C-Peptide: The Insulin Byproduct Biomarker

Pancreatic Peptides

1:1 ratio with insulin

C-peptide is released in equimolar amounts with insulin from pancreatic beta cells, but unlike insulin, it bypasses liver extraction, making it the most reliable measure of endogenous insulin production.

Maddaloni et al., Diabetes Obes Metab, 2022

Maddaloni et al., Diabetes Obes Metab, 2022

View as image

For decades after its discovery, C-peptide was considered biological waste: a 31-amino-acid fragment cleaved from proinsulin during insulin production, released into the bloodstream at a 1:1 ratio with insulin, then filtered out by the kidneys. Clinicians used it as a convenient proxy for insulin secretion because, unlike insulin, C-peptide is not extracted by the liver on first pass. But "inert byproduct" was the label. That changed in the late 1990s and 2000s, when research revealed that C-peptide binds to cell membranes, activates intracellular signaling cascades, and may protect against the microvascular complications of type 1 diabetes. Today, C-peptide sits at the intersection of diagnostic tool and potential therapeutic target, used routinely in diabetes classification while its biological activity remains incompletely understood. This is the full story of a peptide that was overlooked for 30 years, from its role in pancreatic peptide biology to its clinical applications.

What C-peptide is and where it comes from

C-peptide (connecting peptide) is a 31-amino-acid fragment that connects the A and B chains of insulin within the proinsulin molecule. When beta cells in the pancreatic islets of Langerhans process proinsulin into mature insulin, they cleave C-peptide and release both molecules into the portal vein in equimolar quantities. Every molecule of insulin secreted is accompanied by exactly one molecule of C-peptide.

The critical difference for clinical measurement is hepatic extraction. The liver removes approximately 50% of insulin on first pass through the portal circulation. C-peptide passes through the liver essentially unextracted. This means that peripheral blood insulin levels reflect both beta-cell secretion and variable hepatic clearance, while C-peptide levels reflect beta-cell secretion alone. C-peptide also has a longer plasma half-life than insulin (approximately 30 minutes versus 5-10 minutes for insulin), making it a more stable measurement target.^[3]

The kidneys are the primary clearance organ for C-peptide. Renal impairment elevates C-peptide levels independently of beta-cell function, which must be accounted for in clinical interpretation. This is one of several confounders that make C-peptide testing more nuanced than simply drawing blood and reading a number.

C-peptide as a diagnostic biomarker

Classifying diabetes type

The most established clinical use of C-peptide is distinguishing between diabetes types. Type 1 diabetes is characterized by autoimmune destruction of beta cells, leading to absolute insulin deficiency. Type 2 diabetes involves insulin resistance with relatively preserved (or even elevated) beta-cell function in early stages. C-peptide levels reflect this distinction directly.^[4]

Jones and Hattersley (2013) reviewed the clinical utility of C-peptide measurement and established practical thresholds: a random nonfasting C-peptide below 0.2 nmol/L indicates severe insulin deficiency consistent with type 1 diabetes, while levels above 0.6 nmol/L suggest substantial residual beta-cell function more consistent with type 2 diabetes. The intermediate range (0.2-0.6 nmol/L) requires clinical context for interpretation.^[4]

The practical impact is real. Measurement of random C-peptide in people with a clinical diagnosis of type 1 diabetes for 3 or more years allowed reclassification of 6.8% of the tested cohort into different diabetes types. This changes treatment: a patient reclassified from type 1 to monogenic diabetes or late-onset type 2 may be able to switch from insulin to oral agents, improving both quality of life and glycemic control.^[1]

Measuring beta-cell function over time

Beyond diagnosis, serial C-peptide measurements track the decline of beta-cell function in type 1 diabetes and the progression of beta-cell exhaustion in type 2 diabetes. In type 1 diabetes clinical trials, stimulated C-peptide (measured after a standardized mixed-meal tolerance test) is the primary endpoint for disease-modifying therapies. Evidence supports C-peptide as a validated surrogate endpoint for predicting clinical benefits in trials of immunotherapies aimed at preserving beta-cell function.^[1]

In type 2 diabetes, C-peptide levels can guide therapeutic decisions. Lin et al. (2025) reviewed the emerging role of C-peptide in T2D management, noting that C-peptide levels correlate with the likelihood of achieving glycemic control on specific drug classes. Patients with higher C-peptide levels may respond better to GLP-1 receptor agonists, which require functioning beta cells to stimulate insulin secretion. Patients with low C-peptide are more likely to need exogenous insulin regardless of their nominal diabetes classification.^[7]

Testing methods and the standardization problem

C-peptide can be measured in several ways: fasting, random nonfasting, or stimulated (after a glucagon injection or mixed-meal tolerance test). Leighton et al. (2017) reviewed the practical aspects of C-peptide testing and concluded that random nonfasting C-peptide offers the best combination of patient convenience and diagnostic accuracy for most clinical questions. It correlates well with stimulated C-peptide and avoids the cost and complexity of formal stimulation testing.^[3]

