The pharmacokinetics of pioglitazone in patients with impaired renal function

Abstract

Aims

To evaluate the effect of renal impairment on the pharmacokinetics and safety of pioglitazone and its metabolites M-III and M-IV with impaired renal function and normal renal function.

Methods

In a phase-I, open-label, parallel-group study, six healthy subjects with normal renal function (creatinine clearance>80 ml min⁻¹), nine patients with moderate renal impairment (creatinine clearance 30–60 ml min⁻¹) and 12 patients with severe renal impairment (creatinine clearance <30 ml min⁻¹) received single and multiple oral doses of pioglitazone 45 mg. The serum pharmacokinetic profiles of pioglitazone and its metabolites M-III and M-IV were assessed for the first and last dose administered (day 1 and day 12, respectively).

Results

Pharmacokinetic data revealed no significant accumulation of pioglitazone or its metabolites M-III and M-IV in patients with renal impairment. There was no significant difference in the pharmacokinetic profile of pioglitazone in subjects with normal and with moderately impaired renal function. After single oral doses, mean area under the concentration-time curve (AUC) values were decreased in patients with severe renal impairment compared with healthy subjects with normal renal function for pioglitazone (13 476 vs 17 387, P = 0.371; −23%; confidence interval (CI) −57, 38), M-III metabolite (13 394 vs 15 071, P = 0.841; −11%; CI −74, 194) and M-IV metabolite (27 991 vs 49 856, P = 0.006; −44%; CI −62, −17). After repeated oral doses of pioglitazone, mean AUC values (µg.h l⁻¹) were decreased in patients with severe renal impairment compared with healthy subjects with normal renal function for pioglitazone (8744 vs 14,565, P = 0.004; −40%; CI −57, −16), M-III (3991 vs 7,289, P = 0.0009; −45%; CI −60, −25) and M-IV (21 080 vs 25 706, P = 0.181; −18%; CI 39, 10). The tolerability and safety profile of pioglitazone was comparable between groups.

Conclusions

Pioglitazone was well tolerated in patients with varying degrees of renal impairment. Although mean serum concentrations of pioglitazone and its metabolites are increased in patients with severe renal impairment, adjustment of starting and maintenance doses in these patients is probably unwarranted.

Introduction

Many patients with Type 2 diabetes have mild to moderate renal impairment as a late complication of inadequate glycaemic control [1, 2]. Patients with diabetes are 17 times more prone to kidney disease than nondiabetics, with diabetic nephropathy being the most common complication [3]. The management of diabetes in patients with renal impairment has particular clinical concerns as most oral hypoglycaemic agents are eliminated primarily by renal mechanisms. Impaired renal function may result in greatly reduced excretion [4], leading to the accumulation of drug and active metabolites and thus increasing the risk of hypoglycaemia and other adverse events.

Pioglitazone is a novel oral hypoglycaemic agent for the management of Type 2 diabetes [5, 6]. It is a peroxisome proliferator-activated receptor (PPAR-γ) agonist that increases transcription of insulin-responsive genes and thus increases insulin sensitivity [7–9]. In common with other thiazolidinediones, pioglitazone ameliorates insulin resistance associated with Type 2 diabetes without stimulating insulin release from pancreatic β-cells, thus lowering the risk of hypoglycaemia [10].

