Glioma mutational signatures associated with haloalkane exposure are enriched in firefighters

Abstract

Background: Glioma is the most common malignant primary brain tumor and is associated with significant morbidity and mortality. Modifiable risk factors remain unidentified. New advances in exposure assessment, genomic analyses, and statistical techniques permit more accurate evaluation of glioma risk associated with exogenous occupational or environmental exposures.

Methods: By using whole-exome sequencing data from matched germline and glioma tumor samples, the authors compared tumor mutational signatures for 17 persons with glioma and a documented occupational history of firefighting with those of 18 persons with glioma without an occupational history of firefighting. All 35 individuals were participants in the University of California, San Francisco Adult Glioma Study.

Results: There was a positive correlation among firefighters between the median number of sample variants attributable to single-base substitution signature 42, a single-base substitution mutational signature associated with haloalkane exposure (from the Catalogue of Somatic Mutational Signatures in Cancer) and firefighting years (p = .04; R² = 0.29). Among nonfirefighters, the individuals with the highest number of median variants attributable to single-base substitution signature 42 also had occupations that possibly exposed them to haloalkanes, such as painting and being a mechanic.

Conclusions: In summary, the authors identified gliomas that had mutational signatures associated with haloalkane exposure that were enriched in firefighters and other occupations.

Keywords: epidemiology; firefighters; glioma; haloalkane; mutation; occupation; signature.

FIGURE 1

The median number of mutations attributable to SBS42 among 1000 bootstrap resamplings of variant data, with (A) occupations and (B) variants highlighted. (A) Points correspond to firefighters with a nonzero median SBS42 attribution who had <10 firefighting years and nonfirefighters who had the two greatest median attributable SBS42 mutations and had self‐reported occupation. (B) Variants considered significantly mutated are highlighted (Q < 0.1; green text, with median probability [across 1000 bootstrap samplings] that each variant is attributable to SBS42). In addition, four tumors with >10 median attributable SBS42 variants but without a variant considered significantly mutated have a COSMIC tier 1‐curated variant highlighted, (orange text, with the median probability [across 1000 bootstrap samplings] that each variant is attributable to SBS42). COSMIC indicates the Catalogue of Somatic Mutational Signatures in Cancer; SBS42, single‐base substitution signature 42.