Surgery for the Intact Primary and Stage IV Breast Cancer…Lacking “Robust Evidence”

Excerpt

The association of surgical resection of the intact primary in women with de novo stage IV breast cancer with survival has received a great deal of attention in the past decade; this issue of the Annals of Surgical Oncology contains the fourth meta-analysis/systematic review of approximately one dozen retrospective studies examining this association.1 ^– 3 The reasons that this question is of interest biologically and of practical importance to our patients are well recapitulated by Harris and colleagues and have been discussed extensively elsewhere.4 The fact that a fourth analysis of the same data is considered sufficiently interesting and novel for publication in the Annals illustrates the high level of interest that has emerged over the past decade in the question of primary tumor resection in Stage IV solid tumors. Harris et al. provide a summary estimated hazard ratio of 0.69 (95 % confidence interval (CI) 0.54–0.88) pooled from ten studies that favors surgical resection of the primary tumor, acknowledge the potential biases, and conclude their abstract with the statement that, “In the absence of robust evidence, this meta-analysis provides evidence base for primary resection in the setting of stage IV breast cancer for appropriately selected patients.” The devil, of course, is always in the details, and the authors offer no specific recommendations as to what constitutes appropriate selection. Nor can they. The only hint regarding surgical selection of patients in their analysis comes from a lack of association of benefit with large primary tumors and large number of metastatic sites, but they wisely ignore this trend in their Discussion. In fact, another recent meta-analysis where the use of a variable effects meta-regression model was specified, found no difference in the apparent benefit of surgery based on number of sites of disease, rate of bone or visceral metastasis, type of surgery, margin status, and median age.3 (Notably, Harris et al. do not tell us how much between-study heterogeneity they observed, and whether a random or a fixed-effects model was used.) The authors of this and three similar analyses (one meta-analysis and two systematic reviews) therefore are uniformly circumspect in not recommending a basis for patient selection for surgery in any of their Discussions. …