Otsuka Holdings' cancer unit has a number to back its bet on an experimental lung-cancer pill: patients on the drug went nearly six months longer before their disease worsened, compared with chemotherapy alone. In a disclosure filed with Tokyo's stock exchange, Otsuka said the Phase 3 REZILIENT3 trial of zipalertinib, combined with platinum-based chemotherapy, extended median progression-free survival to 14.5 months versus 8.5 months for chemotherapy alone in patients with EGFR exon 20 insertion mutations, a hazard ratio of 0.50 (95% confidence interval 0.34-0.73, P=0.00015).
The drug is developed by Taiho Pharmaceutical, Otsuka's wholly owned subsidiary, together with its US arm Taiho Oncology and partner Cullinan Therapeutics, the Nasdaq-listed biotech running the US development program. The trial randomized 279 patients with no prior treatment for advanced disease, 140 to the combination and 139 to chemotherapy alone, after a six-patient safety lead-in. Objective response rate favored the combination, 65.0% versus 40.3% (P<0.0001), and median duration of response ran longer too, 14.2 months versus 9.9 months. The progression-free survival benefit held even in patients with brain metastases, a subgroup with few good options, where the hazard ratio was 0.38.
| Measure | Zipalertinib + chemotherapy | Chemotherapy alone |
|---|---|---|
| Median progression-free survival | 14.5 months | 8.5 months |
| Objective response rate | 65.0% | 40.3% |
| Median duration of response | 14.2 months | 9.9 months |
| Grade 3+ adverse events | 87.1% | 54.4% |
Overall survival data are still forming. At 30% data maturity, the hazard ratio for death favored the combination at 0.72 (95% confidence interval 0.42-1.23), and the companies said follow-up is continuing to further clarify overall survival and exploratory endpoints.
The efficacy gain came with a heavier side-effect load. Grade 3 or higher adverse events hit 87.1% of patients on the combination versus 54.4% on chemotherapy alone, though Otsuka said most of the excess was manageable blood-related toxicity, 58.6% versus 28.7%. Grade 3 or higher adverse events specific to the EGFR mechanism were uncommon and appeared only in the combination arm: rash in 10.7% of patients and diarrhea in 1.4%. The company reported no new safety signals.
None of this changes the near-term financial picture. Zipalertinib is not approved by any drug regulator anywhere, and Otsuka said the results do not alter its consolidated earnings forecast for the fiscal year ending in December 2026. The companies said they will continue discussions with regulators about the results while working to deliver the treatment to patients who need it. The data were presented September 14 at the Presidential Symposium of the International Association for the Study of Lung Cancer's 2026 World Conference on Lung Cancer in Seoul, by Dr. Daniel Tan Shao Weng of Duke Health, Singapore. The trial is registered as NCT05973773.