We've already discussed prior work on Met as a potentially valuable target in NSCLC, as illustrated by work with ARQ-197 combined with the EGFR inhibitor Tarceva (erlotinib) has demonstrated encouraging early results. More recently, Met was the source of a lot of buzz at the European Society for Medical Oncology Congress this past week because of another agent targeting Met, called MetMAb, that has demonstrated compelling evidence of activity, at least in a subset of patients with advanced NSCLC. And this is a subset that includes about half of the NSCLC population. (click on image to enlarge) Met is a receptor protein that is mutated and overexpressed in many cancers, in which it is typically associated with a worse prognosis, including in NSCLC. Importantly, in the last few years, it has been identified as one of the mechanisms of resistance to EGFR tyrosine kinase inhibitors (TKIs) like Tarceva or Iressa (gefitinib), both from the very start of EGFR TKI therapy (primary resistance), and after an initial response (secondary or acquired resistance). MetMAb is a single-armed monoclonal antibody (typically abbreviated mAb or MAb) that adheres to and blocks the receptor portion on the outside of the cell without leading to the receptor being stimulated and causing growth, migration, and promotion of survival of the cancer cell. In various cell and animal models, it's active against many cancers.

This particular trial was presented at ESMO by my friend Dr. David Spigel, at the Sarah Cancer Cancer Center in Nashville, TN a few days ago, who was kind enough to send me slides from his presentation. The trial enrolled 128 patients who had received one or two prior lines of therapy for advanced NSCLC, any histology, but no prior exposure to an EGFR TKI, and patients had to have tissue available for submission. The phase II study randomized patients to daily Tarceva with MetMAb, administered IV one day every three weeks, or daily Tarceva with placebo on the same schedule. The trial was designed to compare the progression-free survival (PFS) in patients with high Met expression, as well as PFS in the broader, overall population on the trial. The definition of Met high vs. low was based on immunohistochemistry, so measuring protein expression on the cancer cells. This is graded as 0 (no expression), 1+ (mild/minimal), 2+ (moderate), or 3+ (strong) expression of the Met protein: The patient populations were pretty comparable, about 60% men, ~90% Caucasian, about 25% squamous NSCLC, and only 12-16% never-smokers in the two groups, so a pretty typical NSCLC population and not heavily over-representing never-smokers or patients with other factors associated with a high probability of an EGFR mutation (as we might expect for people who didn't receive an EGFR inhibitor earlier in treatment). The EGFR mutation rate was in the 10-12% rangein the two arms, while KRAS mutations were seen in 23% of the patients on both arms. As you'd expect to see from prior work, the two arms were both split pretty evenly between Met high and Met low expression (54% Met high, as noted in figure above). Looking at the entire population, there were no real differences in PFS or OS to suggest a benefit from MetMAb, but the two groups of Met high vs. Met low had remarkably different results. Specifically, the patients in the Met high group had a nearly 50% improvement in both PFS and OS (though not quite reaching statistical significance for either endpoint): But if one group received a benefit while the overall population showed no difference, that means that another group must have done poorly. In fact, that was clearly shown in the Met low population, who had a significantly worse PFS and OS when MetMAb was added to Tarceva: There were no differences in the degree or pattern of benefit in various subgroups based on smoking history, histology, performance status, or even EGFR or KRAS mutation status. But not only were the results different for Met high vs Met low, both PFS and OS went gradually from most beneficial for Tarceva alone to most beneficial for Tarceva/MetMAb as the Met staining pattern went from o to 1+ to 2+ to 3+: To me, these results are reminiscent of the IPASS results that showed the extremely divergent results of whether patients did better with chemo or Iressa depending on whether a patient has an EGFR mutation or not. Importantly, it underscores how important molecular markers have become, and the need to review results by relevant subgroups. Looking at the results only for an overall population not divided by Met expression, these results would have shown no discernable result and likely would have been abandoned. But instead, it appears most likely to me that MetMAb represents a valuable addition to Tarceva, not just in patients with an EGFR mutation, in whom it may restore sensitivity after acquired resistance (though this wasn't tested in the trial described here), and importantly improving results when given with Tarceva to the other 90% of patients in North America with advanced NSCLC who don't have an EGFR mutation, but only in those patients who have high Met expression. For the other half, we see another example of why more is not always better. Right now, Met testing isn't widely available or performed, nor is there a commercially available agent to direct against it, but I expect that we'll see a similar situation arising as was developed for testing for an ALK rearrangement and treating with crizotinib. The future development and trial plans for MetMAb are still in development, but I hope and expect to share more information as soon as those plans are crystallized.