Akeso has shared for the first time overall survival data from a high-profile Keytruda head-to-head trial for its PD-1xVEGF bispecific ivonescimab—and investors in the company's partner Summit Therapeutics aren’t taking the result too well.

Summit’s stock price on the Nasdaq plummeted by nearly 37% following Akeso’s revelation before a trading halt was initiated. 

In announcing (Chinese) ivonescimab’s Chinese approval in first-line PD-L1-positive non-small cell lung cancer on Friday, Akeso noted that the bispecific drug slashed the risk of death by 22.3% versus Keytruda at an interim analysis of the closely watched HARMONi-2 trial.

The number apparently didn’t cross the statistical significance bar, which had an alpha assignment of 0.0001. The analysis was performed when 39% of prespecified deaths for the final analysis were recorded.

The Chinese approval was based on data showing that ivonescimab reduced the risk of progression or death by 49% compared with Keytruda in the Chinese trial. Patients who took the Akeso drug lived a median 11.14 months without disease progression, versus 5.82 months for Keytruda.

The magnitude of the overall survival improvement and failure to meet statistical significance apparently disappointed Summit’s investors, who recently enjoyed a rally after ivonescimab posted another Chinese phase 3 win, beating BeiGene’s PD-1 Tevimbra in their respective combinations with chemotherapy in first-line squamous non-small cell lung cancer on progression-free survival. 

To put the 22.3% overall survival data in perspective, Keytruda’s various indications in first-line NSCLC are mostly based on 30%-plus reductions in the risk of death versus control arms. The one exception is Keynote-042, which is the exact same monotherapy, PD-L1-positive setting as the HARMONi-2 trial. In Keynote-042, Keytruda only reduced the risk of death by 19% versus chemotherapy.

Despite the gloomy market reaction, Citi analyst Yigal Nochomovitz, Ph.D., called the initial overall survival number an “excellent result,” noting that the statistical bar of 0.0001 was “exceedingly high.”

The 0.777 hazard ratio is “solidly below” the 0.8 threshold cited by most oncologists as clinically meaningful, Nochomovitz said in a Friday note. A hazard ratio of 0.8 translates into a 20% risk reduction.

Akeso has not shared the overall survival curves, “which will provide very important clues on [overall survival] separation that the singular [hazard ratio] number cannot reveal,” Nochomovitz noted.

One thing that Nochomovitz didn’t articulate was that the small alpha allocation may suggest that Akeso is spending much more alpha on future analysis to give ivonescimab a better chance to hit statistical significance on overall survival potentially at the final evaluation.

However, some industry watchers have voiced concern that with longer follow-up, the chance of meeting statistical significance is lower as later-line therapies may muddy ivonescimab’s first-line effect.

At the end of the day, HARMONi-2 is a China-only study. Summit is conducting the HARMONi-7 trial, a global phase 3 pitting ivonescimab against Keytruda, in first-line, PD-L1-high NSCLC. That trial is well-powered on overall survival and “would in all likelihood hit the primary OS endpoint with a similar ~0.777 HR,” Nochomovitz said.

“We expect US physicians would seize the opportunity to treat 1L-NSCLC pts with a drug that offers a ~22-23% relative risk reduction in death over the current gold-standard Keytruda, assuming proven with statistical rigor in HARMONi-7,” he added.