Expert Discusses Translating EGFR Mutation Status Into Action in NSCLC

OncologyLive, Vol. 17/No. 5, Volume 17, Issue 5

Mark Socinski, MD, discusses the three generations of drugs are now available to treat patients with EGFR-mutant non–small cell lung cancer, and mutation testing for patients diagnosed with advanced or metastatic disease has been incorporated into clinical practice guidelines.

Mark A. Socinski, MD

Three generations of drugs are now available to treat patients with EGFR-mutant non—small cell lung cancer (NSCLC), and mutation testing for patients diagnosed with advanced or metastatic disease has been incorporated into clinical practice guidelines.

OncologyLive: How important are EGFR mutations in the genomic landscape of NSCLC?

Yet many questions about translating these new tools into clinical practice linger. In this interview with OncologyLive, Mark A. Socinski, MD, tackles some of these issues. Socinski is a professor of Medicine and Cardiothoracic Surgery at the University of Pittsburgh Medical Center (UPMC) and co-director of the UPMC Lung Cancer Center of Excellence. He also is clinical associate director of the center’s Lung SPORE research program.Socinski: Incredibly important. For a subset of largely adenocarcinoma of the lung, the standard of care has changed over the past 5 or 6 years as EGFR TKIs have been shown in multiple phase III trials to be superior to the former standard, which was combination chemotherapy. The population with this subtype, depending upon the patient mix in a practice, is probably between 5%-20%. By that I mean how many women, how many never-smokers, how many adenocarcinomas, how many Asians you may have.

Please briefly discuss the current testing methods and ongoing research in this area.

For that 5%-20%, lung cancer is a different disease now. It’s associated with a much better prognosis, and the median survival in many studies is in excess of 2 years, even up to 3 years. The treatment is taking a pill once a day versus the standard old-fashioned chemotherapy. So the detection and diagnosis of EGFR mutations is critically important in day-to-day practice today.There are multiple different testing methodologies, and they range from old-fashioned Sanger sequencing to next-generation sequencing and PCR-based technologies, all of which have increasing levels of sensitivity. We are also in a time where blood-based testing is a reality. There are a number of tests that look at circulating tumor DNA or circulating tumor cells to make the diagnosis of these mutations. We certainly will see more blood-based testing in the future.

What are the most significant unanswered questions and barriers to effective EGFR-targeted agents in EGFR-mutant NSCLC?

I think it’s important that physicians realize that there may be different levels of sensitivity with different techniques. If, for instance, you have a clinical situation where you have a high suspicion of NSCLC—a never-smoking female with adenocarcinoma who happens to be Asian— this is a population that might have a 50%-60% chance of having an EGFR mutation. But if you get a negative result back on Sanger sequencing, I might in that setting do next-generation sequencing on that patient just to be absolutely sure. I do think that establishing a diagnosis of an EGFR mutation is a game-changer in adenocarcinoma of the lung.I think the biggest challenge is the development of resistance. It’s very gratifying to diagnose EGFR mutations because you have a patient you can comfort by saying you have a better prognosis, and that this is a different type of lung cancer that’s driven by an oncogene for which we have a targeted therapy that is pretty effective. Response rates are usually in the 60%-80% range.

Unfortunately, it’s not a 100% response because we do have some de novo resistance issues. Patients get on oral therapy. There are toxicities that are problematic, but I think the toxicities with all the TKIs are manageable, certainly easier than chemotherapy side effects can be.

The median progression-free survival time is always in the 9- to -13-month range. So, I always think I have about a year with EGFR TKI therapy before patients are going to start to develop resistance. Certainly, we can have the experience where patients are on TKI therapy for 2 to 3 years—which is very gratifying for both doctor and patient—but what to do at the time of resistance is a key question.

A number of studies have determined that the predominant mechanism of resistance is the development of a secondary mutation in exon 20, which is the T790M mutation, in 50%-60% of patients. That means that 40%-50% of patients have other mechanisms of resistance, whether it’s MET amplification or PI3K alterations or another growth pathway. HER2 has been implicated in certain instances, as has BRAF, and histologic transformation is another issue in this population.

Could you briefly discuss the approval of osimertinib and how it adds to the treatment landscape?

