Validating ROS1 rearrangements as a therapeutic target in non-small-cell lung cancer.

Abstract

Initially thought to be prevalent predominantly in hematologic malignancies and sarcomas, chromosomal rearrangements leading to oncogenic gene fusions have now been described across a range of epithelial cancers, including non–small-cell lung cancer (NSCLC). In 2007, rearrangements involving the anaplastic lymphoma kinase (ALK) gene were described in NSCLC. Single-arm phase I and II studies led to accelerated approval of the first-in-class ALK inhibitor crizotinib for ALK-rearranged NSCLC. Two phase III studies confirmed the superiority of crizotinib over chemotherapy, establishing crizotinib as a standard of care for patients with NSCLC whose tumors test positive for ALK gene rearrangements. In the article that accompanies this editorial, Mazieres et al report the therapeutic efficacy of targeting a second gene rearrangement in NSCLC with crizotinib, this time involving the ROS1 gene. ROS1 is an orphan receptor tyrosine kinase encoded by the ROS1 gene that is vulnerable to intrachromosomal or interchromosomal rearrangements, resulting in transforming gene fusions that occur in tumor types including glioblastomas, NSCLC, cholangiocarcinoma, ovarian cancer, gastric cancer, and colorectal cancer. ROS1 gene rearrangements were first recognized in NSCLC in 2007 and have since been described in 1% to 2% of patients with NSCLC. At least 11 fusion partners have been identified in NSCLC, including CD74-ROS1, SDC4-ROS1, EZR-ROS1, and SLC34A2-ROS1, all of which maintain a constant breakpoint in ROS1, preserving the kinase domain and resulting in aberrant ROS1 expression with constitutive kinase activity. These rearrangements may be detected in clinical samples by a variety of techniques, including fluorescent in situ hybridization, immunohistochemistry, reverse-transcriptase polymerase chain reaction, and next-generation sequencing. Patients with ROS1-rearranged NSCLC share many clinical features in common with ALK-rearranged NSCLC in that they are typically younger, are never-smokers, and have tumors with adenocarcinoma histology. ROS1 and ALK share substantial sequence homology in their kinase domains. Crizotinib, also a potent inhibitor of the ROS1 kinase, has activity in preclinical ROS1-rearranged NSCLC models, providing a rationale for its evaluation in patients with ROS1-rearranged NSCLC. Since the initial report published in Journal of Clinical Oncology in 2012 of a dramatic response to crizotinib in a 31-year-old male neversmoker whose tumor tested positive for a ROS1 gene rearrangement, there have been additional case reports indicating activity of crizotinib in this setting. Mazieres et al now provide a substantial addition to the body of evidence indicating efficacy of targeting ROS1 rearrangements in NSCLC with crizotinib. In this retrospective study, conducted in 16 centers in six countries in Europe, results are presented from 32 patients with ROS1 rearrangement–positive NSCLC who received off-label treatment with crizotinib. Consistent with previous descriptions, patients were predominantly young (median age, 50.5 years), were never-smokers, and had tumors with adenocarcinoma histology. An impressive response rate of 80%, with a median progression-free survival (PFS) of 9.1 months, was seen in this European cohort of patients with ROS1-rearranged NSCLC treated with crizotinib. Mazieres et al are to be commended on coordinating an international collaborative study, which although subject to limitations related to sample size, potential for selection bias, and reliance on investigator assessments of response, complements and provides independent validation of a recent report by Shaw et al of patients with ROS1-rearranged NSCLC prospectively treated in the molecularly enriched expansion cohort of the crizotinib phase I trial. This ongoing study conducted in the United States, Korea, and Australia enrolled 50 patients with NSCLC with ROS1 gene rearrangements and reported an objective response rate to crizotinib of 72% (95% CI, 58% to 84%), with a median PFS of 19.2 months (95% CI, 14.4 to not reached). No obvious relationship between ROS1 fusion partner and durability of response to crizotinib was demonstrated. The response rate to crizotinib observed in these two independent studies (80% and 72%) confirms that ROS1 rearrangement– positive NSCLC, like ALK-rearranged or EGFR mutation–positive NSCLC, represents an oncogene-addicted tumor and validates ROS1 as a therapeutic target in NSCLC. Interestingly, although Shaw et al reported a median PFS of 19.2 months, raising the possibility that the durability of responses to crizotinib may be greater in ROS1 rearrangement–positive NSCLC compared with ALK rearrangement–positive NSCLC, Mazieres et al report a more modest PFS of 9 months, similar to what has been observed in ALK-positive NSCLC. Although there are potential differences in the populations under evaluation in the studies, including, for example, the proportion of Asian patients in each study, the differences seem most likely attributable to sample size and duration of follow-up. The PFS data are relatively preliminary in both studies, with many patients still in follow-up; however, the prospective study by Shaw et al had a larger sample size and potentially longer duration of follow-up than the retrospective study of Mazieres et al and may therefore provide a more precise estimate of PFS. JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 33 NUMBER 9 MARCH 2