Goldberg Gives Expert Insight on EGFR-TKI Resistant NSCLC

Sarah B. Goldberg, MD

In September 2017, the NCCN updated its clinical practice guidelines to include frontline osimertinib (Tagrisso) for the treatment of patients with locally-advanced or metastatic EGFR-positive non—small cell lung cancer (NSCLC).

Although osimertinib is still only approved by the FDA in the second-line (the frontline decision date is pending), physicians are already excited about using the treatment in the frontline setting, according to Sarah B. Goldberg, MD. She adds that upfront use of osimertinib is prompting questions related to testing, sequencing, and resistance.

“This is still something that is constantly evolving, and we are still learning about the best ways to understand resistance and manage patients,” Goldberg explains.

OncLive: You are giving a presentation on resistance to EGFR TKI therapy. How has this topic evolved in just over the last year?

In an interview with OncLive ahead of the 5th Annual Miami Lung Cancer Conference, Goldberg, an assistant professor of medicine at the Yale School of Medicine and Yale Cancer Center, discussed the key points of her lecture on testing for and treating resistance to EGFR TKI therapy in NSCLC.Goldberg: There have been a lot of changes in testing for and treating EGFR TKI resistance in the last year. Many of these changes are because we are now starting to treat with osimertinib in the first-line setting, both on trial and off trial. As we start to do that, we see different mechanisms of resistance come up after first-line osimertinib, and that leads to different ways of thinking on how to overcome resistance. Also, as we are sequencing the EGFR inhibitors using first- or second-generation inhibitors first and then using osimertinib; we are learning more about mechanisms of resistance after second-line osimertinib. We are also trying to better understand how to overcome resistance there. What we are starting to understand is that resistance after osimertinib may be different after first- and second-line treatment. That has become such an important area of research.

The lung cancer community is anticipating the FDA’s decision to approve osimertinib in the frontline setting. What are your thoughts on how this potential approval could impact the landscape?

In terms of how to test for it, this is something that started several years ago. Whether to get a liquid biopsy or a tissue biopsy is still an evolving question. Many of us are doing both, and while a liquid biopsy is much easier for patients, it can sometimes limit the information that you get.We saw that the FLAURA trial, which compared first-line osimertinib versus either erlotinib (Tarceva) or gefitinib (Iressa), demonstrated a significant benefit in terms of response rate and progression-free survival (PFS) for osimertinib. Based on those results, there has been a lot of enthusiasm to use osimertinib in the first-line setting. It has become a part of the NCCN guidelines, which a lot of people look to for guiding treatment. Technically it is not yet an approved therapy in the first-line setting. An FDA approval gives oncologists a stronger backing to use a drug, and it becomes easier for patients to get through insurance. It will lead to more people using it as a first-line treatment.

If osimertinib is approved in the first-line setting, will we face sequencing challenges? If we treat with osimertinib upfront, what could you give as a second-line treatment?

It will also be useful to see more data from FLAURA, specifically the overall survival (OS) data. We had some of that information in the initial presentation and paper from the FLAURA trial, but the data for OS were still immature. Over the next few months, that data will be more important. The FDA approval is important, but so are the trial data. We learned from the FLAURA trial that using frontline osimertinib does improve PFS compared with using a first-generation EGFR TKI. If you use the first-generation EGFR TKI first, you can use osimertinib second in about half of the patients who will develop T790M. You can use osimertinib in the first-line setting, but what do you do in the second-line setting? That is the potential sequencing challenge.

Currently, we do not have an option that has been proven to be effective as a second-line targeted therapy after osimertinib. There have been some emerging data that show, depending on the mechanism of resistance, that there may be options in the targeted therapy realm after osimertinib.

For example, if a secondary mutation arises in EGFR called C797S, a first-generation EGFR TKI could potentially be effective after osimertinib. The data for that are just emerging, but it has not been thoroughly investigated. It is something that could be considered as an option, but it has not been studied in a prospective trial. After osimertinib, a standard treatment is probably chemotherapy, and a consideration could be given to immunotherapy as well, which is a whole separate discussion. After osimertinib, it involves thinking about other options besides targeted therapies.

What other resistance mechanisms occur after first-line EGFR-targeted therapy other than T790M?

