Beijing-based biotech unicorn METiS TechBio has rolled out what it calls the world's first AI-powered nano-delivery platform, NanoForge, which could reshape the way medicines are developed and delivered, paving the way for reprogrammable drugs.

At the launch event in Beijing on Tuesday, cofounder and CEO Chris Lai compared the breakthrough to aerospace engineering. "NanoForge is like the SpaceX of drug delivery. It is both the rocket that delivers drugs precisely and the satellite that speeds up new drug development. This could solve one of the toughest problems in medicine, which is to get treatments exactly where they need to go, without harming healthy tissue," Lai said.

Drug delivery has long been one of the industry's hardest puzzles. The rollout of mRNA COVID-19 vaccines in 2020 showed the promise of lipid nanoparticle systems. These microscopic fat droplets can shield fragile molecules like mRNA and ferry them through the body. But applying them across multiple organs and scaling them for broader drug innovation have remained elusive.

Lai explained that traditional drug delivery research relies heavily on trial and error, a slow and costly process with low success rates. For example, scientists may have to sift through tens of thousands of lipid combinations to identify a single viable LNP formula.

NanoForge uses a generative AI model called PhatGPT, which designs new lipid molecules atom by atom, much like ChatGPT generates text word by word. It evaluates multiple parameters at once to optimize performance, then validates results through automated lab experiments. "For nucleic acid drugs, NanoForge significantly improves mRNA encapsulation and targeting. What once required screening tens of thousands of formulas can now be achieved in weeks," said Zhang Xiaoju, head of RNA innovation at METiS.

METiS has developed three core solutions: AiLNP for nucleic acid delivery, AiRNA for mRNA sequence design and AiTEM for small-molecule drug formulation.

So far, the company says Nano-Forge has produced more than 10 million lipid structures and collected around 100,000 data points to train its models. The platform has already shown targeted delivery in eight types of human organs and tissues, including the liver, lungs, tumors and the central nervous system.

To secure its edge, METiS has filed or obtained more than 100 patents.

Its drug pipeline includes more than 10 advanced programs. Seven have been validated in preclinical studies, while four are already in clinical trials. The lead candidate, an oncology therapy, has reached the pre-NDA stage, that is the final step before applying for regulatory approval.

Drug development usually drags on for more than a decade and can cost upward of $1 billion. Founded in 2020, METiS seems to have broken that cycle, moving more than 10 programs into advanced stages within five years by combining AI with nanotechnology.

Still, experts urge caution. "The ultimate goal of innovative drug development is to address unmet clinical needs, deliver strong efficacy and safety data in trials, and secure regulatory approval," said Fang Yang, executive director for healthcare and biotech at CEC Capital. "From that perspective, AI-driven drug platforms still need time and clinical validation."

Lai acknowledged that METiS is in a "critical stage, moving from pre-commercial to commercial", with the focus now on ensuring both the platform and its pipeline generate sustained value.

The business case appears compelling. According to consulting company Frost & Sullivan, the global market for intelligent drug-delivery solutions exceeded $18 billion in 2023 and will reach $52 billion by 2030, growing at a compound annual rate of 17.2 percent. Nano-delivery technologies are expected to make up 42 percent of that market by 2025, or roughly $7.8 billion, underscoring its growing importance in the field.

Investor interest has followed. In August, METiS closed a 400 million yuan ($55.6 million) Series D financing round. The fresh capital will support upgrades for NanoForge, pipeline expansion, global partnerships and new talent recruitment.

lijing2009@chinadaily.com.cn