Cellular Intelligence, a Boston-based AI-native biotech operating under the legal entity Somite Therapeutics, has acquired global rights to Novo Nordisk's (NYSE: NVO) clinical-stage Parkinson's disease cell therapy program, positioning itself to advance one of the few pluripotent stem cell-derived dopaminergic neuron therapies in human trials. Under the terms of the agreement, Novo Nordisk will take an equity stake in Cellular Intelligence and remains eligible for future milestones and royalties. The deal transfers full operational and development control of the program to Cellular Intelligence, which intends to apply its AI foundation model to accelerate manufacturing and clinical development. Financial terms were not disclosed beyond the confirmation of equity investment, milestones, and royalties.
Deal context
The asset, STEM-PD, is an allogeneic, pluripotent stem cell-derived dopaminergic progenitor therapy intended to restore striatal dopamine signaling through neuronal engraftment — a disease-modifying approach that targets the cellular loss underlying Parkinson's motor symptoms rather than managing them pharmacologically. The program is currently in a first-in-human Phase I/II trial and holds US FDA Fast Track Designation. No clinical data have been publicly disclosed.
Parkinson's disease is characterized by the progressive degeneration of dopaminergic neurons in the substantia nigra, and current standard-of-care therapies, including levodopa-carbidopa combinations, address symptoms without halting neuronal loss. Cell replacement strategies aim to restore dopamine-producing capacity directly, but the field has historically been constrained by manufacturing complexity, scalability challenges, and the difficulty of generating consistent, high-quality differentiated cell populations at clinical scale.
Cellular Intelligence's platform is designed to address the manufacturing and process development barriers that have slowed cell therapy programs broadly. The company describes a two-layer system: a capsule-based experimental engine that generates perturbation data from pluripotent stem cells at what the company claims is more than 1,000 times the efficiency of conventional methods, feeding a large-scale AI foundation model trained to predict and engineer cell fate. The capsule technology has been published in the journal Science, providing peer-reviewed validation of the data generation layer, though no drug programs generated by the platform have been previously disclosed.