A special prognostic indicator: tumor mutation burden combined with immune infiltrates in lung adenocarcinoma with TP53 mutation

TP53 mutation (TP53 ^mut) is significantly associated with immunotherapy response in lung adenocarcinoma (LUAD), but not an ideal independent prognostic predictor for it. Here, we investigated a novel potential biomarker and constructed a model for prognostic prediction in LUAD TP53 ^mut patients.

469 LUAD samples retrieved from The Cancer Genome Atlas database were divided into TP53 ^wt (wild-type TP53) and TP53 ^mut groups. TMB values were calculated based on the number of variants/exon lengths, and high- and low-TMB groups were divided by the median value. Differentially expressed genes (DEGs) between the two TMB groups were identified using "limma" package, and functional analyses were performed by Kyoto Encyclopedia of Genes and Genomes, Gene Ontology, and Gene Set Enrichment Analysis. The infiltration ratio of 22 immune cells were calculated with the CIBERSORT algorithm. Survival analyses were estimated by Kaplan-Meier with the log-rank test. Finally a TMB prognostic index (TMBPI) with receiver operating characteristic (ROC) curve was constructed and calculated to evaluate the predictive value in TP53 ^mut LUAD.

There were diverse mutation types in 100% of TP53 mutants, while mutations were present in 86.5% of cases with TP53 ^wt. TP53 ^mut patients had higher TMB levels than TP53 ^wt patients. Overall survival in TP53 ^mut patients with low-TMB levels was significantly shorter than that in high-TMB TP53 ^mut patients. High-TMB patients had higher levels of CD8 T cell and effector B cell, while lower levels of resting memory CD4 T cells, monocytes, activated dendritic cells, etc. than low-TMB patients. Poor survival outcome in TP53 ^mut patients was correlated with lower effector B cell infiltration and higher activated dendritic cell. Survival risk analyses of 121 DEGs showed that good survival outcomes correlated positively with FBXO36 and KLHL35 expression levels, but correlated negatively with that of LINC0054. TMBPI analysis of the TP53 ^mut patients showed that high-TMBPI patients had worse survival outcomes than low-TMBPI patients.

Our findings suggest that the TMB value with immune infiltrates is a novel potential biomarker for prognostic prediction of TP53 ^mut patients. The TMBPI combined with detection of TP53 mutation can be used as a better predictor of prognosis in LUAD.