Deciphering the prognostic value and immunotherapeutic strategy of aggrephagy in melanoma: Integrating single-cell sequencing and machine learning

This study aimed to investigate the role of aggrephagy in cutaneous melanoma (CM) and explore its potential as a biomarker for prognosis and therapeutic targeting. We utilized single-cell sequencing technology and machine learning algorithms to analyze melanoma transcriptome data from the TCGA database and validated our findings using 3 independent datasets from the GEO database. By employing enrichment scoring in single-cell sequencing, we identified characteristic expression patterns of different cell types involved in aggrephagy and constructed an aggrephagy-related signature (ARS). We further evaluated the association of ARS with clinical features, immune cell infiltration, tumor mutational load (TMB), and immune checkpoint gene expression. Additionally, we conducted in vivo experiments by knocking down TPX2, the most critical oncogene in ARS, using shRNA and assessed its effects on tumor proliferation and T-cell growth via subcutaneous tumor formation assays and flow cytometry in mice. The ARS demonstrated robust prognostic predictive power across multiple datasets, with higher ARS scores associated with poorer overall survival (OS) and lower levels of immune cell infiltration. Patients with low ARS scores were more likely to benefit from immune checkpoint inhibitor therapies, while those with high scores exhibited increased sensitivity to 2 common chemotherapeutic agents. Compared to published melanoma prognostic models, our ARS showed higher accuracy and stability. The construction of an ARS-related nomogram further facilitated more accurate clinical decision-making. In vivo experiments confirmed that TPX2 knockdown inhibited tumor proliferation and enhanced T-cell growth, highlighting its critical role in CM progression. Our study highlights the complex functions of the aggrephagy-related signature in cutaneous melanoma, underscoring its potential as a therapeutic target and a valuable tool for prognostic assessment.