Abstract B122: Activity of the novel KIF18A Inhibitor, ATX-295, is enriched in whole genome doubled ovarian and TNBC preclinical models

Abstract Background: KIF18A is a plus-end-directed kinesin known to play a role in mitosis by facilitating chromosome alignment and spindle microtubule dynamics. KIF18A loss inhibits proliferation in TP53 mutant CIN cells; those with ongoing segregation defects, like aneuploid or whole genome doubled cells, are especially vulnerable to mitotic disruption. These findings definitively establish KIF18A as a high-potential precision oncology target, opening new avenues for therapeutic intervention. Materials and Methods: Oncology indications were identified via a large, multiplexed PRISM screen using a selective KIF18A inhibitor (compound 1). Responses in ovarian and triple negative breast cancer (TNBC) models were recapitulated in vitro and in vivo with ATX-295, a clinical-stage KIF18A inhibitor. Mechanistic studies were conducted in selected cell lines. Genomic CIN correlates, fraction genome altered (FGA), aneuploidy score (AS), and whole genome doubling (WGD) were evaluated as predictors of ATX-295 sensitivity using an internal bioinformatics pipeline. Feasibility of AI-based detection was assessed using TCGA H&E slides via Imagene AI. Results: A PRISM screen using KIF18A inhibitor compound 1 showed 18% of 826 cell lines depend on KIF18A for proliferation, supporting its role as a precision oncology target. Sensitivity was observed across tumor types, particularly ovarian and breast cancers with high CIN. Accent Therapeutics identified ATX-295, a selective KIF18A ATPase inhibitor (IC50 = 16 nM), which suppressed proliferation in high-grade serous ovarian cancer (HGSOC) and TNBC models. Among CIN-associated biomarkers, WGD was the strongest predictor of sensitivity. In WGD(+) OVCAR-3 xenografts, ATX-295 induced dose-dependent tumor regression and mitotic arrest (pHH3+), while WGD(–) OVK18 tumors showed no effect. In ovarian patient-derived xenograft (PDX) models, 61% responded to ATX-295, with 73% of responders being WGD(+), including all tumors with regression. These findings support WGD as a predictive biomarker for ATX-295 response. To assess clinical feasibility, an H&E-based WGD detection model was developed using Imagene AI’s OI Suite and TCGA-labeled samples. The model achieved >0.7 AUC on a held-out test set within minutes, demonstrating proof-of-concept for AI-based WGD detection in clinical samples. Conclusions: Together, these results demonstrate that inhibition of KIF18A via ATX-295 is a compelling strategy in chromosomally instable, whole genome doubled tumors such as HGSOC and TNBC and identify WGD as a potential enrichment biomarker of ATX-295 sensitivity in these indications. Furthermore, AI-based detection of WGD from patient H&E-stained tissue slides demonstrates promise as a stratification tool for identifying patients most likely to respond to ATX-295 therapy. ATX-295 is currently being evaluated in a first-in-human, Phase 1/2 open-label study assessing safety, pharmacokinetics, pharmacodynamics, and preliminary antitumor activity in patients with advanced solid tumors and ovarian cancer (NCT06799065). Citation Format: Laura Ghisolfi, Maureen M. Lynes, April Greene-Colozzi, Livia Shehaj, Brian A. Sparling, Matthew G. Rees, Melissa M. Ronan, Jennifer A. Roth, Inbal Gazy, Stuart Ince, Jason A. Sager, Serena J. Silver. Activity of the novel KIF18A Inhibitor, ATX-295, is enriched in whole genome doubled ovarian and TNBC preclinical models [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics; 2025 Oct 22-26; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2025;24(10 Suppl):Abstract nr B122.