Abstract A013: Preclinical Development of a Humanized CD44v9-Directed Antibody-Drug Conjugate for Treatment-Refractory Breast Cancer
Antibody-drug conjugates (ADCs) have transformed the treatment of advanced breast cancer, yet resistance to HER2- and TROP2-directed ADCs is an increasing clinical challenge. New ADC targets should be linked to aggressive disease, retained after current ADC exposure, and sufficiently tumor-selective for therapeutic development. CD44 variant isoform 9 (CD44v9) is an attractive next-generation ADC target because it marks stem-like, invasive, and treatment-refractory breast cancer phenotypes while showing limited expression in most normal tissues. In prior work, we identified CD44v9 as a tumor-selective and therapeutically accessible surface antigen in aggressive breast cancer and demonstrated proof-of-concept efficacy with a chimeric anti-CD44v9-MMAF conjugate. To advance translation, we generated fully humanized anti-CD44v9 antibody variants. CD44v9 expression was assessed in human breast cancer cell lines, patient-derived xenografts, tumor specimens, and normal tissues. Two fully humanized anti-CD44v9 variants, SUM24.2 and SUM24.3, were generated using CUMAb, a structural and energy-focused computational humanization tool, and veneering, a surface humanization approach , respectively. The parental chimeric antibody SUM24.1 and humanized variants were conjugated to monomethyl auristatin F (MMAF). Binding, internalization, and cytotoxicity were evaluated in triple-negative, inflammatory, and ADC-resistant breast cancer models. Antitumor activity was assessed in refractory in vivo models. Humanized anti-CD44v9 variants SUM24.2 and SUM24.3 retained selective recognition of membrane-bound CD44v9 and supported efficient target-dependent internalization (p<0.01). SUM24.2-MMAF induced antigen-dependent cytotoxicity in CD44v9-positive breast cancer cells, whereas no cell-killing effect was observed in antigen-negative controls or human peripheral blood mononuclear cells. CD44v9 expression persisted in models resistant to FDA-approved ADCs, including trastuzumab deruxtecan- and sacituzumab govitecan-resistant models, and CD44v9-directed ADC therapy remained active in these settings, with retained selectivity, supporting a strategy mechanistically distinct from HER2- and TROP2-directed pathways. These results reproduced the activity observed in earlier chimeric proof-of-concept studies. CD44v9 is a compelling ADC target for treatment-refractory breast cancer because it is associated with aggressive, stem-like, and therapy-resistant disease biology and remains targetable in models resistant to current breast cancer ADCs. Humanization preserved target recognition, internalization, and ADC activity, representing a key translational advance beyond chimeric proof-of-concept studies. These findings support further preclinical development of a humanized CD44v9-directed ADC for advanced breast cancer after resistance to existing ADCs. Naoto T. Ueno, Dileep R. Reddy, McKenna E. Flynn, Yasuaki Anami, Zhaoxuan Yang, Minji Seo, Kyoji Tsuchikama, Jennifer A. Maynard, Jangsoon Lee. Preclinical Development of a Humanized CD44v9-Directed Antibody-Drug Conjugate for Treatment-Refractory Breast Cancer [abstract]. In: Proceedings of AACR Drug Discovery and Development (AACR D3) Conference; 2026 Jul 21-24; Boston, MA. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(14_Suppl):Abstract nr A013.