MindWalk announces nanobody discovery platform

MindWalk (HYFT) Holdings “announced B Cell Llama(TM), a platform for the discovery of VHH nanobodies: the single-domain antibody fragments increasingly recognized as the ideal molecular building blocks for bispecific antibodies, multispecific therapeutics, and CAR-T cell therapies. Bispecific antibody sales alone are projected to reach US$50 billion by 2030, with the broader cell therapy market adding tens of billions more. B Cell Llama is designed to help partners capture this opportunity with better starting material and AI-guided precision at every step of the discovery process. The announcement is anchored by a study published in Biomacromolecules (American Chemical Society, 2026), conducted by MindWalk in a grant-funded collaboration with Eindhoven University of Technology and Radboud University Medical Center. Three findings from the study illustrate what modular VHH design can achieve: Potency amplified through assembly. Nanobodies displayed in multivalent formats achieved sub-nanomolar potency, 10 to 25 times greater than the same nanobody in monovalent form; Modular reconfiguration defeats resistance. A trivalent VHH construct neutralized variants that escaped all monovalent formats and defeated a pair of approved antibody therapies, demonstrating that rational reassembly of the same building block can restore activity where single agents fail; A potential immune-priming effect. VHH-nanoparticle complexes were preferentially internalized by immune cells associated with long-term immune memory, raising the possibility that certain VHH formats may not only neutralize a target but help the immune system recognize it in the future. The study also produced a finding with direct relevance to AI-guided drug discovery: the molecule with the strongest binding affinity delivered zero functional activity. Binding strength is not a reliable proxy for efficacy. Some conventional discovery workflows depend heavily on binding metrics to rank and select candidates. LensAI(TM) is designed to prioritize function directly.”