1 Introduction

Diabetes mellitus is becoming a global epidemic, with the number of adults with diabetes expected to rise from approximately 285 million in 2010 to 439 million by 2030 [1]. The complications of diabetes, such as cardiovascular disease, neuropathy, retinopathy and nephropathy, are also expected to increase. Consequently, diabetes is placing an ever-growing burden on society and healthcare systems worldwide. Glycaemic control is a key factor in the management of type 2 diabetes, with the International Diabetes Federation recommending a target glycosylated haemoglobin (HbA1c; Diabetes Control and Complications Trial aligned) level of <6.5 % [2], while the American Diabetes Association/European Association for the Study of Diabetes recommend a target of <7.0 %, although more or less stringent targets may be appropriate for selected patients [3]. Nevertheless, despite continuing advances in pharmacotherapy, many patients do not achieve the target HbA1c. Many of the existing pharmacotherapies for type 2 diabetes are associated with adverse effects such as weight gain, hypoglycaemia and gastrointestinal problems that may limit their use in some patients, or do not provide adequate glucose lowering, meaning multiple pharmacotherapies are required [3].

Sodium-dependent glucose cotransporter-2 (SGLT2) inhibitors are a novel and promising class of oral hypoglycaemic agents. SGLT2 inhibitors work by suppressing glucose re-absorption in the proximal tubule of the kidney, causing the excretion of glucose in the urine. They are proposed as an insulin-independent approach for the treatment of type 2 diabetes [4, 5]. Ipragliflozin (Suglat®) is a potent and selective SGLT2 inhibitor with a C-glycoside structure, discovered in joint research by Astellas Pharma Inc. and Kotobuki Pharmaceutical Co. Ltd [6]. Its chemical structure is (1S)-1,5-anhydro-1-C-{3-[(1-benzothiophen-2-yl)methyl]-4-fluorophenyl}-d-glucitol-(2S)-pyrrolidine-2-carboxylic acid (1:1) (Fig. 1), and its molecular weight as ipragliflozin l-proline is 519.58. The log D value was 2.3–2.4 (37 °C, 1-octanol/water) at pH 1, 3, 5, 7, 9 and 11 [6]. The pKa value was not apparent in a pH range of 1–12. Phase III studies have shown that ipragliflozin as monotherapy [7, 8] or in combination with other antidiabetic drugs [8–11] significantly reduced HbA1c, fasting plasma glucose (FPG), and bodyweight in Japanese patients with type 2 diabetes [8–11]. Based on the positive results of phase III studies, ipragliflozin was recently launched in Japan for the treatment of type 2 diabetes.

Fig. 1
figure 1

Structure of ipragliflozin and its main metabolite M2

Full size image

The aim of this review is to provide a comprehensive overview of the pharmacokinetic and pharmacodynamic properties of ipragliflozin, examining data obtained from studies in healthy subjects, patients with type 2 diabetes and special populations. We also report the results of a thorough QT study, and drug–drug interaction and food effect studies. We performed electronic literature searches of PubMed to identify relevant English-language articles using the keywords ‘ipragliflozin’ and ‘ASP1941’, without date limits. We also searched relevant scientific congresses and our clinical study database. All publications reporting pharmacokinetic and/or pharmacodynamic data on ipragliflozin/ASP1941 in humans were considered for this review.

2 Clinical Pharmacokinetics

2.1 Absorption, Distribution, Metabolism and Excretion (ADME) Properties of Ipragliflozin

To examine the ADME characteristics of ipragliflozin, a single-centre open-label study was conducted in six healthy male volunteers. All subjects were hospitalised until day 7 after drug administration. Blood, plasma, urine, faeces and expired air samples were collected for the analysis of [14C] radioactivity, the study drug and/or for metabolic profiling [12]. After a single oral dose of 100 mg ipragliflozin containing 1.8 MBeq [14C]-labelled drug, means of 67.9 and 32.7 % of [14C] radioactivity were excreted in urine and faeces, respectively. The majority of radioactivity in urine was excreted as metabolites, with ≤1 % of the dose excreted as unchanged ipragliflozin. Metabolite profiling demonstrated that ipragliflozin was the main component in plasma. The plasma protein binding of 14C-ipragliflozin (0.05–200 μg/mL) was almost constant within the concentration range examined, from 94.6 to 96.5 %. The main plasma binding protein was albumin. Furthermore, metabolites M2 (2′-O-β-glucuronide of ipragliflozin) and M4 (3′-O-β-glucuronide of ipragliflozin) were formed, with M2 being the major metabolite (Fig. 1). The ratios of area under the plasma-concentration curve (AUC) to the total AUC of all analytes were 54 and 27 % for the unchanged drug and the M2 metabolite in plasma, respectively. In urine, the metabolites M1 (2′-O-β-glucuronosyl-6-hydroxide of ipragliflozin), M2, M3 (6′-O-β-glucuronide of ipragliflozin), M4, M6 (6′-O-sulphate of ipragliflozin) and M7 (S-oxide of ipragliflozin) were detected. The major metabolite in urine was M2, with metabolite profiling revealing that 0.2 and 33.5 % of the dose was excreted in urine as unchanged drug and the M2 metabolite, respectively. None of the metabolites of ipragliflozin is pharmacologically active. In faeces, ipragliflozin was detected as the main component, and the major metabolite was M3 [13]. Because glucuronide metabolites were detected in humans, a study was performed to determine which UDP-glucuronosyltransferase (UGT) isozymes are responsible for hepatic glucuronidation of ipragliflozin in humans. This study revealed that multiple enzymes are involved, with UGT2B7 playing a major role and UGT2B4 and UGT1A9 potentially having minor roles [14].