There are other mechanisms of resistance that arise after osimertinib, and some of those may be able to be targeted, but that is in the realm of clinical trials. That is a great option for patients to consider after osimertinib in the first-line setting. There is a lot to learn and there are some exciting opportunities there, but again, looking at things such as MET amplification is all in the clinical trial realm. There are a lot of questions on sequencing after osimertinib. If the PFS is so far superior to a first-generation TKI, then it is important what you do second. However, if you think about OS, it might not matter what you do second because you get a much longer benefit with 1 drug over the other.T790M occurs in about half of patients who progress after treatment with first- or second-generation EGFR inhibitors. What about the other half? There are some patients who will develop MET amplification. It is a fairly small number of patients at approximately 5%. However, I believe this is a very important subset of patients because there are now trials with MET inhibitors that are looking exciting in them. It is not yet an approved therapy for the MET-amplified population, but it is something to keep an eye on and, if there is a trial available, it could be a great option for patients.

There are other patients who will develop a small cell transformation after treatment with EGFR inhibitors. Again, this is a fairly small subset of patients, but it is an important one to consider because those patients tend to act like a typical patient with small cell lung cancer (SCLC) in that their tumor can progress very rapidly—and also in that they can respond very well to small cell—directed therapy, which is typically chemotherapy with a platinum-based agent and etoposide. That is something to be aware of, and an important reason to think about doing a tumor biopsy and not just a liquid biopsy; you need to be able to do a histologic diagnosis for SCLC. It is a small number of patients, but I have seen it several times so it is something that we need to keep in mind, especially for those patients who are having more rapid progression on EGFR-directed therapies.

Liquid biopsies are being used more frequently in lung cancer. What is your approach on when to perform a plasma-based biopsy versus standard tissue?

There are other alterations that we see as well, such as HER2 amplification. What to do about those alterations is unknown, but these are all areas of ongoing research. Biopsies are really important in understanding the mechanism of resistance. Fortunately, now we have 2 options. We can get a tissue biopsy, which has been the standard forever, but now we are able to get a liquid biopsy, which is a blood test looking for circulating tumor DNA. Compared with an invasive biopsy, liquid biopsy is much easier for patients, and it often can provide all of the information that we need to treat the patient. It can give the information on the mechanism of resistance—T790M presence or not—and other things that are present in the tumor.

The issue with a liquid biopsy is that it does have a false-negative rate. In the studies looking at progression on a first- or second-generation EGFR inhibitor, the false-negative rate was around 30%—so it missed T790M about 30% of the time. That is a big number. What that tells me is that if you do a liquid biopsy as your first step, and it is negative for T790M, you should really consider doing a tissue biopsy because you do not want to miss those patients who have T790M, as they could receive osimertinib as a second-line treatment.

A liquid biopsy could be considered as a first option for biopsy, followed by a tissue biopsy. The alternative is that you could do a tissue biopsy and if it gives you the information that you need, then that is the end of it. In some ways, you could think of liquid biopsy as sampling multiple sites of resistance. It is not just getting 1 site of resistance, it’s looking at all sites at once. It has become a part of standard treatment, with the knowledge that you may have to give a tissue biopsy.

Can you discuss the FDA expanded approval of afatinib (Gilotrif) for rare EGFR mutations?

You should also think about the issue of SCLC in patients. In patients who do have rapid progression or are declining quickly, I would still consider getting a tissue biopsy as my first step because I want to make sure I am not missing SCLC. I think of tissue biopsy and liquid biopsy as complementary; I use them both in different ways and occasionally do both if I do not want to miss something in a rapidly evolving clinical scenario. They both have benefits and can both be used to analyze the mechanism of resistance in a patient. Utilizing these ensures that you are finding the right next option for the patient.About 10% of EGFR mutations are considered uncommon. That is a subset of mutations lumped together because they aren't the common exon 19 mutations or L858R point mutations. Those mutations are very heterogeneous; some of them are mutations that are sensitized to EGFR inhibitors and some are resistant. There was a recent retrospective study looking at several trials with afatinib, and those trials included some of these uncommon mutations. Investigators were able to see the efficacy of afatinib in those mutations; some of these point mutations were sensitizing to afatinib. That led to an approval for afatinib in several of those mutations.

The mutations that afatinib was not effective in were the ones that we have known about for a long time—the T790M mutation and the exon 20 insertion mutations. Afatinib and first-generation drugs are not effective in those mutations.