Pharmacokinetics and Pharmacodynamics of Dapagliflozin in Obese and Nonobese Healthy Adults Following Multiple Administration Authors: Hae Won Lee, Hyungjin Kim, Ji Seo Park, Jae Hwa Lee, Jin Ju Park, Mi-Ri Gwon, Joo-Youn Cho, Sook Jin Seong, Young-Ran Yoon Published: 2026 ABSTRACT AbstractBackgroundDapagliflozin, a sodium-glucose cotransporter 2 inhibitor, lowers blood glucose levels by preventing renal glucose reabsorption, resulting in urinary glucose excretion. Because dapagliflozin-induced glycosuria produces a modest reduction in body weight, it may benefit patients with type 2 diabetes who are overweight or obese. Pathophysiologic alterations associated with obesity may lead to clinically significant differences in drug pharmacokinetic (PK) and pharmacodynamic (PD) properties.PurposeThis study aimed to evaluate the PK and PD characteristics of dapagliflozin in obese and nonobese healthy adults and assess the effect of obesity on these profiles by comparing outcomes between the two groups following multiple oral doses.MethodsIn this open-label, multiple-dose study, dapagliflozin 10 mg tablets were administered once daily for 5 days to healthy adults. Serial blood samples for PK analysis were collected before and up to 48 h after the last dose on day 5. PK parameters were determined using noncompartmental analysis.Sensitivity analyses were conducted, including correlation analysis of PK parameters with weight or BMI using Pearson’s correlation coefficients (r), as well as analysis of covariance (ANCOVA) with body weight as a covariate, supplemented by bootstrap analysis, to evaluate the effect of obesity on the primary PK and PD parameters.ResultsThirteen nonobese and nine obese participants completed the study. The geometric mean ratio (obese/nonobese) (90% CIs) of the area under the plasma concentration-time curve during a dosing interval at steady state (AUCtau,ss) and maximum plasma concentration at steady state (Cmax,ss) were 0.6855 (0.5503–0.8538) and 0.7673 (0.6145–0.9581), respectively. There was no moderate correlation (r > 0.6) between any of the PK parameters analyzed and weight or BMI. The ANCOVA least squares mean differences between the nonobese and obese groups were 163.6 h*ng/mL, 38.4 ng/mL, 2.4 mg/dL, for AUCtau,ss, Cmax,ss, and serum glucose levels from baseline to maximum level on day 5 (ΔGmax), respectively (p value > 0.05, ANCOVA), which were consistent with the median values of bootstrap estimates.ConclusionThe AUCtau,ss and Cmax,ss values for the obese group were lower than those of the nonobese group (31% and 23%, respectively). The sensitivity analyses revealed no moderate correlation between the primary PK and PD parameters and weight or BMI, nor any statistically significant differences in systemic exposure and ΔGmax between the nonobese and obese groups; however, this finding should be interpreted with caution due to the exploratory nature of the study and its limited sample size. Dapagliflozin was safe and well-tolerated following multiple oral doses.Keywords: dapagliflozin, obesity, pharmacokinetics, pharmacodynamics, type 2 diabetes Introduction Obesity is a chronic, multisystemic, complex disease that is spreading globally at pandemic proportions, representing a major public health challenge.1 Obesity and type 2 diabetes (T2D) frequently co-occur, with a significant portion of individuals with T2D being overweight or obese.2–4 These conditions are closely related due to the multiple overlapping pathophysiological pathways.2 Obesity induces a chronic pro-inflammatory state that promotes insulin resistance and constitutes a major risk factor for T2D.5,6 Prevention of weight gain or modest weight loss in individuals with T2D can improve glucose homeostasis by reducing insulin resistance, lowering cardiometabolic risk factors, and mitigating diabetic complications.7,8 Therefore, the effect of antidiabetic treatment regimens on body weight is a critical consideration, given the benefits of weight reduction and the potential hazards of weight gain in individuals with T2D. Current guidelines recommend glucose-lowering medications that are weight-neutral or promote weight reduction as preferred therapies for overweight or obese patients with T2D.9,10 Sodium-glucose cotransporter 2 (SGLT2) inhibitors, approved for the management of T2D, are associated with weight loss. These agents increase urinary glucose excretion (UGE) by inhibiting renal glucose reabsorption, thereby reducing plasma glucose levels in an insulin-independent manner.11,12. Weight reduction associated with SGLT2 inhibitors is primarily due to energy loss from UGE and fluid loss from osmotic diuresis.12 Dapagliflozin is a potent, selective, and reversible SGLT2 inhibitor that effectively lowers blood glucose, reduces body weight, and exerts beneficial effects on blood pressure, lipid profiles, and hyperuricemia.12,13 Recent clinical trials have demonstrated that dapagliflozin treatment reduces cardiovascular mortality and hospitalization for heart failure, and provides renoprotective effects.12–14 Furthermore, dapagliflozin is generally well-tolerated, presents a minimal risk of hypoglycemia, and exhibits no significant pharmacokinetic (PK) interactions with other