Integrating Machine Learning Algorithms to Construct a Triaptosis-Related Prognostic Model in Melanoma.
Jiaheng Xie1, Min Zhang1, Min Qi2
1Department of Plastic Surgery, Xiangya Hospital, Central South University, Changsha, 410008, People's Republic of China.
Cancer Management and Research
|June 19, 2025
Summary
This study reveals triaptosis, a programmed cell death (PCD) mechanism, is linked to melanoma prognosis and immune cell infiltration. Triaptosis-associated signatures may serve as biomarkers and therapeutic targets for melanoma treatment.
Area of Science:
- Oncology
- Molecular Biology
- Immunology
Background:
- Melanoma is an aggressive skin cancer with high mortality due to metastasis and therapy resistance.
- Programmed cell death (PCD), including apoptosis and ferroptosis, influences tumor progression and treatment response.
- Triaptosis, a recently identified PCD pathway, has unexplored implications in melanoma.
Purpose of the Study:
- To investigate the role of triaptosis in melanoma.
- To identify key triaptosis-related genes and pathways in melanoma.
- To develop a prognostic signature for melanoma based on triaptosis.
Main Methods:
- Integrated single-cell and bulk RNA sequencing data.
- Constructed a prognostic signature using machine learning (SurvivalSVM) with TCGA-SKCM and GEO datasets.
- Performed survival analysis, ROC curve analysis, PCA, and immune infiltration analysis.
Main Results:
- Developed a robust triaptosis-associated signature (TAS) with high predictive performance.
- High-risk patients identified by TAS showed significantly worse overall survival.
- TAS was significantly associated with immune cell populations and tumor microenvironment.
Conclusions:
- Triaptosis-related gene expression patterns correlate with melanoma prognosis and immune infiltration.
- Triaptosis presents a potential biomarker and therapeutic target for melanoma.
- Findings offer strategies to enhance melanoma treatment efficacy and overcome resistance.
