Background: The randomized trials comparing breast-conserving therapy (BCT), i.e., surgery and radiation to the breast, with mastectomy in early-stage breast cancer use a variety of protocols. Meta-analysis may assist in understanding the impact of these differences on survival.
Purpose: To evaluate the possible variations of the relative efficacy of BCT and mastectomy in terms of overall survival according to tumor size, nodal status, and use of adjuvant radiation therapy.
Methods: The most recent published results and, where available, updated patient-level data from randomized controlled trials of BCT and mastectomy for early-stage breast cancer were combined in a meta-analysis using a random effects model. Pooled survival rates and odds ratios were generated according to subgroups of nodal status and tumor size. Five- and 10-year odds ratios were also determined according to adjuvant radiation protocol.
Results: The pooled odds ratio comparing 10-year survival for BCT and mastectomy was 0.91. The odds ratios comparing the two treatment regimens were not significant after grouping according to tumor size and nodal status. When more than 50% of node-positive patients in both the mastectomy and BCT arms received adjuvant radiation, both arms had similar survival rates. When less than 50% of node-positive patients in both arms received adjuvant nodal radiation, the odds ratio was 0.69, and patients receiving BCT had a survival advantage.
Conclusions: Patients allocated to BCT have survival rates at least as high as patients allocated to mastectomy. When all protocols were combined, nodal status and tumor size did not significantly alter the relative survival rates. However, under some conditions, particularly for node-positive patients, BCT may confer a relative survival advantage over mastectomy. In particular, mastectomy without adjuvant radiation appears to be inferior to BCT for node-positive patients.