Previous clinical trials in healthy subjects have shown that pioglitazone is well absorbed after oral administration. At doses ranging from 2 to 60 mg, peak concentrations of pioglitazone in the blood of healthy subjects are achieved approximately 1.5 h after oral drug administration [11]. Pioglitazone is highly bound to plasma proteins (approximately 97%), with a low tissue distribution and slow elimination (half-life approximately 9 h). Pioglitazone is extensively metabolized in the liver, with the majority excreted as inactive metabolites in the faeces. Pioglitazone undergoes significant hepatic metabolism by hydroxylation of aliphatic methylene groups to form three metabolites (M-I, M-II, and M-IV), by the oxidation of the methyl group to form an additional metabolite (M-V), and by oxidation of metabolite M-IV to metabolite M-VI [11]. Three of the metabolites (M-III, M-IV, and to a lesser extent M-II), were shown to have pharmacological activity in diabetic animal models. In rats, the relative hypoglycaemic potency (ED₅₀) of these metabolites was 40–60% of that of pioglitazone. The potency of the triglyceride-lowering effect of M-II is nearly twice that of the parent compound, while the potency of metabolites M-III and M-IV is slightly less than that of pioglitazone [11]. The major active metabolites M-III and M-IV have considerably longer terminal half-lives than the parent compound (approximately 26–28 h) [11]. Because elimination of pioglitazone and its metabolites is primarily hepatic, pioglitazone may have potential value in patients with Type 2 diabetes who have renal impairment. To address the possible use of pioglitazone in such patients, this study aimed to investigate the pharmacokinetics of pioglitazone and its main metabolites M-III and M-IV in patients with impaired renal function compared with people with normal renal function [12].

Patients and methods

In a phase-I, open-label, parallel-group study conducted at two university hospitals, six healthy subjects with normal renal function, nine patients with moderate renal impairment and 12 patients with severe renal impairment received single and multiple doses of pioglitazone. Two patients with impaired renal function had previously been diagnosed with diabetes (one patient with Type 1 diabetes and another with recently diagnosed Type 2 diabetes). No healthy subjects had diabetes at study entry.

Male and female subjects aged 18–70 years and with a body mass index of between 18 and 28 kg m⁻² were eligible to enter the study. Concomitant medications for the basic treatment of renal disease were permitted for the duration of the study. Exclusion criteria included severe organ system disease, except complications directly related to diabetes. Subjects who had previously received pioglitazone or who had participated in another clinical investigation within the previous 30 days were also excluded from the trial.

Eligible subjects were subdivided by degree of renal function into three groups based on creatinine clearance (CL_CR): moderate renal impairment (CL_CR = 30–60 ml min⁻¹); severe renal impairment (CL_CR ≤ 30 ml min⁻¹); and normal renal function (CL_CR> 80 ml min⁻¹) (creatinine clearance SI conversion factor to ml s⁻¹ = × 0.0167). The 24-h clearance of endogenous creatinine was assessed on two occasions prior to the study and at the end of the study. To avoid the risk of a potential accumulation of pioglitazone or its metabolites in patients with severe renal impairment, the group of patients with moderate renal impairment was recruited and investigated first. Patient comedication was to remain unchanged for the duration of dosing.

On day 1, subjects received a single oral dose of 45 mg pioglitazone and drug concentrations were measured at intervals over the following 48-h period (1, 2, 3, 4, 6, 8, 10, 12, 16, 24 and 48 h). On days 3–12, pioglitazone treatment continued (45 mg once daily) and drug concentrations were measured at intervals over the following 120-h period (1, 2, 3, 4, 6, 8, 10, 12, 16, 24, 48, 72, 96 and 120 h). Blood samples were withdrawn from a catheter at the specified times, and centrifuged at 1500 g for 15 min. The serum was decanted and stored at −20°C until assayed. Serum pioglitazone and metabolites M-III and M-IV were quantitatively assayed using an internally standardized high-performance liquid chromatography (HPLC) assay with u.v. detection [13]. The analytes were isolated from serum samples by liquid/liquid extraction under neutral conditions. The organic extract was dried, reconstituted in methanol/mobile phase and an aliquot was injected into the HPLC. The respective peak height ratios determined by reversephase HPLC were used for calibration by weighted (1 y⁻²) linear regression. Calibration standards prepared in human serum were prepared with the patient samples. During the assay period, human serum quality control samples spiked with three different concentrations of pioglitazone, M-III and M-IV were assayed together with the test samples for quality control purposes, and analysis continued if four out of six quality control samples (at least one sample at each concentration) were within a range of ±20% of the spiked concentration. The interassay precision (coefficient of variation), as determined from the analysis of quality control samples, ranged between 4.8% and 9.2% for pioglitazone, between 5.4% and 7.2% for M-III, and between 5.6% and 7.6% for M-IV. The limit of quantification was 50.3 µg l⁻¹ for pioglitazone, 50.5 µg l⁻¹ for M-III and 50.7 µg l⁻¹ for M-IV.

