Future of Traditional Rapid Flu Tests in Doubt as New Review Highlights Improved Performance of New Testing Technologies

With the start of the 2017-2018 flu season approaching, and the upcoming deadline for the U.S. Food and Drug Administration's (FDA's) new minimum performance requirements, there is renewed interest in understanding the performance characteristics of different flu testing methods.

A newly published review in the Annals of Internal Medicine shows that newer-generation influenza tests—including digital immunoassays (DIAs) and rapid nucleic acid amplification tests (NAATs)—perform better than traditional rapid influenza diagnostic tests (RIDTs).

"The results … suggest that traditional RIDTs are likely to be phased out by regulatory agencies like the FDA because of their poor sensitivity, especially in adults," write the authors led by Joanna Merckx, M.D., from McGill University in Canada. "Understanding the performance characteristics of different test methods across different patient populations is important for laboratory directors who must decide on their implementation and clinicians who must interpret their results for patient management."

Rapid diagnosis of flu infection is important to prompt initiate antiviral therapy, limit ancillary diagnostic tests, reduce hospitalizations, institute prompt hospital infection control measures, and cut unnecessary antibiotic use.

"Despite guidance that antivirals should be prescribed for high-risk patients and those hospitalized with clinically suspected influenza before confirmation by diagnostic testing, clinical practice falls far short of the guidelines," writes Michael G. Ison, M.D., from Northwestern University Feinberg School of Medicine in Chicago, Ill. in an accompanying editorial. "The newer DIAs and especially rapid NAATs have the appropriate characteristics and sensitivity to provide actionable results. Increased availability and use of such assays likely will drive more appropriate early use of antivirals, may decrease unnecessary antibacterial therapy, and may improve patient outcomes."

Reverse transcriptase polymerase chain reaction (RT-PCR) is the gold standard for flu diagnosis, but batch testing in specialized laboratories leads to delays in time to results that hamper its clinical usefulness, despite the test's superior sensitivity. Traditional RIDTs that rely on immunoassay to detect viral antigens remain in use, despite concerns over their sensitivity because they can deliver results at the point of care (POC) within 30 minutes. More recently, though, two novel classes of rapid flu assays (automated immunochromatographic antigen detection tests [DIAs] and NAATs) have entered the commercial market with claims of improved sensitivity. The performance of these tests will be scrutinized with the December 2018 implementation of the FDA's new minimum performance standards for rapid flu tests of at least 80 percent sensitivity (with a 95% confidence interval lower bound of 70 percent against an RT-PCR comparator).

The recently published systematic literature review identified published papers comparing commercialized rapid tests to RT-PCR (162 studies, including 130 of RIDTs, 19 of DIAs, and 13 of NAATs). Among the tests evaluated were the BD Veritor System for Flu A+B (BD Diagnostic Systems), the Sofia Influenza A+B Fluorescent Immunoassay (Quidel), the Alere i Influenza A & B (Alere), and the cobas Liat Influenza A/B assay (Roche Diagnostics). Studies covered pediatric and adult populations, as well as POC and hospital testing. Nasopharyngeal swabs were the most commonly used specimens.

The review and meta-analysis found that pooled sensitivities for detecting influenza A were 54.4 percent for RIDTs (below the new FDA minimum performance standards), 80.0 percent for DIAs, and 91.6 percent for NAATs. For influenza B, the pooled sensitivities were 53.2 percent, 76.8 percent, and 95.4 percent, respectively. Pooled specificities were uniformly high at above 98 percent.

"The improved sensitivity of DIAs is likely due to proprietary chemistry innovations and to automated readers that eliminate the subjectivity of an operator visualizing and interpreting test results," the authors note.

Other key clinical takeaways are that:

• Newer-generation tests are acceptable for use in the pediatric populations

• There is a possibility of DIA false negative results in adults and clinicians should consider retesting by RT-PCR if the result could influence patient management

• NAAT testing may be the preferred rapid test for adults.

"Physicians can therefore diagnose influenza with confidence on the basis of a positive RIDT, DIA, or rapid NAAT result," write the authors. "However, the cost of DIAs ($15 to $20 per test) is similar to that of RIDTs, whereas rapid NAATs may cost two- to five-times that amount. Whether the incremental gains in sensitivity of rapid NAATs versus DIAs are worth their added costs will likely depend on the patient populations and clinical contexts in which they are used."

The authors note that industry sponsorship was common in DIA studies evaluated (68.4 percent) and rapid NAAT studies (61.5 percent). Additionally, several review authors reported financial ties to the diagnostics industry.

Takeaway: With improved performance of novel rapid flu tests and the upcoming implementation of the FDA's new minimum performance standards, there are some questions about the long-term commercial viability of traditional immunoassay RIDTs.