Genetic variation and regulation of MICA alters natural killer cell-mediated immunosurveillance in early-onset colorectal cancer.

Abstract

207 Background: One of the most concerning trends in oncology is the rapid increase in the incidence of early-onset cancers, such as colorectal cancer, in patients <50 years old. Patients with early-onset colorectal cancer (EO-CRC) often present with more advanced disease and have worse outcomes than average-onset CRC, leading the American Cancer Society to recommend colonoscopies starting at age 45. While some environmental risk factors have been linked to EO-CRC (i.e. Western diet and changes in microbiota), the genetic and immunologic etiology of EO-CRC is almost entirely unknown. Interestingly, Single Nucleotide Polymorphisms (SNPs) in MHC class I polypeptide-related sequence A ( MICA ) have been associated with non-progression and elite control of HIV. Given the similarities between the immune response to viruses and cancer, we hypothesized that a SNP in MICA could alter immunosurveillance in the colon and contribute to EO-CRC development. Methods: To investigate our hypothesis, we analyzed genotypes in the target region of chromosome 6 harboring MICA in the Colorectal Cancer Transdisciplinary (CORECT) study, a well-characterized, international consortium study of CRC. Since MICA is a stress-induced ligand for the Natural Killer group 2 member D (NKG2D) receptor on immune cells such as Natural Killer (NK) cells, we used RNA-seq and multiplex immunofluorescence to quantify NK cells in the colons of patients with various MICA genotypes. Results: We identified 5 SNPs in MICA that showed a significant difference in MICA RNA expression in normal colonic epithelium. These 5 SNPs were evaluated for CRC association in a discovery set of 39,274 cases and controls from the CORECT consortium study. Using this approach, we identified a SNP in MICA (rs9295988) which encodes for a C to G transversion and is associated with an increased EO-CRC risk (Ratio of Odds Ratio = 1.239). Next, we validated our findings in a larger, independent validation study (76,983 cases and controls) from the FIGI consortium. Lastly, using deconvoluted RNA-seq data and multiplex immunofluorescence on colonic specimens, we found that patients with the G allele at rs9295988, but not the CC genotype, have reduced NK cells in their colon cancers compared to adjacent normal colonic epithelium. Conclusions: Our results suggest that MICA SNP rs9295988 dysregulates NK cell immunosurveillance, potentially contributing to EO-CRC development. By providing the first evidence of an interaction between genetic determinants and innate immune dysfunction in EO-CRC, our work fills a critical gap in knowledge by suggesting that EO-CRC is an innate immune-related disease. This constitutes a major paradigm shift that could open new avenues for addressing the etiology and treatment of EO-CRC.