Extract

Atkins and Kunkler et al. raise many important issues, not all of which can be addressed from results in our article. The main impetus for our analysis was to explore the conclusions of previous meta-analyses, which was that there was minimal or even no statistically significant survival benefit from postmastectomy radiation therapy and that any early benefit was more than offset by late radiation-induced effects. These conclusions have had a substantial effect on clinical practice, leading to the underutilization of adjuvant radiation therapy after mastectomy. Using the same database as these earlier meta-analyses, we have shown that the survival benefits of radiation therapy are greater than previously reported when the quality of the radiation therapy in the different trials is taken into account. This important finding has implications for the design of future trials in which radiotherapy is restricted or omitted and for clinical guidelines.

The risk of recurrence in the axilla or supraclavicular fossa is less related to the adequacy of the clearance than to the actual number of lymph nodes that are found to be involved in the resected specimen and to the presence or absence of clinically significant extracapsular spread. Pisansky et al. ( 1 ) have shown that in lymph node–positive patients with breast cancer who received chemotherapy but not adjuvant radiation therapy after total mastectomy and axillary dissection, locoregional recurrence exceeded 20% in the first 8 years, with one-third of these recurrences being in the axilla. For these reasons, we cannot agree with the comment by Atkins that any benefit of postmastectomy radiation therapy may be limited to patients who do not undergo an adequate surgical lymph node clearance. Nevertheless, the adequacy of surgery may affect the risk of locoregional recurrence and hence the absolute risk reduction of radiation therapy. Even in this circumstance, however, the relative benefit of the radiation therapy associated with survival should be similar, because the adequacy (or inadequacy) of the surgery would presumably have been balanced between the randomized groups. Inadequate surgery essentially moves patients into a higher risk category.

You do not currently have access to this article.