REZILIENT3 Delivers Phase 3 PFS Win, But OS Silence and Amivantamab Entrenchment Define the Real Battle

REZILIENT3 Phase 3 Trial Shows Significant PFS Benefit for Zipalertinib in EGFR Exon 20 NSCLC

Taiho Oncology, Taiho Pharmaceutical, and Cullinan Therapeutics announced positive Phase 3 REZILIENT3 trial results for zipalertinib plus chemotherapy in first-line EGFR exon 20 insertion mutation-positive non-small cell lung cancer (NSCLC). Presented at WCLC 2026, the trial met its primary endpoint, demonstrating a statistically significant and clinically meaningful median progression-free survival (PFS) improvement of 6.0 months (14.5 months vs. 8.5 months) compared to chemotherapy alone (HR=0.50; P=0.00015). The combination also showed significantly higher objective response rates (65.0% vs. 40.3%) and longer duration of response.

• The REZILIENT3 trial showed that zipalertinib combined with platinum-based chemotherapy extended median progression-free survival by 6.0 months, reaching 14.5 months compared to 8.5 months for chemotherapy alone. This benefit was statistically significant (HR=0.50; 95% CI, 0.34-0.73; P=0.00015) and consistent across subgroups, including patients with brain metastases (HR: 0.38).
• Beyond PFS, the combination therapy achieved a significantly higher objective response rate of 65.0% compared to 40.3% for chemotherapy alone (P<0.0001). Patients receiving zipalertinib plus chemotherapy also experienced a longer median duration of response, at 14.2 months versus 9.9 months for the control arm.
• The safety profile of zipalertinib plus chemotherapy was consistent with known individual agent profiles, with no new safety signals. While Grade ≥3 adverse events were more frequent with the combination (87.1% vs. 54.4%), these were primarily manageable hematologic events. The importance of these findings was underscored by their selection for presentation in the prestigious Presidential Symposium 2 at the IASLC 2026 World Conference on Lung Cancer.

Zipalertinib's Significant PFS Benefit in First-Line EGFR Exon 20 NSCLC

EGFR exon 20 insertion (Exon20ins) mutations confer inherent resistance to approved tyrosine kinase inhibitors (TKIs), leaving patients with limited effective options. Amivantamab, an EGFR-MET bispecific antibody with immune cell-directing activity, has demonstrated meaningful clinical activity in this population. In the CHRYSALIS phase I study, amivantamab at the recommended phase II dose produced an overall response rate (ORR) of 40% (95% CI, 29 to 51) in 81 platinum-pretreated patients with EGFR Exon20ins NSCLC, including three complete responses, with a median duration of response of 11.1 months (95% CI, 6.9 to not reached) and a median progression-free survival (PFS) of 8.3 months (95% CI, 6.5 to 10.9). By contrast, mobocertinib — an irreversible, oral EGFR TKI designed to selectively target in-frame EGFR exon 20 insertions — was compared indirectly against a real-world external control group of platinum-pretreated patients from the Flatiron Health database. After inverse probability treatment weighting, mobocertinib achieved a confirmed ORR of 35.1% versus 11.9% for real-world standard-of-care (odds ratio: 3.75 [95% CI: 2.05, 6.89]), a median PFS of 7.3 versus 3.3 months (HR: 0.57 [95% CI: 0.36, 0.90]), and a median OS of 24.0 versus 12.4 months (HR: 0.53 [95% CI: 0.33, 0.83]), demonstrating substantially improved outcomes relative to available therapies in this rare population.

More recently, data presented at the 2024 Annual ESMO meeting from REZILIENT-1, a single-arm phase II study evaluating zipalertinib, reported an ORR of 50% in patients pretreated with amivantamab and 25% in patients pretreated with both amivantamab and an EGFR exon 20 insertion-directed TKI, with the vast majority of these patients also having received platinum-doublet chemotherapy. These findings position zipalertinib as a potentially active option in later lines following amivantamab exposure. In the post-osimertinib and post-platinum setting — though not restricted to Exon20ins — the CHRYSALIS-2 cohort A study of amivantamab plus lazertinib in EGFR exon 19 deletion- or L858R-mutated NSCLC reported a blinded independent central review-assessed ORR of 35% (95% CI: 27–42), a median duration of response of 8.3 months (95% CI: 6.7–10.9), and a median overall survival of 14.8 months (95% CI: 12.2–18.0), illustrating the continued utility of amivantamab-based combinations in heavily pretreated EGFR-mutant disease.

