Biomarker Potential of C-Peptide and HbA1c for Screening Diabetic and Non-Diabetic Individuals: A Cross-Sectional Analytical Study

health challenge, with its prevalence steadily increasing due to lifestyle changes and aging populations [1]. Early identification of individuals at risk is crucial to prevent progression to overt diabetes and associated complications. Glycated hemoglobin (HbA1c) has been widely used as a diagnostic and prognostic biomarker, reflecting mean plasma glucose levels over the preceding 8–12 weeks [2]. However, its accuracy can be influenced by factors such as hemoglobin variants, anemia, and ethnicity, necessitating complementary biomarkers to improve diagnostic precision [3].

 

C-peptide, a cleavage product of proinsulin, provides a direct measure of pancreatic β-cell function and endogenous insulin secretion [4]. Unlike insulin, it is not subject to hepatic extraction, making it a reliable indicator of β-cell reserve. Emerging evidence highlights its utility in distinguishing between type 1 and type 2 diabetes, monitoring residual β-cell function, and identifying metabolic dysfunction even in normoglycemic individuals [5].

 

Recent studies have emphasized that combining C-peptide with HbA1c may enhance screening sensitivity, allowing stratification of prediabetic and non-diabetic individuals based on insulin resistance and secretion profiles [1,5]. Such an integrative approach could refine early diabetes risk assessment and improve targeted interventions. Thus, the present research explores the biomarker potential of C-peptide alongside HbA1c for effective screening and discrimination between diabetic and non-diabetic individuals.

 

 

3.1 Participant Characteristics

A total of 300 participants were enrolled, equally distributed among non-diabetic (Group I), prediabetic (Group II), and diabetic (Group III) cohorts. The mean age was 47.2 ± 9.8 years, with 54% males and 46% females. There were no significant differences in age or gender among groups (p > 0.05).

Biochemical characteristics across the three groups are presented in Table 1. A consistent and statistically significant increase in HbA1c, C-peptide, and fasting glucose levels was observed from non-diabetic to diabetic individuals (p < 0.001). The data reveal a progressive β-cell secretory response (C-peptide rise) corresponding to worsening glycemic control, confirming the metabolic continuum between euglycemia and diabetes.

 

Table 1. Biochemical characteristics among study groups (n = 300).

Parameter	Group I (Non-diabetic)	Group II (Prediabetic)	Group III (Diabetic)	p-value
N	100	100	100	—
HbA1c (%)	5.3 ± 0.2	6.0 ± 0.3	7.8 ± 0.6	<0.001
C-peptide (ng/mL)	1.1 ± 0.3	2.5 ± 0.5	3.9 ± 0.8	<0.001
Fasting Glucose (mg/dL)	88 ± 9	109 ± 12	152 ± 22	<0.001

Values are mean ± SD. ANOVA followed by Tukey’s post hoc test.

 

3.2 Relationship between HbA1c and C-Peptide

Pearson’s correlation analysis demonstrated a strong positive relationship between HbA1c and C-peptide levels (r = 0.68, p < 0.001) across the study population (Figure 1). This suggests that as HbA1c rises, compensatory hyperinsulinemia leads to elevated C-peptide levels in early disease stages. The correlation was most pronounced in prediabetic individuals (r = 0.72), indicating β-cell overactivity preceding decline.

Figure 1. Scatter plot showing correlation between HbA1c and C-peptide concentrations across all study participants (r = 0.68, p < 0.001).

 

3.3 Group-wise Comparison of Biomarkers

Figure 2 illustrates the comparative pattern of HbA1c and C-peptide levels among the three groups. C-peptide concentrations increased proportionally with HbA1c, emphasizing its potential role in differentiating prediabetic and diabetic states.

Figure 2. Comparative analysis of mean HbA1c (%) and C-peptide (ng/mL) among non-diabetic, prediabetic, and diabetic groups.

 

3.4 Diagnostic Performance Analysis

Receiver Operating Characteristic (ROC) curve analysis revealed that HbA1c alone achieved an area under the curve (AUC) of 0.84 (95% CI: 0.78–0.90), while C-peptide alone had an AUC of 0.80 (95% CI: 0.74–0.87). When both biomarkers were combined, the diagnostic accuracy improved significantly (AUC = 0.92; 95% CI: 0.88–0.95; p < 0.001), as illustrated in Figure 3.

 

The optimal threshold values for detecting diabetes were HbA1c ≥ 6.2% and C-peptide ≥ 2.8 ng/mL, yielding a sensitivity of 91% and specificity of 88%.

 

 

 

 

Table 2. Diagnostic performance of HbA1c and C-peptide in diabetes detection.

