CELL INTELLIGENCE · 肿瘤细胞治疗研究

BACKGROUND: Programmed death-ligand 1 (PD-L1) expression is widely used to guide immune checkpoint inhibitor (ICI) therapy in advanced nonsmall cell lung cancer (NSCLC); however, substantial heterogeneity in clinical outcomes persists among patients with high PD-L1 expression. Reliable biomarkers for further stratifying this population are still lacking. This study investigated whether genomic alterations and features of the tumor immune microenvironment could explain this variability. METHODS: We conducted a retrospective single-center study including 91 patients with advanced NSCLC and PD-L1 tumor proportion score (TPS) ≥ 50% who received ICI-based therapy. Clinical outcomes were assessed according to PD-L1 expression levels, genomic profiles, and tumor immune microenvironment characteristics. Tumor-infiltrating lymphocytes (TILs) and tertiary lymphoid structures (TLSs) were evaluated in available tumor specimens from 75 patients. Multiplex immunohistochemistry (mIHC) was performed in a subset of 30 patients to characterize the immune cellular composition underlying TIL-based stratification. Kaplan-Meier survival analysis and multivariable Cox regression were used to identify factors associated with progression-free survival (PFS) and overall survival (OS). RESULTS: At a median follow-up of 19.1 months, the objective response rate was 31.9%, the median PFS was 10.7 months, and the median OS was 43.5 months. No significant survival differences were observed between the TPS 50%-89% and TPS ≥ 90% subgroups. TP53 mutation was associated with shorter PFS and remained an independent adverse prognostic factor. In contrast, high TIL infiltration was independently associated with improved PFS and OS, and this OS benefit was most pronounced in the TPS ≥ 90% subgroup. mIHC analysis revealed that high-TIL tumors exhibited increased stromal CD8 + T-cell infiltration, enriched NK-cell populations, and a more proinflammatory macrophage profile, providing biological support for TIL-based stratification. TLS presence was not significantly associated with survival outcomes. CONCLUSIONS: PD-L1 expression alone does not fully account for the outcome heterogeneity observed in PD-L1-high NSCLC treated with ICIs. Integrating genomic alterations-particularly TP53 mutation status-with tumor immune microenvironment features, such as TIL density, may help refine prognostic stratification in this population.