"Harnessing the Anthrax Toxin Protein Delivery System as a Platform for Targeted Tumor Therapy", Department of Environmental and Occupational Health, School of Public Health. 

Committee: 

• Shihui Liu, (dissertation advisor) Department of Medicine
• Aaron Barchowsky, EOH
• Toren Finkel, Department of Medicine
• Patricia L. Opresko, EOH
• Berthony Deslouches, EOH

Abstract: 

The increasing global burden of cancer underscores the need for safer and more effective targeted therapies. The advent of the anti-HER2 antibody-drug conjugate trastuzumab deruxtecan has expanded HER2-targeted therapy beyond HER2-overexpressing breast cancer to include all other HER2-high and several HER2-low malignancies. These findings support an emerging role for HER2 as a pan-cancer biomarker for targeted therapy. However, current HER2-directed agents are frequently associated with significant on-target and off-target toxicities.
Mutations in the RAS family (KRAS, NRAS, and HRAS) drive 20-33% of all human cancers. However, directly targeting RAS has long faced both intrinsic and extrinsic challenges, primarily due to a lack of binding pockets, its high affinity for GTP/GDP, mutational heterogeneity (e.g., G12C, G13D, and Q61), and rapid adaptive resistance. 
To address these unmet medical needs, we developed a unique platform for cancer treatment by reengineering the anthrax lethal toxin (ALT) delivery system. ALT consists of a delivery vehicle, PA, and an enzymatic effector, LF. Delivery of LF into cells strictly depends on PA activation, which requires both receptor binding and subsequent cleavage by furin proteases on the cell surface. Here, we constructed a HER2-directed PA variant, PA3MzHER2-MMP, which relies on tumor-overexpressed matrix metalloproteinases (MMPs) for activation rather than the ubiquitously expressed furin. When combined with an LF-derived protein synthesis inhibitor, this dual-targeting system was able to target both HER2-high and HER2-low cancers with high tumor specificity and effectively led to the complete regression of most low-HER2 xenograft tumors. In vitro testing showed our toxin’s superior cytotoxicty compared to existing anti-HER2 agents. We then introduced a novel LF-derived effector protein LFnDUF5-L418W, a pan-RAS-specific endopeptidase, which exhibited potent proteolytic activity against RAS proteins when delivered via PA3MzHER2-MMP. The combination of PA3MzHER2-MMP and LFnDUF5-L418W had a high therapeutic index and demonstrated robust antitumor efficacy across a diverse range of xenograft tumors harboring oncogenic RAS mutations. Our system achieved antitumor activity comparable to, or even exceeding, that of current small-molecule RAS inhibitors. These studies establish our platform as a unique, safe, and potent antitumor agent effective for a range of solid tumors and poised for further clinical development.