A cilia-associated signature predicts resistance to anti-PD-1 immune checkpoint blockade therapy in lung adenocarcinoma

BACKGROUND: Targeting immune checkpoints, specifically programmed death 1, has significantly improved clinical outcomes in non-small cell lung cancer. However, only a proportion of patients benefited from immune checkpoint blockades (ICBs) therapy. Thus, identifying novel biomarkers to better select responders to ICBs is critically important in clinical practice. METHODS: This retrospective study included two cohorts, a discovery cohort (n = 24) and a validation cohort (n = 21) of lung adenocarcinoma (LUAD) patients treated with ICB at Shanghai Pulmonary Hospital between April 2016 and November 2019. Formalin-fixed paraffin-embedded tissue samples were collected for RNA sequencing. Additionally, a TCGA dataset (GSE135222) and a clinical study dataset (POPLAR) were utilized as external validation and test sets, respectively. RESULTS: In the discovery cohort, cilia-associated pathways were significantly enriched in LUAD patients with non-durable benefit. A three-gene cilia-associated signature (TMEM100, DNAH3, and NEFH) was established and effectively predicted resistance to ICB treatment (discovery cohort, Hazard Ratio [HR, 95% CI] = 0.21 [0.056-0.791], P = 3.64e-05; validation cohort, HR [95% CI] = 0.200 [0.049-0.830], P = 0.0003). This signature was further validated in an external cohort (HR [95% CI] = 0.461 [ 0.164-1.292], P = 0.067). Further analysis demonstrated that the cilia-associated signature was negatively associated with various types of tumor-infiltrating immune cells, including effector T cells, macrophages, and dendritic cells. CONCLUSION: Our findings suggest that cilia-associated signature may serve as a potential biomarker to identify LUAD patients with reduced likelihood of benefiting from ICB therapy, though further validation in larger cohorts is warranted.