Beyond the Diabetes Control and Complications Trial—Addressing Weight Gain in Type 1 Diabetes

Type 1 Diabetes Treatment Challenge

Type 1 Diabetes Treatment Challenge
At diagnosis, patients with type 1 diabetes are often underweight following a period of glucosuria, osmotic diuresis, and catabolism due to insulin deficiency. The initial weight gain upon commencing insulin treatment is viewed as normalization. However, as patients work to bring their glycated hemoglobin (HbA_1c) level as close to the target level of ²7% as possible, a subset will not just normalize to their pre-diabetes weight but begin to exceed it.^2,3 In the DCCT, weight gain with intensive diabetes therapy was greater than with conventional treatment (5.1 versus 3.7kg; p<0.0001 during the first 12 months of therapy). After 12 months, the intensively treated cohort continued to gain weight that was, on average, 10% above ideal,⁴ and one-quarter of the intensively treated subjects attained an average body mass index (BMI) that exceeded 30kg/m2,² the current cut-off for obesity. The subgroup that gained this excess weight had larger waist-to-hip measurements (central obesity), higher triglyceride levels, and lower high-density lipoprotein (HDL) cholesterol, and required more insulin (on a unit/kg/day basis) to achieve the same glycemic targets as a matched group that did not gain weight with intensive therapy2— all outcomes we now associate with the metabolic syndrome and increased risk for cardiovascular disease. Taken together, these observations suggest that weight gain with intensive therapy of type 1 diabetes should not be overlooked in the everyday clinical setting.

The analysis of this problem raises two questions: how do patients cope with the insulin-induced weight gain, and does this weight gain negate the improvements achieved by lowering blood glucose? Regarding the first question, early in the feasibility study DCCT researchers attempted to limit weight gain through intensive nutritional management.⁵ Despite these efforts, excess weight gain still emerged as a significant complication of intensive management in the full trial.⁶ Instead, it has become evident that to avoid unwanted weight gain some patients with type 1 diabetes deliberately ignore the prescribed insulin advice. A US study of 341 women (aged between 13 and 60 years) with type 1 diabetes reported that 31% intentionally omitted insulin,⁷ with half of them citing weight control as their primary reason. These women had more diabetes-related hospitalizations and microvascular complications and displayed greater psychological distress than patients that adhered to treatment. The authors concluded that: “Patients pre-occupied with weight concerns may also become emotionally overwhelmed by insulin treatment and the associated weight-related consequences, thus reinforcing the desire to omit insulin and maintain elevated blood glucose levels.”

The problem of insulin omission was confirmed in a small UK study of 65 young individuals with type 1 diabetes who were followed during the transition from adolescence to young adulthood.⁸ In that study, 30% of the women admitted to having underdosed insulin to manipulate their weight, while 45% of women who developed microvascular complications had intentionally misused insulin doses to prevent weight gain. Despite the omission, average weight and BMI increased during the 10-year study; women were overweight as both adolescents and adults, while men became overweight as young adults. Concern about bodyweight and form increased significantly with time for both sexes, resulting in increased dietary restraint. Given the benefits of improved glycemic control on both microvascular and macrovascular events in patients with type 1 diabetes,^9,10 such suboptimal glycemic control arising from insulin misuse has worrying prognostic implications.

Traditionally treated group, as mentioned above, a subgroup within the intensively treated group experienced excessive weight gain. Purnell et al. categorized data from the intensively treated DCCT cohort by quartiles of weight gain,² all of whom began the trial with similar BMIs and achieved the same target HbA_1c levels near %. In the first quartile of the study, where BMI remained stable, lipid levels, blood pressure, and insulin doses were all favorable compared with conventional therapy. However, in the fourth quartile, where BMI increased by some 7kg/m2, all parameters of cardiovascular risk (systolic blood pressure and levels of triglycerides, and (HDL) and low-density lipoprotein [LDL] cholesterol) were worse compared with the first quartile and no different than the conventionally treated group. Essentially, this excess weight gain negated the benefits of intensive therapy on these macrovascular risk factors.

Etiology of Weight Gain Associated with

Intensive Treatment
Several mechanisms have been proposed to explain the weight gain associated with insulin therapy. These include decreased glycosuria owing to improved glycemic control, the direct lipogenic effects of insulin on adipose tissue, and increased food intake owing to recurrent hypoglycemia. Purnell et al. sought evidence for yet another possible explanation: genetics. They examined the effect of having a family history of type 2 diabetes on the weight gain of the patients with type 1 diabetes in the DCCT cohorts. With intensive therapy (but not conventional therapy), the type 1 patients with a family history of type 2 diabetes not only gained more weight than the patients who did not, but also displayed characteristic changes in their blood pressures and lipids that are similar to the metabolic syndrome, implying that this metabolic defect can also occur in type 1 diabetes patients.¹¹ The authors reasoned that with near normalization of glycemic control, genetic traits that would otherwise have been suppressed become manifest in these patients.

