- Email: [email protected]
Pharmacokinetics and metabolism of medroxyprogesterone acetate in patients with advanced breast cancer
Vol.31,No. 4A, PP.437-441,1988 Printedin Great Britain.All rightsreserved
J. steroid Biochem.
0022-4731/88 $3.00+ 0.00 Copyright0 1988PergamooPressplc
PHARMACOKINETICS AND METABOLISM OF MEDROXYPROGESTERONE ACETATE IN PATIENTS WITH ADVANCED BREAST CANCER E. UTAAEEE*$B,S. LLJNDGRENt$,S. KVINNSLAND~and A. AAKVAAG* *Departments of Biochemical Endocrinology and toncology, University of Bergen, Bergen, Norway (Received 11 August 1986; receivedfor publication22 April 1988)
Summary--The pha~a~okinetics and metabolism of medroxyprogesterone acetate (MPA) were studied in patients with advanced breast cancer after i.v. injection and oral administration of [‘HIMPA. MPA was distributed very rapidly into three compartments after i.v. injections, revealing half-lives of 47 h. Using a nonlinear model fitting metabolic clearance rates (MCR) were found to be 652 I/day before and 601 l/day during MPA treatment, and dist~bution volumes (V,) 5.9 and 3.4 1 respectively. The major metaboiite of MPA following iv. injection was a glucuronide of MPA, presumably of the 3-enol form. After oral administration the radioactivity in serum increased rapidly and reached a plateau of about I% of the dose per litre serum after approx 2 h. About W-90% of the radioactivity was found in the water phase after hexane extraction, persumably as glucuronides of metabolites more polar than MPA. Radioimmunoassay (RIA) of MPA in untreated serum samples showed 3-8-fold greater MPA values as compared to measurements in hexane extracts of serum. Ethanol extraction did not remove these interfe~ng substances. Extraction of serum with a low polar solvent before RIA of MPA is essential in order to prevent great overestimation, as the glucuronidated polar metabolites most likely will crossreact in an assay with an antiserum raised against a MPA-3-0-carboximethyloxime coupled to bovine serum albumin.
EXPERIMENTAL
INTRODUCTION Medroxyprogesterone acetate (MPA; 6a-methyl-17a hydroxyprogesterone acetate), is a synthetic progestin used in the treatment of advanced breast cancer. Differences in the pharmacokinetics and metabolism of MPA in individual patients might be reasons for the interpatient variations in serum levels after administration of the same dosages fl, 25 The method used for MPA determinations may be crucial. In radioimmunoassays (RIA) the type of antigen used to raise the antiserum and the extraction procedure will influence the validity of the result obtained ]3,41. The present study was initiated to investigate the pharmacokinetics and metabolism of MPA in patients with advanced breast cancer. We also wanted to study if any potential metabolites of MPA might crossreact with the antiserum raised against MPA-3CMO-BSA, used in our RIA after hexane extraction of serum.
fiAddress for correspondence: Bdle Utaaker, Laboratory of Bi~h~i~l Endocrinoloav, Haukeland Sykehus, N5021 Bergen, Norway. -_’ SResearch fellow of the Norwegian Society for Fighting Cancer. PResearch fellow of the Norwegian Cancer society.
Materials [I ,2-” H(N)]MPA, 60.0 Ci/mmol was purchased from New England Nuclear (Boston, MA), crystalline nonradioactive MPA, from Sigma Chem. Co (St Louis, MO), and tablets of 100 mg MPA, Proverag, from Upjohn Co. (Kalamazoo, MI). Antiserum against MPA3-CMG-BSA, No. 15698FAK-53, was a gift from the Upjohn Co. (Kalamazoo, MI). Precoated plates of Silica gel 60 F254 from Merck were used in thin layer chromatography (TLC). /3-Glucuronidase 2000 U/mg (Patella vulgaris) was obtained from David Love Co. (Musselburgh, Scotland), and all solvents from Merck were of analytical grade. Opti-Fluor from Packard instruments was used as liquid scintillation cocktail. Patients and patient protocols
Three women with advanced breast cancer were given a single iv. injection of 13H]MPA (100 &i), in 10 ml 0.9% saline. Serum samples were drawn after 3, 5, 7, 10, 15, 30, 45, 60, 90, 120, 180 and 360min. After 5 days 13H]MPA (lOO@), was administered orally on 5 tablets of 100mg MPA, to the same patients. Serum samples were obtained 2 min before, and 0.5, 1, 2, 3, 4, 6, 8, 12, 24, 36 and 48 h after the 431
