Abstract
Aims
To assess whether hepatic impairment influences the pharmacokinetics of ziprasidone.
Methods
Thirty subjects with normal hepatic function or a primary diagnosis of clinically significant cirrhosis (Child-Pugh A or B) were enrolled into an open-label, multicentre, multiple-dose study. The subjects with chronic, stable hepatic impairment and the matched control subjects received ziprasidone 40 mg day−1, given orally with food, as two divided daily doses for 4 days and a single 20 mg dose on the morning of day 5. Pharmacokinetic variables were determined from multiple venous blood samples collected on days 1 and 5. Liver function was evaluated quantitatively using antipyrine.
Results
On day1 there were no statistically significant differences in the pharmacokinetics (AUC(0, 12 h), Cmax, tmax) of ziprasidone between the two groups. On day 5 there were no statistically significant differences in the Cmax or tmax for ziprasidone between the two groups. The mean AUC(0,12 h) for ziprasidone was statistically significantly greater in the hepatically impaired subjects compared with the normal subjects (590 ng ml−1 h vs 467 ng ml−1 h, P = 0.042). However, the AUC(0,12 h) increased by only 26% in the cirrhotic group compared with the matched control group. The ziprasidone λz in the subjects with normal hepatic function was statistically significantly greater than that in the hepatically impaired subjects (P< 0.001). There was no correlation between antipyrine λz and ziprasidone λz in the subjects with normal hepatic function or in those with hepatic impairment.
Conclusions
The findings of this study indicate that mild to moderate hepatic impairment does not result in clinically significant alteration of ziprasidone pharmacokinetics.
Keywords: ziprasidone, pharmacokinetics, hepatic impairment, metabolism
Introduction
Although the pharmacokinetics of ziprasidone do not vary significantly with age or sex in healthy individuals [1], and do not appear to be influenced by mild to moderate renal impairment [2], it is possible that they are altered by hepatic impairment. As the main site of oxidative metabolism of ziprasidone is in the liver, in particular by the action of CYP3A4 [3], hepatic impairment that results in sufficiently diminished functional capacity of the liver in general, and more specifically, of the cytochrome oxidase system, may increase systemic availability of the drug and reduce oral dosage requirements. As ziprasidone may be used to treat psychotic individuals who have concurrent hepatic impairment, it was therefore necessary to investigate specifically the impact of hepatic impairment on the pharmacokinetics of ziprasidone.
Here we report the findings of a study designed to assess the influence of hepatic impairment on the pharmaco-kinetics of ziprasidone.
Methods
Subjects
The subjects were adults (aged 18–75 years) with either normal hepatic function or hepatic impairment. Those with impaired hepatic function were matched by age (± 5 years) and weight (± 9 kg) with those with normal hepatic function. All subjects provided written informed consent.
The subjects with hepatic impairment had a clinical diagnosis of cirrhosis and satisfied the criteria of Class A or B of the modified Child-Pugh classification [4]. The presence of cirrhosis was confirmed by a previous liver biopsy specimen, by an imaging procedure which demonstrated a grossly nodular liver with portal collaterals and splenomegaly or by a scan showing portal hepatic reuptake with redistribution to spleen and bone marrow. None of the diagnostic records mentioned alcohol as the cause of cirrhosis. In all cases, hepatic impairment was chronic and stable. As such the subjects did not have an acute exacerbation of liver disease (defined as no worsening clinical signs of hepatic impairment, or by a greater than 50% worsening of the serum bilirubin concentration or prothrombin time in the preceding 3 months).
Protocol
This was an open-label, multicentre, multiple-dose study designed to compare the steady-state pharmacokinetics of ziprasidone in subjects with normal hepatic function with those in matched subjects with chronic, stable hepatic impairment. The study protocol was approved by the ethics committees of the Universities of Colorado, Texas and Washington.
Subjects who met all the entry criteria received oral doses of ziprasidone 40 mg day−1, given as two divided daily doses for 4 days and a single 20 mg dose on the morning of day 5. Ziprasidone was given, with 50 ml water, in the morning immediately after eating a standard breakfast and, approximately 12 h later, immediately after eating a standard dinner. All concomitant therapy was administered at least 2 h before ziprasidone dosing.
