Dados do Trabalho

Título

INTERPLAY BETWEEN MUTATIONAL PROFILE AND TMB ON OVERALL SURVIVAL: A POSTHOC ANALYSIS OF 1,662 METASTATIC PATIENTS TREATED WITH IMMUNE CHECKPOINT INHIBITORS.

Introdução

Tumor mutational burden (TMB) represents the total of mutations in the tumor genome, and TMB≥10mut/Mb received FDA agnostic approval for pembrolizumab use.

Objetivo

A more integrative view of this biomarker is needed to select better candidates to immune checkpoint inhibitors (ICI).

Método

We performed a posthoc analysis of overall survival (OS) using clinical data from a unicentric advanced cancer patients cohort treated with ICI and submitted to genomic profiling by MSK-IMPACT platform available as public domain on the cbioportal database. Validation analysis was performed to correlate predefined percentiles used at the original publication with cutoffs used for regulatory approval. Kaplan-Meier (KM) method, multivariate log-rank p value, and Cox regression hazard ratios (HR) for OS were calculated for all patients with TMB ≥20, ≥10, or <10mut/Mb. For a ≥10mut/Mb cutoff, KM method was performed, and log-rank P values with confidence interval (CI) of 95% were reported for each gene tested (N=464). Holm-Bonferroni alfa error adjustment was performed. For all genes exhibiting a possible correlation with survival considering a less rigorous alfa-error level (p<0.05), a Cox multivariable analysis was also performed.

Resultado

A total of 1,662 patients (11 cancer types) were included. ICI type received was anti-PD-(L)1 in 78.7%, CTLA-4 in 6%, and combination regimen in 15.4% of the cases. With a maximum follow-up of 80 months, 49.9% of patients were still alive. Median OS was 42 months for both TMB ≥20mut/Mb and TMB ≥10, and 15mo for TMB <10mut/Mb (multivariate log-rank p<0.005). HR for death was HR 0.44 (95%CI 0.34-0.56; p<0.005) and 0.57 (95%CI 0.49-0.67; p<0.005), for TMB ≥20 and ≥10mut/Mb respectively. From a ≥10mut/Mb cutoff, after log-rank test followed by Holm-Bonferroni alfa-error adjustment (p<0.0009), four genes negatively impacted survival when altered: CCND3, TIMM8B, CDKN2B and STK11 (median OS of 1, 1, 1, and 7 months). From that same TMB cutoff, a Cox multivariate analysis stratified by median TMB revealed that mutations in STK11 (HR 2.95; p<0.0007) and TP53 (HR 1.89; p<0.0032) were associated with a higher risk of death. Otherwise, mutations in TERT (HR 0.54; p<0.0044) and NOTCH3 (HR 0.23; p<0.0355) were protective.

Conclusão

This pan-cancer analysis confirmed the 10mut/Mb TMB cutoff as a positive predictive biomarker to ICI treatment. In addition, for TMB ≥10mut/mB, the mutational profile can additionally contribute to define outcomes to ICIs.

Palavras-chave

Tumor mutational burden; immunotherapy; biomarkers