FDA 510(k) Glossary · regulatory

PCCP - Predetermined Change Control Plan

PCCP (Predetermined Change Control Plan) is an FDA-authorized framework allowing manufacturers of AI-based and software medical devices (SaMD) to pre-specify planned device modifications and their controls - without requiring a new 510(k) submission for each change. Established under Section 524B of the FD&C Act.
Dataset Insight - 510k Database (166,200 records, 1980–2025)
113 PCCP-authorized submissions in the dataset across 12 FDA panels. Microbiology leads with 37 (AI diagnostics), followed by Radiology (32). Final guidance issued December 2024.

Definition

The Predetermined Change Control Plan (PCCP) is a mechanism introduced by the FDA to address the iterative, data-driven nature of AI/ML-based medical devices. Traditional 510(k) submissions assume a static device design. AI/ML devices are designed to adapt based on real-world data. A PCCP allows manufacturers to pre-specify the nature, scope, and controls for planned modifications, with FDA review included in the original 510(k) clearance.

Legal Basis

The PCCP framework was formally established under Section 524B of the Federal Food, Drug, and Cosmetic Act, enacted as part of the Food and Drug Omnibus Reform Act (FDORA) of 2022. The FDA issued its final guidance in December 2024: Marketing Submission Recommendations for a PCCP for AI/ML-Enabled Device Software Functions.

What a PCCP Must Include

  • Description of Modifications: The specific types of changes the manufacturer anticipates making to the device's AI/ML algorithm, software, or performance specifications.
  • Methodology for Implementing Modifications: Data management practices, re-training procedures, and performance evaluation protocols governing each change.
  • Performance Evaluation Methods: How the manufacturer will assess whether each modification maintains or improves device performance.

PCCP Submissions by FDA Panel

Based on 510k Database records enriched from FDA accessdata.gov, PCCP-authorized submissions span more panels than commonly assumed. Microbiology leads with 37 submissions - driven by AI-based rapid diagnostic and molecular identification software - followed by Radiology with 32, reflecting the well-documented surge in AI imaging clearances.

PanelSubmissions
Microbiology37
Radiology32
Chemistry10
Cardiovascular10
Neurology7
Orthopedic5
Anesthesiology3
General & Plastic Surgery3
Pathology2
General Hospital2
Gastroenterology/Urology1
Obstetrics/Gynecology1

Dataset covers all records including unpublished. Total: 113 PCCP-authorized submissions.

Trend

PCCP authorizations have grown steadily since the framework's formal establishment under FDORA 2022. The distribution across 12 FDA panels confirms that PCCP adoption extends well beyond radiology AI into microbiology diagnostics, cardiovascular decision support, and neurology software.

Search PCCP-authorized devices in the database

FDA References

Related Terms

SaMD - Software as a Medical Device - 510(k) Premarket Notification - Substantial Equivalence - De Novo Classification

Frequently Asked Questions

A PCCP (Predetermined Change Control Plan) is a regulatory mechanism allowing manufacturers of AI/ML-based medical devices to specify in advance the types of modifications they intend to make post-clearance. The FDA reviews and authorizes the PCCP as part of the original 510(k) clearance, eliminating the need for a new submission for each covered modification.

PCCPs are primarily relevant for AI/ML-based Software as a Medical Device (SaMD), adaptive diagnostic algorithms, and devices using machine learning models that retrain on real-world data. They are most commonly used in Radiology, Neurology, and Cardiovascular device categories.

The PCCP framework was formally established under Section 524B of the Federal Food, Drug, and Cosmetic Act, enacted through the Food and Drug Omnibus Reform Act (FDORA) of 2022. The FDA issued its final guidance on PCCPs in December 2024.

Based on 510k Database records enriched from FDA accessdata.gov, 91 submissions in the dataset have FDA-authorized PCCP status. This number is expected to grow as AI/ML device submissions increase.