Scientific Rationale for Development of TRK Inhibitors

Marcia Brose, MD, PhD:Can you talk a little bit about how common it is to find anNTRK[neurotrophic tropomyosin receptor kinase] fusion in the presence of another targetable mutation likeEGFRorBRAF?

David Hong, MD:We don’t entirely know, but I think that right now it is unlikely if a patient, for example, has a lung cancer, or an EGFR mutation, or anALK-ROSfusion, that they’re likely going to have anNTRKfusion. We’ve not really seen that in the 153 patients we’ve now enrolled in the larotrectinib trials.

The patients who likely are going to expressNTRKfusion or haveNTRKfusion alterations are those who are not going to have a standardEGFRmutation orALK-ROStranslocations.

Marcia Brose, MD, PhD:Can you talk a little bit about the patients who might be candidates for something like PD-L1 [programmed death-ligand 1] or immunotherapy?

David Hong, MD:There is a subset of patients with MSI [microsatellite instability]—high—particularlycolorectal—that we saw in a series of patients enrolled in the larotrectinib studies that seemed to haveNTRKfusions, and there was a paper approximately 2 years ago that was published by a group in and around Spain that showed in a subset of MSI-high colorectal, the most common molecular alteration, or the most common fusion, wasNTRKfusions.

Corey Langer, MD:NTRKrepresents a unique potential target in broad-spectrum adult and actually pediatric malignancies. For the first time, we have an agent now that has been FDA approved based on its target specificity, and that’s larotrectinib. Again, based on really exceptional activity against all the variousNTRKvariants:NTRK1,NTRK2, andNTRK3.

This target is not specific for lung cancer, which is where I regularly labor, but goes across the board in both adults and kids—some unusual tumors, [including] secretory tumors of the salivary gland and the breast, pediatric tumors, congenital nephromas, infantile fibrosarcomas, GBM [glioblastoma], thyroid carcinoma, and sarcoma in both adults and children. And then in adults, lung cancer, pancreatic cancer,cholangiocarcinoma, colorectal cancer, and melanoma. As I’ve indicated, it is not confined or specific to lung cancer or any other solid organ tumor. It is not generally seen in liquid tumors, at least to the best of my knowledge, and no major identification, at least not yet, in lymphomas or leukemias.

Activation ofNTRKleads to constitutive kinase-mediated tumor growth and ultimately oncogene addiction. And on that basis, these targets likeEGFR,ALK, andROScan be potentially inhibited and can lead to the shutdown of tumor proliferation.

The NTRK rearrangements, as opposed to mutations, are generally mutually exclusive almost completely with one another and also with other typical oncogenic drivers. You don’t tend to see NTRK in the setting of ALK or ROS1, although the data are still unfolding. I would not be surprised if we encounter patients who developed resistance to some of the other more common drivers, and suddenly NTRK materializes.

But by and large, they are mutually exclusive. Remember these abnormalities are quite rare. At least in lung cancer, the incidence—depending on which series you look at—is anywhere from 0.2% to 2.7%, but mostly on the much lower side, probably about 0.3% to 0.5%. That underscores the importance of next-generation sequencing. This is not an abnormality that you would normally specifically test for. It used to be part of the much larger panel where we can really harvest small amounts of tissue optimally to get a lot more information. And in so doing, potentially identify targets that we might not otherwise been aware of.

With respect to PD-L1, I think the data are still out. We see that other oncogenic-driven tumors have varying levels of PD-L1 expression. But of note, at least so far, except for BRAF and KRAS, the checkpoint inhibitors don’t seem to work well in tumors with oncogenic drivers, certainly in EGFR, ALK, ROS1 tumors. The initial step is using a TKI [tyrosine kinase inhibitor] and then later going on to chemo at the time of progression, and only after that considering a checkpoint inhibitor. I suspect the same scenario will apply toNTRK-positive non—small–cell lung cancer.

Marcia Brose, MD, PhD:Can you describe mechanism of action and the scientific rationale for how TRK [tropomyosin receptor kinase] inhibitors were developed?

David Hong, MD:Larotrectinib and entrectinib—more so larotrectinib—are very specific for the NTRK-receptor pocket, ATP [adenosine triphosphate] pocket, and though it has a slow half-life, it binds very tightly to the actual pocket and blocks this constitutive signaling that truly drives many of these cancers.

Transcript edited for clarity.