few exhibited possible associations with one disease subtype in analyses stratified by menopausal status. These study results suggest that most recognized breast cancer risk factors do not operate through HER-2/neu amplification in breast carcinogenesis. Differential effects of long-term breastfeeding by HER-2/neu amplification status have been observed in earlier studies and are provocative; however, the direction and magnitude of the associations have not been consistent.