Reducing Local Therapy in Patients Responding to Preoperative Systemic Therapy: Are We Outsmarting Ourselves?

Abstract

There is growing interest in the use of preoperative (neoadjuvant) chemotherapy for patients with localized breast cancer. Originally, it was hypothesized that such early use of chemotherapy might improve survival, compared with standard postsurgical adjuvant systemic therapy, but this has not proven to be the case. Nonetheless, the primary tumor response is likely a barometer for tumor sensitivity to therapy and may be used to help guide decisions regarding additional systemic therapy. Furthermore, preoperative chemotherapy can increase the fraction of women eligible for breast-conservation therapy. A seemingly logical extension of this latter observation is the hypothesis that one can further reduce local therapies in patients responding well to initial chemotherapy. For example, studies are under way to omit (or limit) radiotherapy (RT) in patients who present with positive axillary nodes and experience pathologic complete response in the nodes to preoperative chemotherapy (eg, NSABP [National Surgical Adjuvant Breast and Bowel Project] B-51/RTOG [Radiation Therapy Oncology Group] 1304). In the NSABP/RTOG study, patients with involved axillary nodes (histologically confirmed) are treated with neoadjuvant chemotherapy. Those who are node negative at subsequent mastectomy are randomly assigned to postmastectomy RT (PMRT) to the chest wall and regional nodes. Similarly, patients who undergo subsequent breast conservation surgery and whose nodes have become negative after preoperative chemotherapy will be randomly assigned to breast RT regional nodal RT. We appreciate the importance of the question being asked in these studies and support their conduct. Defining patient subgroups that do (or do not) benefit from our therapies is an important goal. It is possible that response to preoperative chemotherapy predicts for the potential benefit of local regional RT (and hence theassociatedtherapeutic ratio).Responders topreoperativechemotherapy may have a lesser need for additional local regional RT. However, we believe the converse may be true: that the potential survival benefits of local therapies are likely highest among responders to preoperative chemotherapy. Consider the more typical clinical scenario for node-positive patients who undergo initial mastectomy, followed by adjuvant chemotherapy (Fig 1). In that setting, the addition of PMRT improves overall survival (OS) by approximately 6% to 9%. This group includes both responders and nonresponders to chemotherapy, althoughidentificationofthesesubgroupsisnotpossible,becausechemotherapy is administered postoperatively. If it is true that responders to neoadjuvant chemotherapy derive lesser survival gains with PMRT, it must follow that nonresponders derive greater gains, because the OS benefit should still be 6% to 9% (ie, sequencing of surgery and chemotherapy should not alter this; Fig 1, lower panel). An analogous situation exists for patients undergoing breast conservation. The NCIC Clinical Trials Group MA.20 study reported improvements in locoregional control and distant disease-free survival and a trend in OS with the addition of nodal RT to breast RT in patients undergoing lumpectomy followed by chemotherapy. If responders to preoperative chemotherapy derive lesser gains with nodal RT, nonresponders should derive greater gains (again, to make the math work). As we discuss here, the evidence suggests that the opposite may be true: namely, that responders to systemic therapy might derive the most survival benefit from RT. Mastectomy Chemotherapy 6%-9%