"One to three" or "Four or more"? selecting patients for postmastectomy radiation therapy

In their recent commentary, Marks and Prosnitz1 discussed the problem of selecting patients for postmastectomy radiotherapy (RT) and concluded that improvement of survival with RT involves both patients with many (≥4) positive lymph nodes and patients with just a few (1-3) metastatic lymph nodes. They recommended RT for both groups instead of a "compromise" policy of irradiating only patients at very high risk for locoregional failure (LRF). I would like to support this view with an additional argumentation. Improvement of survival with RT may be expected in patients in whom 1) subclinical foci of cancer remain after mastectomy within the chest wall and/or within regional lymph nodes; and 2) there is no subclinical distant dissemination, or if there is, it is controlled with adjuvant systemic therapies. Furthermore, an increase in survival may be demonstrated if RT eradicates locoregional subclinical disease in a high proportion of patients and does not adversely influence survival (e.g., by increasing the risk of noncancer death). In addition, it is now clear that systemic therapies are not effective in the prevention of LRF. The extent of involvement of axillary lymph nodes remains the best indicator of risk for locoregional recurrence and distant metastases. However, other factors may contribute to better estimation of these risks. In this respect, our study, which involved a group of 1068 patients treated with mastectomy alone, could be of interest.2 Lymph node involvement, tumor size, and histologic grade (Bloom classification) were all found to be independent prognostic factors for survival and LRF in multivariate analysis. Table 1 shows the 10-year actuarial rates of survival and LRF for several prognostic subgroups defined with all three variables. The risk of LRF in some subgroups was very high indeed, and the need of RT to treat these patients has not been questioned even by RT opponents, which led to the above-mentioned "compromise" policy. However, the number of positive axillary lymph nodes is not only an indicator of the risk of failure but is assumed to correlate with subclinical tumor burden (both locoregional and distant). Adjuvant chemotherapy was found to be more effective in improving the survival of patients with 1-3 positive lymph nodes than in improving that of patients with more lymph nodes involved.3 Postmastectomy RT is highly effective in improving local control. Since the early work of Fletcher et al.,4 it has been suggested that a dose of 50 Gy eliminates subclinical disease in >90% of patients. In Table 2, our data on the effectiveness of postmastectomy RT in relation to the number of positive lymph nodes are presented (data on regional lymph nodes and the chest wall are presented separately).5 These data indicate that the effectiveness of RT is lower in preventing chest wall failures than in preventing lymph node failures, and that this effectiveness decreases with the number of positive lymph nodes. This observation is keeping in with more recent reports of Fletcher et al.,6 who also found that 50 Gy to the chest wall was relatively ineffective in patients with massive lymph node involvement, and who even recommended the use of higher doses.7 All the above considerations consistently lead to the conclusion that patients with 1-3 positive lymph nodes are in fact more likely to benefit from postmastectomy RT in terms of improvement of survival than those with 4 or more lymph nodes involved. First, they are at lower risk of distant dissemination; and if it exists, it is more likely to be controlled with adjuvant systemic therapy. Second, RT, due to smaller tumor burden, may be more effective in eradicating subclinical locoregional disease, thus improving disease free and overall survival. Marks and Prosnitz were quite right in stating that "final resolution of the controversy regarding the survival impact of postmastectomy RT would require a very large prospective trial." In such a trial, with the application of modern, "safe" RT technique, an unresolved issue of the optimal combination of RT and adjuvant systemic therapy could also be addressed. A question arises as to whether such trial would be feasible nowadays, when a high proportion of patients are treated with breast-conserving therapy and there is an increasing interest in neoadjuvant approaches. Stanisław Korzeniowski M.D. Ph.D.*