Characterization of tumor immune microenvironment and real-world efficacy of first-line immunotherapy in <i>BRAF</i> -mutant non–small cell lung cancer.

Abstract

e20607 Background: The efficacy of immune checkpoint inhibitors (ICIs) in BRAF-mutant non-small cell lung cancer (NSCLC) remains under debate compared to driver-gene wild-type (WT) populations. To elucidate the biological basis of clinical responsiveness, this study integrated a comprehensive analysis of the tumor immune microenvironment (TIME) with real-world outcomes of first-line ICI-based therapy in BRAF-mutant NSCLC. Methods: In this single-center study, for cohort I, we characterized the TIME of 63 surgically resected NSCLC samples (BRAF V600E, n = 31; WT, n = 32) using RNA sequencing and multiplex immunofluorescence (MIF) stained with CD8, CD68, CD163, CD57, PD-1, and PD-L1 antibodies. In cohort II, we conducted a real-world analysis of 110 advanced NSCLC patients treated with first-line ICI monotherapy or combination therapy. Patients were stratified into BRAF V600E (n = 14), BRAF non-V600 (n = 14), and WT (n = 82) groups to evaluate objective response rate (ORR), disease control rate (DCR), progression-free survival (PFS), overall survival (OS), and treatment-related adverse events (TRAEs). Results: In cohort I, RNA sequencing identified only 77 differentially expressed genes between BRAF V600E and WT. Analysis of immune-related signatures, including most infiltrating immune cell subtypes and tertiary lymphoid structures, indicated that BRAF V600E mutation status had minimal impact on TIME characteristics. MIF revealed significantly higher proportions of CD68+CD163- tumor-associated macrophages in the BRAF V600E group across tumor, stroma, and total regions. However, the proportions of CD8+, CD8+PD-L1+, and CD8+PD-1+ tumor-infiltrating lymphocytes showed no significant disparity between BRAF V600E and WT groups. In cohort II, baseline characteristics were comparable. The BRAF V600E group achieved an ORR of 35.7% and DCR of 85.7%, which were consistent with the BRAF non-V600 (ORR 28.6%, DCR 78.6%) and WT groups (ORR 42.7%, DCR 87.8%). The BRAF-mutant cohort (n = 28) showed no statistically significant difference compared to the WT cohort (n = 82) in median OS (36.73 vs. 55.62 months; P = 0.60) or median PFS (12.75 vs. 18.30 months; P = 0.25). Further subtype stratification revealed comparable OS between BRAF V600E and non-V600 groups (NR vs 35.84 months; P = 0.74). Notably, BRAF V600E patients exhibited a numerically longer median PFS compared to non-V600 patients (25.63 vs. 9.61 months; P = 0.11). Safety profiles were manageable, with Grade ≥3 TRAEs occurring in only 7.1% of V600E patients. Conclusions: The TIME was generally similar between the BRAF V600E mutant and driver gene wild-type groups. Clinically, BRAF-mutant patients derive similar survival benefits from first-line ICI-based therapy as the wild-type population, supporting the use of immunotherapy in this subgroup.