DNA-Based NGS Panel Accurately Detects MET Exon 14 Skipping Mutations in NSCLC

Research has identified that 2% to 4% of patients with non—small cell lung cancer (NSCLC) haveMETexon 14 skipping mutations. These patients are eligible to receive MET inhibitor therapy, such as crizotinib (Xalkori), but pathologists need to identify these mutations in order for them to receive such therapies.

An analysis conducted at the University Medical Center in the Netherlands demonstrated their custom-made DNA-based next-generation sequencing (NGS) panel is capable of detecting 96% ofMETexon 14 skipping mutations in patients with NSCLC. Their panel has been used for routine practice since 2015.

Investigators found that theMETexon 14 skipping mutation was found in 2% of patients with NSCLC that were evaluated for targeted therapy from January 2016 to January 2018. Out of 46 patients with NSCLC analyzed,METexon 14 skipping mutations were most commonly found in adenocarcinomas (68.9%), followed by sarcomatoid carcinoma (17.8%).

Following their detection ofMETexon 14 skipping mutations, 11 patients were treated with MET inhibitors, either crizotinib or others in clinical trials, such as capmatinib and tepotinib. According to Willemina R. Geurts-Giele, PhD, lead investigator, 4 of the 5 patients treated with crizotinib achieved a response.

In an interview withTargeted Oncology, Geurts-Giele, University Medical Center, Rotterdam, The Netherlands, discussed the accuracy of detectingMETexon 14 skipping mutations in patients with NSCLC with DNA-based NGS panels. These findings were presented at The Association for Molecular Pathology (AMP) Annual Meeting and Expo (AMP 2019).

TARGETED ONCOLOGY: How frequently do you see theMETexon 14 skipping mutation in NSCLC?

Geurts-Giele:The frequency we found is about 2% of NSCLCs that we test for targeted therapy. This mutation is quite unique in that in the DNA panel, it is quite a lot of different variants that all result in the sameMETexon skipping [mutation] on the RNA. The RNA is quite easy to detect, but with the DNA, it is quite difficult because you have very small point mutations, but you also have very large deletion insertion mutations, which is difficult, especially with amplicon sequencing, to detect.

TARGETED ONCOLOGY: How is this detected in patients with NSCLC?

Geurts-Giele:You have 2 approaches. You can do it either with DNA or you can do it with RNA. With RNA, you can use a simple polymerase chain reaction because it is always the same event, so it is quite easy. The downside of that is that in [patients with] lung cancer, you usually do not have as much material, so you have limited amounts [of DNA]. The DNA analysis is quite routinely used in pathology departments. RNA is starting to come up, but not all labs are routinely performing RNA analysis. That’s why we focused on optimizing the DNA analysis part because we do that in all of our patient materials. If we can detect all mutations or almost all mutations on the DNA, then at least we are not missing out on a lot. You can have a perfect RNA assay, but if you can only perform it for about 60% of your lung cancers, you are still missing out on a lot. That was our approach.

TARGETED ONCOLOGY: How did you design this analysis, and what were the findings?

Geurts-Giele:What we did is we designed amplicons for theMETexon detection. For the basis of that, we used the paper fromCancer Researh.They used hybrid capture, and it really showed a lot of variability of the DNA level for those mutations, and that we used as a starting point to design our amplicons. Our amplicons covered a splice size on both sides ofMETexon 14, but it also covered the branch sides and the poly-paradigm tracks, which are all important regions for accurate skipping and for accurate splicing. If they are not there anymore or changed, it could lead to skipping. Those are all [covered] in our panel.

We have 4 amplicons, which are overlapping amplicons also because the mutations are so variable that if your amplicons are overlapping, [there’s a] higher chance of finding the actual mutations. We implemented this in 2015 in our routine screening for NSCLCs and with this approach, we detected 2% ofMETexon 14 skipping mutations, which is quite in line with the percentages in the literature. Of course, we are missing some because they also are very large deletions and our amplicons are 50-based. We think it is a minor amount and based on our in silicopredictions, our panel should be able to detect up to 96% of all METexon 14 skipping mutations on DNA levels. We think that is quite good, so we urge people to look at their NGS panel to see what it catches and if it does not catch enough.

There was a paper in 2018 that showed the commercially available panels at that time, and [findings demonstrated that they] detected just up to 63% of theMETexon skipping mutations. That is just not enough. We urge people to look at their panels to see what they are covering. If they are not covering enough, they should improve that.

TARGETED ONCOLOGY: Is this panel used in your study being widely used?

Geurts-Giele:Our panel was a custom-based panel, and it was a panel we already were using. In 2015, we added the extra amplicons for theMETexon skipping detection, so it is not a commercially available panel. It’s really custom amplicons. I don’t know how the panels are now and maybe they are better in their coverage, but if people are using commercially available panels, please check whether it is well covered or otherwise you can add amplicons to that. You can do RNA analysis as well because, of course, that is another option, but with the older commercially available panels, you are missing out on a lot, and that is bad because this is a target that can be treated. People are responding well to these MET-targeting inhibitors. We should be detecting those.

TARGETED ONCOLOGY: What are the next steps to make these panels more widely used?

Geurts-Giele:It would be great if the commercially available panels would expand theirMETexon detection abilities. I think the RNA analysis is also becoming more routinely used in the pathology labs. If we have that and we can do that for 60% of our lung samples, but 90% to 95% you can detect it very easily on the RNA, so that is a very good approach. It should be either a very good, comprehensive DNA panel or you should be able to perform the easy RNA [testing], but you should be performing it on most of your lung samples.

TARGETED ONCOLOGY: What are the biggest challenges in detectingMETexon 14 skipping mutations in NSCLC?

Geurts-Giele:For the amplicons, you just have to be in the right area. They are quite familiar now, which our study shows, but also a recent study published in theJournal of Thoracic Oncologyshow a similar approach with these amplicons. They all have quite good coverage of around 2% with the METexon 14 skipping mutation in NSCLC. I think if you are just aware of the complexity, and you are looking at your panels to try to improve it, I think we are doing quite okay there.

Of course, if you can detect the skipping mutation, these patients should also be able to receive treatment based on those mutations. In our study, we had 11 patients treated with a MET inhibitor. Six of those were in trial, so we cannot say anything about those, but 5 received crizotinib in comparative use. Four of them had a response with the progression-free survival ranging from 4 to 14 months. That is quite good, but I don’t think all of the patients with these mutations are yet receiving therapy, so that is also something we are trying to get treatment for those patients with those mutations.

TARGETED ONCOLOGY: What do you hope the audience here at AMP 2019 takes away from this?

Geurts-Giele: