Groundbreaking LLNL and BridgeBio Oncology Therapeutics collaboration announces start of human trials for supercomputing-discovered cancer drug

Newswise — In a substantial milestone for supercomputing-aided drug design, Lawrence Livermore National Laboratory (LLNL) and BridgeBio Oncology Therapeutics (BridgeBio) today announced clinical trials have begun for a first-in-class medication that targets specific genetic mutations implicated in many types of cancer.

The development of the new drug — BBO-8520 — is the result of collaboration among LLNL, BridgeBio and the National Cancer Institute (NCI)’s RAS Initiative at the Frederick National Laboratory for Cancer Research (FNL). In a first for a Department of Energy (DOE) national laboratory, the drug was discovered through DOE’s leadership in high performance computing (HPC) for mission applications, combined with an LLNL-developed platform integrating artificial intelligence (AI) and traditional physics-based drug discovery, and effective partnership with the FNL and NCI.

The drug candidate has shown promise in laboratory testing for inhibiting mutations of KRAS proteins linked to about 30% of all cancers — targets long considered “undruggable” by cancer researchers. The achievement provides hope for broad impact on cancer patients whose tumors harbor susceptible KRAS mutations. This indicates that a computational/AI drug design approach could unlock new insights into the disease and the future of cancer treatment.

"For the DOE complex, this is the first true example that high performance computing can accelerate drug discovery, which is reinforced by making it to human trials,” said LLNL Biochemical and Biophysical Systems Group Leader Felice Lightstone, principal investigator for the project. “Passing FDA clearance to get to human trials already says that the FDA has looked at the design, and we've met all the criteria to have a real-world application. Not only that, but we're taking our basic research in high performance computing and converting it to an application that industry can find useful — this is a true milestone." For more on the milestone and the collaboration, visit YouTube.

After just three years of development, thanks to a legacy of scientific expertise in small molecule research, LLNL’s unparalleled HPC capabilities and the Livermore Computer-Aided Drug Discovery (LCADD) platform, the U.S. Food and Drug Administration (FDA) cleared BBO-8520 for human trials in Dec. 2023. The trial will focus on patients with KRASG12C mutant non-small cell lung cancer and will test the novel inhibitor for safety and efficacy.

The milestone leverages computational drug design capabilities originally motivated by a longstanding DOE-NCI partnership under the Cancer Moonshot, which aims to apply world-class computing resources at LLNL and other DOE national laboratories to advance cancer research and treatment for the public good. Researchers on the project said the partnership’s success in developing a tangible drug candidate underscores the value of uniting expertise from the DOE national laboratories, biomedical research institutions and innovative companies to solve difficult challenges such as cancer. It also indicates the computational approach could save millions of dollars and valuable time — perhaps years — over the traditional drug discovery process, where even promising compounds can fail before reaching human testing.

"Through our collaboration with LLNL and FNL, we’ve reached a significant achievement in drug discovery for inhibiting KRAS mutations that lead to cancer,” said BridgeBio Oncology Therapeutics Chief Science Officer Pedro Beltran. “The development of BBO-8520 could only have been possible through our partnership approach and exemplifies the power of teamwork and innovation. This drug was able to reach human trials in record time and epitomizes the potential of computational and AI-enabled drug design to save time, money and lab resources on new drug discovery. It's a testament to the bright future of our collaboration and the hope it can bring to patients in need. We are proud to continue working alongside LLNL and FNL to push the boundaries of what's possible in cancer research."

In addition to advancing cancer research, LLNL representatives said the milestone is validation that integrating supercomputing with AI-and physics-based computational platforms has the potential to further accelerate small molecule drug discovery and equip DOE, the National Nuclear Security Administration and LLNL with the ability to quickly and routinely develop medical countermeasures for disease or future pandemics, aligning with broader mission focus areas in biosecurity, bio-resilience and national security.

"Through this pioneering partnership, LLNL, the Frederick National Laboratory for Cancer Research and BridgeBio have unlocked a beacon of optimism in the fight against cancer—and this is only the beginning of what we hope will be a long and fruitful collaboration,” said Pat Falcone, Deputy Director for Science and Technology at LLNL. “This rapid journey from computational simulations to human trials for this drug reflects the transformative power of research at the DOE national laboratories and ensures we are better prepared for the next pandemic. This collaboration stands as a testament to the potential of public-private partnerships, and we invite other institutions and companies to join us in the pursuit of innovative solutions through computational and AI-aided drug design. Together, we have accelerated the pace of discovery. 

