EACR25-1298

Assessment of two new ROS1+ NSCLC patient-derived cell lines as in vitro models for TKI resistance studies

C. Dijkhuizen1, P. Korthuis1, C. Aguado2, S. Garcia-Román2, A. Aguilar-Hernández3, L. Drayer4, M. de Looff1, F. Milder1, A. van der Wekken4, A. van den Berg5
1Hanze University of Applied Sciences, Institute For Life Science And Technology, Research Center Biobased Economy, Groningen, Netherlands
2Dexeus University Hospital, Pangaea Oncology, Barcelona, Spain
3Quirón Dexeus University Hospital, Dr Rosell Oncology Institute, Barcelona, Spain
4University Medical Center Groningen, Department of Pulmonology, Groningen, Netherlands
5University Medical Center Groningen, Department of Medical Biology and Pathology, Groningen, Netherlands
Introduction:

ROS1 fusion gene-positive (ROS1+) is found in 1-2% of the non-small cell lung cancer (NSCLC) patients. Initially these patients are treated effectively with tyrosine kinase inhibitors (TKIs), however patients generally develop resistance to TKI treatment due to on- and off-target resistance mutations. In this study, we assessed TKI sensitivity of two new patient-derived ROS1+ cell lines as in vitro models for investigating the resistance mechanisms to ROS1 TKI treatment.

Material and method:

Two cell lines were generated from pleural effusions obtained from CD74-ROS1+ NSCLC patients. The cell line from patient 1 was derived from cells obtained upon resistance to crizotinib (PC1 cells). The cell line from patient 2 was generated from cells obtained upon resistance to crizotinib followed by resistance to lorlatinib treatment (PC2 cells). The Archer lung fusion cancer and TSO500 panels were used to identify potential resistance mechanisms in these cell lines. To assess efficacy of TKI’s in the patient derived cell lines, IC50 values were compared to CD74-ROS1 Ba/F3 cells. Sensitivity to TKIs crizotinib, entrectinib, lorlatinib, repotrectinib and zidesamtinib was assessed using MTS assays.

Result and discussion:

Presence of the CD74-ROS1 fusion gene was confirmed in all three cell lines. The IC50 values of unmutated ROS1+ Ba/F3 cells for crizotinib, entrectinib, lorlatinib, repotrectinib and zidesamtinib were 31.4; 44.6; 14.3; 14.9 and 29.9 nM, respectively. PC1 cells had a TP53 p.M246I mutation and IC50 values above 3000 nM for all five ROS1 inhibitors. Interestingly, these cells were resistant to lorlatinib, while the patient is – after being treated with lorlatinib+cis-platinum+pemetrexed, now currently successfully treated with lorlatinib mono-therapy. PC2 cells had a ROS1 p.G2032R mutation, along with mutations in PIK3CA, PIK3CG and ARID1A. These cells responded to repotrectinib (IC50 57 nM) and lorlatinib (IC50 556 nM), while response to zidesamtinib, crizotinib and entrectinib was poor with IC50 values above 1500 nM. The p.G2032R mutation is a known resistance mutation for crizotinib and lorlatinib, but was shown be responsive to repotrectinib and the ROS1 specific zidesamtinib. Thus, the results from the PC2 cell line indicate presence of other off-target resistance mechanism that lead to resistance to zidesamtinib.

Conclusion:

In this study we showed that PC1 cells generated from a crizotinib-resistant patient were resistant to all ROS1 inhibitors tested. The G2032R mutated PC2 cells generated from a crizotinib and lorlatinib resistant patient were sensitive to repotrectinib, but not to zidesamtinib. This PC2 cell line can be used to assess off-target resistance mechanism to zidesamtinib.