Thank you for visiting nature.com. The browser version you are using has limited support for CSS. For the best experience, we recommend that you use a newer browser (or turn off compatibility mode in Internet Explorer). At the same time, in order to ensure continued support, we display the website without styles and JavaScript.
Health workers conducted a large-scale screening using rapid antigen testing at a school in France. Image credit: Thomas Samson/AFP/Getty
As the number of coronavirus cases in the UK surged in early 2021, the government announced a potential game change in the fight against COVID-19: millions of cheap, fast virus tests. On January 10, it stated that it would promote these tests nationwide, even for people with no symptoms. Similar tests will play a key role in President Joe Biden’s plan to contain the epidemic that is raging in the United States.
These quick tests usually mix a nasal or throat swab with the liquid on a paper strip to return results within half an hour. These tests are considered infectious tests, not infectious tests. They can only detect high viral loads, so they will miss many people with low SARS-CoV-2 virus levels. But the hope is that they will help contain the epidemic by quickly identifying the most infectious people, otherwise they may spread the virus unknowingly.
However, as the government announced the plan, angry controversy broke out. Some scientists are pleased with the British testing strategy. Others say that these tests will miss too many infections that if they are spread to millions, the harm they might cause outweighs the harm. Jon Deeks, who specializes in testing and evaluation at the University of Birmingham in the United Kingdom, believes that many people may be relieved of negative test results and change their behavior. And, he said, if people manage the tests themselves, instead of relying on trained professionals, these tests will miss more infections. He and his Birmingham colleague Jac Dinnes (Jac Dinnes) are scientists, and they hope they need more data on rapid coronavirus tests before they can be widely used.
But other researchers soon fought back, claiming that the test could cause harm is wrong and “irresponsible” (see go.nature.com/3bcyzfm). Among them is Michael Mina, an epidemiologist at the Harvard T.H. Chan School of Public Health in Boston, Massachusetts, who said this argument delays a much-needed solution to the pandemic. He said: “We still say that we don’t have enough data, but we are in the middle of a war-in terms of the number of cases, we really won’t be worse than at any time.”
The only thing scientists agree with is that there needs to be clear communication about what a quick test is and what the negative results mean. Mina said, “Throwing tools at people who don’t know how to use them properly is a bad idea.”
It is difficult to obtain reliable information for quick tests, because-at least in Europe-products can only be sold based on manufacturer data without independent evaluation. There is no standard protocol for measuring performance, so it is difficult to compare assays and force each country to conduct its own verification.
“This is the wild west in diagnosis,” said Catharina Boehme, CEO of the Innovative New Diagnostics Foundation (FIND), a non-profit organization in Geneva, Switzerland that has re-evaluated and compared dozens of COVID -19 Analysis method.
In February 2020, FIND embarked on an ambitious task to evaluate hundreds of COVID-19 test types in standardized trials. The foundation works with the World Health Organization (WHO) and global research institutions to test hundreds of coronavirus samples and compare their performance with those obtained using highly sensitive polymerase chain reaction (PCR) technology. The technology looks for specific viral genetic sequences in samples taken from a person’s nose or throat (sometimes saliva). PCR-based tests can replicate more of this genetic material through multiple cycles of amplification, so they can detect the initial amount of parvovirus. But they can be time-consuming and require well-trained personnel and expensive laboratory equipment (see “How COVID-19 Testing Works”).
Cheap, fast tests can often work by detecting specific proteins (collectively called antigens) on the surface of SARS-CoV-2 particles. These “rapid antigen tests” do not amplify the contents of the sample, so the virus can only be detected when the virus reaches high levels in the human body-there may be thousands of copies of the virus per milliliter of sample. When people are most infectious, the virus usually reaches these levels at the time of the onset of symptoms (see “Catch COVID-19″).
Dinnes said that the manufacturer’s data on test sensitivity mainly comes from laboratory tests in people with symptoms with high viral loads. In those trials, many quick tests seemed very sensitive. (They are also very specific: they are unlikely to give false-positive results.) However, real-world evaluation results indicate that people with low viral loads exhibit significantly different performance.
The level of virus in the sample is usually quantified with reference to the number of PCR amplification cycles required for virus detection. Generally, if approximately 25 PCR amplification cycles or less are required (called the cycle threshold, or Ct, equal to or less than 25), then the level of live virus is considered high, indicating that people may be infectious-although not yet It is clear whether people have or do not have a critical level of contagion.
