Background: Primary breast cancer is routinely subtyped using immunohistochemistry (IHC) staining and treatment choice is guided by the subtype. Targeted treatment requires the targeted signaling pathway to be active and tumor driving. IHC staining does not provide reliable information on active signaling pathways, and we reported before that clinical response of ER positive breast cancer patients to neoadjuvant aromatase inhibitor therapy appears to be more closely related to ER pathway activity rather than ER/PR IHC staining. To better understand the role of tumor driving signaling pathways within breast cancer subtypes and potentially improve precision medicine we evaluated signaling pathway activity profiles per IHC subtype, using novel assays to quantitatively measure activity of multiple main tumor driving signal transduction pathways, within tissue samples.

Methods: mRNA-based estrogen (ER) and androgen (AR) receptor, MAPK, PI3K, Hedgehog (HH), TGFβ, Notch, Wnt signal transduction pathway assays have been previously described. Pathway activity scores (PAS) were measured on primary untreated breast cancer samples, and compared to normal breast epithelium, using Affymetrix expression dataset GSE65194 (Institut Marie Curie), excluding the duplicate measures, containing 29 Luminal A, 30 Luminal B, 30 HER2, 41 triple negative (TN) breast cancer and 6 normal breast tissue sample data. Pathway activity scores exceeding the 95th percentile of normal were considered highly active.

Results: PAS thresholds for defining highly active pathway activity were for MAPK 23, AR 29, ER 34, PI3K 42, HH 34, Notch 65, TGFβ 48, Wnt 29. Luminal A was characterized by high ER pathway activity (n=25, 86%, mean PAS 45, SD 10); luminal B by high ER (n=10, 33%; mean PAS 32, SD 11) increased PI3K pathway activity (n=27, 90%; mean PAS 53, SD 10) and increased HH pathway activity (n=23, 77% mean PAS 29, SD 7); HER2 by high AR (n=18, 60%; mean PAS 30, SD 6) increased PI3K pathway activity (n=19, 63%; mean PAS 48, SD8) and increased HH pathway activity (n=27, 90% mean PAS 30 SD 7) ; TNBC by high PI3K (n=37, 90%; mean PAS 53, SD 7) plus high developmental pathway activity (Notch, n=6, 15%; mean PAS 53, SD 12; Wnt, n=28, 68%; mean PAS 31, SD 6). All subtypes had elevated MAPK pathway activity.

Discussion: Each breast cancer subtype based on IHC, had in the overall population a typical pathway activity profile. However within each IHC subtype, a substantial range of pathway activities between patients is observed, suggesting that within same IHC subtypes, clear differences in signal pathway activation can be measured that may provide a novel approach to more effectively select patients for specific precision oncology targeted therapy treatments.

Citation Format: Anthony M. Magliocco, Anja van de Stolpe, Martijn Akse, Eveline den Biezen, Laurent Holtzer. Breast cancer IHC subtypes display heterogenous and independent targetable signaling pathway activity profiles [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 533.