Clinical Management of NTRK+ Advanced NSCLC

Ben Levy, MD: There are now 2 TRK [tyrosine receptor kinase] inhibitors that are active, proven and FDA approved for TRK fusion-positive lung cancer. The first is larotrectinib. This was initially published in the New England Journal of Medicine looking at patients with NTRK fusion-positive cancers, not just lung cancer, but multiple cancers showing a response rate of 81%. Because of that, larotrectinib was approved. Entrectinib is the other drug. Of course, entrectinib has activity in ROS, but also has potent activity for NTRK fusions. Now we’ve got NTRK in the data as well showing response rates north of roughly 55% to 65%, and because both drugs have elicited meaningful responses in these patients, both drugs are approved. Both are [also] preferred by the NCCN [National Comprehensive Cancer Network]. Because larotrectinib came first, it is my go-to therapy for patients with NTRK fusions. It’s hard to identify NTRK fusions [because] they are uncommon in common malignancies. You don’t find that often in lung cancer, breast cancer, or colon cancer, but they are common in uncommon cancers, like infantile fibrosarcoma and other types of tumors. It doesn’t mean we shouldn’t look for them; we should. But clearly, they’re rare. As it stands now, larotrectinib and entrectinib are both approved therapies. We need to test for these fusions. Again, these are fusions we’re looking for, not mutations. If you see an NTRK mutation on your next-generation sequencing report, that will not predict efficacy to these therapies. You must identify the fusions. That’s very important.

Mark A. Socinski, MD: NTRK fusions should also be a part of your comprehensive genomic testing at the time of diagnosis. They are much less common, far less than 1% in nonsmall–cell lung cancer. We know that we have 2 agents approved in this area. First was larotrectinib, and the second was entrectinib. These agents are clearly active. When you look at the studies of these 2 agents, they were done in the context of a basket trial. For instance, with larotrectinib, there were well over 100 patients in this experience, but only 20 of them had nonsmall–cell lung cancer. In those 20 patients with nonsmall–cell lung cancer, the overall response rate was 73%. Apparently only 15 of them were valuable for response at the time of the report, that’s why it’s 73%. The durability of that response was excellent—81% of patients were still in a response at 1 year. We also know entrectinib came second as the NTRK targeting agent. Remember, entrectinib also has approval in ROS1-positive nonsmall–cell lung cancer. The data in NTRK, again, shows around a 60% overall response rate with almost half of patients still on a response at 1 year. The toxicity profiles of these agents are similar, but slightly different. Certainly, with entrectinib, the issues with CNS [central nervous system] alterations, mood alterations, and dizziness—these sorts of things—can sometimes be problematic. Weight gain seems to be an issue with both agents. Some myelosuppression with anemia and some hypersensitivity with larotrectinib, too. However, in general, none of these have a high rate of grade 3 or grade 4 toxicity. It’s all far less than 10%. Even then, only the minority of patients are having either grade 1 or grade 2 adverse events of the natures that I mentioned.

Stephen Liu, MD: We’re seeing a lot of basket studies now. When we look at these basket trials, do we look at a specific tumor type or do we take a step back and look at it overall? I focus specifically on the lung outcomes because that’s my clinical focus, but overall, it depends on how we see different patterns. There are some targets that are clearly tumor agnostic, and the case of NTRK is a good one. NTRK fusions predict response to NTRK inhibitors, like entrectinib and larotrectinib, and it is across tumor types. That’s a setting where the alteration is more important than the tumor type, where that defines the tumor. This extends to most other fusions: ALK fusions, ROS1 fusions, RET fusions, for example, but we must watch closely because we’re still learning about this field. There’s a lot less certainty for mutations. A good example is BRAF mutations and KRAS mutations, where they may guide therapy. BRAF predicts therapy in lung cancer and melanoma, but not so much in colon cancer. We have good inhibitors now approved for nonsmall–cell lung cancer with KRAS G12C, but not effective in colon cancer with the same mutation. So there are some alterations where the environment of that tumor, the tissue of origin, is playing an important role. Maybe not yet totally defined right now, it depends, and you must look at it both ways.

Transcript edited for clarity.