2.2 Pharmacokinetics in Healthy Subjects

2.2.1 Pharmacokinetics in Healthy Japanese Subjects

To examine the pharmacokinetics of ipragliflozin in healthy Japanese subjects, a phase I, randomised, placebo-controlled trial was conducted. Forty-eight subjects (36 active drug and 12 placebo) received a single oral dose (1–300 mg) of ipragliflozin and 36 subjects (24 active drug and 12 placebo) received repeated once-daily oral doses (20–100 mg) for 7 days [15].

The plasma pharmacokinetic parameters of ipragliflozin in the single- and multiple-dose groups are shown in Table 1. After a single oral dose under fasting conditions, ipragliflozin was rapidly absorbed and its concentration decreased in a biphasic manner. The mean apparent terminal elimination half-life (t 1/2) was 10.0–13.3 h after the first dose of ipragliflozin at doses ≥3 mg. Maximum plasma concentration (C max) and the AUC from time zero to infinity (AUC∞) increased in a dose-dependent manner after a single dose of ipragliflozin. The plasma ipragliflozin concentration appeared to reach steady state 3 days after starting multiple oral dosing. After multiple dosing, the mean t 1/2 varied from 11.2 to 15.4 h. Ipragliflozin did not significantly accumulate following 7-day multiple dosing compared with a single dose. The median time to reach C max (t max) was consistent across the dose groups. The urinary excretion ratio (fraction of the dose excreted into urine [Ae%]) of unchanged ipragliflozin was low at all doses, varying from 0.65 to 1.18 % after a single dose, and from 1.29 to 1.44 % after multiple doses. The renal clearance (CLR) of ipragliflozin was approximately the same at all doses, varying from 0.09 to 0.13 L/h after a single dose and from 0.14 to 0.15 L/h after multiple doses.

Table 1 Pharmacokinetic parameters in healthy Japanese subjects following single and multiple oral doses of ipragliflozin
Full size table

2.2.2 Pharmacokinetics in Healthy Non-Japanese Subjects

The pharmacokinetics of ipragliflozin in healthy non-Japanese subjects was assessed in a multiple, ascending-dose, double-blind, placebo-controlled trial conducted in 48 healthy male and female volunteers (36 receiving active drug and 12 receiving placebo) [16]. The study subjects were administered a single oral dose (1–600 mg) of ipragliflozin or repeated once-daily oral doses (5–600 mg) for 10 days. Ipragliflozin was rapidly absorbed and the median t max among all ipragliflozin dose groups varied from 1.0 to 1.5 h after the first single dose, and from 1.0 to 2.3 h after the last dose in the multiple-dose study. After reaching the C max, the plasma concentration of ipragliflozin decreased in a biphasic manner with mean t 1/2 values varying from 10.2 to 13.5 h after a single dose, and from 11.2 to 14.6 h after multiple doses. C max and AUC increased in a dose-dependent manner after both single and multiple doses of ipragliflozin. Plasma ipragliflozin concentration appeared to reach steady state 3 days after starting multiple oral dosing. No significant accumulation was observed after the 10-day multiple dosing. The mean peak/trough ratio was approximately 10–20 across all dose groups. Approximately 1 % of the ipragliflozin dose was excreted into the urine as the parent compound at all dose levels. The CLR of ipragliflozin was approximately the same at all doses, varying from 0.12 to 0.15 L/h after a single dose and from 0.13 to 0.17 L/h after multiple doses.

2.3 Pharmacokinetics in Patients with Type 2 Diabetes

2.3.1 Pharmacokinetics in Japanese Patients with Type 2 Diabetes

The pharmacokinetics of ipragliflozin in patients with type 2 diabetes was evaluated in a multicentre, randomised, placebo-controlled, double-blind study [17]. In this study, patients were treated with placebo, 50 or 100 mg ipragliflozin once daily for 14 days. Patients were drug-naïve or receiving monotherapy, aged 20–74 years, diagnosed with diabetes for ≥12 weeks, body mass index (BMI) 20–45 kg/m2, and HbA1c 7.4–10.5 %. Subjects receiving oral antidiabetic drugs underwent a 13-day washout period before starting ipragliflozin. In total, 30 subjects were involved in the study (20 receiving active drug and 10 receiving placebo). Two subjects discontinued treatment because of treatment-emergent adverse events (one each treated with 50 and 100 mg ipragliflozin). Therefore, pharmacodynamic and pharmacokinetic data were analysed for 28 subjects and safety data were analysed for 30 subjects. Ipragliflozin was rapidly absorbed in both dose groups, reaching peak concentrations within 1 h after dosing. The mean plasma C max and AUC24 were approximately 1.7 and 1.9 times higher, respectively, in the 100-mg group than in the 50-mg group. T max, t 1/2, and oral clearance were similar in both groups. In terms of urinary pharmacokinetics, the amount of drug excreted in urine was approximately two times higher in the 100-mg group than that in the 50-mg group. Ae% was very low in both ipragliflozin groups. Overall, the pharmacokinetics in Japanese patients with type 2 diabetes was similar to that previously reported in healthy Japanese volunteers [15].

