Comment on Edelman et al. AUTONOMY: The First Randomized Trial Comparing Two Patient-Driven Approaches to Initiate and Titrate Prandial Insulin Lispro in Type 2 Diabetes. Diabetes Care 2014;37:2132–2140

The AUTONOMY study (1) examines an important issue in the care of patients with type 2 diabetes: When inadequate glycemic control mandates addition of prandial insulin to a basal-only insulin regimen, can the initiation and titration of bolus insulin doses be handled in a patient-driven manner without compromising efficacy or safety? A straightforward, evidence-based strategy that enabled patients to self-manage treatment intensification with prandial insulin would provide guidance for implementing treatment recommendations (2–4).

In this regard, the AUTONOMY study is to be welcomed alongside the recent FullSTEP study (5). Both trials examined patient-driven titration of prandial insulin in type 2 diabetes previously inadequately controlled with basal insulin. The FullSTEP study compared stepwise addition of bolus insulin with a full basal-bolus regimen; AUTONOMY compared a daily and a 3-day protocol for prandial insulin titration. Unfortunately, the AUTONOMY study report appears to misrepresent aspects of the FullSTEP study (5).

The introduction to the AUTONOMY report states that “no randomized, controlled studies in patients with type 2 diabetes have investigated treatment escalation with prandial (bolus) insulin using patient-driven treatment intensification” (1). Similarly, the discussion asserts that in FullSTEP “the prandial dose was adjusted by the study investigators” (1). These two statements are incorrect; FullSTEP examined patient-driven treatment intensification in two arms: three daily prandial doses added to basal insulin from the start of the treatment period and use of one, two, or three prandial boluses added in a stepwise manner. In both arms, patients self-titrated their prandial insulin dose (5). By contrast, in the AUTONOMY study, Edelman et al. added one to three boluses sequentially in both arms (1).

In the stepwise arm of the FullSTEP study, additional boluses were added at predetermined intervals if the target HbA_1c of <7.0% (53 mmol/mol) was not achieved (5). In AUTONOMY, investigators added additional boluses at unspecified time stages “if necessary” (1). In both studies, patients used a simple dose-titration algorithm: the AUTONOMY algorithms involve changes of 1–4 units/day, depending on blood glucose measurements using either a 1- or 3-day adjustment schedule; the FullSTEP algorithm adjusts the prandial dose by 1 unit/day according to blood glucose levels.

The AUTONOMY and FullSTEP investigators report improved glycemic control with low rates of hypoglycemia. In AUTONOMY, there were no clinically meaningful differences between the two adjustment schedules evaluated. In FullSTEP, the stepwise regimen was noninferior to the full basal-bolus regimen with respect to change in HbA_1c from baseline to end of trial. The FullSTEP study also showed greater treatment satisfaction and fewer hypoglycemic events in the stepwise arm compared with the full basal-bolus arm.

Thus, the AUTONOMY study shows that self-management of treatment intensification can be successful with either a 1- or 3-day adjustment schedule. However, it is not the first to investigate patient-driven treatment escalation with prandial insulin. Data from studies such as FullSTEP provide practical guidance to assist health care providers in the implementation of current treatment recommendations (2–4).