Abstract 757: Decoding resistance: Spatial transcriptomics reveals tumor core and edge dynamics in high-risk cutaneous squamous cell carcinoma treated with neoadjuvant cemiplimab

Introduction: Immune checkpoint inhibitor (ICI) therapies have redefined the treatment landscape of advanced cutaneous squamous cell carcinoma (CSCC). NeoPOWER is an ongoing, single-arm, phase II clinical trial studying the role of preoperative cemiplimab in patients with high risk/unresectable CSCC. Based on recent data, patients with CSCC who attain a pathologic complete response with neoadjuvant ICI have a 12-month disease free survival of 92%. Response rates to ICI; however, are approximately 50-60% and CSCC refractory to ICI is associated with an inferior survival, highlighting a need to stratify patients who are most likely to benefit from neoadjuvant cemiplimab. We employ Visium (v2, 10x genomics) to comprehensively analyze the CSCC tumor immune microenvironment of patients with divergent responses to immunotherapy. Methods: We performed Visium spatial transcriptomics (ST) profiling on formalin fixed paraffin embedded biopsy samples from patients with high risk or unresectable CSCC treated with neoadjuvant cemiplimab. Nine patients were included and divided into two groups as defined by presence of residual disease (RD) or no residual disease (NRD). An integrated RD and NRD pre-treatment (pre-iRD/NRD) analysis (n = 8 samples) was performed to identify factors predictive of treatment response. Separately, an integrated RD pre-treatment and post treatment (iRD) analysis was explored to assess for mechanisms of treatment resistance (n = 7 samples). Gene expression data generated by Visium was imported and analyzed in Seurat. Results: We observed that 2 distinct clusters identified as CSCC, histologically and transcriptionally, formed a substructure comprised of an outer and inner layer tumor edge (TE) and tumor core (TC), respectively. This association of spatial adjacency was visually evident in both pre-iRD/NRD and iRD groups. Comparative gene set enrichment analysis between pre-iRD/NRD showed alterations in KRAS signaling pathways and down-regulation of multiple immune response pathways, implying fundamental differences in the TC of both response groups. In contrast, comparisons of the iRD TC showed post-treatment enrichment of epithelial mesenchymal transition, hypoxia, and metabolic pathways. Post treatment TE was similarly enriched in EMT and hypoxia with down-regulation of immune response pathways. Alterations in immune cell proportions post-treatment were notable in the iRD group following ICI, as were differences in baseline B and T cell proportions in pre-iRD/NRD tumor clusters. Ligand receptor analysis highlighted key immune-inhibitory interactions within the TE and TC. Conclusions: ST analysis of the pre- and post-treatment CSCC microenvironment captures tumor edge/core dynamics associated with different response outcomes. Citation Format: Ali Baghian, Hugo Lara-Martinez, Rania Bassiouni, Gino In, Homa Dadrastoussi, Paniz Zarand, Harley Wu, Kaijin Wu, Brittney K. DeClerck, Ashley Gochoco, Kevin Kelly. Decoding resistance: Spatial transcriptomics reveals tumor core and edge dynamics in high-risk cutaneous squamous cell carcinoma treated with neoadjuvant cemiplimab [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 757.