Update on diabetes

Insulin has been around for almost a century and there are now numerous variations with different durations of action. Several new classes of antidiabetic drugs have also been developed, while research continues to seek new ways to understand and treat this common disease

Despite the therapeutic diversity, despite on-going innovations, and despite ranking high on the clinical, political and public heath agenda, diabetes still accounts for about 26,000 additional deaths a year, say Diabetes UK.

People with diabetes are about 32 per cent more likely to die prematurely than their peers in the general population, the charity says. Fortunately for the millions of people with diabetes in the UK, recent research is changing the way we view this common disease and raises the prospect of improved diagnosis and treatment.

Despite the therapeutic diversity, despite on-going innovations, and despite ranking high on the clinical, political and public heath agenda, diabetes still accounts for about 26,000 additional deaths a year, say Diabetes UK.

People with diabetes are about 32 per cent more likely to die prematurely than their peers in the general population, the charity says. Fortunately for the millions of people with diabetes in the UK, recent research is changing the way we view this common disease and raises the prospect of improved diagnosis and treatment.

Range of subtypes

The progressive loss of beta-cell mass, function, or both, that underlies hyperglycaemia and diabetes has numerous potential causes. In addition to autoimmunity and obesity, for example, genetic defects of beta-cell function and insulin action (e.g. mutations in the receptor) can cause diabetes mellitus. Several pancreatic diseases, endocrinopathies – including Cushing’s syndrome (excess secretion of hormones that antagonise insulin’s action) – and certain drugs (e.g. glucocorticoids, atypical antipsychotics, beta-blockers and thiazides) can all cause or contribute to diabetes mellitus.1

Clinically differentiating the various types of diabetes often proves difficult. For example, fewer than 10 per cent of people with diabetes have latent autoimmune diabetes in adults (LADA), which is caused by auto-antibodies against glutamic acid decarboxylase.

This enzyme synthesises gamma-aminobutyric acid (GABA).2 In addition to being an inhibitory neurotransmitter in the brain, GABA regulates pancreatic beta-cells.3 At diagnosis, LADA is clinically indistinguishable from type 2 diabetes but over time, however, it increasingly resembles type 1 diabetes.2

Swedish researchers who analysed 8,980 patients with newly diagnosed diabetes found that 1.5 per cent had type 1 diabetes, 5.3 per cent had LADA and 1.2 per cent had secondary diabetes arising from pancreatic disease. Data was missing on 3.8 per cent. The remaining 88.3 per cent of patients had type 2 diabetes.2

The researchers considered the way in which six variables clustered: age at diagnosis; body mass index (BMI); HbA1c; homoeostatic model assessment (HOMA2), which estimates beta-cell function and insulin resistance; and whether or not the patient expressed glutamate decarboxylase auto-antibodies.

The analysis revealed five clusters (see table 1). For instance, ‘severe autoimmune diabetes’ (SAID) overlapped with type 1 diabetes and LADA. Another two clusters – ‘severe insulin-deficient diabetes’ (SIDD) and ‘severe insulin resistant diabetes’ (SIRD) – “represent two new, severe forms of diabetes” previously considered to be type 2 diabetes.2

The clusters differed in progression and the risk of diabetic complications. For example, people with SAID or SIDD had substantially higher HbA1c at diagnosis (80.0 and 101.9mmol/L respectively) and throughout follow-up than the other clusters. People with SAID or SIDD also showed higher rates of ketoacidosis at diagnosis (31 and 25 per cent respectively) than the other clusters (less than 5 per cent).

SIRD and ‘mild obesity-related diabetes’ (MOD) showed the highest rates of non-alcoholic fatty liver disease (24.1 and 21.9 per cent respectively).2

Early signs of diabetic retinopathy were more common in those with SIDD than in the other clusters with, for instance, the risk 60 per cent higher than in people with ‘mild agerelated diabetes’ (MARD). After adjusting for age and sex, people with SIRD were more than twice as likely to develop chronic kidney disease (CKD) than those with MARD (hazard ratio [HR] 2.41), while 1.9 per cent of SIRD patients developed end-stage renal disease: about five times more than MARD patients (HR 4.98).

The increased risk of CKD in SIRD patients reinforces the link between insulin resistance and renal disease, while the increased incidence of CKD in people with relatively low HbA1c levels may suggest that “glucose-lowering therapy is not the optimum way of preventing this complication”.2

Treatment patterns

Treatment patterns also differed. For example, 41.9 and 29.1 per cent of patients with SAID and SIDD respectively received insulin, compared with less than 4 per cent of patients in the other clusters.

