A new tumor neoantigenicity metric may improve prediction of response to immunotherapy in patients with melanoma, lung cancer, and kidney cancer, a retrospective analysis suggests.
The new metric, known as the Cauchy-Schwarz index of neoantigens (CSiN) score, incorporates both immunogenicity and clonality, according to lead study author Tianshi Lu, a PhD candidate at the University of Texas Southwestern Medical Center in Dallas, and colleagues.
“The major biological insight from this study is that the neoantigen clonal structure in each tumor specimen and the immunogenicity of the neoantigens (represented by the MHC-binding strength in our study) are predictive of response to checkpoint inhibitors and prognosis,” the investigators wrote in Science Immunology.
The study involved 2,479 patients with various cancers, including immunogenic types such as renal cell carcinoma (RCC), and nonimmunogenic types, such as pediatric acute lymphocytic leukemia.
The investigators first evaluated CSiN in relation to clinical outcome among patients with immunogenic cancers who received immunotherapy. Drawing data from multiple cohorts, the investigators found that patients who had better responses to therapy were significantly more likely to have above average CSiN scores than those who had worse responses.
In one cohort of patients with melanoma who received anti–CTLA-4 therapy, those with better responses were more likely to have high CSiN scores (P = .009). In another cohort of melanoma patients who received anti–CTLA-4 therapy, those with higher CSiN scores were more likely to achieve durable clinical benefit (response or stable disease for more than 6 months), compared with patients who had lower CSiN scores (P = .033).
Among patients with clear cell RCC treated with anti-PD-1/PD-L1 therapy, there was a significant positive association between higher CSiN scores and better response (P = .036). Among T effector-high patients with metastatic clear cell RCC, there was a significant association between higher CSiN scores and better response to atezolizumab (P = .028) but not sunitinib (P = .890).