Point-of-Care Molecular RSV Test

Against the FDA-cleared molecular comparator, sensitivity was about 90% and specificity about 95%. The test was positive in about 22% of evaluable specimens, and the large majority of those positives came from children under 5.

In discordant testing, the two alternative FDA-cleared molecular assays disagreed: one detected RSV in nearly all of the false positives, the other in fewer than a third. Both failed to detect RSV in most of the false negatives. The initial invalid rate was about 4%, falling to about 2% after repeat testing per the instructions.

In the near cut-off study, untrained intended operators agreed with expected results in 100% of low-positive and negative samples. In the reproducibility study, agreement with expected results was 100% across all sites, operators and days. FDA found the device substantially equivalent under the dual-track review, and FDA's CLIA database lists the test as waived.

No numeric acceptance criteria for RSV nucleic acid tests are published in 21 CFR 866.3980, so the results are best read on their own terms. Sensitivity of about 90% rests on about 150 comparator positives, enough to keep its confidence interval reasonably tight. Specificity of about 95% rests on about 550 comparator negatives, so its interval is much narrower. FDA's codified influenza antigen criteria and its influenza guidance benchmarks do not formally apply to RSV, so citing them as the bar here would be misleading.

The discordant results show how uncertain the 'truth' is near the detection limit. One alternative assay detected RSV in nearly all of the false positives while the other detected it in fewer than a third, and most of the false negatives were negative on both alternative assays. This suggests that many disagreements involve low viral loads that some molecular methods detect and others do not, rather than gross errors by either the investigational test or the comparator. The primary estimates are reported against the original comparator, with discordant results in footnotes. Because positives were heavily concentrated in children under 5, the overall sensitivity mainly reflects pediatric specimens.

The test was submitted as a combined 510(k) and CLIA waiver package for dual review. It is intended as an aid in diagnosing RSV infection in children and adults with signs and symptoms of respiratory infection, together with clinical and epidemiological risk factors. A nasal swab is eluted and tested on a small dock, and the operator reads the result visually on a lateral-flow strip. Negative results do not rule out RSV infection.

The clinical evidence had to show performance on fresh, prospectively collected nasal swabs tested within one hour of collection, against a high-performance FDA-cleared molecular RSV assay, across a population spanning young children to older adults. Because RSV circulates in a short winter season, the whole positive sample had to be collected within one season at enough sites.

For the waiver, the package also had to show that untrained intended operators could detect RSV at the limit of detection and that results were reproducible across waived and moderately complex sites, operators and days.

Nucleic acid tests for RSV fall under 21 CFR 866.3980, respiratory viral panel multiplex nucleic acid assay. The regulation's identification lists RSV subtypes A and B among the viral targets, and the device is Class II with special controls. The special control named in the regulation is FDA's Class II special controls guidance for respiratory viral panel multiplex nucleic acid assays. Unlike antigen-based influenza tests, which have minimum sensitivity and specificity written into 21 CFR 866.3328, the RSV nucleic acid regulation contains no numeric performance thresholds. Clinical acceptability is judged in review against the special controls guidance and the comparator used.

The clinical pattern of RSV shapes the study. In this study's evaluable population, the test was positive in about 36% of specimens from children under 5 but only about 6% of those aged 6-18 and about 5% of those aged 19 and older. A prospective all-comers study therefore ends up with most of its positives from young children, and with relatively few positives from the adult part of a 'children and adults' claim.

A test aimed at physician offices and urgent care also needs a CLIA waiver, which requires showing that untrained operators get accurate results. The Dual 510(k) and CLIA Waiver by Application pathway, created under MDUFA III in 2012, lets one set of comparison and reproducibility studies run by intended waived users support both decisions. FDA's guidance describes the current form of these studies: representative waived sites, untrained operators with testing built into daily workflow, and reproducibility panels that include a true negative, a sample close to the LoD and a moderate positive.

• On-site testing by non-laboratory personnel at intended-use sites, with the comparator run by trained staff at a central laboratory, produces one comparison dataset that serves both the 510(k) and the CLIA waiver.
• Testing within one hour of collection keeps the investigational result tied to a fresh specimen and reflects point-of-care workflow. Putting the comparator swab in transport media lets a single central laboratory run a high-performance molecular assay consistently across all sites.
• Using two alternative molecular assays for discordant analysis gives more than one view of each disagreement. Here the two assays themselves disagreed on many of the false positives, which a single tiebreaker would have hidden.
• The near cut-off panel used dozens of replicates per sample per site, close to a hundred low positives in total, so a perfect result is meaningful evidence that untrained operators read faint lines at the LoD correctly.
• Including a moderately complex site in the reproducibility study alongside three waived sites shows that results do not depend on setting, which matters for a test labeled for both laboratory and point-of-care use.