Type 1 diabetes diagnosis: antibody tests and C-peptide

Brief Summary

Diabetes presents several forms that require different treatment. When your symptoms are ambiguous, or if you don’t respond to treatment, a way to confirm your diagnosis is to get a C-peptide test combined with a set of antibody tests. While none of these tests provide perfect discrimination:

  • If your C-peptides are high, you are probably unlikely to be suffering from type 1 diabetes right now.

  • If you count 2 or more positive antibody tests, you are at serious risk to develop type 1 diabetes in the future.

  • If you count one positive antibody test, you are at some risk to develop type 1 diabetes in the future.

  • If you are negative to all antibody tests, you can probably stop reading this forum.

  • If your C-peptides are low, your production of insulin is probably low, and you may be suffering from the onset of type 1 diabetes.

    • If you test positive for 2 or more antibody tests, you are very likely to have type 1 diabetes.
    • If you test positive to one antibody test, you are likely but not certain to have type 1 diabetes.
    • If you don’t test positive to any antibody test, you may have type one diabetes – or not.

Different types of diseases

Diabetes as a name covers several related but separate diseases. Among them, Type 1 diabetes is an autoimmune disease where the body’s own immune system attacks and eventually destroys the pancreas’ beta cells that generate insulin, while Type 2 diabetes is a disease where the body responds less and less to insulin (i.e. becomes insulin-resistant). There are other, somewhat related types of diabetes.

Distinguishing between Type 1 and Type 2 diabetes

We used to believe that type 1 diabetes was what you got as a child, while type 2 diabetes was for adults, but reality is very different.

We now know that type 1 diabetes can be acquired at any time in life, and that 50% of patients with type 1 diabetes acquire it after age 30. In the same manner, even very young children can acquire a genetic variant of type 2 diabetes called MODY, and children can acquire regular Type 2 diabetes as well – often in a manner that evolves faster into more dangerous complications. We are now finding that all types of diabetes, with no exception, carry important genetic components.

The early phases of both Type 1 and Type 2 diabetes can show similar symptoms. Yet it is essential to be able to distinguish them quickly, since their treatments are very different: precise diagnosis is critical. The primary way to distinguish between Type 1 and Type 2 diabetes is to test the patient for the presence of antibodies that might show the existence of an autoimmune disease. Since Type 1 diabetes in an autoimmune disease, we would expect to be able to detect the antibody markers of an autoimmune disease.

Antibody tests

Several antibodies appear associated with beta cell destruction, for which tests have been developed. They are:

  • IAA: insulin autoantibodies. These antibodies do not distinguish between endogenous insulin (made by the patient’s body) and exogenous insulin (externally added insulin, typically from injection). So, if the patient has already been exposed to exogenous insulin, this antibody test in not helpful.

  • GAD65/ GAD/ GADA: antibodies to glutamic acid decarboxylase 65, a neuroendocrin protein. This is probably the most common test in use today.

  • ICA: islet cell (cytoplasmic) antibodies.

  • IA-2A: insulinoma-associated protein 2 autoantibodies

  • ZnT8 autoantibodies. Zinc transporter 8 is expressed in the beta-cells and associated with storage and transport of insulin. The ZnT8 test is the most recent test, and is not always available in a test laboratory.

From test results to diagnosis: single antibody test

The essential issue of diagnosis is that none of these tests correlates perfectly with Type 1 diabetes. The frequencies given below are approximate and vary per study.

  • IAA is present in 40-50% if children with diabetes, but rarely in adults. A subset of children where IAA is present never develop diabetes. The presence of IAA can create a false positive response to some ICA tests.

  • GADA is present in approximately 70% of recently diagnosed patients with type 1 diabetes, but is detected less often as time goes on (although for a longer period than ICA). About 8% of glucose-normal people also have GAD antibodies.

  • ICA is present in approximately 70% of newly diagnosed patients with type 1 diabetes, but is detected less often as time goes on (rarely after 10 years). Relatives of Type 1 diabetics showing presence of ICA has 50% chance to develop Type 1 diabetes at 5 years and 90% at 10 years.

  • IA-2A is detected in about 50-60% of children with type 1 diabetes at diagnosis time, but is much less common in patients with type 1 diabetes diagnosed as adults. Because it develops later than IAA and other antibodies, its detection prior to diabetes symptoms is likely to mean rapid progression. IA-2A detection progressively fades with time after diagnosis, but persists longer than ICA. It is found in 2% of glucose-normals.

  • ZnT8 is found in approximately 60-70% of recently diagnosed children with Type 1 diabetes, and 20-30% of adults. It is found much more rarely in long-time patients and is practically never found in 50-year Joslin medalists. It is a particularly interesting test because it can often (3-4%) be found in patients with no other marker, or with one marker: in combination with other tests it makes a significant difference in overall diagnosis accuracy.

