NEW YORK – A research team from the Institute of Cancer Research and the University of Edinburgh has identified hundreds of genes that appear to be influenced by colorectal cancer (CRC) risk variants, offering clues to processes that go awry in CRC and possible treatment strategies for targeting them.
"Our analyses provide a detailed interpretation of CRC risk signals and their underlying basis," senior and corresponding author Richard Houlston, a researcher at the Institute of Cancer Research, and his coauthors wrote in a paper published in Nature Genetics on Monday.
Houlston and colleagues began by using statistical fine-mapping with cross-tissue single-cell RNA sequencing profiles, histone chromatin immunoprecipitation sequences, ATAC-seq data, and long-range chromatin interactions found by Micro-C — coupled with findings from massively parallel reporter assays — to search for suspected causal variants at loci found in prior genome-wide association studies of CRC.
"To prioritize functional variants for the identification of CRC susceptibility genes at risk loci, we systematically scored multiple genetic and functional features as well as assayed allelic transcriptional activity," the authors explained.
Using an integrative score that drew on these data, the team highlighted functional variants at 170 loci previously linked to CRC by GWAS. While 40 loci appeared to contain individual variants with causal ties to CRC, another 98 loci appeared to house more than one suspected causal variant.
From there, the researchers tracked down some 208 genes that appeared to be impacted by that set of suspected causal variants by bringing in normal colon- and CRC-related expression quantitative trait locus profiles from GTEx and PancanQTL and by performing GWAS summary statistics-based randomization analyses.
"By deciphering CRC risk loci, we identify direct links between risk variants and target genes," the authors wrote, "providing further insight into the molecular basis of CRC susceptibility and highlighting potential pharmaceutical targets for prevention and treatment."
In particular, their results suggested that many of the risk loci found by GWAS in the past involve functional variants with regulatory roles, including for gene expression in specific tissue types.
The collection of genes targeted by CRC risk variants included half a dozen that are known to act as CRC drivers, the team noted. They also unearthed new candidate genes in the calcium sensor-related calmodulin superfamily, in pathways involved in immune response and inflammation, signaling pathways, and other processes.
The broader strategy used for the study "provides an outline for a generalized strategy to profile disease-associated GWAS loci using high-throughput variant screening in concert with multilayered functional annotation," the authors reported.