Targeting Acquired Resistance Leads EGFR+ NSCLC Research Efforts

Heather Wakelee, MD, discusses the rationale and findings for two combination regimens in EGFR-mutant non-small cell lung cancer, as well as the implications for our understanding of mechanisms of resistance.

Heather A. Wakelee, MD

Combination strategies with osimertinib (Tagrisso) are in the midst of investigations in an effort to target patients with EGFR-mutant non—small cell lung cancer (NSCLC) with acquired resistance mechanisms, explained Heather A. Wakelee, MD.

One study, which was presented at the 2019 ASCO Annual Meeting, looked at the combination of osimertinib with chemotherapy in previously treated patients with EGFR-mutant NSCLC.1 Of a group of 35 patients, all had progressed on prior osimertinib and 34 had also received a prior first- or second-generation EGFR TKI. Twenty-five of the 35 patients were T790M-positive at progression. The investigators looked at 47 possible regimens of osimertinib plus chemotherapy, with most patients receiving carboplatin and pemetrexed with or without bevacizumab (Avastin).

Results showed that the median overall survival (OS) was 48.4 months (95% CI, 30.7-not reached) from the time of metastatic diagnosis and the median OS was 19.6 months (95% CI, 16.1-32.6) since first receiving osimertinib. The median duration of treatment varied based on chemotherapy regimen.

The most common adverse events across the combination regimens was AST/ALT elevation in 34%, anemia in 71%, neutropenia in 40%, and thrombocytopenia in 66%. Researchers found that osimertinib in combination with chemotherapy was a reasonable and safe option for the treatment of patients with EGFR-mutant NSCLC at the point of progression.

In a second trial, also presented at the 2019 ASCO Annual Meeting, researchers explored the combination of osimertinib plus necitumumab (Portrazza) at the time of resistance in select populations of patients with EGFR-mutant NSCLC. The study included 4 expansion cohorts of 18 patients each, looking at patients with T790M-negative NSCLC who had progressive disease on their last therapy of a first- or second-generation TKI, patients with T790M-negative disease and progression on a third-generation TKI, patients with T790M-positive progression on a third-generation TKI, and also at patients with EGFR exon 20 insertions who had disease progression following chemotherapy.

In patients who progressed on a first- or second-generation TKI, the objective response rate (ORR) with the combination of osimertinib and necitumumab was 29%, and the ORR was 13% among patients who progressed on a third-generation TKI, consisting of responses only in patients who had T790M-positive disease. Researchers concluded that osimertinib plus necitumumab was a viable and tolerable combination for patients with EGFR-mutant NSCLC at the time of resistance.2

In an interview with OncLive, Wakelee, professor of medicine (oncology), Stanford University, discussed the rationale for these two combinations and the findings from both studies, as well as the implications for understanding EGFR mechanisms of resistance.

OncLive: Can you provide some background into the use of EGFR TKIs in patients with EGFR-mutant NSCLC?

Wakelee: For patients with newly diagnosed EGFR-mutated NSCLC, we actually have 5 FDA-approved EGFR TKIs. And we know from long ago that in the setting where there’s a known EGFR mutation, starting with EGFR TKI is definitely better than starting with chemotherapy as far as response and also disease-free or progression-free survival. The question is which of those drugs is going to be the best choice.

Initially, we had gefitinib (Iressa) and erlotinib (Tarceva), [but] we never had a head-to-head [study] on those. And then afatinib, the first of the second-generation drugs was developed, and there was a head-to-head of afatinib (Gilotrif) versus gefitinib conducted, which showed some improvement, [but there was no] statistically significant improvement in survival. Therefore, because of toxicity differences, we still didn't use a whole lot of afatinib. [However], it does have a role, and it is approved.

