We appreciate the thoughtful comments raised by Agrawal and Gupta (1) on our recent article (2) on the efficacy and safety of the glucagon receptor antagonist LY2409021 in patients with type 2 diabetes. We are pleased that they acknowledge that this study has provided further proof of efficacy of glucagon antagonism as a treatment for type 2 diabetes with acceptable adverse effects at a dose of 30 mg (1). Agrawal and Gupta queried how we arrived at the unequal allocation to the treatment groups with the ratio of 1.5:2.1:1:1 (for 60 mg, 30 mg, 10 mg, and placebo, respectively). Our randomization was probability based and computer generated with the goal to optimize the number of patients allocated to the 30 mg dose, selected on the basis of the results of the phase 1 studies (3). Agrawal and Gupta also inquired about the reason for fasting triglycerides >400 mg/dL as an exclusion criterion and wanted to understand the basis for this exclusion since no significant change in triglyceride concentration was noted in the phase 1 study of this molecule (3). The rationale to exclude patients with uncontrolled hypertriglyceridemia in the reported clinical studies was not based on expected side effects of LY2409021, as suspected by Agrawal and Gupta, but rather to exclude a bias of hypertriglyceridemia-induced insulin resistance on the primary efficacy objective. In addition, it is considered ethically inappropriate to include patients with severe hypertriglyceridemia in placebo-controlled studies because of the potential risk of worsening during placebo treatment including secondary adverse effects. The authors also noted that we used weekly change in HbA1c as a secondary outcome measure and were of the opinion that the clinical implications of a weekly change in HbA1c is difficult to comprehend because of the fact that HbA1c levels change gradually over a 3-month period. Although change in HbA1c from baseline to end was used to assess the primary end point in each study, we did not in fact use weekly change in HbA1c as a secondary outcome measure. In both studies, HbA1c was not determined weekly but every other week, and depicting the course of HbA1c in a graphical manner as done in Fig. 1A and B of the article (2) is a common way of reporting the data for diabetes studies. Agrawal and Gupta further pointed out that the efficacy of the treatment is reported to be dose dependent but the least squares mean change in HbA1c from baseline is highest in the 10-mg group and lower in the 30- and 60-mg groups, which was unexplained. We agree that in the phase 2a study, a dose-dependent difference between the 10-mg dose and higher doses (30 and 60 mg) was not apparent, indicating that the higher doses were on the upper end of dose-response curve; this result is consistent with the previously reported exposure-response data for LY2409021 (3). However, there is an obvious dose-dependent change in HbA1c in the phase 2b study between the 2.5-mg and 20-mg dose levels, thus confirming that the 30-mg and 60-mg doses used in the phase 2a study were on the upper end of dose-response curve.
Article Information
Duality of Interest. The study was sponsored by Eli Lilly and Company and/or one of its subsidiaries. C.M.K., R.P.K., P.G., H.F., D.E.W., W.H.L., M.A.D., D.E.M., and T.A.H. are employees and stockholders of Eli Lilly and Company and/or one of its subsidiaries. Y.D. and C.N.L. were employees of Eli Lilly and Company and/or one of its subsidiaries. A.J.L. was an employee of the National Research Institute, Los Angeles, CA, at the time of the study and is currently retired. No other potential conflicts of interest relevant to this article were reported.
