Galux Unveils Breakthrough in AI-Driven Antibody Design with Broad Therapeutic Implications

Galux's Groundbreaking Advances in De Novo Antibody Design

Galux, a pioneering biotech company from South Korea, has announced exciting new findings that significantly enhance its AI-driven antibody design platform, known as GaluxDesign. This breakthrough has showcased the ability to generate unique antibodies that not only meet precise structural requirements but also demonstrate functional efficacy across a diverse array of targets, thereby opening floodgates of potential therapeutic avenues.

In earlier studies, Galux introduced antibody designs targeting six therapeutic indications, marking a substantial step in the field of de novo antibody design---a method that generates antibodies entirely through computational techniques. This new paper extends this study to include two additional targets: CD98hc and IL-11, leading to a total of eight targets that also include established oncological markers like PD-L1, HER2, and EGFR(S468R mutant). Notably, IL-11 is particularly intriguing as it has no known antibody complex structure, thereby underscoring the platform's innovative capability to engage previously unwieldy targets.

One of the most compelling aspects of this study is the validation process. For the designed antibody against PD-L1—dubbed GX-aPDL1-3—Galux utilized cryo-electron microscopy (cryo-EM) to experimentally determine the antibody-antigen bonding structure. The results showed a significant distinction from existing PD-L1 antibodies recorded in the Protein Data Bank (PDB), revealing only a 43% sequence identity in CDR loops. This novelty was coupled with a remarkably close structural match to the experimentally determined form, marked by an interface Root Mean Square Deviation (RMSD) of just 1.1 Å, a testament to the atomic-level precision achievable via this methodology.

The antibodies were initially validated in a single-chain variable fragment (scFv) format before transitioning to the IgG format where their stability and binding characteristics were thoroughly evaluated. Remarkably, favorable results were found across the five primary targets, with subtype specificity and appropriate binding shown to be effectively preserved. The designed solutions targeting FZD7 and EGFR-S468R exhibited this rigor, while the IL-11 binders demonstrated versatility by successfully engaging multiple biologically relevant epitopes.

These technical accomplishments highlight the remarkable potential of Galux's antibody design platform, especially for therapeutic applications targeting complex disease states. The implications of the eight targets span across numerous areas like oncology, metabolic disorders, and immune diseases, pinpointing an opportunity to address challenges associated with conventional antibody development methods.

Galux's CEO, Chaok Seok, expressed optimism regarding these findings. “Our results suggest that antibodies designed entirely through computational means can not only be novel but also possess atomic-level accuracy. This opens the door to broader applications in diverse therapeutic areas,” he noted, emphasizing the company's aim to revolutionize the drug design paradigm.

Moreover, Galux remains committed to pushing the boundaries of its technology to tackle previously undruggable targets, also engaging in discussions with global pharmaceutical companies to explore joint ventures. This effort highlights the ambition of the company to establish a new frontier in medication development—one not based on happenstance but rather on precise and efficient design, ultimately benefiting patients in critical need around the world.

As the world of drug development continues to evolve, Galux’s advances signify a promising future where AI and computational methodologies lead the charge. The full findings were published on BioRxiv, demonstrating a substantial leap forward in biotech innovation and giving hope for more effective treatments in the near future.