Diabetes 'test' needs more work

Behind the Headlines

Monday September 20 2010

The study looked at the decline of cells in the pancreas

A simple blood test is being developed that could save thousands of lives every year by predicting a person’s risk of getting heart disease and diabetes, reported the Daily Mail. According to the newspaper, the test “would revolutionise the diagnosis and treatment of two of Britain’s biggest killers” and may be available in just five years.

The study behind this and other news reports found that the levels and functioning of five particular chemicals in the blood may act as a marker for the onset of type 2 diabetes and peripheral arterial disease. However, this is early research. The “test”, which involved using levels of these chemicals as a predictor of disease, correctly identified only 10 out of 19 people (52%) who went on to develop diabetes, and needs to be refined and repeated in a larger sample of people.

Diabetes and peripheral arterial disease affect a large number people in the UK and cost the NHS a lot of money to manage. The findings of this well-conducted study are promising, but the news is optimistic given the early stage of this research.


Where did the story come from?

The study was carried out by researchers from King’s College London, the University of Cambridge, University Hospital of Verona and Bruneck Hospital in Italy. It was funded by the British Heart Foundation and published in the peer-reviewed medical journal Circulation Research.

The news reports optimistically heralded a test that can predict type 2 diabetes ten years before its onset. The study behind this report is important, but it is too soon to make such claims. A lot of work needs to be done before any such test may be useable.

The study did not identify a test, as such. Instead, the researchers noted that assessing the levels of five particular chemicals in the blood allowed some capacity to predict new-onset diabetes and new-onset peripheral arterial disease. Some newspapers focused on the application of the “test” to identify diabetes, while others discussed heart disease. Any potential application for either or both conditions is clearly important.


What kind of research was this?

Certain chemicals called microRNAs circulate in the body and are responsible for regulating the way proteins and other substances are used in the body. Some microRNAs circulate in the blood and previous research has suggested that these may be markers of disease. Other studies have identified specific tumour microRNAs in patients with cancer, for example. In this study, the researchers set out to determine whether they could identify a unique set of microRNAs in people with type 2 diabetes.

The study was a nested case-control study in a prospective cohort of adult inhabitants, aged 40 to 79 years old, of a town called Bruneck in Italy.


What did the research involve?

A blood sample was collected from 822 participants in 1995. The study examined the microRNA in blood samples from this group, both at the start of the study and 10 years later in 2005.

The researchers were particularly interested in the differences in microRNA between people with diabetes at the start of the cohort study, those who developed diabetes during the course of the study, and a group of people without diabetes (the controls). Not only were they interested in the concentration of these chemicals, but also in their structure and how they clustered together. Eighty participants had a diagnosis of type 2 diabetes in 1995 at the start of the study, while 19 participants, who were originally free of the disease, developed diabetes between 1995 and 2005. The controls were matched to these groups in terms of age and gender.

The profiling and analysis techniques used to investigate the structure of chemical compounds in the blood are complex. Essentially, the researchers investigated whether there was an association between the levels and formation of microRNA between people with existing diabetes, those who developed diabetes and the group without the disease.

The researchers also assessed whether some lifestyle and other factors might be associated with levels of microRNA, including family history of diabetes, alcohol consumption and smoking, BMI, social status and blood levels of the inflammatory protein CRP. If any of these were associated, the researchers would need to adjust for them when analysing the link between microRNA and disease status.

The researchers also investigated whether microRNAs could predict new-onset peripheral arterial disease (which affects the arteries of the limbs, usually the legs). They did this in 785 subjects who did not have this disease at the start of the study and assessed the link between microRNAs and low ankle brachial index. This is a ratio that compares blood pressure in the ankle with that in the arm (a low value would suggest the presence of arterial disease).


What were the basic results?

Initially, the researchers identified 30 microRNAs that differed in people with diabetes. After further analysis, 13 of these microRNAs were considered to be structurally unique. In people who had diabetes at the start of the study, blood levels of 12 of these microRNAs were lower than in healthy subjects, while the level of one of them was higher. After further adjustments, four of these microRNAs remained highly significant, in particular one called miR-126, which is common in endothelial cells (which line blood vessels).

In the 19 people who went on to develop diabetes during the study, blood levels of four of these microRNAs were lower at the beginning of the study while one was higher (the same as in the population who already had diabetes). The researchers used tests to establish whether these five microRNAs could be used to distinguish people with diabetes or who would get diabetes from those without the disease. They found that 92% of the control group and 70% of people who already had diabetes were correctly identified. This test also identified as diabetic 10 out of the 19 people who went on to develop diabetes during the study.

The one microRNA most consistently associated with diabetes was also linked to the development of peripheral arterial disease.


How did the researchers interpret the results?

The study found a set of circulating microRNAs, which the researchers dubbed a “plasma miRNA signature”, with altered activity and concentration in people with type 2 diabetes. They say that this research may ultimately lead to “novel biomarkers” that could be used to estimate risk, and that it may one day be used in microRNA-based therapies for vascular complications associated with disease.



This research has identified a set of microRNAs that are uniquely regulated in people with diabetes, and which could be used to identify the disease or to predict the development of it. Such screening tools are potentially valuable if they can forewarn of the development of a disease in an individual, although more work will need to be done to determine whether this particular approach can be applied to the general population.

Only 19 individuals in this sample who were disease-free at the start went on to develop diabetes during the course of the study. The screening test developed in this study (i.e. using levels of five microRNAs) identified 52% of these people as having diabetes. Nineteen is a small number of people from which to draw strong conclusions, and it will be important to replicate these findings in larger numbers of people.

These findings are promising, although the full potential of this discovery will only be realised with further refinement of the methods and replication of these results in relevant populations, such as the general population of the UK.

Analysis by Bazian

Edited by NHS Choices

Links to the headlines

Blood test to predict diabetes 10 years earlier. The Daily Telegraph, September 19 2010

£2 heart disease test: Blood screening to revolutionise cardiac and diabetes treatment. Daily Mail, September 19 2010

Links to the science

Zampetaki A, Kiechl S, Drozdov I et al. Plasma MicroRNA Profiling Reveals Loss of Endothelial MiR-126 and Other MicroRNAs in Type 2 Diabetes. Circulation Research, September 2010, 107: 810 - 817


How helpful is this page?

Average rating

Based on 5 ratings

All ratings

3  ratings
1  ratings
1  ratings
0  ratings
0  ratings

Add your rating