Circulating proteins to predict COVID-19 severity

Chen‐Yang Su, Sirui Zhou, Edgar Gonzalez‐Kozlova, Guillaume Butler‐Laporte, Elsa Brunet‐Ratnasingham, Tomoko Nakanishi, Wonseok Jeon, David Morrison, Lætitia Laurent, Jonathan Afilalo, Marc Afilalo, Danielle Henry, Yiheng Chen, Julia Carrasco-Zanini, Yossi Farjoun, Maik Pietzner, Nofar Kimchi, Zaman Afrasiabi, Nardin Rezk, Meriem Bouab, Louis Petitjean, Charlotte Guzman, Xiaoqing Xue, Chris Tselios, Branka Vulesevic, Olumide Adeleye, Tala Abdullah, Noor Almamlouk, Yara Moussa, Chantal DeLuca, Naomi Duggan, Erwin Schurr, Nathalie Brassard, Madéleine Durand, Diane M. Del Valle, Ryan C. Thompson, Mario A. Cedillo, Eric E. Schadt, Kai Nie, Nicole W. Simons, Konstantinos Mouskas, Nicolas Zaki, Manishkumar Patel, Hui Xie, Jocelyn Harris, Robert Marvin, Esther Cheng, Kevin Tuballes, Kimberly Argueta, Ieisha Scott, Celia M.t. Greenwood, Clare Paterson, Michael Hinterberg, Claudia Langenberg, Vincenzo Forgetta, Joëlle Pineau, Vincent Mooser, Thomas U. Marron, Noam D. Beckmann, Seunghee Kim‐Schulze, Alexander W. Charney, Sacha Gnjatic, Daniel E. Kaufmann, Miriam Mérad, J. Brent RichardsScientific 2023 · DOI

Abstract

Predicting COVID-19 severity is difficult, and the biological pathways involved are not fully understood. To approach this problem, we measured 4701 circulating human protein abundances in two independent cohorts totaling 986 individuals. We then trained prediction models including protein abundances and clinical risk factors to predict COVID-19 severity in 417 subjects and tested these models in a separate cohort of 569 individuals. For severe COVID-19, a baseline model including age and sex provided an area under the receiver operator curve (AUC) of 65% in the test cohort. Selecting 92 proteins from the 4701 unique protein abundances improved the AUC to 88% in the training cohort, which remained relatively stable in the testing cohort at 86%, suggesting good generalizability. Proteins selected from different COVID-19 severity were enriched for cytokine and cytokine receptors, but more than half of the enriched pathways were not immune-related. Taken together, these findings suggest that circulating proteins measured at early stages of disease progression are reasonably accurate predictors of COVID-19 severity. Further research is needed to understand how to incorporate protein measurement into clinical care.

Introduction

The clinical course of COVID-19 is notoriously unpredictable. While some patients remain asymptomatic, others progress rapidly to respiratory failure or death. Current triaging relies heavily on age, sex, and basic comorbidities, but these are often blunt instruments for predicting individual outcomes. The biological mechanisms driving this variability remain partially obscured. While the "cytokine storm" is a known factor, the broader systemic changes are less understood. This study aimed to identify a proteomic "fingerprint" at hospital presentation that could accurately predict a patient's worst-case outcome, providing both a clinical tool and new insights into disease biology.

Methods and Datasets

The researchers employed a large-scale proteomic approach, measuring the abundance of 4,701 unique human proteins using the SomaScan assay. To handle the high dimensionality of the data, they used L1-regularized logistic regression (LASSO) for feature selection. This allowed them to distill the candidates down to 92 predictive proteins for severe disease and 67 for critical disease.

Study workflow showing training and testing cohorts, model development, and feature selection.

The study utilized two geographically and genetically distinct cohorts:

  • Training Cohort (BQC19): 417 individuals from Montreal, Canada.
  • Testing Cohort (Mount Sinai): 569 individuals from New York City.

Key Results

The protein-based model significantly outperformed standard clinical risk factors. A baseline model using only age and sex achieved an Area Under the Curve (AUC) of 65% in the test cohort, whereas the protein-based model improved this to 86% for severe disease. For critical disease, the protein model achieved an AUC of 0.796 in the test cohort, compared to 0.589 for the baseline model.

