This study was designed to measure treatment side-effects and quality of life (QL) of 47 nonmetastatic breast cancer patients subjected to a dose-intensity increase while receiving a sequential high dose chemotherapy (doxorubicin+cyclophosphamide - 4 cycles). The dose-intensity increase was obtained by shortening the length of cycles from 21 to 14 days. Treatment side-effects were self-assessed in terms of frequency and associated distress in cycles 1 and 3 by using a specific side-effect self-report questionnaire (19 items). Multidimensional QL measurement was performed at inclusion and before the start of cycles 2 and 4, by using the EORTC QLQ-C30. Pain was evaluated by patients on a visual analogue scale at the same times as QL evaluation. Patients' self-ratings indicated that the total number of symptoms, the number of symptoms rated by patients as quite or very distressing, and symptom frequency were comparable whatever the length of cycle. Overall, although underestimating most patients' symptoms, physicians' reports provided similar results. However, analysis of multidimensional QL showed that, in comparison to standard administration of 4 cycles of 21 days, there was a more significant deterioration of the QLQ-C30 global QL score ( P=0.01) at the second cycle of chemotherapy and of the physical functioning score ( P=0.02) at the fourth cycle when the cycle length was reduced. This study, although limited by a small patient cohort, has shown that shortening cycles to increase dose intensity had relatively few consequences on adverse treatment effects but a highly negative impact on patients' quality of life.