This week on Trying to Keep Up, Allison Komiyama and Michael Nilo welcome Alex Cadotte, former FDA reviewer and current Vice President of Digital Health, AI, and Radiology Regulatory Affairs at MCRA, an IQVIA business, for an in-depth conversation about the rapidly evolving world of AI-enabled medical devices.

Alex shares his perspective from both inside FDA and industry on what it takes to bring AI products through regulatory review, why companies still misunderstand Predetermined Change Control Plans (PCCPs), and why FDA's cautious approach to generative AI may actually be exactly what patients need.

The group also discusses one of the first patient-facing large language model (LLM) diabetes management tools, why FDA reviewers sometimes struggle to "let go" of post-market changes, and what companies should consider before investing in a PCCP strategy.

Along the way, they unpack recent De Novo and PMA approvals, cybersecurity expectations, human factors updates, data integrity issues at testing laboratories, and the growing importance of reimbursement planning long before FDA clearance.

As always, there are terrible dad jokes, regulatory tangents, startup war stories, and plenty of honest discussion about where medtech regulation is headed next.

This episode also covers:

• The HepQuant SHUNT® Liver Diagnostic Test PMA

• One of the first patient-facing LLM-enabled diabetes management platforms

• FDA's approach to generative AI-enabled medical devices

• Practical advice for developing AI regulatory strategies

• When PCCPs make sense—and when they don't

• Cybersecurity expectations for modern medical devices

• Human factors considerations for device design

• FDA data integrity notifications and testing laboratory concerns

• New draft guidance on substantial evidence for drugs and biologics

• Breakthrough Device trends and what sponsors should expect

• Why reimbursement strategy should begin before FDA submission