The need for long term studies to establish the best means of treating Diabetes, was underlined by Prof John Cleland from the University of Hull at the ESC Congress in Munich. Prof Cleland listed the latest treatment available for patients and voiced his concern about the side-effects and efficacy of available anti-diabetic drugs.
Treating Patients with Diabetes Mellitus
Treatment should be based, wherever possible, on the results of studies of substantial size and duration that measure outcomes that are meaningful to patients. We know that treatment of hypertension and hyperlipidaemia reduce progression of micro-vascular disease, reduce vascular events associated with large vessel disease and improve the prognosis of patients with diabetes.
In contrast, treatment of blood glucose has not convincingly reduced any of these problems, apart from sub-clinical micro-vascular disease. A major barrier to obtaining evidence that long-term treatments are beneficial is the short and often insecure drug patent-life offered by regulatory agencies.
Short-termism forces regulators to accept surrogates instead of real clinical benefits. This deceives manufacturers, scientists, clinicians and patients into believing that blood glucose control is an important goal of treatment rather than a mere surrogate for real success. Longer patent-life is a pre-requisite for obtaining good evidence that any treatment (for diabetes, blood pressure or lipids) designed to reduce long-term morbidity and mortality is safe and effective. Another alternative is to confine such studies only to patients with established cardiovascular disease and thus high event rates.
Type 1 Diabetes
Originally, diabetes mellitus was conceived as a syndrome of insulin deficiency, mainly affecting younger people and often associated with weight loss, glycosuria, ketoacidosis and often rapidly fatal. This syndrome requires insulin replacement therapy, which is still predominantly given by intermittent subcutaneous injections. The evidence that tight insulin control is superior to a lax regimen is weak.
The largest study, DCCT (n = 1,441), reported no reduction in diabetic keto-acidosis or mortality and only small reductions in vascular events (21 patients difference after 17 years follow-up) with clear benefits only on subclinical micro-vascular disease. This study was not blinded and we know that unblinded studies tend to over-estimate benefit. More intense insulin therapy was associated with substantial weight gain which may obviate any cardiovascular benefit of improved diabetes control. Insulin pumps, inhaled insulin and pancreatic islet cell transplants are potential, but mainly theoretical alternatives. Potentially, very long-term studies are required to establish the best means of treating this disease.
Type II Diabetes
Subsequently, a new population of patients with high insulin levels and hyperglycaemia (insulin-resistance) was identified. These were more commonly older patients, obese and had evidence of other cardiovascular diseases including hypertension and hyperlipidaemia. This population has increased markedly over the last 20 years, partly due to the growing proportion of the population who are elderly, partly because of the increase in obesity and partly because of the reduction in glucose thresholds required for diagnosis. Diabetes of this type does not really represent a distinct disease but rather just one end of the spectrum of the population. Blood glucose is continuously distributed in a similar way to blood pressure or heart rate. For each, there is an ideal natural range. Levels below this range or markedly above it cause acute illness and moderately elevated levels are associated with worse long-term outcome.
It is a giant assumption to suggest that using drugs to get patients back into the normal range is helpful or safe. We know from experience with other diseases that this assumption is no longer tenable. Unfortunately, there is remarkably little evidence that treatment of moderate hyperglycaemia is of benefit to patients and concerns that treating hyperglycaemia associated with obesity, other perhaps than by treating obesity itself, is safe.
Although the micro-vascular complications of diabetes mellitus are of great concern, few older patients with diabetes survive long enough to develop them. For instance, over 10 years follow-up in the 411 patients of UKPDS-34 managed with low intensity regimen (to maintain fasting blood glucose <15mmol/L) less than 5% had developed a micro-vascular complications by 10 years of follow-up, the vast majority of which were asymptomatic. Rates of myocardial infarction or stroke, even when blood pressure and lipid control are poor, in such primary prevention populations do not exceed 1% per year. Substantial visual loss, need for haemodialysis or troublesome neuropathy are rare. On the other hand, diabetes associated with already established vascular disease portends a high rate of vascular events and, if ventricular dysfunction is present, of death.
Oral Drugs for Treating Hyperglycaemia
Two major drug classes that have been in use for some decades are biguanides (metformin) and sulphonylureas (eg:- chlorpropamide, gliclazide and glyburide). Metformin reduces hepatic synthesis of glucose (considered its main effect) and increases glucose uptake (insulin sensitivity) of peripheral tissues. There is more evidence for a therapeutic benefit with this agent than any other but even that evidence is not strong. Sulphonylureas increase pancreatic insulin secretion. There is little evidence that these agents reduce micro- or macro-vascular complications but they do cause obesity.
Thiazolidinediones (eg:- Pioglitazone, Rosiglitazone)
These agents increase glucose uptake (insulin sensitivity) in peripheral tissues. They typically reduce HbA1c by 0.5-2% compared to placebo. Two substantial trials have been reported and several meta-analyses. Overall, these suggest little or no effect on clinically relevant outcomes in patients with diabetes mellitus. However, these agents can cause fluid retention that may cause peripheral and/or pulmonary oedema. A trial of pioglitazone (PROACTIVE) did suggest some clinical benefit but after following more than 5,000 patients for 3 years there was only 9 deaths difference between active and control groups.
Alpha-Glucosidase Inhibitors (Precose and Miglitol)
These agents slow the digestion of complex carbohydrates and reduce HbA1c by 0.5 to 1.0%. No adequate trial has been conducted to show that they improve outcome.
Glinides (Repaglinide, Nateglinide)
These agents enhance the release of insulin in response to glucose and therefore mimic normal human physiology. This reduces HbA1c by 0.5-2% compared to placebo. They may be combined with metformin but are not recommended with sulponylureas, which also stimulate insulin release. These are short-acting agents and should be taken before meals.
Peptidyl peptidase-IV (DDP-IV) Inhibitors (Sitagliptan, Vildagliptin)
These agents block the degradation of incretins (such as GLP-1 see below). This increases insulin release in response to glucose, may improve islet-cell mass and function and improve peripheral glucose uptake. They reduce HbA1c by about 1% and may reduce weight by 1-2kg. There is no evidence yet that these effects translate into clinical benefits.
New Subcutaneous Agents
Glucagonlike Peptide-I (Incretin) Analogues (GLP-1) (Exenatide, Liraglutide)
In healthy people GLP-1 rises along with insulin in response to food intake. Analogues can be used to enhance secretion of endogenous insulin in patients with type II diabetes mellitus. They also delay gastric emptying, may increase beta-cell (insulin producing) mass and enhance uptake glucose in the periphery. Treatment is associated with weight loss of up to 4-5kg over 2 years. HbA1c drops by around 1%. There is no evidence that these effects translate into clinical benefits for patients.
Insulin is co-secreted with a peptide called amylin, which delays the surge in blood glucose by slowing gastric emptying and reducing glucose production by the liver. Unlike insulin it suppresses appetite and causes weight loss. Unfortunately, it can produce a toxic substance called amyloid in animal models. Accordingly, pramlintide, a synthetic analogue of amylin that should not cause this problem, has been developed which may be co-injected with insulin. Studies suggest a 0.5-1.0% reduction in HbA1c associated with a 1-2kg weight loss. There is no evidence that this treatment is effective, or indeed safe, for long-term use in patients.
Results here (PDF)