Ten year results of a randomised trial comparing two conservative treatment strategies for small size breast cancer.

We report the 10-year results of a randomised clinical trial in which two different breast conservation treatment strategies were compared in women with small, non-metastatic primary breast cancer: quadrantectomy, axillary dissection and radiotherapy (QUART) versus tumorectomy and axillary dissection followed by external radiotherapy and a boost with 192Ir implantation (TART). No second surgery was given to women with affected surgical margins. Axillary node positive women received adjuvant medical therapy. From 1985-1987, this trial accrued 705 patients, 360 in the QUART and 345 in the TART arm. Crude cumulative incidence curves for intrabreast tumour recurrence (IBTR) and metastases as first events and mortality curves in each of the two treatment arms were computed. A crude cumulative incidence curve of IBTR as a second event (in women who had already had a local recurrence) was also computed. The two groups were compared in terms of hazard for IBTR, metastases or death occurrence by using Cox regression models, both with and without adjustment for patient age, tumour size, number of metastatic axillary nodes and histology. Possible interactions between the aforementioned prognostic factors and the type of surgery were also investigated. The two groups were well matched for baseline patient and tumour characteristics, the only exception being resection margins, which were more often positive in the TART group. At the Cox model, a significant difference between groups was detected for IBTR (P < 0.0001), but not for distant metastases and overall survival. In particular, 5- and 10-year estimates of crude cumulative incidence of IBTR were 4.7 and 7.4% in the QUART group and 11.6 and 18.6% in the TART group. The difference was not substantially affected by patient or disease characteristics. Likewise, the status of resection margins in women who underwent TART treatment did not significantly influence the risk of occurrence of IBTRs. Finally, the rate of second IBTR occurrence was relatively high, when compared with the rate of IBTR occurrence as first event. In summary, the results of this trial show that a better local control of the disease can be obtained with the more extensive surgical resection, i.e. QUART.