To investigate the effect of a high-fat diet on pharmacokinetics/renal function/RAAS-related parameters after a single dose of empagliflozin in healthy Chinese adults

Abstract

Objectives

Food intake affects pharmacokinetic (PK) parameters of empagliflozin (EMPA). In the treatment of chronic kidney disease, EMPA in combination with renin-angiotensin-aldosterone system (RAAS) inhibitors is widely used. This study aims to investigate how a high-fat diet (HFD) affects the PK parameters, pharmacodynamic (PD) parameters related to renal function, and PD parameters related to RAAS of EMPA.

Methods

Based on a bioequivalence study of healthy Chinese adults, twenty blood sampling points were set up for each participant before EMPA and within 48 hours after 10 mg EMPA administration to measure their plasma concentrations, and then calculate their maximum plasma concentration (C_max) and area under the time-concentration curve (AUC_0~t). Urine and blood glucose, uric acid, blood urea nitrogen, serum creatinine, insulin, urine β2-microglobulin (β2-MG) and α1-MG, plasma renin concentration (PRC), angiotensin II, and aldosterone levels were tested.

Results

The 90% confidence intervals of the geometric mean ratios of the fasting C_max and AUC_0~t for the fed group were (152.29~178.77)% and (127.21~147.29)%, not within the bioequivalence range (80.00%~125.00%). When after taking EMPA compared to before taking EMPA, urine glucoses elevated significantly (fasting: (70.9 ± 29.9) mmol/L vs (0.3 ± 0.2) mmol/L: p < 0.001. fed: (89.6 ± 35.4) mmol/L vs (0.3 ± 0.3) mmol/L: p < 0.001), but β2-MG levels decreased significantly (fasting: (0.080 ± 0.065) mg/L vs (0.148 ± 0.054) mg/L: p < 0.05. fed: (0.094 ± 0.059) mg/L vs (0.145 ± 0.075) mg/L: p < 0.05). After taking EMPA, the elevated urine glucose and decreased β2-MG level exceeded the normal ranges (urine glucose: ˂ 2.8 mmol/24 h. β2-MG: 0.1 ~ 0.3 mg/L). Fasting administration of EMPA increased PRC (p < 0.05) but had no effect on aldosterone levels. Other parameters before and after EMPA administration in both groups had no significant difference and were all within the normal ranges.

Conclusions

The potential effects of long-term HFD on the PK and pharmacological actions of EMPA should be considered. Exploring the relationship between the elevated urine glucose and decreased β2-MG may have certain clinical value.

Clinical trial registration

ChiCTR2400089102, retrospectively registered in https://www.chictr.org.cn/ on 2 September 2024.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40360-026-01099-y.

Introduction

Sodium-glucose transporter 2 (SGLT2) inhibitors (SGLT2i) are a class of oral hypoglycaemic agents initially approved by the United States Food and Drug Administration (FDA) for use in patients with type 2 diabetes mellitus (T2DM) [1]. Currently, several SGLT2i have been approved for clinical use, including dagliflozin, cagliflozin, and empagliflozin (EMPA) [2]. They can exert a direct blood glucose (blood-GLU) lowering effect and this effect is not dependent on insulin [3]. EMPA is the representative medicine of SGLT2i and used mainly in the 10 mg and 25 mg dosage forms clinically [4, 5].

The transit and absorption of drugs in the gastrointestinal tract may differ depending on food intake or fasting. For EMPA itself, on one hand, it exhibits low permeability and solubility in aqueous media with a pH range of 1 to 7.5 [6], and its poor bioavailability limits its use [7]. On the other hand, food consumption produces different physiological changes, such as fluctuations in gastric and intestinal pH, delayed gastric emptying and visceral blood flow [8], can influence the efficacy of EMPA, mainly by affecting its pharmacokinetic (PK) mechanisms [9]. Managing drug-food interactions is essential for optimizing the PK parameters, safety profile, and effectiveness of EMPA.

A literature analysis showed that PK studies of EMPA are currently available in a total of six studies, three of which were conducted with 25 mg, two with 10 mg, and one with 50 mg [10–15]. Due to differences in dosage and experimental protocols, the main parameters of PK [16], including maximum plasma concentration (C_max), time to reach C_max (T_max), and area under time-concentration curve (AUC) of the above studies, exhibited distinct characteristics (Supplementary Fig. 1). These PK studies conducted in different countries have confirmed that a high-fat diet (HFD) can reduce the C_max and AUC of 25 mg and 50 mg EMPA [10–15], but the effect of a HFD on the PK parameters of 10 mg EMPA remained inconsistent across studies [13, 15].

