Overcoming MET TKI Resistance in Patients with Non–Small Cell Lung Cancer

In an interview with Targeted Oncology™, Yonina R. Murciano-Goroff, MD, MSc, DPhil, discussed how MET fusions impact patients with NSCLC and the issue of resistance to targeted therapies that treat MET fusion-positive disease.

In MET fusion-positive non–small cell lung cancer (NSCLC), the newly clinically identified MET G1090A resistance mutation has been characterized as a driver of resistance to type 1 MET tyrosine kinase inhibitors (TKIs) by way of serial targeted genomic sequencing.1

Investigators of an analysis characterized resistance to crizotinib (Xalkori) and evaluated the activity of MET inhibitors with alternative binding modes with structural modeling and in vitro kinase assays. The findings of the analysis show that switching to type 2 MET inhibitors after patients develop resistance to type 1 inhibitors may help overcome resistance to certain mutations. Considering the preclinical nature of the study along with a proof-of-concept clinical case report, further research is needed to confirm the findings.

In an interview with Targeted Oncology™, Yonina R. Murciano-Goroff, MD, MSc, DPhil, a medical oncologist at Memorial Sloan Kettering Cancer Center, discussed how MET fusions impact patients with NSCLC and the issue of resistance to targeted therapies that treat MET fusion-positive disease.

TARGETED ONCOLOGY: Can you kind of discuss the prevalence of met fusions and non–small cell lung cancer and what outcomes are like for these patients?

Murciano-Goroff: This is complex because what we know is that MET fusions are rare oncogenic drivers in non–small cell lung cancer. But on the other hand, they may be somewhat underreported. Part of that is because the standard assays that we use for next-generation sequencing in the clinic are DNA-based assays. And we know from our experience with other fusions, that some fusions can be missed when using DNA-based testing but do get picked up on RNA-based testing. So, they may be underreported. But, even accounting for that, these are rare in non–small cell.

In terms of other histologies, we do see them across a number of different cancer types. And they're especially enriched in gliomas where they've been reported in as many as 12% to 14% of cases. So, MET fusions are important but rare oncogenic drivers. That said, we now have more and more instances of fusions in non–small cell lung cancer that are rare, but if we're able to biologically and clinically characterize them, they can be very targetable. NTRK is a classic example, and that's why we see MET fusions as truly an unmet need, as it were, in terms of trying to better characterize them and trying to improve outcomes through better biological and clinical understandings of this subset of non–small cell.

To date, what clinical questions are there about met fusion positive disease and how to approach treatment?

I would say that there's still a lot of open questions, which makes it an important area. As a field, we are really still working out what is the optimal approach for our patients with MET fusion-positive disease. Most of what we have been doing to date on targeted therapy front in particular, has been trying to translate results from other MET-positive tumors. We know more about exon 14 skipping alterations in non–small cell, and we even know MET amplifications in non–small cell better than we know the MET fusions. So, a lot of the strategy to date has been trying to translate from what we know about that works for patients with other MET alterations to MET fusions.

What does the treatment landscape currently look like for these patients?

Again, we’re still trying to translate things on the targeted therapy front from other MET alterations. But what I would say is that we know at this point that in some cases, MET-directed tyrosine kinase inhibitors do make a difference and help patients with MET fusions.

You coauthored a study around another strategy for targeting MET-altered NSCLC. Can you explain the study design?

This study came out of trying to better characterize resistance and how we can overcome it in patients with MET fusion-positive disease. So, we hadn't had reports of resistance with MET TKIs in MET fusion-positive disease yet. And so, we started with a patient that we were able to identify who had a MET fusion, was treated with a type 1 MET TKI (in this case crizotinib), and did well for some time, but unfortunately, they ultimately progressed. And at that point, we were able to do sequencing and to show a resistance alteration, G1090, a MET alteration that we thought was contributing potentially to the resistance.

So, what we did was conduct molecular dynamics simulations, and showed that the interaction between crizotinib and MET wild type is with the y1230 residue. Importantly, when the G1090 mutation was there, that residue was flipped conformationally by about 180 degrees. So, one can picture exactly why that might alter the interaction between MET and crizotinib. Changing that interaction lead to what we saw in the patient, which was resistance to crizotinib.

Now, having done that, we also did in vitro kinase assays to confirm this result. And then, we wanted to be able to overcome this both for this patient and for future patients. And we had some experience with this: some of the senior authors on this presentation had done prior work on NTRK fusions showing that switching inhibitor class has the potential in some cases to overcome conformational resistance. And so, kind of taking the analogy from NTRK, we started looking at different class 1 and class 2 inhibitors. These included type 1 inhibitors like crizotinib and capmatinib, and then type 2 inhibitors, things like cabozantinib, for example. And what we saw there was that the type 1 inhibitors seem to all have to interact with y1230. They had predicted resistance, both on the molecular dynamics simulations, but then also on the in vitro kinase assays, whereas we saw sensitivity, so the ability to overcome this G1090A-mediated resistance with type 2 inhibitors.

That was a very exciting result for this 1 alteration. And then we figured, well, having done this for the fusion, let's see if we can help patients more broadly, who have resistance alterations to MET TKIs. So, we went through the literature, not just for fusions with resistance, because this hasn't been particularly reported before. But then, looking at patients with other MET alterations who had been treated with MET TKIs, to see what other resistance alterations had been found. We did molecular dynamics simulations there. And excitingly found that there were potentially other mutations where class switching, for example, between crizotinib and cabozantinib, might make a difference.

What were the key findings of your analysis?

I think the key findings are sort of two-fold. One, I would say on the clinical front, starting with 1 case of this rare fusion was really that we were able to show that switching between a type 1 and type 2 inhibitor with this particular resistance alteration made a difference. And actually, we went back to our patient who had developed that G1090A alteration and treated them with cabozantinib (ie. with a type 2 inhibitor) after they had developed the resistance to the type 1 inhibitor. We saw an excellent response in terms of both tumor shrinkage and clinical response, which was very exciting on the clinical front and kind of gives us a lot of hope about what we might be able to do for other patients. Beyond this single patient and beyond this single alteration, we were able to look more broadly at MET resistance alterations, and to say, at least based on our preclinical work based on our molecular dynamics simulations, that we might expect that there may be other alterations where such type switching might make a difference in terms of likelihood of response and in terms of overcoming resistance.

What should oncologists learn from this study?

From a drug development and preclinical perspective, this really lends further support to the idea that we need a more rational design of type 2 inhibitors. I think, again, it lends further support that there may be cases in which that's important. And then on a clinical front, with certain alterations, in certain cases, there may be instances in which type switching of the inhibitor can make a difference in overcoming resistance for patients.


Murciano-Goroff Y, Kannan S, Chang JC, et al. Switching inhibitor class overcomes crizotinib resistance in a MET fusion-positive NSCLC with a novel acquired MET G1090A mutation. Preseneted at: American Association for Cancer Research 2022 Annual Meeting; April 8-13, 2022; New Orleans, LA. Abstract 5239