Deciphering the structural determinants of response and resistance to HER2-targeted antibody-drug conjugates and tyrosine kinase inhibitors for HER2-mutant Non-Small Cell Lung Cancer - Project Summary HER2 activating mutations occur in approximately 3% of NSCLC patients and more than 20 other cancer types, including breast and colon, comprising >10,000 patients annually in the U.S. More than 80 different recurrent activating mutations are observed, and are distinct from HER2 amplifications. There are currently no FDA- approved tyrosine kinase inhibitors (TKIs) for HER2 mutant cancers, and currently the only approved HER2- targeted treatment for lung cancer is the antibody drug conjugate (ADC) trastuzumab deruxtecan (T-DXd). There are two major obstacles limiting the effective use of HER2 targeted therapies for these patients. First, the different mutations vary widely in their sensitivity to HER2 inhibitors, and there is no validated approach for selecting the right HER2 inhibitor for a given mutation. Currently, trials typically select patients based on the exon in which the mutation occurs (exon-based classification, e.g. exon 20-mutant NSCLC). These criteria, however, are not based on biological evidence and we recently reported that even mutations within the same exon can vary widely in terms of their impact on kinase structure and drug sensitivity. Second, tumors may acquire resistance either through additional HER2 mutations (HER2-dependent resistance), or via HER2-independent mechanisms (e.g. resistance to the ADC chemotherapy component such as deruxtecan, so called “payload resistance”). These mechanisms are not yet well characterized, and we do not yet have a rational basis for selecting the most appropriate drug for treating resistant tumors. In a recent publication in Nature, we reported that for a related gene, EGFR, classifying mutations based on how they impact the 3-dimensional structure and drug response (a structure/function-based system) more accurately predicted drug response than a standard exon-based classification. We hypothesize that i) by using a similar structure/function-based approach, we could more accurately match HER2 mutations with effective therapies than a standard exon-based approach, and ii) by investigating HER2-dependent and -independent mechanisms of resistance, we can develop rational strategies for targeting refractory tumors. To investigate these hypotheses, in Aim 1 we will comprehensively characterize the landscape of HER2 mutations associated with response or resistance (both primary and acquired) to TKIs using established preclinical models, including the LentiMutate scanning mutagenesis, and data from clinical trials led by the MPI Dr. Heymach and others (see attached letters of support); in Aim 2 we will similarly characterize the landscape of response and resistance for HER2-ADCs, and determine to what extent resistance is HER2- vs payload-dependent. Finally, in Aim 3, we will develop a structure/function-based classification for HER2 TKIs and ADCs and test strategies for targeting treatment-resistant tumors. The project brings together a world-class team of investigators with expertise in HER2 biology, molecular modeling, and clinical trials for the overarching goal of developing more tailored and effective approaches for HER2-mutant cancers.
