TNBC treated using neoadjuvant chemotherapy (NACT) has a varied prognosis with 50% of patients (pts) having excellent response to treatment (pCR/RCB-I) and excellent prognosis, while 50% have marked residual disease (RCB-II-III) and significantly worse prognosis. Lack of response to an initial NACT regimen also indicates a low chance (5%) of achieving pCR with subsequent chemotherapy, even if drugs are changed.
This randomized study will determine the impact of predicting chemosensitivity to NACT using molecular characterization combined with diagnostic imaging and determine if offering a clinical trial of selected targeted therapy will impact outcomes (as measured by pCR and RCB) in predicted chemo-insensitive disease. The algorithm uses a pre-defined genomic signature (JAMA, 2011;305:1873-81) and response to an intitial course of anthracycline based NACT to determine predicted sensitivity to chemotherapy. Pts undergo biopsy of the primary tumor prior to NACT and are randomized 2:1 to know the molecular testing results versus not (control). For the second phase of neoadjuvant therapy, pts who fit molecular/imaging criteria for chemo-insensitive disease after AC are offered a clinical trial based upon comprehensive molecular profiling (if known) or based upon physician/patient choice if randomized to the control arm. Pts with chemo-sensitive disease continue with taxane based NACT. Success is defined as improvement in the rate of excellent pathologic response (pCR/RCB-I) from 50%– > 64% using the platform. A maximum of 360 pts will be randomized using a group sequential design with one-sided O'Brien-Fleming boundaries, with up to two equally spaced binding interim tests for both futility and superiority and one final test, having an overall Type I error .05 and power .80 to detect a response rate improvement from a null rate of .50 to a target value of .642.
Clinical trial identification
ClinicalTrials.gov Identifier: NCT02276443 (10/21/2014)
Legal entity responsible for the study
MD Anderson Moon Shot Program
All authors have declared no conflicts of interest.