Figure 3. ROC curve comparing diagnostic accuracy of HbA1c alone, C-peptide alone, and combined biomarkers for diabetes classification.

Interpretation

The combined use of HbA1c and C-peptide substantially enhances the discrimination of diabetic, prediabetic, and non-diabetic individuals. This integrative biomarker model improves both sensitivity and specificity, underscoring the clinical value of pairing glycemic and β-cell function markers in population-level diabetes screening.

This study provides strong evidence supporting the combined utility of C-peptide and HbA1c as integrated biomarkers for diabetes screening and classification. The observed significant correlation between these parameters (r = 0.68, p < 0.001) reinforces their complementary physiological roles in glycemic homeostasis. HbA1c represents long-term glucose exposure, while C-peptide directly reflects β-cell secretory capacity and endogenous insulin production [6,7].

 

Our findings are consistent with Huang and Luo (2023) [8], who demonstrated that the combined assay of serum C-peptide and glycated hemoglobin improved diagnostic sensitivity for early diabetes by over 15% compared to either marker alone. Similarly, Lin et al. (2025) [9] and Jones et al. (2024) [10] reported that residual C-peptide activity strongly correlates with metabolic stability and treatment response, supporting its clinical value beyond traditional glucose testing.

 

A novel insight from this study is the enhanced diagnostic performance (AUC = 0.92) of the combined model compared to single-marker use. This echoes findings from Sim et al. (2025) [11], who developed AI-driven biomarker integration frameworks that improved early diabetes prediction accuracy. Furthermore, Toprak et al. (2023) [12] reported that the HbA1c/C-peptide ratio predicts thrombotic burden in diabetic cardiovascular complications, underscoring its prognostic utility.

 

From a pathophysiological standpoint, our data align with Hendriks et al. (2025) [13] and Aanstoot et al. (2024) [14], who highlighted that early β-cell stress—manifested as rising C-peptide despite normoglycemia—marks the prediabetic phase. Zheng et al. (2025) [10] and Li et al. (2025) [15] further linked C-peptide fluctuations to microvascular complications such as retinopathy, emphasizing its systemic biomarker relevance.

 

Standardization remains a limitation; Schleicher et al. (2025) [16] called for unified C-peptide assay calibration to ensure inter-laboratory consistency. Incorporating novel approaches such as urinary C-peptide creatinine ratios [17] and machine-learning–based conversion models [18] could enhance clinical feasibility.

 

Collectively, the evidence suggests that dual-marker screening using HbA1c and C-peptide offers a superior, physiologically grounded diagnostic framework capable of identifying at-risk individuals before overt diabetes develops. Integrating these biomarkers into clinical screening protocols could refine early intervention strategies and enable precision-based glycemic monitoring.

The present study underscores the combined diagnostic potential of HbA1c and C-peptide as complementary biomarkers for screening and stratifying diabetic and non-diabetic individuals. Our findings demonstrate that the integration of C-peptide, a marker of endogenous insulin secretion, with HbA1c, a measure of long-term glycemic control, significantly improves diagnostic accuracy and sensitivity. The combined model achieved an AUC of 0.92, outperforming either biomarker alone, confirming its reliability for early metabolic dysfunction detection.

 

Physiologically, this dual-marker approach provides a more nuanced understanding of β-cell activity and insulin resistance. Elevated C-peptide levels among prediabetic participants reflect compensatory β-cell responses, preceding the deterioration seen in overt diabetes. This aligns with recent global findings suggesting that early β-cell stress and hyperinsulinemia can be detected well before significant hyperglycemia manifests. Therefore, employing both biomarkers enables clinicians to detect metabolic dysregulation at an earlier, reversible stage.

 

From a clinical and public health perspective, the incorporation of C-peptide testing alongside HbA1c in routine screening protocols can improve risk stratification, particularly in ethnically diverse populations where HbA1c variability due to hemoglobinopathies is prevalent. Future longitudinal and multicentric studies are recommended to establish standardized cutoff values and evaluate cost-effectiveness in large-scale screening programs.

Ultimately, this study supports a paradigm shift toward dual-marker metabolic profiling for diabetes screening — a more physiologically accurate, sensitive, and preventive diagnostic strategy in the modern era of personalized medicine.

 

Funding Information:

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of Interest Statement:

No conflicts of interest related to this study.

