NRG-BR003: A randomized phase III trial comparing doxorubicin plus cyclophosphamide followed by weekly paclitaxel with or without carboplatin for node-positive or high-risk node-negative TNBC.

Abstract

LBA509 Background: NRG-BR003 is a phase III, randomized trial evaluating whether the addition of carboplatin (carbo) to an adjuvant chemotherapy regimen of doxorubicin/cyclophosphamide (AC) followed by paclitaxel (P) will improve invasive disease-free survival (IDFS) compared to AC followed by P when administered to patients (pts) with operable node-positive or high-risk node-negative triple-negative breast cancer (TNBC). Methods: Eligible pts had operable node-positive or high-risk node-negative TNBC and were randomized to receive dose-dense (DD) AC every 2 weeks for 4 cycles followed by weekly P (80 mg/m2) for 12 doses or the same regimen with carbo AUC of 5 IV every 3 weeks for 4 cycles. Stratification factors were number of positive nodes (0, 1-3, 4-9, 10+) and BRCA mutation status (positive; negative; or unknown). The study was designed to detect a hazard ratio (HR) in IDFS at 0.67 with the addition of carbo. The stratified log-rank test was used for the primary analysis. Secondary endpoints include DRFI, OS, BCFS, and RFI. Results: 769 pts were randomized to control arm (n=385) and carbo arm (n=384) from June 2015 to May 2022. Patient characteristics include age >50 (66%), primary tumors >2 cm (70%), node-positive (69%), and g BRCA pathogenic variants (9%). Delivery of AC was balanced between arms and P delivery was not compromised by co-administration of carbo with a mean of 11.3 doses (sd=2.1) of P with a relative total dose intensity (RTDI) at 0.97 in the control arm and 11.0 doses (sd=2.3) and a RTDI at 0.95 in the carbo arm. At data cutoff (2/28/25), median follow-up was 79.4 mos. IDFS events were reported in 92 pts (23.9%) in the control group and 76 (19.8%) in the carbo group. The stratified log-rank test p-value was 0.097, not meeting the prespecified significance of 0.049; the HR was 0.77 (95% CI, 0.57-1.05). The 5-year IDFS (95% CI) was 77.8% (73.7%-82.2%) vs 82.9% (79.2%-86.9%), respectively. HR was similar across patient subgroups, including germline BRCA and nodal status. Grade ≥3 treatment-related AE rates were 51.1% in the control group and 72.9% in the carbo group. Grade 5 events were 0.8% vs 0.8%, respectively. Conclusions: The addition of carbo to P following DD AC for adjuvant therapy of node-positive or high-risk node-negative TNBC did not result in a statistically significant improvement in IDFS, DRFI, or OS. However, it increased grade ≥3 treatment-related AE rates. Although not meeting criteria for efficacy across the entire study population, results support planned translational research to identify subsets of pts who may benefit from carbo. Clinical trial information: NCT02488967 . Secondary efficacy results of DRFI and OS. End Points Treatment 5-year Event-free Rate (95% CI) HR (95% CI) DRFI AC → P 84.4% (80.7-88.2) 1 AC → P+Carbo 88.7% (85.5-92.0) 0.74 (0.50-1.10) OS AC → P 84.4% (80.8-88.3) 1 AC → P+Carbo 87.7% (84.4-91.2) 0.81 (0.56-1.16)