The proportion of patients taking metformin was highest in patients with SIDD (77.8 per cent) and lowest in those with SAID (44.7 per cent). Metformin use was just 48.8 per cent in people with SIRD, which, the authors suggest, is the subtype most likely to benefit. They suggest that the “traditional classification is unable to tailor treatment to the underlying pathogenic defects”.2

Finally, the researchers analysed genetic loci (regions of chromosomes) that previous studies had associated with diabetes and related traits. No genetic variant was associated with all five clusters but several cluster-specific associations emerged.2

Despite the strong associations, the authors “cannot at this stage claim that the new clusters represent different aetiologies of diabetes, nor that this clustering is the optimal classification of diabetes subtypes”. For example, studies need to determine whether patients can move between clusters and hone the model by exploring additional autoantibodies and complications, such as hypertension and dyslipidaemia.2

In the future, such models may include other innovations in diagnosis. For example, microRNAs are small sequences of RNA that regulate the expression (turn on or off) of certain genes. Some microRNAs seem to modulate glucose metabolism, for instance, while others seem to regulate insulin signalling.

One study followed 462 people with coronary heart disease, but not diabetes, for a median of 60 months. During this time, 107 patients developed type 2 diabetes.

Changes in circulating levels of six microRNAs in the plasma were associated with an 11.27-fold increase in the likelihood of developing type 2 diabetes.4 The study raises the prospect of using microRNAs to diagnose type 2 diabetes early.

EASD in brief

New research presented at this year’s annual meeting of the European Association for the Study of Diabetes (October 1-5) revealed that:

  • Consumption of low-calorie sweeteners worsen blood sugar control in healthy subjects by disrupting the regulation of glucose uptake and disposal, as well as from changes in the balance of gut bacteria. This highlights the clinical relevance of dietary low-calorie sweetener patterns to overall blood sugar control, the authors say
  • A Japanese study showed that early signs of type 2 diabetes can be identified more than 20 years before diagnosis. The study tracked over 27,000 non-diabetic adults (average age 49 years) between 2005-16 and found that increased fasting glucose, higher BMI and impaired insulin sensitivity were detectable up to 10 years before the diagnosis of diabetes and pre-diabetes.

DiRECTING weight loss

Diagnosing type 2 diabetes early increases the likelihood of being able to make lifestyle changes that reverse the disease. For instance, the recent DiRECT study, sponsored by Diabetes UK, shows that a weight management programme delivered in primary care can maintain remission after a year in almost half of people with type 2 diabetes.

The programme consisted of three phases, beginning with total diet replacement using a formula diet providing 825- 853kcal a day for three months, which could be extended up to five months if the participant wished. During the next two to eight weeks, food was gradually reintroduced. Following this, patients made monthly visits to maintain weight loss. Those patients taking part in the weight loss programme discontinued oral antidiabetic drugs and antihypertensives, which were reintroduced based on national guidelines.5

DiRECT enrolled 306 type 2 diabetes patients with BMIs of 27-45kg/m2, aged 20-65 years, from 49 general practices in Scotland and the North-East. Mean bodyweight fell by 10.0kg in those taking part in the programme and by 1.0kg among controls. Mean HbA1c fell by 0.9 per cent and rose by 0.1 per cent respectively.

At 12 months, 24 per cent of those who took part in the programme lost at least 15kg compared with none of those in the control group.5 Forty-six per cent of those in the programme showed remission of their type 2 diabetes, compared with just 4 per cent of the controls, so the programme increased the likelihood of remission almost 20-fold (odds ratio 19.7).

Across both groups, none of those who gained weight and 7 per cent of those who maintained a loss of 0-5kg entered remission. This proportion increased to 34 per cent with 5-10kg loss, 57 per cent with 10-15kg loss and 86 per cent of those who lost at least 15kg. Quality of life also improved significantly in those who took part in the programme.5

“Putting type 2 diabetes into remission as early as possible after diagnosis could have extraordinary benefits, both for the individual and the NHS. DiRECT is telling us it could be possible for as many as half of patients to achieve this in routine primary care – and without drugs,” said Professor Mike Lean from the University of Glasgow, lead researcher of the DiRECT trial.


Not everyone can change their lifestyle, however, and weight loss won’t reverse type 1 diabetes. We’ve long known that type 1 diabetes is an autoimmune disease that probably arises from several aetiologies. Broadly, however, a person at risk of type 1 diabetes may have certain genes that predispose the pancreas to autoimmune attack.

