AI and pathology reveal higher late recurrence risk in invasive lobular breast cancer

Dr Roberto Salgado speaks to ecancer about clinical outcomes of invasive lobular carcinoma (ILC) versus non-lobular breast cancer (NLC) assessed by expert pathologists, an artificial intelligence (AI) CDH1 classifier, and AI-derived tumour microenvironment (TME) biomarkers in TAILORx.

He says that in this analysis ILC was consistently associated with higher late recurrence and worse survival compared with non-lobular breast cancer, whether identified by expert pathology review or an AI-based CDH1 classifier.

While early outcomes were similar, ILC showed a significantly increased risk between years 5–15, with a nearly 5% overall survival difference at 15 years.

Both manual TIL scoring and a novel AI-derived tumour microenvironment risk score independently stratified recurrence risk beyond standard clinicopathologic factors and the 21-gene recurrence score.

Dr Salgado concludes by saying that these findings highlight the value of AI-driven pathology and TME analysis for long-term risk assessment and support consideration of extended endocrine therapy in patients with ER+/HER2−, node-negative ILC, even when genomic risk is low.