Well Isn’t that Special: An Assessment of the Special Control Associated with Simple Point of Care COVID-19 Antigen TestsMay 19, 2023
As we noted in our previous blog post available here, Quidel’s Sofia 2 SARS Antigen+ FIA, Sofia 2 SARS Antigen+ FIA Control Swab Set’s De Novo, now opens the door for follow-on 510(k) submissions that declare this product as their predicate, if the product meets the established Special Controls as noted in the reclassification order. In this blog we examine the Special Controls put in place to mitigate false results, incorrect interpretation of results, and incorrect operation of the device.
Design Verification and Validation Special Controls
Analytical testing includes many of the same tests already required for an EUA:
- Limit of Detection (LoD)
- Inclusivity, including relevant variants
- Microbial interference
- Interfering substances
- Hook effect
- Inclusivity of relevant circulating variants
- Specimen stability
- Reagent stability
- Flex studies
However, there are some additional studies noted, specifically site-to-site reproducibility, within-lab precision, carryover and cross contamination, and competitive inhibition. Previously, the EUA template only required these studies for Multiplex Respiratory panels in which a new instrument was used with the proposed device; however this De Novo was for a simple point-of-care device to detect SARS-CoV-2 viral targets only.
Final Release Criteria Special Controls
Sponsors will need to provide the final manufacturing release criterion with evidence that lots released at the extremes of the specification will still meet the claimed device performance (i.e., analytical, clinical and stability). The Special Controls are silent on what would constitute appropriate levels of evidence necessary to satisfy this criterion.
Collection Device Special Controls
Sponsor will need to select a collection device that is either FDA-cleared, -approved, or -classified as 510(k) exempt (either as a standalone collection device or as part of a system) or alternatively the collection device can be cleared within the premarket submission for the POC test.
Clinical Testing Special Controls
In the Special Controls, there are two different criteria a sponsor can meet for their clinical study, each resulting in different indications for use:
|If your clinical study establishes a lower bound of the two-sided 95% confidence interval of…|
|Greater than or equal to 80% PPA||Negative results are considered presumptive|
|Greater than or equal to 70% PPA||Negative results are considered presumptive, and sponsors need to include serial testing requirements in the indications for use.|
Serial testing was defined as testing symptomatic individuals twice over three days with at least 48 hours between tests, which is in accordance with study findings from the National Institute of Health referenced here and current EUA labeling for serial testing.
The other requirements, listed below, for the clinical study are similar to previous recommendations included in the Antigen EUA template with some important subtle nuances to consider:
- prospective multi-site clinical study,
- geographically diverse intended use population,
- consistent with the intended use population and intended operators,
- conducted in a representative intended use setting,
- performance estimates derived for each claimed specimen, and
- compare results of the candidate device to an” FDA accepted” molecular comparator method.
For EUAs, FDA recommended the clinical evaluation be done at one to two sites; however, the Special Controls explicitly states the testing should be done in geographically diverse intended use populations, to ensure the device results represent “all present, circulating strains of the SARS-CoV-2.” It is not clear how sponsors are to meet this special control as prospective studies are executed in the months prior to the FDA review, given the time necessary to prepare for and submit a 510(k) in addition to the FDA’s premarket review timelines. Even though only two sites are required, sponsors should consider including a sufficient number of sites to ensure geographic diversity and ability to capture all circulating strains in the study.
The selection of operators is another critical component of the study design and should be operators that would be typically expected to run simple tests in a POC environment and are not hired just to execute a clinical study. We note the Special Controls indicated that the POC study needs to be conducted in the representative intended use setting so consider incorporating the testing into the operator’s existing workload.
Notably absent are details regarding what is considered an “FDA accepted” molecular method. In the EUA template the FDA indicated sponsors should use a “RT-PCR which used chemical lysis step followed by phase extraction of nucleic acid (e.g., silica bead extraction) and reports cycle threshold values and be one of the more sensitive RT-PCR assays.” We note the FDA has granted a De Novo and subsequent clearance for SARS-CoV-2 RT-PCR test which would seem to be the logical comparator of choice, however discussing the proposed comparator or even the use of a composite comparator with the FDA prior to initiating a clinical study is highly recommended. Sequencing can be an important tool to understand the distribution of variants in the clinical study especially given the fact sponsors will be required to include a statement in their labeling regarding the predominant strain in the study.
Software Special Controls
If the device includes software or is an instrument with software, the software required under the special controls is consistent with the typical software documentation provided in 510(k) submissions. In addition, similar to what was required for an EUA, sponsors should evaluate cybersecurity of their system to ensure user and patient safety in the intended use environment.
