Manufacturing Intelligence: What FDA's Real-World Use Case Database Means for CMC

The FDA's expanding embrace of new approach methodologies (NAMs) is forcing pharmaceutical manufacturing and quality teams to rethink how preclinical and CMC data will be generated, validated, and used. Richard Jaenisch, speaking with PharmTech associate editorial director Christopher Cole, lays out both the promise and the unresolved questions that come with it.

Jaenisch, who describes himself as both a liver transplant recipient and a transplant advocate, has previously called the prospect of lab-grown replacement organs "a fantasy." Asked whether the FDA's newly opened pathway for organoid and organ-on-a-chip data changes that timeline, he says the fundamentals haven't moved. "I think we're still pretty unchanged," he notes, pointing to the cost, scarcity, and resource burden of producing something as large and complex as a liver, the body's largest organ. The deeper obstacle, he argues, isn't manufacturing the organ itself but figuring out how to test it before it goes into a patient. Today's 3D-printed organs and organoid/organ-on-a-chip systems are used almost exclusively for drug testing—modeling how a tissue reacts to a compound—not as stand-ins for a transplantable organ. As Jaenisch puts it, once the goal shifts from testing a drug to testing the organ itself, "what happens when the test itself is the test?" With tens of thousands of patients on transplant waiting lists in the United States, he sees real urgency behind the question, even while remaining skeptical that viable replacement organs are closer than roughly a decade out.

Jaenisch also raises a parallel concern about animal testing. Xenotransplantation research—implanting human-derived tissue into animals—has produced mixed results, and he questions its long-term value as organoid and chip-based approaches mature. Cole envisions a path where accumulated data from many printed-tissue drug studies could eventually build enough confidence to extend the same models toward organ-level safety testing, comparing outcomes across dozens or hundreds of trials against real-world human results collected years after a drug's approval. That kind of longitudinal validation, he says, is what might ultimately let regulators and sponsors trust a NAM-derived model enough to support implantation decisions.

On the practical question of what quality and manufacturing teams should do now that FDA's real-world use case database is public and open for comment, Jaenisch pushes back on a wait-and-see posture. He frames the NAMs shift as more than a terminology change, a "rather sweeping" redefinition that alters the funding stream and sequencing of work once handled through traditional preclinical and animal studies. Because NAMs introduce a fundamentally different base material into the development pipeline, CMC teams should examine the roughly dozen practical use-case categories within FDA's broader list of 25 and map each against their own production processes.

Using drug-induced liver injury as an example, he suggests teams ask targeted questions: Do new toxicity data expand or narrow the required testing? Does more data mean more work, or less? Jaenish concludes that the answer depends heavily on data quality, which is precisely why he recommends reviewing the use cases now and submitting comments rather than waiting for the rule to settle on its own terms.