Effects of Shifting From Twice Daily Insulin Glargine or Detemir to Once Daily Insulin Degludec in Type 1 Diabetes

Scientific background In type 1 diabetes mellitus, insulin therapy is a pivotal requirement, in order to achieve and maintain a satisfactory glycaemic control, as well as for the survival of the patient. Attaining a good glycaemic profile also allows limiting the impact of acute and chronic complications. Indeed, it is demonstrated that the maintenance of HbA1c levels as close as possible to 7% (53 mmol/mol), or even lower, can help in preventing or delaying the development of chronic complications. According to International Diabetes Federation (IDF) statements, the HbA1c target for young people with type 1 diabetes should be 6.5% in order to ensure a longer life free from complications. Pursuing this HbA1c goal should possibly be achieved without excess incidence of hypoglycaemia, which can compromise a patient's autonomic reactivity and represents a barrier to good glycaemic control. For this reason, patterns of insulin therapy in type 1 diabetes are intended to reproduce as close as possible the physiological endogenous insulin secretion, both in tonic (baseline) and phasic (prandial) states. This objective is typically achieved through the basal-bolus scheme, by subcutaneous administration of rapid-acting insulin analogue at the 3 main meals, and 1 or more injections of long-acting insulin that supplies inter-prandial and nocturnal insulinization.

Ideally the long-acting insulin should ensure a duration of 24 hours and be administered once a day. Long acting insulin analogues glargine and detemir are being used in type 1 diabetes. Clinical experience suggests that in a certain percentage of patients with type 1 diabetes, when administered at bedtime (h22.00), both glargine and detemir have a duration of less than 24 hours, which is manifested by an increase in blood glucose during the late afternoon (h17.00-20.00) resulting in pre-dinner hyperglycaemia, a phenomenon that plays regularly on different days or may be inconsistent owing to day-by-day variations in absorption. This issue in the clinical management of type 1 diabetes, also known as "sunset phenomenon", does not have a unique solution and still represents an "unmet clinical need". Possible approaches to address the sunset phenomenon include: (1) the control of capillary BG at h16.00-17:00 with eventual administration of an extra shots of rapid-acting insulin to correct the tendency to hyperglycaemia; (2) the use of a second long-acting insulin injection (usually at breakfast or lunch) in addition to that of the evening; (3) the replacement of the ultra-fast insulin at lunch with a pre-mixed insulin containing a mix of rapid-acting and protamine-bound intermediate-acting insulin; and (4) the transition to insulin pump therapy (CSII).

However, all these approaches have limitations: (1) patients may not be able to control daily capillary BG in the afternoon, and are anyhow still subject to an additional injection; (2) the two doses of long-acting insulin can overlap, exposing the patient to an increased risk of hypoglycaemia at other times during the day, being the patient nevertheless subject to an additional injection; (3) using pre-mixed insulin, the variations of the ultra-fast insulin dose according to pre-lunch BG and carbohydrate intake also involves proportional changes in intermediate insulin fraction, entailing the risk of hyper- or hypoglycaemia; and (4) insulin pump therapy is costly, requires long therapeutic education and special efforts by the patient, and is not suitable for all patients with type 1 diabetes. The choice of one of these approaches is carried out at the discretion of physicians, in order to be individualized for specific patient's needs. It is therefore reasonable that the search for other strategies to address the sunset phenomenon should be pursued, in a way which could be more effective, safe and convenient for patients.

In our Centre, one common solution to the sunset phenomenon issue for patients not eligible for the insulin pump therapy is a twice-daily administration of long-acting insulin analogue glargine or detemir. Indeed, the use premixed insulin does not fit well with the flexibility required for the therapy of type 1 diabetes, whilst an extemporaneous bolus administration is often inconstant, making identification of this phenotype of patients more difficult. The use of twice-daily administration of long-acting insulin is currently estimated in 20-25% of patients, while the use of premixed only represents 5-10%.

Insulin Degludec is a new generation ultra-long acting insulin analogue which has been developed by Novo Nordisk A/S and recently marketed with the trade name of Tresiba®. It has been approved by the Regulatory European Agency (EMA) for subcutaneous use in patients with diabetes mellitus and is currently available in some European countries, including Italy, England, Germany, and Switzerland (EMA/689592/2012; EMEA/H/C/002498).