Type 1 diabetes begins when a trigger – such as an infection or change in the microbiome – provokes an immune response mediated by T-cells against beta-cells. A strong immune response starts destroying beta-cells and can trigger changes to the antigens displayed by the cells, which amplifies the attack. If the patient’s control of autoimmune responses is ineffective, this response can become chronic and the person develops type 1 diabetes.6,7

Interrupting this autoimmune pathway seems to be a logical approach to treatment. For example, immunotherapy could, in theory at least, help prevent the progression of type 1 diabetes in people who have yet to develop symptoms but who show auto-antibodies.

This prodromal stage can last several years. When type 1 diabetes emerges, immunotherapy could suppress the autoimmune response, protect remaining beta-cells and, ideally, reverse clinical diabetes. In people with advanced type 1 diabetes, immunotherapy could protect the remaining beta-cells. Even a small number of functioning beta-cells seems to help control glycaemia and limit morbidity.7

Several immunotherapies appeared promising in animal models. Many of these target T-cells, which seem to be responsible for the beta-cell destruction and dysfunction in most type 1 diabetes patients. T-cells may, however, be less important in adult-onset type 1 diabetes than they are in younger patients.7

Until recently, no immunotherapy showed clinical benefits in patients but a recent study has assessed ‘peptide immunotherapy’, which is under investigation for other autoimmune diseases including multiple sclerosis and coeliac disease. Essentially, peptide immunotherapy uses short fragments of the autoantigens that provoke the response. This seems to restore immune tolerance in some patients.8

The study assessed ‘human leukocyte antigen-DR4 (DRB1*0401)-restricted immunodominant proinsulin peptide’ injected intradermally every two or four weeks for six months in 27 adults diagnosed with type 1 diabetes in the previous 100 days. Peptide immunotherapy was well tolerated.8

People who received placebo showed a significant decline in stimulated C-peptide levels (which measure insulin reserve) at three, six, nine and 12 months compared with baseline. But levels did not significantly change in those who received immunotherapy every four weeks or at three, six and nine months in those who received the peptide every two weeks.

Daily insulin use in the placebo group increased by 50 per cent over 12 months, but remained unchanged in those treated with peptide immunotherapy. These and other metabolic changes suggest that this form of “peptide immunotherapy is safe, does not accelerate decline in beta-cell function, and is associated with antigen-specific and non-specific immune modulation”.

Fitting tribute

In four years’ time, it will be a century since insulin first effectively treated diabetes in humans. Perhaps by then, we will have better characterised the various subtypes of diabetes and have new diagnostic tests. We should have more results from DiRECT, not to mention numerous other studies and, perhaps, immunotherapy will have confirmed its promise in type 1 diabetes. Such advances would be a fitting tribute to this landmark discovery.

Surprising diabetes complications

Recent studies highlight the wide range of complications linked to diabetes. For example, a meta-analysis of 21 studies that included 425 controls and 365 people with type 1 diabetes found that hearing loss was 7.7 times commoner in people with type 1 diabetes.9

Another recent meta-analysis confirmed that root canal treatment is commoner in people with diabetes. The authors included data from 50,301 root canal treatments in controls and 4,635 from people with diabetes included in three studies. People with diabetes were 2.44 times more likely to need extraction of root-filled teeth.10

1. Lecture Notes in Endocrinology and Diabetes Wiley-Blackwell
2. The Lancet Diabetes & Endocrinology 2018; 6:361-369
3. Diabetes, Metabolic Syndrome and Obesity: Targets and Therapy 2015; 8:79-87
4. Molecular Therapy – Nucleic Acids 2018; 12:146-157
5. Lancet 2018; 391:541-551
6. Trends in Immun 2013; 34:583-591
7. Frontiers in Immun 2018; 9:DOI: 10.3389/fimmu.2018.01891
8. Science Translational Medicine 2017; 9:DOI: 10.1126/scitranslmed.aaf7779
9. International Journal of Pediatric Otorhinolaryngology 2018; 113:38-45
10. International Endodontic Journal DOI:10.1111/iej.13011

Comments (0)

This website is for healthcare professionals, people who work in pharmacy and pharmacy students. By clicking into any content, you confirm this describes you and that you agree to Pharmacy Network News's Terms of Use and Privacy Policy.

We use essential, performance, functional and advertising cookies to give you a better web experience. Find out how to manage these cookies here. We also use Interest Based Advertising Cookies to display relevant advertisements on this and other websites based on your viewing behaviour. By clicking "Accept" you agree to the use of these Cookies and our Cookie Policy.