Helena Yu, MD:
The third-generation EGFR TKI osimertinib is approved for the first-line treatment of adult patients with metastatic NSCLC whose tumors have EGFR exon 19 deletions (ex19del) or exon 21 L858R mutations.¹ Although osimertinib monotherapy is the global SoC for patients with EGFR-mutated advanced NSCLC, there is significant interest in finding osimertinib-based combinations that might help extend the time of disease control. Prior combination studies with earlier-generation EGFR TKIs such as gefitinib plus CT did show a significant PFS benefit in patients with EGFR-mutated advanced NSCLC.^2,3 More recent data from the phase II OPAL trial suggested that the combination of osimertinib plus platinum-based CT might further enhance antitumor activity in the first-line setting with our current SoC EGFR TKI.⁴

The phase III FLAURA2 trial, one of the most eagerly anticipated studies at WCLC 2023, looked at first-line osimertinib with or without platinum-based CT in 557 patients with EGFR-mutated advanced nonsquamous NSCLC.⁵ Treatment-naive patients with advanced nonsquamous NSCLC harboring EGFR ex19del or L858R mutations were randomized 1:1 to receive either osimertinib monotherapy or osimertinib plus platinum CT (cisplatin or carboplatin with pemetrexed). Of importance, patients with stable untreated central nervous system (CNS) metastases were allowed to enroll. Patients stayed on study until progression, toxicity, or investigator decision. The primary endpoint of the study was investigator-assessed PFS, with key secondary endpoints including ORR, duration of response (DoR), DCR, OS, second PFS, health-related quality of life (QoL), and safety.

Helena Yu, MD:
FLAURA2 met its primary endpoint of investigator-assessed PFS with a clear, clinically meaningful improvement in PFS for patients treated with osimertinib plus CT compared with the osimertinib monotherapy arm (25.5 months vs 16.7 months; HR: 0.62; 95% CI: 0.49-0.79; P <.0001). A similar benefit was observed across all subgroups assessed, including EGFR mutation type and CNS metastases status at baseline. I thought it was very interesting that patients with CNS metastases at baseline seemed to have a more pronounced PFS benefit with the combination vs monotherapy (24.9 months vs 13.8 months; HR: 0.47; 95% CI: 0.33-0.66) compared with those without (27.6 months vs 21.0 months; HR: 0.75; 95% CI: 0.55-1.03).

Helena Yu, MD:
Response rates were similar with osimertinib plus CT vs osimertinib alone (83% vs 76%; adjusted odds ratio: 1.61; 95% CI: 1.06-2.44). Median duration of response was 24.0 months vs 15.3 months, respectively.

Of interest, immature OS data did not show a clear difference between the 2 arms (HR: 0.90; 95% CI: 0.65-1.24; P = .5238).

Helena Yu, MD:
As might be expected with combination therapy, there were more grade ≥3 adverse events (AEs) (64% vs 27%) and those that led to treatment discontinuation (48% vs 6%) or death (7% vs 3%) with osimertinib plus CT compared with osimertinib alone. Of importance, 76% of patients in the combination arm did complete the 4 cycles of induction platinum-based CT, and most of the treatment discontinuations were related to pemetrexed maintenance therapy (43% of patients). 

Helena Yu, MD:
Furthermore, AEs consistent with those expected for osimertinib were similar in both groups, and in the combination arm, there were more AEs traditionally attributed to CT, such as cytopenias.

Helena Yu, MD:
Dr Gubens, I am interested to know your initial impressions of this study. What did you expect, and did it meet those expectations?

Matthew Gubens, MD, MS:
The excellent efficacy of osimertinib along with its tolerable safety profile sets a high bar. Furthermore, many trials have evaluated TKI combination therapy over the years, and although many have shown a PFS benefit vs TKI monotherapy, very few have achieved an OS benefit. Because adding CT on top of osimertinib adds significant toxicity, I await—as I am sure many do—mature OS results to truly reveal whether sequencing osimertinib followed by CT is inferior to combining them up front.

