CDRH Issues Draft Guidance Regarding Test Reports for Nonclinical Bench Studies in Premarket SubmissionsJune 7, 2018
On May 31, 2018, CDRH issued the draft guidance, “Recommended Content and Format of Complete Test Reports for Non-Clinical Bench Performance Testing in Premarket Submissions.” The draft guidance is intended to provide an outline of the type of information to be included in non-clinical bench test reports included in medical device premarket submissions, including 510(k)s, PMAs, and de novos. There’s nothing earth shattering in this guidance. In fact, in my experience, it mirrors the content of test reports I’ve seen from companies with well-established, FDA-compliant systems. Some smaller companies or those new to FDA regulation may find this guidance helpful in creating a model format for its test reports and ensuring that the content of test reports included in its submissions meet FDA’s requirements in an effort to avoid unnecessary questions.
The guidance recommends including full test reports for each non-clinical bench test included in a submission, with the exception of Special 510(k)s or when a declaration of conformity to a recognized standard is included, for example in an Abbreviated 510(k). This recommendation is consistent with FDA’s refuse to accept checklist, which for 510(k)s states that for each performance test a sponsor must include a “full test report . . . for each completed test. A full test report includes: objective of the test, description of the test methods and procedures, study endpoint(s), pre- defined pass/fail criteria, results summary, conclusions.” The topics listed in the RTA checklist are consistent with the draft guidance, with the exception that the draft guidance omits “study endpoint(s)” from its list of suggested elements.
The draft guidance elaborates on the type of information that should be included in each of these sections. For example, test results should contain all data points from a test, a summary of the individual results (e.g., minimum, maximum, average, and standard deviation), data analysis, and a discussion of any protocol deviations. Most of these sections are logical and consistent with the documentation companies use to prepare these types of reports. What this guidance appears to forget is that many “full test reports” are generated as part of the design and development process as part of the design history file, not specifically for the premarket submission.
While the two are related, there is certain information that the engineers performing the tests may not know. For example, in the “description of test methods and procedures” section, the Agency suggests stating whether the device is the final version of the product or if it is not the final version of the product, an explanation as to why the unit tested is representative of the final unit tested. First, non-clinical bench tests are generally part of design verification, not validation, and therefore, testing can be performed on prototype units, which may differ slightly from the final product. Second, at the time engineering is testing a device, they may not know whether it is identical to the finished product for which a company will submit to FDA. It may be the tentative final design or a prototype. We agree with the Agency that this is important information to know as part of a premarket review, but it seems odd to suggest that this information should appear in the full test report itself – it seems more appropriate to appear in a summary report (discussed below).
Another example of this disconnect appears in the conclusions. The draft guidance suggests that the conclusion should explain how the test supports substantial equivalence, in the case of a 510(k) submission. Again, the connection between an engineering test and a planned 510(k) submission may not be known by an engineer performing testing and writing a test report. The conclusion typically discusses how the test is supportive of a company’s design verification activities because that is the purpose of the test. As discussed above, a conclusion as to how a test supports substantial equivalence seems more appropriate for a summary report, which is developed specifically for a premarket submission, than for the test report. For review purposes, the key should be that this information is included in the application, and not that it appears in the test report.
As mentioned above, in addition to the full test reports, FDA also recommends including a summary report for each test in a premarket submission. According to the guidance, the summary test should include the following information:
- Test performed;
- Objective of the test;
- Brief description of test methods/procedures, including sample size, device tested, and any standards utilized;
- Pre-defined pass/fail criteria, including a clinical/scientific justification for such criterion;
- Results summary, including for quantitative tests the mean, maximum, minimum, and standard deviation, whether the acceptance criteria were met, and a brief explanation of any failures or deviations;
- Discussion of conclusions, including for 510(k) submissions only, how the data support substantial equivalence; and
- Location of where the full test report can be found in the submission.
In sum, the guidance appears to be potentially useful to new, small companies that have less experience with FDA regulatory requirements. For larger, more established companies, it may serve as a good reminder to update their template test reports to ensure all of this information is covered.