Dr Joshua Sabari: “Tissue Is the Issue” in NGS for EGFR Exon 20 Insertion+ NSCLC

Joshua Sabari, MD, assistant professor of medicine, Medical Oncology, Perlmutter Cancer Center at NYU Langone Health, New York, New York, discusses topics involving next generation sequencing (NGS) to develop molecular profiles in the treatment of non-small cell lung cancer (NSCLC). When it comes to treating patients with the relatively uncommon EGFR Exon 20 insertion positive mutation in NSCLC, more options are available than in the past. Sabari spoke with The American Journal of Managed Care® (AJMC®).

AJMC®: How has molecular profiling and next generation sequencing (NGS) evolved in the treatment of NSCLC, particularly with Exon 20 mutations?

Sabari: With molecular profiling, broadly speaking, broad panel NGS of both tissue and plasma has been instrumental in identifying patients with driver alterations who may benefit from match targeted therapy. There are now 8 or even 9 different FDA-approved alterations with match targeted therapy. It’s important to not think only of the EGFR and the ALK patient population, which is high on the pretest probability for many clinicians. We also should think of the less common alterations; instead of ordering piecemeal testing, PCR-based assays or single gene testing, it's critical to think about a broad panel. Because tissue is the issue—and oftentimes we run into insufficient tissue when doing these piecemeal tests. It's now common practice to order broad panel NGS to identify all the different alterations.

So, EGFR exon 20 insertion mutations are uncommon. They occur in about 1% to 2% of [patients with] non-small cell lung cancer, most commonly adenocarcinoma. And we know that we miss about 50% or half of these by PCR-based assays or single gene based assays. Thus, obtaining a broad panel NGS is critical here. This is a very important patient population, because now we have 2 FDA approved match targeted therapies in the second-line setting, post-progression on chemotherapy.

AJMC®: Testing for EGFR mutations may be performed using either tissue or blood. Can you discuss the advantages and disadvantages for each of these approaches as well as clinical considerations?

Sabari: Tissue NGS remains the gold standard. It has 99% to 100% sensitivity, in my opinion, if there's adequate tissue or good cellular content. We know that tissue sufficiency remains an issue in about 10% or 15% of our patients' biopsies. But tissue NGS gives us a broad look at the tumor at the site of biopsy. Plasma NGS is interesting in that I think it complements the tissue NGS. And some of the benefits of plasma NGS is the less invasive modality. It's a simple blood test, 2 Streck tubes of 10 cc's of blood each and we get a rapid turnaround time; we generally get results in about 5 to 7 days, compared with compared to tissue NGS, which can take 21 days on average once the tissue is in hand. You can count how long it takes to get the tissue as well from other institutions or the acquiring of the biopsy. That can actually cause patient delays [of] 6 to 8 weeks. Plasma NGS has a rapid turnaround time. It also allows us to identify the heterogeneity of the tumor, whereas tissue NGS really only gives us what's going on at that site of the actual biopsy.

Finally, plasma NGS can help us understand heterogeneity or resistance over time, whereas tissue NGS is more difficult in getting that larger, bigger picture. I think one of the major drawbacks of plasma NGS is the false negative rate. And I never tell patients that their test was negative; rather, it was nondiagnostic, in the sense that if you go fishing in the ocean and you don't catch a fish that day. It’s not that there are no fish out there; there's still fish out there. That's just a nondiagnostic test, and it's important that we speak that way and we keep that language in check when talking with our patients. So, I think these 2, both tissue and plasma complement each other. I order them up front on all patients, and I generally will monitor with ctDNA plasma NGS, particularly in a patient with a driver alteration who I'm concerned about recurrence or acquired resistance alterations.

AJMC®: That makes sense. Let’s talk a bit about treatment selection. So how does the presence of an EGFR mutation influence treatment selection in NSCLC? And of course, in this case, particularly curious about the Exon 20 insertion positive mutations.

Sabari: So, EGFR mutations are quite common. I see them in about 20% to 25% of my patients. And we commonly think of the activating mutations—exon 19 deletion or L858R exon 21—that makes up about 80% of our patients. The next most common group is the atypical or uncommon EGFR mutations, probably makes up about 10% of our population. The third most common is EGFR exon 20 insertion mutations. And it's really important to know what type of EGFR mutation your patient has because activating mutations in exon 21 and exon 19 del, those patients do very well with third generation EGFR [tyrosine kinase inhibitors]. It's interesting, some of the uncommon or atypical mutations, patients might perform better with a second-generation EGFR TKI. With the exon 20 insertion mutations—these are the least common—patients do not do well with third- and second-generation EGFR TKI. Actually, there's recently some data looking at osimertinib, a third-generation EGFR inhibitor. Response rate is in that low 20% range in this patient population. And that's at double the dose of osimertinib, 160 milligrams daily. So generally, we think of this patient population as refractory or resistant to second and third generation TKI. So, understanding the landscape is important.

I commonly use chemotherapy in the frontline setting, so [that would be] carboplatinum and pemetrexed. And at the time of progression, I generally am offering amivantamab, which is an EGFR and c-MET bispecific antibody or mobocertinib, an EGFR HER2 TKI. And [I am] happy to discuss the ins and outs of those drugs, and how we think about them in the clinic.

The 2 drugs are unique; I think amivantamab has a unique mechanism of action. It binds extracellularly, which is unique for this class of agents. Also, it acts as a bispecific, inhibiting both EGFR and MET. It has immune modulating activity through trogocytosis or cellular knowing, and it also has direct ligand inhibition. I think one of the unique things about amivantamab though is that it's an infusion, it's IV and we don't commonly think about targeted therapies as infusion or IV. And one of the unique side effects of amivantamab is the infusion related reaction, which commonly occurs on cycle 1, day 1 in about 60% to 70% of my patients. Oftentimes as grade 1 or 2, patients can have flushing, can have hives, for example, or some abdominal discomfort. And if the drug is stopped or slowed down at a sort of mitigated quickly, patients are then able to tolerate the rest of the infusion or the remainder of the dose on cycle 1, day 2.