2.3.2 Pharmacokinetics in Non-Japanese Patients with Type 2 Diabetes

Schwartz et al. [18] assessed the pharmacokinetic profile of ipragliflozin in non-Japanese patients with type 2 diabetes. In this single-centre, randomised, placebo-controlled trial, 61 patients (49 active drug and 12 placebo) were randomised to receive placebo, 50, 100, 200 or 300 mg ipragliflozin once daily for 28 days. The patients were aged 18–75 years, with BMI 20–45 kg/m2 and HbA1c 7.0–10.0 %, and were diagnosed with type 2 diabetes ≥3 months before the study. Patients were either drug-naïve or could discontinue their medication ≥2 weeks before starting the study drug. The AUC and C max of ipragliflozin were dose proportional at steady state for the tested dose range (50–300 mg). The t max of ipragliflozin was approximately 1 h and was similar on days 1 and 28 of administration. C max was slightly higher on day 28 than on day 1. The percentages of the excretion ratio of ipragliflozin excreted into urine at steady state (Ae24%) were similar across all dose groups.

3 Clinical Pharmacodynamics

3.1 Mechanism of Action

The kidneys play an important role in glucose homeostasis. Glucose is filtered in the glomeruli and is almost completely reabsorbed in the proximal tubules [19]. The SGLTs are a large family of membrane proteins responsible for the transport of glucose, amino acids, vitamins, osmolytes, and electrolytes in the intestine and proximal tubules of the kidney. SGLT2 is a high-capacity low-affinity membrane transporter predominantly expressed in the early proximal tubules, where it is responsible for the reabsorption of approximately 80–90 % of the glucose in the glomerular filtrate [4]. The remainder is reabsorbed via SGLT1. In people with diabetes, the kidneys have an increased capacity to reabsorb glucose [20, 21], meaning that excess glucose that would normally be excreted in the urine is reabsorbed, exacerbating hyperglycaemia. A more recent study has examined the expression profile of glucose transporters using cultured renal proximal tubular cells obtained from humans [22]. The authors noted that SGLT2 expression was much greater in cells from patients with non-insulin-dependent diabetes than in cells from healthy volunteers when exposed to a hyperglycaemic environment. However, it is unclear whether this is related to genetic factors or represents a deleterious physiological response to the hyperglycaemic environment. This role of SGLT2 led to the approach of inhibiting SGLT2 to regulate glucose levels in people with diabetes.

3.2 Urinary Glucose Excretion

Urinary glucose excretion (UGE) has been evaluated in healthy volunteers [15, 16] and in patients with type 2 diabetes [17, 18]. In healthy volunteers, administration of a single oral dose of ipragliflozin increased mean UGE across all dose groups [15, 16]. In a study involving healthy Japanese subjects [15], a dose-dependent increase in UGE occurred with increasing ipragliflozin dose after a single dose, and the increases in UGE could be detected for up to 24 h after administration (Fig. 2a). After the final dose in 7 days of multiple dosing, UGE approached a maximum of approximately 50 g over 24 h with both 50- and 100-mg ipragliflozin doses (Fig. 2b) [15]. In healthy non-Japanese subjects [16], a dose-dependent increase in UGE was also observed after a single dose. In addition, daily UGE tended to continue to increase slightly at doses >100 mg [16]. In Japanese patients with type 2 diabetes receiving multiple doses of ipragliflozin over 14 days, the change in UGE from baseline up to 24 h post-dose (ΔUGE24) increased significantly in the 50- and 100-mg ipragliflozin groups, although the changes were similar in both groups (approximately 81 and 90 g, respectively). There was no significant change in the placebo group. In non-Japanese patients with type 2 diabetes, ΔUGE24 increased in all dose groups (50, 100, 200 and 300 mg) over 28 days’ administration [18]. At day 28, ΔUGE24 was approximately 90 g for 300 mg ipragliflozin.

Fig. 2
figure 2

Mean ± standard deviation cumulative urinary glucose excretion after single doses of 1–300 mg ipragliflozin or placebo (a) or after multiple doses of 20, 50 or 100 mg ipragliflozin or placebo (b) in healthy Japanese volunteers. Both panels are reprinted with permission from Diabetology International

Full size image

It is well known that SGLT2 inhibitors inhibit only 30–50 % of the filtered glucose load [23]. In the case of ipragliflozin, 40–50 % inhibition was observed in Japanese patients with type 2 diabetes (50–60 % of the filtered glucose load was excreted into urine [24]). One possible explanation for this phenomenon is that complete inhibition of SGLT2 forces SGLT1 to reabsorb glucose in full capacity, and therefore only the fraction of filtered glucose that escapes SGLT1 will be excreted in the urine [23].

