David S. Hong, MD: What patients do I consider? That’s a really important question. If you look at the subsets of the population, up to now, in the adult data set in the abstract we’ve now identified at least 13 different tumor types. Likely there will be even more tumor types in an upcoming data set at ESMO [the European Society for Medical Oncology], up to 22 different tumor types. If you look and break down the type of tumors and the type of patients who are seen, there are types of tumors that are almost pathognomonic. That means almost diagnostic of that disease with NTRK. For example, mammary analogue secretory carcinoma, called MASC tumors, are oftentimes misdiagnosed as standard head and neck cancers, salivary gland tumors, or parotid tumors. But they have a distinct feature histologically: invariably if you test them for NTRK, they will have NTRK fusions. Many of these patients present early, and so they are cured by surgical resection. But a number of these patients will eventually develop metastatic disease, and at that point there is really no standard therapy for them. So I think that’s one key point.

There’s also a whole class of histological tumors similar to MASC. Secretory breast, for example, is another type of cancer that probably harbors MASC tumors. Other almost diagnostic or pathognomonic diseases include rare tumors such as pediatric infantile fibrosarcomas, pontine gliomas. They almost invariably have NTRK fusions. Obviously, the community oncologists will not see those patients, so they’ll be oftentimes seen in an academic pediatric setting. But these patients will oftentimes have this.

Then there’s the other, probably largest group of individuals, which are other solid tumors that can range from very common cancers like lung cancer, standard breast cancer, thyroid cancer, and colorectal cancer. There’s not a clear defined picture of who these patients are. With that said, I think it tends toward patients who may be a little bit younger, tends to be patients who if you had standard mutational testing, for example, in lung, they don’t have EGFR mutations. These patients don’t have ALK/ROS translocations.

They’re really truly driven by NTRK fusions. So I think that’s something to keep in light of, that these patients who community oncologists see and they may…have an in-house mutation set of analyte assays and no mutations emerge. It may be worth sending it off, if they don’t have an NGS [next-generation sequencing] platform in-house, to other places like Foundation Medicine, places like Caris Life Sciences, to really see if there is a possibility of an NTRK fusion. Those are really the 2 kinds of populations. Within that population of other common tumor types, there are nuances and details that are probably more detailed than is necessary in this presentation. But in general, those are the patients who we should be looking for.

I recognize in the community not everybody is able to be tested, and it goes back to then identifying these patients as I described. But I think many of my colleagues, including some of the significant authors on the papers that we’ve published and the study chairs such as Dr David Hyman, Dr Shivaani Kummar, and Dr Alexander Drilon MD, would argue that we should do our best to try to test patients, particularly all cancer patients if they have metastatic cancer, not only for larotrectinib, not only for NTRK. This ASCO [American Society of Clinical Oncology] has shown that increasingly, we are seeing more and more effective targeted therapies. Lung has multiple targeted therapies now, and really these large multiplex NGS platforms are really the best and most efficient way to identify these sometimes life-changing targeted therapies for our patients.

Transcript Edited for Clarity