bioavailability and reversible metabolism of prednisone and prednisolone in man

BIOPHARMACEUTICS & DRUG DISPOSITION, VOL. 15, 163-172 (1994)

BIOAVAILABILITY AND REVERSIBLE METABOLISM OF PREDNISONE AND

PREDNISOLONE IN MAN

VARUN GARG AND WILLIAM J. JUSKO*

Department of Pharmaceutics, School of Pharmacy, State University of New York at Buffalo. Buffalo, New York 14260. U.S.A.

ABSTRACT

The pharmacokinetics of prednisone and prednisolone was examined in 12 healthy male subjects to assess the bioavailability and the parameters of reversible metabolism between the two steroids. After an oral prednisone dose of 0.8mg kg-' and an intravenous prednisolone dose of 0.66 mg kg-', the bioavailability was found to be about 62%. The fraction of the dose recovered in the urine as the hydroxylated metabolites of prednisone and prednisolone was lower after the oral prednisone dose, suggesting that poor absorption of prednisone was the main cause of the low bioavailability. There was a high degree of interconversion between prednisone and prednisolone with 76% of the dose being recycled. The formation clearance of prednisolone from prednisone is much greater than the formation clearance of prednisone from prednisolone or the irreversible elimination clearances of the two steroids. The possible dose dependences of bioavailability and interconversion may be important factors in prednisolone therapy.

KEY WORDS Bioavailability Reversible metabolism Prednisone Prednkolone Pharmacokinetics

INTRODUCTION

Prednisone and prednisolone are synthetic glucocorticoids used as anti- inflammatory and immunosuppressive agents for a wide variety of disease states. Prednisone is both a prodrug and a metabolite of the active drug prednisolone. Prednisolone is most commonly administered as prednisone tablets. However, prednisone formulations for intravenous (i.v.) administration are not approved and the bioavailability of prednisone tablets is generally based on the resulting prednisolone concentrations. With low doses of prednisone tablets, the systemic availability seems to be essentially complete.IJ With high-dose tablets, however, some clinical studies have indicated a lower bioavailability (about 60%) of predni~olone.~-~ It was also reported that the systemic availability of prednisone was reduced when 0.8 mg kg-I of prednisone was given as a single oral dose as opposed to four doses of 0.2 mg kg-1.3 Ferry et a1.l conducted a

*Addressee for correspondence.

CCC 0142-2782/94/020163- 10 0 1994 by John Wiley & Sons, Ltd.

Received 30 June I993 Accepted 20 October I993

164 V. GARG AND W. J . JUSKO

four-way crossover study in healthy subjects in which each subject was dosed with 10 mg of prednisone and prednisolone as oral tablets and as 0.5 hour i.v. infusions. They found that the bioavailability values based on prednisone concentrations differed significantly from the corresponding values based on prednisolone concentrations, suggesting that reversible metabolism of prednisone and prednisolone complicates bioavailability assessment.

The purpose of the present study was to investigate whether the decreased bioavailability of prednisolone from high-dose prednisone tablets results from a high first-pass metabolism of prednisolone or from poor absorption. We assessed the fraction of dose excreted in the urine (unchanged or as metabolites) after prednisone tablets in comparison with that obtained after intravenous prednisolone administration. Another objective was to assess the role of interconversion in the disposition of prednisone and prednisolone. In the presence of appreciable interconversion, the traditional methods of calculating bioavailability and other pharmacokinetic parameters are no longer valid. Hence we determined whether appreciable interconversion occurs between prednisone and prednisolone.

MATERIALS AND METHODS

Subjects

Twelve normal healthy male subjects aged 18-45 years and weighing within 10% of their ideal body weight (mean 79-6 kg) volunteered to participate in this study. All subjects were non-smokers and did not take other medications for the two weeks before and during the study.

