SMARCA4 mutations in KRAS-mutant lung adenocarcinoma: a multi-cohort analysis

Abstract

KRAS is a key oncogenic driver in lung adenocarcinoma (LUAD). Chromatin-remodeling gene SMARCA4 is comutated with KRAS in LUAD; however, the impact of SMARCA4 mutations on clinical outcome has not been adequately established. This study sought to shed light on the clinical significance of SMARCA4 mutations in LUAD. The association of SMARCA4 mutations with survival outcomes was interrogated in four independent cohorts totaling 564 patients: KRAS-mutant patients with LUAD who received nonimmunotherapy treatment from (a) The Cancer Genome Atlas (TCGA) and (b) the MSK-IMPACT Clinical Sequencing (MSK-CT) cohorts; and KRAS-mutant patients with LUAD who received immune checkpoint inhibitor-based immunotherapy treatment from (c) the MSK-IMPACT (MSK-IO) and (d) the Wake Forest Baptist Comprehensive Cancer Center (WFBCCC) immunotherapy cohorts. Of the patients receiving nonimmunotherapy treatment, in the TCGA cohort (n = 155), KRAS-mutant patients harboring SMARCA4 mutations (KS) showed poorer clinical outcome [P = 6e-04 for disease-free survival (DFS) and 0.031 for overall survival (OS), respectively], compared to KRAS-TP53 comutant (KP) and KRAS-only mutant (K) patients; in the MSK-CT cohort (n = 314), KS patients also exhibited shorter OS than KP (P = 0.03) or K (P = 0.022) patients. Of patients receiving immunotherapy, KS patients consistently exhibited the shortest progression-free survival (PFS; P = 0.0091) in the MSK-IO (n = 77), and the shortest PFS (P = 0.0026) and OS (P = 0.0014) in the WFBCCC (n = 18) cohorts, respectively. Therefore, mutations of SMARCA4 represent a genetic factor leading to adverse clinical outcome in lung adenocarcinoma treated by either nonimmunotherapy or immunotherapy.

Keywords: KRAS; SMARCA4 mutation; immunotherapy; lung adenocarcinoma; nonimmunotherapy; prognostics biomarker.

Fig. 1

Global somatic mutation landscape of KRAS, TP53, STK11, KEAP1, and SMARCA4 genes in the TCGA, MSK‐CT, MSK‐IO, and WFBCCC cohorts. Comutations were determined with Fisher’s exact test. Black *,**,**: P < 0.05, 0.01, 0.001 for co‐occurrence; red ***: P < 0.001 for exclusive occurrence.

Fig. 2

SMARCA4 mutations are associated with shorter DFS and OS of KRAS‐mutant LUAD patients treated with nonimmunotherapy treatment from the TCGA cohort. Kaplan–Meier survival analysis of survival in (A, C) the KS, KP, and K subgroups and (B, D) in the two‐group comparison between SMRACA4‐mutant and wild‐type KRAS‐mutant patients.

Fig. 3

SMARCA4 mutations are associated with shorter OS of KRAS‐mutant LUAD patients treated with nonimmunotherapy treatment from the MSK‐CT cohort. Kaplan–Meier survival analysis of OS (A) in the KS, KP, and K subgroups and (B) in the two‐group comparison between SMRACA4‐mutant and wild‐type KRAS‐mutant patients.

Fig. 4

SMARCA4 mutations are associated with shorter PFS of KRAS‐mutant LUAD patients treated with immunotherapy treatment from the MSK‐IO cohort. Kaplan–Meier survival analysis of PFS (A) in the KS, KP, and K subgroups and (B) in the two‐group comparison between SMRACA4‐mutant and wild‐type KRAS‐mutant patients.

Fig. 5

SMARCA4 mutations are associated with shorter PFS and OS of KRAS‐mutant LUAD patients treated with immunotherapy treatment from the WFBCCC cohort. Kaplan–Meier survival analysis of survival (A, C) in the KS, KP, and K subgroups and (B, D) in the two‐group comparison between SMRACA4‐mutant and wild‐type KRAS‐mutant patients.

Fig. 6

Tumor microenvironment varied among three groups of patients. KS patients contained the lowest proportions of CD8 and activated CD4 memory T cells than either K or KP patients. The plot for all 22 types of immune cells was shown in Fig. S2. *P<.05; Mann–Whitney U‐test.