A three-protein urinary panel comprising TNFRSF1A, F9, and PRDX3 may provide a noninvasive approach for differentiating acute myocarditis from acute myocardial infarction.
Key Findings
Results
Urinary proteomics identified large numbers of differentially expressed proteins in both acute myocarditis (AM) and acute myocardial infarction (AMI) compared with controls.
The discovery cohort included 42 AM patients and 80 AMI patients.
In the training set, 880 differentially expressed proteins were identified in AM versus controls.
In the training set, 1099 differentially expressed proteins were identified in AMI versus controls.
Urinary proteomes were profiled by data-independent acquisition mass spectrometry.
Results
A three-protein urinary panel (TNFRSF1A, F9, PRDX3) was selected by LASSO regression to discriminate AM from AMI and achieved high diagnostic accuracy in the training set.
The model was built using LASSO with 1000 bootstrap resamplings in the training set.
The panel achieved an AUC of 0.954 in the training set.
The discovery cohort was stratified by age and sex and randomly divided into training and internal validation sets at a 3:1 ratio.
Results
The protein panel outperformed clinical models for distinguishing AM from AMI.
The panel exceeded the clinical models in the training set (all DeLong's P < 0.05).
Results
Adding the protein panel to clinical models further improved diagnostic performance.
Addition of the panel to either clinical model improved performance (both P < 0.001).
Results
The panel's performance was confirmed in internal validation and in an independent cohort using ELISA.
The model was validated internally in the held-out validation set.
It was further verified in an independent cohort using ELISA.
The abstract reports that similar performance was confirmed in the validation cohorts.
What This Means
People who arrive at the hospital with sudden chest pain may have a heart attack (acute myocardial infarction) or inflammation of the heart muscle (acute myocarditis). These conditions can look alike at first, but they require different treatments, and telling them apart without invasive tests is difficult. This study looked for a urine-based way to separate them.
Researchers analyzed the proteins in urine from 42 patients with myocarditis and 80 with heart attack, and compared them with controls. They found hundreds of proteins that differed in each condition. Using a statistical selection method, they narrowed these to a panel of three proteins: TNFRSF1A, F9, and PRDX3. The panel distinguished the two conditions very well (AUC 0.954 in the training data) and did better than models based on standard clinical information. Combining the panel with clinical models improved accuracy further. Similar results were seen in the internal validation data and in a separate group of patients tested with ELISA, a common lab method.
This research suggests a simple urine test might one day help doctors tell myocarditis from heart attack without invasive procedures. However, the findings come from a relatively small number of patients, and the test would need confirmation in larger and more varied groups before it could be used in routine care.
Wang Y, Qin Y, Yang J, Wang Q, Wang A, Zhang J, et al.. (2026). Urinary Proteomic Panel Distinguishes Acute Myocarditis from Acute Myocardial Infarction.. Journal of proteome research. https://doi.org/10.1021/acs.jproteome.6c00106