Experimental protocol

This was a double-blind, randomized two-way crossover study with a seven- day washout period between the two steroid treatments. Each subject fasted overnight (10 h) and in the morning received (with 120ml water) either 0.8 mg kg-I of oral prednisone tablets (Deltasone; The Upjohn Co., Kalamazoo, MI) or i.v. prednisolone sodium phosphate (Hydeltrasol; Merck Sharp & Dohme; West Point, PA) equivalent to 0.66 mg kg- of the free base. The oral tablet dose was correct to the nearest 2.5 mg (the smallest Deltasone tablet). The mean oral prednisone dose was 63 - 5 f 7 . 0 mg and the mean i.v. prednisolone dose was 52.2 k 5 . 8 mg. Following the dose, subjects maintained the fast for another 4 h after which they received a standardized meal.

Blood samples (8ml) were collected at various time points from the contralateral arm using a catheter. When oral prednisone was administered, blood samples were collected immediately prior to the dose and then 0.25,0-5, 1, 1.5,2, 3 , 5 , 9 , 12, 14, 16, 18, 20,22 and 24 h following the prednisone dose. When i.v. prednisolone was administered, samples were collected immediately

PREDNISONE AND PREDNISOLONE IN MAN 165

prior to the dose and then 5 , 10, 15, and 30 min and 1, 1.5, 2, 3, 5 , 9, 12, 14, 16, 18,20,22, and 24 h following the prednisolone dose. The samples were placed in heparinized glass tubes and centrifuged to extract the plasma, which was stored at - 20 "C until assay. Urine was collected prior to and until 24 h after each steroid dosing. The total volume of urine during each 24 h period was recorded and a 30ml aliquot stored at - 20 "C until analysis.

Analytical procedures

Plasma concentrations of prednisone and prednisolone were determined by a previously described high-performance liquid chromatographic (HPLC) m e t h ~ d . ~ Plasma protein binding of prednisolone was determined in each sample by ultrafiltration at 37 "C (Centrifree; Amicon, Danvers, MA) after spiking each plasma sample with trace amounts of 3H-prednisolone. A liquid scintillation counter (Packard model 1900CA; Downers Grove, IL) was used to obtain the free fraction. Urine samples were assayed simultaneously for prednisone, prednisolone, and their major hydroxylated metabolites, viz. 6&, 2001-, 20&hydroxyprednisolone, and 20@-hydroxyprednisone by HPLC.8

Serum and urine creatinine levels were also measured using an automated analyzer (Paramax; Baxter Diagnostics Inc., Irvine, CA) and creatinine clearance was determined for 24 h periods both before and after steroid administration.

Pharmacokinetic analysis

The area under the plasma concentration-time curve (AUC) extrapolated to infinity was calculated for total prednisolone, total prednisone, and unbound prednisolone using a computer program for LaGrange polynomial interp~lation.~ This program also generated the terminal half-life (T%) of prednisone and prednisolone. The bioavailability (fl of prednisone tablets was calculated as follows:

F= (AUC,,/AUCiV) Dosq,/Dose, (1)

where the subscripts indicate the treatment (oral prednisone or intravenous prednisolone).

For drugs undergoing reversible metabolism, the elimination of the parent drug (p) and metabolite (m) is determined by four fundamental clearances. These are the irreversible elimination clearances of the parent and metabolite (CLlo and CLzO), and the interconversion clearance of the drug to the metabolite (CLI2) and vice versa (CL21). Based on mass balance relationships, the following equations have been derived for estimating these clearances:I0

CLlo = (DoseP AUCZ - Dosem AUCL)/ (2)

(AUCF AUCI - AUC; AUCF)


CLzO = (Dosem AUC; - DoseP AUCF)/ (3)

(AUCE AUC; - AUCg AUCF)

CL12 = DosemAUCL/(AUCE AUC: - AUCL AUCF) (4)

