Cell Reports: Methods (Oct 2023)
Multi-omics indicators of long-term survival benefits after immune checkpoint inhibitor therapy
Abstract
Summary: Molecular indicators of long-term survival (LTS) in response to immune-checkpoint inhibitor (ICI) treatment have the potential to provide both mechanistic and therapeutic insights. In this study, we construct predictive models of LTS following ICI therapy based on data from 158 clinical trials involving 21,023 patients of 25 cancer types with available 1-year overall survival (OS) rates. We present evidence for the use of 1-year OS rate as a surrogate for LTS. Based on these and corresponding TCGA multi-omics data, total neoantigen, metabolism score, CD8+ T cell, and MHC_score were identified as predictive biomarkers. These were integrated into a Gaussian process regression model that estimates “long-term survival predictive score of immunotherapy” (iLSPS). We found that iLSPS outperformed the predictive capabilities of individual biomarkers and successfully predicted LTS of patient groups with melanoma and lung cancer. Our study explores the feasibility of modeling LTS based on multi-omics indicators and machine-learning methods. Motivation: Long-term survival (LTS) benefit is the landmark achievement of immunotherapy, but significant questions remain about the underlying molecular mechanism and predictive biomarkers. However, the limited clinical and genetic data on patients with long-term follow up poses challenges for research into ICI LTS biomarkers. Since conducting in vitro/in vivo experiments using cell lines or mouse models to explore the mechanism of human LTS is impractical, in silico bioinformatic analysis with human genomic sequencing data is a needed approach. Here, we conduct a systematic analysis to elucidate LTS predictors, including a panoramic biomarker review, rigorous feature selection, and Gaussian process regression machine-learning model training and testing. We validate model performance with multiple cohorts, discuss the biological interpretation of model findings, and provide illustrative guidance on real-world clinical application.
