oral contraceptive effects on methylprednisolone pharmacokinetics and pharmacodynamics*

Oral contraceptive effects on methylprednisolone pharmacoldnetics and pharmacodynamics

Objective: Oral contraceptive (OC) steroids alter the disposi t ion o f numerous drugs, including corticosteroids. We investigated the pharmacokinetics and pharmacodynaanics o f methylprednisolone. Methods: Twelve women (six women used OC steroids and six women did not) received intravenous methylprednisolone (0.6 m g / k g ideal body weight) . Methylprednisolone disposi t ion was assessed f rom plasma concentrations. Pharmacodynamic parameters measured were plasma cort isol , whole b lood histamine (reflecting basophils) , and b lood helper T lymphocytes. Results: Methylprednisolone clearance was significantly decreased in the women who used OC steroids (0.298 versus 0 .447 L / h r / k g ) , result ing in a longer el iminat ion half-life (2.20 versus 1.72 hours). Wi th use o f indirect response models, significant differences were observed with the cortisol and basophil responses. A larger value for the concentrat ion tha t inhibi ts the zero-order product ion rate by 50% (0.37 versus 0.11 n g / m l ) was observed in the women who used OC steroids for suppression ofcor t i so l secretion, indicat ing less sensitivity to the suppressive effects o f methylprednisolone. Greater net suppression o f basophils was observed in the users o f O C steroids (area under the response curve, 694 versus 401 ng • h r / m l ) . N o differences were observed for helper T-cell responses. Conclusion: O C steroids appear to inhibi t methylprednisolone metabolism. However , mixed changes in several responses occur, indicat ing that women can probably receive similar doses o f methylprednisolone irrespective o f OC steroid use. (CLIN PHAaMACOL TRIER 1996;59:312-21. )

K a t h r y n L. Slayter, P h a r m D , El izabe th A. Ludwig , P h a r m D , Kim H . Lew, P h a r m D , El l io t t M i d d l e t o n , Jr. , M D , James J. Ferry , P h D , and Will iam J. Jusko, P h D Buffalo, N.Y., and Kalamazoo, Mich.

The use of oral contraceptive (OC) steroids is appreciable among women, thereby exposing them to the risks of multiple-drug therapy. ~ These steroid

From the Departments of Pharmaceutics, Pharmacy, and Medi- cine, Schools of Pharmacy and Medicine, State University of New York at Buffalo, and the Departments of Pharmacy and Medicine, Buffalo General Hospital, Buffalo, and the Upjohn Company, Kalamazoo.

Supported by grant GM 24211 from the National Institutes of General Medicine Sciences, National Institutes of Health (Bethesda, Md.), and a grant from The Upjohn Company (Kalamazoo, Mich.).

Presented at the Ninety-sixth Annual Meeting of the American Society for Clinical Pharmacology and Therapeutics, San Diego, Calif., March 15-17, 1995.

Received for publication Feb. 23, 1995; accepted Aug. 31, 1995. Reprint requests: William J. Jusko, PhD, Hochstetter 565, School

of Pharmacy, State University of New York at Buffalo, Buffalo, NY 14260.

Copyright © 1996 by Mosby-Year Book, Inc. 0009-9236/96/$5.00 + 0 13/1/68998

312

hormones share the same hepatic microsomal en- zyme system responsible for metabolism of certain other drugs and therefore inhibition or induction may occur. 2-4 Drugs eliminated by oxidative mechanisms, such as antipyrine, diazepam, caffeine, and theophylline, often exhibit a decreased metabolic clearance and a prolonged elimination half-life (tl/2) when administered concurrently with OC steroids. 5-s In contrast, some drugs that are primarily conjugated by glucuronidation, such as acetaminophen, oxaze- pam, and lorazepam, have increased clearances in users of OC steroids. 9"m

Combination OC steroids alter the pharmacokinetics of prednisolone. ~1~3 Boekenoogen et al. 13 reported an approximate 40% reduction in free prednisolone clearance after 40 mg prednisolone doses in women taking OC steroids compared with control subjects. The increase in prednisolone concentrations occurred through two mechanisms, a decrease in intrinsic clearance and an increase in

CLINICAL PHAI(MACOIA)GY & THEIkAf'EUTICS VOLUMF. 59, NLIMBI-II. 3 Slayter et al. 313

plasma transcortin. Prednisolone binds to transcortin in a concentration-dependent manner, and this protein is elevated in women who use estrogen- containing OC steroids) 4 However, a study involving coadministration of OC steroids and fluocortolone reported no interaction, indicating some specificity in OC steroid effects) 5

Methylprednisolone is an important anti- inflammatory and immunosuppressive agent that differs from prednisolone because of the addition of a 6et-methyl group. This group prevents transcortin binding, and methylprednisolone is therefore bound primarily to albumin in a linear manner, n~' Methyl- prednisolone is also more susceptible than other corticosteroids to certain drug interactions, such as those caused by macrolide antibiotics 17 and ketoconazole. ~ The effect of the use of OC steroids on methylprednisolone pharmacokinetics and pharmacodynamics has not yet been investigated.

