1-2% of non-small cell lung cancer (NSCLC) is driven by constitutive activation of ROS1 due to a ROS1 rearrangement. ROS1+ NSCLC is treated with tyrosine kinase inhibitors (TKI) which inhibit the phosphorylation of ROS1 and as a consequence its downstream pathways. Treatment with TKIs is palliative and will eventually lead to resistance via accumulation of additional genomic aberrations in either the tyrosine kinase domain of ROS1 (on-target) or other genes leading to alternative survival routes (off-target). In this study we aim to generate a repotrectinib resistant cell line by long term treatment with TKIs to study treatment induced resistance mechanisms.

Ba/F3 cells were transfected with a pCDNA3.4 plasmid containing the SLC34A2-ROS1 fusion gene. Sensitivity for repotrectinib was measured by Western blot for phosphorylated ROS1, MTS assays and calculation of the IC50 value. To generate resistant cells, we treated the Ba/F3 SLC34A2-ROS1 cells with an increasing dose of repotrectinib during the course of 4 months until growth at a 100-fold increase of the IC50 value was observed. MTS assays were performed on the resistant cells cultured either one or two weeks with and without repotrectinib before conducting the MTS assay.

IC50 value of the parental Ba/F3 SCL34A2-ROS1 treated with repotrectinib was 0,15nM. Effective treatment was confirmed by western blot for phospho-ROS1 showing a strongly reduced phosphorylation between 0,1 and 1nM of repotrectinib. Resistance cell cultures were made using 5, 10 and a final concentration of 15nM repotrectinib. IC50 values of the resistance levels cultured for one week without repotrectinib pressure were 3,56, 3,69 and 5,16nM for the 5, 10 and 15nM respectively. IC50 values of cells continuing to grow with repotrectinib were higher with 9,32 and 11,07nM for the 10 and 15nM repotrectinib cultures. For the 5nM culture the IC50 was as high as 47nM, this is possibly due to hormesis. These data show that removal of repotrectinib from the medium might make the cell lines more sensitive to repotrectinib again. Cells cultured with repotrectinib were smaller in size compared to the parental cells. When repotrectinib pressure was removed cells regained similar size as the parental cells but did however clump together which is not observed in the parental cell line.

We generated three Ba/F3 SLC34A2-ROS1 repotrectinib resistant cell lines. A short term repotrectinib withdrawal makes the resistant cells more sensitive to repotrectinib. Mechanisms underlying the observed resistance such as mutations in the tyrosine kinase domain or in genes affecting alternative signaling pathways are currently being examined.