However, the 13 most commonly used C-peptide assays show systematic differences of up to 38% between assay pairs. This means a patient tested on one platform might receive a different clinical interpretation than if tested on another. The American Diabetes Association (ADA) and European Association for the Study of Diabetes (EASD) have introduced C-peptide cut-off values in their guidelines, but these thresholds assume standardized measurement, which has not been achieved. International efforts to standardize C-peptide measurement are ongoing but incomplete.^[1]

Screening for insulin resistance

Khan et al. (2018) investigated C-peptide's potential as a screening biomarker for insulin resistance in both diabetic and non-diabetic individuals. Elevated fasting C-peptide levels correlated with insulin resistance, metabolic syndrome markers, and cardiovascular risk factors. The advantage over insulin measurement is the same as for diabetes classification: C-peptide is not confounded by hepatic first-pass extraction and is more analytically stable in blood samples.^[6]

Beyond biomarker: C-peptide's biological activity

The paradigm shift

For 30 years after its discovery, C-peptide was classified as biologically inert. This changed in the 1990s when Wahren and colleagues demonstrated that C-peptide binds to cell membranes (likely through a G-protein-coupled receptor, though the specific receptor has not been conclusively identified) and activates intracellular signaling pathways.

Nordquist et al. (2008) reviewed the molecular mechanisms of C-peptide action, identifying several distinct biological effects: stimulation of endothelial nitric oxide synthase (eNOS) leading to vasodilation, activation of Na+/K+-ATPase in renal tubules (reducing sodium reabsorption and glomerular hyperfiltration), and anti-inflammatory effects through reduction of leukocyte adhesion to endothelial cells. These effects are specifically relevant to the microvascular complications of type 1 diabetes, where C-peptide is absent.^[5]

The biological activity of C-peptide creates a testable hypothesis: type 1 diabetes patients lack not just insulin but also C-peptide, and the absence of C-peptide may contribute to the microvascular complications (neuropathy, nephropathy, retinopathy) that characterize the disease. If true, C-peptide replacement alongside insulin could reduce complications.

Neuropathy evidence

The neuropathy hypothesis has been tested in clinical trials. Ekberg et al. (2007) conducted a double-blind, placebo-controlled study of C-peptide replacement in type 1 diabetes patients with early-stage neuropathy. Patients received subcutaneous C-peptide (1.5 mg/day) or placebo for 6 months. C-peptide treatment improved sensory nerve conduction velocity and reduced vibration perception thresholds compared to placebo, indicating measurable improvement in peripheral nerve function.^[8]

A follow-up study (Ekberg et al., 2008) extended these observations, demonstrating that respiratory heart rate variability (a measure of autonomic nerve function) increased during C-peptide treatment but was unchanged during placebo. Thermal thresholds also improved during treatment. The neuroprotective mechanism appears to involve C-peptide-mediated stimulation of Na+/K+-ATPase activity in nerve fibers, correcting the impaired nerve membrane potential that occurs in diabetes.^[9]

These trials represent some of the strongest evidence that C-peptide has therapeutic activity, not just diagnostic utility. However, no C-peptide replacement therapy has been approved for clinical use. The development of C-peptide as a drug has been complicated by its status as a naturally occurring peptide (difficult to patent), its short half-life requiring frequent injection or continuous infusion, and the pharmaceutical industry's focus on insulin analogs rather than C-peptide replacement.

Kidney protection

C-peptide replacement in type 1 diabetes models reduced glomerular hyperfiltration (one of the earliest signs of diabetic nephropathy) and decreased urinary albumin excretion. Clinical observations showed that during C-peptide treatment, urinary albumin excretion decreased from 58 micrograms/min to 34 micrograms/min over 3 months, and tended to increase during the placebo period.^[5]

The mechanism involves C-peptide's inhibition of Na+/K+-ATPase in renal tubules, which reduces sodium reabsorption and the tubuloglomerular feedback that drives hyperfiltration. By reducing the workload on the glomerulus, C-peptide may slow the progression from microalbuminuria to overt diabetic nephropathy. This is the same enzymatic target that drives the neuroprotective effect, suggesting a unified mechanism of action across multiple microvascular complications.

C-peptide in the broader diabetes picture

Chen et al. (2023) published an updated review of C-peptide's role in diabetes and its complications in Frontiers in Endocrinology, synthesizing evidence across diagnostic, prognostic, and therapeutic applications. They noted that C-peptide measurement is evolving from a specialized research tool to a routine clinical test, driven by recognition that diabetes classification based on age of onset and clinical presentation alone misclassifies a substantial minority of patients.^[2]

The relationship between C-peptide and other pancreatic peptides is complex. Beta-cell destruction in type 1 diabetes eliminates not just insulin and C-peptide but also amylin (IAPP), another co-secreted peptide with roles in satiety and gastric emptying. The combination of insulin, C-peptide, and amylin deficiency may explain why type 1 diabetes complications differ from type 2 complications, even when blood glucose is equivalently controlled.

