The SARS-CoV-2 epidemic has undoubtedly caused extensive damage to societies and the economy. However, the epidemic seems to have allowed molecular biology to develop quite rapidly. Many companies intensively developed a variety of tests, which quickly obtained IVD (in vitro diagnostic) status.
The symptoms of COVID-19 are quite nonspecific, similar to influenza or other infectious diseases; hence, there is a need to perform diagnostic tests to confirm SARS-CoV-2 infection.
2. Method Types
WHO presents NAATs (nucleic acid amplification tests) as a reference
[1]. The classic tests rely on detecting the presence of viral RNA by reverse transcription polymerase chain reaction (RT-PCR). Multiple strategies exist for amplifying target genes, including reverse transcriptase loop-mediated isothermal amplification (RT-LAMP), recombinase polymerase amplification (RPA), helicase-dependent amplification (HDA), strand displacement amplification (SDA), and nucleic acid sequence-based amplification (NASBA)
[2]. Recently, assays based on the isothermal amplification of viral nucleic acids have been developed, in combination with clustered regularly interspaced short palindromic repeat (CRISPR)-based detection methods
[1]. Below is a brief description of the most popular of the abovementioned methods.
2.1. RT-qPCR
Real-time/quantitative reverse transcription polymerase chain reaction (RT-qPCR) is a relatively simple method that allows detecting the presence of a specific fragment of genetic material. After isolating the RNA from the sample, a reverse transcriptase reaction is performed, and then the sequences of the appropriate genes are amplified using their specific primers. Detection of amplicons is possible thanks to the use of fluorescent dyes, with the growth of products being monitored in real time. Real-time RT-PCR remains the most popular molecular method in the diagnosis of COVID-19 in Poland and Europe
[3][4]. Its advantages are its simplicity, high accessibility, high sensitivity, and high specificity, as well as the fact that the tests based on this method can be quantitative.
2.2. Isothermal Amplification Methods
These methods, which do not require thermal cycling, are more rapid than RT-PCR, characterized by comparable sensitivity and specificity, and they are also considered suitable as point-of-care tests for the detection of SARS-CoV-2
[1].
2.2.1. RT-LAMP (Reverse Transcription Loop-Mediated Isothermal Amplification)
RT-LAMP uses a combination of 4–6 primers against different loci of target DNA sequences with the constant reaction temperature varying between 60 and 65 °C. Reverse transcription is performed, followed by loop-mediated isothermal amplification. The advantage is the shorter time, which can only take 30 min, as well as the lack of the need for specialized equipment such as thermal cycler
[5][6].
2.2.2. CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats)
The principle of this method is the activity of endoribonuclease Cas with an sgRNA (single guide RNA) complex followed by probing of the amplified specific fragments of the virus. The test can be designed according to the measurement of fluorescence or lateral flow
[6][7]. COVID-19 diagnosis uses the Cas-13 enzyme, which has the ability to react with RNA; after reverse transcription, the Cas12 or Cas9 enzyme possibly interacts with cDNA
[6]. The advantage of Cas12 and Cas13 over Cas9 is that they have collateral activity, making these enzymes more likely to be used in diagnostics
[8].
2.3. Genomic Sequencing
Genomic sequencing of SARS-CoV-2, mostly based on next-generation sequencing (NGS), is used to identify pathogens through sequencing of nucleic acid fragments, followed by analyzing and comparing biological information and databases with high accuracy
[2]. The method is generally used for genetic mapping rather than diagnostic testing due to its high cost and long detection time
[2].
2.4. An Alternative to NAATs
An alternative to NAATs is to use tests detecting antigens or antibodies.
The sensitivity of an antigen test is strongly dependent on the viral load. If it is high (Ct < 25 on a real-time RT-PCR test), the sensitivity can reach up to 100%. If, on the other hand, the viral load is low (Ct > 35 on a real-time RT-PCR test), the sensitivity of the antigen test can be as low as 22.2%, i.e., more than two-thirds of results may be false negatives
[9].
Immunoassays detecting antibodies against SARS-CoV-2 have the major disadvantage that they cannot be used to detect an infection, but only confirm that there has been a past immune system response to the pathogen. The sensitivity of the immunoassay strongly depends on the time elapsed since the infection, on the patient’s individual characteristics, and on the biochemical features of individual diagnostic kits. One study comparing five tests for IgG antibodies showed that the sensitivity of serum samples collected on average 24 days after the onset of symptoms was between 74.5% and 88.2%
[10]. Earlier, IgM antibodies appear. It should be remembered that both disease and vaccination cause different immune system reactions in each patient; therefore, a negative result of such a test does not necessarily mean that the patient has not been infected.