Pharmacodynamic models have been developed that characterize the suppressive effects of corticosteroids on plasma cortisol concentrations, blood basophils (measured by whole blood histamine), and helper T lymphocyte counts) '~2~ These models were applied to examine whether these representative adrenosuppressive, anti-inflammatory, and immunosuppressive effects differ in women who use OC steroids.

METHODS

Subjects. Six healthy nonobese (within 25% of ideal body weight [IBW]) women between the ages of 30 and 36 years who had been receiving OC steroids for at least 6 months were included in the study. All subjects had their OC steroid regimens changed (with permission of their primary care prac- titioner) to Triphasil 28 (Wyeth-Ayerst Laboratories [Philadelphia, Pa.], 0.05 mg levonorgestrel and 30 &g ethinyl estradiol [days 1 through 6], 0.075 mg levonorgestrel and 40 ~g ethinyl estradiol [days 7 through 11], and 0.125 mg levonorgestrel and 30 lxg ethinyl estradiol, [days 12 through 21]) for at least 1 month before the study and for its duration. Sub- jects were considered to be healthy on the basis of complete medical histories, physical examinations, and blood chemistry and hematologic profiles before entry into the study. No subject was a smoker or was taking any medication that was known to inter- act with methylprednisolone before or during the study. No subject had a history of allergic disease or hypersensitivity to corticosteroids. The comparison

group was six healthy women who were part of another recent study 21 and who were not using OC steroids. The study was approved by the Buffalo General Hospital Institutional Review Board, and informed consent was obtained from each subject.

Procedures. The study was divided into two phases (baseline and methylprednisotone), with each phase separated by a washout period of 1 week. Subjects receiving OC steroids were randomly assigned to initiate either phase and were then crossed over to the remaining phase the following week. Each subject participated in the study between days 7 and 21 of her OC steroid regimen. Each subject in the comparison group underwent the methylprednisolone phase during the luteal phase (the 2-week period after ovulation during which estradiol and progesterone concentrations in the blood are relatively elevated and constant) and underwent the baseline phase during the follicular phase of her menst rua l cycle. 21 Endogenous estradiol and progesterone concentrations were measured on the methylprednisolone study day.

On each study day, a 20-gauge angiocatheter was inserted into an arm vein of each subject for blood sample collections. The device was kept patent with diluted heparin (10 units/mI) solution.

During the baseline phase, plasma cortisol, whole blood histamine, and helper T lymphocyte measurements were obtained over a 24-hour period in the absence of methylprednisolone. Blood samples (5 ml) were drawn from the catheter into heparin- containing tubes every 2 hours for 24 hours. Whole blood (300 Ixl) was removed from the heparinized collection tubes and stored at - 20 ° C for later determination of whole blood histamine concentrations. The remaining blood was centrifuged to har- vest plasma for later determination of cortisol concentrations. The samples were stored at - 20 ° C until assayed. An additional 4 ml blood was drawn into collection tubes that contained ethylenedi- aminetetraacetic acid at 0, 2, 4, 8, 12, 16, 20, and 24 hours for helper T lymphocyte analysis.

For the methylprednisolone phase, subjects were required to fast from 10 pm the evening before and for 2 hours after receiving the methytprednisolone dose. Each subject received an intravenous bolus dose (0.6 mg/kg IBW) of methylprednisolone sodium succinate (Solu-Medrol, The Upjohn Com- pany, Kalamazoo, Mich.) in the arm contralateral to which blood was sampled. Approximately 5 ml blood was collected at 0, I/4, 1/2, 1, 2, 3, and 4 hours and then every 2 hours until 24 hours, and at 28 and

314 Slayter et al. CI.INICAI, I'HAI(MA(X)I.OGY & THI(II2d~ELTTI(2S

MARCH I t)t)('l

32 hours after drug administration to determine plasma methylprednisolone and cortisol concentrations and whole blood histamine concentrations. The blood samples were processed as described above, except that two further T-cell determinations were obtained at 1 and 32 hours. In addition, the plasma harvested from the samples drawn up to 6 hours were immediately placed in an ice bath to prevent in vitro hydrolysis of the methylprednisolone succinate ester.

Assays. Plasma methylprednisolone and cortisol concentrations were determined by the HPLC method of Lew et al. 2~ and Ebling et aI. 22 The lower and upper limits of quantitation were 10 and 750 ng/ml. Intraday coetficients of variation (CV) for cortisol and methylprednisolone were 3.3% and 2.3% at 40 ng/ml and 0.7% and 0.5% at 600 ng/ml. Interday CV values were 2.4% and 4.2% at 40 ng/ml and 5.8% and 4.3% at 600 ng/ml. Cortisol concentrations below the limit of quantitation were assigned a value of 5 ng/ml for pharmacodynamic modeling purposes. Plasma protein binding of methylprednisolone was determined by ultrafiltration at 37 ° C with use of the Centrifree Micropartition Sys- tem (Amicon Inc, Beverly, Mass.).

Whole blood histamine was analyzed by means of a commercial radioimmunoassay procedure (Immu- notech International, Westbrook, Maine). The lower limit of determination was 0.05 ng/ml. Intra- day CV values for low- and high-quality control samples were 3.0% and 5.8%. Corresponding interday CV values were 6.7% and 11.9%.