antihyperglycemic drugs, including metformin and sitagliptin.13,15 Accordingly, dapagliflozin can be used as a valuable add-on treatment option for managing T2D across a broad patient population.13,16 Obesity induces complex pathophysiological changes in multiple tissues and organs, including increased blood volume, cardiac output, and capillary flow; altered liver expression of drug-metabolizing enzymes; elevated glomerular filtration rate; and accelerated gastric emptying and intestinal permeability.17,18 These alterations may significantly influence the PK and pharmacodynamic (PD) characteristics of drugs.19 However, relatively few studies have investigated obesity-related changes in drug PKs, as obese individuals are usually excluded from most bioavailability studies.20 To the best of our knowledge, this is the first study to specifically examine the effect of obesity on the PK and PD profiles of dapagliflozin. This study aimed to evaluate the PK and PD characteristics of dapagliflozin following multiple oral administrations in nonobese and obese healthy adult participants and to assess the effect of obesity on these parameters. Methods Participants and Study Design Male volunteers aged 19 years or older, in good health, with a body mass index (BMI) of 18.0 kg/m2 to ˂ 25.0 kg/m2 or ≥ 30.0 kg/m2 were eligible to participate in the study. Based on BMI, participants were assigned to either the nonobese or obese group. Health status was evaluated during a screening visit by medical history, physical examination, routine clinical laboratory test results, and 12-lead electrocardiography. Key exclusion criteria included: hypersensitivity to the study drug; a history of, or current clinically significant medical illness, including diabetes mellitus, diabetic ketoacidosis, or lactic acidosis; estimated creatinine clearance (Cockcroft-Gault equation) ≤ 80 mL/min; aspartate aminotransferase, alanine aminotransferase, or total bilirubin > 1.5 × upper limit of normal (ULN); or prior participation in another clinical study within 6 months of the first dose of study drug. It was prohibited to take any medication or consume any food that could alter the metabolism of the drug, such as grapefruit juice, for seven days before the initial dose of dapagliflozin until the end of the study. It was also prohibited to drink any xanthine-containing beverages, such as coffee, tea, or Coke, during the admission period. The study complied with the International Conference on Harmonization Good Clinical Practice (ICH-GCP), the Declaration of Helsinki (and subsequent amendments), and all applicable local regulations. The study protocol was reviewed and approved by the Kyungpook National University Hospital (KNUH) Institutional Review Board (Daegu, Republic of Korea; Approval No. 2022–09-016-001). The study was registered in the Clinical Research Information Service, a public trials registry, prior to initiation of the first participant enrollment (https://cris.nih.go.kr/cris/search/list; identifier number KCT0007864). Written informed consent was obtained from all participants before enrollment. The study design is shown in Figure 1. Participants were admitted to the KNUH Clinical Trial Center (CTC) on day −1. On day 1, following an overnight fast, participants received a single oral dose of dapagliflozin 10 mg (Forxiga®; AstraZeneca Korea, Seoul, Korea) with 150 mL of water before discharge. On days 2–4, participants returned to the KNUH CTC each morning to receive the study drug under investigator supervision. On day 4, they were admitted to the KNUH CTC at 20:00 h, and remained for 24 h following the final dose on day 5. On day 5, the last dose of dapagliflozin was administered with 150 mL of water in the morning, and serial blood samples were collected for 24 h before discharge. On day 7, 48 h following the last dose, a follow-up visit was conducted for the final PK sampling. Schematic representation of the study design. Each group underwent a 5-day treatment period following an overnight fast. Drug Administration (▲), administration of dapagliflozin 10 mg once daily for 5 days; PK blood sampling (↑), venous blood samples for PK assessment collected at 0 (predose), 0.33, 0.66, 1, 1.5, 2, 3, 4, 6, 8, 10, 12, 24, and 48 h after the last dose on day 5; Blood glucose monitoring (↑), venous blood samples for PD evaluation at 0 (predose) on day 1 and at 0 (predose), 0.66, 1, 1.5, 2, and 3 h following the last dose on day 5; Urine glucose excretion (↔), cumulative amount of urinary glucose excretion over 12 h before the first dosing on day 1 and over 12 h after the last dosing on day 5. Venous blood samples for PK assessment were collected at the following time points after the last dose on day 5: predose (0), 0.33, 0.66, 1, 1.5, 2, 3, 4, 6, 8, 10, 12, 24, and 48 h. Samples were collected into heparinized tubes, centrifuged at 3000 rpm for 10 min within 30 min of collection, and the plasma was aliquoted into polypropylene tubes (1 mL/tube) and stored at –70°C or lower until analysis. For PD evaluation, venous blood samples were collected at predose (0) on day 1 and at predose (0), 0.66, 1, 1.5, 2, and 3 h following the last dose on day 5. Each 2-mL sample was collected into SSTTM serum