In the midst of the government shutdown and with its accompanying lull in new FDA documents, we thought it would be a good time to update our readers on some guidances that were issued late last year. Two such notable draft guidances were the, “Select Updates for Recommendations for Clinical Laboratory Improvement Amendments of 1988 (CLIA) Waiver Applications for Manufacturers of In Vitro Diagnostic Devices” (CLIA Waiver Guidance) and “Recommendations for Dual 510(k) and CLIA Waiver by Application Studies” (Dual Waiver Guidance). CDRH issued draft versions of these guidance documents on November 29, 2017, and exactly one year later the Center reissued updated drafts of these guidance documents. We blogged on the 2017 drafts here. The overall requirements and framework of the guidances have not changed substantively since 2017.
In a recent pre-submission meeting that we had with CDRH, the Center stated that CLIA categorization is one of the most misunderstood regulatory schemes because it is counterintuitive. It has nothing to do with the importance of the test; rather it focuses on a user’s interaction with the test. Given this context, it was somewhat surprising to see the updated drafts cut down significantly in length with the CLIA Waiver Guidance going from 24 pages to 13 pages and the Dual Waiver Guidance going from 49 pages to 12 pages.
Most notably, CDRH removed virtually all of the examples from the 2017 drafts. We have read that this removal will provide additional “flexibility” for manufacturers. But, as many in industry know, examples can be very helpful. Having a list of requirements for waiver studies, as set out in the draft guidances is useful, but illustrative examples give an additional level of interpretation that can be tremendously informative for companies when they are designing their own studies. Rather than removing all examples, it might have been more useful for CDRH to have retained some examples to give some more concrete indications of FDA’s expectations. We also note that FDA has made much more prominent reference to the CLSI guidelines in the 2018 drafts, moving these references from footnotes, in many cases, to the body of the guidance. It is possible that some of the omitted examples can be found in these CLSI guidelines, but many manufacturers, especially small ones, may not have access to these documents because they can be costly to obtain.
While there were many deletions from the 2017 to 2018 drafts, there were also a couple of noteworthy additions. First, in the Dual Waiver Guidance, in the context of explaining the content requirements for a Dual Submission, CDRH states, “[m]ost 510(k)s and Dual Submissions do not include a clinical performance study.” This is an interesting statement to have added. In our experience, in vitro diagnostic devices, more than most types of devices, include clinical study data in their submissions. It is unclear what purpose this new statement serves other than to state that – at least in theory – a clinical study is not always required for a Dual Submission. While it may appear to give additional flexibility, we are unaware of any situations where FDA has granted CLIA waiver without some form of clinical data.
Second, in the CLIA Waiver Guidance, the 2017 draft guidance contemplates two waiver study designs: (1) by comparison to a traceable calibration method; and (2) without comparison to a traceable calibration method. In the 2018 draft guidance, the Center lists four possible options for demonstrating waiver: (1) comparison of a candidate test in the hands of trained and untrained users; (2) assay migration study design; (3) flex and human factors studies alone without additional comparison studies; and (4) comparison of a candidate test in the hands of untrained users compared to a comparator method in the hands of trained users. These added options should give study sponsors additional flexibility when designing their studies. Examples of how the designs of these studies might look in practice would have been helpful, as discussed above.
CDRH is accepting comments through February 27, 2019. The regulations.gov website currently displays a banner stating that the Federal Register feed will not be processed during the shutdown. It does, however, appear that the “Comment Now” function to comment on existing documents, such as the draft CLIA guidances, is functioning. Although, we are unsure whether comments will be reviewed during the shutdown.