CL2] = DoseP AUCF/(AUCE AUC: - AUCL AUCF) ( 5 )

where the superscripts refer to the dosed compound, and the subscripts refer to the measured compound. It should be noted that the use of equations (2)- ( 5 ) to calculate the four fundamental clearances requires intravenous administration of the parent drug and the metabolite. In the present study, since an oral dose of the parent drug prednisone was given, the equations were modified by multiplying DoseP by the bioavailability (F) obtained using equation (1). In order to assess the relative importance of the interconversion process, the recycled fraction (RF) was calculated as follows: lo

The metabolites excreted in the urine were expressed as percentage of dose, on a molar basis, since the oral prednisone and intravenous prednisolone doses were different. These were compared between the two treatment groups by a two-tailed paired t test. Differences at p c 0.05 were considered to be statistically significant.

RESULTS

The plasma concentration-time profile of prednisone and prednisolone in a representative subject is shown in Figure 1. As seen from this figure, the prednisolone concentrations exceeded the concentrations of prednisone at all times. Prednisolone was formed rapidly from its phosphate ester as well as from prednisone. In the terminal phase, the concentrations of prednisone and prednisolone declined in parallel.

The mean pharmacokinetic parameters of prednisone and prednisolone in the 12 subjects are listed in Table 1. The bioavailability of prednisone tablets based on total and unbound prednisolone concentrations was 62% and 53%. However, based on prednisone concentrations from both oral and intravenous doses, the availability was much higher (84%). The ratio of the AUCs of prednisolone and prednisone was significantly higher after the intravenous prednisolone dose. The terminal half-lives of prednisolone and prednisone in either treatment groups were similar to each other.

The percentages of dose recovered as the various metabolites (or unchanged) after oral prednisone or intravenous prednisolone are given in Table 2. This


3 a U

B E U

8 8 u

0 4 8 12 16 u) 24

Time, h Figure 1 . Plasma concentration-time profiles of prednisolone (0, 0 ) and prednisone ( A , v ) after i.v. prednisolone (closed symbols) or oral prednisone (open symbols) dosing in a representative subject

table also shows the relative importance of the hydroxylated metabolites of prednisone and prednisolone. Thus 20/3-hydroxyprednisolone seems to be the most important metabolite, followed by 60-hydroxyprednisolone. On the other hand, 200-hydroxyprednisone was found to be a minor metabolite. The total dose recovered in the urine unchanged and as hydroxylated metabolites was about 42% after the intravenous prednisolone treatment and 25% after the oral prednisone treatment. For each of the metabolites, the percent of dose recovered in the urine was less after oral prednisone than the intravenous prednisolone treatment. This suggests that the main cause of the low systemic availability

Table 1. Pharmacokinetic parameters of prednisone and prednisolone (Mean k SD, n = 12)

Parameter Prednisone p.0. Prednisolone i.v. (0.8 mg kg-l) (0.66mg kg-')

Total prednisolone AUCo-, (ng h ml-I) 3348 & 513 4459k 636 T% (h) 2.82 k 0- 35 2.67 k 0.33 AUC prednisolone/AUC prednisone 7.96 f 1.2 10.63 kO.64 Bioavailability 0- 62 f 0-08 -

Unbound prednisolone AUCo-, (ng h ml-I) 730f 161 1 1 4 4 f 195 T% (h) 2-63 k 0.36 2-44k 0.35 AUC prednisolone/AUC prednisone 1 -72 f 0.31 2.74 k 0.40 Bioavailability 0.53 f 0.10 -

T , (h)

Prednisone AUCo-, (ng h m1-I) 426 & 67 420 k 59

2-91 f 0.49 Bioavailabilit y 0.84 k 0.13

2.92 k 0-40 -

168 V. GARG AND W. J. JUSKO

Table 2. Percentages of dose recovered in the urine unchanged or as metabolites after administration of oral prednisone (PSON) or i.v. prednisolone (PSLN): (mean ? SD, n = 12)