Continuous Monitoring and Post Market Reporting Special Controls
Within the Special Control for Risk Analysis are details regarding continuous monitoring. The submission will need to include a “detailed description of a protocol for continuous monitoring, identification, and handling of genetic mutations and/or novel isolates or strains.” The Special Controls provides the examples on what sponsors might do to monitor:
- regular review of published literature, and
- periodic in silico analysis of target sequences to detect possible mismatches.
Is this really as straightforward as reviewing literature and doing periodic in silico analysis? When reviewing FDA’s Policy for Evaluating Impact of Viral Mutations on COVID-19 Tests (Revised) dated January 12, 2023, the agency previously stated monitoring the impact on antigen tests is not as straightforward as monitoring the impact on molecular tests because antigen tests do not directly target the genetic sequence. The Special Controls seem to leave it up the sponsor to determine if monitoring is required for all new circulating variants or only if they reach a threshold level of prevalence in the United States. It is unclear what specific testing would be done to demonstrate the impact on established performance. We know that FDA has asked sponsors to conduct Limit of Detection Studies with specific variants, but it is not clear what FDA’s expectations are for demonstrating equivalence.
The monitoring protocol also needs to include plans to update labeling with the additional performance data. This data and any additional FDA-directed data analysis must be provided to the FDA, when requested or be made available during a routine evaluation. FDA can request that sponsors provide this data within 48 hours of the Agency making said request. While it is true sponsors will have the data on file, responding and providing the data within a 48-hour time-period can create stress in any organization. And if you think a 48-hour turn-around time is fast, if novel respiratory pathogen strains or isolates impact the stated performance of the device, the data needs to be sent to the FDA immediately (although the specific trigger for such “immediate” notification is “unclear”). While immediately is not defined, it could suggest a shorter turnaround time than the previously stated 48 hours. The Special Controls is silent on defining to what extent a reduction in performance would require immediate reporting to the FDA. In the prior FDA EUA guidance, a reduction of test performance of 5% or more from the established performance or a drop below the stated performance recommendation in the EUA templates were identified as thresholds for impact on performance, however it is unclear if this same threshold would be applied.
Emergency Analytical Reactivity Post Market Obligations Special Controls
Interestingly the Special Controls also add details regarding post market obligations if a new public health emergency is declared for the analyte detected and if FDA has characterized samples available for testing. In this case, sponsors will have 30 days to complete testing in accordance with an FDA protocol. Within 60 days and continuing for 3 years, sponsors will be required to include the results of that testing in the device’s labeling along with details about the samples used for the testing.
Labeling Special Controls
The labeling may seem like the easiest Special Control for demonstrating compliance. Here we have broken out the requirements for the Indications for Use, Quick Reference Instructional Sheet, and Device Labeling.
Indications for Use
The indications for use should include the following:
- analytes detected,
- specimen types tested,
- indication for testing of respiratory specimens,
- clinical indications for which the test is used,
- specific intended population(s),
- intended use locations including testing location(s),
- statement regarding positive results,
- for symptomatic, number of days of symptom onset,
- statement regarding negative results,
- statement regarding the predominant strain during performance testing, and
- depending on the results of the clinical study serial testing requirement.
Quick Reference Instruction
The Quick Reference Instructions (QRI) should include, at a minimum, the following:
- name of the test,
- intended use of the test,
- easy to follow step-by-step instruction for controls,
- easy to follow step-by-step instructions for each sample type,
- graphic illustrations of each step,
- results interpretation guidance,
- warning and limitations,
- toxicology information and safety considerations for any hazardous materials,
- troubleshooting or frequently asked questions, and
- how to get technical assistance (e.g., Customer Support Help line).
Labeling Instructions for Use
The labeling should include the following:
- detailed device description,
- detailed description of performance characteristics,
- detailed descriptions of the test procedure(s), interpretation of results, and acceptance criteria for QC testing,
- statement regarding positive results and patient management decisions,
- detailed instructions on how to minimize exposure to infectious reagents included in the device,
- detailed instructions on how to minimize false positive results, and
- limiting statements regarding results and performance characteristics with the predominant strain.
As we stated previously, this submission took well over nine (9) months to review and develop the Special Controls. After reviewing the Special Controls, we understand what took so long. Unlike like the definition of “simple test,” the Special Controls are anything but simple. Since the De Novo reclassification order on March 8, 2023, the FDA has recently posted the Quidel’s decision summary, which happed at warp speed as compared to this De Novo from July 2022, which still has yet to have its decision summary posted. Stay tuned for a future post evaluating the details of the Quidel’s decision summary.