(P 275) GIST-BSF1: A NOVEL SCF-DEPENDENT, WILD-TYPE KIT CELL LINE PLATFORM THAT ALLOWS RAPID GENERATION OF ISOGENIC SUBLINES WITH BOTH IMATINIB-SENSITIVE AND IMATINIB-RESISTANT GENOTYPES

Objective: Approximately 80% of Gastrointestinal Stromal Tumors (GISTs) are driven by activating mutations in the proto-oncogene KIT. These mutations result in constitutive activation of KIT, independent of the natural KIT ligand Stem Cell Factor (SCF). Most patients respond well to KIT inhibitor imatinib mesylate (IM) but the majority eventually progress due to secondary mutations in KIT. Despite a broad range of primary and secondary KIT and PDGFRA mutations observed in patients, very few primary cell lines are available for preclinical research. Here, we introduce an isogenic cell line platform allowing rapid generation of primary and consecutive mutations in KIT in a GIST cellular background.

Methods: CRISPR/Cas9-mediated gene editing was used to repair the primary KIT mutation in the well-established GIST cell line GIST-T1. Limiting dilutions fed with SCF were performed to achieve a monoclonal population with only wild-type (WT) KIT, "GIST-BSF1". KIT mutations were reintroduced to GIST-BSF1 by CRISPR/Cas9 or plasmid expression. Effects of treatment with SCF and tyrosine kinase inhibitors (TKIs) were evaluated using sulforhodamine B (SRB) cell proliferation assays and western blot.

Results: GIST-BSF1 cells show dose-dependent growth upon SCF treatment. KIT-WT signaling is stimulated by SCF and effectively abrogated by sunitinib, a known inhibitor of KIT-WT. In comparison to parental GIST-T1, GIST-BSF1 treated with SCF displayed 4.3-29.3-fold higher GR50-values for all tested KIT inhibitors, except for sunitinib (1.4-fold) and nintedanib (3.1-fold). The introduction of KIT mutations either by gene editing or plasmid led to SCF-independent growth. To date we have established a panel of 20 cell lines harboring different primary, secondary and tertiary KIT mutations. Evaluation of a panel of approved and non-approved KIT inhibitors showed inhibitory profiles expected for the respective mutations. GIST-BSF1 sublines with secondary mutations in-cis in both the ATP binding pocket (AP) and activation loop (AL) of KIT, called AP/AL mutations, display the same inhibitory profile towards ripretinib as seen in AP/AL mutations previously modelled in GIST-T1 (GR50 values for AP/AL mutants >20-fold higher than corresponding AL mutants). Parallel gene knock-out of NF1 and PTEN as well as introduction of activating mutations in PIK3CA result in SCF-independent cell growth and confer resistance to TKIs targeting KIT.

Conclusion: GIST-BSF1 is the first KIT-WT, SCF-dependent GIST cell line, which allows study of the impact of KIT inhibitors on WT KIT. Reintroduction of KIT mutations leads to SCF-independent growth and yielded anticipated TKI profiles. This cell line platform allows the rapid validation of novel variants of KIT and other oncogenic kinases in a GIST-specific cellular background.