Assessments of antipyrine clearance and antipyrine metabolite formation were performed during screening. For this purpose, antipyrine 1000 mg was given orally with apple juice. The subjects were asked to empty their bladders immediately before antipyrine dosing.
Pharmacokinetic sampling
Venous blood samples for the determination of serum ziprasidone were collected immediately before and up to 12 h after the morning dose of ziprasidone on day 1, and up to 120 h after the morning dose of ziprasidone on day 5. Before the administration of ziprasidone on the morning of days 1–5, additional venous blood samples were taken for the determination of ziprasidone plasma protein binding.
Quantitative evaluation of liver function (antipyrine sampling)
Saliva samples for the determination of salivary antipyrine concentrations were collected from the subjects with normal hepatic function immediately before and up to 48 h after the antipyrine dosing, and from the hepatically impaired subjects immediately before and up to 96 h after the antipyrine dosing.
Twenty-four hour urine samples for the determination of antipyrine metabolite concentrations were collected on four occasions from the hepatically impaired subjects (0–24, 24–48, 48–72 and 72–96 h after antipyrine dosing) and two occasions (0–24 and 24–48 h after antipyrine dosing) from the subjects with normal hepatic function.
Salivary and urinary concentrations of antipyrine and its metabolites were determined using a standard h.p.l.c. assay [5, 6].
Pharmacokinetic assessments
Serum concentrations of ziprasidone were determined using a validated h.p.l.c. assay with solid-phase extraction and ultraviolet detection. The assay had a dynamic range of 1.0–250.0 ng ml−1[7]. Ziprasidone concentrations below the lower limit of quantification were assigned a value of 0 ng ml−1 in pharmacokinetic calculations.
The plasma binding of ziprasidone was determined by equilibrium dialysis and h.p.l.c. with ultraviolet detection assay with a 25-ng ml−1 lower limit of detection.
The maximum serum concentrations of ziprasidone (Cmax), and the earliest time at which Cmax occurred (tmax) were estimated directly from the experimental data. The area under the serum ziprasidone concentration–time curve from time zero to 12 h postdosing (AUC(0, 12 h)) was estimated using linear trapezoidal approximation. The terminal phase rate constant (λz) for ziprasidone was estimated using least-squares regression analysis of the serum concentration–time data obtained during the log-linear phase. The mean t½,z of ziprasidone was calculated as ln 2/mean λz. The percent of ziprasidone bound to plasma proteins (Fb) was calculated using the following equation:
 |
where Fb = Fraction of bound ziprasidone, Cpe = concentration of ziprasidone in plasma, Cbe = concentration of ziprasidone in buffer, Vpe= volume of plasma at equilibrium, Vpi = volume of plasma before dialysis [8].
Antipyrine assessments
The volume of distribution (Vd) of antipyrine was calculated as the ratio of the dose to the concentration in saliva extrapolated to time zero. The terminal phase rate constant (λz) for antipyrine was estimated using least-squares regression analysis of the saliva concentration–time data obtained during the log-linear phase, and the antipyrine clearance (CL) was calculated as the product of the λz and the Vd.
The three antipyrine metabolites investigated and the cytochrome (CYP) isoenzymes implicated in their metabolism were: norantipyrine (CYP2C9 and CYP1A2), 3-hydroxymethylantipyrine (CYP1A2 and CYP2C9) and 4-hydroxyantipyrine (CYP1A2 and CYP3A4) [9].
Statistical evaluation
The ziprasidone pharmacokinetic parameters were summarized using descriptive statistics. Geometric means and standard deviations were calculated for AUC(0,12 h) and Cmax. Arithmetic means and standard deviations were calculated for tmax, λz and percentage protein binding. Harmonic means were calculated for t½,z. The least-squares means statement in PROC GLM of SAS® was used to estimate the means and their variances and an analysis of variance (anova) was performed to test for a hepatic function group effect. As the study included only two groups, this was equivalent to performing Student’s t-tests to test for differences between the groups. A 5% level of significance was used throughout.