In addition to lowering blood-GLU level, EMPA improves renal outcomes in patients with chronic kidney disease (CKD), including those with diabetic kidney disease (DKD) and non-DKD [17]. In multiple global studies initiated by the EMPA-KIDNEY Collaborative Group, the only dose of EMPA used by CKD patients was 10 mg per day [18–20]. In this context, studying how HFD affects the PK parameters of 10 mg EMPA has certain clinical significance and has become the first research objective of this manuscript.

Several studies have explored renal function-related pharmacodynamic (PD) parameters of EMPA, including blood urea nitrogen (BUN) and serum creatinine (SCr) [21]. On the other hand, in addition to controlling blood-GLU and blood pressure, blocking the renin-angiotensin-aldosterone system (RAAS) has been an important intervention for retarding DKD progression for decades [22]. A new aldosterone synthase inhibitor, used with EMPA, has been reported not only to reduce albuminuria, but also have additional efficacy in the treatment of CKD [23]. Based on a 10 mg EMPA bioequivalence (BE) study conducted by our research center, the second part of this manuscript explores how a HFD affects the PD parameters of EMPA: renal function-related PD parameters and RAAS-related PD parameters.

Materials and methods

Study design

In this study, the influence of HFD on PK/PD/RAAS parameters after EMPA administration is based on a BE study (https://www.cde.org.cn. Registration No. CTR20232603) in a randomized, open-label, two-period, two-sequence, two-way crossover study under fasting and fed conditions. The BE of two types of EMPA was evaluated: a generic 10 mg EMPA as the test drug (EMPA-T, sponsor information not allowed to be displayed according to the study protocol), and 10 mg branded Jardiance®, manufactured by Boehringer Ingelheim (Ingelheim, Germany, batch number 1705039), as the reference drug (EMPA-R).

All trial-related procedures were performed at the Phase I Clinical Trial Research Center of the Drug Clinical Trial Institution of the Shanxi Provincial People’s Hospital. The study protocol has been approved by the Ethics Committee of Shanxi Provincial People’s Hospital and approval Number 2023–045 for the PK study in the BE trial, 2023–364 and 2023–399 for the PD/RAAS study under fasting and fed conditions, respectively. This study is registered in Chinadrugtrials.org.cn (ChiCTR2400089102).

Subjects

The subjects in the BE/PK and PD/RAAS study were healthy Chinese volunteers. Before starting the study, medical and laboratory tests were conducted to ensure the health of volunteers. Smokers, heavy drinkers, those who used drugs that affected kidney metabolism within the previous 4 weeks, those who had taken any medicine within the previous 7 days, those who had a history of medication allergies, hypoglycemia or hypotension, those who had participated in previous clinical studies within the previous three months, and those with any significant clinical abnormality were excluded. In accordance with the trial protocol, 30 volunteers were recruited for the study in the fasting group and 30 in the fed group.

Empagliflozin administration

The day before the study, the participants were asked to avoid eating particular fruits (such as dragon fruit, mango, and grapefruit) and xanthine-rich foods, thus avoiding excessive physical exercise. The participants consumed a uniformly prepared dinner and began fasting from 09:00 PM, without abstaining from water.

On the day of the study, the subjects in the fasting group were administered one tablet of EMPA with 240 mL of water at 8:00 AM (study time point: 0 h). The fed group finished a standard HFD 30 min before EMPA administration (Fig. 1). Taking subject 1 as an example, he started his first bite at 7:30 AM and completed all meals within 29 min. At 8:00 AM, subject 1 took one tablet of EMPA with 240 mL water. Subsequent subjects, according to their number, at 1-min intervals, started their HFD and took EMPA. No food was allowed for 4 h after the dose, and in addition to ambulation, everyone was in a sitting position. Two hours after administration, the subjects each drank 200 mL of water.

Fig. 1.