Total leukocyte counts were performed by an automated hemocytometer (Coulter Counter S-Plus IV, Coulter Electronics lnc; Hialeah, Fla.) on the whole blood samples collected for helper T-cell analysis. Lymphocyte and monocyte proportions were determined microscopically, and the total number of circulating lymphocytes per cubic milli- meter was determined. The whole blood samples were then lysed, reacted with mononuclear antibody (Coulter Cytostat anti-CD4), and analyzed on an automated flow cytometer (FACS 440, Becton Dick- inson Co., Cockeysville, Md.). The total number of circulating helper T cells was determined by multi- plication of the proportion of fluorescent cells by the number of lymphocytes. Estradiol and progesterone were measured by radioimmunoassay (Coat-A- Estradiol and Coat-A-Progesterone, Diagnostic Products Corp., Los Angeles, Calif.).

Pharmacokinetics. The nonlinear least-squares regression computer program PCNONLIN (SCI

Software Inc., Lexington, Ky.) was used to fit the following equation to methylprednisoione plasma concentration (CMp) versus time (t) curves:

kf- Dose CMF'- V" (kf - k) (e-k" - ek' '') (1)

in which kf is the first-order rate constant for the formation of methylprednisolone from methylprednisolone sodium succinate, k is the elimination rate constant for methylprednisolone, and V is the volume of distribution of methylprednisolone. Methyl- prednisolone clearance was calculated from kV and tl/2 from 0.693/k.

Pharmacodynamics. The area under the effect curve (AUC) for each pharmacodynamic parameter for the baseline and methylprednisolone phases from 0 to 24 hours was calculated with use of the LaGrange interpolation. 23 These values were used to calculate the the area between the baseline and effect curve (ABEC). The AUC suppression ratio (AUCsR) was determined as follows:

~4

AUCsI~ = AUC ~%,~li,,~ (2)

Pharmacodynamic models 19"2t for cortisot, whole blood histamine (basophils), and CD4+ cells were used. The following is the general equation:

dR d~ = k~,,. l(t) - tg,,,, • R (3)

in which R is response, kin is a zero-order production rate, and ko~ t is the first-order removal rate. The inhibition function I(t) uses the concentration of methylprednisolone (CMI,) as predicted by each subject's pharmacokinetic parameters (equation 1) and the concentration that inhibits kin by 50% (ICs0):

C M p

l(t) = 1 CMp + ICs0 (4)

This assumes that methylprednisolone can fully inhibit each response measure. The figures and tables have specific symbols for applying the model to each response.

The cortisol and T-cell models use a cosine function to reflect the circadian influx (kin) into blood:

ki, = R,, + Rh'cos" (t - t,.).-2~- (5)

in which R m is the mean, Ru is the amplitude, t is the clock time within a 24-hour cycle, t z is the peak time of the circadian function in relation to time zero

CLINICA[, PHARMACOLOGY & THEIU~.PEUTI(k~. VOLUME 59, NUMBER 3 Slayter et al. 315

Table I. Subject characteristics and endogenous hormone concentrations

Comparison OC steroids Control

Age (yr) TBW (kg) IBW (kg) Dose/IBW (mg/kg) Dose/TBW (mg/kg) Days since start of menses* Estradiol concentration (pg/ml)* Progesterone concentration (ng/ml)*

32.7 +-- 2.8 67.4 ___ 12.9 57.3 _+ 8.1 0.60 +- 0.00 0.52 --+ 0.05

19, 18, 11, 16, 11, 18 <20, <20, <20, 467, 108, <20

0.3, <0.1, 0.2, 0.4, 0.5, 1.1

37.0 - 6.0 56.2 - 5.3 53.0 - 4.1 0.60 +- 0.01 0.57 +-- 0.04

22, 24, 20, 17, 18, 23 343, 128, 35, 9I, 103, <20 15.4, 12.1, 2.5, 4.8, 12.4, 0.7

Data are mean values _+ SD. OC, Oral contraceptive; TBW, lotal body weighl; IBW, ideal body weight, *Parameters are for subjects I through 6. respectively.

( 8 AM) in the study, and 2"rr/24 converts clock time into radians.

The pharmacodynamic models for both baseline (where I(t) -- 1) and methylprednisolone phases (equation 4) were fitted simultaneously with the PCNONLIN program.

Statistical analysis. The nonparametric Wilcoxon rank sum test was used to detect statistically significant differences at a p -<- 0.05 level for all pharmacokinetic and pharmacodynamic parameters between subjects in the OC steroid group and control subjects.

RESULTS Subject characteristics. Six women receiving oral

contraceptives completed the study and were compared with six control subjects (Table I). Based on data of Dunn et al, 24 patients received doses based on IBW, unless actual body weight was lower. There were no significant differences in the weights and doses in the two groups. The endogenous estradiol and progesterone levels are also listed. The values for the control group are in agreement for the normal values (estradiol, 15 to 260 pg/ml; progesterone, 2.5 to 28 ng/ml) exhibited during the luteal phase except for the estradiol value in control subject 1, which is more suggestive of a midcycle (ovulation) value. The estrogen values for OC steroid subjects 4 and 5 and the progesterone value for subject 6 were higher than expected in women who received OC steroids. However, much variation in both estradiol and progesterone occurs in such women, zs

Pharmacokinetics. Fig. I illustrates the plasma concentration-time profile, showing linear decline of methylprednisolone concentrations in a representative OC steroid subject and a control subject. The users of OC steroids eliminated methylprednisolone

more slowly, as indicated by longer ttn values and higher plasma concentrations.

