NEW YORK – In response to encouragement from a number of stakeholders, the US Food and Drug Administration has broadened its emergency use authorization process for tests to detect the SARS-CoV-2 virus that causes novel coronavirus disease COVID-19. High-complexity labs are now permitted to develop their own tests.
High-complexity CLIA labs in the US became eligible under new FDA guidance issued on Feb. 29 to create their own diagnostic test kits, each of which would technically require its own EUA.
The new guidance says that labs must apply for EUA within 15 days of beginning testing, and public health labs can also still obtain test kits from the US Centers for Disease Control and Prevention.
Labs can also choose to purchase test kits from Integrated DNA Technologies, and there is one lot number of test kits so far that has been validated and authorized by the CDC.
The Association of Public Health Labs and the American Association of Clinical Chemistry had been urging FDA to expand its authorization to allow labs to develop tests.
"I'm thrilled that the FDA is allowing high-complexity laboratories to do this testing," said Carmen Wiley, president of AACC, in an interview.
Labs can now develop what are essentially their own LDTs to use on patient samples. Wiley noted that there are approximately 11,000 high-complexity labs in the US.
"When you are certified as a high-complexity laboratory, it means that the types of laboratory professionals you have on your staff are credentialed to do laboratory-developed tests, which are tests that are not FDA-cleared," Wiley explained. These labs have experts who can design, validate, and report out patient results, she said.
Not all of high-complexity labs have expertise in microbiology and virology, however, and others have estimated there are around 400 labs with the necessary staff and equipment to create a SARS-CoV-2 LDT.
Labs in New York State were the first to obtain an EUA for an LDT, with a test authorized for emergency use shortly after the new FDA guidance was issued on Saturday.
A critical component, however, will likely be the test validation. Because this is a novel pathogen, it is not immediately apparent whether there is a ready supply of inactivated virus or positive clinical samples.
The FDA has already been responsive by allowing high-complexity labs to do this testing, Wiley said, but, one area of weakness is that the current guidance requires labs to validate their tests using RNA or inactive SARS-CoV-2.
"This is somewhat of a sticking point, because this is not easy to get a hold of," Wiley said.
Ideally, labs will also have a "true positive" sample to compare against, but each lab will be doing its validation study in its own way.
"I have confidence that these highly-complex laboratories have the trained experts that will know how to do this, and their main focus is making sure that they are generating quality lab tests," Wiley said.
In a webinar with labs on Monday, Timothy Stenzel, director of the Office of In Vitro Diagnostics and Radiological Health at the FDA's Center for Devices and Radiological Health (CDRH) outlined the requirements for labs to validate their tests in order to get EUA.
In the context of the COVID-19 public health emergency, the FDA has determined that prior public participation for this guidance is not feasible, or appropriate, and issued this guidance without prior public comment, Stenzel noted.
The guidance is immediately in effect. CLIA-certified high-complexity labs can begin using lab-developed tests right away. They are instructed to notify the FDA when they begin testing, and then they need to confirm the first five positive and negative samples with an EUA test.
Labs also must indicate in test reports that their tests have been validated but that independent review by the FDA is not yet complete.
Finally, labs need to submit the paperwork and documentation of the validation process for EUA within 15 days.
The FDA recommended in the webinar that labs use pooled negative samples or artificial matrix to determine the limit of detection of their assays. The agency is defining limit of detection as the lowest concentration in which a least 19 out of 20 replicates are positive.
"We recommend testing a dilution series with three replicates at each concentration, in order to find the range of the LOD for your assay," Stenzel said. This final concentration should then be confirmed with 20 replicates.
The agency will waive the requirement to validate every sample type for which a lab seeks EUA.
Instead, labs are permitted to use only the most challenging of the specimen types as a stand in. For example, rather than test nasal, throat, and nasopharyngeal swabs as well as sputum samples, a lab could test only sputum, as it is the most complex matrix with the most potential challenges in terms of PCR inhibitors.
For clinical evaluation, the agency requires at least 30 contrived reactive specimens, and 30 non-reactive specimens. "The contrived reactive specimens can be created by spiking viral RNA or inactivated virus into leftover clinical specimens," Stenzel said.
Acceptable performance will be considered to be 95 percent agreement – or 19 out of 20 samples – with virus spiked in at ranges spanning 1 to 2 times LOD concentration, and 100 percent agreement for 10 samples spiked in at concentrations higher than that.
Labs can perform in silico inclusivity testing against publicly available SARS-CoV-2 sequences, and all published sequences should be detected with the lab's primers and probes. Cross-reactivity testing can also be done in silico, comparing the primers and probes to common respiratory flora and pathogens.
"Once you have validated your assay according to these recommendations and met the performance criteria ... you can begin testing as soon as you notify us," Stenzel told the attendees, with the notification process involving sending FDA an email.
A completed EUA template should then be submitted 15 days after initiating clinical testing.
Stenzel also said that the agency encourages labs to contact their local public health authorities prior to testing patients, as these authorities may need to do confirmatory testing on presumptive positive samples.
Stenzel said FDA, BARDA, and the CDC will prioritize and coordinate shipments of viral materials to labs when they are ready to validate tests.
There is also viral RNA material available from the National Institute of Allergy and Infectious Diseases' BEI Resources Repository, but Stenzel acknowledged that there is a limitation on availability and requested that people reach out to FDA if they have issues obtaining material. "We will work with you," he said.
Furthermore, "The procedures for validations in this new guidance are suggestions and recommendations – if you have ideas for alternative approaches, please reach out" to FDA, he said, particularly for cases where there is not easy access to viral isolates or viral RNA.
Stenzel said that biosafety level 2 labs should not try to culture virus, and only BSL-3 labs should be dealing with cultured, live virus.
In terms of clinical specimens, and particularly in response to questions during the webinar about verification for commercial test developers, Stenzel said the agency understands that this material is limited in the US and asked that developers send an email request to get on a list and get clinical samples as they become available.
"You can also work interactively with our staff in order to determine if there are alternate pathways that you can take to validate your assay," Stenzel said.
Having FDA-authorized commercial tests will be a benefit in the future, Wiley said, particularly ones that can be put on existing instrumentation. But currently, this isn't an option, so, "In order to meet the demand during this health crisis, we really are going to need to depend on laboratory-developed tests," she said.
Allowing high-complexity labs to jump in is a huge stride, she said. However, more testing means the likelihood of finding more positive cases, but it will be critical to help people understand that a likely increase in the number of cases is due to the ability to test more people not necessarily to the virus spreading.
Regarding the CDC test kits, Wiley said the CDC's experience was not unreasonable. "When you take something out of a research-type environment and push it out into clinical practice, it is not uncommon that you find there was a weakness you weren't aware of. They were very timely in correcting that as quickly as possible."
Stenzel also said the same. "We believe the design of the original CDC assay that was reviewed in the EUA application is solid. We have a lot of confidence in that, and that includes the N1, N2, and N3, and the control reaction," he said. "We've worked closely with them and we believe we have resolved the manufacturing issues," Stenzel added, noting that the agency is also working with IDT and other manufacturers to qualify lots of test kits.
"If you purchase kits that are lot qualified, you will not need to submit your own EUA," Stenzel told the labs on the call, and also noted that the IDT kit is identical in design to the CDC kit and includes the N1, N2, and N3 as well as the control reaction.