Abstract Lung cancer has the second highest incidence and leads cancer mortality worldwide, being Non-Small Cell Lung Cancer (NSCLC) the most prevalent subtype. Immunotherapy with checkpoint blockers has shown outstanding benefits in a subset of NSCLC patients, which are not accurately identified due to the lack of robust biomarkers of response. We hypothesize that specific molecular alterations of NSCLC tumor cells impact the immune microenvironment. Defining this association could improve the prediction of response to immunotherapy. This work aims to identify a multiparametric biomarker signature. We studied 178 tumor samples from a cohort of early stage NSCLC with complete clinical annotation. We analyzed their molecular aberrations, calculated their Tumor Mutational Burden and described their immune landscape by immunohistochemistry and a dedicated RNAseq panel. We defined novel subgroups of NSCLC tumors with specific immune, molecular and clinical features, showing an association between their molecular genotype and immune phenotype. Based on these data, subgroups were proposed as pro-immunogenic, pro-tumorigenic or mixed. Adenocarcinomas clustered in 4 groups. The pro-tumorigenic group had a significant higher proportion of alterations in ARID1A, FGF10, ROS1, TP53. One pro-immunogenic group had a significant higher proportion of MET alterations and lower proportion of EGFR alterations. An immune mixed group significantly accrued never smokers and had no alterations in FGF10. Squamous cell carcinomas clustered in 4 groups. The pro-tumorigenic group had a significant higher proportion of alterations in CCND1, FGF3, FGF10, FGF19, NOTCH1, PIK3CA, PIK3CB, TFRC. A pro-immunogenic group had a significant higher infiltration of T CD4+, T CD8+ and B cells and a higher overall survival. Another pro-immunogenic group had a significant absence of MYCN and NF1 alterations. Taken together we identified a set of candidate predictive biomarkers of response to immunotherapy in NSCLC. Genes overexpressed in pro-immunogenic tumors are related to adaptive immune response stimulation (CD28, FCRLA, JCHAIN, LY9, MS4A1, SLAMF7, TNFRSF9, TNFRSF17), antigen presentation (CD1C, CD1D, HLA-A), apoptosis (FAS), immune chemotaxis (CCL20, CCR4, CCR6), and immune regulation (CD53, PDCD1, SH2D1A, ZAP70). On the other hand, genes upregulated in pro-tumorigenic tumors are involved in angiogenesis (VEGFA), cell cycle regulation (BUB1, CCNB2, FOXM1, MAD2L1, TOP2A), cell growth/survival (IGF1R), DNA replication/repair (KIAA0101) and iron metabolism (HMBS, TFRC). These findings are being validated by Digital Spatial Profiling in a cohort of patients with advanced stage NSCLC that were treated with checkpoint blockers. If confirmed, these results could remarkably improve patient selection and the benefit upon immunotherapy. Citation Format: Javier Ramos-Paradas, David Gomez-Sanchez, Aranzazu Rosado, Alvaro Conrado Ucero, Nuria Carrizo, Ana Belen Enguita, Maria Teresa Muñoz, Esther Conde, Luis Paz-Ares, Eva Maria Garrido-Martin. Comprehensive analysis of non-small cell lung cancer identifies molecular genotype-immunophenotype associations and candidate biomarkers predictive of response to immunotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1245.
Read full abstract