The primary pharmacokinetic parameters were area under the serum concentration–time curve (AUC) and maximum serum concentration (C_max). Secondary parameters were time to reach C_max (t_max), terminal half-life (t_1/2), the serum concentration 24 h following last drug intake (C_trough) and total amount of unchanged pioglitazone. Comparisons between study groups were made with analysis of variance (anova) for AUC and C_max. anova models were fitted to log-transformed AUC and C_max values. Estimates of the group means with 95% confidence intervals (CI) were calculated using the mean square error of the anova as estimate of the variance. Furthermore, pairwise comparisons between the group of healthy subjects and the groups with moderate and severe renal impairment were performed. A statistical P-value <0.05 was considered significant. No formal statistical sample size estimation was performed for this study.

All adverse events, including hypoglycaemic events, were recorded. Routine clinical laboratory tests were performed before the study, and then after single and repeat dosing of pioglitazone. The trial was conducted in accordance with the International Conference for Harmonization Guidelines for Good Clinical Practice and performed according to the Revised Declaration of Helsinki. Written informed consent was obtained from all subjects prior to study entry. The above protocol was reviewed and accepted by the relevant local ethics committees (Bioethics Comittee of Jagellonian University, Cracow, Poland, Ethikkommission der Charité zu Berlin, and Freiburger Ethik Komission International).

Results

A total of 27 subjects completed the study, six with normal renal function, nine with moderate renal impairment, and 12 with severe renal impairment. The three groups were comparable in baseline demographic characteristics except for CL_CR, smoking habits and age (Table 1). The lowest prevalence of smoking was observed in patients with severe renal impairment. Owing to parallel recruitment of subjects with severe and normal renal function, a satisfactory age match across groups was not possible. However, healthy subjects and patients with moderate renal impairment were age matched. Mean age was highest in patients with severe renal impairment.

Table 1.

Demographic data of the trial population.

	Normal renal function (n = 6)	Moderate renal impairment (n = 9)	Severe renal impairment (n = 12)
Male/female (n)	3/3	7/2	7/5
Age (years) ± s.d.	35.7 ± 9.4	38.3 ± 10.8	48.7 ± 16.8
Weight (kg) ± s.d.	72.8 ± 13.4	74.6 ± 15.1	69.0 ± 14.1
Height (m) ± s.d.	1.67 ± 0.05	1.76 ± 0.11	1.68 ± 0.09
BMI (kg m−2) ± s.d.	26.0 ± 3.95	24.0 ± 3.83	24.5 ± 4.25
CLCR (ml min−1) ± s.d. (range)	99.8 ± 13.0 (88.3–120)	46.2 ± 9.43 (32.4–58.6)	14.9 ± 8.83 (4.3–28.4)
Smoker (%)	83	67	25

BMI, Body mass index. Normal renal function: CL_CR>80 ml min⁻¹. Moderate renal impairment: CL_CR 30–60 ml min⁻¹. Severe renal impairment: CL_CR <30 ml min⁻¹.

After a single dose of pioglitazone, the mean AUC_0–∞ of pioglitazone, though numerically reduced, was not statistically significantly different between patients with either severe or moderate renal impairment and those with normal renal function. There were no statistically significant differences in C_max, and no apparent differences in t_max or t_1/2 after a single dose (Table 2).

Table 2.

Pharmacokinetic parameters for pioglitazone.