3.3 Effects on Plasma Glucose

The effects of ipragliflozin on plasma glucose levels have been studied in Japanese and non-Japanese healthy volunteers [15, 16]. In both studies, treatment with single or multiple doses of ipragliflozin (1–300 and 5–600 mg/day for Japanese and non-Japanese healthy volunteers, respectively) did not significantly affect plasma glucose levels compared with placebo. The pharmacokinetic/pharmacodynamic characteristics of two oral doses of ipragliflozin were also examined in Japanese patients with type 2 diabetes. In that study, patients were treated with placebo, 50 or 100 mg ipragliflozin once daily for 14 days [17]. In both ipragliflozin dose groups, the plasma glucose levels measured throughout the day were lower on day 14 than at baseline (i.e. day −1 [1 day before the first dose]), whereas plasma glucose profiles were similar on days −1 and 14 in the placebo group (Fig. 3). FPG decreased significantly from day 1 (immediately before study drug administration) to day 15 in both ipragliflozin groups compared with the placebo group (0.3 ± 20.5, −31.6 ± 24.3 and −35.8 ± 29.1 mg/dL in the placebo, 50 and 100 mg groups, respectively). Plasma glucose AUC3h (post-prandial) and AUC24 decreased significantly from day −1 to day 14 in both ipragliflozin groups compared with the placebo group. Post-prandial glucose (PPG) levels measured at 1 and 2 h after each meal were lower on day 14 than on day −1 in both ipragliflozin groups, but not in the placebo group. The mean changes in 1 h PPG were −53.1 ± 34.2 and −58.7 ± 32.4 mg/dL in the 50 and 100 mg ipragliflozin groups, respectively, vs. 3.5 ± 23.8 mg/dL in the placebo group. The mean changes in 2-h PPG were −60.1 ± 45.5 and −62.9 ± 34.3 mg/dL in the 50 and 100 mg ipragliflozin groups, respectively, vs. −1.9 ± 39.8 mg/dL in the placebo group. In the therapeutic dose range, the plasma glucose level is one of the most relevant factors to determine the effect on UGE and renal function. The relationship between baseline plasma glucose levels and UGE24 in Japanese patients with type 2 diabetes after multiple doses of 50 or 100 mg ipragliflozin is shown in Fig. 4a.

Fig. 3
figure 3

Plasma glucose profiles in Japanese patients with type 2 diabetes mellitus on days −1 and 14 in the 50 mg ipragliflozin (a), 100 mg ipragliflozin (b) and placebo (c) groups. Values are shown as the means ± standard deviation

Full size image
Fig. 4
figure 4

a Correlation between baseline fasting plasma glucose (FPG) and urinary glucose excretion up to 24 h post-dose (UGE24) in Japanese patients with type 2 diabetes after multiple doses of 50 or 100 mg ipragliflozin [12]. b Relationship between area under the plasma concentration-time curve (AUC) and change from baseline in urinary glucose excretion up to 72 h post-dose (ΔUGE72) in Japanese patients wih type 2 diabetes [12]. c Correlation between estimated glomerular filtration rate (eGFR) and change from baseline in urinary glucose excretion up to 24 h post-dose (ΔUGE24) in Japanese patients with type 2 diabetes [12]

Full size image

3.4 Effects on Glycosylated Haemoglobin

At a daily dose of 50 mg, ipragliflozin decreased HbA1c levels in patients with type 2 diabetes by nearly 1 % under various conditions [8]. In addition, in studies examining its effect in combination with other oral antidiabetic drugs, ipragliflozin reduced HbA1c levels remarkably when added to metformin [25], pioglitazone [10], or a sulphonylurea [9].

3.5 Effects on Other Parameters

In the study of Japanese T2DM patients, the mean serum insulin levels on day 14 were generally lower than those on day −1 in both ipragliflozin groups, but not in the placebo group [17]. Fasting serum insulin decreased significantly in both ipragliflozin groups but not in the placebo group. (0.54 ± 2.08, −1.31 ± 1.21 and −2.38 ± 1.88 μU/mL in the placebo, 50 and 100 mg groups, respectively). Urine volume increased slightly from baseline in the 50 and 100 mg ipragliflozin groups (224.4 and 203.3 mL, respectively), although the changes were relatively small because the urine volumes measured at baseline were quite large (2,034.4 and 2,557.2 mL, respectively). There was no clear difference in mean water intake in all three groups. In clinical trials, the administration of 50 mg ipragliflozin was associated with a mean reduction in body weight of 2 kg in patients with type 2 diabetes [7, 8]. In addition, administration of 50 mg ipragliflozin for 16 weeks reduced systolic and diastolic blood pressure by 3.2 and 2.5 mmHg, respectively, without causing hypotension [7].

3.6 Relationship between Pharmacokinetics and Pharmacodynamics

UGE increased with exposure of ipragliflozin, showing that level of exposure is an important factor in the pharmacological effects of ipragliflozin (Fig. 4b). As described in Sect. 3.2, with increasing ipragliflozin dose, daily UGE approaches the maximum effect of approximately 50 g in healthy subjects and 81–90 g in patients with type 2 diabetes, after doses varying from 50 to 100 mg, which was chosen as the therapeutic dose range. In the therapeutic dose range, the impact of the increased drug exposure on the pharmacodynamic effect will not be relevant. This also indicates that UGE can be mainly attributed to plasma glucose levels and renal function (glomerular filtration rate [GFR]) within the therapeutic dosage range [24]. The relationship between estimated GFR (eGFR) and the change in UGE24 from baseline (ΔUGE24) after a single dose of 50 mg is shown in Fig. 4c.

