FDA proposes more transparent de novo pathway for medical device
The US Food and drug Administration (FDA) has proposed revisions to the de novo pathway in order to make medical device classification more efficient and transparent.
The De Novo pathway is used to review new, low-to-moderate risk devices and determine predicates that assure certain safety and effectiveness measures.
The regulatory agency have published the new de novo Classification Proposed Rule involving procedures and criteria associated with the process. The rule is intended to enable ‘appropriate’ classification of new types of medical devices.
Specifically, the revision offers structure, clarity and transparency on the format and content of de novo requests, and the processes and requirements for acceptance, declining and withdrawal of the requests.
“The regulatory agency have published the new de novo Classification Proposed Rule involving procedures and criteria associated with the process.”
The move comes shortly after the FDA announced plans to modernise the medical device 510(k) clearance pathway.
FDA commissioner Scott Gottlieb said: “Our goal is to make the de novo pathway significantly more efficient and transparent by clarifying the requirements for submission and our processes for review. As a result, we expect to see more developers take advantage of the de novo pathway for novel devices.
“The proposed regulation we’re issuing today-as well as those steps that we announced last week-will help the FDA regulate new technologies in ways that enable us to protect patient safety while promoting innovations that can advance peoples’ health and function.”
The regulatory agency also formally recognised a public database called ClinGen Expert Curated Human Genetic Data, which comprises information on genes, genetic variants and their relationship to disease.
This is expected to enable the development of new, beneficial genetic tests. Developers will be able to use the information in the database to support the validity of their tests, rather than generating data on their own.