An autoimmune disease that destroys the cells making insulin. It shares a word with type 2 diabetes and almost nothing else — different age, different cause, different genetics, and a much stronger inherited component.
Type 1 diabetes happens when the immune system destroys the beta cells of the pancreas — the only cells in the body that make insulin. Without insulin, glucose cannot get from the blood into the cells that need it, so it accumulates in the blood while the body starves. It is treated by replacing the insulin, for life.
The two conditions share a name because they share a symptom: too much glucose in the blood. Almost everything underneath is different.
The distinction matters practically. Nothing about diet or exercise causes type 1 or prevents it, and a person with type 1 cannot stop needing insulin. Our type 2 page covers the other one, and the genetics on the two pages barely overlap.
Type 1 diabetes has one of the strongest inherited components of any common disease, and the numbers are worth stating precisely. Genetic factors account for roughly half of the total risk. Of that genetic half, about half again comes from a single region: HLA, the cluster of genes that tells the immune system what belongs to the body and what does not.
That is a very unusual shape. In most common diseases the genetic contribution is spread thinly across thousands of positions with no single region dominating. Here one region does most of the work, which is why HLA typing has real predictive value in type 1 diabetes and almost none in type 2.
The rest of the genetic risk is spread across dozens of other regions, each contributing a little. A 2009 study combining a genome-wide scan with earlier work — 7,514 people with type 1 diabetes and 9,045 reference samples — found 41 distinct locations associated with the disease. Most of them sit in or near genes with immune roles, which is what you would expect for a disease the immune system causes.
None of these is in the HLA region, and none of them is why anyone develops type 1 diabetes. They are outside-HLA signals — part of the thinner spread of risk — and they are here because they were replicated properly.
Two come from the Wellcome Trust Case Control Consortium follow-up published in 2007, which took the strongest hits from the original scan and tested them again in 4,000 people with type 1 diabetes, 5,000 controls and 2,997 families. Surviving that is what separates a real association from a lucky one.
The third comes from the 2009 meta-analysis above. All three sit near genes involved in immune signalling, which is the pattern rather than the exception.
It does not tell you whether you will develop type 1 diabetes, and it is not used to decide anything clinically. Diagnosis is made from blood glucose, from autoantibodies against the beta cells, and from the clinical picture — not from these positions.
Where genetics does get used in type 1 diabetes, it is HLA typing, usually in research settings or in families with an affected child, and usually alongside autoantibody testing rather than instead of it. A person with the highest-risk HLA types still very probably does not develop the disease.
The signs of new type 1 diabetes in a child are specific and worth knowing, because it can develop quickly: drinking a great deal, passing urine often, losing weight without trying, and unusual tiredness. That combination is a same-day question for a doctor, not a genetics question. A blood glucose test answers it in minutes.
Type 1 diabetes is a clinical diagnosis supported by laboratory findings, and no part of it depends on the variants listed here.
Islet autoantibodies — against GAD, IA-2, insulin, or ZnT8 — are the marker that distinguishes autoimmune diabetes from other kinds, and they are often present for years before symptoms begin. That gap is the reason screening studies exist in families with an affected relative, and the reason staging language has entered the field: autoantibodies without high glucose, then glucose abnormalities without symptoms, then clinical disease.
Management is insulin replacement, and everything else follows from doing that well: matching doses to food and activity, monitoring glucose, and watching for the two acute failures in either direction — ketoacidosis when insulin is too low, and hypoglycaemia when it is too high. Technology has changed how this is done rather than what it is.
Nothing on this page is a reason to change any of it. If a genotype ever conflicts with a clinician's assessment, the clinician's assessment is the one that reflects the person.
What a 23andMe/AncestryDNA export or raw VCF can and can't tell you about Type 1 Diabetes comes down to these specific, well-studied positions — not a diagnosis.
No. That is type 2 you are thinking of, and the confusion causes real harm — people with type 1 are regularly told they brought it on themselves. Type 1 is an autoimmune disease. Nothing a child ate or did not do caused it.
Partly, and more strongly than most common diseases: genetic factors account for roughly half of the total risk, and about half of that comes from the HLA region alone. But most people who develop it have no affected relative, and most relatives of an affected person never develop it.
No. These are outside-HLA signals, each contributing a small amount to a risk that is mostly decided elsewhere. They are not diagnostic, they are not used clinically, and no combination of them predicts the disease in one person.
HLA typing has genuine predictive value and is used in research and in families with an affected child, generally alongside autoantibody testing. It is not a diagnosis. Diagnosis is made from glucose, autoantibodies and the clinical picture.
Free to reuse. This page's text is original writing from freely-available research, licensed CC BY 4.0 — reuse it, including commercially, with attribution to MyGeneLog. It's general research-derived information, not medical advice or a diagnosis — see Terms of Use.