It is important to be aware of the fact that not all tests are created equal for antibody assays: ask your endocronologist’s practice to discuss test quality prior to scheduling antibody tests in a lab.

From test results to diagnosis: multiple antibody tests

The real power of antibody tests is when all tests are done at once. That is where predictive value is greatest.

95% of patients with Type 1 diabetes have at least one type of antibodies: “If four markers are measured—GADA, IA-2A, IAA, and ICA or ZnT8A—only 2–4% of patients are autoantibody negative, fewer than 10% have only one marker, and around 70% have three or four markers” (Bingley 2010). So, when using multiple antibodies as a criterion for diagnosis, the risk of a false negative is small (4%), although it exists: if you carry mulitiple antibodies and you have diabetes symptoms, you are most likely to have type 1 diabetes.

What about false positives: how likely are you not to have or develop type 1 diabetes when you carry one or more antibodies?

Although we have elements of response, we do not have an absolute answer, because to do so we would need to follow a population for many years to figure out if someone with antibodies eventually develops the disease. Quoting Bingley again: “Those with ICA alone had a 2.2% cumulative risk of diabetes within 5 yr, whereas the risks with one, two, or three additional antibodies were 17, 39, and 70%, respectively. Multiple antibodies are not only specific but also sensitive; 91% of […] relatives who progressed to diabetes had two or three markers in addition to ICA. This message is strikingly consistent between studies, irrespective of the autoantibodies tested.” These numbers cover a 5-year interval after testing. There is some belief that, if you wait long enough, everyone with multiple antibody markers will eventually get type 1 diabetes.

The evidence is a lot less clear with a single antibody. It is probable that some or many people with a single antibody do not develop Type 1 diabetes.

Insulin C-peptide test: a symptom of type 1 diabetes

While antibody tests are investigating the actual cause of the diabetes symptoms you may have, the C-peptide test is investigating a symptom: how much insulin are your beta-cells still producing?

When beta-cells in the pancreas make insulin, the manufacturing process produces one molecule of C-peptide for each molecule of insulin. By measuring C-peptides we can theoretically find out how much insulin is being produced by the pancreas – although the test is not perfect. If C-peptides are high, your body is probably producing a lot of insulin. If C-peptides are low, you can count this as a possible symptom of the onset of Type 1 diabetes.

However, multiple factors can cause this test to fail: discrimination is imperfect, and patients with type 1 diabetes can have detectable C-peptides in the normal range. Different assays and assessment methods give different results. Normal ranges themselves are not clearly established across all test manufacturers. The presence and titration of C-peptides in the blood are influenced by insulin resistance and by kidney function (since kidneys filter out some C-peptides, malfunctioning kidneys artificially increase C-peptides). C-peptide is a relatively good predictor, but not a perfect one.

Final note: T1 or T2?

A study in the UK showed that 12% of T2s tested positive to at least one antibody, and that 4% tested positive to two. In both cases, the need to use insulin came much faster than for other samples. It is likely that a small minority of T2 are misdiagnosed T1s, for whom standard T2 treatment is almost certain to fail at great cost to their health.

If you have been recently been diagnosed as a T2 and have a good reason to believe you may be a T1, or if your T2 treatment is failing, consider getting tested to validate your diagnosis and treatment.

Some reasons you could have for believing you may be a T1 (from van Deutekom, see below):

  • “the presence of other autoimmune disorders”
  • “[absence] of markers of the metabolic syndrome: […] high body mass index, hypertension and [high cholesterol or trigycerides]”
  • “[genetic factors] known to increase the risk of Type 1 diabetes”


Almeida, Mirella Hansen de, et al. “Residual C-peptide in patients with Type 1 diabetes and multiethnic backgrounds.” Clinics 68.1 (2013): 123-126.

Bingley, Polly J. “Clinical applications of diabetes antibody testing.” The Journal of Clinical Endocrinology & Metabolism 95.1 (2010): 25-33.

Borg, Henrik, Per Fernlund, and Göran Sundkvist. “Protein tyrosine phosphatase-like protein IA2-antibodies plus glutamic acid decarboxylase 65 antibodies (GADA) indicates autoimmunity as frequently as islet cell antibodies assay in children with recently diagnosed diabetes mellitus.” Clinical chemistry 43.12 (1997): 2358-2363.

Fourlanos, Spiros, et al. “A clinical screening tool identifies autoimmune diabetes in adults.” Diabetes care 29.5 (2006): 970-975., research link by @britt_j

Insel, Richard A., et al. “Staging presymptomatic type 1 diabetes: a scientific statement of JDRF, the Endocrine Society, and the American Diabetes Association.” Diabetes care 38.10 (2015): 1964-1974.
Nakamura, K., et al. “Trajectories of anti-islet autoantibodies before development of type 1 diabetes in interferon-treated hepatitis C patients. Case reports and a literature review.” Endocrine journal 57.11 (2010): 947.

