Pan-cancer analysis identifies proteasome 26S subunit, ATPase (PSMC) family genes, and related signatures associated with prognosis, immune profile, and therapeutic response in lung adenocarcinoma
Hui Jia,Wen-Jin Tang,Lei Sun,Chong Wan,Yun Zhou,Wei-Zhong Shen
DOI: https://doi.org/10.3389/fgene.2022.1017866
IF: 3.7
2023-01-10
Frontiers in Genetics
Abstract:Background: Proteasome 26S subunit, ATPase gene (PSMC) family members play a critical role in regulating protein degradation and are essential for tumor development. However, little is known about the integrative function and prognostic significance of the PSMC gene family members in lung cancer. Methods: First, we assessed the expression and prognostic features of six PSMC family members in pan-cancer from The Cancer Genome Atlas (TCGA) dataset. Hence, by focusing on the relationship between PSMC genes and the prognostic, genomic, and tumor microenvironment features in lung adenocarcinoma (LUAD), a PSMC-based prognostic signature was established using consensus clustering and multiple machine learning algorithms, including the least absolute shrinkage and selection operator (LASSO) Cox regression, CoxBoost, and survival random forest analysis in TCGA and GSE72094. We then validated it in three independent cohorts from GEO and estimated the correlation between risk score and clinical features: genomic features (alterations, tumor mutation burden, and copy number variants), immune profiles (immune score, TIDE score, tumor-infiltrated immune cells, and immune checkpoints), sensitivity to chemotherapy (GDSC, GSE42127, and GSE14814), and immunotherapy (IMvigor210, GSE63557, and immunophenoscore). Twenty-one patients with LUAD were included in our local cohort, and tumor samples were submitted for evaluation of risk gene and PD-L1 expression. Results: Nearly all six PSMC genes were overexpressed in pan-cancer tumor tissues; however, in LUAD alone, they were all significantly correlated with overall survival. Notably, they all shared a positive association with increased TMB, TIDE score, expression of immune checkpoints (CD276 and PVR), and more M1 macrophages but decreased B-cell abundance. A PSMC-based prognostic signature was established based on five hub genes derived from the differential expression clusters of PSMC genes, and it was used to dichotomize LUAD patients into high- and low-risk groups according to the median risk score. The area under the curve (AUC) values for predicting survival at 1, 3, and 5 years in the training cohorts were all >.71, and the predictive accuracy was also robust and stable in the GSE72094, GSE31210, and GSE13213 datasets. The risk score was significantly correlated with advanced tumor, lymph node, and neoplasm disease stages as an independent risk factor for LUAD. Furthermore, the risk score shared a similar genomic and immune feature as PSMC genes, and high-risk tumors exhibited significant genomic and chromosomal instability, a higher TIDE score but lower immune score, and a decreased abundance of B and CD8 + T cells. Finally, high-risk patients were suggested to be less sensitive to immunotherapy but had a higher possibility of responding to platinum-based chemotherapy. The LUAD samples from the local cohort supported the difference in the expression levels of these five hub genes between tumor and normal tissues and the correlation between the risk score and PD-L1 expression. Conclusion: Overall, our results provide deep insight into PSMC genes in LUAD, especially the prognostic effect and related immune profile that may predict therapeutic responses.
genetics & heredity