Constructing a folate metabolism gene signature for predicting prognosis in pulmonary neuroendocrine carcinomas.

Folate metabolism is a crucial biological process in cell proliferation and exhibits its pro-tumorigenic functions in multiple tumor types. However, its role in pulmonary neuroendocrine carcinomas remains uncertain. Folate metabolism related genes were obtained from previous studies, and the gene expression data and clinical data were collected from GEO database. The expression patterns of folate metabolism related genes were measured across normal and tumor tissues. We subsequently assessed their prognostic role using Kaplan-Meier and univariate Cox regression analysis. The core genes were isolated from 16 prognostic genes through four algorithms. Based on the expression of core genes, patients were divided into two clusters employing consensus clustering algorithm. Furthermore, we evaluated immune infltration level, biological mechanisms, and drug sensitivity. ALDH1L2 was finally identified through qRT-PCR and its pro-tumorigenic function was confirmed via in vitro experiments. The expression patterns of 26 folate metabolism related genes were evaluated between normal lung tissues and PNEC tumor tissues, and 20 of them exhibited differential expression. All of folate metabolism related genes were related to the prognosis of PNECS and 16 genes were identified as prognostic genes. Using SVM-RFE, RF, Xgboost and LASSO algorithm, three core genes were isolated from 16 prognostic genes. Based on the expression patterns of core genes, PNECs patients were divided into two clusters through consensus clustering algorithm. Cluster 1 was characterized by the worse survival, higher immune infiltration level, and sensitivity to chemotherapy. Compared with the HBEC cells, ALDH1L2 was notably overexpressed in NCI-H446 cells (SCLC cell line). ALDH1L2 knockdown significantly repressed the proliferation and migration capacity of tumor cells and increased the cell proportion in S phase. Our results indicated that folate metabolism gene signature is a reliable biomarker for PNECs. Classification based on this signature could be utilized to guide the treatment of PNECs patients and improve its prognosis.