Jones, A. G., and A. T. Hattersley. “The clinical utility of C‐peptide measurement in the care of patients with diabetes.” Diabetic Medicine 30.7 (2013): 803-817.

Orban, Tihamer, et al. “Pancreatic islet autoantibodies as predictors of type 1 diabetes in the Diabetes Prevention Trial–Type 1.” Diabetes care 32.12 (2009): 2269-2274.

Pietropaolo, Massimo, et al. “Combined analysis of GAD65 and ICA512 (IA-2) autoantibodies in organ and non-organ-specific autoimmune diseases confers high specificity for insulin-dependent diabetes mellitus.” Journal of autoimmunity 11.1 (1998): 1-10.

Scott-Morgan, L., et al. “Insulin autoantibodies (IAA) may cause false positive islet cell antibody (ICA) results in the indirect immunofluorescence test.” Diabetes research (Edinburgh, Scotland) 8.1 (1988): 45.

Schmidli, Robert S., et al. “Antibodies to the protein tyrosine phosphatases IAR and IA-2 are associated with progression to insulin-dependent diabetes (IDDM) in first-degree relatives at-risk for IDDM.” Autoimmunity 28.1 (1998): 15-23.

Steck, Andrea K., et al. “Predictors of progression from the appearance of islet autoantibodies to early childhood diabetes: The Environmental Determinants of Diabetes in the Young (TEDDY).” Diabetes care 38.5 (2015): 808-813.

Szepietowska, Barbara, et al. “Latent autoimmune diabetes in adults in a population-based cohort of Polish patients with newly diagnosed diabetes mellitus.” Archives of medical science: AMS 8.3 (2012): 491.

Törn, Carina, et al. “Glutamic acid decarboxylase antibodies (GADA) is the most important factor for prediction of insulin therapy within 3 years in young adult diabetic patients not classified as Type 1 diabetes on clinical grounds.” Diabetes/metabolism research and reviews 16.6 (2000): 442-447.

Towns, Roberto, and Massimo Pietropaolo. “GAD65 autoantibodies and its role as biomarker of Type 1 diabetes and Latent Autoimmune Diabetes in Adults (LADA).” Drugs of the future 36.11 (2011): 847.

van Deutekom, A. W., R. J. Heine, and S. Simsek. “The islet autoantibody titres: their clinical relevance in latent autoimmune diabetes in adults (LADA) and the classification of diabetes mellitus.” Diabetic medicine 25.2 (2008): 117-125.[PDF]

Wallace, Tara M., Jonathan C. Levy, and David R. Matthews. “Use and abuse of HOMA modeling.” Diabetes care 27.6 (2004): 1487-1495.

Winter, William E., and David Pittman. “The clinical application of islet autoantibody testing for the diagnosis of autoimmune diabetes.” MLO. Med Lab Obs 45 (2013): 16-20.

Ziegler, Anette G., et al. “Seroconversion to multiple islet autoantibodies and risk of progression to diabetes in children.” Jama 309.23 (2013): 2473-2479.

Diapedia, Insulin Autoantibodies
Diapedia, GAD antibodies
Diabedia, Islet Cell Antibodies
Diapedia, IA-2 Antibodies
Diapedia, Zinc transporter 8 (ZnT8) autoantibodies
Lab Tests Online: Diabetes Antibody Tests
Quest Diagnostics: ZnT8 antibody tests

Diapedia, C-peptide in type 1 diabetes
MedlinePlus.gov: Insulin C-peptide test
Lab Tests Online: Insulin C-peptide Test

Medscape: differentiating between type 1 and type 2 diabetes
Melitta, Positive Autoantibody Tests Indicate Type 1 Autoimmune Diabetes

End of wiki ---------- comments start here


Wow! Thanks for sharing this great information. One thing is for sure…
T1, T2 or T37.5 Diabetes really sucks!!!


You know how it is – you accumulate info month after month, and, when you end up looking at it, you think – I should really write this up :slight_smile: I hardly ever do, so this is one exception…


Thank you for writing this up and posting along with references. My doctor has changed my medical record from Type 1 to Type 2, because (according to him) adults don’t get Type 1. He has also stated there is no way to differentiate Type 1 from Type 2. Maybe it doesn’t make a difference to him, but it does, and will continue to make a difference to me. I think I may give him a copy of this. I wonder sometimes, if he actually does continuing medical education seminars, and journal reading.


This is UNBELIEVABLE. On both counts!

They have to every year – but he clearly did not pick the sessions on diabetes… This has got to be close to malpractice. It makes me so angry to read your post, I wish I could just go talk to your doctor and tell him what I think.

You don’t have access to another doctor?

I believe this is another case that supports my feeling that most gp’s know squat about diabetes.
@literaturesnob18 I hope you have access to an endocrinologist in the near future.

1 Like

Thank you so much for providing information about peptides and such necessary information about diabetes and other diagonosis.

Research Peptides


Glad you are finding helpful information here. That is why we exist.