Therapeutic Advances for Metastatic NSCLC With EGFR Exon 20 Insertion Mutations

Zofia Piotrowska, MD, is a clinical researcher and lung cancer medical oncologist at Massachusetts General Hospital Cancer Center in Boston, where she cares for patients with thoracic cancers. Dr Piotrowska conducts clinical trials in both non-small cell and small cell lung cancer. She has a particular interest in understanding and overcoming resistance to targeted therapies in lung cancer, particularly among patients with EGFR-mutant NSCLC.

Trends in detection of mutations have changed over time, and the rising uptake of NGS testing of tissue and plasma samples has helped to increase detection of uncommon EGFR mutations in NSCLC, including EGFR exon 20 insertions. These are not as common as EGFR exon 19 or exon 21 mutations, but they — along with other uncommon EGFR mutations — are now being detected with the use of NGS panels.

How common is EGFR-mutant lung cancer overall? How common are the EGFR 20 exon mutations? Is there a subset of patients in whom these are more commonly detected, and is their prognosis poorer?

EGFR mutations comprise 15% to 20% of nonsquamous cell NSCLC. They are more commonly observed in light or never smokers, and studies suggest they are present in 45% to 47% of never smokers.^1,2 The exon 20 insertions are the third most common EGFR mutation that we see in lung cancer, and among the EGFR mutations overall, they make up about 4% to 10% of all EGFR-positive lung cancers.³ Some data show that they are seen in about 2% of lung cancers overall.⁴

This is comparable to some of the other rare mutations we’re used to looking for, like ROS1 and RET, for example.  It’s not a huge number, but when you see so many patients with lung cancer every year, it does add up.

The response of lung cancers with classical EGFR mutations — the L858Rs and the exon 19 deletions, for example — has been quite good when they are treated with first-generation inhibitor drugs like gefitinib and erlotinib, second-generation inhibitors like afatinib, and third-generation inhibitors, as well. But exon 20 mutated disease has not demonstrated the same response rate, and some data have shown that the response to standard first-generation drugs is about 3% to 8%.³ 

Why are these mutations so different and their response to current targeted therapies so poor? What is the standard treatment right now for EGFR exon 20 mutated NSCLC?

The difference in response with first- and second-generation TKIs has to do at least in part with the structural differences between the mutations and how they arise. Their structure is more distinct from the more common exon 19 deletions or the L858R mutation, and historically they have been more challenging to target. These exon 20 insertions occur within or just after the “C-helix” domain of EGFR exon 20 and are predicted to “push” the C-helix into an active conformation and result in steric hindrance that has made them more challenging to target therapeutically.³

Indeed, we have seen that, for the majority of EGFR exon 20 insertions, the rates of response to the first-generation inhibitors (erlotinib and gefitinib) were quite low in the original studies. Responses to the second-generation EGFR inhibitor (afatinib) have also been limited. We presented a poster last year at the American Society of Clinical Oncology® (ASCO®) annual meeting of a small study in which patients who were treated with a high dose of osimertinib (160 mg) had a 25% response rate.⁵ Median progression-free survival was 9.7 months. Although this is a small study, it’s certainly better than what we’ve seen with first-generation EGFR inhibitors, and we are hoping to expand this study to a larger number of patients soon.

The standard of care for these patients, at least as an initial treatment, is platinum doublet chemotherapy, with or without immunotherapy. There is still a great debate in the community as to whether these patients should receive chemoimmunotherapy or whether it has less benefit to the addition of immunotherapy, as we see with other EGFR mutations. This is still an open question and up to the discretion of the treating physician whether or not to add immunotherapy.

here were no approved therapies to treat patients with exon 20 mutations until very recently, when the US Food and Drug Administration (FDA) approved amivantamab for use in this patient population.⁶ The decision was based on data from the phase 1 CHRYSALIS trial (ClinicalTrials.gov Identifier: NCT02609776), making this the only approved targeted therapy to date. The overall response rate was 40%, and the mean duration of response was 11.1 months in pretreated patients with NSCLC harboring EGFR exon 20 insertion mutations. How does this compare to current treatment for these patients; that is, how much of an improvement is it? And where will amivantamab fit into the treatment regimen?

With many of these rare mutations, some of the therapies are getting provisional approval based on single-arm studies, with confirmatory studies to follow. This allows patients access to the drugs, and there are challenges to doing large studies in these populations. The response to amivantamab was 40%, which was quite good and higher than what we’ve seen previously. It was comparable to or better than chemotherapy and other targeted therapies, but it’s fair to say that it still leaves room for improvement and for other drugs to be developed in this space.