	Renal impairment
	Normal renal function (n = 6)	Moderate renal impairment (n = 9)			Severe renal impairment (n = 12)
	Mean ± s.d.	Mean ± s.d.	P-value†	% change vs normal renal function [CI]	Mean ± s.d.	P-value‡	% change vs normal renal function [CI]
Single dose
AUC0-∞ (µg.h l−1) (range)	17 387(11 161–27 034)	14 466(8315–21 159)	0.538	−17 [−55, 52]	13 476(5589–112 343)	0.371	−23 [−57, 38]
Cmax (µg l−1)	1329 ± 667	1337 ± 363	0.626		1123 ± 295	0.665
tmax (h) (range)	2.0 (1.0–4.0)	1.0 (1.0–4.0)			2.0 (1.0–4.0)
t1/2 (h)	13.7 ± 5.7	9.7 ± 3.6			8.0 ± 3.0
Repeated doses
AUC0−24 (µg.h l−1) (range)	14 565(9333–18 395)	12 195(9241–19 698)	0.305	−16 [−41, 19]	8744(4817–16 244)	0.004	−40 [−57, −16]
Cmax (µg l−1)	1587 ± 225	1356 ± 461	0.293		1054 ± 157	0.009*
tmax (h) (range)	3.5 (1.0–4.0)	2.0 (1.0–4.0)			2.0 (1.0–4.0)
t1/2 (h)	11.1 ± 5.5	13.1 ± 5.6			11.4 ± 5.4

^†Normal renal function vs moderate renal impairment.

^‡Normal renal function vs severe renal impairment.

^*Statistically significant. Normal renal function: CL_CR>80 ml min⁻¹. Moderate renal impairment: CL_CR 30–60 ml min⁻¹. Severe renal impairment: CL_CR <30 ml min⁻¹.

The pharmacokinetic profile of pioglitazone after 10 consecutive doses is shown in Figure 1. There was a slight reduction in geometric mean daily AUC on day 10 compared with single-dose pioglitazone in all subject groups (Table 2). Daily AUC on day 10 was lower in patients with severe renal failure than in healthy subjects with normal renal function, and this achieved statistical significance after repeated doses of pioglitazone. Mean C_max tended to decrease with worsening renal failure, which became statistically significant in the severe renal impairment group compared with healthy subjects with normal renal function (P = 0.009; Table 2). Renal impairment had no effect on mean t_1/2 or t_max (Table 2). The half-life of pioglitazone was similar in all subject groups over the course of the study. The apparent clearance of pioglitazone (calculated as dose/daily AUC on day 10) and its relationship to creatinine clearance is shown in Figure 2. Generally, clearance of drug increased modestly with renal failure (correlation coefficient 0.535; P = 0.044). Four patients had considerably higher apparent clearance than the others, although this was not associated with a longer half-life.

Figure 1.

Serum concentration–time profiles of pioglitazone (45 mg OD for 10 consecutive days) in subjects with normal renal function (▴), n = 6; moderate renal impairment (•), n = 9; and severe renal impairment (▪), n = 12.

Figure 2.

Relationship between pioglitazone oral clearance and creatinine clearance following 10 single oral doses of pioglitazone (correlation coefficient 0.535; P = 0.044).

Serum concentrations of metabolites M-III and M-IV

Metabolites M-III (Table 3) and M-IV (Table 4) showed a similar pattern to parent drug, with reduced concentrations (in terms of both AUC and C_max) associated with worsening renal impairment. Similarly, half-life appeared to be unaffected by renal failure.

Table 3.

Pharmacokinetic parameters for the pioglitazone metabolite M-III.

	Renal impairment
	Normal renal function (n = 6)	Moderate renal impairment (n = 9)			Severe renal impairment (n = 12)
	Mean ± s.d.	Mean ± s.d.	P-value†	% change vs normal renal function [CI]	Mean ± s.d.	P-value‡	% change vs normal renal function [CI]
Single dose
AUC0–∞ (µg.h l−1) (range)	15 071(9151–22 247)	8590(5447–12 780)	0.363	−43 [−94, 100]	13 394§(3238–539 408)	0.841	−11 [−74, 194]
Cmax (µg l−1)	183 ± 72	211 ± 66	0.409		136 ± 47	0.196
tmax (h) (range)	12.0(12.0–24.0)	10.0(8.0–16.0)			12.0(6.0–24.0)
t1/2 (h)	51.8 ± 13.2	25.6 ± 9.3			72.2 ± 77.7
Repeated doses
AUC0−24 (µg.h l−1) (range)	7289(4757–11 314)	6549(5178–8539)	0.527	−10 [−36, 26]	3991(2018–8810)	0.0009	−45 [−60, −25]
Cmax (µg l−1)	448 ± 157	381 ± 72	0.294		235 ± 95	0.0004*
tmax (h) (range)	11.0(2.0–48.0)	6.0(3.0–12.0)			5.0(3.0–16.0)
t1/2 (h)	29.1 ± 8.5	33.4 ± 21.4			24.0 ± 6.8

^†Normal renal function vs moderate renal impairment.

