Ivonescimab, a novel bispecific antibody targeting both programmed cell death protein-1 (PD-1) and vascular endothelial growth factor (VEGF) pathways, has shown significant clinical benefit in patients with EGFR-mutated non-small cell lung cancer (NSCLC) who have progressed on third-generation EGFR tyrosine kinase inhibitors (TKIs).

Jonathan Goldman, MD, of UCLA Health, presented this finding and other new results of the global phase 3 HARMONi trial at the World Conference on Lung Cancer (WCLC) 2025.

The study demonstrated that ivonescimab combined with chemotherapy reduced the risk for disease progression or death by 48% compared to chemotherapy alone, with a hazard ratio (HR) of 0.52 (95% CI, 0.41-0.66; P < .001). The overall survival (OS) results showed a favorable trend, but they did not reach statistical significance at the prespecified analysis.

Rationale

Goldman explained that although third-generation EGFR TKIs can improve outcomes for patients with EGFR-mutated NSCLC, acquired resistance is common, leaving limited therapeutic options beyond platinum-based chemotherapy.

“There is a clinical unmet need in patients with advanced NSCLC harboring EGFR mutations in the relapsed or refractory setting, where subsequent management options remain limited,” Goldman said during his talk.

Goldman noted that ivonescimab is a first-in-class bispecific antibody that simultaneously targets PD-1 and VEGF. This dual mechanism aims to address the resistance patterns that emerge after EGFR TKI failure.

In an interview, he further explained the rationale behind this bispecific approach: “The effects of a dual targeted antibody are still being investigated, but they do seem to be different than combining two different antibodies with single targeting. Most likely, you achieve more specificity for cells or tumor microenvironments that overexpress both markers. This may improve the therapeutic index by maximizing the effect at the tumor and minimizing the effects elsewhere.”

Study Design

The HARMONi trial enrolled 438 patients across multiple regions, with 273 patients from China and 165 (38%) from North America and Europe. Patients were randomly assigned 1:1 to receive either ivonescimab (20 mg/kg) or placebo plus standard chemotherapy consisting of pemetrexed (500 mg/m²) and carboplatin every 3 weeks for four cycles, followed by maintenance therapy. Patients received maintenance pemetrexed plus ivonescimab or pemetrexed plus placebo every 3 weeks, with stratification for presence of brain metastases by region.

The study population was well balanced between arms, with a median age of 62 years, and 24.7% having brain metastases at study entry in both arms. Approximately 60% of patients harbored EGFR exon 19 deletions, 35% had L858R mutations, and 5% had noncanonical mutations.

External discussant Suresh Ramalingam, MD, of the Winship Cancer Institute of Emory University, Atlanta, highlighted the complexity of the trial design, noting that it enrolled patients over two different time frames with distinctly different median follow-up durations: 9 months for non-Asian patients compared to 33 months for the Asian population.

Significant PFS Benefit, Mixed OS Results

The primary efficacy endpoint, median progression-free survival (PFS) by independent radiology review, was 6.8 months in the ivonescimab arm and 4.4 months in the control arm. 

“Almost 20% more patients in the ivonescimab arm were progression-free at both the 6- and 12-month landmarks,” Goldman explained during his presentation. The 6-month PFS rate was 54.0% in the ivonescimab arm and 34.7% in the control arm, while the 12-month PFS rate was 25.4% and 8.3%, respectively. 

Goldman emphasized that the PFS benefit was consistent across all predefined subgroups, with particularly notable efficacy in patients with brain metastases, who achieved a hazard ratio of 0.34 (95% CI, 0.20-0.57).

When asked about the mechanism behind central nervous system (CNS) activity in an interview, Goldman noted: “The trial showed benefit that was perhaps accentuated in patients with brain metastases. It has long been theorized that VEGF blockade improves drug delivery to the CNS, possibly due to normalization of pressures within the tumor.”

At the final analysis with a median follow-up of 29.7 months, median OS was 16.8 months in the ivonescimab arm vs 14.0 months in the placebo arm (HR, 0.79; 95% CI, 0.62-1.01; P = .0570), missing the prespecified significance threshold. 

Safety Profile and Tolerability

Goldman reported that grade ≥ 3 treatment-related adverse events (TRAEs) occurred in 50.0% of patients in the ivonescimab arm and in 42.2% of patients in the control arm. The most common TRAEs were laboratory abnormalities, nausea, and decreased appetite.

“The low toxicity rate stood out to me; it was very difficult to tell who was on placebo and who was on the active drug,” Goldman remarked during his presentation. He also noted that VEGF-related TRAEs were primarily driven by reversible hypertension and proteinuria, with grade ≥ 3 VEGF-blocking TRAEs occurring in 7.3% vs 3.2% of patients. Grade ≥ 3 hemorrhage occurred in 0.9% vs 0% of patients, and TRAEs leading to death were numerically lower in the ivonescimab arm (1.8% vs 2.3%).

Clinical Implications and Future Directions

The results from HARMONi position ivonescimab as a potential new treatment option in the post-EGFR TKI setting, although regulatory approval will depend on how agencies interpret the mixed efficacy signals, Goldman noted in his interview. 

“The low serious side effect rate may be an important consideration in treatment selection for these patients,” he added.

Although Ramalingam acknowledged the positive PFS findings, he raised questions about the mechanism of action and clinical significance. 

“The level of PFS benefit observed in HARMONi is comparable to that in studies that have evaluated VEGF inhibitors in NSCLC,” he noted during the discussion part of the session. “In my view, there is no clear evidence of PD-1 blockade as a contributor to the efficacy,” he added, referencing the comparable PFS in patients with positive PD-L1 expression and those without PD-L1 expression.

Ramalingam also emphasized the need for careful consideration of the risk-benefit profile. 

“The relative merits of incremental efficacy vs toxicity should be contextualized to individual patients,” he concluded.

The HARMONi study was supported by Summit Therapeutics and Akeso Biopharma. 

Goldman reported receiving research funding from AbbVie, Advaxis, Amgen, Astellas, AstraZeneca, Bristol Myers Squibb, Eli Lilly, Genentech, GSK, Janssen, Merck, Pfizer, and RayzeBio; honoraria from AbbVie, Amgen, AstraZeneca, Bristol Myers Squibb, Genentech, Gilead, Gritstone, Janssen, Jazz, Lilly, Pfizer, Puma, Regeneron, and Summit. 

Ramalingam reported receiving research funding from Amgen, AstraZeneca/MedImmune, Bristol Myers Squibb, Merck, Pfizer, and Takeda; and honoraria from Amgen, AstraZeneca, Bristol Myers Squibb, Merck, Glaxo SmithKline, Takeda, Genentech, Daiichi Sankyo, and Puma.

Christos Evangelou, MSc, PhD, is a freelance medical writer.