Comprehensive Analysis of a Cancer-Immunity Cycle-Based Signature for Predicting Prognosis and Immunotherapy Response in Patients With Colorectal Cancer

Immune checkpoint blockade (ICB) has been recognized as a promising immunotherapy for colorectal cancer (CRC); however, most patients have little or no clinical benefit. This study aimed to develop a novel cancer-immunity cycle-based signature to stratify prognosis of patients with CRC and predict efficacy of immunotherapy. CRC samples from The Cancer Genome Atlas (TCGA) were used as the training set, while the RNA data from Gene Expression Omnibus (GEO) data sets and real-time quantitative PCR (RT-qPCR) data from paired frozen tissues were used for validation. We built a least absolute shrinkage and selection operator (LASSO)-Cox regression model of the cancer-immunity cycle-related gene signature in CRC. Patients who scored low on the risk scale had a better prognosis than those who scored high. Notably, the signature was an independent prognostic factor in multivariate analyses, and to improve prognostic classification and forecast accuracy for individual patients, a scoring nomogram was created. The comprehensive results revealed that the low-risk patients exhibited a higher degree of immune infiltration, a higher immunoreactivity phenotype, stronger expression of immune checkpoint-associated genes, and a superior response to ICB therapy. Furthermore, the risk model was closely related to the response to multiple chemotherapeutic drugs. Overall, we developed a reliable cancer-immunity cycle-based risk model to predict the prognosis, the molecular and immune status, and the immune benefit from ICB therapy, which may contribute greatly to accurate stratification and precise immunotherapy for patients with CRC.

Automated summary

In this study, a cancer-immunity cycle-based signature was developed to predict the prognosis and efficacy of immunotherapy for colorectal cancer patients. Low-risk patients showed better prognosis and stronger response to immunotherapy.