separation tubes and analyzed at the KNUH Department of Laboratory Medicine. For UGE measurements, urine was collected at predetermined intervals: for 12 h before the first dose on day 1 and for 12 h after the last dose on day 5. Determination of Plasma Concentrations of Dapagliflozin Plasma samples were subjected to liquid–liquid extraction using ethyl acetate and analyzed on a SCIEX TQ4500 system (AB SCIEX, USA) equipped with ultra-high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS, NexeraTM, Shimadzu, Kyoto, Japan). Chromatographic separation was performed using a Luna C18 column (50 × 2 mm i.d.; particle size, 3 μm; Phenomenex, Torrance, CA, USA). The mobile phase consisted of 10 mM ammonium acetate in 0.1% acetic acid (solvent A) and acetonitrile (solvent B). The gradient program delivered the mobile phase at a flow of 0.25 mL/min as follows: 40% B for 0–0.1 min; 40–100% B for 0.1–8.0 min; 100% B for 8.0–8.5 min; 100–40% B for 8.5–8.6 min; 40% B for 8.6–11.0 min. Multiple reaction monitoring transitions were used for detection, with mass-to-charge ratios of 425.968 → 167.000 for dapagliflozin and 236.990 → 193.900 for the internal standard (carbamazepine). The regression equation for the calibration curves was y = 0.00439x + 0.0014, where y is the peak area ratio of dapagliflozin to the internal standard and x is the concentration of dapagliflozin in ng/mL. Calibration curves demonstrated excellent linearity (r ≥ 0.9993) across the concentration range 2–400 ng/mL. The lower limit of quantification (LLOQ) was 2.0 ng/mL. The intra- and inter-run precision, expressed as coefficients of variation (CV%), ranged from 0.99% to 4.80% and 2.07% to 4.00%, respectively. Assay accuracy ranged from 86.75% to 95.51% for intra-run and from 85.54% to 98.02% for inter-run assessments. PK Evaluations PK parameters for dapagliflozin were estimated from individual plasma concentration-time profiles using actual sampling times via non-compartmental models (Phoenix WinNonlin, version 8.4, Certara, Inc., St. Louis, MO, USA). Parameters included the area under the plasma concentration-time curve during a dosing interval at steady state (AUCtau,ss); maximum plasma concentration during a dosing interval (τ) at steady state (Cmax,ss), trough plasma concentration at steady state (Ctrough,ss), time to reach Cmax at steady state (Tmax,ss), elimination half-life (t1/2,ss); clearance at steady state (CLss/F), and apparent volume of distribution at steady state (V/F). PD Measurement PD endpoints included the maximal serum glucose concentration on day 5 (Gmax) and the change in serum glucose concentrations from baseline (day 1, predose) to Gmax after the final dosing on day 5 (ΔGmax). UGE was determined by measuring the total cumulative urinary glucose (Ae) over two intervals: the 12 h before the first dose on day 1 [Ae(−12-0h)] and the 12 h following the final dose on day 5 [Ae(0–12h)]. Safety Assessments Safety was evaluated throughout the study by monitoring adverse events (AEs), vital signs, physical examination, and clinical laboratory parameters. Treatment-emergent adverse events (TEAEs) were defined as AEs occurring after the clinical dose of dapagliflozin and recorded for all participants who received at least one dose. Statistical Analyses As this was an exploratory study, no formal sample size calculation was performed. Instead, a minimal number of participants was considered sufficient to meet the study objectives. The target enrollment was 30 participants (15 nonobese and 15 obese), allowing for attrition while maintaining sufficient numbers assigned to meet the study objectives. Demographic data, PK parameters and PD parameters were summarized using descriptive statistics. Because of the limited sample size and possible non-normal distribution of continuous variables, the results were summarized using the mean ± standard deviation (SD), or median and interquartile range (IQR), except for Tmax,ss values, which were expressed as the median, minimum and maximum values. Differences in demographic data, PK and PD parameters between the two groups were determined as follows: (1) For the parameters that did not satisfy the equality of variances by F-test, the non-parametric Mann–Whitney U-test was used. (2) Because the sample size in the study was less than 30, the normality test using the Shapiro–Wilk test was carried out for the parameters that satisfied the equality of variances. If the data had a normal distribution, the parametric independent t-test was used to compare the differences between the two groups; otherwise, the Mann–Whitney U-test was used. The primary PK parameters were AUCtau,ss and Cmax,ss, and the primary PD parameter was ΔGmax. The geometric mean ratios (GMRs) and 90% confidence intervals (CIs) of log-transformed AUCtau,ss and Cmax,ss of dapagliflozin for the two groups (obese/nonobese) were assessed using a linear mixed-effect model. To compare categorical data that were summarized as frequencies and percentages, Fisher’s exact test was used. A two-sided value of less than 0.05 was considered to be statistically significant. As a sensitivity analysis, differences in dapagliflozin AUCtau,ss and Cmax,ss between the groups were analyzed using analysis of covariance (ANCOVA), with body weight as a covariate. Adjusted least squares mean (LSM) differences and 95% confidence intervals were calculated, and the p-values were reported to assess the effect of body weight. Because of the limited sample size, the stability and robustness of the ANCOVA analysis were examined using a non-parametric bootstrap analysis. The ANCOVA-derived estimates were repeatedly sampled to obtain multiple combinations of 1000 datasets for the bootstrap analysis. ANCOVA-derived results and median bootstrap estimates with 95% CI were then compared. Additionally, Pearson’s correlation coefficient (r) was used to establish the relationships between the PK or PD parameters and total body weight or BMI. Effect sizes were estimated using Cohen’s d with 95% confidence intervals, to quantify the practical magnitude of between-group differences for PD parameters. All statistical analyses were conducted using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA)., with significance defined as p < 0.05. RStudio (version 2026.01.0+403) was used to plot the scatterplots, quantile-quantile plots, and standardized residuals versus leverage plots. Results A total of 24 healthy Korean male participants were enrolled in this study and assigned to one of two groups based on their BMI: 15 in the nonobese group (Group A) and 9 in the obese group (Group B). One participant from Group A voluntarily withdrew consent for personal reasons before administration of the study medication. On day 4, another participant from Group A was withdrawn following a positive COVID-19 test. Consequently, 13 nonobese [median (IQR) BMI, 22.9 (2.6) kg/m2] and 9 obese [median (IQR) BMI, 31.4 (6.8) kg/m2] participants completed the study, as illustrated in the study participant flow diagram (Figure 2). Baseline demographic characteristics of the 22 participants who completed the study are summarized in Table 1. Flow diagram of the study participants. Table 1Demographic Characteristics of the 22 Subjects Who Completed This StudyCharacteristicsTotalGroup AGroup Bp value*No. of subjects22139Age, years Median (IQR)31.5 (5.0)31.0 (6.0)34.0 (3.0)0.8363† Minimum-maximum21-5521-5521-39Height, cm Median (IQR)174.7 (6.3)174.9 (6.3)174.4 (6.6)0.8487† Minimum-maximum161.6–185.3161.6–184.6166.2–185.3Weight, kg Median (IQR)78.1 (25.4)72.4 (14.3)96.1 (33.7)0.0001‡ Minimum-maximum54.2–129.754.2–85.083.1–129.7BMI, kg/m2 Median (IQR)24.9 (8.5)22.9 (2.6)31.4 (6.8)0.0001‡ Minimum-maximum18.8–41.818.8–24.930.0–41.8Notes: Data are given as the median (IQR), with minimum and maximum values. Group A (nonobese): BMI ≥ 18.0 kg/m2 and < 25.0 kg/m2; Group B (obese): BMI ≥ 30.0 kg/m2. *Compared between two groups by independent t-test†or Mann–Whitney U-test‡.Abbreviations: IQR, interquartile range; BMI, body mass index. Demographic Characteristics of the 22 Subjects Who Completed This Study Notes: Data are given as the median (IQR), with minimum and maximum values. Group A (nonobese): BMI ≥ 18.0 kg/m2 and < 25.0 kg/m2; Group B (obese): BMI ≥ 30.0 kg/m2. *Compared between two groups by independent t-test†or Mann–Whitney U-test‡. Abbreviations: IQR, interquartile range; BMI, body mass index. PK Parameters PK analysis included all 22 participants who completed the study (Group A, n = 13; Group B, n = 9). Figure 3A and Figure 3B present the mean plasma concentration-time profiles of dapagliflozin following repeated once-daily oral administration for 5 days, including the detailed early-time profile (0–3 h after the last dose). Table 2 summarizes the PK parameters after multiple oral administration of dapagliflozin 10 mg once daily for 5 days. The GMR (90% CI) values for the primary PK parameters of dapagliflozin following multiple oral administration showed that the obese group had lower AUCtau,ss and Cmax,ss than the nonobese group, as shown in Table 3. In Figure 4, whisker boxplots superimposed on the original data observations are used to graphically compare the AUCtau,ss (A) and Cmax,ss (B) in the obese and nonobese groups. One outlier from the AUCtau,ss values is displayed outside the margin of the upper whisker in Figure 4A. Mean plasma concentration–time profiles of dapagliflozin in obese and nonobese participants: (A) following multiple oral administration of dapagliflozin 10 mg once daily for 5 days, and (B) expanded view of plasma concentrations from 0 to 3 h after the last dose (nonobese, ●; obese, ○). Table 2Pharmacokinetic Parameters of Dapagliflozin Following Multiple Oral Doses of Dapagliflozin 10 mg for 5 Days in Nonobese and Obese SubjectsParametersArithmetic Mean (SD)Median (IQR)Group A (n = 13)Group B (n = 9)Group A (n = 13)Group B (n = 9)AUCtau,ss (h*ng/mL)524.1 (165.9)350.2 (76.9)505.5 (172.1)359.1 (102.9)Cmax,ss (ng/mL)177.5 (58.1)131.1 (25.0)178.0 (105.4)127.1 (27.0)Ctrough,ss (ng/mL)3.7 (2.4)0.5 (1.0)3.1 (1.9)0.0 (0.0)Tmax,ss (h)†0.66 (0.66–2.00)0.66 (0.66–2.00)t1/2, ss (h)9.3 (4.4)7.3 (3.3)7.5 (6.2)6.4 (5.0)CLss/F (L/h)21.1 (7.5)29.9 (7.1)19.8 (7.2)27.9 (8.3)V/F (L)279.3 (169.2)305.8 (148.0)206.8 (115.6)227.0 (190.5)Notes: †Data expressed as arithmetic mean (SD), and median (IQR), except Tmax,ss which is expressed as median (range)†.Abbreviations: SD, standard deviation; IQR, interquartile