P188 Single-cell RNA and T-cell receptor sequencing reveals novel regulatory mechanisms of CD8+ memory T-cell formation and identifies a marker of response to checkpoint inhibitor therapy in melanoma
Goran MicevicAbstract
Melanoma is the deadliest form of skin cancer, with 104 960 new cases and 8430 deaths estimated for 2025 in the USA. Despite significant progress with immune checkpoint inhibitors (ICIs) targeting programmed death (PD)-1 and cytotoxic T-lymphocyte-associated protein (CTLA)-4, many patients with advanced melanoma do not have a durable response to ICIs and are at high risk for recurrence and death. Lack of a durable response is linked to failure of T-cell memory formation. However, firstly, our understanding of memory T-cell formation during ICI therapy is limited, and secondly there are currently no markers to identify patients predisposed to impaired memory T-cell formation and poor response to ICIs. Here, we use a Braf V600E-driven mouse model of melanoma, and patient samples to characterize the single-cell transcriptional and TCR profiles. The analyses were designed to identify CD8+ T-cell clonal expansion with distinct transcriptomic and epigenetic signatures associated with response to CTLA-4 and PD-1 blockade in vivo. In patients responding to therapy, we found expansion of a dominant tumour-specific CD8+ T clone in the tumour microenvironment and lymph nodes. In addition to interleukin-7 receptor (IL-7R) pathway activation, samples from responders showed upregulated cytotoxic and natural killer cell-like properties, expression of killer cell lectin receptor family members, and interferon-induced transmembrane proteins in CD8+ T cells. In contrast, nonresponders exhibited polyclonal exhausted CD8+ populations (P < 0.005). Increasing IL-7R reduced exhaustion and increased response to therapy. IL-7R hypomethylation was associated with longer overall survival (P < 0.001) and is an epigenetic marker of therapy response associated with a fourfold prolongation (P < 0.05) of response and differentiates between responders and nonresponders (hazard ratio 3.1). We confirmed our findings in a separate cohort of patients with melanoma. These analyses uncovered an IL-7R-dependent regulatory network that orchestrates multiple transcriptional programmes. The network contributes to response to ICI therapy in melanoma and highlights a marker that may be clinically useful to identify nonresponders to ICIs.