Results of randomized phase II trial of neoadjuvant carboplatin plus docetaxel or carboplatin plus paclitaxel followed by AC in stage I-III triple-negative breast cancer (NCT02413320).

516 Background: Addition of neoadjuvant carboplatin (Cb) to paclitaxel (T) followed by doxorubicin + cyclophosphamide (AC) improves pathologic complete response (pCR) rate compared to T/AC in TNBC. An anthracycline-free regimen of Cb plus docetaxel (D) also yields high pCR rates in TNBC, and patients achieving pCR with this regimen demonstrate excellent 3-year outcomes without adjuvant anthracycline. This study was designed to compare the efficacy of neoadjuvant regimens CbT→AC and CbD in TNBC. Methods: In this multicenter study, eligible patients with stage I–III TNBC were randomized (1:1) to either paclitaxel 80 mg/m2 every week X 12 + carboplatin (AUC 6) every 3 weeks X 4, followed by doxorubicin 60 mg/m2 + cyclophosphamide 600 mg/m2 every 2 weeks X 4 (CbT→AC, Arm A), or to carboplatin (AUC 6) + docetaxel (75 mg/m2) every 21 days X 6 cycles (CbD, Arm B). The primary endpoint was pCR (no evidence of invasive tumor in the breast and axilla). The two regimens were compared for differences in pCR, residual cancer burden (RCB), treatment delivery, and toxicity. Results: Between 2015 and 2018, 100 patients were randomized; 48 to Arm A and 52 to Arm B. Median age was 52 years, median tumor size was 2.7 cm, 30% were lymph node-positive and 17% carried a BRCA1/2 mutation. Baseline demographic and tumor characteristics were balanced between two arms. pCR was 55% (95%CI: 41%-59%) in Arm A and 52% (95%CI: 39%-65%) in Arm B, p = 0.84. RCB 0+1 rate was 67% in both arms. Grade 3/4 adverse events were more common in Arm A compared to Arm B (73% vs 21%, p < 0.0001), with most notable differences in rates of G3/4 neutropenia (Arm A = 60%, Arm B = 8%, p = 0.0001), febrile neutropenia (Arm A = 18%, Arm B = 0%, p = 0.0001), and G3/4 anemia (Arm A = 46%, Arm B = 4%, p = 0.0001). 81% of Arm A patients completed all 4 doses of AC and 4 doses of Cb, and 71% completed > 9 doses of T. 90% of Arm B patients completed all 6 doses of CbD (p = 0.034). Conclusions: The non-anthracycline platinum regimen of CbD yields pCR and RCB 0+1 rates similar to 4-drug regimen of CbTàAC, but with a more favorable toxicity profile and higher treatment completion rate. The CbD regimen should be further explored as a way to de-escalate therapy in TNBC. Clinical trial information: NCT02413320.