Targeting Refractory EGFR-Driven Tumors By Induction Of Dominant-Negative EGFR Splicing Variants

ABSTRACT We recently identified a number of novel secreted soluble decoy RTK isoforms (sdRTKs), generated by an alternative splicing/intronic polyadenylation (IPA) mechanism, which can act as potent natural inhibitors of aberrant RTK signaling, a key driving aspect of a large fraction of cancers, including lung cancer. Further, we developed an antisense-based method to effectively induce expression of the inhibitory soluble decoy RTKs in vitro and in vivo. These compounds represent a new class of drugs that carry the advantage of simultaneously knocking down the pathological targets while introducing natural dominant-negative isoforms, thus greatly increasing efficacy. We propose to apply this antisense-based splicing redirection strategy to induce endogenous, soluble decoy EGFR inhibitory variants (sdEGFR) in EGFR-dependent tumors. sdEGFR variants, which are normally expressed at low levels, encode the extracellular ligand-binding domain but lack the intracellular kinase domain. We hypothesize that sdEGFR-inducing compounds inhibit EGFR signaling by a three-fold mechanism: A) removal of full-length EGFR receptor, B) sequestration of EGF and other ligands, and C) neutralizing hetero- dimerization with residual receptors (EGFR, HER2/3, etc.). We further posit that, as the resulting products are secreted, they affect signaling in both targeted and bystander cells alike and interfere with paracrine and autocrine loops, thus exerting an amplified effect. Since its mechanism of action is different from that of current therapies, this strategy should be effective in all refractory tumors where the mechanisms of resistance still relies on EGFR activity, both when such activity is being driven by wild-type EGFR amplification or EGFR-activating mutations. To this end, we will pursue the following specific aims: 1) Characterization of natural inhibitory sdEGFR splicing variants. 2) Optimization of splicing redirection compounds targeted to EGFR. 3) Modulation of endogenous EGFR splicing and polyadenylation in treatment-resistant tumor models. We propose to use non-small cell lung cancer (NSCLC) as our initial cancer model. NSCLC is the most frequent cause of cancer fatalities, with over 130,000 deaths per year, a quarter of which are associated to, and depend on, aberrant EGFR signaling. Notwithstanding the initial efficacy of targeted drugs, such as third-generation tyrosine kinase inhibitor Osimertinib, against most activated EGFR, resistance typically emerges within 2 years. Thus, the development of additional, alternative approaches to treat refractory EGFR-dependent cancers remains an unfulfilled need. Importantly, whereas we propose to use NSCLC as our main model system both in vitro and in vivo, our findings will also directly apply to any cancer where EGFR plays an important role and is frequently mutated or amplified, such as glioblastoma, colorectal cancer and others. Furthermore, analogous compounds can be simply designed against other RTKs or other targets responsible for various mechanisms of drug-resistance and pathological signaling, thus adding a powerful new set of tools to understand such phenomena, and to inform a multitude of therapeutic leads.