부산대학교 도서관

Functional gene expression signatures (FGES) from pretreatment biopsy samples have been used to predict the responses of metastatic melanoma to immune checkpoint blockade (ICB) therapies. However, there are no predictive FGE signatures from patients receiving treatment. Here, using the Elastic Net Regression (ENLR) algorithm, we analyzed transcriptomic and matching clinical data from a dataset of patients with metastatic melanoma treated with ICB therapies and produced an FGE signature for pretreatment (FGES-PRE) and on-treatment (FGES-ON). Both the FGES-PRE and FGES-ON signatures are validated in three independent datasets of metastatic melanoma as the validation set, achieving area under the curve (AUC) values of 0.44–0.81 and 0.82–0.83, respectively. Then, we combined all test samples and obtained AUCs of 0.71 and 0.82 for the FGES-PRE and FGES-ON signatures, respectively. The FGES-ON signatures had a higher predictive value for prognosis than the FGES-PRE signatures. The FGES-PRE and FGES-ON signatures were divided into high- and low-risk scores using the signature score mean value. Patients with a high FGE signature score had better survival outcomes than those with low scores. Overall, we determined that the FGES-ON signature is an effective biomarker for metastatic melanoma patients receiving ICB therapy. This work would provide an important theoretical basis for applying FGE signatures derived from on-treatment tumor samples to predict patients' therapeutic response to ICB therapies. [ABSTRACT FROM AUTHOR]