FAQs
FDA Submission and Regulatory Pathway
Q4: What clinical evidence does FDA require for an IVD submission?
​
The FDA’s clinical evidence requirements for an IVD submission depend on the device’s risk classification, intended use, and regulatory pathway. For most 510(k) submissions, the FDA expects analytical performance and clinical performance studies. Analytical performance studies include precision, accuracy, and interference/cross-reactivity data. Clinical performance studies demonstrate that the device performs as claimed in the intended use population. Using FDA-recognized consensus standards (e.g., CLSI Guidelines) for analytical and clinical study designs is highly recommended. Consensus study designs may be modified based on the device’s technological features and intended use, but technical/statistical justifications should be provided for modified study designs.
​
Clinical performance data can often come from studies using clinical specimens archived rather than prospective trials. This is typically appropriate when the specimens represent the intended-use population. For companion diagnostics (CDx) and devices with novel intended use claims, the FDA typically requires prospective clinical evidence with pre-specified statistical analysis plans. The FDA recently shifted its policies to accept Real-World Evidence or Real-World Data more, depending on the relevance and reliability of the data, which could limit the size and scope of prospective studies.
​
The evidence package is expected to support every aspect of the intended use statement. This includes the patient population, specimen type, clinical setting, and performance thresholds. Evidence that doesn’t map precisely to the labeled claim creates gaps in the FDA review.
​
The best ways to ensure robust evidence packages are to research the FDA’s Establishment Registration & Device Listing database for similar or predicate devices, design analytical and clinical studies, prepare a regulatory strategy or road map, and conduct a pre-sub meeting to obtain FDA feedback. The pre-sub process is non-binding; however, it is rare for the FDA to change its guidance unless risk profiles for the device have significantly changed.