作者
Priyanka Sharma,Bruce F. Kimler,Anne O’Dea,Lauren Nye,Yen Y. Wang,Rachel Yoder,Joshua M. Staley,Lindsey Prochaska,Jamie L. Wagner,Amanda L. Amin,Kelsey E. Larson,Christa Balanoff,Manana Elia,Gregory Crane,Sheshadri Madhusudhana,Marc Hoffmann,Maureen K. Sheehan,Roberto Rodríguez,Karissa Finke,Rajvi Shah,Deepti Satelli,Anuj Shrestha,Larry Beck,Richard McKittrick,Robert Pluenneke,Vinay Raja,Venkatadri Beeki,Larry Corum,Jaimie Heldstab,Stephanie LaFaver,Micki Prager,Milind A. Phadnis,Dinesh Pal Mudaranthakam,Roy A. Jensen,Andrew K. Godwin,Roberto Salgado,Kathan Mehta,Qamar J. Khan
摘要
Addition of carboplatin (Cb) to anthracycline chemotherapy improves pathologic complete response (pCR), and carboplatin plus taxane regimens also yield encouraging pCR rates in triple-negative breast cancer (TNBC). Aim of the NeoSTOP multisite randomized phase II trial was to assess efficacy of anthracycline-free and anthracycline-containing neoadjuvant carboplatin regimens.Patients aged ≥18 years with stage I-III TNBC were randomized (1:1) to receive either paclitaxel (P) weekly × 12 plus carboplatin AUC6 every 21 days × 4 followed by doxorubicin/cyclophosphamide (AC) every 14 days × 4 (CbP → AC, arm A), or carboplatin AUC6 + docetaxel (D) every 21 days × 6 (CbD, arm B). Stromal tumor-infiltrating lymphocytes (sTIL) were assessed. Primary endpoint was pCR in breast and axilla. Other endpoints included residual cancer burden (RCB), toxicity, cost, and event-free (EFS) and overall survival (OS).One hundred patients were randomized; arm A (n = 48) or arm B (n = 52). pCR was 54% [95% confidence interval (CI), 40%-69%] in arm A and 54% (95% CI, 40%-68%) in arm B. RCB 0+I rate was 67% in both arms. Median sTIL density was numerically higher in those with pCR compared with those with residual disease (20% vs. 5%; P = 0.25). At median follow-up of 38 months, EFS and OS were similar in the two arms. Grade 3/4 adverse events were more common in arm A compared with arm B, with the most notable differences in neutropenia (60% vs. 8%; P < 0.001) and febrile neutropenia (19% vs. 0%; P < 0.001). There was one treatment-related death (arm A) due to acute leukemia. Mean treatment cost was lower for arm B compared with arm A (P = 0.02).The two-drug CbD regimen yielded pCR, RCB 0+I, and survival rates similar to the four-drug regimen of CbP → AC, but with a more favorable toxicity profile and lower treatment-associated cost.