Normal Insulin Secretion
The increased risk for hypoglycemia associated with trying to maintain euglycemia during insulin therapy reflects the inability of exogenous insulin regimens to reproduce physiological insulin profiles. Normal insulin secretion consists of two major components: a chronic basal release plus meal-induced surges in secretion.¹² The role of the low-level basal insulin secretion is to modulate the rate of overnight hepatic glucose production and glucose output between meals. Thus, basal glucose levels are maintained within a narrow range. Meal-related insulin secretion controls post-prandial increases in blood glucose, which seldom rise above 5.5mmol/l for more than 30 minutes.¹³

Insulin Therapy
Ideally, insulin replacement should reproduce both the basal and prandial/post-prandial secretion profile to achieve stable glycemic control. The latest strategy is known as basal–bolus therapy, since it attempts to reproduce both the basal and meal-induced components of normal insulin secretion. Unfortunately, the mean pharmacokinetic insulin profiles produced by basal–bolus therapy are not perfect recreations of normal physiological secretion profiles.¹⁴ There is variability in the absorption, and hence the pharmacodynamic profiles, of subcutaneously injected insulin, which means that the intended profile is seldom recreated by any given injection.^15–17 Thus, there is often an unpredictable imbalance between insulin supply and physiological need, with periods of both over-supply, leading to hypoglycemia, and under-supply, leading to post-prandial hyperglycemia, during each day. Imperfect mean pharmacodynamic profiles and variability are particularly problematic with traditional basal insulins and as these are generally dosed in the evening, hypoglycaemia often occurs at night, when patients are asleep and unable to react. This becomes another limiting factor for adherence with insulin doses, and may compound weight gain through defensive snacking.

Apart from the pharmacodynamic profiles, the route of administration of exogenous insulin can impact the weight gain with insulin therapy.¹⁸ In normal physiology, insulin is secreted into the portal vein and arrives at the liver first, where it suppresses endogenous glucose production (EGP) and up to 60% is cleared via receptor interaction. Only the remaining 40–50% continues into the systemic circulation to act on peripheral muscle and fat tissue to increase glucose uptake and suppress lipolysis. However, when insulin is given subcutaneously, the situation is very different. Here, the absorbed insulin first circulates systemically, so it has a disproportional influence on muscle and adipose tissue.¹⁹ In effect, the liver is ‘under-insulinized’ and the periphery ‘overinsulinized,’ which could at least partly explain the disproportionate increase in fat mass typically reported with insulin therapy.²⁰

Approaches to Reduce Weight Gain with Intensive Therapy

Increasing Insulin Sensitivity
Two broad strategies can be employed to limit the weight gain that accompanies intensive therapy: reducing the insulin dose requirement and employing alternative administration methods that more accurately reflect the dynamic physiological needs for insulin. Metformin has been associated with weight loss in type 2 diabetes, and has both insulin-sparing and cardioprotective properties.²¹ In a placebo-controlled three-month study of 27 adolescents taking more than one unit of insulin/kg/day, the addition of metformin enabled a 0.12U/kg dose reduction compared with placebo (p<0.05), and achieved a relative HbA1c reduction of 0.6%. ²² BMI remained unchanged in this short-term study, but longer-duration studies are planned to quantify the potential benefits of adjunct insulin-sparing therapy and to determine whether specific subgroups stand to gain any particular benefit.

Other studies have examined the addition of medications that work on the central nervous system to reduce food intake and cause sustained weight loss maintenance. Pramlintide, discussed further below, is an analog to the islet-hormone amylin with receptors in the brainstem that enhance satiety.²³ When added to insulin therapy in type 1 diabetes,^24,25 a pooled analysis of three studies showed pramlintide to be associated with reductions in both HbA_1c (by 0.3%; p<0.001) and bodyweight (1.8kg; p<0.001) over six months without an increase in severe hypoglycemia.²⁵

Appropriate Timing
By providing insulin when needed, glycemic control can be improved and the risks of adverse effects minimized. Moreover, the appropriate timing of insulin availability may enable greater dose-efficiency, and this too should help avoid weight gain. The advantages of increasing early post-prandial insulin availability have been demonstrated in studies involving patients with type 2 diabetes.²⁶ Insulin was shown to have a significantly greater overall blood-glucose- lowering effect when administered as a bolus early in the prandial setting than when delayed or protracted.