The great advantage of antigen and serological tests is that the test can be relatively easily designed to be a so-called rapid diagnostic test (RDT) or a point-of-care test (POCT), to be performed quickly, without specialist equipment or training
[11].
2.5. Diagnostic Windows
It should also be borne in mind that the diagnostic window for molecular, antigen-based, and antibody-based methods is quite narrow. The narrowest is for antigen tests, whereas it is relatively wide for molecular tests, and the broadest window exists for tests detecting IgG
[11]. In general, RT-PCR tests show the highest sensitivity between days 5 and 14 after contact with the SARS-CoV-2 coronavirus, and this is usually the time when a patient experiences the first symptoms of infection. Thus, it is sometimes possible to achieve a false-negative molecular test result for SARS-CoV-2 due to the test being performed before the virus reaches a detectable number of copies in the epithelial cells. Likewise, if the test is carried out too late, the opposite may happen, as false-positive results may be obtained because the RNA of the virus can remain in the epithelium for a long time. Individual differences are important here, because the immune system will react differently in each patient; hence, the dynamics of virus replication will be different
[11]. In some patients infected on the same day, the molecular test may be positive after 2 days, whereas it may take longer in others.
3. What Genes Are Analyzed in PCR for the Detection of SARS-CoV-2?
The SARS-CoV-2 genome is nearly 30,000 nucleotides long, making it one of the largest RNA viruses. The longest fragment here is organized into two overlapping ORFs (open reading frames), ORF1a (11–13 kb) and ORF1b (7–8 kb), which codes for RNA-dependent RNA polymerase (RdRp). Its sequence is highly specific; therefore, it is often chosen for detection in PCR tests for COVID-19 diagnostics
[1]. The first RT-PCR protocols for the detection of SARS-CoV-2 targeted genes related to RdRp, nucleocapsid (N), and envelope protein (E)
[12]. The tests based on the detection of RdRp used two probes; one probe called “Pan Sarbeco-probe” detected bat-related SARS coronaviruses and other coronaviruses, while the other probe called RdRp-p2 was specific only for SARS-CoV-2
[12]. Currently, tests are being developed to detect very diverse sequences of the entire genome of the SARS-CoV-2 virus, in addition to the abovementioned sequences for RdRp, the spike protein (S), and the membrane (M) protein
[13].
To minimize the risk of false positives related to the presence of non-SARS-CoV-2 virus, WHO recommends the detection of at least two different targets on the COVID-19 virus genome. Despite the large proportion (38% of commercially available tests) being based on analyzing only one gene, most of them are reflex tests, detecting specific variants in positive samples. Only a few of them were developed in the beginning of the pandemic and are still available, but they are generally not used in Europe.
4. Do Diagnostic Kits Detect Current SARS-CoV-2 Variants?
SARS-CoV-2, like other RNA viruses, is prone to genetic evolution. Even a single-nucleotide substitution can cause a change in virulence properties relative to the parent virus variance
[14]. The changes may concern the interaction of the virus with host proteins or the ability to avoid mechanisms of the host’s immunity, which undermines the effectiveness of vaccination. In addition, changes can also cause problems in the diagnosis of infections, because molecular tests may not detect altered sequences
[15].
During the pandemic, many variants have been distinguished, i.e., variants of concern (VOCs): (1) Alpha (B.1.1.7; September 2020); (2) Beta (B.351, B.1.351.2, and B.1.351.3 in December 2020); (3) Delta (B.1.617.2, AY.1, AY.2, AY.3, and AY.3.1 in December 2020); (4) Gamma (P.1, P.1.1, and P.1.2 in January 2021); Omicron (B.1.1.529 in November 2021)
[15]. In addition, several variants of interest (VOIs) have arisen, e.g., Zeta (P.2), Kappa (B.1.617.1), and Mu (B.1.621 and B.1.621.1). In those cases, specific genetic markers were found that are predicted to affect transmission, diagnostics, therapeutics, or interaction with the immune system
[15].
The Omicron variant is an example of a variant with multiple mutations, from which the potential diagnostic problems that may arise in the future can be extrapolated. The mentioned variant has over 30 mutations in a key gene for the S protein, resulting in a 13-fold increase in viral infectivity and substantial escape from neutralizing antibodies induced by vaccination
[15][16]. It can also be suspected that some tests based on the detection of the S protein gene will show false-negative results for the Omicron variant. Analyses in the direction of changes in the test effectiveness depending on the virus mutation can be carried out experimentally, as well as in silico.
Metzger et al. analyzed 39 test kits commercially available in Switzerland and Liechtenstein, and they exchanged kits for S protein sequences so as to address the potential of not detecting Omicron. Only two of the eight assays targeting the S gene appeared to show S-gene dropout with the Omicron variant. However, it is emphasized that these data are preliminary, based on in silico experience, and that each set should be thoroughly evaluated
[17].