Research flow chart. Notes: RAAS: renin-angiotensin-aldosterone system; PK: pharmacokinetics; C_max: peak concentration; AUC: area under time-concentration curve; PD: pharmacodynamics; GLU: Glucose; UA: uric acid; SCr: serum creatinine; BUN: blood urea nitrogen; eGFR: estimated glomerular filtration rate; α1-MG: α1-microglobulin; β2-MG: β2-microglobulin

The component list of HFD is shown in Supplementary Table 1. If the participant is unable to complete the HFD before receiving EMPA, he or she will withdraw from the trial.

Blood samples of BE/PK measurements

According to the current design methods of BE studies [10–15], the venous blood samples (3 mL per sample) were obtained before administering the medication (0 h) and at 0.25, 0.5, 1, 1.25, 1.5, 1.75, 2, 2.25, 2.5, 2.75, 3, 4, 6, 8, 10, 12, 24, 36, and 48 h after EMPA administration under both fasting and fed conditions. Samples were centrifuged at 2500 × g for 10 min, and the plasma was extracted and frozen at −60 °C for 60 min. Plasma concentrations of EMPA were tested by HPLC-MS/MS (Waters Xevo TQ-XS, America) in Beijing Scinovo Laboratories Pharmaceutical Technology Co., Ltd. (Beijing, China).

Blood samples of PD/RAAS measurements

According to the absorption and action characteristics of EMPA: it can be rapidly absorbed, reaching C_max in about 1.33–3.0 hours [24], followed by an increase in urinary glucose excretion by 27 times within 3 hours [25]. The blood samples were collected three times to measure the PD and RAAS-related parameters. The first time in the fasting group was carried out from 7:30 AM to 7:50 AM on the day of the study (i.e., before EMPA administration). In the fed group, the first collection time was conducted from 7:00 AM to 7:20 AM, that is, before the HFD and EMPA administration. The second and the third time was after 2 h and 4 h of taking EMPA, respectively (study time points: 2 h and 4 h). All blood samples were centrifuged at 3800 rpm at 4 °C for 8 min. The protocol is illustrated in Fig. 1.

Blood-GLU levels were measured using a Beckman Coulter AU5800 fully automatic biochemical analyzer (Beckman Coulter International Trading Co., LTD, USA). Insulin levels were analyzed using chemiluminescent immunoassay (Roche Cobase 601; Roche Diagnostics GmbH, Germany). Renal function-related parameters, including BUN, SCr, and blood uric acid (blood-UA) and RAAS-related parameters, including plasma renin concentration (PRC), angiotensin II (Ang II), and aldosterone (ALD), were measured using AutoLumo A2000 Plus (Zhengzhou Antu Biological Engineering Co. LTD, China). eGFR was determined using the following formula [26].

Urine samples of PD measurements

Urine samples were collected twice. The first time was after waking up at 6:00 AM and before 8:00 AM on the study day (study period: −2~0 h). The second time was within 4 h of EMPA administration (study period: 0 ~ 4 h). Urine from each subject was collected separately on his own in a clean 1000 mL cylinder. Urine samples (8 mL) were tested for urine glucose (urine-GLU) and urine uric acid (urine-UA) using a Beckman Coulter AU5800 fully automatic biochemical analyzer (Beckman Coulter International Trading Co., LTD, USA). Urine samples (8 mL) were centrifuged at 1500 rpm at room temperature for 5 min, and the levels of urine α1-microglobulin (α1-MG), and β2-microglobulin (β2-MG) were measured using an automatic specific protein analyzer, BIOSYSTEM BA400 (Biosystems S.A. Costa Barcelona, Spain).

Safety observation

Potential adverse events (AEs) and vital signs (systolic and diastolic blood pressure, body temperature, and pulse rate) were assessed throughout the study. AEs were described according to the “Common Terminology Criteria for Adverse Events (version 2025”). Researchers recorded AEs in terms of their seriousness, intensity, time course, outcome, and relationship with the study drug.

Statistical analysis

PK parameters included C_max, T_max, and area under the time-concentration curve from administration to the last observed concentration at time t (AUC_0~t), area under the time-concentration curve extrapolated to infinity (AUC_0-∞), percentage of the residual area of the AUC (AUC_{_%Extrap}), terminal half-life (t_1/2), and elimination rate constant (λ_z). The PK study employed a non-compartmental method using Statistical Analysis System version 9.4. Blood concentration-time data were collected after fasting or fed administration. The PK parameters were statistically analyzed. The AUC_0~t for EMPA was calculated using the trapezoidal method. If the 90% CI for the geometric mean ratio (GMR) of AUC_0~t, AUC_0~∞, and C_max fell within the statistical range of 80.00% to 125.00% as proposed by the NMPA, then the fasting and fed conditions are considered bioequivalent [27].

