Diagnostic testing is almost the only thing that the world has to fight the novel coronavirus, COVID-19. Most nations with confirmed cases of the virus are braving these dark times without vaccine or proven remedy by turning their to expedient identification and quarantine for countermeasures. Naturally, global attention has focused on South Korea, which has established large-scale testing since the initial proliferation of the virus. Subsequently, requests for diagnostic kits flooded in from various countries, in spite of President Trump’s comments that the US has “done more tests in eight days than South Korea has done in eight weeks.” The market is showing signs of overheating. Search for “COVID testing kits” online and the first relevant keyword is “related stocks.”
Given the fierceness of attention towards diagnostic tests and kits, the heat sometimes tends to detract from what is truly important. Disagreements over test accuracies sometimes lead to political quarrels and overzealous media optimism give false hope to the public. To be sure, diagnostic tests are not the end-all solution that can end the COVID-19 situation. However, it is certainly a pivotal starting point for initiating medical action against the virus to stop its spread and treat patients. This report contains the A to Z on diagnostic tests, our only sword and shield at least until a COVID-19 vaccine and/or cure have been developed.
■ From swab to real-time PCR
If a diagnostic test is called for a suspected infection by the novel coronavirus, a myriad of worries crosses one’s mind. The most immediate worry is the invasion of the cotton swab. Sample collection for COVID-19 testing takes place at the upper (nose, mouth) and lower (lungs) respiratory tracts. While the upper tract sample is widely known as mucus, exactly speaking it consists of the epithelial cells of the respiratory organs. A 15cm-long swab is pushed into the inner wall of the nose and mouth. The shared experience of those who have been tested is that “the pain immediately makes one shed tears.” The upper tract sample from the nose and mouth is then encased in a collection container, swab and all. A separate container receives the lower tract sample, which is the phlegm produced by the testee. Samples are triple-wrapped and sent to a testing institution.
The sample does not head directly for the diagnostic equipment, because the nucleic acid has to be extracted from the sample first. Clinical technologists clad in PPE (personal protective equipment) extracts the nucleic acid in a room equipped with negative pressure equipment. Extraction takes approximately an hour to an hour and a half. The diagnostic methodology for COVID-19 is called ‘real-time polymerase chain reaction (real-time PCR). This method involves multiplying specific parts of the DNA to a multitude in the tens of thousands to examine whether the coronavirus is present (positive) or not (negative). The reagent, which is the chemical used in the diagnosis, is what is widely referred to as the diagnostic kit. The kit consists of 5 to 6 tubes of medical compounds. These compounds are introduced into the nucleic acid extracted from a sample.
An essential portion of the reagent is the primer, which only adheres to DNA specific to COVID-19. The multiplication only occurs when the primer has adhered to the COVID-19 DNA. The design of the primer and the chemical concentration of a reagent differ across manufacturers, and thus can be understood as industrial know-how. This is also where the accuracy of the test itself diverges. The targeted DNA also differs across kits. 5 companies produce kits in South Korea; Kogene Biotech’s kit targets DNA E and RdRP, Solgent’s kit targets ORF1a and DNA N, and so on. According to WHO guidelines, 2 or more DNAs need to be identified to judge a sample COVID-19 as positive.
For a time, controversy has been raised on the point that South Korea’s testing methodology deviates from the US CDC (Center for Disease Control and Prevention) and lacks accuracy. Conversely, however, a number of errors have been identified in February during diagnostic testing with kits developed by the US CDC. The CDC’s real-time PCR testing only targets DNA N. The South Korea CDC (KCDC) considered adopting the American methodology, but in the end decided to set targeting DNA E and RdRp as the base guideline based on the judgment that DNA N is more subject to mutation. Test kits produced by Roche, which received the first emergency approval by the FDA (Food and Drug Administration) also target DNA other than N.
Now, it is time to feed the sample into the testing equipment. While the figure varies by manufacturer, a typical real-time PCR equipment can test samples from about 50 individuals at once, and it takes around 2 hours to complete. The overall testing time, including the nucleic acid extraction and equipment running time, is around 3 hours. From the point of sample collection to reporting, the entire process takes about 6 hours.
■ False positive, false negative, undecided
The accuracy of diagnostic tests has come under scrutiny following a train of cases where cured patients released back into society tested positive for the virus again. A number of possibilities need to be examined in addition to that of an error in testing. First is the possibility of re-infection. A number of cases have been reported in China, the country with the most clinical data, where former patients have not formed antibodies against the novel coronavirus. The second possibility is reactivation of the virus. Professor Lee Hyuk-min at Yonsei Severance Hospital’s Department of Clinical Pathology explains that COVID-19 has strange characteristics. “While not the norm, there are cases where the patient transitions completely to negative, yet tests positive again after 1 to 2 weeks. Such cases have also been reported overseas.” In consideration of this peculiarity, the Chinese government has cured patients - those that test negative - self-quarantine for 2 weeks and be tested 2 and 4 weeks following their release.