Initial research and funding for the work came from the RAS Initiative, established by NCI to explore novel approaches to attack proteins encoded by mutant forms of RAS genes and to ultimately create effective, new therapies for RAS-related cancers. Among all forms of RAS proteins, KRAS is the most desired target for therapeutic treatment because of the prevalence of mutant (oncogenic) KRAS in pancreatic, lung and colorectal cancer, researchers said. Mutations of the RAS family of genes are implicated in 20-30% of all cancers — including 95% of pancreatic cancers and 45% of colorectal cancers.

“The Frederick National Laboratory’s longstanding relationships with our colleagues at Lawrence Livermore through the NCI-DOE collaboration set the stage for this public-private partnership with BridgeBio which represents a distinct model for drug development that encompasses world-class computing capabilities,” said Dwight Nissley, cancer biologist and head of the RAS Initiative team at FNL. “By leveraging each partner’s unique strengths, we realized a long-standing goal of the NCI, to develop novel therapeutic options for RAS-driven cancers.”

Normally, RAS proteins receive and follow signals to switch between active and inactive states. Like a light switch, mutated RAS proteins such as KRASG12C can become stuck in an “always on” state that causes cells to grow out of control; forming tumors and the cancer itself to spread to other parts of the body. While existing drug treatments aim to turn this switch off, cancerous cells sometimes still find ways to turn it back on again — making the treatments less effective over time. Unlike most current KRASG12C drugs, BBO-8520 targets KRASG12C in both its active and inactive states, effectively blocking its function and hindering cancer progression. This dual mechanism is believed to offer advantages over existing therapies by addressing some mechanisms of resistance that can develop over time.

Preclinical studies demonstrated promising results for BBO-8520, showing statistically significant inhibition of tumor growth in various cancer models, even in cases where resistance has emerged against other medications targeting the same protein. By binding only to the “off” conformation, existing treatments “result in a patient developing resistance quite quickly,” resulting in cancer progression, said Anna Maciag, a cancer biochemist who led the FNL’s preclinical work on BBO-8520. “This dual inhibitor hits the ‘on’ and ‘off’ states simultaneously, providing 100 percent coverage of the target,” Maciag added.

The studies suggest that BBO-8520 has the potential to be more effective in treating certain types of cancer, particularly those driven by mutations in the KRASG12C gene, which plays a critical role in cancer cell growth and proliferation and for decades has been considered undruggable by the scientific community.

While a typical drug discovery program synthesizes many thousands of promising compounds to design and develop a drug, the team behind this research initially synthesized only a few hundred compounds to discover BBO-8520 — using the AI- and physics-based LCADD platform to narrow down the number of possible successful small molecules for synthesis. With thousands of compounds modeled in silico, the team synthesized about 850 compounds for the entire project.

The joint effort stems from Cooperative Research and Development Agreements (CRADAs) between LLNL and BridgeBio Oncology Therapeutics and between BridgeBio and FNL, aimed at advancing discovery of novel RAS inhibitors for the treatment of cancer. The CRADAs brought the three institutions together to tackle the most challenging aspects of cancer therapy: the long timeframe to bring a drug to market and the high rate of failure for promising drug candidates. LLNL also saw the problem as an important use case for advanced computing. In the coming years, LLNL and BridgeBio will continue to seek new compounds for other targets identified by the RAS Initiative, experimentally testing and validating these compounds, analyzing interactions between compounds and targets and optimizing existing compounds.

“This is a prime example of a truly collaborative spirit with a shared vision: to develop innovative therapies that target RAS-driven cancers with precision and efficacy to give new hope to cancer patients,” said LLNL’s Yue Yang, a computational chemist and a lead researcher on the project. “We want to thank BridgeBio for trusting us to become part of their team, and to have an impact on future candidates as well as their ongoing drug discovery efforts. BridgeBio’s willingness to integrate us with their plans was key to the success of this drug.”

Yang added that the LCADD platform used to discover BBO-8520 is broad and customizable to a wide range of ailments, including other forms of cancer and infectious disease. By collaborating with U.S. industry, scientists could test and improve the platform to combat future biological threats with the assurance of a validated system.

LLNL Innovation and Partnerships Office (IPO) Business Development Executive Yash Vaishnav negotiated the CRADA with BridgeBio subsidiary Theras, as well as the license agreement for the drug candidate with BridgeBio Oncology Therapeutics. Vaishnav also manages the intellectual property portfolio of KRAS inhibitors developed under the CRADA and the relationship with his counterparts at BridgeBio. IPO is the focal point for LLNL’s engagement with industry and aims to accelerate U.S. competitiveness by identifying new economic opportunities and solutions and transferring those to the private sector through licensing or partnerships.