Among tumors with a high mutational burden, there are differences in the nature or flavor of the mutations involved, reflecting differences in underlying mutagenic processes.

— Shridar Ganesan, MD, PhD

Tweet this quote

Shridar Ganesan, MD, PhD, of Rutgers Cancer Institute of New Jersey, discussed the study by Cristescu et al at the 2017 ASCO-SITC Clinical Immuno-Oncology Symposium. He noted that data from many sources have shown that point mutation burden correlates with response rate to checkpoint blockade.

“However, among tumors with a high mutational burden, there are differences in the nature or flavor of the mutations involved, reflecting differences in underlying mutagenic processes,” he said. “For example, in melanoma and non–small cell lung cancer, high mutational burden is associated with exposure to exogenous mutagens, such as ultraviolet light and smoking.”

“Both exposure to exogenous mutagens and endogenous defects in DNA repair and replication can result in higher mutational burden in different tumor types,” he continued. “Data from KEYNOTE, involving a diverse group of tumors, also show that a correlation between mutational load and response exists.”

No Easy Explanations

Why this should be so is not easily explained. The “classic answer,” stated Dr. Ganesan, relates to the presence of peptide neoantigens encoded by somatic mutations; however, this is disputed by observations that the burden of neoantigens, as calculated by bioinformatics algorithms, is not superior to the overall mutational burden in predicting response or progression-free survival. This has been borne out by the KEYNOTE analysis in diverse tumor types.

Why would estimated neoantigen load not be more predictive? It is possible that the current “algorithms” do not distinguish well among classes of mutations or that subclones are being counted in these algorithms when only truncal mutations are relevant. There may be other host immune factors at play, or mutational load may be a marker of another process that is immunogenic, suggested Dr. Ganesan.

The findings that correlate gene-expression profile to response and progression-free survival are consistent with other studies showing that inflammatory signatures and mutational load were independent predictors , yet again, the reasons are unclear. It is possible that, since the tumor microenvironment is “spotty,” sampling may capture a regions of either high or low tumor purity, which may have opposing effects on ability to reliably detect either mutation load or an inflammatory gene signature. Alternatively, there are likely tumors where inflammatory signature may be driven by other events, such as viral infection, that may be truly independent of mutation load. ■

Disclosure: Dr. Ganesan has relationships with Inspirata, Novartis, and Merck.