Clonal hematopoiesis (CH) is described as the phenomenon where hematopoietic cells acquire advantageous somatic mutations that result in a dominant subclonal population. These subclones are linked to various adverse outcomes such as elevated risk of hematologic malignancies and cardiovascular disease. Few studies, however, address the presence and role of CH in clinically-relevant advanced and metastatic solid tumors or how progression of CH is altered via immune selection.
Here, we identified and quantified CH mutations in 883 patients with 64 different advanced solid cancer subtypes. The age of patients ranged from 1 to 96, with the average age being 66.7 years. In this cohort, whole exome sequencing (WES) of tumor and matched peripheral blood was conducted through the Englander Institute of Precision Medicine. CH was identified through mutations in the peripheral blood that were absent in the tumor. Probable SNPs were filtered using population allele frequency data from the ExAC database. Tumor mutation burden (TMB) and microsatellite instability (MSI) scores were calculated for each patient. MHC Class I alleles for HLA types A, B, C were inferred for each patient based on WES.
CH was identified in 12% of patients and 19 different solid advanced and metastatic cancer subtypes and was associated with increased age (P < 0.001). Cancer subtypes with the highest frequency of CH mutations were urothelial carcinoma, glioblastoma, adenocarcinoma, and transitional cell (urothelial) carcinoma. The most commonly mutated genes were DNMT3A, TET2, ASXL1, and BCOR1. We restricted further analysis to the most common CH genes: DNMT3A, TET2, and ASXL1. 60% of patients with CH had only one mutation, 13% had two mutations, and 4% had 3 or more mutations. When adjusting for covariates including tumor subtype and age, the presence of CH was suggestive of elevated MSI scores (P < 0.1) but did not impact TMB.
We next sought to determine if MHC class impacted susceptibility to the emergence of CH. Fisher's exact test was used to identify associations between HLA alleles and the presence of CH. We identified MHC alleles conferring hyper-susceptibility to CH development by as much as 8-fold (OR 8.0, 95% C.I. 2.1 - 39.1, P < 0.001). We further identified 52 probable protective alleles. These data suggest that certain HLA alleles are significantly more permissive to CH development, while others may be protective against CH.
Overall, our study is the first to systematically evaluate and suggest that patient-specific hyper-susceptibility HLA alleles impact evolutionary processes that underlie clonal sculpting leading to the emergence of CH. The presence of CH may also impact the evolution of co-occurring solid tumors, a hypothesis requiring further evaluation. This study indicates that MHC alleles and thus personal genetics place immunity-driven selective pressure for the emergence of CH, supporting the role of inherited determinants of CH.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.