Cullinan reports six-month median PFS gain for zipalertinib combination in Phase 3 REZILIENT3

Cullinan Therapeutics and its Taiho partners reported detailed interim Phase 3 results showing that zipalertinib plus chemotherapy extended median progression-free survival to 14.5 months, compared with 8.5 months for chemotherapy alone, in previously untreated advanced non-small cell lung cancer with EGFR exon 20 insertion mutations. The six-month difference anchors the REZILIENT3 readout, which also showed higher response rates and longer median response duration with the combination. The September 13 announcement describes a clinical trial result, not an FDA approval, and the interim overall-survival findings do not establish a survival benefit.

What REZILIENT3 showed

Cullinan Therapeutics and its Taiho partners reported detailed interim Phase 3 results showing that zipalertinib plus chemotherapy extended median progression-free survival to 14.5 months, compared with 8.5 months for chemotherapy alone, in previously untreated advanced non-small cell lung cancer with EGFR exon 20 insertion mutations. The six-month difference anchors the REZILIENT3 readout, which also showed higher response rates and longer median response duration with the combination. The September 13 announcement describes a clinical trial result, not an FDA approval, and the interim overall-survival findings do not establish a survival benefit.

The trial had previously been announced as meeting its primary progression-free survival endpoint. The newly disclosed detail puts numbers around that result: the PFS hazard ratio was 0.50, with a 95% confidence interval of 0.34 to 0.73 and P=0.00015. The pre-planned interim efficacy analysis followed 122 PFS events. The hazard ratio and the six-month difference in medians describe different aspects of the result; neither should be read as a guarantee that an individual patient receives six additional months without progression.

Trial design and who was studied

REZILIENT3 is a multicenter, randomized, controlled, open-label global study. Following a six-patient safety lead-in, investigators randomized 279 patients: 140 to zipalertinib plus platinum-based chemotherapy and 139 to chemotherapy alone. The release's description of 285 enrolled adults includes that lead-in and should not be confused with the randomized population. Zipalertinib was given at 100 mg twice daily in the combination arm. The comparison directly addresses adding zipalertinib to chemotherapy in this first-line setting. It does not establish superiority over treatments outside the two trial arms.

Reported baseline characteristics were roughly balanced. Age was 66.5 years in the combination arm and 64 years in the chemotherapy arm; women accounted for 65.7% and 63.3%, respectively. Brain metastases were present in 31.4% and 31.7%. The companies reported a PFS hazard ratio of 0.38 in the brain-metastases subgroup and described benefit as consistent across subgroups. That subgroup observation adds context to the overall endpoint, but the supplied disclosure summary does not provide its confidence interval or event count. It should therefore be read with less precision than the overall PFS estimate.

Responses and duration

Tumor-response measures also favored the combination. The objective response rate was 65.0%, versus 40.3% with chemotherapy alone, with P<0.0001. That is a calculated difference of 24.7 percentage points. Median duration of response was 14.2 months with zipalertinib plus chemotherapy and 9.9 months with chemotherapy alone, a calculated difference of 4.3 months. Those measures show both a greater proportion of patients responding and a longer reported median response duration, while remaining distinct from the question of whether treatment extends overall survival.

Overall survival still immature

The overall-survival analysis remains preliminary. At 30% event maturity, the hazard ratio for death was 0.72, with a 95% confidence interval of 0.42 to 1.23. Because that interval crosses 1, the reported analysis does not demonstrate a statistically significant overall-survival benefit. Follow-up is ongoing, and the final survival outcome remains unknown. The favorable direction of the interim point estimate cannot substitute for a mature result or be presented as a confirmed reduction in deaths.

Safety trade-off

Safety is a material part of the readout. Grade 3 or higher adverse events occurred in 87.1% of patients receiving the combination, compared with 54.4% receiving chemotherapy alone, a calculated increase of 32.7 percentage points. The release described these events as primarily manageable hematologic adverse events, reported at 58.6% versus 28.7%. In the combination arm, reported Grade 3 or higher rash and diarrhea rates were 10.7% and 1.4%, respectively. The supplied summary does not include a complete Grade 3/4 safety table or matching chemotherapy-arm rates for those two toxicities.

The companies said the adverse-event profile was generally consistent with the known profiles of the individual agents and that no new safety signals were observed. Those statements are the sponsors' characterization. They sit alongside, rather than remove, the higher reported frequency of severe adverse events with combination treatment. The efficacy and safety results both matter when assessing the potential first-line role of this regimen; the release alone does not resolve every clinical treatment-selection question.

How investigators framed the result

Study investigator Helena A. Yu of Memorial Sloan Kettering characterized the PFS improvement as statistically significant and clinically meaningful and said the higher response rates supported the regimen's potential as a first-line option. Cullinan Chief Medical Officer Jeffrey Jones similarly framed the magnitude of the PFS benefit, response rate and response duration as supporting a potential first-line role. These are attributed investigator and company assessments, paraphrased from the company release rather than reproduced as verbatim quotations.

WCLC Presidential Symposium context

The results were designated for Presidential Symposium 2, PL03, at the IASLC 2026 World Conference on Lung Cancer in Seoul. The listed session time was Monday, September 14, at 8 a.m. Korea Standard Time in Plenary Hall D2 3F, with Dr. Daniel Tan Shao Weng as presenting author. That conference context is separate from the September 13 press-release date and the September 14 U.S. filing date. The symposium selection provides scientific presentation context; it is not a regulatory decision.

Filing path and unresolved questions

Taiho Oncology Chief Medical Officer Harold Keer said the company intended to discuss the results with health authorities and work toward making zipalertinib available in a timely manner. The disclosure does not establish a submission date, regulatory acceptance, approval or commercial launch timing. It identifies Taiho Oncology's U.S. collaboration with Cullinan, but supplies no basis here for additional partner-economics assumptions. The next unresolved issues are the final overall-survival outcome and the regulatory path, while the current evidence establishes an interim PFS benefit accompanied by a higher severe-adverse-event rate. Cullinan disclosed the announcement in its Form 8-K under accession number 0001193125-26-389844, with the joint press release attached as Exhibit 99.1.