A refined classification system creating five phenotypic ‘clusters’ for type 2 diabetes heralds a major step towards precision medicine in diabetes, researchers say in the Lancet.
A Swedish study identified five distinct sub-groups of diabetes in adults on the basis of variables including glutamic acid decarboxylase antibodies (GADA), age, BMI, HbA1c and estimates of β-cell function at the time of diagnosis. The model was then applied to three independent cohorts with patient outcomes data.
Cluster 1 or severe auto-immune diabetes (SAID) represented about 6% of the almost 9,000 patients and was characterised by insulin deficiency, the presence of GADA, relatively low BMI and early onset disease.
The similar but more common (17.5%) sub-group, labeled severe insulin-deficient diabetes (SIDD), was antibody negative and associated with a high rate of diabetic retinopathy.
Severe insulin-resistant diabetes (SIRD) represented about 15% of patients and was associated with a high BMI, the highest prevalence of non-alcoholic fatty liver disease and risk for chronic kidney disease despite ‘reasonably low’ HbA1c.
A fourth cluster called mild obesity-related diabetes (MOD) in about 22% of patients was characterised by obesity but not insulin resistance while a large fifth cluster (39%) nominated as mild age-related diabetes (MARD), showed only modest metabolic derangements.
“While SAID overlapped with type 1 diabetes and LADA, SIDD and SIRD represent two new, severe forms of diabetes previously masked within type 2 diabetes,” the researchers said.
It would be reasonable to target individuals in these clusters with intensified treatment to prevent diabetic complications.”
Professor Sophia Zoungas from Monash University, president of the Australian Diabetes Society (ADS) and a director of Diabetes Australia, told the limbic the research was ‘well conducted and really interesting’ but may not yet be ready for prime-time application.
“We are certainly moving more towards a greater number of categories. Yes, we traditionally had type 1 and type 2 diabetes but now we also have latent autoimmune diabetes (LADA) as well as MODY and are recognising secondary forms of diabetes such as after pancreatectomy or pancreatitis or diabetes associated with cystic fibrosis or after transplantation. So we have been expanding the spectrum of classification of diabetes and this is a further step.”
She said the hypothesis generating research certainly suggested a rethink in the way diabetes was diagnosed, categorised and managed.
“A Swedish registry is likely to represent a much more homogenous population so how this applies to an ethnically diverse population such as we have in Australia is yet to be determined,” said Professor Zoungas.
“I don’t think it is something we will be able to implement immediately because some of the testing is not part of routine care and we have to understand how it applies to our population, how management decisions need to alter and whether it will actually improve health outcomes.”
“Also what this study doesn’t tell us is if an individual can change categories over time,” she said.
In the interim Professor Zoungas advised clinicians to have a high index of suspicion for insulin deficiency or severe insulin resistance, particularly among younger people with poor metabolic control and at high risk of long term diabetes complications.