TRENDS & PERSPECTIVES
At the end of January, FDA released its “Guidance for Industry and FDA Staff: Recommendations for Clinical Laboratory Improvement Amendments of 1988 (CLIA) Waiver Applications for Manufacturers of In Vitro Diagnostic Devices.” While FDA has made some revisions in the requirements for applying for a CLIA waiver, industry analysts believe the agency has not done enough to lower the overall burden for obtaining such a waiver.
“There do not appear to be many changes from the prior version,” says Leif E. Olsen, director of regulatory sciences at Hogan & Hartson (Washington, DC). “While I consider the few changes that were made to be an improvement to the 2005 document, I don’t believe the agency took into account many of the stakeholders’ suggestions.”
FDA had released a previous draft of the CLIA guidance in September 2005. In this latest version, FDA has implemented a number of changes including the following: additional emphasis on quality control procedures; recognition that reference methods may not be available for all device types; greater emphasis on intended users during device testing; updated study recommendations; and greater emphasis on scientifically based flex studies and validation and verification studies linked to each device’s risk assessment.
“With this guidance, FDA wants to provide more flexibility, but at the same time a scientifically and medically sound document,” says Carol Benson, associate director at FDA’s division of chemistry and toxicology devices. “FDA also wants more emphasis on the risk analysis in asking applicants to assess what the risks are and how they can be mitigated. In addition, FDA wants to meet the Medical Device User Fee and Modernization Act (MDUFMA) commitments to allow pilot programs in which 510(k) and CLIA waiver reviews can occur simultaneously.”
According to FDA officials, one of the primary differences between the previous and current drafts includes simplifying the graph for allowable total error (ATE) and limits of erroneous results (LER), and removing the Clarke error grid since it was too specific and confusing.
Analysts noted that the new guidance retains the simple empirical method of evaluating accuracy against total error limits, instead of combining separate bias and precision estimates. The description of the error grid approach has also been improved. Instead of using the Clarke error grid as the example of how to plot and classify results into ATE and LER zones, a simpler figure was developed to illustrate how to implement the error grid concept. However, the definition of the ATE zone seems incorrect.
“The guidance says values that fall within the ATE zone are values that can be tolerated without invalidating the medical usefulness of the waived test, which is actually the definition of the zone between the ATE and LER zones,” says Donald M. Powers, PhD, president and principal consultant at Powers Consulting Services (Pittsford, NY). “The ATE zone accommodates the expected uncertainty of the test results, which is measured by comparing the waived method to a reasonably accurate comparative method. The differences are due not only to the inherent analytical error (imprecision and bias) of the waived method, but also the analytical error inherent in the comparative method measurements.”
Another primary difference includes allowing more flexibility for using more banked and prepared samples in clinical studies for obtaining a CLIA waiver. While many in the IVD industry would have preferred to allow half the samples as banked rather than a third, this approach may allow for better positive predictive value (PPV) calculations. However, the question that FDA needs to address in the next iteration is whether PPV calculations are even needed for a waived test.
“If a waived test does in fact perform in a similar fashion to a laboratory test, a case could be made for more banked samples and a limitation in the labeling stating that PPV has not been calculated for this test,” says Glen P. Freiberg, president of RCQ Consulting (San Diego). “The IVD industry trade groups should negotiate further with FDA on this subject with the position that users of this test do not need PPV data, and establishing independent PPV is not one of the criteria for a waiver determination.”
Analysts added that there is nothing in the latest iteration of the CLIA waiver guidance that could not have been documented and proposed 10 years ago. The IVD industry has long held that clearing the air and the path to the waiver category has been a priority for sponsors and patients.
“While this may have been a resource issue with FDA, to the IVD industry, it has been poor prioritization of resources and lack of desire on the part of OIVD and CDC to move more tests into an environment where the benefits outweigh the risks,” says Freiberg.
This guidance can be accessed via FDA’s Web site at www.fda.gov/cdrh/oivd/guidance/1171.pdf.
