Patients with non–small cell lung cancer harboring MET exon 14 skipping mutations have commonly been treated with the multikinase inhibitor crizotinib, but more MET-specific therapies are being developed. Two such agents, capmatinib and tepotinib, have demonstrated preliminary efficacy in ongoing phase II trials.

Roughly 4% of patients with nonsmall cell lung cancer (NSCLC) harbor MET exon 14 skipping mutations (METex14), and the ALK/ROS1/MET inhibitor crizotinib (Xalkori; Pfizer) is a common treatment, albeit not currently FDA-approved for this disease subtype. More MET-specific therapies may be on the horizon, however, as reported at the 2019 American Society of Clinical Oncology (ASCO) Annual Meeting in Chicago, IL, May 31–June 4.

Targeting METex14 is tricky, explained Karen Reckamp, MD, of City of Hope Comprehensive Cancer Center in Duarte, CA, because “more than a hundred genomic variants have been described,” from deletions and insertions to base substitutions. The result is constitutive MET signaling, and new therapies are much needed for a patient population that tends to be older (median age 74 years), with increased comorbidities and more treatment-refractory tumor histologies—chiefly sarcomatoid carcinomas, Reckamp said.

At ASCO, Juergen Wolf, MD, of University Hospital Cologne in Germany, presented data from the phase II GEOMETRY mono-1 trial, which is evaluating the selective MET inhibitor capmatinib (Novartis) in METex14 NSCLC. Two cohorts were highlighted: 69 patients who had received prior treatment, including chemotherapy; and 28 untreated patients. The objective response rates (ORR) with capmatinib were 41% and 68%, respectively. The median duration of response (DOR) in each cohort—10 months and 11 months—“appears to be independent of prior lines of therapy,” Wolf observed.

Of 13 patients with brain metastases, seven saw their lesions shrink significantly, or disappear in four cases, with capmatinib. The drug's efficacy was independent of any specific METex14 variant, Wolf noted, and treatment was well tolerated, with the main side effects being peripheral edema and nausea.

Another investigational MET inhibitor discussed at ASCO was tepotinib (Merck). Paul Paik, MD, of Memorial Sloan Kettering Cancer Center (MSKCC) in New York, NY, discussed data from the phase II VISION study, reporting that among 48 patients whose METex14 alterations were determined by liquid biopsy, the ORR with tepotinib was 50%. In 51 patients who underwent a tissue biopsy instead, the ORR was 45.1%. The median DOR was 12.4 months and 15.7 months, respectively, and consistent tumor shrinkage was observed regardless of the previous number of therapies.

Evidence of tepotinib's activity in the central nervous system has been largely preclinical, Paik said, “but in our trial, we saw that patients with brain metastases at baseline appeared to benefit equally from treatment, in terms of durability of response.” Tepotinib's toxicity profile was much the same as that of capmatinib, and treatment was also well tolerated.

“These are preliminary efficacy data in small cohorts of patients, but the response rates look promising,” said Reckamp of both trials. “Due to numerous variants, I don't know that there's a gold standard for the [METex14] testing platform at this point. But it's important to have learned, from VISION, that the results are very similar whether we look at liquid or tissue biopsy.”

Meanwhile, Robin Guo, MD, also of MSKCC, presented findings from a retrospective analysis aimed at better understanding the development of resistance to MET inhibitors. Focusing first on primary resistance, he observed that in 16 patients with METex14 alterations identified by next-generation sequencing, surprisingly, almost one third did not have detectable MET protein expression via an antibody assay or mass spectrometry. This was counterintuitive, he said, “because biologically, these alterations have been thought to result in increased MET levels.”

Guo then reported that confirmed responses to MET inhibition—mainly crizotinib in this study—were observed only when protein expression, not just the genomic variant, was detected in patients. Interestingly, though, “the amount of MET was not associated with the degree of response to therapy,” he said.

Next, examining paired pretreatment and post-progression tumor samples from 14 patients, “we saw that on-target acquired resistance was uncommon,” Guo said. Of several secondary MET kinase domain mutations published in the literature, only one, D1228N, was identified in these patients. Another culprit, also rare and not previously reported, was amplification of the MET ligand, HGF.

On the other hand, bypass pathways of acquired resistance were more frequently seen, Guo noted. He and his colleagues uncovered two previously unreported routes: EGFR amplification, and inactivation of RASA1—a negative regulator of RAS signaling—through a truncating mutation. Their putative roles need further research, he said, “but this profile of resistance may be relevant when selecting the next therapy for patients whose disease has progressed on crizotinib.” As well, the study findings “underscore the importance of testing for MET protein levels” alongside METex14, he added.

More molecular screening efforts are warranted in NSCLC as a whole, Reckamp agreed. “There are now cost-effective ways to do so, and if patients aren't being tested for these alterations, they won't be given the appropriate therapy.”

The FDA has granted orphan drug and breakthrough therapy designations to capmatinib; tepotinib is on the agency's fast track path. A third inhibitor, savolitinib (AstraZeneca), is also in development. Overall, “with regards to bringing MET-specific drugs into clinical practice, we're on our way,” Reckamp said. –Alissa Poh