The PD/RAAS statistical analysis was performed using SPSS 22.0. Due to the small sample size (n < 30), the Shapiro-Wilk test was used to analyze all variables for normal distribution. Data that conformed to a normal distribution were analyzed using an independent samples t-test, while data that did not conform were analyzed using the Mann-Whitney U test. All continuous variable data were expressed as the mean ± standard deviation. All reported p values were two-tailed, and statistical significance was set at p < 0.05.

Results

The participants included in the analysis

Of the 30 volunteers enrolled in the fasting group, one subject withdrew due to syncope before the first period of administration, and the other two subjects withdrew at the second period. In the fed group, one voluntarily withdrew after the end of the first period and one did not complete HFD at the second period.

According to the sensitivity analysis, the plasma concentrations of the participants included in the PK parameter set and the BE set for analysis are as follows: EMPA-R (n = 28) and EMPA-T (n = 28) in the fasting group and EMPA-R (n = 29) and EMPA-T (n = 29) in the fed group (Table 1).

Table 1.

Participants

Situations of the participants	Fasting group	Fed group
The enrolled participants	n = 30	n = 30
The participants who withdrew or dropped out	EMPA-T and EMPA-R: n = 1
	EMPA-T: n = 1	EMPA-T: n = 1
	EMPA-R: n = 1	EMPA-R: n = 1
The participants who included in the analysis	EMPA-R: n = 28	EMPA-R: n = 29
	EMPA-T: n = 28	EMPA-T: n = 29

Notes: EMPA-R, a branded empagliflozin was as the reference. EMPA-T, a generic empagliflozin was the test

Our research confirmed that the 90% CI for the GMR of AUC_0~t, AUC_0~∞, and C_max is within the statistical range of 80.00%~125.00% when 10 mg EMPA-T compared to EMPA-R under fasting (Supplementary Table 2) and fed conditions (Supplementary Table 3).

In the study of the effect of HFD on PK parameters, the plasma concentrations included in the analysis are as follows: fasting group (n = 56, EMPA-R (n = 28) and EMPA-T (n = 28)) and fed group (n = 58: EMPA-R (n = 29) and EMPA-T (n = 29)) according to the sensitivity analysis. Table 1 lists these details.

In the PD/RAAS study, 42 healthy subjects have signed an informed consent form (fasting, n = 19; fed, n = 23) and completed the entire study.

The influence of HFD on plasma concentrations

Table 2 shows plasma concentrations at 20 blood sampling points before and 48 h after EMPA administration. A total of 16 time points had higher concentrations in the fasting than in the fed from 0.25 h to 12 h (all: p < 0.01). At 0.25 h after dosing, the plasma concentration ratio of fasting to fed was highest at 3.3-fold, followed by 3.0-fold at 0.5 h and 2.3-fold at 1 h.

Table 2.

Comparisons of plasma concentrations

Serial number	Time points of sampling (h)	Plasma concentrations (ng/mL)				CV of the plasma concentrations (%)
		Fasting (n = 56)	Fed (n = 58)	P value	Fasting/fed	Fasting (n = 56)	Fed (n = 58)
1	0	BQL	BQL	/	/	/	/
2	0.25	26.9 ± 31.9	8.2 ± 13.4	<0.001*	3.28	118.45	162.73#
3	0.5	102.6 ± 61.0	34.2 ± 31.0	<0.001*	3.00	59.44	90.61#
4	1	165.7 ± 60.6	72.9 ± 49.5	<0.001*	2.27	36.59	67.83#
5	1.25	167.5 ± 44.1	87.4 ± 39.2	<0.001*	1.92	26.34	44.90#
6	1.5	163.2 ± 35.6	95.7 ± 33.7	<0.001*	1.71	21.83	35.25#
7	1.75	157.9 ± 32.4	94.1 ± 28.1	<0.001*	1.68	20.49	29.86#
8	2	151.8 ± 27.3	94.2 ± 26.3	<0.001*	1.61	18.02	27.88#
9	2.25	148.2 ± 28.2	91.5 ± 22.6	<0.001*	1.62	19.02	24.71#
10	2.5	138.0 ± 21.9	85.3 ± 18.8	<0.001*	1.62	15.86	22.05#
11	2.75	130.5 ± 18.5	83.9 ± 16.0	<0.001*	1.55	14.18	19.04#
12	3	123.3 ± 17.9	79.4 ± 15.0	<0.001*	1.55	14.55	18.86#
13	4	104.4 ± 16.4	74.4 ± 11.1	<0.001*	1.40	15.73	14.92
14	6	72.0 ± 12.5	53.1 ± 8.8	<0.001*	1.36	17.39	16.57
15	8	52.6 ± 9.1	40.5 ± 7.6	<0.001*	1.30	17.28	18.77#
16	10	40.4 ± 8.2	32.7 ± 6.7	0.002*	1.24	20.42	20.60#
17	12	30.7 ± 6.5	25.4 ± 4.9	0.004*	1.21	21.20	19.43
18	24	10.1 ± 3.0	9.3 ± 1.8	0.281	1.09	29.27	19.63
19	36	3.5 ± 1.4	3.2 ± 1.2	0.483	1.09	39.44	35.97
20	48	2.2 ± 0.9	2.2 ± 0.8	0.839	1.03	40.19	34.66