The mean pharmacokinetic parameters for both groups are listed in Table II. Clearance (normalized for IBW) was significantly slower (33%) in the subjects receiving OC steroids. There were no differences in the absolute V or V normalized for IBW. Consistent with the clearance values, the elimination rate was significantly lower (25%) and the tt/2 was prolonged (30%) in subjects in the OC steroid group. No difference in protein binding was observed between the two groups, with the subjects in the OC steroid group exhibiting a mean ___ SD free fraction of 23.7% +_ 2.0% versus 24.7% +_ 2.9% in the control subjects. These data are consistent with previous binding results for methylprednisolone in humans. ~f'

Pharmacodynamics. Representative cortisol profiles for baseline and methylprednisolone phases are presented in Fig. 2. The plasma cortisol concentrations exhibited a normal circadian variation, with peaks occurring in the morning and nadirs in the late evening during baseline conditions. After administration of methylprednisolone, the normal circadian rhythm of cortisol was suppressed, as shown by the rapid first-order decline in cortisol concentrations followed by prolonged suppression. Cortisol concentrations approached baseline by 32 hours.

Table III lists the mean pharmacodynamic parameters for the cortisol model. These data excluded two control subjects and two women in the OC steroid group. One subject in the OC steroid group had a large discrepancy between her initial cortisol concentrations for the two study phases, apparently due to stress. The suppressive response after the dose of methylprednisoione was thus incomplete. One control subject showed no cortisoi suppression

316 Slayter et at.

m

E 03 E

O I - < r~ t--- z uJ 0 z O o

1000

100

10 0

OC

, I , I , 1 4 8 12

TIME, h

Fig. 1. Plasma methylprednisolone versus time profiles for a typical user of oral contraceptive (OC) steroids and control subject. Symbols represent experimental data; lines indicate the least-squares regression fittings. Time 0, Time of the administration of the dose; kf, first-order rate constant for the formation of methylprednisolone from methylprednisolone sodium succinate; CMp, methylprednisolone concentration; V, volume of distribution of methylprednisolone; CL, clearance.

after methylprednisolone administration. A second control subject and subject in the OC steroid group had peak baseline levels in the midafternoon, which are consistent with a reversed sleep/wake cycle and was unlike the typical peak cortisol levels achieved in the morning. This may have occurred because the subject receiving OC steroids worked a night shift before the morning of the study. However, the whole blood histamine and helper T-cell data for the four women excluded were typical and were therefore included. The R m value, which is indicative of the mean baseline input rate of cortisol secretion, although not significantly different, was 40% higher in subjects in the OC steroid group. The amplitude (Rb) was higher (70%) in the OC steroid group, indicating greater circadian fluctuation in cortisol concentrations, although this did not reach statistical significance. It was also observed that the elimination constant (k¢) was lower in subjects in the OC steroid group (35%; p < 0.05), indicating slower disposition of cortisol. These values reflect the find-

CLIN,CAI. PHA,la~..IACOLOGY & "FHEILa.PEUT,CS MARCH L996

CORTISOL

3oo[ 200 iI ~

(tz) Ulcso

Control

El

~=Ec~ 1oo " .o. m o. ~," ° 'm . . . . " "

< nl I--- 400 z

300

o o

200

100

0

OC O

" . O O . - - .

O O . b O • • •

,

8 16 24 32

TIME, h

Fig. 2. Plasma cortisol concentration versus time profiles for a woman receiving oral contraceptives (bottom panel) and a control subject (top panel). Symbols show experimental data; lines represent the fittings to the depicted pharmacodynamic model. Open ~ymbols and broken line, Baseline phase; closed symbols and solid line, methylprednisolone phase. R~,,r,, Net cortisol production rate; Rb, amplitude; t~, peak time of the circadian function in relation to time zero (8 AM) in the study; Rm, mean; IC5,, concentration that inhibits kin by 50%; kc, elimination constant.

ings of higher cortisol concentrations observed over- all in the OC steroid group.

The IC5, value, which measures intrinsic sensitivity to the suppressive effects of methylprednisolone on adrenal cortisol secretion, was significantly higher in subjects in the OC steroid group (0.37 -+ 0.20 versus 0.11 _ 0.09 ng/ml; p < 0.05). The remaining parameters were not significantly different.

Blood basophils (measured as whole blood histamine) for a representative subjects are depicted in Fig. 3. Baseline blood histamine concentration profiles are typically flat, with occasional variability. After methylprednisolone administration, there is a

CLINICAl. PHARMACOLOGY & THEIL~.PEUTICS VOLUME 59. NUMBER 3 Slayter et al. 317

HISTAMINE (BASOPHILS)

75,

50

25

Z - 0 0

,<

}_ 100 Z q UJ ~-) 75 Z O O 50 (

25

kr

I CMP' lc£O

Control

i I.l r l 0 17 OC] O

.... = I , l , I ~ I

OC

0

p.o._.o . . . . 0 . 0 . 0 o o

\ooo

o 0 s 16 24 32

TIME, h

Fig. 3. Whole blood histamine concentration versus time profiles for a woman receiving oral contraceptives and a control subject. Panels, symbols, and fines are defined in the legend of Fig. 2. H, Histamine concentration; k h, efflux rate constant; k,, blood influx rate constant.

steady decline in blood histamine concentrations until methylprednisolone is eliminated, with a nadir occurring at about 8 to 10 hours and return to baseline by 32 hours.