Patients in the stage of recovery from their time with COVID-19 sometimes produce test results that are inconclusive, because the virus output can vary. “Depending on the amount of the virus, there is a zone around a certain level where the possibilities are 50-50 between positive and negative. This is a limitation owing to the fact that the test is man-made (Professor Lee).” While even positive cases have little infectivity at this stage, such cases are declared inconclusive following a reading by a clinical pathology specialist, and the patients are retested in a day or two.
Controversy flared around the results of a 17-year-old patient who had passed away from pneumonia on March 18th in Daegu. The patient had come to the hospital for high fever and was tested for COVID-19 12 times. The result was negative every time, but the 13th test conducted post-mortem produced an inconclusive result. To be declared COVID-19 positive, a case must test positive for two or more target DNAs. In the case of this particular patient, one target DNA tested weakly positive. This development was spun in some media, who produced headlines that the patient “tested positive post-mortem.”
Later, the KCDC investigated the Youngnam University Hospital inspection facility and confirmed that some degree of contamination had occurred during testing. As evidence, KCDC pointed to the fact that the negative control sample used during the 13th test at Youngnam University Hospital on March 18th also produced positive results. During real-time PCR tests, a control sample - which should always return negative results - is run with the patient sample to verify that there has not been any testing error or contamination. When Seoul National University Hospital and Yonsei University Severance Hospital tested the sample of the patient in question on March 19th, the tests produced negative results.
■ Rapid Test Kits - the hot potato
Can COVID-19 testing be done easily at the comfort of one’s home like a pregnancy test? If rapid test kits that produce results in 10 minutes have been developed, why are they not in use? Currently, the KCDC limits its emergency approval to kits that follow the real-time PCR methodology. Why? Accuracy. The real-time PCR testing used in South Korea is molecular testing, which identifies DNA in the sample. On the other hand, rapid test kits are largely based on testing antibodies or antigens. Antigen refers to the invading virus within the body, and an antibody is a substance produced by the immune system to defeat invaders. Antigen testing refers literally to testing whether the subject has antigens, and antibody testing tests whether the subject has antibodies. Testing for antigens involves feeding antibodies into the antigen testing kit, and conversely testing for antibodies involves feeding antigens into the kit.
The accuracy of both antigen and antibody testing lags behind that of molecular testing. The metrics utilized to estimate accuracy of such tests are sensitivity and specificity. The former indicates how well a testing method is able to identify positive cases. Identifying 5 out of 8 infected across 10 tested cases makes for a relatively inaccurate test. The latter measures how well a testing method can refrain from false positives. If a testing method identified 8 positives among 10 samples despite there only being 5 actual positives, the method lacks specificity. In other words, lacking sensitivity lets real patients go home and lacking specificity hospitalizes the wrong people.
Real-time PCR testing clocks in at over 95% for both sensitivity and specificity. However, the same metrics for antigen testing is around 50 to 70%. In the case of antibody testing, sensitivity can be taken beyond 95%, but its specificity is uncertain. Furthermore, an infection does not immediately result in antibodies. Rather, antibodies are produced only 7 to 28 days after the virus infiltrates the body, which renders the test useless during early stages of infection.
Professor Hong Ki-jong of Konkuk University, was in charge of diagnostic testing of infectious diseases and reviewed various kits and equipment from home and abroad during his time at the Korean National Institute of Health.
Professor Hong said unambiguously that rapid test kits cannot be used at this time. “South Korean companies also do a good job in making rapid testing kits. So it’s not that the quality is falling behind, but rapid types (like antibody and antigen testing) inherently fall behind the real-time PCR testing in accuracy. I am aware that some local governments have reviewed using these kits, but this can potentially bring quarantine management to its knees.”
Regardless of the possibilities of the kits actually being adopted, the market has ballooned in anticipation. Stock prices have jumped for diagnostic kit manufacturers that received emergency use approval, spreading the perception among investors that “this is a winning pot.” There was a deluge of media reports on diagnostics companies lacking domestic approval that have succeeded, or may, succeed in exporting their products. As of April 2nd, the number of kit manufacturers approved by the Ministry of Food and Drug Safety is 18. The list can be found in the Korea Health Industry Development Institute notice board.
Recently, it has become known that a number of rapid test kits developed by American companies have received emergency approval from the US FDA. One kit utilizes antibody·antigen testing, and two feature drastically decreased testing time based on molecular testing (as of April 2nd). Cepheid promotes 45 minutes, Abbote 13 minutes, Cellex’s antibody·antigen testing 20 minutes as turnaround time.
Professor Hong was dismissive about the need to utilize the American kits in South Korea. “If you consider the testing time itself, the real-time PCR equipment can also be sped up to conclude a round of testing in an hour and a half. So that makes for 6 hours to have the sample depart from the testing center to reporting the results. Even if Abbote’s kits are shorter on the turnaround, they fall relatively behind on accuracy. So why did the US approve those kits? Because they don’t have enough of them.“
translated by Hyo Sung Joo
translation supervised by Franz Maier, Sumi Paik-Maier