Notes: *p < 0.01, plasma concentration of the fasting vs the fed. ^#CV of the fed was higher than that of the fasting. CV, coefficient of variation

Within 3 h after administration, the plasma concentration variability in the fed group, expressed as the coefficient of variation (CV), was higher than that in the fasting group, suggesting that the HFD may affect the absorption of EMPA among individuals [28].

The influence of HFD on PK parameters

The calculated PK parameters, including T_max, C_max, AUC_0~t, AUC_0~∞, AUC_{_%Extrap}, t_1/2 and λ_z are presented in Table 3.

Table 3.

Comparisons of pharmacokinetic parameters

Parameters	Units	Geometric mean and ratio			90%CI of the ratio (%)
		Fasting	Fed	Fasting/Fed (%)
Tmax	h	1.31	1.67	78.44	49.05~107.46
Cmax	ng/mL	186.70	112.79	165.53	152.29~178.77
AUC0-t	ng·h/mL	1305.15	950.93	137.25	127.21~147.29
AUC0-∞	ng·h/mL	1331.45	977.89	136.16	125.86~146.45
AUC_%Extrap	%	1.80	2.57	70.04	44.06~96.16
t1/2	h	8.18	8.34	98.08	85.81~110.26
λz	1/hr	0.08	0.08	100.00	96.41~107.58

Notes: C_max, maximum plasma concentration. T_max, time to reach C_max. AUC_0~t, area under time-concentration curve from administration to last observed concentration at time t. AUC_0-∞, area under time-concentration curve from administration extrapolated to infinity. AUC__%Extrap, percentage of residual area of AUC. t_1/2, terminal half-life. λ_z, elimination rate constant. CI, confidence interval

The GMR for T_max, C_max, AUC_0~t, AUC_0~∞ and AUC_{_%Extrap} between the fasting and the fed were 78.44%, 165.53%, 137.25%, 136.16% and 70.04%, respectively, and their corresponding 90% CI were (49.05~107.46)%, (152.29~178.77)%, (127.21~147.29)%, (125.86~146.45)% and (44.06~96.16)%, respectively, which were not within the bioequivalent range (80.00~125.00)% [29, 30], suggesting that HFD may delay T_max and decrease C_max and AUC of 10 mg EMPA.

The influence of HFD on PD parameters

Before taking EMPA, the median time for urine sample collection was 6:44 AM for the fasting group and 6:33 AM for the fed group. After taking the medication, the median time for urine collection was 10:50 AM for the fasting group and 10:59 AM for the fed group.

As shown in Fig. 2A, when after taking EMPA compared to before taking EMPA, urine-GLU was significantly higher (fasting: (70.9 ± 29.9) mmol/L vs (0.3 ± 0.2) mmol/L: p < 0.001. fed: (89.6 ± 35.4) mmol/L vs (0.3 ± 0.3) mmol/L: p < 0.001). Additionally, within 4 h after EMPA administration, the fed group exhibited a significantly higher urine-GLU level than the fasting group (p < 0.05).

Fig. 2.