From a network meta-analysis perspective encompassing 9 randomized controlled trials and 2,534 patients with EGFR-TKI-resistant NSCLC, amivantamab plus chemotherapy ("ChemoAmi") significantly prolonged PFS compared with platinum-based doublet chemotherapy alone (HR, 0.48 [95% CI, 0.32–0.71]) and improved ORR (OR, 3.13 [95% CI, 1.64–5.96]). However, "ChemoAmi" was associated with a significantly higher incidence of grade 3–5 treatment-related adverse events compared with chemotherapy alone (OR, 3.71 [95% CI, 1.08–12.7]). Across the amivantamab clinical program, the most common adverse events include rash, infusion-related reactions, and paronychia, with thrombosis and pneumonitis representing clinically important but less frequent toxicities requiring structured management. Taken together, the emerging data across investigational agents — amivantamab, mobocertinib, and zipalertinib — consistently demonstrate superior efficacy over historical standard-of-care benchmarks in EGFR Exon20ins NSCLC, though toxicity profiles and sequencing strategies remain active areas of clinical investigation.

Addressing Unmet Needs in EGFR Exon 20 Insertion NSCLC

EGFR exon 20 insertion (EGFRex20ins)-positive NSCLC represents a molecularly distinct subset of disease that has historically been resistant to standard EGFR-targeted therapies, creating a significant unmet need. While newer agents such as amivantamab and mobocertinib have demonstrated meaningful clinical activity, several challenges persist in optimizing outcomes for this population.

• Intrinsic resistance to conventional EGFR-TKIs. EGFRex20ins mutations are relatively resistant to existing EGFR tyrosine kinase inhibitors, limiting the utility of the broad first-, second-, and third-generation TKI armamentarium that has transformed outcomes in classical EGFR-mutant NSCLC.

• Substantial toxicity burden with targeted agents. Amivantamab, the current standard of care for this population, is associated with treatment-related adverse events (TRAEs) of any grade in 68.8% of real-world patients, with grade 3–4 TRAEs in 10.9%. Prominent toxicities include skin rash (56%, G3 9.4%), infusion-related reactions (9.4%), asthenia (9%), and peripheral edema (8%). Dose interruptions, reductions, and discontinuations due to TRAEs occurred in 20.3%, 12.5%, and 3.1% of patients, respectively. Thrombosis was also a notable adverse effect, observed even in patients receiving anticoagulant prophylaxis, and pneumonitis, while less common, was severe when it occurred.

• Modest and variable efficacy in real-world settings. In the Italian ATLAS real-world registry, single-agent amivantamab achieved an objective response rate of 37.5% and a median progression-free survival of 9.6 months in a heavily pretreated population. Mobocertinib demonstrated an ORR of 25% in real-world compassionate-use data, consistent with the phase 2 EXCLAIM cohort, underscoring that a substantial proportion of patients do not achieve an objective response.

• Limited intracranial activity. Among patients with brain metastases treated with amivantamab in the ATLAS registry, only 13% achieved a partial intracranial response, with 43.5% achieving stable disease and an intracranial median PFS of 11.6 months. Real-world mobocertinib data similarly indicated limited intracranial activity, with the greatest benefit observed in patients without brain metastases (median duration of treatment 14.8 months versus 5.4 months in those with brain metastases, p=0.01).

• Acquired resistance and bypass signaling. As with other EGFR-targeted strategies, resistance to treatment remains a significant hurdle. Mechanisms including secondary EGFR mutations, MET amplification, and activation of bypass signaling pathways have been documented across the broader EGFR-mutant NSCLC landscape, and the evolution of resistance in the EGFRex20ins setting continues to pose challenges for long-term disease control.