Metabolite After PSON PO After PSLN i.v. p value Ratio

PSON 1.55 20.52 2.12?0*79 <0.05 0.73 PSLN 9. lo? 3 *27 20*70?6.15 < O . o 0 0 5 0.44 20P-(OH)PSLN 5.51 ? 1.77 7*31?2-11 <Om05 0.75 6P-(OH)PSLN 4-70? 1.82 6-29? 2-43 NS 0.75 20a-(OH)PSLN 3-24? 1.30 5*14? 1.77 <Om01 0.63 2OP-(OH)PSON 0.55 k 0.23 0.62k0.26 NS 0.89 Total 24.65 & 7.80 41.88+11.72 <0.005 0.59

of prednisolone from oral prednisone is poor absorption rather than a high first- pass effect. The creatinine clearances following the oral prednisone dose and the intravenous prednisolone dose were not significantly different (1 33 k 27 and 138 k 30 ml min- l) suggesting that the renal function was not different between the two collection periods.

The mean values of the four fundamental clearances of prednisone and prednisolone (based on total prednisolone AUCs) and the recycled fraction are given in Table 3. As seen from this table, the formation clearance of prednisolone from prednisone (CL12) is about tenfold greater than the formation clearance of prednisone from prednisolone (CLZ1). The irreversible elimination clearance of prednisone (CL,,) was found to be smaller than the other three clearances. This implies that most of the prednisone is converted to prednisolone before being eliminated, and agrees with the urine data, as prednisone and 200- hydroxyprednisone were found to be present only in small fractions of the dose. The recycled fraction (RF) was estimated to be 0.76, suggesting that a great deal of interconversion takes place between prednisone and prednisolone before they are irreversibly eliminated.

DISCUSSION

Our study confirms previous finding^,^-^ which indicated that at high doses of oral prednisone, the bioavailability of prednisolone is reduced. Since we were

Table 3. Fundamental clearances of prednisone and prednisolone and the recycled fraction

Parameter Mean k S.D.

CL,, (ml min-l) 53*9?51 CL,, (ml min-') 196 2 39 CL,, (ml min-') 8822 ? 6445 CL,, (ml min-') 836 ? 606 RF 0.76 & 0.09


able to quantify the major metabolites of these steroids in the urine, we determined the percentage of the dose recovered unchanged and as metabolites after i.v. prednisolone and oral prednisone doses to understand the cause of the low bioavailability. Based on the consistently low metabolite fractions in the urine after the oral dose compared to the intravenous dose, we speculate that poor absorption rather than high first-pass metabolism may be the main cause of the observed low bioavailability of prednisone tablets. If high first- pass metabolism were responsible, we would have seen a higher dose fraction of metabolites after the oral dose than after the i.v. dose.

It should be noted that only about 42% of the i.v. dose and about 25% of the oral dose was recovered in the urine unchanged and as hydroxylated metabolites as shown in Table 2. Several other metabolites of prednisone and prednisolone have been identified, which may account for the remainder of the dose.I2 However, our assay was not capable of detecting these metabolites.s In order to determine the contribution of poor absorption versus first-pass metabolism in causing a low bioavailability of prednisone tablets more precisely, a larger fraction of the dose must be accounted for. Our conclusions are based on the assumption that other metabolites are not recovered in a much greater fraction in the urine following the oral dose compared to the i.v. dose.

Frey et d 3 found a higher AUC of total and unbound prednisolone when 0.8mg kg-1 of oral prednisolone was given compared to oral prednisone. From their data, the bioavailability of prednisolone from prednisone tablets and prednisolone tablets can be estimated to be 61% versus 80% (based on total prednisolone) or 55% versus 90% (based on unbound prednisolone). If high first-pass metabolism were the main reason for poor systemic availability, one would expect prednisolone tablets to have less bioavailability since prednisone first converts to prednisolone before being sequentially metabolized. Since prednisolone is less lipophilic than prednisone, it is conceivable that high-dose prednisolone tablets will be absorbed better as a result of faster dissolution compared to prednisone tablets.