For type 2 diabetes, the picture is different. C-peptide levels are typically normal or elevated in early T2D (reflecting the compensatory hyperinsulinemia that accompanies insulin resistance) and decline as beta cells exhaust. This trajectory makes C-peptide useful for staging type 2 diabetes and predicting when patients will need insulin therapy. The emerging evidence that C-peptide levels predict response to GLP-1 receptor agonists adds a therapeutic dimension to this prognostic value.^[7]

Current limitations and open questions

Receptor identification. Despite strong evidence that C-peptide binds to cell membranes and activates intracellular signaling, the specific receptor has not been conclusively identified. A G-protein-coupled receptor (GPR146) has been proposed as a candidate, but the evidence is not definitive. Without a confirmed receptor, the molecular pharmacology of C-peptide cannot be fully characterized.

Failed drug development. Cebix, a pharmaceutical company, developed a long-acting C-peptide analog (CBX129801/Ersatta) for type 1 diabetes complications. A Phase II trial for diabetic peripheral neuropathy failed to meet its primary endpoint in 2014, and the company ceased operations. This setback dampened commercial interest in C-peptide therapeutics, despite the positive results of earlier academic trials.

Assay standardization. The 38% inter-assay variability undermines the clinical thresholds established in guidelines. Two patients with identical beta-cell function could receive different diagnoses depending on which assay their laboratory uses. International standardization efforts are in progress but not complete, limiting the utility of guideline-specified cutoffs.^[1]

Biological activity debate. Some researchers have questioned whether C-peptide's biological effects are clinically meaningful at physiological concentrations, or whether the observed effects in trials reflected supraphysiological dosing. The Cebix Phase II failure could support either interpretation: the drug did not work, or the trial design was inadequate to detect the effect.

The Bottom Line

C-peptide has evolved from a presumed inert byproduct of insulin processing to a clinically valuable biomarker for diabetes classification and beta-cell function assessment, with evidence of biological activity in neuropathy, nephropathy, and vascular function. Its diagnostic applications are expanding as guidelines incorporate C-peptide thresholds for diabetes subtyping and treatment decisions, though assay standardization remains an unresolved barrier. The therapeutic potential demonstrated in early clinical trials has not been successfully translated to an approved drug, leaving C-peptide in an unusual position: widely used as a measurement but not available as a treatment.

Frequently Asked Questions

What does a C-peptide test measure?

A C-peptide test measures the blood level of connecting peptide, a 31-amino-acid fragment released from pancreatic beta cells in a 1:1 ratio with insulin. Because C-peptide is not extracted by the liver (unlike insulin), it provides a more accurate measure of endogenous insulin production. It is used to classify diabetes type, assess beta-cell function, and guide treatment decisions (Leighton et al., 2017).

What is a normal C-peptide level?

Fasting C-peptide typically ranges from 0.3-0.6 nmol/L in healthy individuals. Levels below 0.2 nmol/L indicate severe insulin deficiency consistent with type 1 diabetes, while levels above 0.6 nmol/L suggest preserved beta-cell function more consistent with type 2 diabetes (Jones and Hattersley, 2013). These thresholds vary by assay due to a standardization gap of up to 38% between testing platforms.

Is C-peptide biologically active?

Yes. Once considered inert, C-peptide binds to cell membranes, stimulates nitric oxide synthase, activates Na+/K+-ATPase, and has vasodilatory effects on muscle, skin, and kidney blood vessels (Nordquist et al., 2008). In clinical trials, C-peptide replacement improved sensory nerve conduction velocity in type 1 diabetes patients with neuropathy (Ekberg et al., 2007). No C-peptide replacement therapy is currently approved.

Why test C-peptide instead of insulin?

C-peptide is a more reliable measure of beta-cell secretion because the liver extracts approximately 50% of insulin on first pass, while C-peptide passes through unextracted. C-peptide also has a longer half-life (30 minutes vs. 5-10 minutes for insulin), making it more analytically stable. Additionally, C-peptide measurement is not affected by exogenous insulin therapy, allowing assessment of residual beta-cell function in insulin-treated patients.

Can C-peptide levels predict which diabetes treatment will work?

Emerging evidence suggests yes. Higher C-peptide levels in type 2 diabetes correlate with better response to GLP-1 receptor agonists, which require functioning beta cells to work. Low C-peptide predicts the need for exogenous insulin regardless of diabetes classification (Lin et al., 2025). C-peptide is also being explored as a predictor of successful insulin withdrawal in T2D patients on GLP-1 therapy.

Does C-peptide protect against diabetic complications?

In type 1 diabetes patients, C-peptide replacement reduced sensory neuropathy symptoms, improved autonomic nerve function, and decreased urinary albumin excretion (a marker of kidney damage) in clinical trials (Ekberg et al., 2007, 2008). The mechanism involves stimulation of nitric oxide production and Na+/K+-ATPase activity. A Phase II trial of a long-acting C-peptide analog failed in 2014, and no C-peptide therapy is approved for complication prevention.