^‡Normal renal function vs severe renal impairment.

^*Statistically significant.

^§Includes one patient with a solitary unexplained high concentration at last sampling point. Mean value excluding this point is 8883 (range 3288-31 068). This patient was not an outlier on repeat dosing. Normal renal function: CL_CR>80 ml min⁻¹. Moderate renal impairment: CL_CR 30–60 ml min⁻¹. Severe renal impairment: CL_CR <30 ml min⁻¹.

Table 4.

Pharmacokinetic parameters for the pioglitazone metabolite M-IV.

	Renal impairment
	Normal renal function (n = 6)	Moderate renal impairment (n = 9)			Severe renal impairment (n = 12)
	Mean ± s.d.	Mean ± s.d.	P-value*	% change vs normal renal function [CI]	Mean ± s.d.	P-value†	% changevs normal renal function [CI]
Single dose
AUC0–∞ (µg.h l−1) (range)	49 856(27 664–109 987)	39 489(22 130–62 562)	0.257	−21 [−48, 20]	27 991(14 030–47 061)	0.006	−44 [−62, −17]
Cmax (µg l−1)	632 ± 161	756 ± 163	0.163		687 ± 170		0.5107
tmax (h) (range)	12.0(10.0–12.0)	10.0(8.0–24.0)			12.0(8.0–24.0)
t1/2 (h)	60.4 ± 28.4	36.5 ± 21.6			32.0 ± 31.6
Repeated doses
AUC0−24 (µg.h l−1) (range)	25 706(14 914–39 114)	26 513(22 084–33 958)	0.840	3 [−25, 41]	21 080(11 158–39 514)	0.181	−18 [39, 10]
Cmax (µg l−1)	1402 ± 456	1364 ± 175	0.947		1151 ± 413		0.1654
tmax (h) (range)	11.0(4.0–16.0)	4.0(1.0–8.0)			6.0(3.0–12.0)
t1/2 (h)	26.9 ± 7.6	21.8 ± 4.0			23.6 ± 5.9

^*Normal renal function vs moderate renal impairment.

^†Normal renal function vs severe renal impairment. Normal renal function: CL_CR>80 ml min⁻¹. Moderate renal impairment: CL_CR 30–60 ml min⁻¹. Severe renal impairment: CL_CR <30 ml min⁻¹.

There were no serious adverse events reported in this study and no patients discontinued due to an adverse event. After 10 doses of pioglitazone 45 mg OD, adverse events occurred in four subjects in the normal renal function group, three patients in the moderate renal impairment group, and eight patients in the severe renal impairment group. A total of 47 adverse events occurred, of which eight were reported by six subjects during the single-dose treatment regimen and 39 adverse events were reported by 15 subjects during the multiple-dosing treatment regimen. Adverse events were mild to moderate in nature. Oedema occurred in one individual with normal renal function, one with moderate renal impairment and three with severe renal impairment. In each case it was of mild to moderate severity, and in only two cases was it described as continuous. In all cases except one, the oedema resolved or improved despite continued treatment with pioglitazone. There were no cases of heart failure. Hypoglycaemia did not occur. No clinically significant changes in laboratory tests (liver function, haematology, electrolytes and creatinine clearance) were observed in any subject during the study.

