Benjamin Levy, MD: Let’s move on and focus briefly on EGFR exon 20 mutations. Lyuda, I’m so confused. Exon 20 mutations are not exons 19 and 21. They don’t seem to be as sensitive to TKIs [tyrosine kinase inhibitors] as the exon 19 and 21 mutations. They’re challenging to treat. To make things even more complicated, all these new therapies that have come out to be offered or tested in the exon 20 mutation patient population. Can you briefly walk us through some of these trials and some of the strategies that are ongoing?

Lyudmila Bazhenova, MD: Sure. We just finished discussing all the data on what we call the EGFR-sensitizing mutations. In the EGFR pathway, the type of mutation matters. If you look at all the EGFR mutations, you can divide them into groups. There is the sensitizing mutation, which is either traditional deletion 19 or L858R. You can also look at the atypical mutations, which are mutations in exon 18, in general. Then you have your patients in exon 20.

A mutation in exon 20 that you might be familiar with is the T790M point mutation, which is also in exon 20. We have these emerging data on the ability to drug that EGFR exon 20 insertion. When you look at the molecular pathology report, the phrase EGFR mutations detected is no longer valuable. You have to know what EGFR mutation was detected to make your decision.

As Ben correctly pointed out, traditional EGFR TKIs, both first generation and second generation, usually have minimal activity in patients with EGFR exon 20 insertions with a response rate in the single digits at about 3% to 8% and median PFS [progression-free survival] of about 2 to 2.9 months. At the first restaging, the patients will usually progress.

At ASCO [the American Society of Clinical Oncology Annual Meeting], they have updated us on 3 compounds. One is called amivantamab. It’s easier for me to call it JNJ-372, and I’ll continue to do that for the duration of the talk. JNJ-372 is a bispecific antibody against EGFR and MET. It works by inhibiting the signaling through the EGFR receptor and then downmodulating the receptor. It also creates some immune-directed antitumor activity.

There was a large study, and cohort D of that study specifically looked at EGFR exon 20 insertions. Overall, about 2% of your patients will have an EGFR exon 20 insertion, so it’s a pretty sizable population. It is very similar to what you expect for RAT, ROS, or MET.

There was a small number of patients in that cohort—only 50. Thirty-nine were response evaluable. The majority of the patients were previously treated with platinum doublets. There’s a second-line-and-above study. The overall response rate was 36%. That is much better than what you would expect from a traditional EGFR TKI. The clinical benefit rate was about 67%. Responses were durable, at about 10 months’ median duration of response. Toxicities were pretty good. The majority of the toxicities were grades 1 and 2. The majority of the toxicities were infusion reactions, which you would expect with antibodies—rash and paronychia. There were almost no grade 3 toxicities. Currently, this drug has an FDA breakthrough designation for patients with EGFR exon 20 insertions.

The second drug that was presented was osimertinib from the ECOG-ACRIN [Cancer Research Group] study, EA5162. We desired to study osimertinib because we believe the first- and second-generation TKI possibly did not work because we cannot increase the dose high enough to inhibit the receptor without running into our traditional EGFR toxicity.

Osimertinib is definitely better tolerated than your first- and second-generation EGFR TKI. The idea was to give osimertinib at a higher dose, of 160 mg, and see if that has an ability to inhibit the EGFR exon 20 insertion.

The results were a bit mediocre, in my view. Only 15 patients were looked at in the second line and above. The overall response was 24%. The disease-control rate was 82%, and median PFS was 9.6. Using 160 mg is still very tolerable and toxicities were as expected, without many grade 3 toxicities. It’s a little less effective than JNJ-372 but certainly more effective than a traditional EGFR TKI.

The last 1 is the abstract that gets my “most disappointing abstract of ASCO” title, and that is poziotinib. It looked at 115 patients, second line and above, with EGFR exon 20 mutation. Overall response was 14%, disease-control rate was 68%, and median PFS was 4 months. Twenty-five percent had grade 3 diarrhea, and grade 3 diarrhea is something that’s not compatible with life, in my opinion. Twenty-eight percent had grade 3 rash, and 9% had grade 3 stomatitis. With grade 3 stomatitis, your patient is unable to eat. To me, this doesn’t have a place, at the current dose, in the patient with EGFR exon 20 insertion.

What was interesting in that abstract is that the efficacy of poziotinib depends on where your insertion is. Loop insertions did better than helical insertions. Eventually they might even subdivide the EGFR insertion patients more. Even though it wasn’t assigned to me, we cannot talk about EGFR exon 20 insertions without mentioning TAK-788, which evaluated mobocertinib. The last time we heard an update on that trial was ASCO 2019. It’s also an EGFR TKI but differently designed, and it showed a response rate of 43%. For the EGFR exon 20 insertion, our 2 front-runners are JNJ-372 and TAK-788. Both of them have FDA breakthrough designation for that disease.

Transcript Edited for Clarity