range; AUCtau,ss, area under the plasma concentration-time curve during a dosing interval (τ) at steady state; Cmax,ss, maximum plasma concentration at steady state; Ctrough, ss, trough plasma concentration at steady state; Tmax,ss, time to reach Cmax,ss; t1/2,ss, terminal elimination half-life at steady state; CLss/F, clearance at steady state; V/F, apparent volume of distribution. Pharmacokinetic Parameters of Dapagliflozin Following Multiple Oral Doses of Dapagliflozin 10 mg for 5 Days in Nonobese and Obese Subjects Notes: †Data expressed as arithmetic mean (SD), and median (IQR), except Tmax,ss which is expressed as median (range)†. Abbreviations: SD, standard deviation; IQR, interquartile range; AUCtau,ss, area under the plasma concentration-time curve during a dosing interval (τ) at steady state; Cmax,ss, maximum plasma concentration at steady state; Ctrough, ss, trough plasma concentration at steady state; Tmax,ss, time to reach Cmax,ss; t1/2,ss, terminal elimination half-life at steady state; CLss/F, clearance at steady state; V/F, apparent volume of distribution. Table 3Geometric Mean, Geometric Mean Ratio and 90% Confidence Intervals for the Log-Transformed AUCtau,ss and Cmax,ss of Dapagliflozin Following Repeated Oral Administration of Dapagliflozin in Healthy Adult SubjectsParametersGeometric MeanGMR (Obese/Nonobese; 90% CIs)Group A (n = 13)Group B (n = 9)AUCtau,ss (h×ng/mL)499.5342.40.6855 (0.5503–0.8538)Cmax,ss (ng/mL)168.1129.00.7673 (0.6145–0.9581)Notes: Group A (nonobese): BMI ≥ 18.0 kg/m2 and < 25.0 kg/m2; Group B (obese): BMI ≥ 30.0 kg/m2.Abbreviations: GMR, geometric mean ratio; CI, confidence interval; AUCtau,ss, area under the plasma concentration-time curve during a dosing interval (τ) at steady state; Cmax,ss, maximum plasma concentration at steady state; BMI, body mass index. Geometric Mean, Geometric Mean Ratio and 90% Confidence Intervals for the Log-Transformed AUCtau,ss and Cmax,ss of Dapagliflozin Following Repeated Oral Administration of Dapagliflozin in Healthy Adult Subjects Notes: Group A (nonobese): BMI ≥ 18.0 kg/m2 and < 25.0 kg/m2; Group B (obese): BMI ≥ 30.0 kg/m2. Abbreviations: GMR, geometric mean ratio; CI, confidence interval; AUCtau,ss, area under the plasma concentration-time curve during a dosing interval (τ) at steady state; Cmax,ss, maximum plasma concentration at steady state; BMI, body mass index. Whisker boxplots superimposed on the original data observations comparing the AUCtau,ss (A) and Cmax,ss (B) in obese and nonobese groups. The horizontal lines of the boxplot represent the first and third quartiles (Q1 and Q3, respectively) as well as the median (Q2). The borders of the whiskers, which terminate at an observed data point, are defined as 1.5 times the distance between Q1 and Q3. The outliers are the symbols beyond the whiskers. Table 4 compares the results of the ANCOVA analysis using body weight as a covariate and the bootstrap estimation. The LSM differences of AUCtau,ss and Cmax,ss between the nonobese and obese groups in the ANCOVA analysis were close to the median values of bootstrap estimates (166.7 h*ng/mL and 37.2 ng/mL, respectively), indicating the stability and robustness of the ANCOVA analysis results. Table 4The ANCOVA Least Squares Mean Difference Values for Primary PK (AUCtau,ss and Cmax,ss) and PD (ΔGmax) Parameters of Dapagliflozin and Bootstrap AnalysisANCOVA analysisBootstrapParametersLSM difference95% CIp value*Median95% CIAUCtau,ss (h*ng/mL)163.6−45.7–373.00.1183166.7−40.5–383.4Cmax,ss (ng/mL)38.4−34.2–110.90.282537.2−33.4–90.0ΔGmax (mg/dL)2.4−5.6–10.30.53772.5−4.6–9.7Notes: *p values were determined by analysis of covariance with treatment as a fixed effect and body weight as a covariate.Abbreviations: ANCOVA, analysis of covariance; LSM, least squares mean; CI, confidence interval, AUCtau,ss, area under the plasma concentration-time curve during a dosing interval (τ) at steady state; Cmax,ss, maximum plasma concentration at steady state; ΔGmax, change in serum glucose concentration from baseline (day 1) to maximum serum glucose concentration on day 5. The ANCOVA Least Squares Mean Difference Values for Primary PK (AUCtau,ss and Cmax,ss) and PD (ΔGmax) Parameters of Dapagliflozin and Bootstrap Analysis Notes: *p values were determined by analysis of covariance with treatment as a fixed effect and body weight as a covariate. Abbreviations: ANCOVA, analysis of covariance; LSM, least squares mean; CI, confidence interval, AUCtau,ss, area under the plasma concentration-time curve during a dosing interval (τ) at steady state; Cmax,ss, maximum plasma concentration at steady state; ΔGmax, change in serum glucose concentration from baseline (day 1) to maximum serum glucose concentration on day 5. PD Parameters The PD analysis included all 22 participants who completed the study (nonobese, n = 13; obese, n = 9). As summarized in Table 5, the Gmax and ΔGmax values did not differ significantly between obese and nonobese participants (p value = 0.6789 and 0.9971, respectively). The ANCOVA LSM difference (95% CI) for Gmax, using baseline glucose concentration on day 1 as a covariate, between the nonobese and obese groups was −0.5 mg/dL (−4.4–3.4) (p value = 0.8007, ANCOVA). Using body weight as a