Ivonescimab significantly extends overall survival compared with pembrolizumab for patients with PD-L1-positive advanced non-small cell lung cancer, according to a prespecified interim analysis of the phase 3 HARMONi-2 trial. The data, presented at the International Association for the Study of Lung Cancer 2026 World Conference on Lung Cancer, show the first-in-class PD-1/VEGF bispecific antibody offering a survival advantage over the standard-of-care pembrolizumab. Summit Therapeutics markets this investigational agent under the name AK112.
Survival Data and Trial Design
The HARMONi-2 study randomized 398 treatment-naive patients with locally advanced or metastatic NSCLC, requiring a PD-L1 tumor proportion score of at least 1% and no EGFR or ALK alterations. Patients received either ivonescimab at 20 mg/kg every three weeks or pembrolizumab at 200 mg every three weeks. At the August 20, 2026, data cutoff, 234 overall survival events had occurred. The median overall survival was approximately 30.8 months with ivonescimab versus 22.6 months with pembrolizumab, with a hazard ratio of 0.73 (95% CI, 0.57-0.95; P = 0.009). This represents a statistically significant survival benefit, indicating that patients treated with the bispecific antibody lived longer on average than those receiving the monoclonal antibody.
This benefit appeared across prespecified subgroups, including patients with squamous histology (HR 0.65; 95% CI, 0.45-0.95) and those with a PD-L1 TPS of 1% to 49% (HR 0.85; 95% CI, 0.61-1.18). The trial previously showed ivonescimab extended progression-free survival to 11.1 months versus 5.8 months for pembrolizumab. That earlier finding supported regulatory approval of ivonescimab in China in April 2025. This approval was based on the demonstrated ability of the drug to improve outcomes when used as monotherapy in this specific patient population.
Read Also: Cancer rates rise among 9/11 survivors
Tolerability and Clinical Implications
Safety profiles remained comparable between the two treatments during the extended follow-up period. Any-grade treatment-related adverse events occurred in 93.4% of ivonescimab patients versus 84.9% with pembrolizumab, and serious TEAEs were reported in 29.9% versus 21.6%, respectively. Discontinuations due to TEAEs were similar at 4.1% and 5.0%, respectively. Immune-related adverse events also showed no significant difference at 34.0% versus 33.7%. The high rate of adverse events without a corresponding increase in discontinuations suggests the side effects are generally manageable and do not prevent patients from continuing their therapy.
For clinicians in China, where ivonescimab is already approved, these results may solidify its role in standard regimens. However, the slightly raised rate of adverse events without a corresponding increase in discontinuations or severe immune reactions suggests that tolerability remains manageable even as survival data mature. In the United States and other markets where ivonescimab is still investigational, pharmacy teams should treat these findings as a signal to monitor rather than a current formulary option. The trial is registered under ClinicalTrials.gov identifier NCT05499390, and the results were presented in Seoul, Republic of Korea, at the IASLC 2026 conference.