ROC curves for predicting critical COVID-19, comparing the protein-based model (AUC=0.796) against the baseline model (AUC=0.589) in the test cohort.

The model for "severe" COVID-19 (requiring any oxygen) achieved a Positive Predictive Value (PPV) of 88.9%, indicating high reliability in identifying patients who would eventually need respiratory support.

Biological Insights

Analysis of the selected proteins revealed that more than half of the enriched biological pathways were not immune-related. This suggests that COVID-19 severity is driven by a complex interplay of lung tissue damage, metabolic shifts, and vascular changes. A top predictor was SFTPD (Surfactant Protein D), a lung-specific protein whose presence in the blood likely signals early alveolar damage. Interestingly, adding standard clinical risk factors like diabetes or hypertension to the protein model did not significantly improve accuracy, suggesting the proteome already captures the biological impact of these conditions.

Discussion and Future Work

The model's performance remained stable across different genetic ancestries, suggesting the identified proteins represent a universal biological response to the virus. However, clinical implementation faces hurdles regarding cost and infrastructure. Measuring thousands of proteins is currently impractical for routine hospital use; a smaller, specialized assay focusing on the 92 identified proteins would be necessary. Future research should also include longitudinal sampling to track how these protein levels evolve during patient recovery or decline.

Summary updated

Abstract

Predicting COVID-19 severity is difficult, and the biological pathways involved are not fully understood. To approach this problem, we measured 4701 circulating human protein abundances in two independent cohorts totaling 986 individuals. We then trained prediction models including protein abundances and clinical risk factors to predict COVID-19 severity in 417 subjects and tested these models in a separate cohort of 569 individuals. For severe COVID-19, a baseline model including age and sex provided an area under the receiver operator curve (AUC) of 65% in the test cohort. Selecting 92 proteins from the 4701 unique protein abundances improved the AUC to 88% in the training cohort, which remained relatively stable in the testing cohort at 86%, suggesting good generalizability. Proteins selected from different COVID-19 severity were enriched for cytokine and cytokine receptors, but more than half of the enriched pathways were not immune-related. Taken together, these findings suggest that circulating proteins measured at early stages of disease progression are reasonably accurate predictors of COVID-19 severity. Further research is needed to understand how to incorporate protein measurement into clinical care.
BibTeX
@inproceedings{su2023circulating,
  author = {Chen‐Yang Su and Sirui Zhou and Edgar Gonzalez‐Kozlova and Guillaume Butler‐Laporte and Elsa Brunet‐Ratnasingham and Tomoko Nakanishi and Wonseok Jeon and David Morrison and Lætitia Laurent and Jonathan Afilalo and Marc Afilalo and Danielle Henry and Yiheng Chen and Julia Carrasco-Zanini and Yossi Farjoun and Maik Pietzner and Nofar Kimchi and Zaman Afrasiabi and Nardin Rezk and Meriem Bouab and Louis Petitjean and Charlotte Guzman and Xiaoqing Xue and Chris Tselios and Branka Vulesevic and Olumide Adeleye and Tala Abdullah and Noor Almamlouk and Yara Moussa and Chantal DeLuca and Naomi Duggan and Erwin Schurr and Nathalie Brassard and Madéleine Durand and Diane M. Del Valle and Ryan C. Thompson and Mario A. Cedillo and Eric E. Schadt and Kai Nie and Nicole W. Simons and Konstantinos Mouskas and Nicolas Zaki and Manishkumar Patel and Hui Xie and Jocelyn Harris and Robert Marvin and Esther Cheng and Kevin Tuballes and Kimberly Argueta and Ieisha Scott and Celia M.T. Greenwood and Clare Paterson and Michael Hinterberg and Claudia Langenberg and Vincenzo Forgetta and Joëlle Pineau and Vincent Mooser and Thomas U. Marron and Noam D. Beckmann and Seunghee Kim‐Schulze and Alexander W. Charney and Sacha Gnjatic and Daniel E. Kaufmann and Miriam Mérad and J. Brent Richards and J. Brent Richards},
  title = {{Circulating proteins to predict COVID-19 severity}},
  booktitle = {Scientific},
  year = {2023},
  doi = {10.1038/s41598-023-31850-y},
}
Figure