Abstract A4: DNA methylation profiles associated with response to anti-PD-1 immunotherapy in non-small cell lung cancer

Abstract

Abstract Understanding the molecular mechanisms of regulating anti-PD-1 efficacy is key to improving cancer immunotherapy. Epigenetic modulation of tumors, particularly via DNA methylation, is a key regulatory strategy in immune evasion of cancer. Although there are several types of research on the association between genomic features or transcriptomic features and cancer immunotherapy, tumor methylomes are relatively unexplored. The study cohort of patients with NSCLC was established by recruiting patients from Yonsei Cancer Center, Seoul, Korea. Each patient was treated with anti-PD-1 inhibitors. Patients were discriminated as responder or nonresponder by RECIST version 1.1. DNA methylation data were obtained by using the Infinium Methylation EPIC Array (850K CpG sites). Methylation data were processed by Minfi and RnBeads packages. Differentially methylated regions (DMR) were obtained by DMRcates. Enhancer regions overlapping with DMR were obtained from enhancer-promoter interaction network from lung cancer cell line. RNA-seq was also performed by HiSeq 2500 and processed by STAR-2.5.2a, FeatureCounts, and DESeq2. Validation experiment was performed by pyrosequencing of candidate markers. Survival analysis was performed by the Kaplan-Meier plot with a log-rank test. Here, we profiled all DMRs overlapping promoters (pDMRs) and enhancers (eDMRs) between responders and nonresponders to anti-PD-1 inhibitors. As a result, 1,007 pDMRs and 607 eDMRs are associated with anti-PD-1 response. We also found 1,109 and 1,173 target genes of pDMRs and eDMRs, respectively. We found eDMRs to be more associated with the regulation of immune systems than pDMRs by pathway analysis. For example, previous studies reported that upregulation of MHC expression in the tumor correlates with anti-PD-1 response. We found that the methylation level of MHC-II enhancers differs between responders and nonresponders, which suggests an association between methylation of MHC enhancers and anti-PD-1 response. Moreover, those MHC-II enhancers are overlapped with super-enhance of lung tissue, implying that those enhancers are highly active. In addition, cancer-fitness genes or cancer-risk genes are associated with target genes of DMR. We also identified two hypomethylated pDMR that can predict responders to anti-PD-1 therapy with higher accuracy than PD-L1 expression. In addition, those two markers showed significantly better prognosis by survival analysis. In summary, we identified numerous methylome profiles of the anti-PD-1 response associated with both promoters and enhancers and found that epigenetic regulation of enhancers is a major strategy of immune modulation or cancer-related genes affecting immunotherapy efficacy. Two pDMR showed superior accuracy for predicting clinical outcomes of PD-L1 expression, suggesting that using methylation-based biomarkers is a robust and cost-effective strategy for patient stratification for anti-PD-1 therapy in NSCLC. Citation Format: Jae-Won Cho, Min Hee Hong, Sang-Jun Ha, Young-Joon Kim, Insuk Lee, Hye Ryun Kim. DNA methylation profiles associated with response to anti-PD-1 immunotherapy in non-small cell lung cancer [abstract]. In: Proceedings of the AACR Special Conference on Tumor Immunology and Immunotherapy; 2019 Nov 17-20; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2020;8(3 Suppl):Abstract nr A4.