Abstract 157: Exploratory analysis of tumor microenvironment using scRNA, scTCR, and spatial transcriptomics in salivary gland cancer with surgical sample after neoadjuvant immuno-chemotherapy

Abstract Background: High-grade resectable salivary gland cancer (SGC) is a rare malignancy, with surgery as the standard primary treatment. However, patients with high-grade histology face a significant risk of regional and distant metastasis, often leading to treatment failure. As part of the ONO-4538-X78 clinical trial, three cycles of neoadjuvant nivolumab combined with cytotoxic chemotherapy were administered prior to surgery. This study aimed to conduct exploratory biomarker analyses using surgical samples to investigate tumor microenvironment (TME) changes after neoadjuvant immunotherapy. Methods: Single-cell RNA sequencing (scRNA-seq) and single-cell T cell receptor (TCR) sequencing were performed on 17 samples from 14 SGC patients, including 3 adjacent normal tissues, 2 pre-treatment biopsies, and 12 post-treatment surgical specimens. Additionally, spatial transcriptomics (Xenium) and histological analyses (H&E via Lunit SCOPE IO) were conducted on 16 post-treatment surgical samples. Responders were defined as patients exhibiting ≤10% viable tumor in surgical specimens. Results: scRNA-seq analysis identified 32, 489 T and NK cells, which were subclustered to reveal cell types associated with treatment response. Non-responders exhibited a predominance of CD4+ memory T cells, while responders demonstrated enrichment of CD8+ dysfunctional T cells and CD8+ memory T cells. Despite statistical insignificance of increased CD8+ T cell subtypes in responders, accompanied by elevated TCR clonality and reduced TCR diversity, suggestive of clonal expansion. Myeloid cell analysis revealed a predominance of tumor-associated macrophages in non-responders, while responders were enriched in dendritic cells and neutrophils. Morphological profiling via Lunit SCOPE IO identified endothelial cells, fibroblasts, lymphocytes, macrophages, tumor cells, and other cell types. Integration of scRNA annotations with Xenium data validated gene expression patterns and identified detailed subtypes. In cancer regions, responders exhibited significantly higher densities of "other" cell populations, predominantly myeloid and stromal cells, including monocyte-derived macrophages, neutrophils, and smooth muscle cells. Conclusions: This study employed an integrative approach combining scRNA-seq, Xenium spatial transcriptomics, and advanced morphological profiling to characterize the TME and its association with neoadjuvant therapy response in SGC. This multi-modal methodology provided enhanced resolution, enabling the identification of distinct cellular compositions and spatial patterns in cancer areas that differed between responders and non-responders. These findings offer valuable insights into the mechanisms of neoadjuvant therapy and may inform future therapeutic strategies for SGC. Citation Format: Hyunsu Kim, Sehhoon Park, Jin Woo Oh, Soohyun Hwang, Jinyoung Kim, Eun-hye Kim, Nayeon Choi, Junhun Cho, Hyun-Ae Jung, Dongryul Oh, Se-Hoon Lee, Yong Chan Ahn, Han-Sin Jeong, Chang Ho Ahn, Chan-Young Ock, Myung-Ju Ahn. Exploratory analysis of tumor microenvironment using scRNA, scTCR, and spatial transcriptomics in salivary gland cancer with surgical sample after neoadjuvant immuno-chemotherapy [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 157.