Machine learning-derived immunosenescence index for predicting outcome and drug sensitivity in patients with skin cutaneous melanoma

Abstract

The functions of immunosenescence are closely related to skin cutaneous melanoma (SKCM). The aim of this study is to uncover the characteristics of immunosenescence index (ISI) to identify novel biomarkers and potential targets for treatment. Firstly, integrated bioinformatics analysis was carried out to identify risk prognostic genes, and their expression and prognostic value were evaluated. Then, we used the computational algorithm to estimate ISI. Finally, the distribution characteristics and clinical significance of ISI in SKCM by using multi-omics analysis. Patients with a lower ISI had a favorable survival rate, lower chromosomal instability, lower somatic copy-number alterations, lower somatic mutations, higher immune infiltration, and sensitive to immunotherapy. The ISI exhibited robust, which was validated in multiple datasets. Besides, the ISI is more effective than other published signatures in predicting survival outcomes for patients with SKCM. Single-cell analysis revealed higher ISI was specifically expressed in monocytes, and correlates with the differentiation fate of monocytes in SKCM. Besides, individuals exhibiting elevated ISI levels could potentially receive advantages from chemotherapy, and promising compounds with the potential to target high ISI were recognized. The ISI model is a valuable tool in categorizing SKCM patients based on their prognosis, gene mutation signatures, and response to immunotherapy.