Primary systemic treatment of breast cancer with cytotoxics yields a high response rate and allows conservative surgical procedures in bulky tumours. In order to maximise local control of disease, two innovations were introduced in a pilot study. The first was to identify the good responders after three cycles of chemotherapy and to treat them with three additional cycles. The second was to also give this group of patients a full dose of radiotherapy before surgery with the aim of verifying the rate of pathological complete remissions in view of a possible treatment of breast primary with chemoradiotherapy only. Patients were treated with doxorubicin 60 mg/m2 and cyclophosphamide, 600 mg/m2 both intravenously on day 1, every 21 days for three courses. Partial or complete responders received three more courses followed by radiotherapy (50 Gy plus a 10 Gy boost). The others underwent immediate surgery. A total of 32 patients (median age, 50 years; range 28-69 years); performance status, 0-1; T2 22, T3 8, T4 2) were enrolled and were evaluable for response and side-effects. 9 patients had only three cycles of chemotherapy due to absence of response and 23 patients had six cycles of chemotherapy. Overall, 7 patients had a complete remission, 16 a partial remission and 9 had stable disease, for an overall response rate of 72% (95% confidence interval 53-86%). In the group of patients that completed the programme, two complete pathological remissions were observed and 5 patients had only microfoci of tumour. No toxic death or grade III-IV toxicities were observed. Mild or moderate side-effects included mucositis, nausea/vomiting and leucopenia. In conclusion, our results indicate that the addition of radiotherapy to pre-operative chemotherapy did not significantly enhance the incidence of pathological complete remissions. New primary treatment approaches should be explored in this subset of patients in order to improve outcome.