Schrödinger Halts Cancer Drug Trial after Patient Deaths

Schrödinger suspends its cancer drug program after two patients die in a Phase I trial, raising safety concerns over its AI-driven oncology pipeline.

Schrödinger Inc., a biotech based in New York, has paused work on an experimental cancer drug after two patients in an early trial died. The drug, aimed at blocking a newly identified cancer pathway, was in a Phase I trial that was mainly testing safety and dosage. The company hasn’t said what caused the deaths. Investigators noted both patients died soon after their doses were increased, which led the trial’s monitoring board to launch an immediate review. The setback highlights how risky cancer drug research can be, especially when a brand-new type of therapy is tested in people for the first time.

The pause is a blow to Schrödinger’s push to use physics-based AI models to speed up drug design and cut down on failures in clinical trials. The company works with big pharma partners and has drugs in development for cancer, brain diseases, and immune disorders. It has tried to brand itself as a leader in AI-powered biotech. But the pause shows the limits of computer models once a drug is tested in people’s bodies. Phase I trials are mainly about safety, but they’re also where unexpected side effects are most likely to appear especially with new kinds of drugs that animal models can’t always predict. Investors are paying close attention, since cancer drugs make up a big share of Schrödinger’s future revenue hopes.

The FDA and other regulators will review the trial data to decide if the drug can be adjusted or if it has to be pulled for good. Patient groups say stopping a study can slow access to new drugs, but quick action is needed to keep patients safe and protect trust in the process. Schrodinger said it’s staying transparent and will put its efforts into other cancer drugs still in the lab and early testing. The case raises safety concerns and adds fuel to the debate over how AI should be used in drug discovery and what safeguards are needed as machine learning spreads through biotech.