A randomized, 52-week, treat-to-target trial comparing insulin detemir with insulin glargine when administered as add-on to glucose-lowering drugs in insulin-naïve people with type 2 diabetes.

Background:  Previous studies have shown a glycemic advantage to using basal insulin as add-on therapy.  In addition, basal insulin may have advantages over NPH insulin, in that it causes less hypoglycemic episodes and may decrease the amount of weight gain associated with insulin use. 

Objective:  The objective of this study was to compare treatment with insulin detemir (once or twice daily) and insulin glargine (once daily) as add-on therapy to oral glucose-lowering agents in insulin-naïve patients with type 2 diabetes. 

Methods:  This study was open-label and used a non-controlled experimental, parallel-group trial, with randomization to each group.   582 patients were randomized to receive either insulin detemir (291) or insulin glargine (291).  The patients were Recruited/Included based on the following criteria:  ≥ 18 years old, ≥ 12 months disease duration, BMI ≤ 40.0 kg/m2, HbA1c 7.5%-10%, one or two oral agents (metformin, insulin secretagogues, alpha-glucosidase inhibitors) for ≥ 4 months on at least ½ the max recommended dose.  The patients were excluded if they were on thiazolidinedione treatment (due to labeling restrictions in Europe), more than two oral agents within 6 months, had hypoglycemic unawareness, or other medical conditions likely to interfere with trial conduct.  Withdrawal criteria included:  pregnancy, HbA1c > 11.0% after the first 12 weeks of treatment, and initiation of medication interfering with glucose metabolism.  Basal insulin was given once daily in the evening at 12 U and titrated according plasma glucose levels.  Detemir was given with pen-injector one hour before or after dinner, and glargine was given with pen-injector (EU) or by syringe (US) at bedtime.  A second dose of insulin detemir was allowed to be given within 30 minutes of breakfast, if the pre-dinner plasma glucose was greater than 7.0 mmol/L and pre-breakfast plasma glucose was less than 7.0 mmol/L, or nocturnal hypoglycemia occurred (major episode or less than or equal to 4 mmol/L).  The treatment length was 52 weeks and no meal time insulin was allowed during that time.  The primary outcome was baseline-adjusted HbA1c level at the end of the trial.  The secondary outcome measures included:  clinic fasting plasma glucose, within-participant variation in plasma glucose, ten-point self-measured plasma glucose profiles, proportion of participants with HbA1c less than or equal to 7.0% with and without hypoglycemia, change in body weight, incidence of hypoglycemia, adverse events, and standard safety parameters.  The study reported a power of 95%, and used intention-to-treat analysis. 

Results:  483 patients, 231 (79%) of the insulin detemir group and 252 (87%) of the insulin glargine group, completed the study.  The HbA1c was 7.12 (SE 0.11) in the once daily detemir group, and 7.06 (SE 0.10) in the twice daily detemir group, and for the insulin glargine group was 7.12 (SE 0.08), (95% CI -0.11, 0.21).  The clinic FPG was 7.27 (SE 0.31) and 6.73 (SE 0.25) for the once and twice daily insulin detemir, respectively; and for insulin glargine was 6.98 (SE 0.21), (95% CI -0.26, 0.58).    The differences in within-participant variation, 10 point SMBG profiles, and the proportion of participants achieving HbA1c levels at or below 7.0% were not statistically significant.  Hypoglycemic events occurred at a low rate of 5.8 (detemir) vs. 6.2 (glargine) episodes per patient year with a relative risk of 0.94 (96% CI 0.71 to 1.25).  Nocturnal hypoglycemia had 1.3 episodes per patient year for both insulin groups.  Weight gain was lower with insulin detemir compared to insulin glargine in both intention to treat analyses (2.7 vs. 3.5 kg, p=0.03) and per protocol analyses (2.8 vs. 3.5 kg, p=0.04).  In the comparison of insulin detemir vs. insulin glargine, it was found that they are both clinically significant and have similar improvements in glycemic control.  Less weight gain was noticed with once daily detemir compared to twice daily, but may be due to differences in diet and eating patterns.  Overall insulin dose was higher with insulin detemir due to twice daily dosing, and both insulin analogs have a low risk of hypoglycemia. 

Strengths:  The randomization method was clearly described and appropriate for this study, in that it evenly distributed the patients on mono- and combination oral therapy.  Also, the primary and secondary outcome measures were clearly defined, appropriate for the study, and adequately measured. 

Weaknesses:  The study was funded and monitored by the study drug manufacturer.  The design of the study was open label and used standard error to report the variability.  The within-group statistics involving the age, weight, and number of years with diabetes had large standard deviation values, resulting in large amounts of variation between the patients.  Compliance of the patients’ current medications or the study drugs were not specifically addressed, but could have played an important role when determining study drug efficacy.  Also, no control for diet and exercise was used in the study. 

Conclusion:  Further study is needed in order to properly evaluate the differences between once and twice daily dosing of insulin detemir, as well as, between the two insulin analogs.  I feel that this study was poorly conducted, and that the comparison between the insulin analogs would have been more accurate had the study allowed only once daily dosing with insulin detemir, or offered the twice daily option in both groups.  By using insulin detemir twice daily, the kinetics of the drug are no longer directly comparable to insulin glargine because the twice daily detemir profile is more similar to NPH insulin rather than  insulin glargine.  Also, it raises the question of how far did they titrate the insulin detemir before they switched to twice daily dosing?  Insulin detemir and insulin glargine may fit in as add on therapy in type 2 diabetes, but it is not known which if either is more effective at lowering HbA1c levels until further study is completed on their head to head comparison. 

Rosenstock J.  A randomised, 52-week, treat-to-target trial comparing insulin detemir with insulin glargine when administered as add-on to glucose-lowering drugs in insulin-naive people with type 2 diabetes.  Diabetologia. 2008 Mar;51(3):408-16.

Prepared by:  Marcella Hoyland, Pharm. D. Candidate, 2009