More than 29 million Americans—about 9% of the U.S. population—have diabetes, according to the American Diabetes Association (ADA). Racial and ethnic minority groups have higher rates of the disease, which, when not managed effectively, leads to debilitating complications like cardiovascular disease, kidney disease, stroke, and blindness. As it is, however, some patients are well down the road to developing these sequelae before being diagnosed formally with diabetes. Since point-of-care (POC) tests that measure HbA1c are well-established tools for monitoring and managing long-term glycemic control, some healthcare professionals believe using them for diagnosis would catch individuals earlier in the diabetes disease process, enabling timelier treatments and better outcomes.
The Food and Drug Administration (FDA) is now considering an application for what would be the first approved POC HbA1c diagnostic test in the U.S. Waltham, Massachusetts-based Alere’s Afinion HbA1c Dx test would be considered a moderate-complexity test, meaning that sites offering it would be required to perform proficiency testing and adhere to other quality control conditions. Alere currently markets a version of Afinion that is a CLIA-waived test for monitoring HbA1c and not subject to as much scrutiny.
An Ongoing Debate
POC HbA1c diagnostic tests have been an ongoing topic of debate. Advocates say these assays will enable clinicians to make quicker treatment decisions and facilitate productive conversations with patients before they leave their medical appointments, while others—especially the ADA—currently disapprove of POC HbA1c tests for diagnosing diabetes. Since 2010 when ADA first endorsed HbA1c testing as a means of diagnosing the disease, the organization has looked askance at using POC tests for this purpose, initially due to concerns about POC accuracy and later because of lack of required proficiency testing. ADA’s influential Standards of Medical Care in Diabetes in 2016 once again did not recommend POC HbA1c assays because proficiency testing is not mandated for their use. The 2017 update continued this stance, but ADA added that it would consider a future recommendation if proficiency testing is performed and documented (Diabetes Care 2017;40:S11–24).
The International View
Even as the U.S. has yet to take the plunge, entities elsewhere have accepted POC HbA1c tests for diagnosing diabetes. The World Health Organization (WHO) did so in 2011 “provided that stringent quality assurance tests are in place and assays are standardized to criteria aligned to the international reference values, and there are no conditions present which preclude its accurate measurement.” WHO calls for a second test to confirm diagnosis. Also in 2011, an expert group commissioned by the United Kingdom’s National Health Service to analyze issues surrounding POC HbA1c came to a similar conclusion. More recently, in June 2016 Australia’s Department of Health released a protocol that also allows diagnostic POC HbA1c testing with appropriate quality control and external proficiency program participation, and emphasizes the need for test users to get proper training and demonstrate competency.
At a July 2016 meeting of FDA’s Clinical Chemistry and Clinical Toxicology Devices Panel, an ADA leader praised POC HbA1c tests’ role in aiding therapeutic changes and patient education during the ongoing management of diabetes, but explained why ADA does not recommend the assays for diagnosis. “The concerns here are lack of reproducibility and imprecision, the lot-to-lot variations in the reagents and the calibration, and in particular, the lack of proficiency testing at waived sites,” said Robert Ratner, MD, ADA’s chief scientific and medical officer. Those waived sites may include community settings outside of physicians’ offices, where people with positive POC results may not seek or have access to appropriate follow-up testing and care, Ratner added.
A recent commentary by Irish authors noted the global influence of ADA recommendations and challenged the group’s stance against use of POC HbA1c assays for diagnosing diabetes. “Recommending that POC machines should not be used to measure HbA1c in the diagnosis of diabetes, but can be used to monitor patients with diabetes, is illogical,” wrote Seamus Sreenan and William Tormey. They also pointed out that ADA guidelines cite evidence that some POC instruments—including Afinion—meet generally accepted analytic performance criteria. Sreenan and Tormey called upon ADA to update its guidelines to reflect technical improvements when POC assays have been shown to perform to recommended standards and demand obligatory participation in quality assurance programs. This stance would require reversal of CLIA-waived status for all HbA1c tests, the authors wrote (Ann Clin Biochem 2016;53:620). FDA spokeswoman Tara Goodwin would not comment on the status of the Afinion DX application. Alere also declined to speak with CLN.
