Construction of a Lung Adenocarcinoma Prognostic Model Utilizing Serine and Glycine Metabolism-Related Genes

The objective of this study was to construct a prognostic
model
by utilizing serine/glycine metabolism-related genes (SGMGs), thus
establishing a risk score for lung adenocarcinoma (LUAD). Based on
the TCGA-LUAD and SGMG data set, two subtypes with different SGMG
expression levels were identified by clustering analysis. Thirteen
differential expression genes were used to construct RiskScore by
Cox regression. GSE72094 data set was used for validation. The survival
characteristics, immune features, and potential benefits of chemotherapy
drugs were analyzed for two risk groups. RiskScore was constructed
based on the genes ABCC12, RIC3, CYP4B1, SFTPB, CACNA2D2, IGF2BP1,
NTSR1, DKK1, CREG2, PITX3, RGS20, FETUB, and IGFBP1. Patients in the
low-risk (LR) group exhibited a superior overall survival. In addition,
aDCs, iDSs, mast cells, neutrophils, HLA, and type II IFN were more
abundant in the LR group with higher IPS scores and lower TIDE scores.
In contrast, NK cells, APC coinhibition, and MHC-I were more common
in the high-risk (HR) group, which may be more sensitive to chemotherapy
drugs such as cisplatin, oxaliplatin, and nilotinib. RiskScore was
a promising biomarker that can be used to distinguish LUAD prognosis,
immune features, and sensitivity to chemotherapy drugs.