Then, osimertinib, a third-generation agent was developed. It initially was approved in the setting of resistance to first- or second-generation drugs. In that setting, first- or second-generation resistance is usually [caused by] the T790M [mutation]. [For patients who] have a T790M mutation, the third-generation drug osimertinib works—the others don't. So, it makes sense to substitute one for the other and switch over to osimertinib. And that was our standard practice for a while. But, of course, there was a head-to-head, osimertinib versus either erlotinib or gefitinib, and that showed a statistically significant improvement in progression-free survival—we were out at 18.9 months as opposed to less than a year. With that, [clinical] practice changed quite a bit.

There's still some debate, because, if [a patient] is on erlotinib, gefitinib, or afatinib, and they have a prolonged progression-free survival of well over a year then develop T790M, then they have to go on osimertinib post-resistance. The total amount of time [with both drugs] is equivalent to or potentially even longer that the 19 months. The challenge is that we don't know which patients are going to get T790M and only about 50% to 60% do. For the other resistance mechanisms, osimertinib is not going to be a better approach by itself. So, that's the reason why a lot of times we switch and start using osimertinib in the first line. We also have dacomitinib and dacomitinib is another drug where there was there was a clear improvement in survival compared to gefitinib in a head-to-head study, however, it's a second-generation drug. It doesn't seem to have to same brain activity. It does have increased toxicity compared to osimertinib, and it just hasn't been widely adopted despite the great data.

What was the rationale for exploring osimertinib in combination with concurrent chemotherapy in this patient population?

At ASCO, we presented a retrospective [analysis] of what people are doing in real practices in some of the academic medical centers. For patients who have developed resistance to osimertinib, there are a lot of different mechanisms of resistance that come into play. The C797S resistance mutation is one, MET amplification in other. And then there are a lot of others as well that aren't fully understood, so there’s a lot of work being done there. We've done multiple clinical trials [in an effort to understand these resistance mechanisms], there was a combination of osimertinib and necitumumab that was a poster Johnathan Riess, MD, MS, presented.

One of the things that's done in practice sometimes is, at the time of resistance, we start chemotherapy. We know from earlier data, such as the IMPRESS trial, that if someone is on gefitinib, a first-generation drug, and they progress—this was before the era of osimertinib—our only option was chemotherapy. [At that time] there was a debate [about whether to send patients] to chemotherapy or chemotherapy plus continuing the TKI, and there were lots of strong opinions there. In the end, the trial showed that if patients stayed on gefitinib at the time of chemotherapy versus if they didn’t and just got chemotherapy, [the patient] had better outcomes if you stopped.

There’s a small subset of patients—maybe 5% or so, maybe 10%, depends on the series—where there's a flare. Even though [these patients] are having resistance, there are still cancer cells that are active because of the EGFR mutation and [the cells] are suppressed because of the EGFR. Even though there are other cell clones growing, there’s still a subset being suppressed by the EGFR TKI. When you stop [the TKI], those cells start to grow rapidly, and patients become symptomatic rapidly. Again, it’s a small subset, but you know [those patients] because within a few days of stopping the drug, they feel terrible. So, for that subset, continuing the EGFR TKI along with the chemotherapy is something that we all do in practice. It’s kind of small numbers. That’s part of the group that was receiving chemotherapy.

The other subset [includes patients] who are on osimertinib and have good control of their brain disease but are starting to progress in the body. In our practice, where that’s been done the most is a patient who is doing well on osimertinib who initially had symptomatic brain metastases, those are still controlled but they’re having growth elsewhere, so it’s different clones, so we'll add the chemotherapy [to control that progression]. Our poster was showing that when you combine the 2, it's actually relatively safe. With osimertinib as opposed to the other EGFR TKIs, you do end up with cytopenias, and neutropenias and thrombocytopenias can be seen, so the concern was that would not pair well with chemotherapy. In practice, the toxicity was reasonable. Now this is not to say that people should be doing this routinely. We are not advocating that osimertinib plus chemotherapy together is the right approach, but it is now something that's going to be looked at further in clinical trials.

