Testing in NTRK Fusion-Positive Lung Cancer

Luis Raez, MD: The prognosis for a patient with an NTRK alteration is very interesting because these patients may respond, like anybody else, to palliative chemotherapy. But if you find a targetable alteration like this, it likely gives the patient an extra amount of life. For example, if you put the patient on an NTRK inhibitor, like larotrectinib or entrectinib, the patient may stay on this targeted agent for 1 year, 1½ years, or maybe 2 years. If he fails a targeted agent and we don’t yet have a second targeted agent, then the patient can start palliative chemotherapy like anybody else. The patient really has the benefit of 1 or 2 extra years of life. That’s the way I feel we can easily explain the benefit of these agents, which provide very good quality of life. That’s very important. The quality of life on these types of agents is much better than the quality of life on palliative chemotherapy.

The way we identify these fusions is not easy. There are many options. Historically, the first way we identified fusions was in breast cancer with HER2/neu testing. That was very popular. The way that we identified HER2—because it’s only 1 type of genetic alteration—was with FISH [fluorescence in situ hybridization] testing, and we did that for years. It was an expensive test. However, in lung cancer nowadays, we have a lot of fusions. NTRK fusions are not the only fusions. For example, we have ALK fusions and ROS fusions, and soon we’re going to have RET fusions with a newly approved targeted agent.

If you have a patient with lung cancer and you want to perform FISH testing, you will have to order FISH tests for ALK, ROS, and RET, and 3 more FISH tests for NTRK1, NTRK2, and NTRK3. You still have to do an analysis for other mutations, such as EGFR or BRAF. That’s why FISH testing is impossible: it requires too many probes. That’s why I think it’s not traditionally the standard for fusions. In lung cancer, it’s better to do genetic sequencing, and the standard today is DNA sequencing. We do genetic DNA sequencing, and we have been doing that for the last 3 or 4 years.

Now we need to do RNA sequencing because we have these NTRK3 fusions that are not easy to find with DNA sequencing. For that reason, in some institutions we perform the DNA sequencing, and we include the RNA sequencing as a complement, so we have a better picture and a smaller chance of missing 1 of these genetic alterations.

There is a possibility to use immunohistochemistry. Immunohistochemistry is very attractive because it’s very cheap and something that may be accessible for any hospital in the United States or in the world. However, we need to validate antibodies for each of these fusions, and there are many different barriers that make this situation more complicated. However, we already have studies analyzing a pan-TRK panel with transcriptome analysis or DNA analysis that, for example, have been published by Memorial Sloan Kettering Cancer Center. And the pan-TRK immunohistochemistry test is not so bad. Its results are very close to genetic sequencing and in some cases may be useful.

For example, if you have 1 of these rare tumors, such as mammalian salivary gland cancer or secretory breast cancer, we know that these tumors have a very high incidence of NTRK fusions. In these tumors, maybe you don’t need to obtain the whole genetic sequencing. That’s why there are many other ways to profile. But in general, for a work-up of a new adenocarcinoma of the lung, I feel it is best to continue obtaining DNA sequencing and, if possible, complement that with RNA sequencing. That would be the short answer.

Not all tests capture all fusions, and that’s part of the problem. That’s why there is not 1 test that is the only test to detect NTRK fusions. The most popular tests we use are DNA sequencing in tissue and DNA sequencing in blood. Now, we know that DNA sequencing in blood misses the capture of some NTRK3 fusions, and we also know that DNA sequencing in tissue has the same problem. That’s why we are advocating the use of RNA sequencing as a complement. That is controversial. Not everybody will agree with me. Why should we have to do double sequencing? It’s prohibitive. As I mentioned before, we need to test more than 100,000 patients with metastatic lung cancer in the United States. These are things that we have to balance when we approach these patients until we get a test that can detect all these genetic alterations.

Transcript edited for clarity.

Case: A 67-Year-Old Man With NTRK Fusion-Positive Metastatic Non-Small Cell Lung Cancer

Initial presentation

A 67-year old man presented with a 2-month history of cough and dyspnea on exertion

PMH/SH: hypercholesterolemia, never smoker

PE: right-sided wheezing on auscultation

Clinical workup

Labs: WNL

Chest X-ray showed a right-side mass ~2.5 cm

Chest/abdomen/pelvic CT showed a 2.7-cm solid pulmonary lesion in the right lobe, ipsilateral mediastinal lymph node involvement

CT‐guided core needle biopsy of the lung lesion and lymph node revealed lung adenocarcinoma, grade 3

Contrast‐enhanced MRI of the head showed a small lesion (0.6 cm); indicating CNS metastasis

Molecular and genomic testing:NTRK+, BRAF-, EGFR-, ALK-, ROS1-,KRAS-, PD-L1 0%

Stage T1cN2M1b; ECOG PS 1

Treatment and Follow-Up

Larotrectinib 100 mg PO BID was initiated; treatment was well-tolerated

Stereotactic radiosurgery of the brain was deferred due to location and increased risk of post-operative morbidity

Imaging at 2 months showed stable disease; sustained response upon follow-up

Imaging at 18 months showed decreased size of pulmonary and brain lesions

Repeat genomic testing: NTRK+