IMDRF Issues Two-Part Guidance for Personalized Medical Devices

November 29, 2022

One of the key definitional considerations is that the PMD encompasses custom-made medical devices – which are for the sole use of a particular individual and address the specific anatomic and physiological features of the patient – and patient-matched devices. Patient-matched devices are matched to the patient’s anatomy within a specific design envelope developed by device scaling by anatomic reference or by the use of the patient’s full anatomic features obtained via imaging. An adaptable medical device also falls under the definition of a PMD under the IMDRF policy, a product defined as a device that consists of mass-produced components that are assembled and/or shaped at the point of care (POC) to suit the needs of the patient/user.

There may be a dearth of national regulatory requirements for PMDs at present, a gap the IMDRF explicitly seeks to fill. The Medical Device Coordination Group released a guidance for custom-made devices in 2021, although this is exclusive of PMDs. There is little indication that the U.K.’s Medicines and Healthcare Products Regulatory Authority has released a guidance on PMDs, although Australia’s Therapeutic Goods Administration published a PMD guidance in August 2022. The FDA has been somewhat active in this area since 2014, when it posted a guidance for custom device exemptions. The FDA also released a discussion paper on point-of-care additive manufacturing of custom devices in 2021.

There are two IMDRF documents that are implicated in this guidance, the N49 guidance for definitions and the N58 for regulatory pathways for PMDs. While the ability to manufacture PMDs at the point of care is a boon for patients, there are risks associated with PMDs that are not associated with conventional mass-produced devices. The scope of this IMDRF guidance is exclusive of devices that incorporate materials of a biological origin or which incorporate a drug substance. Active device components and devices that incorporate software or are software as a medical device (SaMD) are also excluded.

Much of the focus in this guidance is on V&V of the device design envelope, which is defined as the minimum and maximum device dimensions and the limits of the mechanical performance. Among the elements of the design envelope specified in this guidance are:

  • Structural parameters (dimension, area, volume, specific geometric features, and locations of and distances between any screw holes);
  • Material parameters (all raw materials along with their biological, physical and chemical properties;
  • Manufacturing parameters (including post-manufacturing processing, cleaning/sterilization, and packaging/labeling); and
  • Clinical environment parameters.

IMDRF recommends that performance parameters also be evaluated to address any potential differences in device behavior due to differences in device size within the design envelope, but any miscellaneous parameters that could affect device performance should also be subject to V&V. Regarding material parameters, IMDRF recommends that manufacturers take several aspects into account during V&V, such as the potential reuse of materials that are not incorporated into a device. The reuse of materials such as uncured resins and un-sintered powders should be validated by assessing any changes to the material properties due to exposure to elements such as ultraviolet light, heat and oxygen.

The V&V demands on a device may vary between implanted and non-implanted PMDs because of the difference in risk. The worst-case design for a PMD implant may require clinical data from literature reviews, clinical experience data from comparable devices, and non-clinical testing, such as bench testing. Worst-case test sample selection may have to address both inter- and intra-lot variability by demonstrating the reproducibility of the manufacturing process across lots or manufacturing runs when this is feasible.

The manufacturer may have to identify more than one worst-case design within a given design envelope in order to demonstrate that risks are appropriately controlled. V&V must also take into consideration any interface with another device or a non-device product.

Clinical Trial Requirements Likely to Vary

There is a low-level academic debate over the appropriate structure of clinical trials for products related to personalized medicine, including but not limited to this article in Social Science and Medicine. The IMDRF document does little to guide national regulatory authorities on the question of whether a randomized, controlled trial (RCT) would be necessary, although the guidance states that a single-arm study may be permissible when the condition under study is sufficiently rare to preclude the use of an RCT.

The document states broadly that direct clinical evidence from the use of a PMD in humans will be required in order to demonstrate compliance with the essential principles in the context of a high-risk device and a device for which little or no clinical experience is available to draw upon. The wording would seem to give national authorities some leeway to require that manufacturers conduct only animal studies and/or bench testing studies in some instances to validate and verify the design and fabrication of the device. We would point out, however, that IMDRF proposals are not compulsory, and consequently human trials may be required for moderate-risk PMDs under some national regulatory frameworks.