Efficacy of Crizotinib After Entrectinib Resistance Due to MET Polysomy in ROS1-Rearranged NSCLC: A Case Report

Resistance to ROS1 tyrosine kinase inhibitors is inevitable, but it has been unclear whether crizotinib might be effective after the development of entrectinib resistance. We here present a case of ROS1-rearranged NSCLC that responded to crizotinib after tumor progression due to MET polysomy during entrectinib treatment. This case suggests that crizotinib is an effective option for patients with MET polysomy, even after disease progression on entrectinib.


Introduction
Tumors with ROS1 rearrangement account for approximately 1% to 2% of NSCLC cases, 1 and entrectinib and crizotinib have markedly improved outcomes for patients with ROS1-rearranged NSCLC. Nevertheless, resistance to entrectinib inevitably develops. The mechanisms of such resistance remain unclear in most cases, and it is not known whether crizotinib is effective after the development of entrectinib resistance. We now report a case of resistance to entrectinib that was characterized by next-generation sequencing (NGS) analysis and overcome by crizotinib.

Case Presentation
A 64-year-old woman who had never smoked presented with dyspnea. A computed tomography (CT) scan revealed a 29-mm nodule in the right upper lobe of the lung, enlargement of the right hilar lymph node and bilateral supraclavicular lymph nodes, and pleural effusion. She was diagnosed with having stage IV lung adenocarcinoma accompanied by malignant pleural effusion. She received first-line treatment with carboplatin plus pemetrexed for four cycles, resulting in a confirmed partial response as the best response ( Fig. 1). After 11 months of chemotherapy, a CT scan revealed tumor progression in the right lung. The patient received second-line treatment with docetaxel for four cycles and then again manifested disease progression. A biopsy performed by bronchoscopy for gene testing revealed neither EGFR mutation nor ALK translocation. She received nivolumab as a third-line treatment and achieved a partial response. After 25 months, a CT scan identified tumor progression in the right lung. S-1 was then administered as a fourth-line treatment for two cycles, after which disease progression was again apparent. A bronchoscopic biopsy from the upper lobe of the right lung was again performed, and the tissue was subjected to NGS (Archer FusionPlex CTL panel), which revealed the presence of an SLC34A2-ROS1 fusion gene.
The patient received entrectinib (600 mg/d) as a fifth-line treatment, and she achieved a partial response (À30%) in the upper lobe of the right lung at 3 months (Fig. 2). The entrectinib dose was reduced to 400 mg/ d as a result of the development of peripheral edema at 9 months. After 34 months of treatment, a CT scan revealed tumor progression in the upper lobe of the right lung. She therefore received crizotinib as a sixth-line treatment, and a CT scan revealed a partial response (À30%) in the upper lobe of the right lung at 2 months.
We performed NGS analysis of tumor specimens obtained both before and after entrectinib treatment to evaluate the mechanism of crizotinib action. This analysis revealed a MET gene copy number of 3 in both specimens, but neither secondary mutations in the ROS1 kinase domain nor activation of bypass signaling pathways was detected. Fluorescence in situ hybridization (FISH) analysis of MET amplification revealed MET polysomy with probes for MET and CEP7. The pretreatment specimen revealed 5.8 signals per cell for MET and indicated a MET-to-CEP7 ratio of 1.1, whereas the posttreatment specimen revealed 6.1 signals per cell for MET and indicated a MET-to-CEP7 ratio of 1.1 (Fig. 3).

Discussion
As far as we are aware, this is the first report to explore the mechanism of a response to crizotinib in a patient with ROS1-rearranged NSCLC after tumor progression during treatment with entrectinib. Indeed, only one case of a crizotinib response after disease progression during entrectinib treatment has been described. 2 KRAS G12C mutation, KRAS amplification, FGF3 amplification, MET amplification, and the ROS1 mutations G2032R and F2004C/I have been identified as mechanisms of resistance to entrecinib. [3][4][5][6] Our report suggests that crizotinib could be an effective option for patients with MET amplification or polysomy, even after disease progression during entrectinib treatment.
In the present case, although the MET gene copy number determined by NGS analysis was 3, FISH analysis revealed a MET signal per cell value of approximately 6. An increase in MET copy number has been classified on the basis of the MET to centromere ratio as polysomy (MET-to-centromere ratio of <1.8) or low (>1.8 to <2.2), intermediate (>2.2 to <5.0), or high (>5.0) amplification. 7 In the FISH assay for the present patient, signals from 50 nonoverlapping nuclei in representative fields were enumerated for MET and CEP7 copy numbers. The analysis revealed MET polysomy on the basis of a copy number gain (5.8-6.1 signals per cell) and low ratio of MET to CEP7 (1.1). Both entrectinib and crizotinib are tyrosine kinase inhibitors for ALK and ROS1. In contrast to entrectinib, however, crizotinib does not inhibit TRK but is a potent inhibitor of MET. A patient with NSCLC having a MET copy number gain without a high MET-to-CEP7 ratio was previously reported to experience a rapid response to crizotinib. 8 In addition, a phase 2 study of crizotinib  efficacy for NSCLC found that patients with greater than or equal to six copies of MET as determined by FISH, including those with MET polysomy, had the best objective response rate (32%) and median progressionfree survival (3.2 mo). 9 Indeed, tumor shrinkage was observed in some patients with MET polysomy in these previous studies, 7,9 suggesting that some patients with MET polysomy benefit from crizotinib. MET polysomy might therefore have the potential to be an oncogenic driver that is targetable by crizotinib.
Although the current patient had MET polysomy before entrectinib, she had an initial response to entrectinib treatment. It has been proposed that intratumoral heterogeneity is associated with a differential drug response. 10 Although the growth of most cells harboring the ROS1 rearrangement in the present patient might have been inhibited by entrectinib, a small subpopulation of cells also harboring MET polysomy might have survived during continuous entrectinib treatment and become the dominant population in the entrectinibresistant tumor.

Conclusions
Our report suggests that crizotinib is an effective option for patients with MET amplification or polysomy, even after disease progression during entrectinib treatment. Further study is warranted to understand the mechanisms of entrectinib resistance in ROS1-rearranged NSCLC and the efficacy of crizotinib in NSCLC with MET polysomy.