The company was aiming for approval in the second half of 2028, paving the way for a market launch in 2029
Hangzhou-based biotech start-up MindRank has progressed to Phase III clinical trials for its weight-loss drug, making it China’s first artificial intelligence-assisted Category 1 new drug to reach this stage.
MindRank announced last month that it had initiated a Phase III clinical trial in China for MDR-001, a small molecule GLP-1 receptor agonists designed with the help of AI. GLP-1 receptor agonists mimic natural hormones to regulate blood sugar and appetite.
According to Niu Zhangming, MindRank’s founder and CEO, the drug was the first AI-assisted, Category 1 new drug to reach this advanced stage in China. The company was aiming for approval in the second half of 2028, paving the way for a market launch in 2029.
The new drug development process to date has taken about four and a half years compared to the seven to 10 typically required to reach this stage, according to Niu, who added that efficiency gains from AI had effectively slashed overall research and development costs by at least 60 per cent.
According to Niu, human specialists can specify a target, typically a protein molecule within the human body directly linked to a disease, and then leverage its proprietary AI tools to rapidly generate potential drugs. This workflow allows researchers to shortlist and select the most promising candidates from an AI-generated pool.
“It’s like overseeing an automated assembly line,” Niu said.
MindRank also developed a specialised biomedical system by leveraging open-source large language models (LLMs) integrated with Retrieval-Augmented Generation (RAG) technology. RAG allows an LLM to look up real-time information from external documents.
This helped improve target research accuracy from an industry average of around 85 per cent to over 97 per cent, according to Niu. Target research accuracy measures how effectively viable targets for treating a disease are identified.
AI can also assess a drug’s safety and efficacy through advanced predictive models, tackling complex calculations that were previously beyond human capability, according to Niu.
However, despite AI’s advancements, humans remain an indispensable part of the process. Many intermediate steps still necessitate manual software operations. “It still takes a human to tie the entire process together,” Niu noted.
Experts still steer high-level strategic planning, making critical decisions such as which targets to prioritise and whether to optimise existing compounds or engineer entirely new ones from scratch, according to Niu.
Last December, US-based Generate:Biomedicines announced plans to launch two global Phase III clinical trials for an AI-engineered antibody to treat asthma.
Despite the progress, Niu warned that at the current stage AI may not disrupt the life sciences in the same way it has other fields.
“The core issue is that the trial-and-error cycle is far too long,” said Niu, lamenting a problem that plagued biopharmaceutical behemoths and start-ups alike, as evidenced by the industry’s protracted validation timelines.
“For AI4S to deliver meaningful impact in life sciences, we still need to go through a longer cycle of testing and evaluation,” Niu said.
来源：《南华早报》