The American Diabetes Association provides guidelines for the management of diabetes and utilizes a general step-wise approach for adding on therapy to achieve glycemic goals that ends with multiple daily doses of insulin therapy. However, the ADA does not give a preference regarding the combination of therapies to use.1 When the need for insulin is reached, previous studies have exhibited that patients are more likely to achieve therapeutic goals with a basal-bolus regimen compared to premixed insulin therapies, though the approach to titrating insulin regimens to achieve goal hemoglobin A1c varies greatly.2,3,4 Though it is common in clinical practice to consider self-titration regimens for basal insulin therapies, it is not as common to use this method for prandial insulin adjustments. The current literature is limited regarding the use of self-titrated prandial insulin regimens in type 2 diabetic patients (T2DM).5
The AUTONOMY trial was designed to investigate the potential for Type 2 diabetic patients, who were not controlled with oral antidiabetic drugs (OADs) and basal insulin, to safely and effectively use a self-titration algorithm for insulin lispro therapy. This multicenter, randomized, controlled, open-label, parallel trial was conducted in 14 countries around the world. Two independent studies were conducted to validate the self-titration algorithm. Study A and Study B enrolled a total of 1112 patients who were 18 to 85 years of age with T2DM, a body mass index (BMI) of less than 45kg/m2, an A1c between 7 and 12%, and who had been on basal and OADs therapy for at least three months. If patients were previously on prandial insulin therapy, excessively insulin resistant, morbidly obese, pregnant or planning on becoming pregnant, or had a history of cancer, severe hypoglycemia, or moderate to severe conditions that would make the patient more high risk, they were excluded from the trial. Patients were also excluded if they were on glucagon-like peptide-1 receptor agonists, dipeptidyl peptidase-4 inhibitors (except for sitagliptin), recent rosiglitazone use, or recent glucocorticoid therapy at screening. 5
A six-week lead-in period was available to convert patients to an appropriate insulin glargine regimen, before randomization. Eligible patients were then randomized to receive either a daily (Q1D) or every three-day (Q3D) algorithm for self-titration based on pre-prandial blood glucose values to achieve a goal range of 85 to 114mg/dL. The algorithm also allowed for a decrease in insulin dose if blood glucose values were below goal. Insulin lispro was added sequentially as needed starting with the breakfast meal, then lunch and dinner, if needed. The subsequent pre-meal blood glucose was used for titration. The primary outcome was the change in A1c from baseline to the end of the study at week 24. Secondary outcomes such as percentage of patients achieving A1c goal, weight gain, and hypoglycemia were monitored as well. A sample size calculation was conducted for the primary outcome measure to establish non-inferiority. 5
Baseline demographic information was similar between treatment groups. There was a statistically significant ...