The antipyrine pharmacokinetic parameters were also summarized using descriptive statistics. Arithmetic means and standard deviations were calculated for λz, Vd and CL, and for the antipyrine renal clearance, norantipyrine formation clearance, 3-hydroxymethylantipyrine formation clearance and 4-hydroxyantipyrine formation clearance. Between-group differences for the antipyrine pharmacokinetic parameters were tested for in the same manner as for the ziprasidone pharmacokinetic parameters.
Results
Subjects
Thirty subjects entered the study and 26 were evaluable for ziprasidone pharmacokinetics; 13 in each group. All 26 of these subjects were included in the statistical comparisons of ziprasidone pharmacokinetics, excepting two subjects for whom Fb could not be determined, either due to a possible leak during equilibrium dialysis or poor chro-matography.
Ziprasidone pharmacokinetics
Mean serum ziprasidone concentrations for days 1 and 5 are shown as a function of time in Figures 1 and 2, respectively.
Figure 1.
Mean serum ziprasidone concentrations on day 1 in subjects with normal (----, n = 13) or impaired (Child-Pugh A cirrhosis (….., n = 7), Child-Pugh B cirrhosis (–––, n =6) and Child-Pugh A or B cirrhosis (-·-·, n =13) hepatic function.
Figure 2.
Mean serum ziprasidone concentrations on day 5 in subjects with normal (----, n =13) or impaired (Child-Pugh A cirrhosis (….., n =7), Child-Pugh B cirrhosis (–––, n =6) and Child-Pugh A or B cirrhosis (-·-·, n =13) hepatic function.
On day 1 there were no statistically significant differences in the pharmacokinetics (AUC(0, 12 h), Cmax, tmax) of ziprasidone between the group with normal hepatic function and the group with hepatic impairment (Table 1). Visual inspection of mean serum ziprasidone concentrations indicated that steady-state was reached by day 3 (data not shown).
Table 1.
Summary of pharmacokinetic parameters of ziprasidone on day 1.
| Hepatic impairment | |||||
|---|---|---|---|---|---|
| Pharmacokinetic parameter | Normal hepatic function (n =13) | Child-Pugh A or B (n =13) | Child-Pugh A (n =7) | Child-Pugh B (n =6) | Ratio of (AUC(0, 12h) and Cmax) or difference in (tmax) means (95% CI) |
| AUC (0,12h)a (ng ml−1 h) | 317±57 | 338±72 | 326±91 | 352±47 | 107% (91, 126) |
| Cmaxa (ng ml−1) | 52±14 | 50±11 | 47±12 | 54±9 | 96% (79, 118) |
| tmaxb (h) | 5±2 | 6±2 | 6±3 | 7±2 | 1 (0, 3) |
Geometric means and standard deviations.
Arithmetic means and standard deviations.
CI – confidence interval.
By day 5 there was a statistically significant increase in the mean AUC(0,12 h) for the cirrhotic group compared with the matched controls (95% CI: 101%, 158%: P = 0.0422). However, the mean AUC(0,12 h) in the cirrhotic group was increased by only 26% compared with the matched controls. In addition, when individual AUC(0,12 h) values were plotted, there was considerable overlap between the cirrhotic and matched control groups (Figure 3). There was no apparent difference in the AUC(0,12 h) values between subjects with Child-Pugh A cirrhosis and those with Child-Pugh B cirrhosis. A statistically significant difference between the λz for the cirrhotic group and that for the matched controls was observed (95% CI: −0.070, −0.028: P< 0.001) (Table 2). However, this corresponded to a mean increase of only 2.3 h in the t½,z (7.1 h vs 4.8 h). There was no apparent difference in the percent protein binding between the two groups.
Figure 3.
Individual AUC(0,12 h) values on day 5 in subjects with normal or impaired hepatic function.
Table 2.