In contrast to amivantamab, mobocertinib does not have that infusion related reaction. It's an oral therapy, an EGFR HER2 TKI. And the common side effects that you see with mobocertinib are the EGFR wild type inhibition, rash, and diarrhea. So, a significant rate, [have] grade 3 [events]; about 20%, 25% have rash and diarrhea. So, I think you need to have a discussion with your patient in the office. When we look at the efficacy data for amivantamab, the response rates were 40% with a median progression free survival of 8.3 months. When we look at the efficacy of mobocertinib, the response rate was lower at 28% with a median progression free survival of 7.3 months. It is interesting to note that the duration on treatment was longer for mobocertinib at 17.5 months. So, those who are able to have a response and are tolerating the therapy do seem to benefit. I've just found it very hard to give in the clinic due to the toxicity profile. When you look at amivantamab outside of the cycle 1, day 1 infusion reaction, the rate of rash and diarrhea are quite low, in my hands, less than 5% grade 2 to 3.

AJMC®: And then just kind of expanding upon that, how would you say the treatment landscape of EGFR exon 20 insertion positive non-small cell lung cancer has evolved over time? Kind of comparing where we are today versus where we were even just few short years ago.

Sabari: Until the 2 approvals of amivantamab and mobocertinib in the second-line setting, we had no FDA-approved match targeted therapy for patients with EGFR exon 20 insertion non-small cell lung cancer. So, I think we've come a long way. That being said, we need to do more and we need to do better. When we think about targeted therapies, we're used to seeing response rates in the 80% range, progression free survival in [the range of] 18 to 20 months. And we really want to see a shift in overall survival. I think that's the key at the end of the day. So, one of the major unmet needs, in my opinion, the EGFR exon 20 space is central nervous system (CNS) metastases or intracranial disease. Neither amivantamab or mobocertinib in my hands have good CNS activity. I think we need to develop therapeutics that have better CNS activity—either newer compounds or compounds in combination with current existing therapies. It's been an exciting time in the exon 20 space. We've seen 2 new therapies that have shown data or presented data over the last 6 months, and particularly [the data for] DZD9008, which looks really exciting in some cohorts of patients, with response rates upwards of 50%. And we also saw data from an agent CLN-081 presented recently at [the American Society of Clinical Oncology] by Helena Yu, with response rate about 43%, and a very early sign of potentially seeing some CNS activity with that agent.1

Moving forward, it would be really interesting to see combinations. We have 2 new agents that are CNS specific, EGFR exon 20 TKI, the BLU-451 therapeutics agent, as well as the ORIC-114 agent. Both of these are in phase 1 dose escalation [trials], and preclinically seem to have very, very nice activities. This is a really a burgeoning field, an exciting time for our patients with EGFR exon 20. I really hope that when we have this discussion in a year or 2, we'll have many more FDA approved match targeted therapies for our patients.

AJMC®: It will definitely be exciting to see where it goes. You mentioned CNS penetration is a very large area of unmet need for this patient population. Are there any other things that come to mind when we talk about unmet need in this patient population?

Sabari: I think acquired resistance alterations is not a well-known or well understood phenomenon in this patient population. And so, for example, if we're talking about EGFR exon 19 or 21 and a patient treated with a third-generation EGFR TKI such as osimertinib, we have a well-defined resistance mutation profile. MET amplification is occurring in 10% to 15% of patients, C797S in 10% of patients; in the other alterations that we can see, a very heterogeneous group of resistance alterations are bypass tracks. In exon 20 we don't have that data yet, and I think that's a huge unmet need. Why do patients progress on amivantamab? Why do patients progress on mobocertinib? For mobocertinib we think some of it may be MET mediated, and therefore using an EGFR and MET bispecific may have increased activity. Again, that we don't have that prospective data. We're looking at sequencing these agents. So, that's a huge unmet need in my opinion, now that you have 2 FDA-approved therapies. Can one work after the other? And we're now seeing that being done in the clinic. There are no prospective data to my knowledge that exists at the moment. But I think resistance profiling and resistance mutations remain a significant unmet need for this patient population.

AJMC®: To conclude, what are you working on in EGFR exon 20 insertion positive non-small cell lung cancer that you're excited to share with colleagues?

Sabari: So, I think first off, the fact that EGFR exon 20 insertion is now as something that people think about and talk about. Prior to 2 years ago where we didn't have really good therapeutics, it was something that was really not looked for or mentioned commonly. I remember talking to a colleague who was still doing EGFR by IHC for L858R and for exon 19 deletion. And his response was, look, I do that because those are the patients I can match to targeted therapy. Now we have FDA-approved match targeted therapies in exon 20 as well as the uncommon mutations. So, it really behooves all clinicians to order broad panel NGS. I think for me, the most exciting opportunities are thinking about combination strategies of a TKI, a small molecule, particularly those with CNS activity in combination with amivantamab and EGFR c-MET bispecific. I think they complement each other mechanistically. I think the concern though is the potential for toxicity, particularly skin and [gastrointestinal] toxicity. I think that is probably something that we're going to be obviously looking into in the next year or 2, to benefit our patients with EGFR exon 20 insertion mutant NSCLC.

Reference

Yu HA, Tan DSW, Smit EF, et al. Phase (Ph) 1/2a study of CLN-081 in patients (pts) with NSCLC with EGFR exon 20 insertion mutations (Ins20). J Clin Oncol. 2022;40(16_suppl):9007. DOI: 10.1200/JCO.2022.40.16_suppl.9007