The pharmacodynamic parameters for whole blood histamine generated during the simultaneous computer fitting of baseline and methylprednisolone phases are summarized in Table IV. The difference in ABEC between subjects receiving OC steroids and control subjects was significant (694 _+ 225 versus 401 +_ 199 ng- hr/ml; p < 0.05), The AUCsR (which accounts for differing initial concentrations) was smaller in subjects in the OC steroid group (0.375 -4- 0.128 versus 0.546 ___ 0.081;p < 0.05). Both of these measures indicate greater net suppression of histamine in women receiving OC steroids. There were no significant differences in any remaining parameters, although a trend for an increased produc-

T-HELPER LYMPHOCYTES

E

1200

800

"~ 4oo o

0 I - o <

I-- 12oo z uJ o z 0 8oo o

400

o o

I Control

I I I I

OC

O ' ' O ' o - . o •

0 . " 0 " . 0 - . ° - - . . . . . . . .

I I I ~ 8 16 24 32

TIME, h

Fig. 4. Blood helper T lymphocyte concentration versus time profiles for a woman receiving oral contraceptives and a control subject. Panels, symbols, and lines are defined in the legend of Fig. 2. TH, Helper T-cell concentration; kt, blood etflux rate constant.

tion rate was observed in subjects in the OC steroid group.

Helper T lymphocytes in two subjects shown in Fig. 4 exhibit a circadian rhythm during baseline conditions, with peak concentrations occurring in the evening and nadirs during the day. Methylpred- nisolone produces a monoexponential decline in helper T cells, with a nadir occurring between 5 and 8 hours and a return to baseline by 24 hours. The pharmacodynamic parameters obtained from modeling these lymphocyte data did not differ between groups (Table V).

DISCUSSION Pharmacokinetics. OC steroids have an apprecia-

ble effect on methylprednisolone disposition. Our subjects exhibited a reduced clearance (33%) and a longer t~/2, which is consistent with data for pred-

3 1 8 Slayter e t al. CLINICAL PHARMACOLOGY & THERAPEUTICS

MARCH 1996

Table II. Pharmacokinetic parameters for methylprednisolone

Parameter Definition OC steroids Control p Value

AUC (ng- hr/ml) Area under the plasma 2145 _+ 604 1443 _+ 426 <0.05 concentration-time curve

CL (L/hr) Clearance 17.1 +- 5.3 23.3 +- 5.8 NS CL/IBW (L/hr/kg) CL/ideal body weight 0.298 _+ 0.080 0.447 + 0.124 <0.05 V (L) Volume of distribution 53.4 +- 11.3 56.1 +- 7.7 NS V/IBW (Idkg) Volume/IBW 0,94 +- 0.17 1.07 -+ 0.18 NS k (hr-~) Elimination rate constant 0.310 -- 0.050 0.413 -+ 0.070 <0.05 t~/2 (hr) Half-life 2.20 _+ 0.33 1.72 - 0.29 <0.05 kf (hr-t) Formation rate constant 4.82 -+- 2.48 3.14 - 2.02 NS F u (%) Unbound fraction 23.7 + 2.0 24.7 + 2.9 NS

Data are mean values _+ SD (n = 6). NS, Not significant.

Table III. Pharmacodynamic parameters for cortisol suppression


R,, (ng/ml/hr) Mean production rate 24.5 --- 8.0 18.0 --+ 3.7 NS R~ (ng/ml/hr) Amplitude of R m 22.6 -+ 6.6 13.3 --- 2.6 NS t~ (24-hour clock) Time of maximum 6.76 --- 1.07 7.02 -+ 1.65 NS k c (hr -~) Elimination rate constant 0.180 +-- 0.040 0.276 __- 0.045 <0.05 IC.so (ng/ml) Drug concentration producing 50% inhibition 0.37 --- 0.20 0.11 _-+ 0.09 <0.05 ABEC (rig- hr/ml) Area between the baseline and drug effect curve 1491 -+ 724 933 --- 348 NS AUCsR Suppression ratio 0.579 --- 0.155 0.430 -+ 0.133 NS

Data are mean values -+ SD (n = 4).

nisolone (40%). 1°"13 Of interest, the elimination of cortisol was also slowed in users of OC steroids, as indicated by the smaller k e values (Table III).

