Back

Role of Anti-SARS-CoV-2 antibodies in different cohorts: Can they provide clues for appropriate patient triaging?

Mutnal, M. B.; Mohammad, A. A.; Arroliga, A. C.; Hua, Y.; Wang, L.; Koss, W.; Rao, A.

2020-06-29 microbiology
10.1101/2020.06.26.174672 bioRxiv
Show abstract

The emergence of coronavirus disease 2019 (COVID-19) has become a major global health crisis. Currently, diagnosis is based on molecular techniques, which detect the viral nucleic acids when present at detectable levels. The serum IgG response against SARS-CoV-2 was examined by using an ELISA-based assay. Serum samples, along with nasopharyngeal specimens were collected from various cohorts and analyzed by ELISA and rRT-PCR, respectively. A total of 167 serum samples were tested for serum IgG antibodies against SARS-CoV-2 in outpatient cohorts, 15 (8.9%) were positive by rRT-PCR and the remaining 152 (91%) were negative. We used these data to generate two different assay cutoffs for serum IgG assay and investigated percent concordance with rRT-PCR test results. The emergency department data revealed, out of 151 nasopharyngeal swabs, 4 (2.6%) were positive by rRT-PCR and 18 (11.9%) were positive for serum IgG assay. Among the 18 patients that were positive for serum IgG, 13 (72.2%) exhibited 1-3 symptoms of COVID-19 and 5 (27.7%) patients did not present with any COVID-19 related symptoms, per CDC criteria. All 4 (100%) patients that were positive by rRT-PCR had symptoms of COVID-19 disease. A longitudinal study from the inpatient population suggested there was a sharp increase in the serum IgG titers in 5 patients, a moderate increase in 1 patient and a plateau in 3 patients. Sero-prevalence of COVID-19 disease in pre-procedure patients was 5.5%. Our findings suggest serological tests can be used for appropriate patient triaging when performed as an adjunct to existing molecular testing.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.