Longitudinally monitored immune biomarkers predict the timing of COVID-19 outcomes
The clinical outcome of SARS-CoV-2 infection varies widely between individuals. Machine learning models can support decision making in healthcare by assessing fatality risk in patients that do not yet show severe signs of COVID-19. Most predictive models rely on static demographic features and clinical values obtained upon hospitalization. However, time-dependent biomarkers associated with COVID-19 severity, such as antibody titers, can substantially contribute to the development of more accurate outcome models. Here we show that models trained on immune biomarkers, longitudinally monitored throughout hospitalization, predicted mortality and were more accurate than models based on demographic and clinical data upon hospital admission. Our best-performing predictive models were based on the temporal analysis of anti-SARS-CoV-2 Spike IgG titers, white blood cell (WBC), neutrophil and lymphocyte counts. These biomarkers, together with C-reactive protein and blood urea nitrogen levels, were found to correlate with severity of disease and mortality in a time-dependent manner. Shapley additive explanations of our model revealed the higher predictive value of day post-symptom onset (PSO) as hospitalization progresses and showed how immune biomarkers contribute to predict mortality. In sum, we demonstrate that the kinetics of immune biomarkers can inform clinical models to serve as a powerful monitoring tool for predicting fatality risk in hospitalized COVID-19 patients, underscoring the importance of contextualizing clinical parameters according to their time post-symptom onset.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2022 |
---|---|
Erschienen: |
2022 |
Enthalten in: |
Zur Gesamtaufnahme - volume:18 |
---|---|
Enthalten in: |
PLoS computational biology - 18(2022), 1 vom: 18. Jan., Seite e1009778 |
Sprache: |
Englisch |
---|
Links: |
---|
Anmerkungen: |
Date Completed 09.02.2022 Date Revised 03.11.2023 published: Electronic-eCollection Citation Status MEDLINE |
---|
doi: |
10.1371/journal.pcbi.1009778 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM33576374X |
---|
LEADER | 01000naa a22002652 4500 | ||
---|---|---|---|
001 | NLM33576374X | ||
003 | DE-627 | ||
005 | 20231225230613.0 | ||
007 | cr uuu---uuuuu | ||
008 | 231225s2022 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1371/journal.pcbi.1009778 |2 doi | |
028 | 5 | 2 | |a pubmed24n1119.xml |
035 | |a (DE-627)NLM33576374X | ||
035 | |a (NLM)35041647 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Lasso, Gorka |e verfasserin |4 aut | |
245 | 1 | 0 | |a Longitudinally monitored immune biomarkers predict the timing of COVID-19 outcomes |
264 | 1 | |c 2022 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Completed 09.02.2022 | ||
500 | |a Date Revised 03.11.2023 | ||
500 | |a published: Electronic-eCollection | ||
500 | |a Citation Status MEDLINE | ||
520 | |a The clinical outcome of SARS-CoV-2 infection varies widely between individuals. Machine learning models can support decision making in healthcare by assessing fatality risk in patients that do not yet show severe signs of COVID-19. Most predictive models rely on static demographic features and clinical values obtained upon hospitalization. However, time-dependent biomarkers associated with COVID-19 severity, such as antibody titers, can substantially contribute to the development of more accurate outcome models. Here we show that models trained on immune biomarkers, longitudinally