Discussion

Many patients with Type 2 diabetes have mild-to-moderate renal impairment as a complication of inadequate glycaemic control. Renal impairment may result in decreased elimination of the oral hypoglycaemic agents, increasing both the duration of drug action and the risk of hypoglycaemia [14]. Hence, the only treatment option for many patients has been insulin. Pioglitazone is a new oral agent used to treat Type 2 diabetes. It is a thiazolidinedione PPAR-γ activator, which improves insulin sensitivity [15]. Hypoglycaemia does not occur with insulin sensitizers in monotherapy [16] and the clearance mechanisms of pioglitazone are primarily hepatic in animals [17] and man. Thus, pioglitazone may be an important alternative in patients with renal impairment. In a previous bioavailability study [11], pioglitazone was shown to have an absolute bioavailability of>80%, a small volume of distribution (0.2–0.3 l kg⁻¹) and an estimated clearance of 1.7–4.2 l h⁻¹. Given that pioglitazone is almost exclusively eliminated by hepatic oxidation and is a drug with a low hepatic extraction ratio, its clearance will be sensitive to changes in the intrinsic capacity of the liver to metabolic drugs and the free fraction of drug in plasma.

This study examined the effect of renal impairment on the pharmacokinetics of pioglitazone and its metabolites after single and repeated oral doses of pioglitazone 45 mg in 27 subjects with different degrees of renal impairment. There was no statistically significant difference in the mean serum concentration of pioglitazone between those with normal and those with moderately impaired renal function. However, after single and multiple oral doses, the mean maximum serum concentrations of pioglitazone and its metabolites M-III and M-IV were reduced in patients with severe renal impairment compared with subjects with normal renal function. Examination of AUC values of pioglitazone and of metabolites M-III and M-IV indicated that the serum concentrations of metabolites M-III and M-IV were proportional to the serum concentrations of pioglitazone, irrespective of the degree of renal failure. This constant relationship is consistent with the assumption that intrinsic metabolic clearance mechanisms are unchanged in subjects with different degrees of renal function.

After repeated doses, the mean serum concentration of pioglitazone and its metabolites M-III and M-IV did not increase, indicating that renal impairment does not cause accumulation of pioglitazone or its metabolites.

Overall, subjects with normal renal function achieved higher serum concentrations of pioglitazone and its metabolites M-III and M-IV than patients with moderate or severe renal impairment. Mean serum concentrations of pioglitazone were lower with increasing renal impairment. This may be explained by reduced protein binding, which is common in patients with renal impairment, resulting in increased free pioglitazone [12]. For drugs such as pioglitazone, with a low hepatic extraction ratio, the overall rate of drug elimination is related to unbound plasma concentration, and the intrinsic capacity of the liver to metabolize the drug. Assuming that the intrinsic capacity of the liver remains unchanged, total clearance will increase, as was observed in this study [18]. However, free drug concentration would not be expected to change, although this was not measured in this study. This effect may not hold true for drugs with a high hepatic extraction ratio, where renal failure has been shown to influence hepatic clearance [19, 20]. Such a mechanism has been suggested on an empirical basis to be a problem for drugs metabolized extensively by CYP2D6, whereas drugs metabolized by CYP3A4 are less affected [19]. Pioglitazone is metabolized mainly by CYP3A4 and CYP2C8/9 [11]. An alternative mechanism to explain our results might be a change in systemic bioavailability caused by reduced gastrointestinal absorption, which has been demonstrated in patients with severe renal impairment for cyclosporine, ibuprofen and levocabastine [21–23]. This could only be shown by a study estimating bioavailability in renally impaired patients.

Pioglitazone was well tolerated. The adverse event profile of pioglitazone in patients with renal impairment was similar to that observed in those with normal renal function [16]. There appeared to be no relationship between adverse events and use of concomitant medication.

After oral administration, the major route of pioglitazone elimination is via hepatic metabolism. Pioglitazone is converted to metabolites M-III and M-IV, which are then excreted through the bile, and little pioglitazone clearance is attributable to renal mechanisms. The results of this study have confirmed that renal impairment has only minimal effects on pioglitazone pharmacokinetics.

In summary, no accumulation of pioglitazone and its metabolites M-III and M-IV in patients with renal impairment was found. Varying degrees of renal impairment resulted in minimal pharmacokinetic effects on pioglitazone, although mean serum values in patients with severe renal impairment were reduced. These effects probably do not warrant a change of dose in patients with severe renal impairment [12]. However, caution should still be observed in the use of any drug in patients with renal failure, and the dose of pioglitazone should be titrated in such patients.