As the debate continues, some healthcare professionals remain uncomfortable with using POC HbA1c assays for diagnosis. “I have previously been cautious in recommending their use as past devices didn’t perform well, in part because the types of quality control that are applied in the laboratory setting are not applied uniformly with POC assays,” said David Nathan, MD, director of the Massachusetts General Hospital Diabetes Center and professor of Medicine at Harvard Medical School in Boston. “Moreover, POC devices run out of doctors’ offices are often performed by nonprofessionals, making their conduct and results suspect.”
Clinical laboratorians share Nathan’s concerns while being cautiously welcoming of Afinion DX, so long as it would be used appropriately. “The Afinion diagnostic test, if properly done in a moderate-complexity setting with adequate quality control and proficiency testing, is OK,” said Randie R. Little, PhD, professor of pathology and anatomical sciences, director of the Diabetes Diagnostic Laboratory at the University of Missouri in Columbia, and faculty for a recent AACC webinar that explored using POC HbA1c tests for diabetes diagnosis.
While some POC tests perform as well or better than lab methods, they come with concerns common to all POC assays, Little added. “There’s little oversight in many settings that use them, so people may not do a good job of quality control, proficiency testing, and training personnel.”
Little also has specific reservations about the Afinion diagnostic test. It would be “essentially the same test” as the monitoring product, so some healthcare settings might be tempted to use the waived monitoring products for diagnosis, she said.
Alere has recognized a potential for confusing the diagnostic and monitoring tests and their inappropriate use and would take steps to prevent this. The two tests would have different packaging and ordering codes. Additionally, the tests would run on separate analyzers for moderate-complexity and CLIA-waived tests, respectively, said Frank Frantzen, director of research and development at Alere Technologies in Oslo, Norway, during the 2016 FDA panel meeting.
Alere has said FDA approval of Afinion DX would prompt the company to seek a subsequent approval for the assay as a waived test not subject to CLIA requirements, a situation that amplifies Little’s concerns, she said.
One clinician welcomes the prospect of FDA-approved POC diagnostics but shares some of Little’s worries. “I think the [POC] test would facilitate diagnoses of diabetes in many settings. But I’m an academic clinician and would say there are valid concerns,” said Steven D. Wittlin, MD, clinical director of the endocrine-metabolism division and director of the diabetes service at University of Rochester Medical Center in New York. Like Little, he emphasized the need for proper proficiency testing and training for staff administering the test and sees the potential for misuse of the already waived monitoring assay as a diagnostic test.
HbA1c assays, POC or not, are convenient because they can be done at any time of day and without regard to whether a patient has fasted, but they are not good diagnostic tests for certain individuals, Wittlin added. “HbA1c, fasting glucose, and two-hour glucose tolerance tests all identify different populations. These are overlapping but slightly different groups of people,” he explained. “HbA1c, depending on the cutoffs you use, will miss a not-insubstantial number of people.” These include those with hemoglobin disorders and certain hemoglobin variants, and some non-Caucasians whose normal HbA1c concentrations are different.
For these patients, “HbA1c in the normal range is not convincing,” Wittlin noted. Neither are HbA1c concentrations “at the margins,” he said. “If you see a patient with an HbA1c of 6.4% or 6.6%, you should use a confirmatory test to diagnose diabetes.” Indeed, results at or near the widely recognized 6.5% HbA1c cutoff for diagnosing diabetes would be Wittlin’s “major concern” with a POC diagnostic. Otherwise, he added, “I think the test would facilitate diagnosis of diabetes in many settings.”
Dr. Little was a consultant to Alere while the company prepared its submission to FDA.
Dr. Nathan is participating in a study supported by Alere.
Deborah Levenson is a writer based in College Park, Maryland. +Email: firstname.lastname@example.org