Highlights
– HARMONi‑6 (presidential symposium, ESMO) showed ivonescimab (PD‑1/VEGF bispecific) + paclitaxel/carboplatin improved IRRC‑assessed mPFS 11.1 vs 6.9 months vs tislelizumab+chemo (HR 0.60; P<0.0001).
– Objective response rate (ORR) 76% vs 54%; median duration of response (DoR) 11.2 vs 8.4 months. Benefit occurred irrespective of PD‑L1 status.
– Safety profile comparable to PD‑1+chemo; grade ≥3 bleeding low (1.9%). Regulatory review in China underway with priority status.
Background and Unmet Need
Squamous non‑small cell lung cancer (sq‑NSCLC) remains a therapeutically challenging subtype. Actionable driver mutations are uncommon (<5%), limiting targeted therapy options. The predominant tumor location is central with frequent cavitation and vascular encasement, conditions that increase hemorrhagic risk from anti‑angiogenic agents and have historically excluded many patients from trials. PD‑1 inhibitors combined with platinum‑doublet chemotherapy established a standard first‑line option, but median progression‑free survival (mPFS) has plateaued around 6–7 months and PD‑L1–negative patients often derive more limited benefit. Older patients, those with prior hemoptysis or cardiovascular comorbidity are frequently underrepresented in clinical trials, leaving uncertainty about tolerability in a realistic population.
Study Design — HARMONi‑6 Overview
HARMONi‑6 was a randomized, head‑to‑head, phase 3 trial comparing ivonescimab plus paclitaxel/carboplatin against tislelizumab (a PD‑1 monoclonal antibody) plus the same chemotherapy backbone in previously untreated advanced or metastatic sq‑NSCLC (main endpoint: IRRC‑assessed PFS). The trial intentionally enrolled a clinically representative population with substantial proportions of central tumors (63%), cavitation (8.8%), major vascular encasement (17.5%) and prior hemoptysis (30%). Key efficacy outcomes reported in The Lancet include mPFS, ORR, DoR and subgroup analyses by PD‑L1 expression; safety outcomes included grade ≥3 adverse events (AEs) and bleeding events.
Key Results
Progression‑Free Survival
Ivonescimab + chemotherapy achieved a median PFS of 11.1 months versus 6.9 months for tislelizumab + chemotherapy (hazard ratio [HR] 0.60; P<0.0001), representing a 40% relative reduction in risk of progression or death by IRRC assessment. The result was statistically robust and clinically meaningful in a setting where incremental PFS gains have been difficult to achieve.
Response and Durability
Objective response rate was higher with ivonescimab: 76% versus 54% with tislelizumab+chemo. Median duration of response was 11.2 months versus 8.4 months, indicating not only more frequent responses but also deeper and more durable tumor control for many patients.
PD‑L1 Subgroups
Preplanned subgroup analyses showed benefit regardless of PD‑L1 expression: HRs for PFS were 0.66 for PD‑L1 ≥1% and 0.55 for PD‑L1 <1%. These data support initiating ivonescimab‑based therapy without waiting for PD‑L1 testing in clinical scenarios requiring immediate treatment decisions.
Safety
Grade ≥3 treatment‑related AEs occurred in 63.9% of ivonescimab patients versus 54.3% with tislelizumab+chemo — within the expected range for immunotherapy plus chemotherapy regimens. Importantly, clinically significant bleeding (grade ≥3) was low (1.9% vs 0.8%) and no signal of fatal hemoptysis attributable to treatment was reported despite the trial enrolling patients with central tumors and prior hemoptysis. The investigators highlighted that Fc engineering of ivonescimab — intended to reduce Fcγ receptor binding and platelet/endothelial overactivation — may have contributed to the favorable bleeding profile.
Mechanistic Rationale: Why a Single‑Molecule Dual Antibody?
Ivonescimab is a tetravalent bispecific antibody designed to bind both PD‑1 (and thereby block PD‑1/PD‑L1/PD‑L2 immune checkpoints) and VEGF/VEGFR2 pathways (anti‑angiogenic). Biologic rationale is threefold: (1) immune activation — enhanced T cell reinvigoration through PD‑1 blockade; (2) vascular normalization — modulation of abnormal tumor vasculature to improve chemotherapy and immune cell delivery and to reduce myeloid‑derived suppressor cell infiltration; (3) safety optimization — Fc ‘silencing’ via engineering to reduce adverse platelet‑endothelial activation and bleeding risk. Preclinical models reportedly showed superior antitumor activity compared with separate PD‑1 inhibitor + bevacizumab combination, with a toxicity profile resembling PD‑1 monotherapy, providing plausible biological basis for the clinical findings.
