| Abstract: | Between January 1977 and December 1982, 66 consecutive patients have been treated for unilateral, rapidly progressing, non metastatic breast cancer. They were divided into three groups: Group A (n = 10): tumor whom volume had increased during the 2 months before diagnosis; Group B (n = 30): inflammatory signs (erythema, skin oedema, elevated local temperature) involving less than one half of the breast; Group C (n = 26): inflammatory signs involving more than one half of the breast. All patients where managed similarly: 3 to 4 courses of chemotherapy (CMF: n = 24; AVCF: n = 42), then loco regional irradiation therapy with cobalt 60, followed by maintenance chemotherapy, only if the first chemotherapy had proved effective (CMF: n = 13; AVCF: n = 27). Nine patients with residual tumor after radiotherapy underwent mastectomy with axillary dissection. The actuarial 5 years survival for the whole group is 29%, and respectively 49%, 38% and 9% for the groups A, B and C. The median disease free intervals are 43, 29 and 12 months respectively. Fifteen prognostic factors likely to influence overall survival or disease free survival were evaluated. With univariate analysis, 8 factors were found to be of individual prognostic value: extent of initial erythema, erythema present after initial chemotherapy, erythema present after radiotherapy, non menopaused status, tumor diameter greater than 10 cms, residual breast tumor (clinical or radiographic) after maintenance chemotherapy, supra clavicular adenopathy (N3). Age at the diagnosis, type of chemotherapy, or performance of a radical mastectomy did not influence the prognosis. Multivariate analysis using the Cox-model isolated 3 factors of bad prognosis: erythema involving the whole breast at initial diagnosis, erythema present at the end of initial chemotherapy, N3. |