Summary of pharmacokinetic parameters of ziprasidone on day 5.
| Hepatic impairment | |||||
|---|---|---|---|---|---|
| Pharmacokinetic parameter | Normal hepatic function (n = 13) | Child-Pugh A or B (n =13) | Child-Pugh A (n =7) | Child-Pugh B (n =6) | Ratio of (AUC(0, 12h) and Cmax) or difference in (tmax) means (λz and Fb) means (95% CI) |
| AUC (0, 12h)a (ng ml−1 h) | 467±110 | 590±184d | 558±195 | 628±178 | 126% (101, 158%) |
| Cmaxa (ng ml−1) | 68±17 | 71±20 | 67±21 | 76±20 | 105% (84, 130%) |
| tmaxb (h) | 4±1 | 4±2 | 5±2 | 4±2 | 1 (−1, 2) |
| λz (h−1) | 0.145±0.026 | 0.097±0.025e | 0.104±0.028 | 0.086±0.019 | −0.049 (−0.070, −0.028) |
| t1/2c,z (h) | 4.8 | 7.1 | 6.6 | 8.1 | – |
| Fb (% bound) | 99.92±0.03 | 99.87±0.08 | 99.84±0.11 | 99.91±0.04 | −0.05 (−0.1, 0) |
Geometric means and standard deviations.
Arithmetic means and standard deviations.
Calcualted as ln 2/mean λz.
Statistically significantly different from normal hepatic function group, P =0.042.
Statistically significantly different from normal hepatic function group, P<0.001. CI – confidence interval.
Antipyrine pharmacokinetics
Analysis of the pharmacokinetics of antipyrine demonstrated that the group with hepatic impairment did indeed have a level of hepatic function that was demonstrably poorer than that in the control group; as shown by the fact that the antipyrine mean λz in the subjects with normal hepatic function was statistically significantly greater than that in the hepatically impaired subjects (95% CI: −0.0384, −0.0156: P< 0.001) (Table 3). The mean antipyrine renal CL in the subjects with normal hepatic function was also statistically significantly greater than that in the hepatically impaired subjects (95% CI: −2824, −939; P< 0.001). As expected, there was no statistically significant difference between the antipyrine Vd in the two groups. When the antipyrine λz values of individual subjects in both treatment groups were plotted against their individual ziprasidone λz values, there was only a very weak correlation (r2= 0.395) (Figure 4). As in the ziprasidone pharmacokinetic analysis, there was no apparent difference between the group with normal hepatic function and that with hepatic impairment regarding the formation of antipyrine metabolites (Table 3).
Table 3.
Summary of antipyrine pharmacokinetic parameters.
| Pharmacokinetic parameter | Normal hepatic function (n=14) | Hepatic impairment (n=16) | Difference in means (95% CI) |
|---|---|---|---|
| λz (h−1)a | 0.0577 ± 0.0192 | 0.0306 ± 0.0107b | −0.027 (−0.0384; −0.0156) |
| Vd (lkg−1)a | 0.72 ± 0.22 | 0.62 ± 0.08 | −0.1 (−0.22; 0.02) |
| CL (mlh−1) | 3316 ± 1751 | 1434 ± 542b | −1881 (−2824; −939) |
| Antipyrine formation CL (h−1) | 0.0619 ± 0.0156 | 0.0619 ± 0.0127 | 0.0001 (−0.0105; 0.0106) |
| Norantipyrine formation CL (h−1) | 0.1829 ± 0.0927 | 0.1334 ± 0.1189 | −0.0495 (−0.1301; 0.0311) |
| 3-hydroxymethylantipyrine formation CL (h−1) | 0.1803 ± 0.1260 | 0.1382 ± 0.0926 | −0.0421 (−0.1241; 0.04) |
| 4-hydroxyantipyrine formation CL (h−1) | 0.4606 ± 0.2912 | 0.3694 ± 0.2282 | −0.0912 (−0.2856; 0.1032) |
Arithmetic means and standard deviations.
Statistically significantly different from normal hepatic function group, P<0.001.
CI – confidence interval.
Figure 4.
Comparison of individual antipyrine λz at screening with ziprasidone λz on day 5 in subjects with normal (▪) or impaired (Child-Pugh A (□), Child-Pugh B (○) cirrhosis) hepatic function.
