A new metabolic signature contributes to disease progression and predicts worse survival in melanoma

Abstract

Metabolic reprogramming is a common hallmark of tumor cells and is a crucial mediator of resistance toward anticancer therapies. The pattern of a metabolism-related signature in melanoma remains unknown. Here, we explored the role of a multi-metabolism-related gene signature in melanoma.We used the training and validation sets to develop a multi-metabolism-related gene signature. Cox regression analysis and the least absolute shrinkage and selection operator (LASSO) method were used for constructing a model. The predictive role of the metabolic signature with clinicopathological features of melanoma was also analyzed. Functional analysis of this metabolic signature was also investigated.A ten metabolism-related gene signature was identified and can stratify melanoma into high- and low- risk groups. The signature was correlated with progressive T stage, Breslow thickness, Clark level, and worse survival (all Ps< 0.01). This metabolic signature was shown as an independent prognostic factor and was also a predictive indicator for worse survival in various clinical and molecular features of melanoma. Furthermore, the metabolic signature was implicated in immune responses such as the regulation of T cell activation and cytokine activity. The metabolic signaturewas associated with the progression and worse survival of melanoma. Our study offered a valuable metabolism-targeted therapy approach for melanoma.

Keywords: Metabolism; immune; melanoma; prognosis; progression.

Figure 1.

A metabolism-related gene signature in melanoma. (a) T classification; (b) Breslow thickness; (c) Clark level; (d) Age; (e) Gender; (f) Tumor stage; (g) Lymph node metastasis; (h) Distant metastasis

Figure 2.

Survival prediction of the metabolism-related gene signature in the training set, the validation set, and the whole set of melanoma samples. (a) Samples sorted by risk score and the corresponding survival status. (b) ROC curve of the metabolism-related gene signature. (c) The Kaplan–Meier curves of metabolism-related gene signature (the high- and low-risk groups)

Figure 3.

Outcome prediction of the metabolism-related gene signature for melanoma based on age, gender, and tumor stage. (a) ≥ 60 years; (b) < 60 years; (c) Male samples; (d) Female samples; (e) Stage 3 − 4 melanoma; (f) Stage 0 − 2 melanoma

Figure 4.

Outcome prediction of the metabolism-related gene signature for melanoma based on T classification, lymph node metastasis, and distant metastasis. (a) T 3 − 4 melanoma; (b) T 0 − 2 melanoma; (c) samples with distant metastasis; (d) samples without distant metastasis; (e) samples with lymph node metastasis; (f) samples without lymph node metastasis

Figure 5.

Outcome prediction of the metabolism-related gene signature for melanoma based on Breslow thickness and Clark level. (a) Breslow thickness >1.5 mm; (b) Breslow thickness ≤1.5 mm; (c) Clark IV−V; (d) Clark I− III

Figure 6.

The functional analysis of the metabolism-related gene signature. MF: molecular function; BP: biological process; CC: cellular component