Abstract PS1-13-10: Elucidating Stemness-Associated Regulatory Pathways to Guide Personalized Treatment for Ductal Carcinoma in Situ

Abstract Introduction: Ductal carcinoma in situ (DCIS), the most common form of non-invasive breast cancer, is considered a precursor to invasive ductal carcinoma (IDC). A critical gap remains in identifying which DCIS lesions are likely to progress to lethal and invasive disease and tailoring treatment accordingly. Methods: We evaluated the role of cellular plasticity and stem-like traits as drivers of DCIS progression. CytoTRACE, an algorithm that uses single-cell RNA sequencing (scRNA-seq), was utilized to evaluate cellular stemness (PMC7694873). Forty-three patient samples were subjected multi-omic technologies including scRNA-seq analysis of 17 patient-derived DCIS MIND models (8 progressed and 9 non-progressed), spatial transcriptomic analysis of 16 pure DCIS and DCIS with associated IDC (DCIS/IDC) and scATAC/RNA-seq analysis of 10 DCIS and DCIS/IDC samples. We utilized cell annotations published by Kumar and colleagues (PMC10168262). Spatial Graph Convolutional Network (SpaGCN) was used to map the location of stem-like cells. Ligand-receptor (LR) interaction analysis and transcription factor motif enrichment analysis were used to identify LR and regulatory networks within stem-like cells. Statistical tests including Mann Whitney U Test (MWU) was used to compare differences in stemness and a mixed-effect regression model (LMM) was used to assess tumor volume changes in control vs. treated animals. Results: Our preliminary data show that HER2-positive DCIS exhibits significantly higher enrichment in stem-like cells, which correlates with an increased risk of invasive and metastatic progression (MWU; p<0.0001). Analysis of the spatial transcriptomics dataset revealed a significant decrease in the proportion of Basal and Luminal secretory (LumSec) cell subclusters from benign to DCIS and IDC regions. This was mirrored by a significant increase in Luminal Hormone Responsive (LumHR) cells across the same progression (MWU; p<0.01), suggesting that LumHR cells are major contributors to DCIS, with their contribution significantly increasing with the transition to IDC. Consistent with this observation, CytoTRACE analysis confirmed that LumHR cells exhibited the highest stemness scores among all epithelial cell subclusters (MWU; p<0.01). ScRNA-seq and spatial transcriptomics further corroborated that DCIS with invasive potential displayed significantly higher stemness compared to DCIS with non-invasive potential (MWU; p<0.01). These included patient DCIS cells before they showed invasiveness in the MIND models. Therefore, stemness may serve as a signature of DCIS with a risk for future invasive potential. SpaGCN analysis identified a unique DCIS niche that was enriched with high-stem like cells. Further, LR analysis revealed interactions between CEACAM6 on high-stem-like LumHR cells with EGFR on myeloid and lymphatic cells. Finally, scATAC/RNA sequencing revealed an enrichment of FOXA1 transcription factor binding motifs in the cell clusters with the highest stemness scores in HER2+ DCIS. To target FOXA1, we employed bomedemstat, an inhibitor of lysine-specific demethylase 1 (LSD1). LSD1 inhibitors have been shown to target FOXA1 by blocking its demethylation and transcriptional activation (PMC7541538). In vitro, treatment of HER2+ SUM225 cells with bomedemstat decreased the expression of HER2, CEACAM6, and FOXA1. In vivo, bomedemstat showed a significant reduction in tumor volume over time in a HER2+ metastatic breast cancer PDX model (LMM; p<0.01). Conclusion: Cellular stemness may serve as a predictor of DCIS with a future risk for invasive progression. By targeting the key biological drivers of cellular stemness, our goal is to distinguish aggressive from indolent disease, ultimately reducing overtreatment and improving outcomes for patients with high-risk DCIS. Citation Format: F. B. Behbod, E. Schueddig, V. Kochat, E. Arslan, Y. Dallas, P. Yang, R. Madan, Z. Y. Li, T. Fields, J. L. Wagner, K. Larson, C. Balanoff, A. Aripoli, A. Huppe, O. Winblad, J. Peterson, M. Hill, C. Smith, C. Fabian, E. J. Jeffers, D. C. Koestler, A. K. Godwin, L. J. Kilgore, M. T. Lewis, S. A. Khan, A. M. Thompson, K. Rai. Elucidating Stemness-Associated Regulatory Pathways to Guide Personalized Treatment for Ductal Carcinoma in Situ [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS1-13-10.