Dados do Resumo

Título

Molecular signature of four genes forecasts immunotherapy efficacy and prognosis in advanced melanoma

Introdução

Despite promising results, not all melanoma patients benefit from immune checkpoint blockade therapies. Improving the precision of predictive tools is essential for guiding personalized treatment strategies, optimizing patient selection for these therapies, and ultimately enhancing clinical outcomes by identifying individuals most likely to respond to immunotherapy.

Objetivo

This study aimed to develop and validate a gene signature that could accurately predict the response of melanoma patients to anti-programmed cell death-1 (PD-1) immunotherapy.

Métodos

The study intergrated transcriptomic data from the NanoString nCounter Immunology v2 Panel, using formalin-fixed, paraffin-embedded primary tumor samples, with clinical data from 35 melanoma patients treated with anti-PD-1 therapy at Barretos Cancer Hospital between 2016 and 2021. Treatment response was evaluated according to the Response Evaluation Criteria in Solid Tumors for immunotherapy (iRECIST) guidelines. Statistical analysis was performed using nSolver v4.0 in the advanced analysis module, IBM SPSS v23.0, and the R v4.2 environment to identify a gene expression signature associated with immunotherapy response. To validate the proposed gene signature, RNA-sequencing data from two independent cohorts were accessed from the Gene Expression Omnibus database. The study was approved by the Institutional Research Ethics Committee (#1772/2019).

Resultados

Clinicopathological variables were similar across the anti-PD-1 response groups (p>0.05), suggesting treatment response differences were unlikely due to baseline factors, pointing instead to a potential molecular basis. A total of 21 differentially expressed genes were identified between responders (n = 18) and non-responders (n = 17), with 18 genes overexpressed in tumors of non-responder patients. Multivariate logistic regression analysis revealed 4 genes (CD24, NFIL3, FN1, KLRK1) as key outcome predictors. A score incorporating the expression levels of these genes was calculated and demonstrated high accuracy in predicting response (area under the curve [AUC] = 0.935, p<0.001), with higher scores associated with primary treatment resistance. Moreover, higher signature scores were linked to reduced progression-free (p<0.001) and overall survival (p<0.001). Validation in two independent cohorts confirmed the robustness of the proposed signature (AUC = 0.758 and 0.839, respectively).

Conclusões

The 4-gene signature emerges as a potential tool for personalized medicine in melanoma treatment. Implementing this signature could facilitate the optimization of therapeutic strategies, promoting the development of effective and cost-efficient treatment approaches for integration into healthcare systems.

Financiador do resumo

This work was supported by the Grupo Español Multidisciplinar de Melanoma (GEM); the São Paulo Research Foundation (#2019/07111-9 and #2019/03570-9); and the Researchers Assistance and Incentive Program (PAIP - Barretos Cancer Hospital).

Palavras Chave

Metastatic melanoma; immunotherapy; prediction of treatment response