monitored throughout hospitalization, predicted mortality and were more accurate than models based on demographic and clinical data upon hospital admission. Our best-performing predictive models were based on the temporal analysis of anti-SARS-CoV-2 Spike IgG titers, white blood cell (WBC), neutrophil and lymphocyte counts. These biomarkers, together with C-reactive protein and blood urea nitrogen levels, were found to correlate with severity of disease and mortality in a time-dependent manner. Shapley additive explanations of our model revealed the higher predictive value of day post-symptom onset (PSO) as hospitalization progresses and showed how immune biomarkers contribute to predict mortality. In sum, we demonstrate that the kinetics of immune biomarkers can inform clinical models to serve as a powerful monitoring tool for predicting fatality risk in hospitalized COVID-19 patients, underscoring the importance of contextualizing clinical parameters according to their time post-symptom onset | ||
650 | 4 | |a Journal Article | |
650 | 4 | |a Research Support, N.I.H., Extramural | |
650 | 4 | |a Research Support, Non-U.S. Gov't | |
650 | 4 | |a Research Support, U.S. Gov't, Non-P.H.S. | |
650 | 7 | |a Antibodies, Viral |2 NLM | |
650 | 7 | |a Biomarkers |2 NLM | |
650 | 7 | |a Spike Glycoprotein, Coronavirus |2 NLM | |
650 | 7 | |a spike protein, SARS-CoV-2 |2 NLM | |
700 | 1 | |a Khan, Saad |e verfasserin |4 aut | |
700 | 1 | |a Allen, Stephanie A |e verfasserin |4 aut | |
700 | 1 | |a Mariano, Margarette |e verfasserin |4 aut | |
700 | 1 | |a Florez, Catalina |e verfasserin |4 aut | |
700 | 1 | |a Orner, Erika P |e verfasserin |4 aut | |
700 | 1 | |a Quiroz, Jose A |e verfasserin |4 aut | |
700 | 1 | |a Quevedo, Gregory |e verfasserin |4 aut | |
700 | 1 | |a Massimi, Aldo |e verfasserin |4 aut | |
700 | 1 | |a Hegde, Aditi |e verfasserin |4 aut | |
700 | 1 | |a Wirchnianski, Ariel S |e verfasserin |4 aut | |
700 | 1 | |a Bortz, Robert H |c 3rd |e verfasserin |4 aut | |
700 | 1 | |a Malonis, Ryan J |e verfasserin |4 aut | |
700 | 1 | |a Georgiev, George I |e verfasserin |4 aut | |
700 | 1 | |a Tong, Karen |e verfasserin |4 aut | |
700 | 1 | |a Herrera, Natalia G |e verfasserin |4 aut | |
700 | 1 | |a Morano, Nicholas C |e verfasserin |4 aut | |
700 | 1 | |a Garforth, Scott J |e verfasserin |4 aut | |
700 | 1 | |a Malaviya, Avinash |e verfasserin |4 aut | |
700 | 1 | |a Khokhar, Ahmed |e verfasserin |4 aut | |
700 | 1 | |a Laudermilch, Ethan |e verfasserin |4 aut | |
700 | 1 | |a Dieterle, M Eugenia |e verfasserin |4 aut | |
700 | 1 | |a Fels, J Maximilian |e verfasserin |4 aut | |
700 | 1 | |a Haslwanter, Denise |e verfasserin |4 aut | |
700 | 1 | |a Jangra, Rohit K |e verfasserin |4 aut | |
700 | 1 | |a Barnhill, Jason |e verfasserin |4 aut | |
700 | 1 | |a Almo, Steven C |e verfasserin |4 aut | |
700 | 1 | |a Chandran, Kartik |e verfasserin |4 aut | |
700 | 1 | |a Lai, Jonathan R |e verfasserin |4 aut | |
700 | 1 | |a Kelly, Libusha |e verfasserin |4 aut | |
700 | 1 | |a Daily, Johanna P |e verfasserin |4 aut | |
700 | 1 | |a Vergnolle, Olivia |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t PLoS computational biology |d 2005 |g 18(2022), 1 vom: 18. Jan., Seite e1009778 |w (DE-627)NLM15722645X |x 1553-7358 |7 nnns |
773 | 1 | 8 | |g volume:18 |g year:2022 |g number:1 |g day:18 |g month:01 |g pages:e1009778 |
856 | 4 | 0 | |u http://dx.doi.org/10.1371/journal.pcbi.1009778 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 18 |j 2022 |e 1 |b 18 |c 01 |h e1009778 |