Clinical and Translational Implications
The HARMONi‑6 results are notable for several reasons: they demonstrate that a single bispecific agent can outperform an established PD‑1 antibody when each is combined with the same chemotherapy backbone. The magnitude of PFS benefit, higher ORR and longer DoR suggest improved disease control that could translate into quality‑of‑life gains and possibly overall survival (OS), although OS data were not reported or remain immature. The effect across PD‑L1 subgroups simplifies early treatment decisions and could expand benefit to patients irrespective of biomarker status.
Operational advantages of a single‑molecule strategy include simplified logistics compared with separate agent coadministration, potential pharmacokinetic advantages and possibly lower manufacturing or distribution complexity. However, economic implications for payers and access will depend on pricing and health‑economic assessments versus existing combinations.
Limitations and Outstanding Questions
Key limitations to consider:
- OS data are not yet mature and remain the definitive endpoint for assessing long‑term clinical benefit. Without OS, changes in PFS and response may not predict survival or cure.
- Although HARMONi‑6 enrolled many patients with high‑risk features (central tumors, cavitation, prior hemoptysis), trial participants are still a selected population relative to all comorbid real‑world patients; broader tolerability and safety data will be needed.
- Cross‑trial comparisons are imperfect. Head‑to‑head design versus tislelizumab+chemo is a strength, but global generalizability — across regions, ethnicities and healthcare settings — will require data from ongoing global studies (for example, HARMONi‑3, NCT05899608) and longer follow‑up.
- Biomarker correlates beyond PD‑L1 (tumor mutation burden, immune gene signatures, angiogenesis markers) were not detailed in the primary report; predictive markers would help refine patient selection and cost‑effectiveness.
Regulatory and Developmental Context
Ivonescimab has moved rapidly through development. The HARMONi‑A trial supported approval of ivonescimab + chemotherapy in EGFR‑TKI‑failed nsq‑NSCLC (accepted May 2024). HARMONi‑2 reported superiority of ivonescimab monotherapy vs pembrolizumab in PD‑L1–positive NSCLC and was published in The Lancet (2025). HARMONi‑6 adds evidence for the squamous histology across PD‑L1 strata. A supplemental NDA for first‑line ivonescimab + chemotherapy in sq‑NSCLC has been accepted by China’s CDE with priority review and an estimated decision in 2026 Q1. Global phase 3 trials (HARMONi‑3 and HARMONi‑7) are underway to provide broader OS, safety and biomarker data needed for potential global registration and guideline incorporation.
Expert Perspective and Practice Considerations
For clinicians treating advanced sq‑NSCLC, HARMONi‑6 suggests a potential new standard that combines immune checkpoint blockade and anti‑angiogenesis within a single agent. If regulatory approval and longer‑term outcomes corroborate the reported PFS and response benefits without unacceptable toxicity, ivonescimab‑based regimens could be particularly attractive for patients with central tumors or prior hemoptysis where historical concerns limited use of anti‑VEGF strategies.
Until OS and mature safety data are available, clinicians should weigh individual patient factors, comorbidities and local regulatory guidance. Participation in ongoing clinical trials and real‑world observational studies will be important to define long‑term outcomes, rare toxicities, and subgroup effects (elderly, poor performance status, active cardiovascular disease).
Conclusion
HARMONi‑6 provides compelling phase‑3 evidence that ivonescimab, a PD‑1/VEGF bispecific antibody, when combined with paclitaxel/carboplatin, significantly improves PFS, ORR and DoR compared with PD‑1 antibody (tislelizumab) plus the same chemotherapy in first‑line advanced sq‑NSCLC, while maintaining a manageable safety profile and a low rate of major bleeding. These data mark an important step in applying single‑molecule dual‑pathway inhibition to a difficult‑to‑treat subtype and justify the ongoing global studies and regulatory review that will determine whether ivonescimab becomes a new standard of care.
Funding and ClinicalTrials.gov
Primary trial funding and sponsor details are reported in the Lancet publication. Ongoing trials include HARMONi‑3 (NCT05899608) and HARMONi‑7; the sponsor‑reported regulatory filings include a priority sNDA review in China for first‑line sq‑NSCLC.
References
1. Ahn MJ, Lu S, Chen J, et al. Ivonescimab plus chemotherapy versus tislelizumab plus chemotherapy in first‑line advanced squamous NSCLC: the phase 3 HARMONi‑6 study. Lancet. 2025 Oct 19. doi: 10.1016/S0140-6736(25)00601-3.
2. Xiong A, Wang L, Chen J, et al. Ivonescimab versus pembrolizumab for PD‑L1‑positive non‑small‑cell lung cancer (HARMONi‑2): a randomised, double‑blind, phase 3 study. Lancet. 2025;405(10481):839-849. PMID: 40057343.
