AUTHOR=Wu Zhiqiang , Wang Jing , Li Yanan , Liu Jianmin , Kang Zijian , Yan Wangjun TITLE=Characterization of a lactate metabolism-related signature for evaluation of immune features and prediction prognosis in glioma JOURNAL=Frontiers in Neurology VOLUME=13 YEAR=2023 URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2022.1064349 DOI=10.3389/fneur.2022.1064349 ISSN=1664-2295 ABSTRACT=Background

Glioma is one of the most typical tumors in the central nervous system with a poor prognosis, and the optimal management strategy remains controversial. Lactate in the tumor microenvironment is known to promote cancer progression, but its impact on clinical outcomes of glioma is largely unknown.

Methods

Glioma RNA-seq data were obtained from TCGA and GCGA databases. Lactate metabolism genes (LMGs) were then evaluated to construct an LMG model in glioma using Cox and LASSO regression. Immune cell infiltration, immune checkpoint gene expression, enriched pathways, genetic alteration, and drug sensitivity were compared within the risk subgroups. Based on the risk score and clinicopathological features, a nomogram was developed to predict prognosis in patients with glioma.

Results

Five genes (LDHA, LDHB, MRS2, SL16A1, and SL25A12) showed a good prognostic value and were used to construct an LMG-based risk score. This risk score was shown as an independent prognostic factor with good predictive power in both training and validation cohorts (p < 0.001). The LMG signature was found to be correlated with the expression of immune checkpoint genes and immune infiltration and could shape the tumor microenvironment. Genetic alteration, dysregulated metabolism, and tumorigenesis pathways could be the underlying contributing factors that affect LMG risk stratification. The patients with glioma in the LMG high-risk group showed high sensitivity to EGFR inhibitors. In addition, our nomogram model could effectively predict overall survival with an area under the curve value of 0.894.

Conclusion

We explored the characteristics of LMGs in glioma and proposed an LMG-based signature. This prognostic model could predict the survival of patients with glioma and help clinical oncologists plan more individualized and effective therapeutic regimens.