In a point-of-care setting, waiting a week for quality control on engineered cells is not feasible. The FDA appears open to a new regulatory paradigm: the processing machine is treated as a medical device and the RNA cargo as the drug. This bypasses the need for QC on the cell output, as long as the machine operates correctly.
The primary obstacle to creating sophisticated AI models of cells isn't the AI itself, but the data. Existing datasets often perturb only one cellular variable at a time, failing to capture the complex interactions that arise from simultaneous changes. New platforms are needed to generate this multi-dimensional data.
AI isn't just for designing RNA sequences. Its real value is in creating predictive models of complex cellular functions. This allows scientists to determine the precise set of instructions (RNAs) needed to make a cell perform a complex series of tasks, like targeting a brain tumor.
Current cell therapies like CAR-T involve permanent genetic modifications, a risk acceptable only for last-resort cases. By using transient RNAs that disappear after a few days, this new approach eliminates long-term genetic risk, making cell therapies safe enough to be considered for first-line treatment.
Surprisingly, Portal.bio's primary engagement with large pharma companies isn't for cell therapy. Instead, pharma uses the platform to deliver normally impermeable small molecules and peptides into cells for early-stage R&D. This provides a key revenue stream while the cell therapy market is in a "trough of disillusionment."