Crizotinib (Xalkori) produced a high response rate and durable responses in patients with ROS1-rearranged non–small cell lung cancer (NSCLC), according to a study reported in The New England Journal of Medicine.1 Lead authors Alice T. Shaw, MD, PhD, of Massachusetts General Hospital Cancer Center, and Sai-Hong I. Ou, MD, PhD, of the University of California at Irvine, contributed equally to this article.
Chromosomal rearrangements in ROS1, which encodes the proto-oncogene receptor tyrosine kinase ROS1, define a distinct molecular subgroup in NSCLC. In addition to inhibiting ALK, crizotinib inhibits ROS1 and MET. As noted by the investigators, oncogenic ROS1 fusions may account for approximately 15,000 of the worldwide 1.5 million new cases of NSCLC each year. ALK and ROS1 rearrangements are infrequently found within the same tumor. Both are more common in patients with a history of never or light smoking and in adenocarcinoma.
Study Details
A total of 50 patients with advanced NSCLC were enrolled between October 2010 and August 2013 in the ROS1 expansion cohort of a phase I study of crizotinib. Patients received crizotinib at the standard dose of 250 mg twice daily.
Patients had a median age of 53 years; 56% were female; 54% were white and 42% were Asian; 78% were never-smokers; 98% had adenocarcinoma; 44% and 54% had an Eastern Cooperative Oncology Group performance status of 0 and 1, respectively; and 44% had received at least one prior regimen for advanced disease.
A break-apart fluorescence in situ hybridization assay was used to identify ROS1 rearrangement in 49 of the 50 patients. An atypical ROS1 fluorescence in situ hybridization pattern was observed in 1 of the 49 patients; subsequent next-generation sequencing showed nonrearranged ROS1. Another patient was tumor fluorescence in situ hybridization–positive for both ROS1 and ALK rearrangement; next-generation sequencing showed an EML4-ALK fusion with no ROS1 rearrangement. Of an additional 32 ROS1-positive tumors tested for ALK rearrangement, none was positive. Of 15 ROS1 fluorescence in situ hybridization–positive tumors tested for MET amplification by fluorescence in situ hybridization, 1 was positive.
Responses
The objective response rate among the 50 patients was 72% (95% confidence interval [CI] = 58%–84%), with a complete response in 3 (6%), a partial response in 33 (66%), and stable disease in 9 (18%). The median time to first response was 7.9 weeks (range, 4.3–32.0 weeks). At the time of data cutoff, 23 of the 36 responses (64%) were ongoing. The estimated median duration of response was 17.6 months (95% CI = 14.5 months to not reached).
Three patients (6%) had evidence of progressive disease on the first restaging scans. One was the patient with atypical fluorescence in situ hybridization results found to be negative for ROS1 rearrangement. A second patient had discontinued crizotinib for 6 weeks before the first restaging scans due to bowel perforation, which was believed to be associated with glucocorticoid use and preexisting diverticular disease; when this patient resumed crizotinib, a 62% reduction in tumor burden was observed. The third patient exhibited a 26% increase in tumor burden on restaging.
The median duration of treatment was 64.5 weeks (range, 2.3–182.0 weeks), with 30 patients (60%) continuing to receive crizotinib after data cutoff. The median progression-free survival was 19.2 months (95% CI = 14.4 months to not reached), with 25 patients (50%) still in follow-up for disease progression at data cutoff. The median follow-up for overall survival was 16.4 months; overall survival at 12 months was 85% (95% CI = 72%–93%), with the median not reached.
Toxicity
Of all treatment-related adverse events, 94% were grade 1 or 2. The most common treatment-related adverse events of any grade were visual impairment (82%), diarrhea (44%), nausea (40%), peripheral edema (40%), constipation (34%), vomiting (34%), elevated aspartate transaminase levels (22%), and fatigue (20%). No treatment-related grade ≥ 4 adverse events were reported. The most common grade 3 events were hypophosphatemia (10%), neutropenia (10%), and elevated alanine aminotransferase levels (4%). All visual-impairment events were grade 1. One patient (2%) discontinued crizotinib due to a treatment-related event (nausea).
Fusion Partners
A total of 30 samples were tested by next-generation sequencing (n = 27) or reverse transcriptase polymerase chain reaction (n = 3). Of the 27 tested by next-generation sequencing, 22 had specific ROS1 rearrangements, 4 were negative for ROS1 rearrangement, and the assay failed in 1; 1 of the 4 negative samples was positive for EML4-ALK.
ROS1 fusion partners consisted of the gene-encoding CD74 (11 of 25 samples), SDC4 (4 samples), EZR (4 samples), SLC34A2 (3 samples), and TPM3 (1 sample); all of them have previously been identified as ROS1 fusion partners. Two new partners—LIMA1 (LIM domain and actin-binding 1) and MSN (moesin)—were identified by next-generation sequencing. There was no apparent association between the type of ROS1 rearrangement and the response to crizotinib or the duration of crizotinib therapy.
The investigators concluded:
… ROS1 rearrangement defines a second molecular subgroup of NSCLC for which crizotinib is highly active. In the majority of patients, crizotinib induced durable clinical responses and was associated with grade 2 or lower toxic effects. These results highlight the importance of screening for this genetic alteration in patients with advanced NSCLC. Although fluorescence in situ hybridization was used in this study, other diagnostic methods have been proposed, and further work is required to establish the most effective screening strategy for ROS1 rearrangement. ■
Disclosure: This study was funded by Pfizer and by grants from the National Cancer Institute, Uniting Against Lung Cancer, Swedish Research Council, and Be a Piece of the Solution. For full disclosures of the study authors, visit www.nejm.org.
Reference
1. Shaw AT, Ou S-HI, Bang YJ, et al: Crizotinib in ROS1-rearranged non-small-cell lung cancer. N Engl J Med 371:1963-1971, 2014.
