Multiple protease-activated probody-drug conjugates for treating CD147-positive ovarian cancer with limited toxicity

Bo Wang & Sihe Zhang et al. · 2026-01-29

Traditional antibody-drug conjugates (ADCs) that target antigens expressed not only on tumor cells but also on nonmalignant cells are often associated with unavoidable on-target off-tumor toxicities. Probodies are masked antibody prodrugs that remain inactive until proteolytically activated in the tumor microenvironment (TME). However, most probodies are produced on the basis of a monoresponsive design and achieve a narrow therapeutic index owing to tumor heterogeneity and nonspecific payload-conjugation. Here, we generated different probodies targeting the cluster of differentiation 147 (CD147) antigen based on the design of multiple-protease-activated linker peptide and HcHAb18 antibody epitope-derived masking peptides. Three anti-CD147 probody-drug conjugates (PDCs) were produced via site-specific conjugation with cytotoxic monomethyl auristatin E (MMAE) through mild cysteine-selective chemical reactions. The created probodies and PDCs can be activated through cleavage by the proteases legumain, matrix-metalloproteinases 9, and urokinase-type plasminogen activator, but exhibit different CD147-targeting potentials. Importantly, PDC1, one of the conditional antibody architectures, exhibits highly selective targeting and strongest cytotoxicity to ovarian cancer cells. More importantly, PDC1 demonstrated promising targeting selectivity and improved the tumor-inhibition efficiency in ovarian cancer-xenograft mouse models without systemic toxicity. This multiple protease-activated, disulfide-bridging PDC strategy provides a novel, precise and safe ADC-targeted therapeutics against ovarian cancer.
TL;DR

Importantly, PDC1, one of the conditional antibody architectures, exhibits highly selective targeting and strongest cytotoxicity to ovarian cancer cells, and PDC1 demonstrated promising targeting selectivity and improved the tumor-inhibition efficiency in ovarian cancer-xenograft mouse models without systemic toxicity.

AI-generated by Semantic Scholar

Authors
Bo Wang, Qiangzhe Zhang, Yuqing Yang, Chenhui Wang, Guiyu Deng, Ying Chen, Zichang Xu, Zhinan Chen, Chuanzheng Zhou, Sihe Zhang
Funding

National Natural Science Foundation of China

82373028