Binding-Induced Enzyme Activation for Prodrug Therapy of HER2 Variants

Strikingly cancer patients with a variant form of HER2 have a median progression free survival (PFS) of 4.9months a stark contrast to generic reports that HER2 targeted therapy gives 90% of patients a PFS of 5+ years.Most resistance is due to HER2 variants structural mutants that prevent drug binding or drastically decreasethe internalization of HER2. Accordingly most standard and emerging treatments fail for these patients. Thereis an immediate need for therapeutic approaches that can be specifically designed to treat HER2 variants.Antibody-directed enzyme prodrug therapy (ADEPT) a strategy of conjugating prodrug-activating enzymes toantibodies that target cell surface receptors nearly reached clinical success. It was limited in large part bypremature prodrug activation in circulation. We propose to re-engineer the ADEPT strategy addressing previousshortcomings to tackle the pressing clinical need resulting from HER2 variants. In our strategyComplementation Dependent Enzyme Prodrug Therapy (CoDEPT) split-enzyme fragments are fused to twoseparate antibodies against HER2. The binding of these antibodies to HER2 at distinct epitopes brings the split-enzyme fragments into proximity facilitating reconstitution of the active enzyme. The enzyme subsequentlyactivates prodrug at the tumor and should thereby drastically reduce off-site activation and related toxicity.We hypothesize that CoDEPT constructs can reduce tumor growth in a murine breast cancer model. In Aim 1we will engineer and characterize a library of split--lac constructs targeting HER2 variants and select leadCoDEPT constructs. In Aim 2 we will determine kinetic parameters and concentration thresholds that willestablish foundational requirements for CoDEPT prodrug delivery and activation. In Aim 3 we will evaluatetherapeutic efficacy of select CoDEPT constructs in vitro and in vivo.