Helena Yu, MD:
I agree that many of us are waiting to see a survival benefit before adding this treatment regimen into our clinical practice. What is really interesting about FLAURA2 is that we are combining 2 active treatments that often are sequenced as first-line and second-line therapy. This is compared with combinations such as the addition of a VEGF inhibitor, which is not active as a single agent, to EGFR TKI therapy. For this reason, we do not just want to see additive PFS with increased toxicity, but rather a synergistic PFS improvement and/or a clear improvement in OS. Although the data are immature and we obviously do not want to make too much of it, there was not a survival benefit with this initial look.

Matthew Gubens, MD, MS:
I am also curious if there is a subset of patients for whom the combination might have more benefit, and I think it will be key in the future to figure out strategies to select these patients. As you pointed out, those with CNS metastases at baseline had a more pronounced difference in median PFS between the 2 arms compared with those without. However, there are patients with EGFR-mutated advanced NSCLC who receive early stereotactic radiosurgery who end up having long survival with first-line osimertinib, and perhaps those would be patients for whom the CT would just add more toxicity without benefit.

Dr Yu, you are leading a really important trial (NCT04410796) evaluating clearance of EGFR ctDNA as a predictor of therapy benefit in patients with EGFR-mutated NSCLC that might be a better way to select a high-risk patient population with enhanced benefit to more intensive therapy than any baseline characteristic.

Helena Yu, MD:

I also agree that patient selection is key and that ctDNA clearance may be a very important biomarker for this purpose. In the FLAURA trial, 20% to 25% of patients show persistent detectable EGFR ctDNA 3 weeks after starting osimertinib monotherapy, and this group has a much shorter PFS and OS with osimertinib monotherapy.⁹¹ As you suggested, adding CT to osimertinib might be particularly beneficial in this group because they do not do as well with monotherapy. Our ongoing trial is looking to escalate care for patients with persistent EGFR ctDNA after osimertinib initiation. I also want to mention other subgroups that might benefit from treatment escalation: patients with co-occurring mutations in TP53 or RB1. We also know that atypical EGFR mutations that were not included on this study sometimes do worse with single-agent EGFR TKI therapy, so combining 2 active therapies may make sense in this group, as well—but we really do not know. We will need to officially evaluate in clinical trials whether these higher-risk subgroups are the ones that actually have more benefit with the combination.

Matthew Gubens, MD, MS:
There are also trials evaluating other EGFR TKI combinations in this setting, which if approved will give us multiple options and make finding a way to choose the optimal subgroup of patients for each strategy all the more important. For example, at the 2023 European Society for Medical Oncology Congress, we just saw positive results for the phase III MARIPOSA trial evaluating first-line amivantamab-vmjw, a bispecific EGFR-MET antibody, in combination with lazertinib, a third-generation EGFR TKI.⁶

Helena Yu, MD:
That is a great point, and I would say that one interesting biomarker for the amivantamab-vmjw plus osimertinib combination is MET dependency. There might be patients who benefit from amivantamab-vmjw plus lazertinib instead of CT plus osimertinib.

That said, Dr Gubens, based on the FLAURA2 data so far, if osimertinib plus CT was approved by the FDA and reimbursed, is there anyone in your practice to whom you would give this regimen?

Matthew Gubens, MD, MS:
Perhaps upon discussion with a patient who has multiple issues, such as CNS metastases, particularly bulky disease, and maybe a co-occurring mutation such as TP53, and who is inclined to be particularly aggressive in the treatment of their disease, I would consider it. However, I will not be using it routinely in my clinic yet. How about you?

Helena Yu, MD:
I agree. If reimbursed and approved, I think these data are a patient discussion point, and there may be certain patients who would be amenable to treatment escalation like this. However, there are also patients who are not interested in IV therapy that requires them to come see me more frequently. I think balancing QoL, toxicity, and efficacy, as well as patient wishes, is going to be key for choosing who gets which of these first-line treatments.

