Abstract LB233: Single-cell transcriptional analysis of embryonic melanoblasts pinpoints embryonic cell states that underlie targeted therapy and immune checkpoint blockade resistance

Abstract Melanoma is a malignant neoplasia of pigment-producing melanocytes notable for its ability to metastasize early during primary tumor growth. Melanomas display typically high mutation burden—a hallmark associated with immune checkpoint blockade therapy (ICBT) sensitivity—but can still evade ICBT. These melanoma phenotypes likely relate to embryonic neural crest cells, a population of cells from which melanocytes arise. For instance, BRAF inhibition-evading melanoma cells adopt neural crest-like states. However, the exact embryonic lineage underlying BRAFi resistance remains unknown as does its relationship with any neural crest-like states that underlie ICBT resistance. To discover such states, we used a mouse fluorescent reporter model of melanoblast development in which melanocyte-specific GFP is driven by the dopachrome tautomerase promoter (iDct-GFP). Embryonic day (E) 11.5 and 15.5 melanoblasts were isolated and sequenced using single-cell transcriptomics (scRNA-seq). These time-points enriched for cells migrating from the dermis to epidermis (at E11.5) or for cells colonizing the hair follicle (at E15.5). These characteristics could help metastatic progression, namely migration across the basement membrane in early invasive melanomas and colonization of metastatic sites. We classified these embryonic melanocytes and their precursors into multiple clusters which in turn were grouped into four key lineages: Neural Progenitor, Schwann Cell Progenitor (SCP), Mesencyhmal (Mes) and Melanocytes (Mel). To detect the presence of these states in tumors, we derived developmental gene modules (DGMs): sets of genes that were preferentially co-expressed in each embryonic cell state. DGMs stratified TCGA samples; SCP-high and Mes-high subtypes were associated with worse survival compared to Mel-high tumors. Moreover, SCP-high tumors had lower T cell infiltration, and were associated with higher NGFR expression, a known marker of ICBT resistance. High expression of SCP DGMs was associated with poor response to anti-PD1 ICBT. Targeting factors underlying the SCP state might therefore sensitize tumours to ICBT. Mes DGMs on the other hand were associated with targeted therapy resistance. Cross-referencing DGMs with published RNA-seq data from melanoma cell lines suggest that SCP and Mes DGMs are induced by pro-inflammatory cytokines secreted by immune cells in the tumor microenvironment. Together, these observations suggest that the SCP state is implicated in the phenomenon of inflammation-induced de-differentiation, a process by which melanomas are able to evade ICBT. Our identification of the SCP and Mes states, along with a deeper understanding of their transcriptional programs and signaling pathways, may reveal clinically actionable targets to disrupt immuno- or targeted-therapy resistance. Citation Format: Vishaka Gopalan, Chun Wai Wong, Chi-Ping Day, Eva Perez-Guijarro, Maxwell P. Lee, Yuhong Jiang, Howard H. Yang, Maira Alves-Constantino, Antonella Sassano, Cari Smith, Mark Simpson, Sung Chin, Jessica Ebersole, Charli Gruen, Emily Wu, Sridhar Hannenhalli, Adam Hurlstone, Glenn Merlino, Kerrie Marie. Single-cell transcriptional analysis of embryonic melanoblasts pinpoints embryonic cell states that underlie targeted therapy and immune checkpoint blockade resistance [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr LB233.