

A two-in-one cancer drug just did something unprecedented: it beat an established immunotherapy on overall survival in a head-to-head Phase 3 lung cancer trial. The results could reshape how we treat one of oncology's toughest cancers.
Imagine bringing a Swiss Army knife to a sword fight. Your opponent has one blade. You have two. That's roughly what just happened in one of the most competitive arenas in oncology.
A bispecific antibody called ivonescimab just delivered something the lung cancer world has been waiting years to see: a survival advantage over a modern PD-1 inhibitor, head to head, in a rigorous Phase 3 trial. The drug cut the risk of death by 34% compared to tislelizumab (a well-established PD-1 inhibitor) when both were paired with chemotherapy in first-line squamous lung cancer.
That's not beating a placebo. That's not beating chemo alone. That's beating the current playbook.
For the past several years, the standard first-line treatment for advanced squamous non-small cell lung cancer (NSCLC) has been simple: give patients a PD-1 inhibitor plus chemotherapy. PD-1 inhibitors are drugs that release the brakes on the immune system, letting T cells attack tumors. Pembrolizumab (Merck's Keytruda) dominates this space globally.
Nothing has beaten a PD-1 inhibitor plus chemo on overall survival in this setting. Until now.
The HARMONi-6 trial, run by Chinese biotech Akeso, enrolled 532 patients with previously untreated advanced squamous NSCLC across sites in China. It randomized them to receive either ivonescimab or tislelizumab, both layered on top of carboplatin and paclitaxel chemotherapy. The trial was double-blind, meaning neither doctors nor patients knew which drug they were getting.
The results, presented at the ASCO 2026 Plenary Session on May 31, 2026, were striking.
Patients on ivonescimab lived a median of 27.9 months, compared to 23.7 months for the tislelizumab group. The hazard ratio was 0.66 (95% CI: 0.50–0.87; p=0.0017), which translates to that 34% reduction in death risk.
But the gap widened over time, which is often the more telling signal. At the , 64.7% of ivonescimab patients were still alive, versus just 48.6% in the control arm. That's a in survival at 24 months. Think of it this way: for every 100 patients treated, roughly 16 more people were alive at two years with the new drug.

The Pentagon just labeled WuXi AppTec a "Chinese military company," and the ripple effects could force hundreds of biotech and pharma companies to rethink how their drugs get made. The world's largest CDMO is now at the center of a geopolitical earthquake.


