Use NEJM Group’s information and services to prepare to become a doctor, accumulate knowledge, lead a healthcare organization and promote your career development.
It’s time to change our view of the sensitivity of the Covid-19 test. The U.S. Food and Drug Administration (FDA) and the scientific community are currently almost exclusively focusing on detection sensitivity, which measures the ability of a single detection method to detect viral proteins or RNA molecules. Crucially, this measure ignores the context of how to use the test. However, when it comes to the widespread screening that the United States so desperately needs, context is crucial. The key question is not how good a molecule can be detected in a single sample, but can the infection be effectively detected in the population by reusing the given test as part of the overall detection strategy? The sensitivity of the test plan.
Conventional testing programs can act as a kind of Covid-19 filter by identifying, isolating and filtering out currently infected people (including asymptomatic people). Measuring the sensitivity of a test plan or filter requires us to consider the test in the context: frequency of use, who is used, when it works during the infection process, and whether it is effective. The results will be returned in time to prevent spread. 1-3
A person’s infection trajectory (blue line) is shown in the context of two surveillance programs (circles) with different analytical sensitivity. Low analytical sensitivity assays are often performed, while high analytical sensitivity assays are rare. Both test schemes can detect the infection (orange circle), but despite its lower analytical sensitivity, only the high-frequency test can detect it within the propagation window (shadow), which makes it a more effective filter Device. The polymerase chain reaction (PCR) detection window (green) before infectivity is very short, and the corresponding window (purple) that can be detected by PCR after infection is very long.
Thinking about the effects of repeated use is a concept familiar to clinicians and regulatory agencies; it is invoked whenever we measure the efficacy of a treatment plan rather than a single dose. With the accelerated development or stabilization of Covid-19 cases around the world, we urgently need to shift our attention from narrow attention to the analytical sensitivity of the test (the lower limit of its ability to correctly detect the concentration of small molecules in the sample) and the test The program is related to the sensitivity of detecting infections (infected persons understand the possibility of being infected in time to filter them out of the population and prevent spreading to others). The point-of-care test, which is cheap enough and can be used frequently, has high sensitivity for detecting infections that take timely action without having to reach the analytical limit of the baseline test (see figure).
The tests we need are fundamentally different from the clinical tests currently in use, and they must be evaluated differently. The clinical test is designed for people with symptoms, does not require low cost, and requires high analytical sensitivity. As long as there is a test opportunity, a definite clinical diagnosis can be returned. In contrast, tests in effective surveillance programs to reduce the prevalence of respiratory viruses in the population need to return results quickly to limit asymptomatic transmission, and should be cheap enough and easy to perform to allow frequent testing—multiple times a week. The spread of SARS-CoV-2 appears to occur a few days after exposure, when the viral load reaches its peak. 4 This point in time increases the importance of high testing frequency, because testing must be used at the beginning of infection to prevent continued spread and reduce the importance of achieving the very low molecular limit of standard testing.
According to several criteria, the benchmark standard clinical polymerase chain reaction (PCR) test fails when used in surveillance protocols. After collection, PCR samples usually need to be transported to a centralized laboratory composed of experts, which increases costs, reduces frequency, and may delay results by one to two days. The cost and effort required to test using standard tests means that most people in the U.S. have never been tested, and the short turnaround time means that even if current surveillance methods can indeed identify infected persons, they can still spread the infection for several days. Previously, this limited the impact of quarantine and contact tracking.
The Centers for Disease Control and Prevention (CDC) estimates that by June 2020, the number of Covid-19 cases detected in the United States will be 10 times the number of cases detected. 5 In other words, despite monitoring, today’s testing schemes can only detect a sensitivity of 10% at most and cannot be used as a Covid filter.
In addition, after the transmissible stage, the RNA-positive long tail is clearly described, which means that, if not most, many people use high analytical sensitivity to detect infection during routine surveillance, but they are no longer infectious at the time of detection. Detection (see picture). 2 In fact, a recent survey by The New York Times found that in Massachusetts and New York, more than 50% of infections discovered through PCR-based surveillance have a PCR cycle threshold in the middle of 30s to 30s. , Indicating that the viral RNA count is low. Although low counts may indicate early or late infection, the longer duration of RNA-positive tails indicates that most infected people have been identified after the infection period. Crucial to the economy, it also means that even though they have passed the infectious transmission stage, thousands of people are still quarantined for 10 days after the RNA-positive test.
