Comparison of pharmacokinetics and the exposure-response relationship of dapagliflozin between adolescent/young adult and adult patients with type 1 diabetes mellitus

Abstract

Aims: To quantitatively compare pharmacokinetics (PK) and the exposure-response (ER) relationship of the sodium-glucose cotransporter-2 inhibitor, dapagliflozin, between adolescents/young adults and adults with type 1 diabetes mellitus (T1DM).

Methods: Data from 2 clinical studies for dapagliflozin were analysed using a non-linear mixed-effects approach. The PK and the relationship between dapagliflozin exposure and response (24-hour urinary glucose excretion) were characterized. PK was evaluated using a 2-compartment model with first-order absorption while the exposure response-relationship was analysed using a sigmoidal maximal-effect model. The 24-hour median blood glucose, estimated glomerular filtration rate (eGFR), sex, age and body weight were evaluated as covariates.

Results: A 2-compartment model with first order absorption provided a reasonable fit to the dapagliflozin PK data. Body weight was found to be a significant covariate on dapagliflozin exposure. The ER relationship was best described by a sigmoidal maximal effect model with 24-hour median blood glucose and eGFR as significant covariates on maximal effect. In accordance with the observed data, model-predicted urinary glucose excretion response following 10 mg dapagliflozin dose was higher in the study in adolescents/young adults (138.0 g/24 h) compared to adults (70.5 g/24 h) with T1DM. This is linked to higher eGFR and 24-hour median blood glucose in this trial.

Conclusions: Dapagliflozin PK and ER relationship were similar in the 2 analysed studies after accounting for covariate effects. These results suggest that no dose adjustment is required for adolescent patients with T1DM.

Keywords: covariates; dapagliflozin; exposure-response; paediatric; pharmacokinetics; type 1 diabetes mellitus.

Figure 1

(A) AUC₂₄ and (B) 24‐hour UGE with 1^st, 2^nd, 3^rd quartile and predictions/observations outside quartile range and (C) 24‐hour UGE vs AUC₂₄ in adolescent/young adult and adult patients with T1DM following single‐dose oral administration of 1, 2.5, 5, 10 mg and placebo (0). Dots represent observed data. AUC₂₄, model predicted area under the dapagliflozin plasma concentration–time curve from time 0 to 24 hours on day 1; T1DM, type 1 diabetes mellitus; UGE, urine glucose excretion

Figure 2

Covariate effects on (A) relative steady‐state AUC₂₄ (AUC₂₄/reference AUC₂₄) and (B) relative 24‐hour UGE (24‐hour urine glucose/reference 24‐hour urine glucose). High and low covariate values represent the 5^th and 95^th percentile of the study population, error bars represent the 90% confidence interval. AUC₂₄, area under the dapagliflozin plasma concentration–time curve from time 0 to 24 hours; eGFR, estimated glomerular filtration rate; SMPG, self‐monitored plasma glucose; UGE, urinary glucose excretion. Reference group: body weight, 72.1 kg; eGFR, 94.5 mL/min/m²; median 24‐hour blood glucose, 164 mg/dL

Figure 3

Deterministic simulation of ER profiles for typical adult (solid line) and adolescent/young adult (dotted line) patients with T1DM. Squares and circles represent actual observations from adolescent/young adult and adult patients, respectively. AUC24, model predicted area under the dapagliflozin plasma concentration–time curve from time 0 to 24 hours on day 1; ER, exposure–response; T1DM, type 1 diabetes mellitus; UGE, urinary glucose excretion

Figure 4

Simulated absolute change in dapagliflozin 24‐hour UGE from placebo based on the ER model: (A) before, (B) after normalization for blood glucose, and (C) after normalization for both blood glucose and baseline eGFR. Median UGE response is represented by dashed lines for adolescents/young adults and by double‐dashed lines for adults. Grey ribbons are 90% confidence intervals for the median predictions. AUC24, model predicted area under the dapagliflozin plasma concentration–time curve from time 0 to 24 hours on day 1; eGFR, estimated glomerular filtration rate; ER, exposure–response; UGE, urine glucose excretion. 24‐hour UGE response was simulated using population‐specific median blood glucose and eGFR in (A), pooled‐data median blood glucose and population‐specific eGFR in (B) and pooled‐data blood glucose and eGFR in (C). See Methods for details