Background: We previously described a predictive signature for trastuzumab benefit which was validated in the adjuvant setting in an independent cohort within NSABP B-31 (the 8-gene signature) (Pogue-Geile et al JNCI, 2013) and in Alliance/NCCTG N9831 (SABCs 2017). The 8-gene signature subtyped B-31 patients into three trastuzumab benefit groups: high HR=0.27, intermediate HR=0.56 and no benefit HR=1.56 based on disease free survival. The 8 gene signature was also predictive of trastuzumab benefit in N9831. HRs were 0.47, P<0.001, 0.6, P=0.02, and 1.54, P=0.375 in the predicted-high, -intermediate and -no benefit groups, respectively based on recurrence free survival (SABCS 2017). The interaction P-value was significant at 0.019 in adjusted Cox models. The RFS at 10 years for trastuzumab-treated pts was 83%, 83% and 72% in the high, intermediate and no benefit groups, respectively. Now we have tested the association of the 8-gene signature groups with pCR in FB-7 which was a 3 arm neoadjuvant study testing the pCR rate of HER2+ breast cancer patients treated with paclitaxel in combination with trastuzumab (T) or neratinib (N) or the combination (T + N).

Methods: RNA-Seq data from FB-7 pretreatment biopsies was used to predict the trastuzumab benefit groups (high, intermediate, and no) for each patient's tumor using the 8 gene signature using methods and cut-offs as previously described (Pogue-Geile et al 2013). The pCR rates (percentages) were tested for treatment interaction with a chi-square test.

Results: The pCR rates were 75%, 53%, and 22%, in the high (N=12), intermediate (N=32) and no benefit groups (N=9), respectively, when analyzed without regard to treatment arm. The pCR rates for the no benefit group and the high benefit groups were significantly different (p=0.030) and there was a significant treatment interaction with the 8-gene benefit group (intp=0.0081). The predicted low and intermediate groups were combined to test whether the 8 gene signature could identify a group of patients whose pCR rates might improve by adding N to T, and referred to it as the low benefit group. This was necessary due to the small numbers of patients in each group. The pCR rate in the low benefit group was higher in patients treated with T+N (9/15, 60%) than in the T arm (6/11, 45%) but these differences were not significant.

Conclusions: This is the first test of the 8-gene signature in the neoadjuvant setting and interpretations of these data should be interpreted cautiously due to the small numbers. However, if these results were validated in another neoadjuvant trial then the 8 gene signature could provide a rationale for selecting patients who would be appropriate for the addition of neratinib or other TKIs to trastuzumab and chemotherapy.

SUPPORT: PUMA Biotechnology, NCI U10CA180868, -180822, UG1-189867, and U24-196067; The Pennsylvania Department of Health. The Department specifically disclaims responsibility for any analysis,interpretations, or conclusions.

Citation Format: Pogue-Geile KL, Wang Y, Srinivasan A, Gavin PG, Kim RS, Song N, Feng H, Lipchik C, Costantino JP, Wolmark N, Lucas PC, Paik S, Jacobs SA. The fully validated NSABP/NRG 8-gene signature which predicted the degree of benefit in the adjuvant setting (B-31 and NCCTG N9831) associates with pCR in the neoadjuvant setting in NSABP clinical trial FB-7 [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P3-10-04.

©2019 American Association for Cancer Research.
2019
American Association for Cancer Research.