Combinatorial immune biomarkers of efficacy on buparlisib and paclitaxel in patients with SCCHN: BERIL-1 sub-analysis.

Denis Soulieres Hotel-Dieu Hospital, Montreal, QC, Canada info_outline Denis Soulieres, Justin Lucas, Sherry Xu, Sunny Lu, Kevin Dreyer, Nanhai He, Tom Tang, Lars E Birgerson, Lisa F. Licitra, Sandrine Faivre

Authors Denis Soulieres Hotel-Dieu Hospital, Montreal, QC, Canada info_outline Denis Soulieres, Justin Lucas, Sherry Xu, Sunny Lu, Kevin Dreyer, Nanhai He, Tom Tang, Lars E Birgerson, Lisa F. Licitra, Sandrine Faivre Organizations Hotel-Dieu Hospital, Montreal, QC, Canada, Adlai Nortye USA Inc., North Brunswick, NJ, Adlai Nortye, Bridgewater, NJ, Adlai Nortye, North Brunswick, NJ, Head and Neck Cancer Medical Oncology 3 Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Department of Oncology and Hemato-Oncology, University of Milan, Milan, Italy, Hôpital St-Louis, Paris, France Abstract Disclosures Research Funding Pharmaceutical/Biotech Company Adlai Nortye Background: Previous analysis of the BERIL-1 Phase 2 clinical trial (Soulieres et al, CCR, 2018), evaluating the efficacy of Buparlisib + Paclitaxel, was one of the largest genomic landscape analyses to date in metastatic SCCHN. With the recent approval of PD-1 in this indication, we sought to expand the analysis, focusing on combinatorial biomarkers of immune infiltration in addition to specific genomic alterations. The goal is to evaluate hierarchically prognostic and predictive biomarkers to select predefined analysis in the current Phase 3 clinical trial, BURAN (NCT04338399) post immune modulatory therapy. Methods: The results of this analysis focused on combination of genomic alterations in both tumor and plasma samples. Alterations found in tumor versus plasma revealed over 50 differentially altered genes. Combination of these sample types is more comprehensive of the patients’ true mutational and copy number status; no significant differences in alteration frequency between treatment and placebo treatment arms were identified. We evaluated correlations between select biomarkers and improvement in efficacy end points of Progression Free Survival (PFS) and Overall Survival (OS) between the treatment and control arms and within the treatment arm. Results: The most frequently altered genes were, TP53 (49%), NOTCH1 (21%), PIK3CA (19%), FAT1 (15%), and CDKN2A (13%). Previous analysis demonstrated that TP53 mutations as well as high immune infiltration (> 10% intratumoral or stromal TILs) showed improvement in PFS & OS. Further improvement to OS was observed in subjects with high immune infiltration and TP53 (p = .031, HR = 0.52), PIK3CA (p = .005, HR = 0.34), NOTCH1(p = .046, HR = 0.13) or FAT1(p = .007; HR = 0.28) alterations when compared to non-altered subjects in the treatment arm. Further improvement to PFS was observed in subjects with high immune infiltration and TP53(p = .031, HR = 0.52) or FAT1(p = .006, HR = 0.23) alterations when compared to non-altered subjects in the treatment arm; NOTCH1 alterations showed strong trends in improved PFS. Additionally, high tumor mutational burden (TMB) combined with alterations in the TP53 pathway showed significant OS improvement (p = .052, HR = 0.44) compared with the placebo arm. Conclusions: This analysis highlights unique insights into the mechanism of Buparlisib and Paclitaxel in of metastatic SCCHN patients; a significant improvement in efficacy observed when specific alterations are considered in the context of immune infiltration. These observations will be further explored in the ongoing Phase 3 clinical trial, BURAN. Additional analysis to be presented at the meeting include prognostic and predictive value of each biomarker and combination, differences between tumor and plasma samples, and comparison with end of treatment samples.

Clinical