1031 Background: According to observational studies, obesity is an unfavourable prognostic factor in breast cancer (BC), regardless of menopausal status and treatment received. Information collected in clinical trials should confirm this effect and serves to test its homogeneity by pathologic subtype. Methods: We retrospectively analysed 5,683 operable BC patients enrolled in four randomized clinical trials (GEICAM/9906, 9805, 2003–02, and BCIRG 001) evaluating adjuvant anthracyclines and taxanes. Our primary aim was to assess the prognostic effect of body mass index (BMI) on disease recurrence, breast cancer mortality (BCM), and overall mortality (OM). A secondary aim was to detect differences by BC subtypes (ER/PR-positive/HER2-negative, HER2-positive, triple-negative). Cox models were fitted for each end-point, adjusted by potential confounders. Results: Analyses adjusting for age, tumor size, nodal status, menopausal status, surgery, grade, hormone receptor and HER2 status, chemotherapy regimen, and undertreatment showed that obese patients (BMI 30.0–34.9) had similar prognoses to that of patients with a BMI<25 (reference group) in terms of recurrence (HR 1.08 [95% CI 0.9–1.3]; p=0.41), BCM (HR 1.02 [0.81–1.29]; p=0.85), and OM (HR 0.97 [0.78–1.19]; p=0.747). Patients with severe obesity (BMI≥35) had a significantly increased risk of recurrence (HR 1.26 [1.00–1.59]; p=0.05), BCM (HR 1.32 [1.00–1.74]; p=0.05), and OM (HR 1.35 [1.06–1.71]; p=0.02) compared to our reference group (Table). The prognostic effect of severe obesity did not vary by subtype. Conclusions: Severely obese patients treated with anthracyclines and taxanes present a worse prognosis regarding recurrence, BCM, and OM than patients with a BMI<25. The magnitude of the harmful effect of BMI was similar across subtypes. [Table: see text]