4 Clinical Pharmacokinetics and Pharmacodynamics in Special Populations

4.1 Age and Gender

The effects of age and gender on glucose levels, pharmacokinetics, safety, and tolerability of multiple oral doses of ipragliflozin were examined in a phase I, double-blind, placebo-controlled, randomised trial [12]. The trial included 65 male and female healthy volunteers aged 18–45 or ≥65 years who were administered 100 mg ipragliflozin once daily for 18 days. In female subjects, the geometric mean ratio (GMR; 90 % confidence interval [CI]; older vs. younger volunteers) for AUC24 was 1.45 (1.27, 1.67) while that for C max was 1.25 (1.06, 1.49) after multiple doses. In male subjects, the ratio for AUC24 was 1.21 (1.06, 1.38), whereas C max was similar between younger and older male volunteers. In older volunteers, the GMRs (90 % CI; female vs. male subjects) for AUC24 and C max were 1.39 (1.22, 1.59) and 1.35 (1.15, 1.58), respectively. In younger volunteers, the corresponding ratios were 1.16 (1.01, 1.33) and 1.06 (0.89, 1.25). The GMR (older vs. younger volunteers) for ΔUGE24 after a single dose was 0.73 (0.66, 0.80) and the GMR for ΔUGE24 after multiple doses was 0.69 (0.59, 0.80). The GMRs (female vs. male subjects) for ΔUGE24 after a single dose and for ΔUGE24 after multiple doses were 0.75 (0.68, 0.82) and 0.71 (0.61, 0.82), respectively. The increase in ΔUGE24 was smaller in older than in younger volunteers and in female than in male subjects, despite the greater drug exposure in older volunteers and female subjects, which may be because of differences in renal function among these subgroups.

4.2 Renal Impairment

Renal impairment is a common complication of type 2 diabetes [26] and, because of their mechanism of action, the efficacy of SGLT2 inhibitors can be affected by renal impairment. Therefore, two multicentre open-label studies were conducted to examine the effects of ipragliflozin in subjects with renal impairment; one in Japanese patients and the other in European patients and healthy volunteers [27, 28].

The objective of the Japanese study [28] was to investigate the pharmacokinetics and pharmacodynamics (UGE) after a single oral dose of 50 mg ipragliflozin in Japanese patients with type 2 diabetes with varying degrees of renal impairment. In total, 25 patients were categorised into one of the following groups: normal renal function (eGFR ≥90 mL/min/1.73 m2; n = 8), mild renal impairment (eGFR 60 to <90 mL/min/1.73 m2; n = 9) and moderate renal impairment (eGFR 30 to <60 mL/min/1.73 m2; n = 8). eGFR was calculated using the following equation for Japanese individuals [29]: eGFR (mL/min/1.73 m2) = 194 × serum creatinine (mg/dL)−1.094 × age (years)−0.287 × 0.739 (if female). C max and AUC∞ were 1.17 times and 1.21 times higher, respectively, in patients with type 2 diabetes with moderate renal impairment than in patients with normal renal function (Table 2) [28].

Table 2 Effects of intrinsic factors on the pharmacokinetics of ipragliflozin
Full size table

The European study [27] evaluated the effects of a single dose of 100 mg ipragliflozin in eight healthy volunteers and 32 patients with type 2 diabetes. The patients with type 2 diabetes were classified into those with normal renal function (eGFR ≥90 mL/min/1.73 m2; n = 8), mild renal impairment (eGFR 60 to <90 mL/min/1.73 m2; n = 8), moderate renal impairment (eGFR 30 to <60 mL/min/1.73 m2; n = 8) or severe renal impairment (eGFR 15 to <30 mL/min/1.73 m2; n = 8). eGFR was estimated based on the Modification of Diet in Renal Disease formula [30]: eGFR (mL/min/1.73 m2) = 186 × serum creatinine (mg/dL)−1.154/age (years)−0.203 × 1.212 (if black) × 0.742 (if female). Consistent with the Japanese study, in patients with type 2 diabetes, C max and AUC∞ were increased in patients with type 2 diabetes with moderate or severe renal impairment. In particular, AUC∞ was 1.40 times higher in patients with type 2 diabetes with moderate renal impairment and 1.47 times higher in patients with type 2 diabetes with severe renal impairment than in patients with type 2 diabetes with normal renal function (Table 2) [27, 28]. In both studies, patients with type 2 diabetes with renal impairment had a smaller body size than patients with type 2 diabetes with normal renal function, which might partially explain the differences in exposure. The results of unpublished in vitro studies (data on file, Astellas Pharma Inc.) suggested that ipragliflozin is primarily metabolised in the liver, but is also metabolised in the kidney, which may also contribute to increased exposure in patients with type 2 diabetes with moderate or severe renal impairment. The European study [27] also evaluated the pharmacokinetics of ipragliflozin metabolites (M1, M2, M3, M4 and M6). Of these, exposure of M2 was greatest, second only to exposure of unchanged ipragliflozin; exposure of the other metabolites was much lower. For all metabolites, AUC∞ increased with increasing severity of renal impairment, by 2.1–3.0 times in patients with type 2 diabetes with moderate renal impairment and by 3.3–5.4 times in patients with type 2 diabetes with severe renal impairment compared with patients with type 2 diabetes with normal renal function.