In order to effectively stop this pandemic Covid filter, we need to test it to enable a solution that catches most infections but is still infectious. Today, these tests exist in the form of rapid lateral flow antigen tests, and rapid lateral flow tests based on CRISPR gene editing technology are about to appear. Such tests are very cheap (<5 USD), tens of millions or more of tests can be performed every week, and can be performed at home, opening the door to an effective Covid filtering solution. The lateral flow antigen test has no amplification step, so its detection limit is 100 or 1000 times that of the benchmark test, but if the goal is to identify people who are currently spreading the virus, this is largely irrelevant. SARS-CoV-2 is a virus that can grow rapidly in the body. Therefore, when the benchmark PCR test result is positive, the virus will grow exponentially quickly. By then, it may take hours instead of days for the virus to grow and reach the detection threshold of cheap and fast instant testing currently available. After that, when people get positive results in both tests, they can be expected to be infectious (see figure).
We believe that surveillance testing programs that can cut off enough transmission chains to reduce community transmission should supplement rather than replace our current clinical diagnostic tests. An imaginative strategy can take advantage of these two tests, using large-scale, frequent, cheap and rapid tests to reduce outbreaks, 1-3 using a second rapid test for different proteins or using a benchmark PCR test to confirm positive result. The public awareness campaign must also convey any kind of negative test bill that does not necessarily imply health, in order to encourage continued social distancing and the wearing of masks.
The FDA’s Abbott BinaxNOW Emergency Use Authorization (EUA) in late August is a step in the right direction. It is the first fast, instrument-free antigen test to obtain an EUA. The approval process emphasizes the high sensitivity of the test, which can determine when people are most likely to spread the infection, thereby reducing the required detection limit by two orders of magnitude from the PCR benchmark. These rapid tests now need to be developed and approved for home use in order to achieve a true community-wide surveillance program for SARS-CoV-2.
Currently, there is no FDA pathway to evaluate and approve the test for use in a treatment plan, not as a single test, and there is no public health potential to reduce community transmission. Regulatory agencies still only focus on clinical diagnostic tests, but if their stated purpose is to reduce the community prevalence of the virus, new indicators can be applied to evaluation tests based on the epidemiological framework. In this approval approach, the trade-offs between frequency, detection limit and turnaround time can be anticipated and appropriately evaluated. 1-3
In order to defeat Covid-19, we believe that the FDA, CDC, National Institutes of Health and other agencies must encourage structured evaluation of tests in the context of planned test programs to find out which test program can provide the best Covid filter . Frequently using cheap, simple, and fast tests can achieve this goal, even if their analytical sensitivity is much lower than that of benchmark tests. 1 Such a scheme can also help us prevent the development of Covid.
Boston Harvard Chenchen School of Public Health (MJM); and University of Colorado Boulder (RP, DBL).
1. Larremore DB, Wilder B, Lester E, etc. For COVID-19 surveillance, test sensitivity is second only to frequency and turnaround time. September 8, 2020 (https://www.medrxiv.org/content/10.1101/2020.06.22.20136309v2). Preprint.
2. Paltiel AD, Zheng A, Walensky RP. Evaluate the SARS-CoV-2 screening strategy to allow the safe reopening of university campuses in the United States. JAMA Cyber Open 2020; 3(7): e2016818-e2016818.
3. Chin ET, Huynh BQ, Chapman LAC, Murrill M, Basu S, Lo NC. The frequency of routine testing for COVID-19 in high-risk environments to reduce workplace outbreaks. September 9, 2020 (https://www.medrxiv.org/content/10.1101/2020.04.30.20087015v4). Preprint.
4. He X, Lau EHY, Wu P, etc. Time dynamics of virus shedding and COVID-19 transmission capacity. Nat Med 2020; 26:672-675.
5. Centers for Disease Control and Prevention. Transcript of CDC’s updated telephone briefing on COVID-19. June 25, 2020 (https://www.cdc.gov/media/releases/2020/t0625-COVID-19-update.html).
A person’s infection trajectory (blue line) is shown in the context of two surveillance programs (circles) with different analytical sensitivity. Low analytical sensitivity assays are often performed, while high analytical sensitivity assays are rare. Both test schemes can detect the infection (orange circle), but despite its lower analytical sensitivity, only the high-frequency test can detect it within the propagation window (shadow), which makes it a more effective filter Device. The polymerase chain reaction (PCR) detection window (green) before infectivity is very short, and the corresponding window (purple) that can be detected by PCR after infection is very long.
Post time: Mar-11-2021