Abstract PS2-10-16: Distinct Immune Landscapes in Inflammatory and Metaplastic Breast Cancer: Insights from Transcriptomic Profiling

Abstract Background: Inflammatory breast cancer (IBC) and metaplastic breast cancer (MpBC) are rare subtypes associated with poor prognosis and limited therapeutic targets. While both subtypes exhibit resistance to immune checkpoint inhibitors (ICIs), the mechanisms of immune evasion and immune cell exclusion are not fully understood. To address this, we performed a comparative immune transcriptomic analysis of IBC and MpBC tumors using a large institutional dataset, aiming to define distinct features. Methods: A total of 341 breast cancer tumors were evaluated using BostonGene’s comprehensive AI-driven transcriptomic analysis platform. The cohort included: IBC (n=60), MpBC (n=4), invasive lobular carcinoma (ILC) (n=84), and invasive ductal carcinoma (IDC) controls (n=193). Gene expression profiling was conducted using a panel of 504 cancer;immune-related genes, including key immune checkpoint regulators (PD-L1/CD274), epithelial-mesenchymal transition (EMT) markers (VIM, SNAI2), and antibody-drug conjugate (ADC) targets such as TROP2 (TACSTD2). Immune findings are derived from the RNA-seq data provided by BostonGene but were not categorized within a predefined immune-related gene set. Pathway-level difference analysis using Gene Set Expression Analysis and PAM50 intrinsic subtyping were also conducted. Due to the small number of MpBC cases, results were interpreted in the context of biological trends rather than statistical significance. Results: Transcriptomic profiling revealed differences in immune/stromal features between IBC and MpBC. IBC demonstrated upregulation of JAK/STAT and NF-κB signaling pathways, consistent with an inflamed tumor milieu. Our analysis showed no significant difference in expression of PD-L1 (CD274) between IBC and IDC. PD-L1 expression was low among MpBC tumors, although not statistically significant. In contrast, MpBC tumors displayed a enrichment in EMT signatures, including high expression of VIM. This phenotype is consistent with an immune-cold TIME, where T cell infiltration is minimal and suppressed checkpoint molecule expression. Expression of TROP2 (TACSTD2) was elevated in both IBC and MpBC, though MpBC samples exhibited greater heterogeneity. PAM50 subtyping revealed that IBC and MpBC were predominantly basal-like, with ILC cases largely luminal and IDC displaying mixed profiles. Chi-squared analysis showed no significant difference between IBC and IDC subtypes. Conclusions: IBC and MpBC exhibit distinct but clinically significant immune landscapes. IBC shows features of an inflamed but immunosuppressed microenvironment, characterized by active cytokine signaling and checkpoint expression. MpBC expresses higher level of EMT signaling and stroma related genes. These findings suggest IBC may respond to ICIs, particularly when PD-L1 expression or T cell infiltration is evident, while MpBC may require combination strategies that overcome EMT-mediated immune suppression and stromal exclusion. Additionally, the heterogeneous TROP2 expression may predict response to TROP2-targeted ADCs, such as sacituzumab govitecan. While future validation is needed, our data highlight the importance of immune phenotyping in rare breast cancer subtypes to guide therapeutic strategies and trial design. Acknowledgement: The transcriptomic analyses were performed via collaboration with BostonGene and Morgan Welch IBC Research and Clinic Program (MW IBC program). MW IBC program is funded by the State of Texas Rare and Aggressive Breast Cancer Grant. Citation Format: E. R. Lopez, A. Nasrazadani, O. Baranov, V. Smirnov, A. Love, F. Paradiso, M. Hensley, R. Layman, J. Mouabbi, C. Yam, B. Nelson, S. Sadia, V. Valero, M. Stauder, W. Woodward, A. Lucci, S. Sun, G. Whitman, M. Patel, H. Le-Petross, Y. Lu, A. Marx, A. Alexander, C. Yajima, M. Kai, L. Villarreal, H. Lopez, S. H. Giordano, J. K. Litton, L. Bora. Distinct Immune Landscapes in Inflammatory and Metaplastic Breast Cancer: Insights from Transcriptomic Profiling 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 PS2-10-16.