Background: Insulin glargine and lispro mix 75/25 (75% insulin lispro protamine suspension and 25% insulin lispro injection [LM75/25]) represent 2 common starter insulin regimen classes: basal and premixed. After initiation of starter insulin therapy, if patients with type 2 diabetes mellitus (DM) are unable to achieve a glycosylated hemoglobin (HbA1c) level <7.0%, insulin intensification may be indicated. The DURABLE (Assessing Durability of Basal Versus Lispro Mix 75/25 Insulin Efficacy) trial was designed to compare initiating insulin therapy with analogue basal insulin versus premixed analogue insulin in patients unable to achieve good glycemic control while taking multiple oral antihyperglycemic drugs (OADs)., Objective: To provide objective information about insulin intensification, the DURABLE trial also included a substudy evaluating a systematic approach to advancing insulin therapy in those patients who did not achieve glycemic control with their initial insulin regimen. This substudy, the results of which are reported here, tested the hypothesis that advancing insulin therapy with premixed insulin is noninferior to basal-bolus therapy (BBT) in patients with type 2 DM unable to achieve an HbA1c level < or = 7.0% after 6 months of starter insulin therapy., Methods: In the main DURABLE study, 2091 patients (age range, 30-80 years) with type 2 DM and HbA1c values >7.0% receiving > or = 2 OADs were randomized to receive insulin glargine (n = 1046) or LM75/25 (n = 1045), both in combination with prestudy OADs. After 6 months, patients with HbA1c levels >7.0% could enter this intensification substudy; OADs except sulfonylureas were continued. Patients originally receiving insulin glargine were enrolled in intensification arm A and were randomized to receive BBT (insulin glargine once daily plus mealtime insulin lispro TID) or LM75/25 BID. Patients originally receiving LM75/25 were enrolled in intensification arm B and randomized to receive BBT or mealtime 50% insulin lispro protamine suspension and 50% insulin lispro injection (LM50/50) TID. Insulin doses were adjusted based on preprandial plasma glucose levels. The primary end point was noninferiority of premixed therapy versus BBT with respect to end-point HbA1c. Secondary end points included change in HbA1c and weight, percentage of patients reaching HbA1c target levels, total daily insulin dose, and rates of hypoglycemia. The safety profile was also assessed., Results: Of the 475 patients in the insulin glargine + OAD arm of the main study who had HbA1c levels >7.0% at 6 months, 399 (84%) entered intensification arm A. The mean age was 57 years, 53% of the patients were male, and mean (SD) HbA1c was 8.0% (1.0%) at baseline. Of those patients, 199 were randomly assigned to receive BBT and 200 were assigned to receive LM75/25. Of the 411 patients in the LM75/25 + OAD arm of the main study who had an HbA1c level >7.0% at 6 months, 345 (84%) entered intensification arm B. The mean age was 55 years, 51% of the patients were male, and mean (SD) HbA1c was 8.0% (0.9%) at baseline. Of those patients, 171 were randomly assigned to receive BBT and 174 were assigned to receive LM50/50. At end point, noninferiority of LM75/25 or LM50/50 to BBT was supported, with a 95% CI of -0.10 to 0.37 and -0.25 to 0.25, respectively. At 6 months, HbA1c did not differ significantly from baseline in any group. Regardless of treatment group, <20% of patients achieved an HbA1c level <7.0%. There were no significant differences between groups in total daily insulin dose, weight gain, incidence or rate of hypoglycemia, or incidence of serious adverse events., Conclusions: No group had significant improvement from baseline in HbA1c. Our study results suggest that premixed therapy, dosed 2 times per day (LM75/25) or 3 times per day (LM50/50), was noninferior to BBT (4 injections/d) in this population of adult patients with type 2 DM previously uncontrolled with OADs plus basal insulin or twice-daily premixed insulin. Clinical-Trials.gov identifier: NCT00279201., (Copyright 2010 Excerpta Medica Inc. All rights reserved.)