The soldier had no COVID-19-associated symptoms before, during or after his stay at the training centre. Nevertheless, he had a positive rapid antibody test on enrolment, positive serological assays with slightly increasing anti-SARS-CoV-2 antibody levels, and a weakly positive PCR test for SARS-CoV-2 RNA at the last testing session in week 6.
The rising anti-SARS-CoV-2 antibody levels in our patient are difficult to interpret. IgG/IgM antibodies are reported to develop between a few days and possibly up to a week after symptom onset in cases of COVID-19. The time course is somewhat different for IgG and IgM, but the concentration reaches a maximum several weeks after the patient first had symptoms, partly as a result of antibody avidity (2, 3). The serodynamics in asymptomatic patients have yet to be fully characterised.
The patient's positive serology results for SARS-CoV-2 may indicate a history of infection prior to arrival at the training centre, which was not detected in the anamnesis or via the first PCR tests. The test used (Roche Elecsys Anti-SARS-CoV-2) has a sensitivity of more than 95 % and a specificity of more than 99 % (4).
Non-specific reactivity or cross-reactivity with antibodies against other coronaviruses may have led to false positive serological results. However, the S/Co (signal-to-cut-off ratio) would most likely have been lower if that were the case, and would not have shown successive increases over the brief period in which antibody levels were measured.
The weakly positive PCR result for SARS-CoV-2 RNA in the last routine test in week 6 could not be reproduced. This might mean that the result was a false positive, or it could reflect the presence of non-viable virus or of very small amounts of viral RNA (5).
The results of the PCR tests for the first two samples and the control samples might have been false negatives. Other have also reported cases of a positive PCR test after two previous negative tests 24 hours apart, and have proposed that this may be due to false negative results, reinfection or reactivation (6). Another possible explanation is that the virus was strongly bound to receptors in the lower respiratory tract and was not expelled as the patient had no cough. Antibody coating of the virus can also make it difficult to detect antigens, but will not affect the detection of RNA. Furthermore, shedding of the SARS-CoV-2 virus has yet to be fully characterised, especially late in the disease course (7). A mix-up of samples is a theoretical possibility, but is highly unlikely given the stringent procedures followed.
Overall, we believe that the results of the serological assays and PCR tests together suggest a previous asymptomatic infection prior to arrival at the training centre.
The fact that our patient did not infect others may suggest low transmissibility. However, it is not a given that patients with asymptomatic infection will be less likely to transmit the virus than symptomatic patients. The high Ct (cycle threshold) value of over 35 in the positive PCR test indicates that there must have been little viral RNA present in the sample. Another study has shown that SARS-CoV-2 isolated from mucosal specimens with high PCR-Ct values cannot be cultured and are therefore no longer viable (8). The absence of infection transmission may also reflect the effects of good infection control measures or may indicate that the patient was not infectious.
This case report demonstrates that a repeat PCR test can detect viral RNA even if previous tests have been negative and there has been no intercurrent disease. It also shows concordance between positive results for antibody-based rapid tests and tests on venous blood samples, and illustrates how serological assays can be helpful when interpreting PCR results. Finally, this case highlights the various challenges that can arise when screening asymptomatic individuals on a large scale.