Helena Yu, MD:
Next we will talk about patritumab deruxtecan, a HER3-directed ADC composed of a human anti-HER3 IgG1 monoclonal antibody linked to a high-potency topoisomerase I inhibitor payload,⁷ in the setting of previously treated EGFR-mutated advanced NSCLC. EGFR-mutated advanced NSCLC is a subgroup of lung cancers that appears to have enrichment for HER3 expression, with almost all having some degree of HER3 expression.⁸ An initial phase I dose-escalation/dose-expansion study suggested safety and efficacy for patritumab deruxtecan in patients with EGFR TKI–resistant, EGFR-mutated advanced NSCLC.⁹  

At WCLC 2023, I was fortunate to present data from the registrational phase II HERTHENA-Lung01 study evaluating single-agent patritumab deruxtecan in patients with EGFR-mutated (ex19del or exon 21 L858R mutations) advanced NSCLC who had received prior EGFR TKI and platinum-based CT.^8,10  

The initial study design randomized patients to 2 cohorts: one receiving a fixed dose of 5.6 mg/kg of patritumab deruxtecan every 3 weeks and one with an uptitration of patritumab deruxtecan from 3.2 mg/kg to 6.4 mg/kg on Day 1 over 3 cycles. However, soon after the study started, emerging phase I data suggested the fixed 5.6-mg/kg dose to be the most appropriate, resulting in the prompt closure of the uptitration arm.

The primary endpoint of this study was confirmed ORR by blinded independent central review (BICR), with a key secondary endpoint of DoR by BICR.

Helena Yu, MD:
In total, 225 patients received patritumab deruxtecan at the 5.6-mg/kg fixed dose.¹⁰ Patients were heavily pretreated (median of 3 prior lines of systemic therapy), and the vast majority (93%) had received a prior third-generation EGFR TKI. All patients had received platinum-based CT, 40% had received prior immunotherapy (IO), and 51% had a history of CNS metastasis at baseline.

Helena Yu, MD:
In all patients treated with patritumab deruxtecan, the confirmed ORR was 29.8%, with a median DoR of 6.4 months, a median PFS of 5.5 months, and a median OS of 11.9 months. The forest plot of confirmed ORR suggested a similar benefit among most evaluated subgroups.

Helena Yu, MD:
The waterfall plot shows that the majority of patients had tumor shrinkage with patritumab deruxtecan and that this was regardless of whether the mechanism of resistance to EGFR TKI therapy was EGFR dependent, EGFR independent, or unidentified. 

Helena Yu, MD:
As ADC therapy enters the therapeutic landscape for NSCLC, we are interested in trying to identify an appropriate biomarker for selection of patients who are most likely to benefit from these treatments. For the HER3-targeted ADC patritumab deruxtecan, HER3 is the most obvious candidate biomarker to evaluate. 

In HERTHENA-Lung01, HER3 membrane H-score was assessed in all treated patients, with results suggesting no difference in HER3 membrane H-score between those who responded to patritumab deruxtecan and those who did not.

Helena Yu, MD:
In a population where approximately one half of patients end up developing CNS metastases, as is the case with NSCLC, it is important for us to assess the intracranial activity of these novel therapies. Intracranial responses were assessable in 30 patients with measurable brain metastases at baseline who had not received prior radiotherapy. In this small subset, the intracranial response rate was 33.3% (95% CI: 17.3%-52.8%), which was quite similar to the systemic response rate. This result suggests that patritumab deruxtecan has reasonable CNS efficacy.

Helena Yu, MD:
When assessing toxicity of an investigational agent, among the most important criteria are TEAEs that lead to treatment discontinuation. In HERTHENA-Lung01, the rate of TEAEs leading to treatment discontinuation was relatively low (7.1%).

