BACKGROUND: Weight change during chemotherapy is reported to be associated with a worse prognosis in breast cancer patients, both with weight gain and weight loss. However, most studies were conducted prior to the common use of anthracycline-base chemotherapy and on North American populations with a mean BMI classified as overweight. Our study was aimed to evaluate the prognostic value of weight change during anthracycline-based chemotherapy on non metastatic breast cancer (European population) with a long term follow-up. METHODS: Patients included 111 women diagnosed with early stage breast cancer and locally advanced breast cancer who have been treated by anthracycline-based chemotherapy regimen between 1976 and 1989. The relative percent weight variation (WV) between baseline and postchemotherapy treatment was calculated and categorized into either weight change (WV > 5%) or stable (WV < 5%). The median follow-up was 20.4 years [19.4 - 27.6]. Cox proportional hazard models were used to evaluate any potential association of weight change and known prognostic factors with the time to recurrence and overall survival. RESULTS: Baseline BMI was 24.4 kg/m2 [17.1 - 40.5]. During chemotherapy treatment, 31% of patients presented a notable weight variation which was greater than 5% of their initial weight.In multivariate analyses, weight change (> 5%) was positively associated with an increased risk of both recurrence (RR 2.28; 95% CI: 1.29-4.03) and death (RR 2.11; 95% CI: 1.21-3.66). CONCLUSIONS: Our results suggest that weight change during breast-cancer chemotherapy treatment may be related to poorer prognosis with higher recurrence and higher mortality in comparison to women who maintained their weight.