Rebiopsy has kind of become the standard. So I would say that the biggest challenge and one of the biggest questions is: With the growing number of TKIs that we have available and now that we have a T790M—targeting drug, what’s the optimal way to sequence these drugs? How can you get the greatest amount of disease control with oral therapy for as long as you can?I think it’s a huge addition. Prior to the approval of osimertinib, when patients failed a TKI they essentially went to chemotherapy. We would try to string them along as long as we could on the initial TKI—we would tolerate asymptomatic progression longer than we typically would with chemotherapy, before switching to chemotherapy. The approval of osimertinib, at least in the T790M-positive population, creates another oral TKI option and, since osimertinib is a third-generation EGFR TKI, it has more selectivity for mutant EGFR as well as activity in patients with T790M alterations. It spares wild-type EGFR, so you tend to see fewer toxicities relative to the first- and second-generation inhibitors, and that’s a clear advantage. This approval really creates an option that we didn’t have prior to a few months ago.

There are other drugs in development. Rociletinib is probably the next furthest along. That certainly has activity preferentially in T790M-positive patients, but has some activity, as does osimertinib, in T790M-negative patients. So, that’s a drug we should keep our eye on. There’s another drug, HM61713, also an EGFR mutant—selective inhibitor, with some degree of activity, and that’s in development. We’re working with a Pfizer compound [PF- 06747775] that’s a T790M-specific drug. That’s certainly not a complete list, but a number of TKIs are being developed.

However, it is kind of a good news—bad news situation. For patients who seemingly don’t have the T790M resistance mutation, the path forward is a little less clear in that population. We don’t have a lot of data on strategies that specifically target the mechanism of resistance in these patients.

So in the T790M-negative population, we’re stuck with being unsure about the best way forward. Obviously, we have the option of chemotherapy, but it would be nice to have a more targeted approach. I’m hopeful that as we begin to unravel the best way forward here that we will have some better strategies.

EGFR-Targeting Therapies Approved by the FDA in NSCLC

EGFR indicates epidermal growth factor receptor; NSCLC, non—small cell lung cancer; TKI, tyrosine kinase inhibitor.

aGefitinib initially was approved in 2003 but withdrawn from the US market in 2005 after failing to improve survival in molecularly unselected patient populations in confirmatory trials.

What are the key areas of research in terms of the further development EGFR-targeting therapies?

There are some data for T790M-targeting drugs in the T790M-negative population. With rociletinib, for example, the response rates are in the 30%-35% range, which is pretty much what we might expect with chemotherapy, so you might argue that rociletinib could be a reasonable option prior to chemotherapy in the T790M-negative population. I don’t want to discount these drugs in the T790M-negative population, but it’s a much more heterogeneous population in terms of response.I think a key question is going to be what differentiates these new drugs. For example, is there activity of one T790M drug after another one or are there subsets you can define where osimertinib is preferred or rociletinib is preferred? One of the issues is, now that we have first-, second-, and third-generation drugs approved, is it possible that the third-generation drugs represent better first-line options? Those studies are ongoing, comparing rociletinib and osimertinib with first-generation drugs like erlotinib.

Are there any circumstances where you would not use a targeted therapy as first-line treatment for an EGFR-positive patient?

My personal bias is that they aren’t going to necessarily be better. When you’re comparing a drug like osimertinib to, for example, erlotinib in the first-line setting, the implication is that you’d like to replace the first-generation inhibitor with the third-generation inhibitor in this setting. You know the first- and second-generation drugs are really decent drugs, patients tolerate them, and they work in the majority of patients. Some patients can be on them for years, so I think the challenge of trying to get a drug moved from second line to first line is quite a significant one and I don’t know that it’s necessary.For EGFR-mutant disease, if the patient doesn’t have a classic exon 19 or 21 mutation, you kind of have to read the fine print. Some of the exon 18 aberrations are sensitive to EGFR TKIs and some of the exon 20 are more resistant than other mutations.

A doctor really has to make a decision in the first-line setting whether this patient is best served with a TKI or chemotherapy, and there are certain instances where, even though you have an EGFR mutation, it may not be a known sensitivity mutation. If it’s a patient who’s very symptomatic or has a heavy disease burden, I would probably favor using chemotherapy first. For instance, if you have an exon 20 insertion mutation where the response rate is very low and the patient’s not likely to respond to a TKI, that’s a situation where a patient would probably get more benefit from standard chemotherapy rather than deploying an EGFR TKI.

So, for the 10% of patients with uncommon mutations, it really bears going the extra step to make sure that there has been experience with that mutation and some evidence that a TKI is likely to be more effective than chemotherapy.