ADC therapy often is associated with AEs common to CT, including cytopenias, nausea, and alopecia, which were among the most frequent AEs observed with patritumab deruxtecan. Another key AE that is a class effect of ADCs is interstitial lung disease (ILD)/pneumonitis, which was adjudicated as treatment related by an independent review committee in 5.3% of patients on HERTHENA-Lung01. Although this AE is a potential issue with patritumab deruxtecan, it was not as common as seen with some of the other ADCs (eg, 26% with fam-trastuzumab deruxtecan-nxki).¹¹

Helena Yu, MD:
I will turn it to Dr Gubens and ask, what are your thoughts about this data, and can you envision patritumab deruxtecan being used in this patient population?

Matthew Gubens, MD, MS:
First of all, congratulations, Dr Yu. This is such an impactful study, and I look forward to the potential approval of this drug because it really meets an unmet need for treatment options in the later-line setting for EGFR-mutated advanced NSCLC. We have very strong third-generation TKIs in the frontline setting, and some patients with EGFR TKI–resistant disease respond to CT, but there is a dearth of options after that point.

We often encourage biopsies at progression on osimertinib, but there are not many actionable targets, which makes the potential of patritumab deruxtecan particularly promising in that this trial showed it does not require a biomarker and thus will be available to a broad swath of patients after CT failure. It is also relatively safe: The deruxtecan payload has not led to undue pneumonitis, and the discontinuation rate was lower than what I would expect for CT. Furthermore, in a patient population where brain metastases are a significant issue, there was meaningful intracranial response, despite these being bulky molecules. All told, I think patritumab deruxtecan is going to be a really important third-line or later option for our patients with EGFR-mutated NSCLC.

Helena Yu, MD:
I agree with you, and I am excited about all of the ADCs under development in this space. Because there are no approved targeted therapies for use after osimertinib, we eagerly await more options. For our patients with EGFR-mutated NSCLC, the HER3-directed patritumab deruxtecan has been most extensively assessed, but TROP-2–directed and MET-directed ADCs also are under development, so at some point we may have various options for our patients.

The biomarker question that you brought up is very interesting because we are so used to having biomarkers to select patients with driver mutations. The data thus far suggest it is different with these ADCs, where you may need only a little protein expression for them to be effective. As long as it is proven with rigorous testing, I think it will be a benefit for these ADCs to be applicable to a larger population in a biomarker-agnostic manner.

Helena Yu, MD:
MET amplification is one of the most common mechanisms of acquired resistance in a subset of patients with EGFR-mutated disease who have been treated with first-line osimertinib.¹² In recent years, EGFR and MET inhibition has been a strategy under investigation for overcoming acquired resistance to osimertinib in EGFR-mutated NSCLC.¹³

The phase II INSIGHT 2 trial evaluated combining the selective MET inhibitor tepotinib with osimertinib in patients with MET-amplified, EGFR-mutated advanced NSCLC following progression on first-line osimertinib.¹⁴ MET amplification was identified either by fluorescence in situ hybridization (FISH) in tumor biopsy and/or by next-generation sequencing (NGS) in blood.

The primary endpoint was ORR by independent review committee (IRC) in patients with MET amplification detected by FISH in tumor biopsy. Key secondary endpoints included ORR in patients with MET amplification detected by blood-based NGS, DoR, PFS, OS, health-related QoL, safety, and biomarkers.

Helena Yu, MD:
This was a sizable study with 128 patients, with 59.4%, 34.4%, and 6.3% having an EGFR ex19del mutation, an EGFR L858R mutation, or other EGFR mutations including L861Q. Approximately one third of patients had brain metastases at study start.

Helena Yu, MD:
After treatment with tepotinib plus osimertinib, the ORR in patients with MET amplification identified by FISH in tumor tissue biopsy was 50.0% (95% CI: 39.7%-60.3%), which is quite significant in my opinion.