In November last year, the British government released the results of preliminary studies conducted at Porton Down Science Park and Oxford University. All the results that have not yet been peer-reviewed were published online on January 15. These results indicate that although many rapid antigen (or “lateral flow”) tests “do not reach the level required for large-scale population deployment,” in laboratory trials, 4 individual brands had Ct values or lower 25. FIND’s reassessment of many rapid test kits usually also shows that the sensitivity at these virus levels is 90% or higher.
As the virus level drops (ie, the Ct value rises), rapid tests begin to miss infection. Scientists at Porton Down paid special attention to Innova Medical’s tests in Pasadena, California; the British government has spent more than 800 million pounds ($1.1 billion) to order these tests, an important part of its strategy to slow the spread of the coronavirus. At a Ct level of 25-28, the sensitivity of the test is reduced to 88%, and for a Ct level of 28-31, the test is reduced to 76% (see “Rapid Test Finds High Viral Load”).
In contrast, in December, Abbott Park, Illinois, Abbott Laboratories evaluated the BinaxNOW rapid test with unfavorable results. The study tested more than 3,300 people in San Francisco, California, and obtained 100% sensitivity for samples with Ct levels below 30 (even if the infected person did not show symptoms)2.
However, different calibrated PCR systems mean that Ct levels cannot be easily compared between laboratories, and it does not always indicate that the virus levels in the samples are the same. Innova said that the UK and US studies used different PCR systems, and that only a direct comparison on the same system would be effective. They pointed to a British government report written by Porton Down scientists in late December that pitted the Innova test against the Abbott Panbio test (similar to the BinaxNOW kit sold by Abbott in the United States). In just over 20 samples with a Ct level below 27, both samples returned 93% positive results (see go.nature.com/3at82vm).
When considering the Innova test trial on thousands of people in Liverpool, England, the nuances regarding Ct calibration were crucial, which only identified two-thirds of cases with Ct levels below 25 (see go.nature.com) /3tajhkw). This suggests that these tests missed a third of potentially infectious cases. However, it is now believed that in a laboratory that processes samples, a Ct value of 25 is equal to the much lower virus level in other laboratories (perhaps equal to a Ct of 30 or higher), said Iain Buchan, a researcher in Health and Informatics at American University . Liverpool, presided over the trial.
However, the details are not well known. Dix said that a trial conducted by the University of Birmingham in December was an example of how a rapid test missed an infection. More than 7,000 asymptomatic students there took the Innova test; only 2 tested positive. However, when university researchers used PCR to recheck 10% of the negative samples, they found six more infected students. Based on the ratio of all samples, the test may have missed 60 infected students3.
Mina said these students have low levels of the virus, so they are not contagious in any way. Dix believes that although people with lower levels of the virus may be in the late stages of a decline in infection, they may also be becoming more contagious. Another factor is that some students do not do well in collecting swab samples, so not many virus particles can pass the test. He worries that people will mistakenly believe that passing a negative test can ensure their safety-in fact, a quick test is just a snapshot that may not be infectious at that moment. Deeks said the claim that testing can make the workplace completely safe is not the right way to inform the public about its efficacy. He said: “If people have a wrong understanding of security, they may actually spread this virus.”
But Mina and others said that Liverpool pilots advised people not to do that and were told they might still spread the virus in the future. Mina emphasized that frequent use of testing (such as twice a week) is the key to making testing effective to contain the pandemic.
The interpretation of test results depends not only on the accuracy of the test, but also on the chance that a person already has COVID-19. It depends on the infection rate in their area and whether they show symptoms. If a person from an area with a high COVID-19 level has typical symptoms of the disease and gets a negative result, it may be a false negative and needs to be carefully checked using PCR.
Researchers also debate whether people should test themselves (at home, school or work). The performance of the test may vary, depending on how the tester collects the swab and processes the sample. For example, using the Innova test, laboratory scientists have reached a sensitivity of nearly 79% for all samples (including samples with very low viral loads), but the self-taught public only gets a sensitivity of 58% (see “Quick Test: Is it suitable for home?”) -Deeks believes this is a worrying drop1.
Nevertheless, in December, the British drug regulatory agency authorized the use of Innova testing technology in the home to detect infections in asymptomatic people. A DHSC spokesperson confirmed that the trademarks for these tests came from the country’s National Health Service, designed by the Ministry of Health and Social Care (DHSC), but purchased from Innova and produced by China’s Xiamen Biotechnology Co., Ltd. “The horizontal flow test used by the British government has been rigorously evaluated by leading British scientists. This means that they are accurate, reliable, and able to successfully identify asymptomatic COVID-19 patients.” The spokesperson said in a statement.
A German study4 pointed out that self-administered tests can be as effective as those done by professionals. This study has not been peer-reviewed. The study found that when people wipe their noses and complete an anonymous quick test approved by the WHO, even if people often deviate from the instructions for use, the sensitivity is still very similar to that achieved by professionals .