Molecular modifications in insulin degludec monomer include the addition of a hexadecanoyl chain of fatty-diacid to Lysine-29 in B-chain (B29), and the deletion of Threonine at B30.

Degludec is a basal insulin that forms soluble multi-hexamers upon subcutaneous injection, resulting in a depot from which insulin degludec is continuously and slowly absorbed into the circulation leading to a flat and stable glucose-lowering-effect of degludec (see figure 1). During a period of 24 hours with once-daily treatment, the glucose-lowering effect of degludec, in contrast to insulin glargine, was evenly distributed between the first and second 12 hours. Thus, the duration of action of degludec is beyond 42 hours within the therapeutic dose range.

In phase 3 clinical trials involving patients with type 1 diabetes, insulin degludec showed similar glycaemic control in terms of HbA1c, but with a significant reduction in hypoglycaemia rates, especially at night, compared to glargine. This can also be explained by the lower day-to-day within-subject variability rate of insulin degludec respect to glargine. Given the longer duration of action of degludec, the timing of administration can be more flexible than glargine and detemir. As a result, it has been reported that the quality of life in patients with type 1 diabetes can significantly improve with insulin degludec use.

The continuous glucose monitoring system (CGMS) allows a nonstop BG level monitoring over 24 hours for up to 7 days. This tool can help patients to better understand how their food intake, physical activity and medications affect blood sugar levels, often giving them the possibility to better manage diabetes. Many CGMS devices have alarms that indicate when BG levels are going to be above or below a predetermined level. Devices are also equipped with a small glucose sensor inserted subcutaneously in the abdomen, which detects interstitial glucose levels every 5 minutes and sends a wireless signal to a terminal that can be easily applied to the belt or held in the pocket. When used in blind mode, CGMS recordings are only accessible to the diabetologist, but not to the patient, thereby allowing doctors to early detect problems in glycaemic trends (both hypo- or hyperglycaemic) 7 days of patient's normal daily life. It has been demonstrated that CGMS is capable of detecting hyper- or hypoglycaemic trends normally unidentifiable by the traditional SMBG provided by a glucometer. These systems are thus a valuable scientific and clinical support to define in greater detail 24-h glycaemic profiles, and represent the gold-standard for the evaluation of basal insulin efficacy and safety.

Scientific Purpose The present study aims to evaluate whether switching from an insulin regimen with twice-daily administration of long-acting insulin glargine / detemir (in combination with ultra-fast acting insulin analogue at main meals) to insulin degludec administration once-a-day can improve 24-h glucose profiles (from CGM) and glycaemic control (HbA1c and FBG) in patients with type 1 diabetes mellitus which are in suboptimal glycaemic control (baseline HbA1c >6.5%). Since the present research protocol was inspired by a clinical problem without a conclusive solution (unmet clinical need), that is the sunset phenomenon, it was not considered appropriate to provide a control group by randomization to insulin degludec vs. continuation of the previous insulin regimen. This protocol represents therefore an observational prospective study. A pseudo-control group will be created by including those patients who are not willing to shift from twice daily glargine/detemir to once daily degludec, and by the same patients of the "degludec group" observed in the 3 months before changing therapy.

Plausibility and clinical relevance Results of phase 3 clinical trials investigating insulin degludec in patients suffering from type 1 diabetes mellitus have shown that a treatment regimen characterized by the use of once-daily insulin degludec vs. once-daily glargine, in association with the same rapid-acting inulin analogue at meals, results in a similar glycaemic control in HbA1c, with significantly lower frequency of hypoglycaemia in degludec arms. Patients with the sunset phenomenon, taken as representative sample of our Centre, show HbA1c average levels of approximately 8-8.5% despite therapy with twice daily administration of glargine or detemir. Therefore, it can be expected that the transition to a regimen with once-daily ultra-long acting insulin degludec could improve glycaemic control, as well as 24-h profiles, without increase in hypo rates. This question is of particular clinical relevance, because will be addressed in the context of everyday clinical practice and not in a randomized-controlled trial. To date, there is scarcity of CGMS derived data on the efficacy of insulin Degludec. Therefore, the present study has the potential to better clarify the benefits of this new basal insulin.