Emerging Biomarkers for Targeted Agents Show Promise in NSCLC

Alex A. Adjei, MD, PhD, a professor of oncology and pharmacology at Mayo Clinic, discussed the history of five target genes: NTRK, RET, MET, BRAF, and HER2, as well as the exciting data coming out for the lung cancer subset.

Alex A. Adjei, MD, PhD

As advancing technologies and novel agents continue to expand treatment strategies in non—small cell lung cancer (NSCLC), the quest to develop more effective therapies targeting NTRK, RET, MET, BRAF, and HER2 is underway, said Alex A. Adjei, MD, PhD.

"In some lung cancers, there are abnormalities in the genes controlling proteins," said Adjei. "We have drugs that can target these abnormalities. Patients have a significant response to these treatments and live much longer. It is a big advantage over standard chemotherapy or, in some cases, immunotherapy.”

In an interview during the 2019 OncLive State of the Science Summit on Non—Small Cell Lung Cancer, Adjei, a professor of oncology and pharmacology at Mayo Clinic, discussed the history of these five target genes as well as the exciting data coming out for the lung cancer subset.

OncLive: What recent data have demonstrated effective therapies against NTRK?

Adjei: There are a number of agents [available] against NTRK, but two of them have the most data. Larotrectinib (Vitrakvi) is quite significant, in that it is the first tumor-agnostic drug approved—and it is approved for all ages. Traditionally, we test these drugs in older patients and study the pediatric population following approval; however, the NTRK gene has fusions in childhood and adult malignancies. Also, the response rates were impressive. If you take all of the patients together, there was a response of about 80%. It is a new paradigm.

Entrectinib (Rozlytrek) is a very similar drug; the data coming out are similar numerically, though the initial response rate is a little bit lower than larotrectinib. We believe part of the reason for this is that larotrectinib, in this setting, had a lot of pediatric patients who had a very robust response. Entrectinib was [tested] just in adults.

This is exciting but I have to say, in lung cancer, these gene fusions are rare. If you do not do a broad sequence, it is going to be cost prohibitive. If you take the tumor and look for NTRK fusions, most of your samples are going to be negative.

What is the status with BRAF as a marker?

BRAF is an old gene; BRAF mutations were one of the first to be discovered in the early part of genomics. The discovery led to the [successful] BRAF inhibitor vemurafenib (Zelboraf), but all of that information was for melanoma. It was much later that we found BRAF mutations in lung cancer.

We found that if you combine a MEK inhibitor with a BRAF inhibitor, you get much higher responses than with a BRAF inhibitor alone. BRAF inhibitors alone show a response rate of about 30% to 33%. When you combine trametinib (Mekinist) and dabrafenib (Tafinlar), the response rate is 66% and it works in first-line and second-line therapy.

Toxicity [from the combination] does need to be watched carefully. Although most of the toxicities are common with kinase inhibitors, diarrhea, nausea, skin rashes, and high fever are troublesome. Normally, nonsteroidal medications do not work [so steroids are prescribed to lower fevers]. Dose reductions do not work either, so you actually have to treat the fever. Interestingly, we are finding that colchicine (Mitigare; Colcrys), a drug that is used to treat [patients with] gout and familial Mediterranean fever, actually works quite well.

What data exist with RET-targeted therapies?

RET is also a fusion gene with mutations that has been described for a long time, mainly in thyroid cancers. [Originally], it was thought drugs that would affect the RET protein did not exist, but now we have a couple. Once we found fusions in lung cancers, we began testing with agents that had activity against RET. A number of them have been approved for other diseases, but not lung cancer such as vandetanib (Capresla), cabozantinib (Cabometyx), sorafenib (Nexavar), and regorafenib (Stivarga).

Initial studies were disappointing because we are used to the oncogenic driver tumors having response rates in the 60% range. Early studies with vandetanib [elicited] response rates between 15% and 20%, maybe even 30%. These drugs are not very good RET inhibitors.

In the last few years, we have selected potent RET inhibitors, such as LOXO-292 [selpercatinib] and BLU-667. Both are giving us response rates in the 60% range for RET-aberrant tumors.

Finally, what agents are showing activity against HER2?

HER2 is obviously an old gene because we have known about HER2 amplification in breast cancer for a long time. In lung cancer, we have tried trastuzumab (Herceptin), but it does not work very well. In the last few years, with next-generation sequencing, we found that there are a number of mutations in HER2 that are naturally shared with EGFR.

The most common and most difficult to treat are inserted in exon 20 of HER2 and EGFR. Initially, we treated these with pan-HER inhibitors, such as afatinib (Gilotrif) and dacomitinib (Vizimpro), but they have modest activity. We also tried with trastuzumab/pertuzumab (Perjeta) and ado-trastuzumab emtansine (TDM1; Kadcyla) [to no avail]. In the last few years, there are a couple of kinase inhibitors that show more activity. Right now, poziotinib has the most data. It has a lot of activity but a fair amount of toxicity—skin rash, diarrhea, and fatigue.

An antibody-drug conjugate [fam-] trastuzumab deruxtecan (DS-8201), has also been tested. It is like T-DM1, but the toxin attached to the antibody is different. It seems to have broad activity in HER2 aberrant tumors where there is expression of mutation.

Is it feasible to test pan-cancer microsatellite instability via liquid biopsy?

Liquid biopsies are certainly going to be the future. Taking a blood sample is very easy, so it will be easy to do across the board, particularly with lung cancer where it is difficult to get tissue. At this point, the gold standard is still tissue, but liquid biopsies are getting better and better.

Liquid biopsies are quite useful when looking for a particular gene aberration. One study I discussed involved the MET inhibitor tepotinib. In this study, [clinicians] looked at patients who had the MET aberration identified by liquid biopsy versus tissue; the results were about the same.

It is going to be the future, but [a caveat] is that some patients who receive EGFR inhibitors on progression have conversion to small cell lung cancer. For that, you are going to need tissue to tell [if that is happening] and the liquid biopsy won't [show that]. With more research, we might be able to address those issues and it may become the standard in the future.

Is tumor mutational burden (TMB) an effective biomarker?

At this point, we don't know for sure if TMB is an effective biomarker. We have a lot of different assays with different ways of measuring it. Also, part of the problem is that we do not have a uniform definition of what high-TMB is. Tumors with TMB >50 [mutations/megabase] tend to correlate very well with efficacy, but a lot of those patients have other biomarkers we are using.

At this point, it is still early. We are trying to sort out how to measure TMB and what a reasonable cutoff might be.

Is there anything else you would like to mention?

On a broader level, one thing we don't talk enough about is that will all this technology, exciting data, and improvements, how many patients in the United States are actually getting next-generation sequencing? Is insurance paying for them? If not, is it affordable?

If we go globally, the majority of the world population with lung cancer is not getting these tests and new drugs because they are too expensive. We don't talk about this enough. We get focused on the research and the exciting results, but we don't often stop to ask ourselves how many patients are benefitting from them? I don't have any answers, but we have a blueprint. It will make the whole story complete, more satisfying, more ethically sound, and it will be good for all of us.