The users of OC steroids have a diminished capability to eliminate methylprednisolone. Most corticosteroids are eliminated predominantly in the liver, probably by mixed-function oxidases involving the P4503A isozymes. 26"27 The oral contraceptive Triphasil contains both estrogen and progesterone components. Both compounds have a 17ot-ethinyl side chain, which has been shown to bind irrevers- ibly to rat hepatic cytochrome P450, impairing its metabolic capability. 28 The oxidation of the ethinyl group of ethinyl estradiol may produce reactive in- termediates that lead to the destruction ("suicide inactivation") of the heme group of cytochrome P4503A. 29 Other human data suggest an ability of ethinyl estradiol to impair drug metabolism through a competitive inhibition, with the progesterone com- ponent having no effect on oxidized substrates. 3° Methylprednisolone elimination is impaired by various drugs that inhibit P4503A substrates (erythro- mycin, troleandomycin, and ketoconazole), 3x thus suggesting that the present mechanism is similar. It is also of interest that methylprednisolone clearance

is affected by gender, and one study 32 found increased CYP3A activity in human hepatic micro- somes from women.

No clear relationship was observed between endogenous hormone levels and methylprednisolone clearance in users of OC steroids, although subjects 4 and 5 in the OC steroid group had the highest endogenous hormone levels and largest methylprednisolone clearances.

Pharmacodynamics. The effects of oral contraceptives on the pharmacodynamics of methylprednisolone were examined. The elevation of R m and a decrease in kc account for the higher mean cortisol concentrations observed in users of OC steroids (136 versus 65 ng/ml). Other studies also report increased endogenous cortisol concentrations in women receiving OC steroids. Ira3 OC steroids decrease the apparent clearance of total cortisol, partly because of increased corticosteroid binding globulin concentrations. 33 Methylprednisolone is not bound to corticosteroid binding globulin26; thus its protein binding was unaffected (Table I). An increase in free cortisol also occurs, suggesting an impairment in the intrinsic clearance of cortisol. 34

Uncontrolled clinical observations suggest that

CLINICAl. PHAIIddACOLOGY & THERAPEUTICS VOLUME 59, NUMBER 3 Slay ter e t al. 3 1 9

Table IV. Pharmacodynamic parameters for basophil trafficking (whole blood histamine)


ktr ~ (ng/ml/hr) Blood influx rate constant 12.26 - 8.22 kh (hr-~) Efflux rate constant 0.245 _+_ 0.107 I(25o (ng/ml) Drug concentration producing 50% inhibition 4.47 --- 3.44 ABEC (ng. hr/ml) Area between the baseline and drug effect curve 694 +- 225 AUCsR Suppression ratio 0.375 _ 0.128

7.98 _-. 3.66 NS 0.227 + 0.055 NS

6.64 --- 5.30 NS 401 _+ 199 <0.05

0.546 --. 0.081 <0.05

Data are mean values +_ SD (n = 6).

Table ¥. Pharmacodynamic parameters for helper T-cell trafficking


Rm (cells/mm3/hr) Mean blood influx 498 -+ 165 RI, (cells/mm~/hr) Amplitude of R m 114 --- 5l tz (24-hour clock) Time of maximum 0.0 +- 1.3 k t (hr-~) Blood efflux rate constant 0.573 -+ 0.064 IC5o (ng/ml) Drug concentration producing 50% inhibition 18.10 -+- 15.76 ABEC (cells. hr/mm 3) Area between the baseline and drug effect curve 9060 +- 4188 AUCsR Suppression ratio 0.508 - 0.129

525 _ 117 NS 156 -+ 79 NS

21.6 +- 2.0 NS 0.613 _+ 0.102 NS 14.87 +_ 20.36 NS 9261 .+- 4513 NS

0.575 _ 0.123 NS

Data are mean values _+ SD (n = 6).

the addition of estrogens to cortisol therapy produces a reduction in the requirement for corticosteroids in various chronic inflammatory skin diseas- es. 35 Furthermore, in subjects with diabetes, the glucosuric effect of glucocorticoids was enhanced by ethinyl estradiol. 36 These studies also document in- creases in unbound cortisol concentrations in the plasma of estrogen-treated subjects.

The cortisol 1(25o value was significantly higher in the subjects in the OC steroid group compared with the control group, indicating a decreased sensitivity to the effects of methylprednisolone on cortisol suppression. This finding is consistent with the hypoth- esis that estrogens or corticosteroid binding globulin concentrations cause a resetting of the level at which the pituitary or hypothalamus is inhibited by free cortisol, 37 and perhaps this effect also occurs with exogenous corticosteroid administration. Estrogens and corticosteroids are members of a receptor su- perfamily and the possibility of receptor interactions may exist. Turner 38 showed that estrogen was asso- ciated with a reduction in corticosteroid receptor number and increased pituitary cellular transcortin. They concluded that the pituitary gland of the female rat had reduced sensitivity to the negative feedback by circulating glucocorticoids. This was explained by the observation that corticosteroid re-

ceptors in female rats are subject to a 20% down- regulation by circulating estrogen.

Oral contraceptives reduce methylprednisolone clearance and increase the IC5o for adrenal suppression; both changes are toward values found previ- ously in male subjects. 2~ However, the changes are variable because methylprednisolone clearance is nearly identical in users of OC steroids (0.298 L/hr/ kg) and men (0.288 L/hr/kg), basophil and helper T-cell IC50 values remain unaffected, but the IC5o for cortisol suppression remains relatively low in users of OC steroids (0.37 ng/ml) compared with men (1.69 ng/ml).

