113MO - Spatial transcriptomics-derived classification of invasive lobular carcinoma: Associations with clinical, genomic characteristics, and prognosis

Background

Invasive lobular carcinoma (ILC) is the second most common histological breast cancer subtype. Here, we classified ILC into four subtypes based on tumor microenvironment (TME) heterogeneity and investigated their association with clinical characteristics and prognosis.

Methods

Spatial transcriptomics (ST, Visium 10X) was performed on 43 hormone receptor-positive, HER2-negative ILC frozen tumor samples. By integrating morphology and sequencing data, ILCs were classified based on TME heterogeneity. Using gene signatures, these subtypes were retrieved in METABRIC (n=122) and SCAN-B (n=853) datasets. Chi-square, Kruskal-Wallis tests, and Cox proportional hazard models analyzed associations with clinical features and survival.

Results

Four ILC subtypes were identified using ST: proliferative (P, enriched in tumor cells and proliferation-related pathways), normal-stroma enriched (NSE, enriched in fibroblasts, carcinoma in situ and EMT-related pathways), metabolic (M, enriched in endothelial cells, metabolic pathways and AR gene expression), and metabolic-immune enriched (MIE, enriched in adipocytes, macrophages, endothelial cells and metabolic-related pathways). In METABRIC, P linked to higher tumor grade (p=0.0187) and more copy number aberrations (p=0.0159). In SCAN-B, P and NSE were associated with larger and smaller tumors, respectively (p<0.001). P was also linked to higher tumor grade (p<0.001), lymph node involvement (p=0.02), and high Ki67 (p<0.001). Univariate analysis in METABRIC showed NSE and P subtypes associated with good and poor prognosis, respectively, for relapse-free interval (RFI) (HR=0.56, p=0.027; HR=1.8, p=0.019). Multivariable analysis confirmed NSE's good prognosis (HR=0.47, p=0.03). In SCAN-B, NSE was linked to longer RFI (HR=0.42, p=0.0018) and P to shorter RFI (HR=2.2, p=0.0014). At the multivariate, NSE remained associated with longer RFI (HR=0.54, p=0.035).

Conclusions

Spatial transcriptomics revealed four ILC subtypes reflecting TME heterogeneity, which were validated in external datasets and linked to distinct clinical outcomes, enhancing prognosis accuracy in ILC.

Editorial acknowledgement

Clinical trial identification

Legal entity responsible for the study

The authors.

Funding

FNRS, Télévie, Association Jules Bordet.

Disclosure

F.P. Duhoux: Financial Interests, Institutional, Advisory Board: Roche, Pfizer, AstraZeneca, Lilly, Novartis, Amgen, Daiichi Sankyo, Pierre Fabre, Gilead, Seagen, MSD; Financial Interests, Institutional, Invited Speaker: Novartis, Pfizer, MSD, Roche, MSD, Boehringer Ingelheim, Pfizer, Novartis, Lilly, AbbVie, Seagen, Gilead, AstraZeneca, Menarini, Immutep; Financial Interests, Institutional, Expert Testimony: Seagen, Novartis, MSD. C. Sotiriou: Financial Interests, Institutional, Advisory Board: Astellas, Vertex, Seattle Genetics, Amgen, INC, Merck & Co; Financial Interests, Personal, Advisory Board: Cepheid, Puma; Financial Interests, Personal, Invited Speaker: Eisai, Prime Oncology, Teva; Financial Interests, Institutional, Other, Travel: Roche; Financial Interests, Institutional, Other, Internal speaker: Genentech; Financial Interests, Personal, Other, Regional speaker: Pfizer; Financial Interests, Institutional, Invited Speaker: Exact Sciences; Financial Interests, Personal, Advisory Board, Stock options: Signatur Biosciences. All other authors have declared no conflicts of interest.