The correlation between eGFR and ΔUGE24 after administration in Japanese patients with type 2 diabetes is presented in Fig. 4c. ΔUGE24 tended to decrease with decreases in eGFR. In patients with type 2 diabetes with moderate renal impairment, ΔUGE24 was only 54 % of that seen in patients with normal renal function, despite the greater exposure in patients with type 2 diabetes with renal impairment (Table 3). This can be attributed to a decrease in the amount of filtered glucose owing to decreased renal function. However, in the Japanese study, FPG at baseline was higher in patients with normal renal function than in patients with type 2 diabetes with renal impairment, which may have also contributed to the difference in UGE among subgroups.

Table 3 Daily urinary glucose excretion before and after a single dose of 50 mg or 100 mg ipragliflozin stratified by severity of renal impairment
Full size table

The mean changes from baseline in plasma glucose levels at 24 h after administration were −24.9 mg/dL in Japanese patients with type 2 diabetes with normal renal function, −11.9 mg/dL in patients with type 2 diabetes with mild renal impairment and −4.1 mg/dL in patients with type 2 diabetes with moderate renal impairment. These data indicate that the mean change from baseline decreased with increasing severity of renal impairment [28].

Considering the results of the studies presented here and the results of other clinical studies [12], it was not considered necessary to reduce the dose of ipragliflozin in patients with renal impairment from pharmacokinetic and safety perspectives. However, ΔUGE tended to decrease with decreasing renal function, and the glucose-lowering effects of ipragliflozin in patients with moderate renal impairment tended to be weaker than in patients with normal renal function or patients with mild renal impairment [31]. These results suggest that the use of ipragliflozin should be carefully evaluated in patients with moderate renal impairment because the drug may not be fully effective in these patients. Furthermore, ipragliflozin should not be administered in patients with severe renal impairment or in dialysis patients with end-stage renal failure because it cannot be expected to be effective in these patients.

4.3 Hepatic Impairment

The effects of moderate hepatic impairment on the pharmacokinetics of ipragliflozin and its metabolites were investigated in an open-label, single-dose, parallel-group study conducted in the USA [32]. In total, eight subjects with moderate hepatic impairment (Child-Pugh score 7–9) and eight healthy matched controls received a single oral dose of 100 mg ipragliflozin. The GMRs [for moderate hepatic impairment vs. normal subjects (90 % CI)] for C max and AUC∞ were 1.27 (0.93, 1.73) and 1.25 (0.94, 1.66), respectively. No clear differences were observed in the half-life or protein binding of ipragliflozin between subjects with moderate hepatic impairment and normal subjects. For M2, the major metabolite, the C max and AUC∞ were similar in both groups, with GMRs of 0.95 (0.68, 1.33) and 1.00 (0.77, 1.30), respectively [32]. No clear differences were observed in ΔUGE24. Overall, the study demonstrated that moderate hepatic impairment did not have clinically relevant effects on the pharmacokinetics of a single dose of ipragliflozin or its major metabolite, M2.

5 Cardiac Safety

The International Conference on Harmonisation recommends that the proarrhythmic potential of developmental drugs is assessed by characterising the QT/corrected QT (QTc) interval [33]. A phase I, double-blind, placebo- and active-controlled, four-way crossover study was conducted in 88 healthy subjects [34], in which the subjects received four treatments (placebo, 100 mg ipragliflozin, 600 mg ipragliflozin and 400 mg moxifloxacin). Moxifloxacin was used as an active control because of its effects on the QT/QTc interval. For the placebo and ipragliflozin periods, subjects received the study drug for 7 days; those in the moxifloxacin group received the study drug on day 7 only. The study showed that treatment with ipragliflozin at the therapeutic (100 mg) or supratherapeutic (600 mg) doses for 7 days did not have clinically meaningful effects on the QTc (Fridericia’s formula) interval.

6 Drug–Drug Interactions

6.1 Antidiabetic Agents

6.1.1 Metformin

The effects of ipragliflozin on the pharmacokinetics of metformin in patients with type 2 diabetes were investigated in a randomised, double-blind, placebo-controlled study conducted at four sites in Europe [35]. In this study, 36 patients stable on metformin therapy (850, 1,000 or 1,500 mg twice daily) were randomised to receive either ipragliflozin (300 mg; n = 18) or placebo (n = 18) once daily for 14 days. The GMRs (metformin + ipragliflozin vs. metformin alone; 90 % CI) for C max and AUC10 were 1.11 (1.03, 1.19) and 1.18 (1.08, 1.28), respectively. The increase in metformin exposure was considered minor. In a Japanese study examining ipragliflozin in combination with metformin [11], no special safety problems were noted, and tolerability was favourable when ipragliflozin was used in combination with metformin for 52 weeks, including when the dose was increased from 50 to 100 mg at week 24. In addition, no safety concerns were identified in a European/US phase II combination study [25]. Thus, the difference in exposure was not considered to be clinically significant.