The basophil trafficking model showed a statistically significant increase in ABEC and a decrease in AUCsR in the OC steroid group. These changes, indicating greater net basophil suppression in the OC steroid group, and the similar IC5o values are consistent with the slower elimination of methylprednisolone. There was also a trend for an increased production rate, which is reflective of the higher baseline histamine concentration. Thus, for a fully suppressive methylprednisolone dose, greater net suppression can occur in users of OC steroids.

To assess whether methylprednisolone had any different immunobiological effects in women receiving OC steroids, helper T-celt responses were eval-

3 2 0 Slayter et aL CLINICAL PHARMACOI.OGY & THEKAPEUTI(kS

MARCH 1996

uated. No statistically significant difference in parameters was observed. This may have occurred for several reasons. There was more variability in the data, compared with the o ther dynamic parameters , and fewer sample points were collected. This is consistent with our previous studies. 21 However , the results contrast with data from Frey and Frey, 39 who used a bioassay to examine the influence of prednisolone in p lasma from users of OC steroids on mixed lymphocyte cultures. In vitro inhibition was more pronounced in women taking O C steroids.

This study shows a reduced clearance of methylprednisolone in users of OC steroids. Interestingly, there is a paradoxical change in selected biological responses. For cortisol, the IC5o value was higher, showing less sensitivity to the suppressive effects on cortisol; however, net responses were similar and values from some subjects had to be deleted. Sub- jects in the OC steroid group had greater basophil suppression, which is reflected in a decreased A U C suppression ratio and increased ABEC, This was probably caused by the lower clearance of methylprednisolone. It appears that OC steroids can cause impaired methylprednisolone disposition and di- verse effects on cortisol, basophil, and T-cell suppression.

From this study it is apparent that women receiving OC steroids have a similar net response to cortisol suppression compared with control subjects; therefore no changes in dose may be indicated. However, women receiving OC steroids do have an increased net basophil suppression response and an unaffected helper T-cell response, implying equal or enhanced therapeutic responses to methylprednisolone.

The technical assistance of Ms. Nancy Pyszczynski and Ms. Mary Bushway was greatly appreciated, as was the clinical assistance provided by Ms. Jan Pennington and the IV Nursing Team at the Buffalo General Hospital. The Chemistry and Hematology Laboratories at the Buf- falo General Hospital provided clinical chemistry, cell counting, and flow cytometry measurements.

References 1. Back DJ, Orme ML. Pharmacokinetic drug interac-

tions with oral contraceptives. Clin Pharmacokinet 1990;18:472-84.

2. Teichman AT. Influence of oral contraceptives on drug therapy. Am J Obstet Gynecol 1990;163:2208-13.

3. Guengerich FP. Metabolism of 17a-ethynylestradiol in humans. Life Sci 1990;47:1981-8.

4. Mackinnon MM, Sutherland E, Simon FR. Effects of

ethinyl estradiol on hepatic microsomal proteins and the turnover of cytochrome P-450. J Lab Clin Med 1977;90:1096-106.

5. O'Malley K, Stevenson IH, Crooks J. Impairment of human drug metabolism by oral contraceptive steroids. CUN PHARMACOL THER 1972;13:552-7.

6. Abernethy DR, Greenblatt D J, Divoll M, Arendt R, Ochs HR, Shader RI. Impairment of diazepam metabolism by low-dose estrogen-containing oral- contraceptive steroids. N Engl J Med 1982;306:791-2.

7. Abernethy DR, Todd EL. Impairment of caffeine clearance by chronic low-dose oestrogen-containing oral contraceptives. Eur J Clin Pharmacol 1985;28:49.25-8.

8. Tornatore KM, Kanarkowski R, McCarthy TL, Gard- ner M J, Yurchak AM, Jusko WJ. Effect of chronic oral contraceptive steroids on theophylline disposition. Eur J Clin Pharmacol 1982;23:129-34.

9. Mitchell MC, Hanew T, Meredith CG, Schenker S. Effects of oral contraceptive steroids on acetaminophen metabolism and elimination. CLIN PHARMACOL THER 1983;34:48-53.

10. Patwardahan RV, Mitchell MC, Johnson RF, Schen- ker S. Differential effects of oral contraceptive steroids on the metabolism of benzodiazepines. Hepa- tology 1983;3:248-53.

11. Meffin PJ, Wing MH, Sallustio BC, Brooks PM. Al- terations in prednisolone disposition as a result of oral contraceptive use and dose. Br J Clin PharmacoI 1984; 17:655-64.

12. Legler UF, Benet LZ. Marked alterations in dose- dependent prednisolone kinetics in women taking oral contraceptives. CLIN PHARMACOL THER 1986;39:49-5-9.

13. Boekenoogen S J, Szefler S J, Jusko WJ. Prednisolone disposition and protein binding in oral contraceptive users. J Clin Endocrinol Metab 1983;56:702-9.

14. Mills IH, Schedl HP, Chen PS, Bartter C. The effect of estrogen administration on the metabolism and protein binding of hydrocortisone. J Clin Endocrinol Metab 1960;20:515-28.

15. Legler UF. Lack of impairment of fluocortolone disposition in oral contraceptive users. Eur J Clin Phar- macol 1988;35:101-3.

16. Szefler S J, Ebling WF, Georgitis JW, Jusko WJ. Methylprednisolone versus prednisolone pharmacokinetics in relation to dose in adults. Eur J Clin Phar- macol 1986;30:323-9.

