Revealing the significance of tissue-resident memory T cells in lung adenocarcinoma through bioinformatic analysis and experimental validation.

PURPOSE: To investigate the functions of lung T(RM) cells in the development and treatment of lung adenocarcinoma (LUAD). METHODS: R-language bioinformatics analysis was applied to obtain differentially expressed (DE) lung T(RM) cell-specific genes and a related prognostic signature, which were further validated using external datasets, immunohistochemical staining images, and biological experiments. RESULTS: A total of 130 DE lung T(RM) cell-specific genes were identified, 14 of which were involved in the prognostic signature, including SLC16A3, ARHGAP11A, PTTG1, DTL, GPRIN1, EXO1, GAPDH, TYMS, DAPK2, CCL20, HLA-DQA1, ADAM12, ALOX5AP and OASL. The signature was efficient and robust in predicting the overall survival and anti-PD-1/PD-L1 immunotherapeutic outcomes of patients with LUAD. The AUCs for predicting the 1-, 3-, and 5-year survival rates were 0.688, 0.698, and 0.648, respectively, in the training cohort, and were 0.867, 0.662, and 0.672, respectively, in the validation cohort. The signature also had predictive value for the sensitivity of patients to chemical drugs. TYMS was a hub gene in the prognostic signature, and was strongly associated with LUAD progression and cell proliferation in the experimental validation. CONCLUSIONS: The lung T(RM) cell-related prognostic signature is an effective tool for predicting the prognosis and therapeutic outcomes of patients with LUAD.