On April 2, 2019, MDIC’s
External Evidence Methods (EEM) program hosted an Executive & Fellows meeting with about 70 subject matter experts from FDA, industry, and non-profit organizations to discuss the existing methods and identify gaps for the use of external data.
MDIC’s EEM program aims to establish a more predictable pathway for use of external evidence methods, such as new, innovative (Frequentist/Bayesian) methods and the cataloging of existing methods for evidence fusion from data external to a clinical trial. External trial data includes but is not limited to real-world data (RWD), real-world evidence (RWE), engineering modeling and simulation, and similar device clinical trial data to support regulatory medical device decisions and other stakeholder decisions.
The meeting started with updates about MDIC and the program, followed by morning sessions led by FDA representatives highlighting the utility of the Medical Device Development Tools (MDDT) program in promoting innovation in medical device development and regulatory science to help bridge the gap between the research of medical devices and the delivery of devices to patients. Attendees also heard about the progress made through NESTcc initiative towards sustainable generation and use of timely, reliable, and cost-effective RWE throughout the total product lifecycle, using RWD. NESTcc’s Data Network Director Robert Zusterzeel also outlined the efforts made by NESTcc Data Quality and Methods Subcommittees towards developing data quality and methodological standards, design processes for demonstrating conformance to the standards, and recommendations for their implementation.
The morning concluded with a demonstration of one of the novel industry-FDA collaborative projects aimed at transforming the medical device innovation ecosystem with InSilico clinical trials and hearing ase studies focused on leveraging external data to generate evidence. Attendees were also able to learn how FDA and industry collaborated on a mock IDE submission sponsored by MDIC to augment a clinical study with Virtual Patient models. All resources related to this mock submission can be found here.
In the afternoon, the SMEs were split into four breakout sessions, as shown below – each group with equal distribution of FDA and industry members – and addressed different buckets of external data.
- Historical Clinical study data (US/OUS); Led by Chris Mullin (NAMSA)
- Registry Data; Led by Yun-Ling Xu (FDA)
- Modelling and Simulation; Led by Tarek Haddad (Medtronic)
- Admin Data (EHR, Claims etc.); Led by Ted Lystig (Medtronic)
The groups discussed characterizing data type (data quality; access; relevance), cataloging existing statistical methods that can be utilized (Frequentist/Bayesian), describing limitations of existing methods/data type, identifying areas where new methods need to be developed, and identifying examples for the use of external data as evidence.
Please follow MDIC’s EEM webpage for the latest updates.
