Highlights
- The Atherosclerosis Risk in Communities (ARIC) study has identified five distinct plasma proteins—PTK7, CHAD, LRIG1, FBLN5, and CILP2—specifically associated with heart failure (HF) risk in individuals with diabetes.
- Four of these markers (PTK7, CHAD, LRIG1, FBLN5) are novel discoveries in the context of diabetic heart disease, while CILP2 reinforces known pathways.
- Pathways involving lipid metabolism, chronic inflammation, and brown adipose tissue were significantly over-represented in the diabetic HF proteomic signature.
- While NT-proBNP (NPPB) remains a universal marker for HF, these specific proteins offer a unique window into the pathophysiology of ‘diabetic cardiomyopathy’ independent of general cardiac strain.
Background
Heart failure (HF) remains a leading cause of morbidity and mortality among individuals with diabetes mellitus, who face a two- to four-fold higher risk compared to the general population. While hypertension and coronary artery disease contribute significantly, the existence of a specific ‘diabetic cardiomyopathy’—a primary dysfunction of the myocardium in the absence of other discernible causes—has long been hypothesized. However, the exact molecular transition from metabolic derangement to overt cardiac failure has remained elusive.
Traditional clinical biomarkers, such as NT-proBNP, are effective for general HF screening but lack the specificity to distinguish the unique metabolic and structural drivers of cardiac dysfunction in the diabetic milieu. Proteomics, the large-scale study of proteins, provides a transformative approach to identifying these underlying mechanisms. By analyzing thousands of proteins simultaneously, researchers can map the ‘proteomic signature’ of a disease state, moving beyond single-protein associations toward systems-level understanding.
Key Content
The ARIC Proteomics Framework: Methodology and Discovery
The Atherosclerosis Risk in Communities (ARIC) study represents one of the most robust longitudinal cohorts for cardiovascular research. In this recent analysis, researchers assessed 10,189 participants who were free of HF at baseline (mean age 57 years). The cohort was remarkably diverse, including 56% women and 22% Black adults, with 14% of the population having a baseline diagnosis of diabetes.
Using a high-throughput platform, the study measured 4,955 plasma proteins. The discovery phase focused on 993 individuals with diabetes. Over a 24-year follow-up period, 2,417 HF events were recorded, with 605 occurring among those with diabetes. Using Cox regression models adjusted for age, sex, and race, the team identified proteins specifically associated with the development of HF within the diabetic subgroup.
Specific Proteomic Signatures: The ‘Diabetic Five’
The most significant finding of the study was the identification of five proteins that were uniquely related to HF risk in those with diabetes, but not in those without the condition. This specificity suggests that these proteins represent pathways activated by the diabetic state rather than general cardiac injury.
1. PTK7 (Inactive Tyrosine-Protein Kinase 7)
A novel marker in this context, PTK7 is an evolutionary conserved member of the receptor tyrosine kinase superfamily. While it lacks catalytic activity, it acts as a co-receptor in the Wnt signaling pathway, which is critical for tissue homeostasis and repair. In the diabetic heart, alterations in PTK7 might reflect aberrant tissue remodeling or impaired reparative capacity of the myocardium.
2. CHAD (Chondroadherin)
CHAD is a cartilage-associated protein that mediates cell adhesion. Its presence in the plasma as a predictor of HF in diabetes suggests that extracellular matrix (ECM) turnover and fibrotic changes may be specifically accelerated or altered by hyperglycemia and insulin resistance.
3. LRIG1 (Leucine-rich repeat and immunoglobulin-like domain-containing nogo receptor-interacting protein 1)
LRIG1 is known for its role in regulating growth factor signaling, particularly the Epidermal Growth Factor Receptor (EGFR) family. Its association with HF in diabetes points toward dysregulated growth signaling, which can lead to pathological hypertrophy.