covariate, the ANCOVA LSM difference (95% CI) for ΔGmax between the nonobese and obese groups were 2.4 mg/dL (−5.6–10.3) (p value > 0.05, ANCOVA), which was around the bootstrap median value (Table 4). Table 5Dapagliflozin Pharmacodynamics Following Multiple Oral Doses of Dapagliflozin 10 mg for 5 Days in Nonobese and Obese SubjectsParametersGroup A (n = 13)Group B (n = 9)p value*Gmax (mg/dL)94.0 (12.0)96.0 (5.0)0.6789†ΔGmax (mg/dL)−1.0 (6.0)−1.0 (8.0)0.9971†Ae(−12-0h) (g)0.02 (0.03)0.03 (0.02)0.0528‡Ae(0–12h)(g)33.4 (8.9)39.3 (15.2)0.2650†Notes: Data are given as the median (IQR). *Compared between two groups by independent t-test† or Mann–Whitney U-test‡. Group A (nonobese): BMI ≥ 18.0 kg/m2 and < 25.0 kg/m2; Group B (obese): BMI ≥ 30.0 kg/m2.Abbreviations: Gmax, maximum serum glucose concentration; ΔGmax, change in serum glucose concentration from baseline (day 1) to Gmax (day 5); Ae(−12-0h), cumulative amount of urinary glucose excretion over 12 h before the first dosing on day 1; Ae(0–12h), cumulative amount of urinary glucose excretion over 12h after the last dosing on day 5; IQR, interquartile range. Dapagliflozin Pharmacodynamics Following Multiple Oral Doses of Dapagliflozin 10 mg for 5 Days in Nonobese and Obese Subjects Notes: Data are given as the median (IQR). *Compared between two groups by independent t-test† or Mann–Whitney U-test‡. Group A (nonobese): BMI ≥ 18.0 kg/m2 and < 25.0 kg/m2; Group B (obese): BMI ≥ 30.0 kg/m2. Abbreviations: Gmax, maximum serum glucose concentration; ΔGmax, change in serum glucose concentration from baseline (day 1) to Gmax (day 5); Ae(−12-0h), cumulative amount of urinary glucose excretion over 12 h before the first dosing on day 1; Ae(0–12h), cumulative amount of urinary glucose excretion over 12h after the last dosing on day 5; IQR, interquartile range. Comparative median (interquartile range) serum glucose concentration-time profiles at predose (0) on day 1 and over 3 hours on day 5 following oral dapagliflozin 10 mg once daily for 5 days in 9 obese and 13 nonobese subjects are displayed in Figure 5A. The ΔGmax values in the obese and nonobese groups are graphically compared in Figure 5B using whisker boxplots placed on the original data observations. (A) Median serum glucose concentration-time profiles in obese and nonobese participants at predose (0) on day 1 and over 3 hours on day 5 following oral dapagliflozin 10 mg once daily for 5 days. The interquartile range (IQR) is displayed by error bars for each of the median serum glucose concentrations. (B) Whisker boxplots superimposed on the original ΔGmax values in obese and nonobese groups. Safety Twenty-three participants who received at least one dose of the study drug were included in the safety analysis. The participant who was withdrawn on day 4 due to a positive COVID-19 test was included in the safety evaluation. Dapagliflozin was well tolerated, with no serious AEs reported or observed. A single AE was reported (COVID-19 positive test), which was deemed unrelated to the study drug and resolved by the follow-up visit. Discussion This study evaluated the PK and PD of dapagliflozin following multiple oral administrations for 5 days in obese and nonobese healthy adult participants, and assessed the effect of obesity on these parameters. In the main analysis of the current study, the AUCtau,ss and Cmax,ss values in obese participants were approximately 31% and 23% lower than those in nonobese participants, when body weight was not accounted for. The Tmax,ss values were similar between groups (median [range], 0.66 [0.66–2.00] h), and consistent with those from previous PK investigations (0.75 [0.50–4.00] h).21,22 After an ANCOVA analysis using body weight as a covariate was conducted as a sensitivity analysis, these exposure differences did not achieve statistical significance. Similarly, obese participants did not exhibit statistically significant differences in PD parameters compared with nonobese participants. However, given the specific condition of this exploratory study conducted in a healthy population with a small sample size, these findings should be interpreted cautiously. Accordingly, the basic diagnostic tools, such as scatterplots, quantile-quantile (Q-Q) plots, and standardized residuals versus leverage plots for primary PK and PD parameters, were employed to visually evaluate normality and identify outliers in the data, supplemented by a bootstrap analysis. As illustrated in Supplementary Figure S1, scatterplots were employed to identify the type of relationship between the participant’s weight and the primary endpoints (A, AUCtau,ss; B, Cmax,ss; C, ΔGmax). The scatterplots showed no linear relationship between participant weight and the primary endpoints. The Q-Q plots were used to determine if the standardized residuals of the primary endpoints in our study have a normal distribution, as shown in Supplementary Figure S2. The plots for AUCtau,ss and ΔGmax (A and C, respectively) followed an approximately straight line, suggesting that the data have a roughly normal distribution. In contrast, there are more points that deviate from the line in the plot for Cmax,ss (B), indicating a reduced probability of a normal distribution. None of the points are