6.1.2 Sitagliptin

The drug–drug interactions between sitagliptin (a dipeptidyl peptidase-4 inhibitor) and ipragliflozin were determined in an open-label, randomised, two-way, crossover study in 64 healthy subjects [36]. The primary objective was to investigate the effects of multiple doses of ipragliflozin (150 mg) on the pharmacokinetics of a single dose of sitagliptin (100 mg), and vice versa. The GMRs (sitagliptin + ipragliflozin vs. sitagliptin alone; 90 % CI) for C max and AUC∞ were 0.924 (0.828, 1.031) and 1.001 (0.969, 1.035), respectively. Similarly, multiple doses of 100 mg sitagliptin did not alter the pharmacokinetics of a single dose of 150 mg ipragliflozin in terms of C max (0.965 [0.904, 1.031]) or AUC∞ (0.950 [0.934, 0.966]).

6.1.3 Pioglitazone

The study by Smulders et al. [36] also examined drug–drug interactions between ipragliflozin and pioglitazone. The primary objectives of the study were to investigate the effects of repeated doses of 150 mg ipragliflozin on the pharmacokinetics of a single dose of 30 mg pioglitazone, and vice versa. The GMRs (pioglitazone + ipragliflozin vs. pioglitazone; 90 % CI) for C max and AUC∞ were 0.986 (0.877, 1.108) and 1.017 (0.966, 1.070), respectively. Similarly, multiple doses of pioglitazone did not affect the pharmacokinetics of a single dose of ipragliflozin in terms of AUC∞ (1.000 [0.981, 1.020]) or C max (0.935 [0.863, 1.012]). In addition, ipragliflozin did not seem to affect the pharmacokinetics of the two active metabolites of pioglitazone (ketopioglitazone and hydroxypioglitazone).

6.1.4 Glimepiride

Smulders et al. [36] also examined the potential for interactions between 150 mg ipragliflozin and glimepiride, a sulphonylurea. At 1 mg glimepiride, the GMRs (glimepiride + ipragliflozin vs. ipragliflozin; 90 % CI) for AUC∞ and C max were 0.991 (0.966, 1.016) and 0.973 (0.892, 1.062), respectively. After multiple doses of 150 mg ipragliflozin, the GMRs for AUC∞ and C max were 1.051 (1.013, 1.090) and 1.100 (1.019, 1.188), respectively, for 2 mg glimepiride. The GMRs for AUC∞ and C max for 5-OH glimepiride, an active metabolite of glimepiride, were 0.998 (0.966, 1.031) and 1.008 (0.941, 1.079), respectively.

6.1.5 Miglitol

The α-glucosidase inhibitor miglitol can be absorbed from the upper small intestine via SGLT1 [37]. Therefore, the potential for drug–drug interactions in the gastrointestinal absorption of miglitol was assessed in an open-label, randomised, six-sequence, three-way, crossover study in 30 healthy Japanese subjects [38]. Subjects received combinations of 100 mg ipragliflozin, 75 mg miglitol, or both, in varying sequences over three treatment periods. For ipragliflozin, the GMRs (miglitol + ipragliflozin vs. ipragliflozin; 90 % CI) for C max and AUC∞ were 1.034 (0.944, 1.132) and 1.015 (0.988, 1.043), respectively. The corresponding GMRs for miglitol (miglitol + ipragliflozin vs. miglitol) were 0.761 (0.672, 0.861) and 0.796 (0.719, 0.881). Co-administration of miglitol did not affect the pharmacokinetics of ipragliflozin, whereas the C max and AUC∞ of miglitol were decreased by approximately 20–25 % by ipragliflozin. The absorption of miglitol through SGLT1 might be partially inhibited by ipragliflozin in the intestine. However, this effect was not considered to be clinically relevant because the change was small and because miglitol acts in the gastrointestinal tract, rather than systemically.

6.1.6 Mitiglinide

Mitiglinide is a short-acting insulinotropic agent. It has a rapid stimulatory effect on insulin secretion and reduces PPG levels in patients with type 2 diabetes [39]. Glucuronidation is the main metabolic pathway for both ipragliflozin and mitiglinide. UGT1A3 and UGT1A9 are the primary isozymes in mitiglinide metabolism, while UGT1A9 may also be involved in the metabolism of ipragliflozin. The potential for drug–drug interactions between mitiglinide and ipragliflozin was assessed in an open-label, randomised, crossover study in 60 healthy male Japanese volunteers [12, 40]. The GMRs (mitiglinide + ipragliflozin vs. mitiglinide alone; 90 % CI) for C max and AUC∞ were 0.871 (0.769, 0.986) and 1.011 (0.994, 1.029), respectively. Following multiple doses of 10 mg mitiglinide three times daily, the GMRs relative to 100 mg ipragliflozin for C max and AUC∞ were 0.946 (0.896, 0.999) and 1.004 (0.982, 1.026), respectively.