Multiple injection or continuous subcutaneous insulin infusions (CSII) regimens make efficient use of the total insulin doses.^27,28 Kulenovic et al. found that patients with type 1 diabetes on ‘conventional’ twice-daily insulin regimens had a higher mean BMI than those on multiple injection insulin treatment (23.2 versus 21.2kg/m2; p<0.01). There was also a high proportion of people who were overweight (27.3 versus 0%; p<0.012).²⁹ In addition, insulin doses were higher in the conventionally treated patients (0.84 versus 0.77IU/kg: p<0.05), but glycemic control was poorer. Insulin Analogs
The development of insulin analogs has allowed insulin replacement timing that is compatible with physiological need. The rapidly absorbed analogs insulin aspart, insulin lispro, and insulin glulisine have pharmacokinetic profiles that more closely match the physiological prandial insulin secretion than soluble human insulin preparations, and are well suited to CSII. Clinically, these analogs significantly improve post-prandial glycemic control compared with exogenous soluble human insulin, and also reduce hypoglycemic risk concomitant with minor improvements in HbA_1c. Their short duration of action often requires an increase in the dose of basal insulin when used in basal–bolus therapy.³⁰ Insulin glargine is the first available long-acting human insulin analog, and has been shown to induce a smoother metabolic effect than neutral protamine Hagedorn (NPH) insulin, and therefore mimics more closely the basal and prandial/post-prandial insulin secretion profile to achieve stable glycemic control.³¹

Although insulin analogs have not been unequivocally proved to limit weight gain, one in particular—insulin detemir—has yielded promising results. Insulin detemir is an insulin–fatty acid hybrid, in which myristate has been acylated t the terminus of the B chain, thus enabling reversible albumin binding. Binding to albumin is thought to be responsible for the agent’s smooth nd prolonged absorption profile, with low variability in pharmacodynamic response from injection to injection.^32,33 Insulin detemir was approved in the EU in 2004 and in the US in 005.³⁴ It has consistently displayed a weight-sparing effect in seven trials of patients with type 1 diabetes where either a mean weight loss^35–40 or a mean weight gain of <0.25kg was observed.^39,41

Insulin detemir treatment decreases the relative risk of nocturnal hypoglycemia compared with NPH insulin.^35,38,40,41 In patients with type 1 diabetes receiving detemir as the basal component of basal–bolus therapy, weight was unaltered or showed minor non-significant reductions over periods of up to 12 months. In contrast, NPH insulin was associated with weight gain of 0.7–0.8kg over six months^{33,35,37,40,41} and 1.2–1.4kg over 12 months.^36,39 Importantly, the effects of insulin detemir versus NPH insulin are maintained for longer. Results from a six-month open-label comparison of insulin detemir and NPH insulin in 747 patients with type 1 diabetes indicated that those who received insulin detemir lost 0.23kg over the course of the trial versus a 0.31kg weight gain for patients treated with NPH insulin (p=0.024).³⁸

Insulin Adjuncts
The autoimmune process that leads to the dysfunction of pancreatic b cells is the main pathogeneses of type 1 diabetes. However, insulin is not the only hormone depleted by b-cell destruction: amylin, a 37 amino acid hormone, is co-secreted with insulin by b cells in response to meals to suppress postprandial glucagon secretion. Pramlintide, a synthetic amylin analog, has the same biological effect and potency in regulating glucose metabolism as human amylin, but was developed to be non-aggregating and non-adhesive.42 By substituting for amylin as an insulin adjunctive therapy, in addition to its satietyenhancing effects pramlintide aids in the absorption of glucose by slowing gastric emptying and inhibiting secretion of glucagon, the counter-regulatory hormone that opposes the effects of insulin. Pramlintide is taken by subcutaneous injection in conjunction with mealtime insulin. It is indicated for patients with type 1 and 2 diabetes who have failed to achieve adequate glucose control despite optimal therapy with mealtime insulin alone.⁴³ The studies on insulin detemir continue. An interventional, randomized, double-blind, pilot study trial that is currently recruiting will explore the weight-neutral effects of insulin detemir compared with insulin glargine. The primary objective is to determine if type 1 diabetes patients consume fewer calories when allowed to eat to satiety while treated with insulin detemir compared with insulin glargine.

The efficacy of pramlintide in prevention of weight gain in early onset of type 1 diabetes will be evaluated
by an interventional, randomized, open-label pilot study conducted by the University of Texas (currently recruiting). Early onset is defined as those who are diagnosed with type 1 diabetes six to 12 months prior to entry in the study.

Summary and Conclusion
Excessive weight gain with intensive diabetes therapy may hinder patients’ compliance with treatment and predispose patients to serious cardiovascular risks. Insulin-sparing adjunctive therapies aimed at improving insulin sensitivity, therapies that reduce food intake via central nervous system mechanisms, and insulin analogs designed to maintain insulin at physiological levels hold promise to prevent or limit this excess weight gain. Ongoing and future studies will hopefully elaborate on the performance of these agents and their effect on both micro- and macrovascular disease in patients with type 1 diabetes.