Back

Germline cancer gene expression quantitative trait loci influence local and global tumor mutations

Liu, Y.; Gusev, A.; Kraft, P.

2022-08-24 genetic and genomic medicine
10.1101/2022.08.23.22279002 medRxiv
Show abstract

Somatic mutations drive cancer development and are relevant to patients response to treatment. Emerging evidence shows that variations in the somatic genome can be influenced by the germline genetic background. However, the mechanisms underlying these germline-somatic associations remain largely obscure. We hypothesized that germline variants can influence somatic mutations in a nearby cancer gene ("local impact") or a set of recurrently mutated cancer genes across the genome ("global impact") through their regulatory effect on gene expression. We integrated tumor targeted sequencing from 12,413 patients across 11 cancer types in the Dana-Farber Profile cohort with germline cancer gene expression quantitative trait loci (eQTL) from the Genotype-Tissue Expression Project. We identified variants that upregulate ATM expression which are also associated with a decreased risk of having somatic ATM mutations across 8 cancer types (P = 3.43 x 10-5). We also identified GLI2, WRN, and CBFB eQTL that are associated with global tumor mutational burden of cancer genes in ovarian cancer, glioma, and esophagogastric carcinoma, respectively (P < 3.45 x 10-6). An EPHA5 eQTL was associated with the number of mutations in cancer genes specific to colorectal cancer, and eQTL associated with expression of APC, WRN, GLI1, FANCA, and TP53 were associated with mutations in genes specific to endometrial cancer (P < 1.73 x 10-5). Our findings provide evidence for the germline-somatic associations mediated through expression of specific cancer genes and open new avenues for research on the underlying biological processes, especially those related to immunotherapy responses.

Matching journals

The top 10 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.