Back

Cross-sectional IgM and IgG profiles in SARS-CoV-2 infection

Ozturk, T.; Howell, C.; Benameur, K.; Ramonell, R. P.; Cashman, K.; Pirmohammed, S.; Bassit, L.; Roback, J.; Marconi, V. C.; Schinazi, R. F.; Wharton, W.; Lee, F. E.-H.; Hu, W. T.

2020-05-14 infectious diseases
10.1101/2020.05.10.20097535 medRxiv
Show abstract

BackgroundAccurate serological assays can improve the early diagnosis of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, but few studies have compared performance characteristics between assays in symptomatic and recovered patients. MethodsWe recruited 32 patients who had 2019 coronavirus disease (COVID-19; 18 hospitalized and actively symptomatic, 14 recovered mild cases), and measured levels of IgM (against the full-length S1 or the highly homologous SARS-CoV E protein) and IgG (against S1 receptor binding domain [RBD]). We performed the same analysis in 103 pre-2020 healthy adult control (HC) participants and 13 participants who had negative molecular testing for SARS-CoV-2. ResultsAnti-S1-RBD IgG levels were very elevated within days of symptom onset for hospitalized patients (median 2.04 optical density [OD], vs. 0.12 in HC). People who recovered from milder COVID-19 only reached similar IgG levels 28 days after symptom onset. IgM levels were elevated early in both groups (median 1.91 and 2.12 vs. 1.14 OD in HC for anti-S1 IgM, 2.23 and 2.26 vs 1.52 in HC for anti-E IgM), with downward trends in hospitalized cases having longer disease duration. The combination of the two IgM levels showed similar sensitivity for COVID-19 as IgG but greater specificity, and identified 4/10 people (vs. 3/10 by IgG) with prior symptoms and negative molecular testing to have had COVID-19. ConclusionsDisease severity and timing both influence levels of IgM and IgG against SARS-CoV-2, with IgG better for early detection of severe cases but IgM more suited for early detection of milder cases.

Matching journals

The top 6 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.