The need for self-administered surveillance testing finally has a few candidates, thanks to labs and test developers across the globe, and the U.S. FDA is keen on exploiting the opening. Tim Stenzel, director of the FDA’s Office of In Vitro Diagnostics and Radiological Health, said on the agency’s Sept. 2 testing town hall that the agency is interested in a test intended to be self-administered multiple times compared to a test validated under a single test approach, a flexibility that may prove critical in advancing the U.S. approach to testing for the COVID-19 pandemic.
Stenzel said the agency is working on a “serology non-lab or home collection template,” and on a serology home test/non-lab test template in addition to the raft of testing templates already in place. He said the recommendations for validation seen in the existing templates are “good starting points for discussion,” but that the agency is open to other approaches to validation.
For home use/non-lab molecular and antigen tests, the FDA is open to serial testing protocols, which would boost percent positive agreement (PPA) or assay sensitivity. Stenzel said this applies to any test to be administered in the first five to seven days after the onset of symptoms, although the agency is interested in a serial testing approach for asymptomatic individuals as well.
A test that could be done routinely in the home may be useful for picking up asymptomatic carriers as well as those who are symptomatic, but whose symptoms are not clearly those of the COVID-19 rather than another respiratory virus. Stenzel said the agency’s interest in this scenario hinges on the notion that a single administration of a test that might not perform well enough for the FDA to authorize as a one-off test might nonetheless prove useful if the combined results of two passes with that tests proved sufficiently accurate. This could be done on consecutive days or on non-consecutive days over three days.
Stenzel said the direct antigen tests the agency has authorized to date have often been used in those with high viral loads, and these tests have all exceeded the agency’s benchmarks for sensitivity and specificity. For a test that is administered on consecutive days or with a single elapsed day between tests, the agency is open to granting an EUA for that test with less vigorous performance numbers than has been the case to date for direct antigen tests.
One of the callers to the Sept. 2 town hall inquired into EUAs that were granted for antigen tests where the sensitivity was determined from samples obtained in high-prevalence settings, such as greater than 30%, despite that the presumed national average is closer to 8% to 10%. Stenzel said prevalence for direct antigen test may not have much impact on performance characteristics or the test’s specificity. Validating this type of test in a high-prevalence environment allows developers to more rapidly accumulate positive and negative information, which allows FDA to make decisions more quickly “and get those tests out there,” Stenzel said.
The FDA can do a better job of evaluating a test in this scenario if patient-level data are obtained that include the number of days elapsed between the onset of symptoms and the administration of the test, however. That elapsed-days figure gives the FDA staff some idea of what the expected detection rate is for a test, although Stenzel said the agency will want confirmation of a positive result via molecular testing.
Stenzel noted that the FDA does not require that test developers evaluate the Spanish-language versions of their instructions for use, although the sponsor should evaluate their IFUs in English with an individual who speaks English as a native tongue.
NCI straining under validation load
One of the callers said the National Cancer Institute (NCI) had previously turned around its review of serology test validation in an average of 17 days, but that this figure might be closer to eight weeks at this point. Another caller said his company’s test has been in queue at the NCI for more than three months, and Stenzel said the average time from receipt of the test kit to generating results is 26 days, while average time from receipt to public posting is 58 days. He said the FDA contacts the developer before the NCI posts the data, but recommended the sponsor contact the FDA if the device has been at the NCI for more than 26 days.
NIH announces new testing contract awards
The U.S. National Institute of Health said Sept. 2 that it has awarded nearly $130 million to test developers under the Rapid Acceleration of Diagnostics (RADx) Initiative, a sum that will be divided among nine test developers. The nine recipients will use the awards to further develop and scale up production of novel testing technologies, such as a real-time polymerase chain reaction (RT-PCR) test that runs on batteries in a unit smaller than a computer printer.
Another test that will be aided by the awards is a lateral flow immunoassay strip characterized as similar to a home pregnancy test, and a sample concentration method that can be deployed to improve the sensitivity of conventional testing tools.
The NIH said this round of awards followed a previous round unveiled July 31, which funded the efforts of seven companies to the tune of nearly $249 million. The latest round of awards includes one for the Broad Institute of Cambridge, Mass., for scaling up PCR test processing volume from 25,000 tests per day to 100,000 per day, and an award to Ceres Nanosciences Inc., of Manassas, Va., for the Nanotrap particle for viral concentration.
The NIH statement indicated that the awards should have a significant impact on testing volumes in the U.S. in the month of September.