Objective
The aim of this study is to find out whether new blood tests can help diagnose type 1 diabetes in adults at the time of diagnosis. This could lead to more accurate diagnoses and ensure people receive the most appropriate treatment from the outset.
We will build on earlier research that identified eight blood markers which can reliably distinguish type 1 diabetes from other types in people who have had diabetes for many years. Early results suggest these markers also perform well at the time of diagnosis.
In this study, we will validate and expand our early findings in a large group of people with newly diagnosed adult-onset diabetes. We will assess how well these tests improve diagnosis at the time diabetes is first recognised. Importantly, we will also examine how these new markers perform alongside existing tests, such as islet autoantibodies, to understand their added value in current clinical practice.
Finally, we will investigate how these new markers relate to other blood tests to better understand their role in the onset of type 1 diabetes.
Background Rationale
Although type 1 diabetes is often thought of as a condition that begins in childhood, most people who develop type 1 diabetes are adults. Diagnosing type 1 diabetes in adults is challenging because it often shares characteristics with the much more common type 2 diabetes. For example, features typically associated with type 2 diabetes such as older age and higher body weight are also common in adults developing type 1 diabetes. Even adults with lower body weight are statistically more likely to have type 2 than type 1 diabetes, simply because type 2 is so prevalent. As a result, it is often difficult for clinicians to determine the correct diagnosis at the time of presentation. This challenge is becoming even greater as obesity rates continue to rise worldwide.
Due to this diagnostic uncertainty, around one in three adults who develop type 1 diabetes are initially misdiagnosed as having type 2 diabetes. They are started on glucose-lowering tablets that are ineffective for type 1 diabetes, told to lose weight, and are not prescribed insulin. Without insulin, they may become seriously unwell. Even when insulin is eventually introduced, they often miss out on vital support, including structured education and access to technologies such as continuous glucose monitoring, multiple daily injections, or insulin pumps. At the same time, around one in six adults diagnosed with type 1 diabetes actually have another form of diabetes. These individuals may be treated with insulin unnecessarily for life and miss out on alternative treatments that are more appropriate for their condition—for example, newer medications that reduce the risk of heart and kidney complications in people with type 2 diabetes.
At present, the main blood test used to support a diagnosis of type 1 diabetes in those recently diagnosed with diabetes is the islet autoantibody test. However, this test is far from perfect. Around 10–15% of adults with new-onset type 1 diabetes have negative islet autoantibodies, while a small proportion of people with type 2 diabetes can test positive. Another test, C-peptide, which measures the body’s ability to make insulin, is more useful several years after diagnosis, but has limited value at the time of diagnosis. There is therefore a clear need for new diagnostic tools to support accurate diagnosis of type 1 diabetes in adults, particularly at the time the disease is first identified.
We used data on nearly 3000 blood markers from UK Biobank, a large population-based study of thousands of people with long-standing diabetes, and performed detailed analysis to identify new blood test for type 1 diabetes. We found eight new blood markers that help distinguish type 1 diabetes from other forms. When used together, these markers showed high diagnostic accuracy, even in individuals whose diabetes type was difficult to determine. In a smaller study of 88 adults newly diagnosed with diabetes, we found that these same markers also performed well at the time of diagnosis and appeared to complement the existing tests.
These findings raise the exciting possibility that one or more of these blood markers could be used to improve diagnosis at diabetes onset, helping to ensure people receive the right diagnosis and the most appropriate treatment. However, before these markers can be implemented in clinical practice, they need to be rigorously evaluated in a larger group of people with newly diagnosed adult-onset diabetes. We need to determine whether they improve diagnosis beyond current tools, and how they can best be used alongside existing tests in real-world clinical settings.
Description of Project
The purpose of this project is to find out whether new blood tests can help us accurately diagnose type 1 diabetes in adults, so that more people receive the correct diagnosis and the most effective treatment from the outset.
Diagnosing type 1 diabetes in adults is often challenging. Its features can overlap with type 2 diabetes, which is much more common. As a result, around one in three adults with new-onset type 1 diabetes are initially misdiagnosed as having type 2. Without insulin treatment, they may become seriously unwell. Even once they start insulin, many miss out on the education, technology, and support specifically designed for people with type 1 diabetes. At the same time, about one in six adults diagnosed with type 1 diabetes have another type of diabetes. This misdiagnosis can lead to years of unnecessary insulin use and a missed opportunity to benefit from other effective treatments.
