AI Designs Cancer Cure: BBO-10203 Enters Clinical Trials
A new cancer drug candidate — BBO-10203 — designed with the help of artificial intelligence and supercomputers effectively stops tumor growth without causing toxic side effects. A study published in Science shows that the drug targets a difficult-to-attack molecular mechanism in cancer while avoiding the side effects typical of PI3K therapy, such as hyperglycemia.
Scientists from three US institutions — Lawrence Livermore National Laboratory (LLNL) , BridgeBio Oncology Therapeutics (BBOT) and Frederick National Laboratory for Cancer Research (FNLCR) — have announced the development of a new cancer drug candidate: BBO-10203 . The drug blocks tumor growth by interfering with the interaction of two key proteins, RAS and PI3Kα , without causing one of the most common side effects of these therapies, hyperglycemia .
“This is a precise, targeted attack on long-term cancer susceptibility,” said Felice Lightstone , LLNL study co-author.
What makes BBO-10203 unique is how it was created . Instead of years of trial and error in the lab, the team used the advanced LCADD platform, which combines artificial intelligence, molecular modeling, and DOE supercomputers (such as Ruby and Lassen), significantly shortening a process that would normally take years.
- It's about doing things faster without cutting corners, Lightstone explains. We combine cutting-edge DOE supercomputers with cutting-edge chemistry and biology and deliver results.
RAS and PI3Kα are known therapeutic targets, but until now, attempts to block them have been associated with serious side effects. BBO-10203 acts as a “ breaker ”—it cuts off the signal that promotes cancer growth without interfering with insulin signaling , which is new in this class of drugs.
- We have discovered a first-in-class way to block this interaction in tumors without affecting insulin signaling, emphasizes Dr. Dhirendra Simanshu , principal investigator at FNLCR.
In laboratory tests and animal models, BBO-10203 slowed tumor growth in several types of cancer, including HER2-positive and those with PIK3CA and KRAS mutations. It also enhanced the effects of standard therapies used to treat breast, colon and lung cancers.
This gives BBO-10203 the potential to be used as a combination therapy , improving the effectiveness of current treatment in patients with cancers resistant to other methods.
The collaboration began in 2018. Initially, the researchers worked on the so-called “ molecular glue ” that stabilizes the RAS–PI3Kα interaction. However, they soon realized that a better direction would be to break it — that’s how the idea for the “breaker” was born. Over 50 crystal structures were solved during the project, and AI tested millions of potential compounds until the best candidate was selected.
“We’ve built a powerful drug design engine – and we’re just getting started,” Lightstone says.
BBO-10203 is currently being tested in a Phase 1 clinical trial in patients with advanced breast, lung, and colon cancers to assess safety, optimal dose, and preliminary efficacy.
This is the second drug candidate developed by LLNL/BBOT/FNLCR to enter clinical trials — the first, BBO-8520 , began human testing in 2024 and targets KRASG12C mutations.
- We are excited about these results and the potential to expand treatment options for patients with many types of previously incurable cancers, concludes Pedro Beltran , Scientific Director of BBOT.
Source: medicalxpress.com
Updated: 07/07/2025 08:00
politykazdrowotna
