Abstract P2-07-13: High-dimensional, single-cell analysis and transcriptional profiling reveal novel correlatives of response to PARP inhibition plus PD-1 blockade in triple-negative breast cancer

Abstract Background: TOPACIO was a phase I/II study evaluating the PARP inhibitor (PARPi) niraparib in combination with the anti-PD-1 antibody pembrolizumab in patients with locally advanced and metastatic triple-negative breast cancer (TNBC, n=55) and ovarian cancer irrespective of BRCA mutation status. In the efficacy-evaluable population (n=47) the objective response rate (ORR) was 21% and disease control rate (DCR) 49%. Although activity was greater in patients with BRCA mutations (7/15, ORR=47% and 12/15, DCR=80%), durable clinical benefit was seen in patients with wild-type BRCA tumors (3/27, ORR=11% and 9/27, DCR=33%). In a limited cohort of 20 patients with durable clinical benefit, there were 8 BRCA wildtype patients, four of whom had mutations in genes associated with the homologous recombination repair and other DNA damage repair pathways. Pre-treatment tissues were collected and evaluated for tumor PD-L1 status. Patients with PD-L1 positive tumors (28/47, 60%) had a higher response rate (9/28, ORR=32%) than those with PD-L1 negative tumors (1/13, ORR=8%; 6 tumors had unknown PD-L1 status). It remains unstudied whether the tumor’s gene expression profile or immune status in baseline biospecimens is predictive of treatment response. In this study we conducted exploratory biomarker analyses to test the hypothesis that gene expression patterns and immune status are associated with treatment response. Methods: Transcriptional profiling of baseline samples was performed using the BC360 (n=41) and PanCancer IO360 (n=42) panels (Nanostring) and multigene signatures were used to measure tumor and immune activities as well as relative immune cell abundance. Transcriptional analysis was paired with high-dimensional, single-cell cyclic immunofluorescence (CyCIF) of samples that had adequate tissue for analysis (n=19) to characterize the composition and topology of the immune microenvironment at single-cell resolution. Results: Nanostring transcriptional analysis revealed that PAM50 genes stratified tumor samples into 4 subgroups with distinct histology as determined by CyCIF. Each subgroup was capable of responding to niraparib plus pembrolizumab. Multiple genes involved in WNT signaling (WNT5B, TANKS1, TANKS2, PARP4, and NET02) were associated with favorable clinical responses. Low neuropilin and tolloid-like protein 2 (NETO2) gene expression was strongly correlated with favorable progression free survival (PFS; R=-0.61, p=0.0008, Spearman’s correlation), suggesting it may be a predictive biomarker of therapeutic response. Nanostring gene expression signatures for tumor inflammation, apoptosis, and inflammatory chemokines also distinguished responders from non-responders (p<0.05). CyCIF analysis performed on whole tissue sections accounting for 2.97 million single cells revealed 43 distinct cell-states comprising the tumor microenvironment. PD1+CD4+ T cells were significantly correlated with extended PFS (R=0.65, p=0.006, Spearman’s correlation). However, PD1+CD4+ T cells were less abundant in patients who continue to respond to the therapy (2.7-fold reduced, p=0.004), suggesting two groups of responders. Conclusion: WNT signaling, NETO2 and PD1+CD4 T cells are candidate biomarkers for predicting response to niraparib plus pembrolizumab. Further studies are underway to characterize the biological underpinnings of these correlative findings. Citation Format: Jennifer L Guerriero, Gregory J Baker, Jia-Ren Lin, Yu-An Chen, Ricardo Pastorello, Tuulia Vallius, Janae Davis, Clarence Yapp, Sarah E Church, Eric Miller, Anniina Färkkilä, Shaveta Vinayak, Melinda L Telli, Giulia Fulci, Alan D'Andrea, Geoffrey I Shapiro, Sara M Tolaney, Sandro Santagata, Peter K Sorger, Elizabeth A Mittendorf. High-dimensional, single-cell analysis and transcriptional profiling reveal novel correlatives of response to PARP inhibition plus PD-1 blockade in triple-negative breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P2-07-13.