Join thousands of biotech professionals who start their day with our free, daily briefing.
The benefit held up across every PD-L1 subgroup tested. PD-L1 is a protein on tumors that helps predict who responds to immunotherapy. Whether patients had high PD-L1, low PD-L1, or no PD-L1 expression at all, the hazard ratios stayed remarkably consistent (ranging from 0.64 to 0.68). Patients with three or more metastatic sites saw an even larger benefit, with a hazard ratio of 0.47.
These OS results confirmed what the trial had already shown on its primary endpoint: progression-free survival (the time before tumors started growing again) was 11.1 months with ivonescimab versus 6.9 months with tislelizumab. Nearly doubling time-to-progression is impressive on its own. Adding a survival benefit on top seals the deal.
So what makes ivonescimab different from a regular PD-1 inhibitor? It's a bispecific antibody, meaning it's a single molecule engineered to grab two different targets at once.
One arm blocks PD-1, the immune checkpoint that tumors exploit to hide from T cells. The other arm grabs VEGF-A, a protein that tumors use to grow new blood vessels and feed themselves. Blocking VEGF also helps remodel the area around the tumor, making it easier for immune cells to get in and do their work.
The clever part is the cooperative binding. When VEGF is nearby, ivonescimab grips PD-1 about 18 times more tightly. When PD-1 is present, its hold on VEGF strengthens too. The drug essentially becomes more potent exactly where the tumor is most active. It's like a lock that gets tighter the harder you try to open it.
Doctors have tried combining separate PD-1 inhibitors with separate VEGF blockers (like pembrolizumab plus bevacizumab) before. But a bispecific bundles both functions into one molecule, potentially concentrating the dual attack right at the tumor site rather than flooding the whole body with two drugs.
Analysts were broadly positive. Stifel's Dara Azar called HARMONi-6 a "clear win" and argued the data strongly support the case for ivonescimab's upcoming global trial. Citi's Yigal Nochomovitz, who had previously flagged that ivonescimab's earlier results were "anything but a one-hit wonder," sees HARMONi-6 as confirmation that the bispecific mechanism works across multiple NSCLC settings.
But there's a catch. The 4-month median survival gain, while statistically significant and clinically meaningful, reportedly fell slightly short of what some bullish investors had hoped for. Biopharma Dive noted that the absolute OS delta "appears to have fallen short of investor expectations."
This is the classic biotech tension: a drug can be genuinely practice-changing for patients while underwhelming investors who had priced in a bigger win.
Here's the part that keeps this story from being a simple victory lap. HARMONi-6 beat tislelizumab, not Keytruda. And while both are PD-1 inhibitors, Keytruda is the undisputed heavyweight champion of lung cancer treatment globally.
The real test comes in HARMONi-3, a global Phase 3 trial where ivonescimab plus chemo goes head to head against Keytruda plus chemo in a non-Chinese population. That's the trial that will determine whether ivonescimab can break into the U.S. and European markets, where Keytruda's grip is ironclad.
HARMONi-6 does something important for HARMONi-3, though: it de-risks the biology. Showing that a PD-1/VEGF bispecific can deliver hard survival benefits over an active PD-1 comparator (not just a placebo) makes a compelling case that the mechanism works. It's the difference between a startup showing product-market fit in one city before expanding nationally.
Summit Therapeutics, which holds the rights to ivonescimab outside China (through a deal worth up to $5 billion in upfront and milestone payments), is banking everything on HARMONi-3. Akeso, the drug's developer, already has the FDA reviewing a BLA (a biologics license application) for ivonescimab in EGFR-mutated, locally advanced or metastatic non-squamous NSCLC after prior EGFR TKI therapy, with a decision expected in November 2026.
Beyond the ivonescimab story, HARMONi-6 is a landmark for an entire drug class. It's among the first demonstrations that a PD-1/VEGF bispecific can beat an active PD-1 inhibitor on overall survival in a controlled Phase 3 trial. That validates the whole concept of dual-targeting bispecifics in lung cancer.
The competition is already heating up. Pfizer has partnered with 3SBio on SSGJ-707, another PD-1/VEGF bispecific heading toward Phase 3 in NSCLC. BioNTech and Bristol Myers Squibb are advancing pumitamig (a PD-L1/VEGF bispecific) in late-stage trials. Merck itself has in-licensed LM-299, a PD-1/VEGF bispecific from LaNova, essentially hedging against the possibility that its own Keytruda could be outgunned by this new class.
Analysts estimate there are approximately 9–10 PD-1/VEGF bispecifics in clinical development globally, with at least five still having regional licensing rights available. But as one industry report noted, that window is "closing rapidly."
Three things to watch:
First, ivonescimab's FDA decision in November 2026 for EGFR-mutated non-squamous NSCLC. If approved, it becomes the first PD-1/VEGF bispecific available outside China, setting the pricing and regulatory benchmark for everyone behind it.
Second, the HARMONi-3 readout. This is the make-or-break moment. Beating tislelizumab in China is impressive. Beating Keytruda in a global, diverse patient population would be seismic. It would mark the first time any drug has displaced Keytruda in a major first-line lung cancer segment on survival.
Third, the safety story. VEGF inhibitors carry known risks (hypertension, bleeding, protein in the urine). If ivonescimab's safety profile stays manageable, the survival benefit speaks for itself. If toxicity concerns emerge at scale, regulators and doctors will have to weigh the tradeoffs carefully.
Squamous lung cancer is notoriously tough to treat. It's a cancer that doesn't carry the targetable mutations (like EGFR or ALK) that give oncologists other options. For these patients, PD-1 plus chemo has been the best available answer for years. HARMONi-6 suggests there might finally be a better one.
The Swiss Army knife, it turns out, might actually beat the sword.
For the first time ever, a Phase 3 trial shows a finite therapy can functionally cure chronic hepatitis B, with one in four selected patients staying virus-free after stopping all treatment. The 300-million-patient hep B market just got its biggest shakeup in decades.