In the United States, the Food and Drug Administration (FDA) has approved emergency use permits for 13 antigen tests, but only one-the Ellume COVID-19 home test-can be used for asymptomatic people. According to Ellume, a company based in Brisbane, Australia, the test has detected the coronavirus in 11 asymptomatic people, and 10 of these people have tested positive by PCR. In February, the US government announced that it would purchase 8.5 million tests.
Some countries/regions that do not have sufficient resources for PCR testing, such as India, have been using antigen testing for many months, just to supplement their testing capabilities. Out of the concern for accuracy, some companies that perform PCR testing have only begun to introduce quick alternatives to a limited extent. But the government that implemented large-scale rapid testing called it a success. With a population of 5.5 million, Slovakia was the first country to try to test its entire adult population. Extensive testing has reduced the infection rate by nearly 60%5. However, the test is done in conjunction with strict restrictions not implemented in other countries and the government’s financial support for people who test positive to help them stay at home. Therefore, experts say that although the combination of testing and restriction appears to reduce infection rates faster than restriction alone, it is not clear whether the method can work elsewhere. In other countries, many people may not want to take the rapid test, and those who test positive may lack motivation to isolate. Nonetheless, because commercial rapid tests are very cheap-only $5-Mina says cities and states can buy millions at a fraction of the government losses caused by the epidemic.
A health worker quickly tested a passenger with a nasal swab at a train station in Mumbai, India. Image credit: Punit Parajpe / AFP / Getty
Rapid tests may be particularly suitable for asymptomatic screening situations including prisons, homeless shelters, schools and universities, where people may gather anyway, so any test that can catch some additional cases of infection is useful. But Deeks cautions against using the test in a way that might change people’s behavior or prompt them to relax precautions. For example, people may interpret negative results as encouraging visits to relatives in nursing homes.
So far, in the United States, large-scale rapid testing procedures have been launched in schools, prisons, airports and universities. For example, since May, the University of Arizona in Tucson has been using the Sofia test developed by Quidel in San Diego, California to test its athletes on a daily basis. Since August, it has tested students at least once a month (some students, especially those in dormitories with outbreaks, are tested more frequently, once a week). So far, the university has performed nearly 150,000 tests and has not reported a surge in COVID-19 cases in the past two months.
David Harris, a stem cell researcher in charge of Arizona’s large-scale testing program, said that different types of tests serve different purposes: rapid antigen tests should not be used to assess the prevalence of the virus in the population. He said: “If you use it like PCR, you will get terrible sensitivity.” “But what we are trying to do-prevent the spread of infection-antigen testing, especially when used multiple times, seems to work well. ”
A student from Oxford University in the UK took a rapid antigen test provided by the university and then flew to the United States in December 2020.
Many research groups around the world are designing faster and cheaper test methods. Some are adjusting PCR tests to speed up the amplification process, but many of these tests still require specialized equipment. Other methods rely on a technique called loop-mediated isothermal amplification or LAMP, which is faster than PCR and requires minimal equipment. But these tests are not as sensitive as PCR-based tests. Last year, researchers at the University of Illinois at Urbana-Champaign developed their own rapid diagnostic test: a PCR-based test that uses saliva instead of a nasal swab, skipping expensive and slow steps. The cost of this test is $10-14, and results can be given in less than 24 hours. Although the university relies on on-site laboratories to perform PCR, the university can screen everyone twice a week. In August of last year, this frequent testing program allowed the university to detect a surge in campus infections and control it to a large extent. Within a week, the number of new cases fell by 65%, and since then, the university has not seen a similar peak.
Boehme said that there is no one test method that can meet all needs, but a test method that can identify infectious people is essential to keeping the world economy open. She said: “Tests in airports, borders, workplaces, schools, clinical settings-in all these cases, rapid tests are powerful because they are easy to use, low cost, and fast.” However, she added That said, large test programs should rely on the best tests available.
The EU’s current approval process for COVID-19 diagnostic tests is the same as other types of diagnostic procedures, but concerns about the performance of certain testing methods prompted the introduction of new guidelines last April. These require manufacturers to produce test kits that can at least perform COVID-19 testing at the latest state of the art. However, since the effect of testing performed in the manufacturer’s test may be different from that in the real world, the guidelines recommend that member states verify it before launching the test.
Boehme said that, ideally, countries would not have to verify every measurement method. Laboratories and manufacturers around the world will use common protocols (such as those developed by FIND). She said: “What we need is a standardized test and evaluation method.” “It will not be any different from evaluating treatments and vaccines.”
Post time: Mar-09-2021