Résumé : Background Breast cancer is composed of distinct subtypes defined mainly based on the expression of hormone receptors (HR) and HER2. For years, reproductive factors were shown to impact breast cancer risk but it is unclear whether this differs according to tumor subtype. In this meta-analysis we evaluated the association between parity, age at first birth, breastfeeding and the risk of developing breast cancer according to tumor subtype. Methods PubMed and Embase were searched to identify epidemiological studies that evaluated the impact of parity and/or age at first birth and/or breastfeeding on breast cancer risk with available information on HR and HER2. Tumor subtypes were defined as: luminal (HR-positive, HER2-negative or HER2-positive), HER2 (HR-negative, HER2-positive) and triple-negative (HR-negative, HER2-negative). Summary risk estimates (pooled OR [pOR]) and 95% confidence intervals (CI) were calculated using random effects models. The MOOSE guidelines were applied. Results This meta-analysis evaluated 15 studies, including 21,941 breast cancer patients and 864,177 controls. Parity was associated with a 25% reduced risk of developing luminal subtype (pOR 0.75; 95% CI, 0.70–0.81; p < 0.0001). Advanced age at first birth was associated with an increased risk of developing luminal subtype (pOR 1.15; 95% CI, 1.00–1.32; p = 0.05). Ever breastfeeding was associated with a reduced risk of developing both luminal (pOR 0.77; 95% CI, 0.66–0.88; p = 0.003) and triple-negative (pOR 0.79, 95% CI, 0.66–0.94; p = 0.01) subtypes. Conclusions The reproductive behaviors impact the risk of developing breast cancer but this varies according to subtype.