Discussion
This open-label, multicentre, multiple-dose study, involving 13 subjects with normal hepatic function and 13 subjects with hepatic impairment, evaluated the influence of hepatic impairment on the pharmacokinetics of ziprasidone. The principal findings from this study indicate that mild to moderate hepatic impairment, resulting from clinically significant cirrhosis (Child-Pugh A or B), does not result in clinically significant alterations in the pharmacokinetics of ziprasidone at steady-state. The determination that mean antipyrine λz and CL values were highly statistically significantly lower in the hepatically impaired group than in the group with normal hepatic function, provides additional evidence to support the clinical diagnoses which indicated clinically relevant differential hepatic function between the two groups.
There were no statistically significant differences between the two groups in the three ziprasidone pharmacokinetic parameters assessed on day 1 (AUC(0, 12 h), Cmax, tmax), suggesting that mild to moderate hepatic impairment does not influence the first-dose pharmacokinetics of ziprasidone. Even when the group with hepatic impairment was divided into two subgroups according to Child-Pugh classification, hepatic impairment did not appear to influence the first-dose pharmacokinetics of ziprasidone.
As with the first-dose pharmacokinetics, there were no statistically significant differences between the two groups with respect to the Cmax or tmax assessed at steady state on day 5. The mean AUC(0,12 h) value on day 5 in the hepatically impaired group was marginally larger than that in the group with normal hepatic function (26%; 590 vs 467 ng ml−1h), and this difference only just reached statistical significance (P = 0.042). There was a statistically significant difference in the λz for ziprasidone on day 5, but this corresponded to an increase of only 2.3 h in the t½,z in the hepatically impaired group. The small increase in exposure observed in the cirrhotic group is likely to be related to a decrease in metabolic clearance of ziprasidone. This postulate is supported by the relatively longer t½,z for this group. Hepatic function, expressed as the antipyrine λz, only weakly correlated (r2= 0.395) with the ziprasidone λz. This further supports the pharmacokinetic findings that suggest that hepatic impairment has no clinically significant effect on ziprasidone.
As the two primary metabolites of ziprasidone, ziprasidone sulphoxide and ziprasidone sulphone, are formed by the action of CYP3A4 [3] it is possible that the gut is implicated in the elimination of ziprasidone. CYP3A4 is present in the gut and much of the first-pass elimination of orally administered substrates of this isoenzyme is due to metabolism in the gut wall [10, 11]. Indeed, CYP3A4 accounts for 70% of the CYP enzymes in the gut but only 30% of those in the liver [12]. Although unproven, it is therefore reasonable to speculate that CYP3A4-mediated metabolism of ziprasidone may occur in the gut and thereby explain why hepatic impairment does not influence the pharmacokinetics of ziprasidone as much as might otherwise be expected.
As hepatic impairment can lead to the altered production of plasma proteins, and ziprasidone is highly bound to plasma proteins [13, 14], it is noteworthy that the mean ziprasidone Fb values on day 5 in the two groups were almost identical numerically and were not statistically significantly different (99.92%vs 99.87%).
The pharmacokinetic findings of this study contrast with those from a study in which the pharmacokinetics of the atypical antipsychotic, risperidone, and its active metabolite, 9-hydoxyrisperidone, in healthy subjects were compared with those in individuals with hepatic impairment [14]. In that study, the presence of cirrhosis was associated with statistically and clinically significant decreases in the unbound fractions of risperidone and 9-hydoxyrisperidone, and a dramatically reduced Cmax for 9-hydroxyrisperidone. As a consequence, the authors of this study recommended that risperidone be titrated carefully in cirrhotic patients.
In conclusion, the findings of this study indicate that mild to moderate hepatic impairment does not have a clinically significant impact on the steady-state pharmacokinetics or safety of ziprasidone.
