A four-gene signature predicts survival and anti-CTLA4 immunotherapeutic responses based on immune classification of melanoma.
Ying MeiMei-Ju May ChenHan LiangLi MaPublished in: Communications biology (2021)
Cutaneous melanoma is the most malignant skin cancer. Biomarkers for stratifying patients at initial diagnosis and informing clinical decisions are highly sought after. Here we classified melanoma patients into three immune subtypes by single-sample gene-set enrichment analysis. We further identified a four-gene tumor immune-relevant (TIR) signature that was significantly associated with the overall survival of melanoma patients in The Cancer Genome Atlas cohort and in an independent validation cohort. Moreover, when applied to melanoma patients treated with the CTLA4 antibody, ipilimumab, the TIR signature could predict the response to ipilimumab and the survival. Notably, the predictive power of the TIR signature was higher than that of other biomarkers. The genes in this signature, SEL1L3, HAPLN3, BST2, and IFITM1, may be functionally involved in melanoma progression and immune response. These findings suggest that this four-gene signature has potential use in prognosis, risk assessment, and prediction of anti-CTLA4 response in melanoma patients.
Keyphrases
- skin cancer
- end stage renal disease
- risk assessment
- immune response
- ejection fraction
- genome wide
- newly diagnosed
- chronic kidney disease
- copy number
- prognostic factors
- peritoneal dialysis
- machine learning
- squamous cell carcinoma
- genome wide identification
- young adults
- patient reported
- human health
- transcription factor
- free survival
- lymph node metastasis