In this cohort, median DoR was 8.5 months (95% CI: 6.1 to not evaluable [NE]), median PFS was 5.6 months (95% CI: 4.2-8.1), and median OS was 17.8 months (95% CI: 11.1-NE). The majority of patients receiving this combination did have tumor shrinkage, as evidenced in the waterfall plot.

Helena Yu, MD:
When looking at the forest plot, it appears that the majority of subgroups did have some benefit with the combination of tepotinib plus osimertinib.

Helena Yu, MD:
As mentioned above, understanding intracranial activity of new therapeutic strategies in NSCLC is of paramount importance. In 24 patients with measurable, evaluable brain lesions, the intracranial ORR was 29.2% (95% CI: 12.6%-51.1%), with an intracranial DCR of 79.2% (57.8%-92.9%) and intracranial PFS of 7.8 months (95% CI: 3.9-NE).

Helena Yu, MD:
We have a pretty good sense of the individual toxicity profiles for tepotinib and osimertinib and did not see an increased safety signal or any new toxicities with the combination. As might be expected, we did observe MET-dependent in-class AEs, likely attributable to tepotinib, including edema (any grade: 40.6%; grade ≥3: 4.7%) and hypoalbuminemia (any grade: 18.0%; grade ≥3: 0.8%).

Helena Yu, MD:
Dr Gubens, I am interested to hear your thoughts about INSIGHT 2 in particular and the combination of EGFR and MET inhibitors in general. But before you answer, I would like to call out that the definition of high MET amplification has historically varied, depending on the MET detection assay (ie, FISH or NGS), type of biopsy (ie, blood or tissue), and cutoffs used to identify eligible patients, with high MET amplification ranging from as low as 5% to as high as 30% in clinical trials.^15,16

Matthew Gubens, MD, MS:
As you alluded to above, this story has gone on for a while; we have seen data demonstrating a benefit with MET TKIs (savolitinib and capmatinib) and the aforementioned EGFR-MET bispecific antibody amivantamab-vmjw when combined with an EGFR TKI therapy in patients with advanced EGFR-mutated NSCLC and MET amplification.^17-20 INSIGHT 2 is a good study that adds to this story, and tepotinib plus osimertinib is an approach I would use in a patient with MET amplification, however we define it, rather than moving on to a nontargeted treatment.

Moreover, I think these data are yet another reason for us to seriously consider a biopsy at the point of progression on an EGFR TKI. However, your point about the challenges of defining high MET amplification is on target—it can be difficult for those of us who care for these patients to read a report and know when the addition of a MET-targeting agent, whether it be a TKI or monoclonal antibody, will benefit a given patient. My hope is that clinical guideline organizations collaborate with diagnostic companies to help end users of the reports understand when it is useful to give these drugs.

Dr Yu, how do you approach MET amplification, and how do you explain it to your patients and physicians who consult with you?

Helena Yu, MD:
I agree, and I have used the combination of tepotinib plus osimertinib off-label for numerous patients. I think there is a clear appetite for patient- and tumor-directed specific recommendations in the setting of osimertinib resistance, and I like it when there is a clear target to aim for, as is the case with MET amplification. Furthermore, patients like the MET TKI plus EGFR TKI combinations because they are all oral and relatively well tolerated, which of course is desirable.

Like you, I also always biopsy people after first-line osimertinib. There is a significant percentage of histologic transformation that requires tumor tissue for detection, and I also always look for MET amplification, which, when I find it, I target because I believe it is a real driver of disease.

We must keep in mind that targeting acquired resistance mechanisms is a bit different from targeting the initial driver mutation. When we target the initial driver mutation and we have a good drug, we see excellent response rates and really durable disease control. This is because initial driver mutation is clonal and present in all of the tumor cells. The challenge with targeting any of these resistance mechanisms is that they are subclonal rather than clonal, so they may be present in only a portion of the tumor cells. So, although we may see decent response rates, we do not see such a durable time on treatment. That said, I would say that these challenges are not a reason not to try.