Effects of high-fat diet on pharmacodynamics parameters of empagliflozin in healthy subjects. (A) Urinary glucose (Urine-GLU) levels. (B) Urinary uric acid (Urine-UA) levels. (C) Insulin levels. (D) Blood glucose (Blood-GLU) levels. (E) Blood uric acid (Blood-UA) levels. (F) Serum creatinine (SCr) levels. (G) Blood urea nitrogen (BUN) levels. (H) Estimated glomerular filtration rate (eGFR) levels. (I) α1-microglobulin (α1-MG) levels. (J) β2-microglobulin (β2-MG) levels. Notes: Data are presented as mean ± standard deviation. *p < 0.05, in the fasting group, post-dose (2 h and 4 h) vs. pre-dose (−2~0 h or 0 h/-1 h). #p < 0.05, in the fed group, post-dose (2 h and 4 h) vs. pre-dose (−2~0 h or 0 h/-1 h). 0 h/-1 h: before empagliflozin administration. 2 h: 2 h after empagliflozin administration. 4 h: 4 h after empagliflozin administration. −2~0 h: 2 to 0 h before empagliflozin administration. 0 ~ 4 h: 0 to 4 h after empagliflozin administration. Fasting group: n = 19. Fed group: n = 23. Urine samples: green indicates the fasting group and yellow indicates the fed group. Blood samples: red indicates the fasting group and blue indicates the fed group

Figure 2C shows that in the fasting group, blood-GLU levels significantly decreased after 2 h ((4.8 ± 0.3) mmol/L) and 4 h ((4.7 ± 0.3) mmol/L) of EMPA administration when compared to before EMPA administration ((5.1 ± 0.4) mmol/L, p < 0.05). However, the two decreased fasting blood-GLU levels were still within the normal range ((4.0 ~ 6.0) mmol/L).

Figure 2D shows that in the fasting group, the trend of insulin levels before and after EMPA administration was similar to that in Fig. 2C. Compared with the insulin level before EMPA administration ((7.0 ± 3.7) mU/L), insulin levels significantly decreased after 2 h ((5.0 ± 2.2) mU/L) and 4 h ((4.7 ± 3.2) mU/L) of EMPA administration (p < 0.05).

Compared with EMPA administration, significant decrease in blood-UA ((342.6 ± 57.6) μmol/L vs (369.2 ± 62.9) μmol/L: p < 0.05) and a significant increase in urine-UA ((2751.5 ± 1018.1) mmol/L vs (2223.8 ± 891.1) mmol/L: p < 0.05) after 4 h of EMPA administration could be observed in the fed group in Fig. 2B and E.

The results for SCr, BUN, and eGFR are shown in Fig. 2F~H. In the fed group, 4 h after taking EMPA, compared with before taking EMPA, there had significant increase on SCr level ((71.5 ± 9.7) μmol/L vs (66.0 ± 8.4) mmol/L: p < 0.05) and BUN level ((5.5 ± 1.0) mmol/L vs (4.4 ± 1.0) mmol/L: p < 0.05), but significant decrease on eGFR level ((128.3 ± 20.7) ml/min/1.73 m² vs (141.6 ± 22.0) ml/min/1.73 m²: p < 0.05)). EMPA did not cause the above three renal function-related parameters to exceed their normal ranges (SCr: (57.0~97.0) μmol/L. BUN: (2.3~7.0) mmol/L. eGFR: ≥ 90.0 ml/min/1.73 m²).

The results for α1-MG are shown in Fig. 2I. In the fasting group, α1-MG level significantly decreased after EMPA administration compared to that before EMPA administration ((1.9 ± 1.8) mg/L vs (2.9 ± 1.7) mg/L: p < 0.05). However, after taking EMPA post-meal, the α1-MG level did not show a significant difference.

The results of β2-MG can be seen in Fig. 2J. When after taking EMPA compared to before taking EMPA, β2-MG levels decreased significantly (fasting: (0.080 ± 0.065) mg/L vs (0.148 ± 0.054) mg/L: p < 0.05. fed: (0.094 ± 0.059) mg/L vs (0.145 ± 0.075) mg/L: p < 0.05). Additionally, the decreased β2-MG level exceeded the normal range (0.1~0.3 mg/L).

The influence of HFD on RAAS-related parameters

Figure 3A shows that in the fasting group, PRC levels were significantly higher at 2 h ((52.2 ± 28.2) pg/mL) and 4 h ((51.5 ± 23.7) pg/mL) after EMPA administration compared to before EMPA administration ((37.3 ± 25.8) pg/mL) (p < 0.05), and these elevated PRC levels exceeded the normal range (4.0~38.0 pg/mL).

