No abstract

Genome-wide measures of genetic disruption such as tumour mutation burden (TMB) and mutation signatures are emerging as useful biomarkers to stratify patients for treatment. Clinicians commonly use cancer gene panels for tumour mutation burden estimation, and whole genome sequencing is the gold standard for mutation signature analysis. However, the accuracy and cost associated with these assays limits their utility at scale. Using in silico library simulations we demonstrate that restriction enzyme associated DNA sequencing (RADseq) may be a cost-effective solution to improve accuracy of TMB estimation and derivation of mutation profiles when compared to a derived FDA approved cancer gene panel TMB score. Using simulated immune checkpoint blockade (ICB) trials, we show that inaccurate tumour mutation burden estimation leads to a reduction in power for deriving an optimal TMB cutoff to stratify patients for immune checkpoint blockade treatment. Additionally, prioritisation of APOBEC hypermutated tumours in these trials optimises TMB cutoff determination for breast cancer. Finally, the utility of RADseq in an experimental setting is also demonstrated, based on characterisation of an APOBEC mutation signature in an APOBEC3A transfected mouse cell line. In conclusion, our work demonstrates that RADseq has the potential to be used as a cost-effective, accurate solution for TMB estimation and mutation signature analysis by both clinicians and basic researchers.