FDA Expects GCP Compliance from All Medical Device Trials
​​[Posted on: Thursday, February 22, 2018]

This week FDA finalized a rule originally proposed exactly five years ago to the date that makes it mandatory for all clinical trials conducted with new medical devices to be compliant with GCP rules similar to those for drugs and biologics. Although the notification refers to non-US clinical trials, but the details of the rule make it clear that this rule applies to trials conducted in the US as well. Data from clinical trials with devices will only be accepted by FDA if the trial was conducted after getting IRB or IEC approval, included informed consent from each study participant prior to participation, and which FDA can verify in an audit. Clinical trials are needed for most PMA and a few 510k devices, and, in the US, are mostly conducted under an IDE and follow GCP. However, many pilot studies with medical devices are conducted without an IDE. Data from pilot studies used to support an IDE for the pivotal studies or to support a 510k or a PMA will only be accepted if it meets the GCP requirements. This condition is a particularly big hurdle for non-US clinical data. Many countries in the world, including those in Europe, do not require GCP compliant clinical trials for approval of most medical devices. Most medical device trials are conducted without IEC approval and without informed consent, making all the data from such trials to be unacceptable to FDA. There is an international standard for medical device clinical trials, ISO 14155:2011, that has multinational acceptance but it is not mandatory in any country. Medical devices are treated almost like drugs under the US laws, in terms of the safety and efficacy data needed to support market approval. All Class III and some Class II devices need data from randomized controlled clinical trials which are practically identical to those for drugs in terms of the operational and regulatory requirements. However, almost all other countries either do not require controlled clinical trials or have relaxed clinical data requirements for medical device approvals. This rule creates a challenge for manufacturers relying on non-US clinical data. Many non-US medical device clinical trials are conducted as post-market trials that do not require regulatory approval. Since the device is approved for use, its application in a patient does not require informed consent as well in those countries. But this data may still provide valuable information about the safety and efficacy of the device. Under the new rule, FDA will only accept any of such post-market data as supporting, hypothesis-building data but not as the basis for market approval. Manufactures will need to conduct new GCP compliant clinical trials with their medical devices for approval in the US. Trials can be conducted outside the US, so long as they are GCP compliant as defined by FDA. The sponsor will need to follow both the local rules and GCP for such trials. FDA said that it will grant waiver from the GCP requirements on a case-by-case basis if the manufacturer can provide a rationale for why the medical device trial was not conducted in compliance with GCP, but such waivers would be hard to come because FDA does not believe GCP compliance is hard for any trial. With this new rule, FDA has harmonized its clinical data requirements for medical devices with those for drugs and biologics. FDA stated in its announcement that it has historically rejected data from trials whose integrity was doubtful on a case by case basis and it is only now formalizing that practice. Although the rule will be applicable a year later, medical device manufacturers would need to reconfigure their clinical data strategies for US approval going forward.