Adults with diabetes could benefit from better treatment if the condition was categorized into five types, rather than just two. This is the conclusion of a new study published in The Lancet Diabetes & Endocrinology.

The research was led by Professor Leif Groop, of the Lund University Diabetes Centre in Sweden and the Institute for Molecular Medicine Finland in Helsinki.

In the United States alone, around 30.3 million people are living with diabetes.

Excluding gestational diabetes—diabetes that develops during pregnancy—there are two main types: type 1 and type 2.

In type 1 diabetes, the beta cells of the pancreas—which produce insulin, the hormone that regulates blood sugar levels—are mistakingly attacked and destroyed by the immune system.

Type 2 diabetes is the most common form, accounting for around 90%–95% of all cases. This occurs when the body's cells stop responding to insulin, or the beta cells are unable to produce sufficient amounts of the hormone.

In both forms of the condition, blood sugar levels can become too high—a condition known as hyperglycemia. Unless controlled, this can lead to a number of complications, including kidney disease, cardiovascular disease, and nerve damage.

The heterogeneity of diabetes

A diabetes diagnosis is normally made using the fasting plasma glucose (FPG) test or the A1C test. The FPG test assesses a person's blood glucose level at a single time point, while the A1C test measures average blood glucose levels over the previous 3 months.

When it comes to determining which type of diabetes a person has, health-care professionals might look for diabetes-related autoantibodies in the blood. These are proteins produced by the immune system that can attack the body's own cells.

The presence of such autoantibodies is an indicator of type 1 diabetes. If a person does not have these autoantibodies, they are considered to have type 2 diabetes.

But, as Professor Groop and colleagues note, the classification guidelines for diabetes have not been updated for 20 years—despite increasing evidence that diabetes has high heterogeneity.

"Diabetes is a group of chronic metabolic disorders," says Dr. Rob Sladek, of the McGill University and Génome Québec Innovation Centre in Canada, in an editorial linked to the study, "that share the common feature of hyperglycemia, meaning that, in principle, diabetes can be diagnosed via measurement of a single blood component."

"However, elevations in blood glucose can be caused by a number of genetic and acquired factors that reduce the circulating concentrations of insulin or decrease its effectiveness, leading to heterogeneity in the clinical presentation and progression of the disease."

Professor Groop and his team say that a "refined classification" of diabetes based on its heterogeneity could help health-care professionals better predict which individuals are most likely to develop complications and allow a more personalized approach to treatment.

In their study, the researchers propose that diabetes should no longer be categorized as two types. Instead, they say that the condition should be classified into five distinct types.

The five 'clusters' of diabetes

The researchers came to their proposal by analyzing the data of four study cohorts. These included a total of 14,775 adults from Sweden and Finland, all of whom had been newly diagnosed with diabetes.

As part of the analysis, the scientists looked at six measures in each subject that each represent different features of diabetes.

These measures were: BMI; age at diabetes diagnosis; hemoglobin A1C (HbA1C), a measure of long-term blood sugar control; beta cell functioning; insulin resistance; and the presence of diabetes-related autoantibodies.

As well as conducting genetic analyses of the participants, the researchers also compared their disease progression, complications, and treatment.

The study revealed five distinct forms of diabetes, three of which were severe and two that were mild. The team categorized these as follows:

• Cluster 1: severe autoimmune diabetes (currently known as type 1 diabetes), characterized by insulin deficiency and the presence of autoantibodies. This was identified in 6%–15% percent of subjects.
• Cluster 2: severe insulin-deficient diabetes, characterized by younger age, insulin deficiency, and poor metabolic control, but no autoantibodies. This was identified in 9%–20% percent of subjects.
• Cluster 3: severe insulin-resistant diabetes, characterized by severe insulin resistance and a significantly higher risk of kidney disease. This was identified in 11%–17% of subjects.
• Cluster 4: mild obesity-related diabetes, most common in obese individuals. This affected 18%–23% of subjects.
• Cluster 5: mild age-related diabetes, most common in elderly individuals. This was the most common form, affecting 39%–47% of subjects.

The researchers note that each of these five types "were also genetically distinct," meaning that there were no genetic mutations that were shared across all five clusters.

A 'step toward precision medicine'

When the researchers assessed the treatment received by adults in each of the five clusters, they noticed that some were being treated inappropriately.

As an example, the team points out that just 42% of patients in cluster 1 and 29% of patients in cluster 2 received insulin therapy from the point of disease onset.

They say that this indicates that the current classifications of diabetes fail to target the underlying features of the disease.

As such, Professor Groop and colleagues propose that diabetes should be categorized into five distinct types.

While further research is required to refine these five clusters—by using biomarkers and genetic risk scores, for example—the team believes that this study is a great stride toward tailored treatments for diabetes.

"Existing treatment guidelines," concludes Professor Groop, "are limited by the fact they respond to poor metabolic control when it has developed, but do not have the means to predict which patients will need intensified treatment.

"This study moves us towards a more clinically useful diagnosis, and represents an important step towards precision medicine in diabetes."

Professor Leif Groop