Fig. 3.

Effects of high-fat diet on RAAS-related parameters of empagliflozin in healthy subjects. (A) Plasma renin concentration (PRC) levels. (B) Angiotensin II (Ang II) levels. (C) Aldosterone (ALD) levels. Notes: Data are presented as mean ± standard deviation. *p < 0.05, in the fasting group, post-dose (2 h and 4 h) vs. pre-dose (0 h). ^#p < 0.05, in the fed group, post-dose (2 h, and 4 h) vs. pre-dose (0 h). 0 h: before empagliflozin administration; 2 h: 2 h after empagliflozin administration. 4 h: 4 h after empagliflozin administration. Fasting group: n = 19; fed group: n = 23. Green indicates the fasting group and yellow indicates the fed group

Figure 3A also shows that in the fed group, PRC initially increased and then decreased after taking EMPA, ultimately resulting in no significant difference in PRC 4 h after taking EMPA compared to that before taking EMPA ((30.4 ± 18.9) pg/mL) vs (32.5 ± 22.4) pg/mL): p = ns).

Like PRC in the fasting group, Ang II level in the fasting group significantly increased after EMPA administration compared to that before taking EMPA ((88.2 ± 10.3) pg/mL vs (81.9 ± 9.9) pg/mL: p < 0.05). Not like PRC in the fed group, after EMPA administration, Ang II level in the fed group significantly decreased when compared to that before EMPA administration ((82.6 ± 23.3) pg/mL vs (91.6 ± 26.7) pg/mL: p < 0.05). Anyway, the changed Ang II level was still in its normal range ((49.0~252.0) pg/mL). The above results are presented in Fig. 3B.

ALD levels are shown in Fig. 3C. There had no significant differences when after EMPA administration compared to that before taking EMPA, fasting and fed group.

Safety observation

All the AEs were described in the instructions for Jardiance® and were Grade 1 in severity, recovered spontaneously without any intervention, and no subject withdrew from the trial due to AEs (Supplementary Table 4).

Discussion

Lifestyle, including exercise and diet, can alter several physiological functions of the human body. Food intake can cause various physiological changes in the human gastrointestinal tract, such as changes in the gastrointestinal pH, gastric emptying time, and metabolic enzymes [9]. These changes can affect the PK characteristics of certain drugs by altering their release, absorption, distribution, metabolism, and/or excretion [31]. PK and PD interactions with food can lead to reduced drug efficacies. This is known as the negative food effect, which refers to a reduction in bioavailability when a drug is ingested [32].

What is the effect of food on the PK parameters of EMPA? We first analyzed four studies among the existing six PK studies that used warm water to administer EMPA. The results of three studies showed that HFD decreased both C_max and AUC, which were conducted at doses of 10, 25, and 50 mg, whereas another study on 10 mg showed the opposite result, that is, HFD increased both C_max and AUC of EMPA [10–15] (Supplementary Figure 1).

The HFD used in the aforementioned studies refers to foods with a total calorie intake of (800~1000) kcal, and calories from fat should provide (500~600) kcal [27]. It has been confirmed that a HFD has more significant physiological impacts on the gastrointestinal tract, therefore regulatory authorities, including FDA and China National Medical Products Administration, have clearly state that whether it is the impact of food on a single medicine or on multiple medicines used in combination, high-calorie and high-fat meals are the recommended type of food [33].

Our study showed that a HFD delayed the T_max of 10 mg EMPA while reducing C_max and AUC, which was consistent with most previous studies [10–14]. As for the only paper that was inconsistent with the above results, its authors provided the following explanation: this discrepancy might be attributed to individual differences, ethnicity, environmental factors, and dietary habits [15].

In healthy subjects, drug-drug interactions between EMPA and other hypoglycemic medicines such as evogliptin [34], metformin and lobeglitazone [35] were explored based on the PK properties of EMPA. The effect of food on PK parameters often begins in phase I clinical trials, based on observations in healthy subjects [36]. With increasing clinical evidence of EMPA in patients with non-DKD [17], focusing on the direct effect of EMPA on renal function-related indicators in healthy people may lay a foundation for further exploration of its improvement in patients with CKD.

