Abstract 1222: Spatial transcriptomics of trastuzumab deruxtecan-treated metastatic breast cancer identifies candidate response and resistance factors.

Abstract Background: Trastuzumab deruxtecan (T-DXd) is a HER2-targeting antibody-drug conjugate approved for metastatic breast cancer (mBC) with HER2-positive, low, and “ultralow” expression. However, reliable tissue biomarkers for predicting response or resistance are poorly understood. We applied GeoMx® digital spatial profiling to 41 T-DXd-treated mBC samples from Yale New Haven Hospital to identify spatially-informed gene expression markers linked to therapy outcomes. Methods: Biopsies from 41 mBC patients treated with T-DXd (5 HER2+, 28 HER2-/HR+, 8 TNBC) were collected. Patients were classified as responders (time-on-treatment >6 months, N=26) or non-responders (≤6 months, N=15). Tissue transfer arrays from these samples were prepared for digital spatial profiling. Spatial profiling was performed on tumor (CK+), immune (CD45+), and stroma (CK- and CD45-) segments using the GeoMx® human whole transcriptome atlas RNA assay. Sequencing data was processed on the Nanostring platform. For each segment, differentially expressed genes (DEGs) between responders and non-responders were identified using a linear mixed model with Benjamini-Hochberg correction. Over-representation analysis used Enrichr gene sets; gene set enrichment analysis (GSEA) used KEGG and Reactome pathways. Results: Over 10,000 genes in 511 segments from 41 patient biopsies passed quality control. In tumor segments, 20 genes upregulated in responders included early estrogen response (MUC1, MLPH, MAST4), extracellular matrix (ECM1), and integrin signaling pathways (COL4A5, NTN4). Non-responders showed upregulation of 37 genes enriched for ubiquitin-specific proteases (ADRM1, PSMB1, PSMA7, UBA52). GSEA indicated increased APC/Cdc20 mediated degradation pathways in non-responder tumor segments. In immune and stroma segments, 12 upregulated genes in responders included negative regulators of immune cell trafficking (RGS1, RGS5). Non-responders upregulated 53 genes, including markers of tumor-associated macrophages (APOE, APOC1, SPARC, SERPINA1) and cancer-associated fibroblasts (FN1, POSTN, COL1A1/2). GSEA showed enrichment for immunoglobulin-mediated immune response, antigen processing, and ECM organization pathways in responder immune and stroma segments. Conclusions: Spatial transcriptomics revealed DEGs associated with T-DXd response and resistance in mBC. Responders were enriched for early estrogen response and ECM/integrin signaling pathways, whereas non-responders showed upregulation of ubiquitin-proteasome pathways and tumor-associated macrophage/cancer-associated fibroblast gene signatures. Validation of these findings is underway with the goal of finding biomarkers that provide predictive value beyond measurement of ADC target. Citation Format: Charles J. Robbins, Mengni He, Julia Benanto, Thazin Nwe Aung, Yalai Bai, Ian E. Krop, David L. Rimm. Spatial transcriptomics of trastuzumab deruxtecan-treated metastatic breast cancer identifies candidate response and resistance factors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 1222.