Characterization and application of a lactate and branched chain amino acid metabolism related gene signature in a prognosis risk model for multiple myeloma

Zhengyu Yu; Bingquan Qiu; Hui Zhou; Linfeng Li; Ting Niu

doi:10.1186/s12935-023-03007-4

Characterization and application of a lactate and branched chain amino acid metabolism related gene signature in a prognosis risk model for multiple myeloma

Cancer Cell Int. 2023 Aug 14;23(1):169. doi: 10.1186/s12935-023-03007-4.

Authors

Zhengyu Yu^#¹, Bingquan Qiu^#², Hui Zhou^#¹, Linfeng Li¹, Ting Niu³

Affiliations

¹ Department of Hematology, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan, China.
² Department of Biochemistry and Biophysics, School of Basic Medical Sciences, Peking University Health Science Center, Beijing, China.
³ Department of Hematology, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan, China. niuting@wchscu.cn.

^# Contributed equally.

Abstract

Background: About 10% of hematologic malignancies are multiple myeloma (MM), an untreatable cancer. Although lactate and branched-chain amino acids (BCAA) are involved in supporting various tumor growth, it is unknown whether they have any bearing on MM prognosis.

Methods: MM-related datasets (GSE4581, GSE136337, and TCGA-MM) were acquired from the Gene Expression Omnibus (GEO) database and the Cancer Genome Atlas (TCGA) database. Lactate and BCAA metabolism-related subtypes were acquired separately via the R package "ConsensusClusterPlus" in the GSE4281 dataset. The R package "limma" and Venn diagram were both employed to identify lactate-BCAA metabolism-related genes. Subsequently, a lactate-BCAA metabolism-related prognostic risk model for MM patients was constructed by univariate Cox, Least Absolute Shrinkage and Selection Operator (LASSO), and multivariate Cox regression analyses. The gene set enrichment analysis (GSEA) and R package "clusterProfiler"were applied to explore the biological variations between two groups. Moreover, single-sample gene set enrichment analysis (ssGSEA), Microenvironment Cell Populations-counter (MCPcounte), and xCell techniques were applied to assess tumor microenvironment (TME) scores in MM. Finally, the drug's IC50 for treating MM was calculated using the "oncoPredict" package, and further drug identification was performed by molecular docking.

Results: Cluster 1 demonstrated a worse prognosis than cluster 2 in both lactate metabolism-related subtypes and BCAA metabolism-related subtypes. 244 genes were determined to be involved in lactate-BCAA metabolism in MM. The prognostic risk model was constructed by CKS2 and LYZ selected from this group of genes for MM, then the prognostic risk model was also stable in external datasets. For the high-risk group, a total of 13 entries were enriched. 16 entries were enriched to the low-risk group. Immune scores, stromal scores, immune infiltrating cells (except Type 17 T helper cells in ssGSEA algorithm), and 168 drugs'IC50 were statistically different between two groups. Alkylating potentially serves as a new agent for MM treatment.

Conclusions: CKS2 and LYZ were identified as lactate-BCAA metabolism-related genes in MM, then a novel prognostic risk model was built by using them. In summary, this research may uncover novel characteristic genes signature for the treatment and prognostic of MM.

Keywords: Branched-chain amino acids; Drug prediction; Lactate; Multiple myeloma; Prognosis; Tumor microenvironment.