Plasma Proteomics Significantly Enhances Heart Failure Risk Prediction in Type 2 Diabetes Patients
Background
Individuals with type 2 diabetes (T2D) face a substantially elevated risk of developing heart failure (HF), a severe and often debilitating complication. Current methods for predicting HF risk in T2D patients, which typically rely on clinical variables, polygenic risk scores, and N-terminal prohormone of brain natriuretic peptide (NT-proBNP) levels, often fall short in providing sufficiently precise stratification. There's a critical need to identify novel biomarkers and develop more robust predictive models to enable earlier identification of high-risk individuals and facilitate timely preventive interventions, thereby addressing a significant gap in current clinical practice.
Study Design
This prospective cohort study analyzed 2198 participants with type 2 diabetes from the United Kingdom Biobank. Researchers employed Cox proportional hazards models to investigate associations between 2920 plasma proteins and incident heart failure. Clinical and protein predictors were systematically selected using the least absolute shrinkage and selection operator (LASSO) method, based on 10-fold cross-validation. The predictive performance of the resulting models was rigorously evaluated using multiple metrics, including Harrell's C-index, calibration slope, net reclassification improvement (NRI), integrated discrimination improvement (IDI), decision curve analysis, and calibration plots to assess accuracy and utility.
Results
Over a median follow-up of 13.1 years, 298 participants developed incident heart failure. A substantial 455 proteins were significantly associated with HF risk (447 positively, 8 inversely), primarily involved in crucial biological pathways such as cell adhesion, extracellular space organization, signaling receptor activity, and cytokine-cytokine receptor interaction. The protein most strongly associated with increased HF risk was whey acidic protein (WAP) 4-disulfide core domain protein 2, showing a per-SD increment hazard ratio (HR) of 1.90 (95% CI: 1.65, 2.19). Conversely, apolipoprotein C-I was the top protein inversely associated with HF risk, with a per-SD increment HR of 0.75 (95% CI: 0.66, 0.85).
Key Findings
- 455 plasma proteins were significantly associated with incident heart failure risk in type 2 diabetes patients.
- The top protein increasing HF risk was WAP 4-disulfide core domain protein 2 (HR 1.90 per SD increment).
- Apolipoprotein C-I was the top protein inversely associated with HF risk (HR 0.75 per SD increment).
- A 17-protein risk score improved HF prediction, increasing the
C-indexby 0.091 to a maximum of 0.833. - Proteomic data enhanced prediction beyond clinical variables, polygenic risk, and
NT-proBNP.
Why It Matters
Identifying a robust proteomic signature for heart failure risk in type 2 diabetes patients represents a significant advancement for personalized medicine. This study demonstrates that a 17-protein risk score can substantially improve risk prediction beyond traditional clinical markers, offering a powerful new tool for clinicians. Integrating plasma proteomic profiling could enable earlier, more precise identification of T2D patients at high risk for HF, allowing for targeted preventive strategies and potentially delaying or preventing HF onset. While not a direct peptide intervention, this research provides a framework for developing diagnostic panels that could guide treatment decisions, including the judicious use of cardio-protective therapies like GLP-1RAs or SGLT2 inhibitors, by identifying the patients who stand to benefit most from early intervention. Further validation in diverse cohorts is needed before clinical implementation.
proteomics
heart-failure
type-2-diabetes
risk-prediction
biomarkers
cohort-study