Autophagy-related-gene signature predicts survival outcomes and the tumor immune landscape in skin cutaneous melanoma

Abstract

Introduction: Skin cutaneous melanoma (SKCM) is a malignant skin cancer characterized by heterogeneous outcomes. Autophagy exerts dual effects on tumor progression and immune modulation. Objective: This study aims to investigate whether autophagy-related genes can stratify melanoma patients with different prognoses and to examine their association with the immune microenvironment. Methods: We identified 371 autophagy-related genes from the Human Autophagy Database and analyzed transcriptomic and clinical data from 405 SKCM cases from The Cancer Genome Atlas (TCGA). An autophagy-based prognostic signature was initially screened by univariate Cox regression and then refined using Least Absolute Shrinkage and Selection Operator Cox regression. Risk scores derived from this signature classified patients into high- and low-risk groups, and the predictive power of the signature was validated in TCGA and Gene Expression Omnibus datasets. Functional enrichment, differential expression, and immune deconvolution analyses (Cell-type Identification by Estimating Relative Subsets of Known RNA Transcripts) were used to explore molecular and immune differences between groups. Results: We constructed a 102-autophagy-related-gene signature, which robustly stratifies patients with distinct survival outcomes in different cohorts. Low-risk tumors exhibited immune-activating transcriptional programs and higher infiltration of cluster of differentiation (CD)8+ T lymphocytes, natural killer cells, plasma cells, and memory CD4+ T lymphocytes. High-risk tumors were associated with oncogenic signaling pathways and immunosuppressive subsets, such as M2 macrophages and Tregs. Canonical immune checkpoints, including PDCD1, CD274, CTLA4, LAG3, and TIGIT, were upregulated in the low-risk group and correlated significantly with effector immune infiltration, presenting an “immune-inflamed” phenotype. Conclusion: This autophagy-related-gene signature distinguished prognostic subgroups and highlighted biological and immune diversity in melanoma, providing potential directions for future autophagy research.