considered outliers, as the absolute values of the standardized residuals remain below 3. The standardized residuals versus leverage plots, as depicted in Supplementary Figure S3, show the leverage and influence for AUCtau,ss (A), Cmax,ss (B), and ΔGmax (C). Any point whose leverage value exceeded 2p/n (0.273 for this study), where n is data points; and p is the number of parameters, was considered as having high leverage.23 Cook’s D values lower than 0.5 indicate points that are not influential on the regression model slope coefficients.24 Accordingly, the leverage and influence assessed for the primary endpoints of this study revealed that one data point was identified as having high leverage, while no points were deemed influential. The correlations between total body weight or BMI and the PK and PD parameters were determined using Pearson’s correlation coefficient (r). Although a statistically significant correlation was found between some PK parameters (AUCtau,ss and CLss/F) and weight or BMI, Supplementary Table 1 shows that there was no moderate correlation (r > 0.6) between any of the PK parameters analyzed and weight or BMI.25 The observed CLss/F values in obese participants were 42% higher than those in nonobese participants. Clearance is primarily determined by organ blood flow and the organ’s ability to eliminate the drug. Obesity is associated with alterations in hepatic blood flow and hepatic clearance; however, its effect on renal clearance remains unclear.26 Dapagliflozin is predominantly metabolized by uridine diphosphate-glucuronosyltransferase (UGT) 1A9 in the liver to form its major metabolite dapagliflozin 3-O-glucuronide, which is primarily excreted via the kidneys.27 Previous studies suggest that obese individuals may exhibit higher clearance of drugs primarily metabolized by UGT than nonobese individuals.19,28 When the Ctrough,ss values of dapagliflozin between the two groups in our study were compared using Mann–Whitney U-test, it was significantly greater in nonobese than in obese participants (p = 0.0009). Eight participants (36.4% of 22 subjects) had Ctrough,ss values of 0 ng/mL [1 (7.7% of 13 subjects) and 7 participants (77.8% of 9 subjects) in nonobese and obese groups, respectively]. Due to the limited sample size and the prevalence of Ctrough,ss values at 0 ng/mL in our study, between-group comparisons of Ctrough,ss values were performed using Fisher’s exact test, following the categorization of Ctrough,ss values based on the presence or absence of plasma concentrations below the LLOQ (BLQ), which were substituted with 0 ng/mL. Based on the Fisher’s exact test result, the obese group had a significantly higher frequency of Ctrough,ss value with 0 ng/mL (p = 0.0015, Fisher’s exact test). The potential factors contributing to the observed difference include variations in volume of distribution values or drug metabolizing process in the two groups, linked to differences in body composition. Obese individuals possess greater adipose tissue than nonobese individuals, which may result in a larger and more variable volume of distribution for lipophilic drugs, such as dapagliflozin.17,23,29 However, given the absence of statistically significant differences in PD parameters, including Gmax and ΔGmax, between the two groups, the observed difference in Ctrough, ss is unlikely to be clinically relevant. According to the latest FDA guidelines, the study samples with concentrations listed BLQ should be reported as BLQ rather than the actual values.30 There are various methods for dealing with plasma concentrations BLQ in model-based PK analyses.31 According to the standard operating procedure for the bioanalytical method in KNUH CTC, the BLQ values in our study were replaced as zero, if they were found before Tmax. However, if the BLQ value was found after Tmax, it was evaluated as a “missing” value. In our study, eight participants (obese:nonobese = 7:1) had nine BLQ values (obese:nonobese = 8:1) prior to Tmax, while seven people (obese:nonobese = 4:3) had seven BLQ values following Tmax. It is crucial how BLQ values are handled because their contribution may significantly impact any estimation of the AUC when there are multiple BLQ values with low concentration values above them.32 Accordingly, sensitivity analyses were performed by substituting all BLQ data with either 0 ng/mL or LLOQ/2 (1 ng/mL in our study) using simple imputation to determine whether the methods of handling BLQ data produce different results.33 As shown in Supplementary Table 2, the results of the sensitivity analysis for AUCtau,ss have been compared with those of the primary analysis, including the GMRs (90% CIs) of the two groups. Since Ctrough,ss values are one of the endpoints where BLQ values are prevalent, altering BLQ values to zero or LLOQ/2 may also affect the Ctrough,ss values. Therefore, we compared the Ctrough,ss values from the primary analysis in our study with those that result from replacing all BLQ data with either 0 or 1 ng/mL, as indicated in Supplementary Table 3. The p value indicates the existence of a statistically significant difference between groups, while effect sizes provide quantitative assessments of the magnitude of actual difference between