MA09.02 SAKK 16/14 - T-Cell Receptor Repertoire Metrics Predict Response to Neoadjuvant Durvalumab in Patients With Stage IIIA(N2) NSCLC

Abstract

T-cell receptor (TCR) repertoire assessment in blood and tissue has emerged as a novel predictive marker for response to immune checkpoint inhibitor therapy in advanced stage cancers. However, its relevance and predictive significance in the setting of resectable stage IIIA(N2) non-small cell lung cancer (NSCLC) has yet to be demonstrated. Here, we performed TCR sequencing in patients from the phase 2 trial SAKK 16/14 undergoing neoadjuvant chemotherapy with three cycles of cisplatin/docetaxel followed by treatment with the PD-L1 antibody durvalumab. A total of 127 peripheral blood samples and 67 formalin-fixed paraffin-embedded (FFPE) tissue samples were processed from 67 patients before and after neoadjuvant sequential chemo-immunotherapy treatment in the trial SAKK 16/14. Total RNA was extracted from peripheral blood and FFPE samples and used for TCR sequencing with the Oncomine TCR Beta-LR and SR Assays, respectively. TCR evenness, Shannon diversity, and TCR richness were calculated and correlated with the primary endpoint event-free survival (EFS) after 1 year, major pathological response (MPR), and nodal clearance. Tumor mutational burden (TMB) was assessed in tissue samples from extracted genomic DNA using the FoundationOne test or the Oncomine Comprehensive Assay Plus. Association between TCR metrics and clinical outcome data were analyzed using Mann-Whitney-Wilcox test. Analysis was performed using R (R Core Team, 2014). TCR repertoire could be assessed in a total of 97 peripheral blood (47 pre- and 50 post-treatment) and 64 FFPE (15 pre- and 49 post-treatment) samples. In pre-treatment peripheral blood samples, TCR evenness (p=0.032) was associated with 1 year EFS. In FFPE post-treatment samples, 1 year EFS as well as MPR were significantly associated with increased TCR richness (p=0.0168 and 0.0134) and Shannon diversity (p=0.0278 and p=0.0334). Furthermore, nodal clearance was significantly associated with TCR richness and Shannon diversity in post-treatment tissue samples (p=0.0015, p=0.0087). In contrast, TMB was not associated with EFS, MPR or nodal clearance (p=0.91, p=0.47, p=0.52). Our results show that TCR repertoire measured in peripheral blood samples and tumor tissue may provide a useful tool for predicting risk of recurrence after neoadjuvant sequential chemo-immunotherapy with durvalumab in patients with resectable stage IIIA(N2) NSCLC.