Genomic mechanisms of RET inhibitor sensitivity and resistance in RET-fusion positive cancer cells

e21029 Background: RET fusions occur in 2-3% of lung cancer cases, and there are currently two FDA-approved RET-specific inhibitors (RETi), selpercatinib and pralsetinib. RET fusions are heterogeneous and the impact of specific fusion partners and their unique breakpoints on drug sensitivity is poorly understood. Moreover, while patients may be initially sensitive to RETi treatment, therapeutic resistance eventually emerges, and studies have shown that RET-dependent mechanisms of resistance include acquired resistance mutations that alter drug binding. We sought to investigate mechanisms of sensitivity and resistance to RET inhibitors in RET-fusion positive lung cancer in effort to improve treatment options and clinical outcomes in this patient population. Methods: We characterized the landscape of RET-fusions and -mutations in non-small cell lung cancer (NSCLC). KIF5B-RET, CCDC6-RET, and NCOA4-RET were the most frequently observed RET fusions in NSCLC. To investigate the drug sensitivity profile of individual RET fusion variants, we generated a panel of Ba/F3 cells expressing various RET fusion variants and evaluated in vitro sensitivity to RETi including selpercatinib, pralsetinib, vandetinib, cabozantinib, TPX-0046, lovatinib, and ponatinib. We observed that unique RET fusions impart differential sensitivities to RET inhibitors. For example, cells expressing KIF-RET (K15, R12) were highly sensitive to selpercatinib, pralsetinib, and ponatinib, whereas cells expressing KIF-RET (K22, R12) were likewise sensitive to selpercatinib and pralsetinib but were less sensitive to ponatinib. We next established a cell line (MDA-L-30) from a patient harboring a RET-fusion (KIF-RET; K15, R12) and observed a RETi sensitivity profile similar to that observed using the BaF3 system. Results: Next, we employed the LentiMutate approach to rapidly identify genomic mechanisms of RETi resistance. In cells expressing the KIF5B-RET (K15, R12) fusion, the G810S mutation was associated with acquired resistance to both selpercatinib and pralsetinib, consistent with prior reports and clinical observations. However, V804M/E and M918T mutations were associated with acquired resistance to selpercatinib but not pralsetinib, and L730I/V and E732K mutations were associated with resistance to pralsetinib but not selpercatinib. Conclusions: Taken together, these data indicate that individual RET fusion variants with diverse fusion partner breakpoints have distinct drug sensitivity profiles and that secondary resistance mutations may be non-overlapping between RETi. These findings support the need to comprehensively characterize RET-dependent mechanisms of resistance in order to provide therapeutic recommendations for treating RET-fusion driven cancers.