Abstract B046: Predictive biomarkers and mechanistic insights in stage III/IV melanoma patients undergoing first-line immune checkpoint inhibitors therapy

Abstract Immune checkpoint inhibitor therapy leads to a strong and durable response in some metastatic melanoma patients, but some patients remain refractory to treatment. Identifying predictive biomarkers for therapeutic response is pivotal in advancing immunotherapy for melanoma. In this study, we analyzed 52 stage III and IV melanoma patients (14 non-responders and 38 responders) undergoing first-line immune checkpoint inhibitors therapy, of which 47 patients received anti-PD1 + anti-CTLA-4 and 5 patients received anti-PD-1 alone. Our main objective was uncovering factors associated with therapeutic response. Methods included multiplex bead array for plasma analytes, flow cytometry for immune profiling in blood and tumors, exome and transcriptome tumor sequencing, and scRNA-Seq for characterizing leukocytes in blood and tumor samples. Among 31 cytokines evaluated, CXCL9 emerged as a strong predictor of response, distinguishing responders prior to treatment and sustaining elevated levels throughout therapy in responders. Flow cytometry revealed responder blood samples enriched with CD8+ T lymphocytes expressing activation and cytotoxic markers, in addition to CXCR3, the receptor for CXCL9. Notably, TIM-3 correlated with complete versus partial response. On the other hand, non-responders showed elevated levels of CD161, PD-1, CD95, and PD-L2. Genomic characterization linked responder tumors with higher tumor mutational burden and lower tumor heterogeneity. Transcriptomic analysis demonstrated higher expression of CXCL9, CXCR3 and IFNG in responder tumors, mirroring systemic immune status. Lastly, scRNA-Seq data showed responder T cells differentiating into populations expressing GNLY, GZMA, GZMB, PRF1, IFNG, IFNGR1, and CXCR3. Conversely, non-responder T cells exhibited markers like TGFBI, ENTPD1, STAT6, ITGAE, IRF4, and KLRB1. Our findings suggest CXCL9 and CXCR3-related markers as potential predictive biomarkers and indicate distinctive immune pathways in responders, providing insight into the mechanisms underlying clinical outcomes in melanoma immunotherapy. Citation Format: Amanda B Figueiredo, Guilherme FB Evangelista, Milton JB Silva, Larissa M Kuil, Clara M Cavalcanti, Katia LP Morais, Juliana S Apostolico, Robert Balderas, Rodrigo P Lopes, Israel T Silva, Alexandre Defelicibus, Renan Valieris, James Turner, Nayane AL Galdino, Iasmim P Santos, Stephanie MI Ferreira, Ananda D Lopes, Clovis AL Pinto, Rubens Chojniak, Joao P Duprat-Neto, Kenneth J Gollob. Predictive biomarkers and mechanistic insights in stage III/IV melanoma patients undergoing first-line immune checkpoint inhibitors therapy [abstract]. In: Proceedings of the AACR IO Conference: Discovery and Innovation in Cancer Immunology: Revolutionizing Treatment through Immunotherapy; 2025 Feb 23-26; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Immunol Res 2025;13(2 Suppl):Abstract nr B046.