Two blood tests, C-peptide and islet autoantibodies, are currently used to help identify type 1 diabetes. However, these tests can give overlapping results in people with newly diagnosed diabetes, and do not always provide a clear answer. There is therefore an urgent need for better tools to improve diagnostic accuracy in adults.
In previous work using the UK Biobank study, which includes data from thousands of people living with diabetes for many years, we identified eight blood markers that strongly distinguish type 1 from other types of diabetes. When used together, these markers have high accuracy, even in people whose diabetes type is difficult to classify. In a smaller study of 88 adults newly diagnosed with diabetes, we found that these same markers also performed well at the time of diagnosis. This raises the exciting possibility that these tests could support earlier, more accurate diagnosis.
We now need to fully evaluate these tests in large number of people at the time of diagnosis, to understand whether they provide additional benefit over current diagnostic tools, and how they could be used in real-world clinical settings.
In this project, we will analyse already collected blood samples and clinical data from 800 adults who were newly diagnosed with either type 1 or type 2 diabetes. We will assess how well each of the eight markers works, both individually and in combination, and whether they improve diagnostic accuracy beyond existing information such as age, body weight, and current blood tests like islet autoantibodies. We will also explore the biological role of these markers in type 1 diabetes and test other blood markers previously linked to diabetes, to see if they could further improve diagnosis.
The outcome of this research will be a clear understanding of whether these new markers can be used to improve the diagnosis of type 1 diabetes in adults at the time of diagnosis. If successful, these markers could be developed into simple, low-cost tests for use in healthcare laboratories helping ensure people are correctly diagnosed and receive the most appropriate care for their type of diabetes.
Anticipated Outcome
This research will show whether new protein tests can help clinicians more accurately diagnose type 1 diabetes in adults who are newly diagnosed with diabetes. We aim to identify which blood markers are most useful for improving diagnosis, how they perform compared to current tests, and in which situations they add the most value.
We will also gain a better understanding of the biological role of these markers and how they relate to both type 1 and type 2 diabetes. This could provide new insights into how type 1 diabetes develops.
Our findings will be critical in deciding whether these new tests can be used in real-life healthcare settings to improve how we diagnose and treat adults developing diabetes. If successful, the next step will be to develop inexpensive tests for these proteins that can be used in laboratories worldwide. This will expand the benefit globally and help ensure that people receive the correct diagnosis from the start and with it, the right treatment, education, and support for their type of diabetes.
Relevance to T1D
This research has strong potential to improve the diagnosis and ultimately the treatment of adults developing type 1 diabetes. It may also benefit children in future by improving how we recognise and manage different types of diabetes at diagnosis.
At present, there is a major unmet need in the diagnosis of type 1 diabetes in adults. Many adults are misdiagnosed as having type 2 diabetes, leading to delays in insulin treatment, inappropriate use of tablets, and missed opportunities for access to education and diabetes technologies. Our study directly addresses this gap. By testing whether new blood markers can improve diagnosis at the time diabetes develops, we aim to support more accurate and timely treatment decisions for people who might otherwise be misdiagnosed.
If our study shows that these new blood markers are accurate and add value beyond existing tools, they could be developed into inexpensive tests for healthcare laboratories. Blood protein markers are generally simple to test and much cheaper than many other technologies, such as genetic testing. This makes them ideal for widespread use. This would help ensure that people developing diabetes receive the correct diagnosis and the most appropriate treatment from the start. It could also reduce the risk of serious illness caused by delayed insulin treatment in type 1 diabetes.
Although this research focuses on adults, the tests may also prove useful in other settings. For example, they could help diagnose difficult or rare forms of diabetes in children, support identification of genetic types of diabetes, or even be used to detect people at risk of developing type 1 diabetes before symptoms begin opening the door to early monitoring or future prevention.
Our work also responds to the broader lack of research into type 1 diabetes developing in adults. While the condition is well studied in children, adult-onset type 1 diabetes is poorly understood, and very few studies follow people from the point of diagnosis. By analysing blood markers (using a method called proteomics) in people tracked from diagnosis for up to nine years, we can explore why type 1 diabetes develops in adults and why some people lose insulin quickly while others continue producing it for much longer.
Together, this research addresses an urgent clinical need and offers the potential for real-world improvements in diagnosis and care for people with type 1 diabetes.