On the other hand, the data in healthy subjects are limited, as shown in our current study, where a single HFD had a significant effect on the PK parameters of a single 10 mg EMPA, but HFD did not affect EMPA on renal function-related PD parameters in the short term. Given the close relationship between PK and PD [9], and considering that the unique advantage of SGLT2i lies in improving long-term renal outcomes [37], subsequent protocols should incorporate varied dietary conditions and extended EMPA dosing periods to elucidate their effects on both PK parameters and renal function-related PD parameters.

In our study, two parameters exceeded the normal range, namely elevated urine-GLU and decreased β2-MG. Both α1-MG and β2-MG are regarded as markers of renal tubular impairment [38, 39]. Decreasing effects of EMPA on α1-MG were observed in patients with acute decompensated heart failure [40], patients with renal proximal tubulopathy [41] and patients with DKD [39]. Previous studies only observed the effect of EMPA in reducing α1-MG and β2-MG in disease states, but this study also observed this effect in healthy volunteers. Is there a connection between elevated urine-GLU and decreased β2-MG after EMPA administration? A study on how EMPA affects broad biological systems through proteomics suggested a role for β2-MG [42], but further exploration of the specific details is needed.

Augmenting RAAS inhibitors with new drug classes, such as SGLT2i, has the potential to improve clinical outcomes in a broad range of patients with CKD [43]. Focused on the effect of EMPA on RAAS axis in the dietary state, our results suggested that taking EMPA on an empty stomach in healthy subjects increased the PRC level but ultimately had little effect on the ALD level. If EMPA was administered after a meal, the three indicators on the RAAS axis remained within the normal range. This suggested that HFD might play a limited role on the effect of EMPA on RAAS parameters.

In our study, we also observed the previously reported decrease in blood-UA [44] and its correlation with increased urine-UA [45] and urine-GLU after EMPA administration [46]. A balance between the increase in hepatic glucose production and the increase in urine-GLU excretion immediately after SGLT2 inhibition has been observed in a clinical study [47]. A similar balance between the increase in urine-GLU excretion and the increase in blood-GLU after a meal in the fed group was also observed in our study.

The glucose-lowering mechanism of EMPA ensures it have clinical safety [48]. We observed that a significant reduction in blood-GLU and insulin simultaneously in the fasting group at the corresponding time points. This might prevent the induction of hypoglycemia [49] and the safety of EMPA was confirmed by the blood-GLU test results within 4 h of taking EMPA under fasting condition, which was to maintain the reduced blood-GLU within the normal range [50].

There still have some limitations of this study: 1. This study was an extension of a BE study, and all participants were healthy individuals, which differs from the physiological characteristics of patient populations. 2. The protection of EMPA on CKD may be achieved by natriuresis and changes in fluid balance. We did not compare urine volume and sodium ions after taking EMPA in a fasting state and after a HFD. We hope that we could improve these limitations in our future work.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization: Yi Jin and Hui Liao; Data curation: Yi Jin and Lu Bai; Formal analysis: Yi Jin and Wenyan Zhao; Funding acquisition: Hui Liao; Investigation, Yi Jin, Wenyan Zhao, Hui Wu and Lu Bai; Methodology, Sunqi Ding and Hui Liao; Project administration: Hui Liao; Resources, Zhaodi Han and Hui Liao; Software, Qian Li and Shuangshuang Tian; Supervision, Hui Liao; Validation, Qian Li and Zhaodi Han; Visualization, Yi Jin and Hui Wu; Writing – original draft: Wenyan Zhao,Sunqi Ding, Qian Li, Shuangshuang Tian and Hui Liao; Writing – review and editing: Wenyan Zhao and Hui Liao.

Funding

This study was supported by Demonstration Project on Reformation and Quality Development of Public Hospitals (No.SCP-2023-8), the Local Science and Technology Development Funds Projects Guided by Central Government (No. YDZJSX2021C027), Basic Research Program of Shanxi Province (No.202103021224370).

Data availability

Some datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Ethics Committee of Shanxi Provincial People’s Hospital (BE protocol code: 2023-045, approval date: 27-7-2023, PD protocol code: 2023–364 (fasting) and 2023–399 (fed), approval date:29-8-2023 (fasting) and 28-9-2023 (fed)).

Informed consent

Informed consent was obtained from all participants involved in the study. Written informed consent has been obtained from the participants to publish this paper.

Competing interests

The authors declare no competing interests.