6.2 Diuretics

The effects of co-administration of furosemide and ipragliflozin were assessed in a three-period, randomised, open-label, crossover, multiple-dose study [12]. The primary objective of the study was to assess urine sodium excretion following multiple doses of ipragliflozin + furosemide vs. furosemide alone in 24 healthy subjects. The effects on pharmacokinetics and other electrolytes (i.e. calcium, chloride, potassium, phosphate and magnesium in blood and urine) were also assessed as secondary parameters. Furosemide did not affect the pharmacokinetics of ipragliflozin or its major metabolite, M2. In this study, the GMR (ipragliflozin + furosemide vs. furosemide alone; 90 % CI) for C max was 1.07 (0.88, 1.30). The GMR for AUCtau was 1.06 (0.95, 1.19). On the first day of treatment, the urinary excretions of sodium, chloride and potassium were increased by 14, 11 and 8 %, respectively, following administration of furosemide + ipragliflozin relative to furosemide alone. Ipragliflozin did not affect the urinary excretion of these electrolytes after 5 days of treatment with furosemide + ipragliflozin. Furthermore, ipragliflozin did not affect urinary excretion of calcium during co-administration with furosemide. Although ipragliflozin decreased urinary excretion of magnesium by 18 % and phosphate by 14 % after the initial dose, this effect was of short duration, and urinary excretion of both electrolytes returned to baseline levels within 5 days of treatment. Ipragliflozin did not affect the serum levels of sodium, chloride, potassium or calcium during co-administration with furosemide. The effects of ipragliflozin on the electrolyte changes induced by furosemide were small and in general of short duration, with most effects having returned to baseline after 5 days of continuous treatment. No remarkable differences were observed on UGE and urine volume.

7 Food Effects

The effects of food on the pharmacokinetics and pharmacodynamics of ipragliflozin were assessed following a single oral dose of 50 mg ipragliflozin in non-elderly healthy Japanese male subjects in a six-sequence three-period crossover study [12]. Thirty healthy male subjects received 50 mg ipragliflozin under fasting conditions, and before and after ingestion of a high-fat meal. Subjects were required to fast from 22:00 on the day prior to drug administration. In the fasting condition, subjects received ipragliflozin without breakfast. In the before-meal condition, subjects began to eat a meal 5 min after drug administration. In the after-meal condition, ipragliflozin was administered within 10 min after meal ingestion. AUClast (i.e. from the time of dosing to the last measurable concentration) was comparable among the three treatment conditions. The GMR (90 % CI) for C max for ipragliflozin was approximately 1.231 (1.138, 1.332) before the meal, and 0.820 (0.758, 0.888) after the meal compared with C max measured under fasting conditions. The mean t max was 2.17 h when ipragliflozin was administered after the meal, 1.25 h when ipragliflozin was administered in fasting conditions and 0.65 h when ipragliflozin was administered before the meal. ΔUGE24 after a single dose of 50 mg ipragliflozin did not differ remarkably between the fasting condition, and the before- or after-meal conditions.

8 Summary

The SGLT2 inhibitors are a novel class of oral hypoglycaemic agents that offer several advantages over other classes of oral antidiabetic agents, including a low risk of hypoglycaemia together with bodyweight reductions during treatment. Because their mode of action is independent of insulin, SGLT2 inhibitors are thought to be able to provide steady glucose control without a high risk of hypoglycaemia. Based on the results summarised in this review, ipragliflozin showed a favourable pharmacokinetic profile similar to that of other SGLT2 inhibitors, including canagliflozin, dapagliflozin and empagliflozin [41–43].

Only approximately 1 % of the ipragliflozin dose was excreted into urine, and the CLR of ipragliflozin was approximately 0.1 L/h (\( { \fallingdotseq } \) 2 mL/min), which is lower than the unbound fraction (fu) × GFR. In patients with type 2 diabetes, the average unbound drug concentration at steady state was around 10 ng/mL (= fu × AUC24/24) after multiple oral doses of 50 mg ipragliflozin. This value of 10 ng/mL (\( { \fallingdotseq } \) 25 nM) was somewhat higher than the IC50 value for ipragliflozin against human SGLT2 in vitro (7.38 nM) [44]. These data suggest that unbound ipragliflozin is freely filtered and inhibits renal glucose reabsorption across the luminal membrane of the proximal tubule rather than across the cytosolic membrane. Furthermore, even though the renal excretion of ipragliflozin is relatively low, the tubular concentration of free ipragliflozin is adequate to provide effective inhibition of SGLT2-mediated glucose transport.

Similar to other SGLT2 inhibitors, we observed no clinically relevant effects of age, gender or food on the exposure of ipragliflozin. Indeed, a long-term phase III study showed comparable efficacy and safety of administering ipragliflozin once daily before or after a meal [45], even though the C max was 33 % higher when administered after the meal than when administered before the meal in the food effect study [12]. These findings suggest that differences in C max are unlikely to affect the efficacy and safety of ipragliflozin, compared with AUC, making it suitable for once-daily administration. Additionally, the pharmacokinetics of ipragliflozin and other oral antidiabetic drugs were not affected by their co-administration in drug–drug interaction studies. Likewise, concomitant administration of a diuretic did not markedly affect the pharmacokinetics of ipragliflozin, UGE or urine volume, although these results are from healthy volunteers.

Ipragliflozin enhanced UGE, with effects that lasted for up to 24 h after administration, suggesting that ipragliflozin could be administered once daily. UGE also increased with an increasing dose of ipragliflozin, approaching a maximum effect at therapeutic doses of 50–100 mg in Japanese patients with type 2 diabetes. This indicates that UGE is associated with plasma glucose levels and renal function at therapeutic doses. ΔUGE tended to decrease with decreasing renal function, especially in patients with type 2 diabetes with moderate or severe renal impairment. This implies that the administration of ipragliflozin to those patients should be carefully considered, as is the case for other SGLT2 inhibitors.