It should be noted that for drugs undergoing reversible metabolism, such as prednisone and prednisolone, the estimation of bioavailability by conventional methods used here may also be improper. For the true estimation of bioavailability of the parent (Fp) and the metabolite (Fm), the following equations have been derived: l3

Fp = (CLlo AUC;.Po + CLzo AUCCP0) DoseP.iv/

(CLlo AUC;.’’ + CL20 AUC2”) DosePIPO (7)

F,,, = ( C L , ~ A U C ~ O + CLzO AUC:3Po) Dosem.iv/

(cL,,, AUC;’’~ + CL20 AUC;.”) Dosem-po


However, these calculations require that both the parent and metabolite be given intravenously in order to calculate CLlo and CL2[). In order to assess the relative error caused by the calculation of bioavailability using equation (1) instead of equation (7) or (8), we used the data of Ferry et al. to calculate the values of F with the three equations. These authors conducted a four-way crossover study in normal subjects in which each subject was given lOmg of prednisone and prednisolone as oral tablets and as 0.5 h i.v. infusions. The bioavailability of prednisone tablets was 0 - 92 using equation (1) (prednisolone concentrations and i.v. prednisolone as reference) and 0.94 using equation (7). Similarly, for prednisolone tablets, F was estimated to be 0.92 (prednisolone concentrations and i.v. prednisolone as reference) using equation (l), while equation (8) yielded a value of 0.97. However, when prednisone concentrations were used in equation (l), the F values of prednisone tablets were 1 -08 (i.v. prednisolone reference) and 0.64 (i.v. prednisone reference). The corresponding values for prednisolone tablets were 1-27 and 0.76. This indicates that prednisolone rather than prednisone concentrations yield bioavailability values closer to the true values given by equations (7) and (8).

Based on urine concentrations, the bioavailability of prednisolone from prednisone was estimated to be (ratio of percentage of dose excreted as prednisolone) about 44% (Table 2). However, the ratio of the total percentage of dose excreted (unchanged or as metabolites) yields a value of 59'70, which is similar to the AUC ratio. This discrepancy may be due to non-linear renal clearance of prednisolone.2

The conversion of prednisolone sodium phosphate to prednisolone after i.v. administration is assumed to be rapid and complete. Our assay methods were not adequate to measure prednisolone phosphate and if the hydrolysis of this ester were incomplete, the estimates of bioavailability would be overestimates. Based on studies in the literature, however, prednisolone phosphate seems to be rapidly hydrolyzed in V ~ V O . ~ J ~ J ~

Interconversion allows for efficient removal of high concentrations of the active drug, yet conservation of the drug in the inactive form in the body. Thus reversible metabolism helps augment biological exposure to prednisolone (and some other steroids) by serving as a dampening compartment to reduce fluctuation in serum concentrations and to prolong the half-life because of recycling. The recycled fraction (RF) represents the probability of a molecule being converted to its metabolite (prednisone to prednisolone or vice versa) and being back converted at least once. The larger the value of RF, the greater is the role of reversible metabolism in the disposition of the drug and its metabolite. Thus a value of 0.76 means that reversible metabolism plays a major role in the disposition of prednisone and prednisolone in man. From the data of Ferry et al,' the value of RF was estimated to be 0.52. In rats, this value was found to be much lower, between 0.11 and 0.16, increasing with the dose."

Finally, as pointed out by Ebling and Jusko,lO for drugs undergoing appreciable reversible metabolism, the more relevant clearance parameters are


the ones that are described in equations (2)-(5) and listed in Table 3. For comparison, data from the Ferry et al.' study yielded mean values of CLlo = 165, CLm = 106, CL12 = 1178, and CLZl = 157 ml min- I . The differences in these values may be due to the lower dose used in the latter study but both investigations show that CLI2 dominates over the other clearances. The effect of disease states and drug interactions on prednisone and prednisolone can be evaluated more meaningfully if their effect on bioavailability and the fundamental clearance parameters are known. Since in this paper it has been shown that reversible metabolism between prednisone and prednisolone is appreciable, future studies on these drugs and other corticosteroids must consider this factor.