The approval of amivantamab is for patients with EGFR exon 20 insertions who have received prior platinum-based therapy, so this will be an option for the second line and beyond. I found the data to be very clinically meaningful. The drug was generally well tolerated, although there were some skin toxicities and infusion reactions, which can occur. I do anticipate that we will use it in the clinic when it becomes available. There is an ongoing phase 3 study looking at it in the first line; the PAPILLON study (ClinicalTrials.gov Identifier: NCT04538664) is assessing amivantamab given in combination with standard of care carboplatin-pemetrexed vs chemotherapy alone.7 The toxicity profile of the combination will be an important focus.

EGFR exon 20 insertions are biologically different from traditional EGFR mutations, and as such, there is a need to develop novel therapies that specifically target this mutation. The FDA has granted Fast Track designation for poziotinib, osimertinib has also reached phase 2 trials, and mobocertinib has been given Breakthrough Therapy designation. However, studies show that the overall response rates are also fairly low for all of these new agents. Is the bar being lowered, so to speak, since this is an unmet need and new agents are urgently needed? If any of them are approved, where will they fit into the armamentarium for treating this population?

There are a number of drugs in development to target EGFR exon 20 insertions. Poziotinib was one of the first of these to be developed and, while it showed promising activity initially, later studies showed a response rate of 15% in a larger cohort of patients.⁸ And the fact that mobocertinib has received Breakthrough Therapy designation from the FDA indicates that it’s a promising investigational agent. The latest data on mobocertinib showed a response rate of about 30% among patients with EGFR exon 20 insertions.⁹ Unfortunately, what we are seeing with both of these agents is that they are associated with significant side effects, most commonly skin toxicities and diarrhea, related to the inhibition of wild-type EGFR. These effects can be really challenging for patients and can lead to high rates of dose reductions and treatment discontinuation, which can limit clinical activity.

These drugs are promising, but there’s room for us to improve and develop better therapies. I think it’s too early to know which one of these drugs will eventually gain FDA approval. Mobocertinib may be next given its Breakthrough Therapy designation, but it isn’t yet approved by the FDA.

Finally, I think it is important to note that these drugs are oral targeted therapies that have a mechanism of action different from amivantamab, which is an intravenously administered, bispecific antibody against EGFR and MET. That is an important point because it’s nice to have different types of drugs available to offer our patients, and I hope this will mean that we’ll have multiple options that can be used sequentially to help extend patient survival.

Are there other novel TKIs coming down the pipeline that may be able to target EGFR exon 20 mutations? Amivantamab is a fully human EGFR-MET bispecific antibody, but are there any other antibodies under investigation at this time, or anything interesting going on in the preclinical setting?

There are several others to be aware of and keep an eye on. One is CLN-081, in development from Cullinan Pearl, which is an oral EGFR inhibitor with high selectivity for EGFR exon 20 insertion mutant vs wild-type EGFR in preclinical models. We recently presented some updated data on this agent, which is still in dose-escalation but is showing early signs of activity.¹⁰

Results were also presented at ASCO this year for another product, DZD9008, a selective, irreversible EGFR exon 20 insertion inhibitor that is being studied in 2 ongoing phase 1/2 studies and is under development by Dizal Pharmaceuticals.¹¹

BDTX-189 from Black Diamond Therapeutics is a potent, selective, irreversible inhibitor of EGFR and human epidermal growth factor receptor 2 (HER2) that is being evaluated in the ongoing MasterKey-01 trial (Clinical.Trials.gov Identifier:  NCT04209465). Early results were presented this year at ASCO.¹² I look forward to seeing the results of these ongoing trials.

Has NGS testing now become the standard over polymerase chain reaction (PCR)-based testing, and are patients with NSCLC being tested routinely before beginning therapy? Guidelines from various organizations, with some variance, recommend testing for EGFR, ALK, MET, ROS1, and so on. Has EGFR exon 20 been included in recommendations, and are patients being routinely tested for it along with the more common EGFR mutations?

I think we, as a field, have to move to broad, multigene panel testing rather than single-gene PCR tests for our patients with NSCLC. We have 9 different mutations with approved targeted therapies in 2021, but these drugs are only effective if we can identify patients with these mutations and select them for the appropriate treatment. Thus, it has become incredibly important to test all patients with NSCLC, and I believe that broad NGS testing is becoming a standard of care.

We saw some data at ASCO this year from the MYLUNG Consortium study highlighting the fact that rates of testing are still inadequate around the country, as well as some data demonstrating the significant racial disparities in biomarker testing.^{13, 14} These data really emphasize the fact that we as a community need to do better in making sure we are testing newly diagnosed patients now that we have all of these targeted therapies.