 

1.     Alkandari H, Al‐Mousa A, Shaltout A, et al. Are children with familial type 1 diabetes from Kuwait different? Report on prevalence, clinical, biochemical, and immunological characteristics. Prim Care Diabetes. 2025;19(5):533–539. doi:10.1016/j.pcd.2025.06.001

2.     Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ. Translating the A1C assay into estimated average glucose values. Diabetes Care. 2008;31(8):1473–1478. doi:10.2337/dc08-0545

3.     Zhang Y, Liu L, Xu J, et al. Predictive factors and diagnostic value of vascular markers for carotid artery disease in type 2 diabetes mellitus patients. Medicine (Baltimore). 2025;104(40):e44333. doi:10.1097/MD.0000000000044333

4.     Głowińska-Olszewska B, Bossowski A. Residual beta-cell function and its relationship with C-peptide and betatrophin in children with type 1 diabetes. Front Endocrinol (Lausanne). 2025;16:1602777. doi:10.3389/fendo.2025.1602777

5.     Carlsson A, Samuelsson U, Lindblad B, et al. Clinical characteristics in Swedish children with and without autoantibodies at the time of type 1 diabetes diagnosis. Diabetes Care. 2025 Oct 15. doi:10.2337/dc25-0840

6.     Mohan V, Usha S, Sandeep S, et al. C-peptide and insulin resistance markers in the transition from prediabetes to diabetes. Diabetes Res Clin Pract. 2021;176:108868. doi:10.1016/j.diabres.2021.108868

7.     Huang Z, Luo X. Evaluation of combined serum C-peptide and glycated hemoglobin in the clinical diagnosis of diabetes. J Biomater Tissue Eng. 2023;13(12):3452–3459.

8.     Lin YY, McCrimmon RJ, Pearson ER. Exploring the potential role of C‐peptide in type 2 diabetes management. Diabet Med. 2025;42(4):e15469.

9.     Jones AG, Shields BM, Oram RA, Dabelea DM. Clinical prediction models combining routine clinical measures have high accuracy in identifying youth-onset type 2 diabetes defined by maintained endogenous insulin secretion. Diabetes Care. 2024;47(12):2110–2119.

10.   Sim H, Ha HH, Kim H. Integrated biomarker analysis and next-generation AI for precision diabetes prediction. Toxicol Res. 2025;41(2):123–134. doi:10.1007/s43188-025-00312-0

11.   Toprak K, Kaplangöray M, Memioğlu T. HbA1c/C-peptide ratio is associated with angiographic thrombus burden and short-term mortality in patients presenting with ST-elevation myocardial infarction. Blood Coagul Fibrinolysis. 2023;34(9):678–685.

12.   Hendriks AEJ, Marcovecchio ML, Evans ML. Early detection of β-cell decline using home dried-blood-spot C-peptide levels in new-onset type 1 diabetes. Diabetes Care. 2025;48(3):456–463. doi:10.2337/dc25-0214

13.   Aanstoot HJ, Varkevisser RDM, Mul D, Dekker P. Cohort profile: The “Biomarkers of Heterogeneity in Type 1 Diabetes” study — a national prospective cohort study. BMJ Open. 2024;14(6):e082453. doi:10.1136/bmjopen-2023-082453

14.   Zheng Z, Liu N, Zhang Y, Gu X, Li H. Postprandial C-peptide to glucose ratio as a promising systemic marker of diabetic retinopathy in type 2 diabetes. Transl Vis Sci Technol. 2025;14(7):27. doi:10.1167/tvst.14.7.27

15.   Li Y, Liu B, Xue M, Zhang Y, Jia X, Guo K, Hao W. Correlation between C-peptide and diabetic retinopathy in patients with type 2 diabetes mellitus: a cross-sectional study. Sci Rep. 2025;15:19714. doi:10.1038/s41598-025-19714-z

16.   Schleicher E, Kabytaev K, Nauck M. Call for standardization of C-peptide measurement. J Diabetes Sci Technol. 2025;19(4):336–342. doi:10.1177/19322968251362848

17.   Alshamsi F, Pathan A, Yasin J, et al. Urinary C-peptide creatinine ratio as a non-invasive diagnostic tool for differentiating type 1 from type 2 diabetes mellitus in adult Emirati population. Front Endocrinol (Lausanne). 2025;16:12571621.

18.   Iwamoto Y, Kimura T, Sugisaki T, Dan K, Iwamoto H. Assay-specific machine-learning models converting insulin to C-peptide or C-peptide index and C-peptide to HOMA2-IR: single-center retrospective observational study. Clin Chim Acta. 2025;573:464–472. doi:10.1016/j.cca.2025.07.004

19.   Baranwal A, Bansal V, Shukla R. Emerging biomarkers and nanobiosensing strategies in diabetes. Biosensors (Basel). 2025;15(10):639. doi:10.3390/bios15100639

20.   Fortin E. Novel biomarkers in cardiovascular disease and dysglycaemia. Karolinska Institute Open Archive. 2025. Available from: https://openarchive.ki.se