17. Szefler SJ, Rose JQ, Ellis EF, Spector SL, Green AW, Jusko WJ. The effect of troleandomycin on methylprednisolone elimination. J Allergy Clin Immunol 1980;66:447-51.

18. Glynn AM, Slaughter RL, Brass C, D'Ambrosio R, Jusko WJ. Effects of ketoconazole on methylprednisolone pharmacokinetics and cortisol secretion. CLIN PHARMACOL THEn 1986;39:654-9.

19. Wald JA, Salazar DE, Cheng H, Jusko WJ. Two- compartment basophil trafficking model for methyl-

CI.INICAI. PHAILMAC()IA)(;Y & THEILa, PEU'I'ICS VOLUME 59, NUMBI'~R X Slayter et al. 321

prednisolone pharmacodynamics. J Pharmacokinet Biopharm 1991; 19:521-36.

20. Fisher LE, Ludwig EA, Jusko W J. Pharmacoimmu- nodynamics of methylprednisolone trafficking of helper T lymphocytes. J Pharmacokinet Biopharm 1992;20:319-31.

21. Lew KH, Ludwig EA, Milad MA, et al. Gender-based effects on methylprednisolone pharmacokinetics and pharmacodynamics. CuN PHARMACOL THER 1993;54: 402-14.

22. Ebling WF, Szefler S J, Jusko WJ. Analysis of cortisol, methylprednisolone and methylprednisolone hemisucci- nate: absence of effects of troleandomycin on ester hydrolysis. J Chromatogr 1984;305:271-80.

23. Rocci ML, Jusko WJ. LAGRAN program for area and moments in pharmacokinetic analysis. Comput Programs Biomed 1983;16:203-16.

24. Dunn TE, Ludwig EA, Slaughter RL, Camara DS, Jusko WJ. Pharmacokinetics and pharmacodynamics of methylprednisolone in obesity. CUN PHARMACOL THER 1991;49:536-49.

25. Tafurt CA, Sobrevilla LA, Estrada R. Effects of progestin-estrogen combination and progestational contraceptives on pituitary gonadotropins, gonadal steroids, and sex hormone-binding globulin. Fertil Sterit 1980;33:261-6.

26. Jusko WJ, Ludwig EA. Corticosteroids. In: Evans WE, Schentag J J, Jusko W J, eds. Applied pharmacokinetics: principles of therapeutic drug monitoring. 3rd ed. Vancouver, Washington: Applied Therapeu- tics, 1992:[Chapter 2711-34.

27. Kinirons MT, O'Shea D, Downing TE, et al. Absence of correlations among three putative in vivo probes of human cytochrome P4503A activity in young healthy men. CuN PHARMACOL THER 1993;54:621-9.

28. Ortiz DE, Montellano PR, Kunze KL, Yost GS, Mico BA. Self- catalyzed destruction of cytochrome P-450: covalent binding of ethynyl sterols to prosthetic heme. Proc Natl Acad Sci USA 1979;76:746-9.

29. Guengerich FP. Minireview: metabolism of t7c~- ethynylestradiol in humans. Life Sci 1990;47:1981-8.

30. Chambers DM, Jefferson GC, Chambers M, Loudon NB. Antipyrine elimination in saliva after low-dose combined or progestogen-only oral contraceptive steroids. Br J Clin Pharmacot 1982;13:229-32.

31. Halpert JR, Guengerich FP, Bend JR, Correia MA. Selective inhibitors of cytochromes P450. Toxicol Appl Pharmacol 1994;125:163-75.

32. Hunt CM, Westerkam WR, Stave GM. Effect of age and gender on the activity of human hepatic CYP3A. Biochem Pharmacol 1992;44:275-83.

33. Burke CW. Biologically active cortisol in plasma of oestrogen-treated and normal subjects. Br Med J 1969;2:798-800.

34. Plager JE, Schmidt KG, Staubitz WJ. Increased unbound cortisol in the plasma of estrogen-treated subjects. J Clin Invest 1964;43:1066-72.

35. Spangler AS, Antoniades HN, Sotman SL, Inderbiti- zin TM. Enhancement of the anti-inflammatory ac- tion of hydrocortisone by estrogen. J Clin Endocrinol Metab 1969;29:650-5.

36. Nelson DH, Tanney H, Mestman G, Gieschen VW, Wilson LD. Potentiation of the biological effect of administered cortisol by estrogen treatment. J Clin Endocrinol Metab 19+63;23:261-5.

37. Doe RP, Dickinson P, Zinneman, HH. Elevated non-protein bound cortisol (NPC) in pregnancy, during estrogen administration and in carcinoma of the prostate. J Clin Endocrinol Metab 1969;29: 757-66.

38. Turner BB. Sex difference in glucocorticoid binding in rat pituitary is estrogen dependent. Life Sci 1990;46: 1399-406.

39. Frey BM, Frey FJ. The effect of altered prednisolone kinetics in patients with the nephrotic syndrome and in women taking oral contraceptive steroids on human mixed lymphocyte cultures. J Clin Endocrinol Metab 1985;60:361-9.

oral contraceptive effects on methylprednisolone pharmacokinetics and pharmacodynamics*

Documents