ACKNOWLEDGEMENTS

This work was supported by Grant No 41037 from the National Institutes of General Medical Sciences, NIH, and by Pfizer, Groton, CT, U S A . The authors thank Ms Suzette Mis for technical assistance.

REFERENCES

1. J. J. Ferry, A. M. Horvath, I. Bekersky, E. C. Heath, C. F. Ryan, and W. A. Colburn, Relative and absolute bioavailability of prednisone and prednisolone after separate oral and intravenous doses. J. Clin. Pharmacol., 28, 81-87 (1988).

2. J. Q. Rose, A. M. Yurchak, and W. J. Jusko, Dose dependent pharmacokinetics of prednisone and prednisolone in man. J. Pharmacokinet. Biopharm., 9, 389-417 (1981).

3. F. J. Frey, M. K. Ruegsegger, and B. M. Frey, The dose-dependent systemic availability of prednisone: one reason for the reduced biological effect of alternate-day prednisone. Br. J. Clin. Pharmacol., 21, 183-189 (1986).

4. F. J. Frey, F. F. Horber, and B. M. Frey, Altered metabolism and decreased efficacy of prednisolone and prednisone in patients with hyperthyroidism. Clin. Pharmacol. Ther., 44,

5. R. M. Zurcher, B. M. Frey, and F. J. Frey, Impact of ketoconazole on the metabolism of prednisolone. Clin. Pharmacol. Ther., 45, 366-372 (1989).

6. A. E. Stuck, B. M. Frey, and F. J. Frey, Kinetics of prednisolone and endogenous cortisol suppression in the elderly. Clin. Pharmacol. Ther., 43, 354-362 (1988).

7. J. Q. Rose and W. J. Jusko, Corticosteroid analysis in biological fluids by high-performance liquid chromatography. J. Chromatogr., 161, 273-280 (1979).

8. V. Garg and W. J. Jusko, Simultaneous analysis of prednisone, prednisolone, and their major hydroxylated metabolites in urine by high-performance liquid chromatography. J. Chromatogr., 567, 39-47 (1991).

9. M. L. Rocci Jr and W. J. Jusko, LAGRAN program for area and moments in pharmacokinetic analysis. Comput. Programs Biomed., 16, 203-216 (1983).

10. W. E. Ebling and W. J. Jusko, The determination of essential clearance, volume, and residence time parameters of recirculating metabolic systems: the reversible metabolism of methyl- prednisolone and methylprednisone in rabbits. J. Phamacokinej. Biopham., 14,557-599 (1986).

11. M. L. Huang and W. J. Jusko, Nonlinear pharmacokinetics and interconversion of prednisone and prednisolone in rats. J. Pharmacokinet. Biopharm., 18, 401-421 (1990).

12. F. J. Frey, Kinetics and dynamics of prednisolone. Endocr. Rev., 8, 453-473 (1987). 13. H. Cheng and W. J. Jusko, Pharmacokinetics of linear reversible metabolic systems. In Advanced

Methoak of Pharmacokinetics and Pharmacodynamic Systems Analysis, D. D'Argenio (Ed.), Plenum, New York, (1991), pp. 13-20.

510-521 (1988).


14. J. C. Melby and M. St Cyr, Comparative studies on absorption and metabolic disposal of water- soluble corticosteroid esters. Metabolism, 10, 75-82 (1961).

15. P. A. Greenberger, M. J . Chow, A. J. Atkinson, J. J . Ambre, and R. Patterson, Comparison of prednisolone kinetics in patients receiving daily and alternate-day prednisone for asthma. Clin. Pharmacol. Ther., 39, 163- 168 (1 986).

bioavailability and reversible metabolism of prednisone and prednisolone in man

Documents