groups.34 Effect size can be measured by the standardized mean difference between two groups using Cohen’s d. In our study, the magnitude of the difference between groups showed a small effect size for both Gmax (Cohen’s d = 0.18, 95% CI: −0.67–1.03) and ΔGmax (Cohen’s d = 0.00, 95% CI: −0.85–0.85).35 The results from Pearson’s correlation coefficient (r) revealed a weak correlation between ΔGmax and weight or BMI.25 Furthermore, the serum glucose concentrations over 3 hours on day 5 were modeled using a linear mixed-effects model, with fixed effects of group, time, and baseline serum glucose concentration. Random intercepts were included to account for within-participant variability. The model was executed utilizing the lme4 package in R version 4.4.0. The LSM differences (95% CI) of serum glucose concentrations from the mixed-effects modelling indicated no significant group differences (p > 0.05) over time, as shown in Supplementary Table 4. Daily administration of dapagliflozin for 5 days was well tolerated, with no serious AEs reported. The present study has several limitations that should be acknowledged. First, as the study was intended to be exploratory, the number of the subjects enrolled in the study was relatively small. As a result, the limited sample size increases the substantial variability in the data and reduces the statistical power to detect the difference. Second, this exploratory study only included healthy, normoglycemic participants. The results of the current study should not be generalized to patients with type 2 diabetes since hyperglycemic patients may not exhibit the same pharmacodynamic effects as those in healthy subjects. Thirdly, due to a short duration of dapagliflozin administration in our study, the pharmacodynamics effects seen in our study should be interpreted cautiously. To generalize the results, more research involving diabetic patients following long-term dapagliflozin therapy in a clinical practice setting might be necessary. Fourthly, participants were selected using the BMI criteria set by World Health Organization.36 BMI-based classification may not have accurately represented true adiposity because it is unable to differentiate between lean muscle mass and fat tissue. Rubino et al recommended that obesity, defined as excess accumulation of adipose tissue, should be confirmed using either direct body fat measurement or at least one anthropometric measure (eg., waist circumference) in conjunction with BMI.37 Conclusions This study evaluated the PK and PD profiles of dapagliflozin in healthy adult male participants, classified as nonobese and obese, following 5-day multiple oral doses. The AUCtau,ss and Cmax,ss values for the obese group were 31% and 23% lower, respectively, than those of the nonobese group. The sensitivity analyses revealed no moderate correlation between the primary PK and PD parameters and weight or BMI, nor were there statistically significant differences in systemic exposure and ΔGmax between nonobese and obese group. However, given that this was an exploratory study, conducted on a healthy population with a limited sample size, this finding warrants cautious interpretation. Multiple oral doses of dapagliflozin administered over 5 days were safe and well-tolerated. Acknowledgments The study parfticipants are much appreciated by the authors. Funding Statement This study was supported by the National Research Foundation (NRF) of Korea, funded by the Korean Government (MSIT) (No. RS-2022-NR068419). Abbreviations PK, pharmacokinetic; PD, pharmacodynamic; ANCOVA, analysis of covariance; AUC, area under the plasma concentration-time curve; AUCtau,ss, AUC during a dosing interval at steady state; Cmax,ss, maximum plasma concentration at steady state; Gmax, maximum serum glucose level; ΔGmax, change in serum glucose level from baseline to Gmax on day 5; T2D, type 2 diabetes; SGLT2, sodium-glucose cotransporter 2; UGE, urinary glucose excretion; BMI, body mass index; KNUH, Kyungpook National University Hospital; CTC, clinical trial center; LLOQ, lower limit of quantification; Ctrough,ss, trough plasma concentration at steady state; Tmax,ss, the time to reach Cmax at steady state; t1/2,ss, elimination half-life at steady state; CLss/F, clearance at steady state; V/F, apparent volume of distribution at steady state; Ae, total cumulative amount of glucose excretion in urine; Ae(−12-0h), Ae over a 12-hour period before the first dose on day 1; Ae(0-12h), Ae over a 12-hour period after the last dose on day 5; AEs, adverse events; TEAEs, treatment-emergent AEs; SD, standard deviation; IQR, interquartile range; GMR, geometric mean ratio; CI, confidence interval; Q-Q plots, quantile-quantile plots; UGT, uridine diphosphate-glucuronosyltransferase; BLQ, below the LLOQ. Data Sharing Statement We, the authors, intend to share individual de-identified participant data. Young-Ran Yoon should be contacted for the sharing of the data. Disclosure The authors report no conflicts of interest regarding the content of this article. References Associated Data Data Availability Statement We, the authors, intend to share individual de-identified participant data. Young-Ran Yoon should be contacted for the sharing of the data.