Tatiana Yakovleva (1), Joanna Parkinson (2), Robert C. Penland (3), David W. Boulton (4), Kirill Peskov (1,5), Victor Sokolov (1), Weifeng Tang (4)
(1) M&S Decisions LLC, Moscow, Russian Federation, (2) Clinical Pharmacology & Quantitative Pharmacology, Clinical Pharmacology and Safety Sciences, R&D, AstraZeneca, Gothenburg, Sweden, (3) Clinical Pharmacology & Quantitative Pharmacology, Clinical Pharmacology and Safety Sciences, R&D, AstraZeneca, Boston, USA, (4) Clinical Pharmacology & Quantitative Pharmacology, Clinical Pharmacology and Safety Sciences, R&D, AstraZeneca, Gaithersburg, USA, (5) I.M. Sechenov First Moscow State Medical University, Moscow, Russian Federation
Introduction: Dapagliflozin is a highly potent and selective inhibitor of sodium-glucose cotransporter 2 originally developed for the treatment of type 2 diabetes mellitus (T2DM). In addition, dapagliflozin was approved in Europe and Japan for the treatment of T1DM as an adjunct to insulin therapy based on the results of Phase 3 DEPICT clinical programme [1, 2, 3]. While HbA1c reduction under dapagliflozin treatment in subjects with T2DM is well-characterized, the understanding of the relationship between exposure and HbA1c in T1DM is confounded by the fundamental differences in underlying pathophysiological mechanisms between the diseases.
Objectives: The aim of this work was to characterize the joint effect of dapagliflozin and total daily insulin dose on average daily glucose and HbA1c using a semi-mechanistic exposure-response model, as an additional confirmatory evidence on the effectiveness of dapagliflozin as an insulin adjunct in T1DM.
Methods: Data from one Phase 2 (MB102-072, NCT01498185: n=70; placebo or 1, 2.5, 5, 10 mg QD for 7 days) and one Phase 3 (DEPICT-2, NCT02460978: n=813 respectively; placebo or 5, 10 mg QD for 24 weeks) clinical studies with dapagliflozin were used to develop a non-linear mixed effects model of HbA1c response in patients with T1DM. External model qualification was performed on the data from Phase 3 DEPICT-1, NCT02268214 (n=778, placebo or 5, 10 mg QD for 24 weeks) trial.
The following baseline covariates were evaluated during the model development: age, sex, diabetes duration, body weight, BMI, fasting plasma glucose, haemoglobin, HbA1c.
The model was developed in a step-wise fashion to consecutively characterize total daily insulin dose (TDID) as a function of dapagliflozin exposure (1), continuous glucose measurements (CGM) response to the changes in TDID (2), dapagliflozin effect on CGM (3), and the link between CGM and HbA1c (4). The forward covariate search procedure as well as the mixture of distributions based on a categorical covariate were performed at the last stage of model development process. The models were evaluated by convergence, precision/stability of estimates, improvements in likelihood-based objective functions, and visual predictive checks.
The Monolix software (version 2018R2) was used for non-linear mixed effects modelling. Data visualization and forward simulations were performed in R software (version 3.5.1).
Results: The decrease of TDID with dapagliflozin treatment was successfully characterized by an Imax equation, with estimated Imax = 0.176 and IAUC50 = 320 ng/mL*h. The gradual increase of CGM with time (0.186 mg/dL/week) and the impact of TDID on CGM were calibrated using placebo data of MB102072 and DEPICT-2 studies, while the effect of dapagliflozin exposure was estimated using the data from the active arms. Due to poor convergence of the model, IAUC50 was fixed at the previously defined value for TDID (320 ng/mL*h). The changes in CGM alone or in combination with haemoglobin, with or without baseline covariates, could not adequately describe the HbA1c response at the concluding stage of the stepwise model development process. The adequate description of HbA1c response and successful model validation on DEPICT-1 study data was achieved when glucose-independent direct impact of dapagliflozin was introduced to the model as an Imax equation (Imax = 0.05). This indicates that dapagliflozin has an additional impact on HbA1c response which is independent of its glucose-lowering mechanism.
Conclusions: In this work, we have successfully established a quantitative link between the dapagliflozin exposure, TDID, plasma glucose and HbA1c in patients with T1DM. An additional glucose-independent decrease of HbA1c was identified during model development.
References:
[1] Dandona P et al. Efficacy and safety of dapagliflozin in patients with inadequately controlled type 1 diabetes (DEPICT-1): 24-week results from a randomised controlled trial. Lancet Diabetes and Endocrinol. 2017:5;846-17
[2] Mathieu C et al. Ef?cacy and Safety of Dapagli?ozin in Patients With Inadequately Controlled Type 1 Diabetes (the DEPICT-2 Study): 24-Week Results From a Randomized Controlled Trial. Diabetes Care 2018;41:1938–1946
[3] Dandona P et al. Efficacy and safety of dapagliflozin in patients with inadequately controlled type 1 diabetes: The Depict-1 52-week study. Diabetes Care 2018 Dec; 41(12): 2552-2559
Reference: PAGE () Abstr 9386 [www.page-meeting.org/?abstract=9386]
Poster: Oral: Drug/Disease Modelling