Abstract PD3-18: The NSABP/NRG 8-gene signature accurately predicts degree of benefit from trastuzumab in Alliance/NCCTG N9831: Validation of the 8-gene signature in an independent clinical trial

BACKGROUND Clinical trials NRG/NSABP B-31 and Alliance/NCCTG N9831 demonstrated that trastuzumab added to chemotherapy extended disease-free survival (DFS) in breast cancer patients (pts) in the adjuvant setting (Romond et al). However, the degree of trastuzumab benefit varies among pts. We previously described an 8-gene model that was predictive for trastuzumab benefit, which was validated in an independent cohort of B-31 patients (Pogue-Geile 2013). The 8-gene signature subtyped B-31 pts in an independent validation cohort into three trastuzumab benefit groups: one with large benefit HR=0.27, one with medium benefit HR=0.56, and one with little to no benefit HR=1.56. The purpose of this study was to validate the 8-gene model in pts enrolled into N9831, which tested the efficacy of adding trastuzumab to doxorubicin plus cyclophosphamide → paclitaxel by randomizing pts into one of three arms: chemotherapy only (Arm A), trastuzumab given after chemotherapy (Arm B), or trastuzumab given concurrently with chemotherapy (Arm C). METHODS NCounter data, consisting of 8 predictive genes and 4 house-keeping genes, for 1,379 patients enrolled into N9831 were used to assign each patient to one of the three benefit groups. Assignments to one of the three trastuzumab benefit groups based on the 8-gene model were made by the NSABP/NRG Pathology Laboratory Biostatistician, who was blinded to outcome data. Predictions were sent to N9831 investigators and accuracy of predictions were tested in pts enrolled into arms A and C (N=892) using Cox models adjusted for age, nodes, ER/PR status, tumor size, and grade. Recurrence-free survival (RFS) and DFS were used as endpoints. RESULTS Tumors from N9831were placed into one of the three trastuzumab benefit groups based on the 8-gene signature. The N9831 pts in the predicted-large benefit group had a hazard ratio (HR) of 0.47, P=0.0006, pts in the predicted-medium benefit group had an HR of 0.6, P=0.02, and the predicted-low benefit group had an HR of 1.54, P=0.375. The interaction P value was significant at 0.019 in adjusted Cox models. The RFS at 10 years for trastuzumab-treated pts was 72%, 83%, and 83% in the low, medium, and large benefit groups, respectively. CONCLUSIONS The 8-gene signature developed in a discovery cohort and validated in an independent cohort of B-31 pts has now been validated in N9831. Many anti-HER2 therapies such as lapatinib, afatinib, neratinib, pertuzumab, and TDM-1 have been approved for treatment in metastatic breast cancer pts and although these agents have shown responses, the actual improvements in outcomes in pts is variable and it is not known which pts actually receive benefit. Thus this signature may be clinically useful in identifying pts who may benefit from additional treatment beyond trastuzumab or for stratification of pts enrolled into clinical trials testing new anti-HER2 therapies. SUPPORT U10CA180868, -180822, UG1CA189867, U24CA196067, the PA DOH, which disclaims certain responsibilities, Genentech, Inc., and the Breast Cancer Research Foundation. Citation Format: Pogue-Geile KL, Song N, Serie DJ, Thompson EA. The NSABP/NRG 8-gene signature accurately predicts degree of benefit from trastuzumab in Alliance/NCCTG N9831: Validation of the 8-gene signature in an independent clinical trial [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr PD3-18.