Precision neoadjuvant treatment with AI-assisted subtyping in HR+/HER2-breast cancer: A randomized, phase 2 trial FASCINATE-N.

In this phase 2 trial (NCT05582499), patients with stage II-III hormone-receptor-positive/human epidermal growth factor receptor 2-negative (HR+/HER2-) breast cancer were categorized into endocrine-based and targeted-based groups based on previously proposed similarity network fusion (SNF) subtyping by digital pathology classification and were then randomly assigned in a 1:1 ratio to receive precision or control treatment. The primary endpoint was pathological complete response (pCR). Among 251 randomized patients, 49 were classified into an endocrine-based group and 202 into a targeted-based group. Patients receiving precision treatment had a significantly higher pCR rate (11.1% [90% confidence interval (CI), 6.8-16.8] vs. 4.0% [90% CI, 1.6-8.2]; p = 0.033), with the benefit mainly derived from the targeted-based group (13.9% in precision vs. 4.0% in control; p = 0.026). Toxicity was manageable. Our findings highlight that guiding neoadjuvant precision treatment by SNF subtyping in patients with HR+/HER2- breast cancer showed potential clinical benefits, with improvement of pCR in the targeted-based group and better tolerance in the endocrine-based group.