We also saw data at this ASCO with the combination of gefitinib plus chemotherapy from a study done in India, and last year we had gefitinib plus chemotherapy in a study done in Japan. The Japanese study was really interesting because they showed a survival benefit with that upfront combination of all of the drugs, and they did it really nicely where patients either started gefitinib or they started gefitinib plus chemotherapy. The group on gefitinib at the time of progression they went and got the chemotherapy. What they were able to show in that study from Japan is that if you looked at the progression-free survival 2 (PFS2), which was after [the patient has] had both the gefitinib and the chemotherapy, either altogether or separately, [the survival] was almost the same.

So, whether [the patient] got gefitinib and then chemotherapy or got everything altogether in the beginning, the PFS2 was the same, but the overall survival favored the combination, so the patients who gefitinib plus chemotherapy together in the beginning should benefit. It looks like the study done in India had some similar positive results. There’s also now a trial in development, osimertinib plus chemotherapy versus osimertinib alone to see if that can prevent some of the other resistance mechanisms from developing. Our poster was supporting that there’s safety to that, and that yes it’s done in practice sometimes, but usually in the setting of needing better brain control from the osimertinib and needing control elsewhere in the body from chemotherapy.

Could you discuss the ETCTN California cancer consortium phase I study?

That was a study again in the setting of EGFR-mutant NSCLC. This was a study looking at the combination of osimertinib with necitumumab. Going back in time there were data with afatinib plus cetuximab, and this was the first time we had data showing activity in the setting of secondary resistance to an EGFR TKI. This study was for patients who had been on erlotinib or gefitinib, progressed, and then got a combination of the second-generation drug, afatinib plus an EGFR antibody cetuximab. That combination was somewhat toxic as far as rash and diarrhea, but patients could get through it. It actually showed activity with or without T790M. So, for a while, [that combination] was the only thing we had at the time of resistance. Then, the third-generation drugs came along, osimertinib was there, people stopped doing that. But history repeats itself, and now when [patients] develop resistance to the third-generation drug, now there are thoughts about going back to that combination of an second-generation drug plus an EGFR antibody. So, that was the impetus for our study.

We actually had some pretty encouraging data, but it was somewhat dependent on the T790M. The study was designed in an era where a lot of patients were still starting on a first- or second-generation drug and at the time of resistance going to osimertinib. There’s a big cohort of the trial for patients that had osimertinib as their second EGFR drug. Then there’s a smaller subset for patients who were on osimertinib first line at the time of resistance, but we don’t have all of those data yet.

In a setting where patients had a first- or second-generation drug, developed T790M, and then went on osimertinib, whether or not they maintained that T790M changed the outcomes For patients who lost the T790M, we didn't see that the osimertinib with necitumumab was very active. If they retained T790M, [the combination was active]. Then there were some interesting subsets, [in patients with] EGFR exon 20 we’re seeing good responses, it’s only a handful of patients so far but EGFR exon 20 is a particularly difficult one to target and we were seeing activity there as well.

I think it's still early, but there’s some encouraging signals. There are some other studies ongoing looking at combinations of an EGFR TKI plus antibody, but at least from our California cancer consortium; we are excited about the data.

What are some unanswered questions that you hope will be addressed with future data?

We still have a long way to go in understanding EGFR resistance and also being able to overcome it. We do know that MET amplification is another big story. And, although I'm not involved in any of those studies, there are some data coming out looks pretty encouraging combining MET inhibition with EGFR inhibition in patients who have MET as their resistance mechanism. There has also been some work done in the past, not all of it positive, combining MET inhibition and EGFR inhibition de novo, starting with those treatments. So, I think we are still trying to figure all of that out. [We also need to understand] some of those other pathways and what we can do.


  1. Joel W. Neal, Daniel Hausrath, et al. Osimertinib with chemotherapy for EGFR-mutant NSCLC at progression: Safety profile and survival analysis. J Clin Oncol. 2019;37(suppl; abstr 9083).
  2. Jonathan W. Riess, Susan G. Groshen, et al. Osimertinib (Osi) plus necitumumab (Neci) in EGFR-mutant NSCLC: An ETCTN California cancer consortium phase I study. J Clin Oncol. 2019;37(suppl; abstr 9057).