References
- 1.Wilner KD, Tensfeldt TG, Baris B, et al. Single- and multiple-dose pharmacokinetics of ziprasidone in healthy young and elderly volunteers. Br J Clin Pharmacol. 2000;;49(Suppl. 1):15S–20S. doi: 10.1046/j.1365-2125.2000.00148.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Aweeka F, Jayesekara D, Morton M, et al. The pharmacokinetics of ziprasidone in subjects with normal and impaired renal function. Br J Clin Pharmacol. 2000;;49(Suppl. 1):27S–33S. doi: 10.1046/j.1365-2125.2000.00150.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Prakash C, Kamel A, Cui D, et al. Identification of the major human liver cytochrome P450 isoform responsible for the primary metabolites of ziprasidone and prediction of possible drug interactions. Br J Clin Pharmacol. 2000;;49(Suppl. 1):35S–42S. doi: 10.1046/j.1365-2125.2000.00151.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Pugh RNH, Murray-lyon IM, Dawson JL, et al. Transection of the oesophagus for bleeding varices. Br J Surgery. 1973;60:646–649. doi: 10.1002/bjs.1800600817. [DOI] [PubMed] [Google Scholar]
- 5.Teunissen MWE, Meerburg van er Touen JE, Vermeulen NPE, et al. Automated HPLC determination of antipyrine and its main metabolites in plasma, saliva and urine including 4,4′-dihydroxyantipyrine. J Chromatogr Biomed Appl. 1983;278:367–378. [PubMed] [Google Scholar]
- 6.Danhof M, deGroot van de vis Breimen DD. Assay of antipyrine and its primary metabolites in plasma, saliva and urine by high pressure liquid chromatography and preliminary results in man. Pharmacology. 1978;18:210–223. doi: 10.1159/000137254. [DOI] [PubMed] [Google Scholar]
- 7.Janiszewski JS, Fouda HG, Cole RO. Development and validation of a high-sensitivity assay for an antipsychotic agent, CP-88,059, with solid-phase extraction and narrow-bore high-performance liquid chromatography. J Chromatogr. 1995;668:133–139. doi: 10.1016/0378-4347(95)00071-p. [DOI] [PubMed] [Google Scholar]
- 8.Boudinot FD, Jusko WJ. Fluid shifts and other factors affecting plasma protein binding of prednisolone by equilibrium dialysis. J Pharm Sci. 1984;6:774–780. doi: 10.1002/jps.2600730617. [DOI] [PubMed] [Google Scholar]
- 9.Engel G, Hofmann U, Heidemann H, et al. Antipyrine as a probe for human oxidative drug metabolism: identification of the cytochrome P450 enzymes catalysing 4-hydroxyantipyrine, 3-hydroxymethylantipyrine and norantipyrine formation. Clin Pharmacol Ther. 1996;59:613–623. doi: 10.1016/S0009-9236(96)90001-6. [DOI] [PubMed] [Google Scholar]
- 10.Lown KS, Kolars JC, Thummel KE, et al. Interpatient heterogeneity in expression of CYP3A4 and CYP3A5 in small bowel. Lack of prediction by the erythromycin breath test. Drug Metab Dispos. 1994;22:947–955. [PubMed] [Google Scholar]
- 11.Fromm MF, Busse D, Kroemer HK, et al. Differential induction of prehepatic and hepatic metabolism of verapamil by rifampin. Hepatology. 1996;24:976–801. doi: 10.1002/hep.510240407. [DOI] [PubMed] [Google Scholar]
- 12.Kolars JC, Lown KS, Schmiedlin-ren P, et al. CYP3A4 gene expression in gut epithelium. Pharmacogenetics. 1994;4:247–259. doi: 10.1097/00008571-199410000-00003. [DOI] [PubMed] [Google Scholar]
- 13.Pacifici GM, Viani A, Taddeucci-brunelli G, et al. Effects of development, ageing and renal and hepatic insufficiency as well as hemodialysis on the plasma concentrations of albumin and α1-acid glycoprotein: implications for binding of drugs. Ther Drug Monit. 1986;8:259–293. doi: 10.1097/00007691-198609000-00003. [DOI] [PubMed] [Google Scholar]
- 14.Snoeck E, Van Peer A, Sack M, et al. Influence of age, renal and liver impairment on the pharmacokinetics of risperidone in man. Psychopharmacology. 1995;122:223–229. doi: 10.1007/BF02246543. [DOI] [PubMed] [Google Scholar]