Abstract B012: Clinical Outcomes and Characterization of the Tumor Microenvironment in Fusion-Positive Cutaneous Melanoma

Abstract Background: Oncogenic gene fusions in cutaneous melanoma are uncommon and their clinical relevance in terms of prognosis and treatment has been poorly delineated. Outcomes to therapy, and characterization of the immune landscape associated with these fusion events have not been previously explored. Methods: Comprehensive genomic profiling (DNA + RNA sequencing) was performed on 13,835 melanoma cases submitted to Caris Life Sciences (Phoenix, AZ) between 2015 and 2025. Pathogenic and likely pathogenic fusions were identified using an RNA-based fusion detection pipeline and integrated with mutational, clinical, and therapeutic outcome data. Immune cell infiltration in the tumor microenvironment was estimated by deconvolution of the RNAseq data using Quantiseq. Real-world overall survival (OS[AE1] ) was obtained from insurance claims data, calculated from start of biopsy or immune checkpoint inhibitors (ICI) to last contact. Mann-Whitney U and X2/Fisher-Exact tests were applied, with p-values adjusted (p < 0.05). Results: A total of 388 melanoma cases (2.8%) harbored pathogenic or likely pathogenic fusions. Across melanoma subtypes, BRAF fusions were most prevalent (39%), followed by RAF1 (14%) and ALK (9%) fusions. Overall fusion prevalence was significantly higher in BRAF/NRAS/NF1 wild-type (WT) compared with mutant cohorts (9.7% vs 0.7%; p<0.0001) and was further enriched in triple-WT tumors with TERT promoter mutations (14.3% vs 9.7%; p<0.01). High tumor mutational burden (310 mutations/Mb) was significantly less frequent in fusion-positive melanoma cases compared to fusion-negative (26.0% vs 59.0%, p<0.0001). No survival advantage was observed based on the oncogenic fusion partner, although a trend toward prolonged survival was observed for patients with MET fusions (median OS =NR, p=0.18). BRAF fusions were mutually exclusive from other MAPK associated mutations (log odds ratio (OR)<1, p<0.05), while CDK2 fusions were concurrent with NRAS and NF1 mutations (OR>2; n =3, p<0.05). Neutrophils and T-regulatory cell fractions were also higher in BRAF fusion-positive cases when compared to triple-WT and BRAF-mutated cases. OS of patients with BRAF fusion-positive cutaneous melanomas (median OS = 29.2 months) was worse compared to those with BRAF-mutated (47.4 months, p<0.01) but similar to triple-WT cases (34.0 months, p=0.202). Conclusions: To our knowledge, this is the largest dataset describing the incidence patterns, immune infiltrative landscape and clinical outcomes of patients with fusion-positive cutaneous melanomas. We demonstrate here that BRAF fusions have worse OS from initiation of ICI compared to BRAF mutant cases. Additionally, BRAF fusion cases tend to have a more immunosuppressive tumor microenvironment suggesting reduced sensitivity to immune checkpoint blockage and potential poorer clinical outcomes. Citation Format: Matthew J. Hadfield, Tolulope Adeyelu, Andrew Elliott, Hassan Ghani, Farah Abdulla, Ari Vanderwalde, Ryan J. Sullivan, Sapna P. Patel. Clinical Outcomes and Characterization of the Tumor Microenvironment in Fusion-Positive Cutaneous Melanoma [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Fusion-Positive Cancer: From Discovery to Therapy; 2026 Jan 13-15; Philadelphia PA. Philadelphia (PA): AACR; Cancer Res 2026;86(1_Suppl):Abstract nr B012.