Decoding the REZILIENT3 Trial Design and Key Endpoints

The CHRYSALIS phase I study remains the pivotal trial characterizing clinical activity of amivantamab-vmjw in EGFR exon 20 insertion mutation-positive NSCLC following platinum-based chemotherapy. Its open-label, dose-escalation and dose-expansion design established the recommended phase II dose and generated the efficacy and safety data supporting FDA approval of the first bispecific antibody targeting this mutation.

Zipalertinib's Safety Profile and Future Treatment Landscape

Zipalertinib has been evaluated across phase 1/2a and phase I/II studies in patients with EGFR exon 20 insertion (ex20ins)-mutant NSCLC previously treated with platinum-based chemotherapy. In the phase 1/2a study, 73 patients received zipalertinib at dose levels of 30, 45, 65, 100, and 150 mg orally twice daily. The most frequently reported treatment-related adverse events of any grade were rash (80%), paronychia (32%), diarrhea (30%), and fatigue (21%). Notably, no cases of grade 3 or higher drug-related rash or diarrhea were observed at 100 mg twice daily or below, supporting the tolerability of the recommended dose level.

In the larger REZILIENT1 phase I/II trial (NCT04036682), 244 patients received zipalertinib 100 mg twice daily. The most common grade ≥3 treatment-related adverse events were anemia (7%), pneumonitis and rash (2.5% each), and diarrhea, ALT increased, and platelet count decreased (2% each). This profile was characterised as manageable across a heavily pretreated population that included patients who had received prior amivantamab with or without additional ex20ins-targeted therapies. The low frequency of high-grade diarrhea and rash across both studies is consistent with zipalertinib's novel pyrrolopyrimidine scaffold, which confers enhanced selectivity for EGFR ex20ins-mutant versus wild-type EGFR — a mechanistic distinction from earlier-generation agents associated with more pronounced on-target toxicities.

Regarding chemotherapy specifically, the published data do not report a dedicated safety and tolerability analysis of zipalertinib in combination with chemotherapy across its studied indications. The knowledge base does not have sufficient information on this aspect.

Zipalertinib: Reshaping First-Line EGFR Exon 20 NSCLC Treatment

The recent announcement of positive Phase 3 REZILIENT3 trial results for zipalertinib in combination with chemotherapy marks a pivotal moment for patients battling EGFR exon 20 insertion mutation-positive non-small cell lung cancer (NSCLC). This subtype has long presented a formidable challenge, historically managed with platinum-based chemotherapy offering a median progression-free survival (PFS) of only 6-7 months. The 14.5-month median PFS achieved with zipalertinib plus chemotherapy represents a substantial leap forward, significantly outperforming chemotherapy alone (8.5 months) and even surpassing the 11.4-month PFS seen with amivantamab plus chemotherapy, which was previously a notable advance in this space.

This robust efficacy, coupled with zipalertinib's profile as an oral, selective EGFR tyrosine kinase inhibitor, positions it as a strong contender for the new first-line standard of care. The convenience of an oral therapy, compared to intravenous options, could significantly enhance patient quality of life and adherence, driving its adoption. Furthermore, its enhanced selectivity for the exon 20 insertion mutation over wild-type EGFR suggests a potentially more favorable tolerability profile, mitigating common TKI-related side effects.

However, the journey ahead is not without its complexities. The therapeutic landscape for EGFR exon 20 insertion NSCLC is dynamic, with other promising oral selective TKIs like sunvozertinib and furmonertinib emerging. These agents could introduce significant competitive pressure, necessitating continuous demonstration of zipalertinib's sustained benefits. Moreover, the nuanced impact of insertion location—near-loop versus far-loop—on TKI sensitivity is a critical consideration. While zipalertinib may show greater potency against near-loop insertions, understanding its real-world efficacy across the full spectrum of insertion types will be crucial. As with all targeted therapies, the long-term safety profile and the eventual development of resistance mechanisms will require ongoing vigilance and research to ensure durable patient outcomes. This evolving field underscores the relentless pursuit of precision medicine, offering renewed hope for patients with this challenging disease.