4. FBLN5 (Fibulin-5)
Fibulin-5 is essential for elastogenesis. In the context of the diabetic heart, FBLN5 might serve as a marker for arterial stiffness and myocardial non-compliance, hallmarks of heart failure with preserved ejection fraction (HFpEF), which is highly prevalent in diabetic populations.
5. CILP2 (Cartilage Intermediate Layer Protein 2)
Unlike the others, CILP2 had been previously implicated in cardiac processes, but its strong specific association with diabetic HF reinforces the importance of cartilage-like proteins in the cardiac ECM under metabolic stress.
Validation and Generalizability: The MESA Comparison
To ensure these findings were not unique to the ARIC cohort, the researchers performed external validation using the Multi-Ethnic Study of Atherosclerosis (MESA), which included 5,233 participants (633 with diabetes). Six of the internally validated proteins from ARIC, including the five specific markers, were replicated in MESA with high statistical significance (FDR q < 0.05). This cross-cohort validation in a multi-ethnic population enhances the clinical relevance and generalizability of the identified proteomic signature.
Mechanistic Insights and Pathway Over-representation
Beyond individual proteins, the study utilized pathway analysis to identify the biological processes driving the risk. Three themes emerged as dominant:
- Lipid Metabolism: Proteins involved in fatty acid transport and oxidation were highly prevalent. This aligns with the ‘lipotoxicity’ theory of diabetic cardiomyopathy, where the heart shifts away from glucose utilization toward inefficient and damaging fatty acid metabolism.
- Chronic Inflammation: The proteomic signature confirmed that low-grade, systemic inflammation is a critical mediator in the diabetic heart, likely contributing to microvascular dysfunction and fibrosis.
- Brown Adipose Tissue (BAT) Activity: Surprisingly, pathways related to BAT were identified. This suggests a systemic metabolic crosstalk where the thermogenic capacity of adipose tissue might influence cardiac energy balance in diabetes.
Expert Commentary
The findings from the ARIC study represent a significant leap toward precision medicine in metabolic cardiology. By identifying proteins that are uniquely elevated or predictive in the diabetic population, we move closer to developing targeted screening tools that go beyond the ‘one-size-fits-all’ approach of NT-proBNP.
From a clinical perspective, the discovery of PTK7 and FBLN5 as novel markers is particularly intriguing. These proteins are involved in structural integrity and signaling rather than just acute strain. This suggests that the diabetic heart undergoes a unique form of structural remodeling long before clinical symptoms appear. If these markers can be validated in clinical trials, they may serve as early warning signs, allowing clinicians to intensify treatment with SGLT2 inhibitors or GLP-1 receptor agonists—therapies known to provide significant cardioprotection in diabetes.
However, several limitations remain. As an observational study, ARIC cannot definitively prove that these proteins cause heart failure; they may simply be markers of the underlying disease process. Furthermore, while the external validation in MESA is a strength, the clinical utility of measuring these five proteins in routine practice still requires prospective cost-benefit analysis and the development of standardized assays.
Conclusion
The ARIC study has successfully identified a unique proteomic ‘barcode’ for heart failure risk in individuals with diabetes. By highlighting five specific proteins—PTK7, CHAD, LRIG1, FBLN5, and CILP2—the research provides both novel biomarkers for risk stratification and new mechanistic targets for therapeutic intervention. Future research should focus on whether modulating these protein pathways can arrest the progression of diabetic cardiomyopathy and whether these markers can be used to monitor the efficacy of modern anti-diabetic therapies in preventing heart failure.
References
- Echouffo-Tcheugui JB, et al. Proteomic Signatures of Cardiac Dysfunction Among People With Diabetes: The Atherosclerosis Risk in Communities Study. Circulation. Heart failure. 2026;19(3):e013171. PMID: 41569286.
- Ndumele CE, et al. Diabetes and Heart Failure: Pathophysiology, Epidemiology, and Emerging Therapeutic Targets. Circulation. 2023.
- Selvin E, et al. Proteomics in Large-Scale Cohorts: Lessons from ARIC. Journal of Clinical Investigation. 2024.