438
E. UTAAKER et al.
administration. In two of the patients the i.v. injection &as repeated after 1 month of oral MPA treatment (lOOOmg/day). Serum samples were drawn as described for the first i.v. injection, One woman with advanced breast cancer was given [3H]MPA (500 PCi), on five tablets of 100 mg MPA. Urine was collected for 3 days after each administration of [3HIMPA. Steroid extraction and fractionation Total radioactivity was measured in aliquots of serum. 6ml serum was then extracted 3 times with 2 vol of diethylether after addition of authentic MPA (5Ogg). The extracts were dried with Na,SO, and evaporated. The residues were dissolved in 0.5 ml dichlormethane and 1 ml methanol. Aliquots were obtained for scintillation counting. The remaining extracts were evaporated and defatted by partitioning between 70% aqueous metanol and hexane. Following evaporation the extracts were subjected to TLC, using ethylacetate-cyclohexane (4: 1, v/v) as mobile phase. The chromatographic lanes were cut in pieces of 1 cm, the silica gel scraped off, and the radioactivity eluted with methanol and aliquots were obtained for scintillation counting. The serum proteins in the residual water phase were precipitated with ethanol (80%), and removed by centrifugation. The ethanol extracts were evaporated, dissolved in 10 ml 0.15 M acetate buffer with a pH of 4.5, and incubated with 10,000 U fi-glucuronidase at 37°C for 20 h. After addition of authentic MPA (5Opg), the hydrolysed steroids were extracted twice wtih 1 vol of diethylether. The combined ether extracts were washed once with 0.1 N NaOH and twice with distilled water, dried with NazSO1, evaporated and dissolved in methanol. Aliquots were obtained for scintillation counting and the extracts fractionated by TLC as described previously. The radioactivity corresponding to MPA on TLC were acetylated over night at room temperature in 0.5 ml pyndine-acetic anhydride (1: 1, v/v>, and subjected to TLC as described above. The radioactivity corresponding to MPA after acetylation, was combined with 25 mg authentic MPA and crystallised to constant specific activity. The crystals were obtained from methanol-water, acetone-water, ethylacetate-hexane, and ethanolwater. Radioimmunoassay RIA of MPA was carried out as described by Ortiz et al. [5], with some modifications. A phosphate-BSA buffer (PBSA), at pH 7.5 (0.05 M potassium phosphate, 0.1 M sodium chloride, 0.2% BSA and 0.05% sodium azide) was used. Approximately 1000 cpm of [‘HIMPA was added in 300 ~1 buffer to 300 ~1 serum and extracted once with 5 ml hexane. The extracts were evaporated and the residues dissolved in 1 ml PBSA. One aliquot
(400 ~1)~ was obtained for scintillation counting to correct for extraction loss, and aliquots of 50-200 ~1 were used in the RIA. RIA measurements were also done on serum directly and on 80% ethanol extracts of serum. In both cases a 50 ~1 aliquot of a 1:30 dilution of serum in buffer was used for analysis. The aliquots were adjusted to 20011, and [3H]MPA (5000 cpm) in 100 ~1 buffer, and 100 ~1 antiserum with a final dilution 1: 16,000 in buffer was added to the incubation tubes. Samples and standards (62.5 to 4000 pg of MPA) were run in duplicates and incubated overnight at 4°C. Ice-cold dextran coated charcoal suspension (750 ~1) [0.25% charcoal Norit A and 0.025% dextran] was added to separate free and antibody-bound MPA. The antibody-bound [‘HIMPA in the supernatant was obtained for liquid scintillation counting. A nonlinear fit was applied to construct standard curves and calculate MPA concentrations in unknown samples. Scintillation counting The radioactivity was counted in a Mark III /Icounter from Searle Analytic Inc., U.S.A. All samples were corrected for quenching based on the “Sample Channels Ratio” (SCR). RESULTS AND DISCUSSION
Metabohim
of [3H]MPA
i.v. injection. The concentration of radioactivity in serum, in the ether extract of serum, and in the residual water phase from the patients before and during MPA treatment, showed only minor differences between individual patients. In the initial phase the total radioactivity declined rapidly, thereafter it increased slightly before it again declined slowly (Fig. 1). Using a computer program which applies the Marquardt algorithm [6] and successive approximations to find the least square best fit multiexponential function describing a curve, we found that the disappearance of ether extractable radioactivity (mean fraction of dose per litre serum) was most accurately described by a function which was the sum of three exponentials both before and during MPA treatment (Fig. 2). The coefficients found from the functions describing the disappearance curves were applied in the formulas given by Gupta et al. [S], and the metabolic clearance rates (MCR) calculated to be 652 l/day before and 601 l/day during MPA treatment. The distribution volumes (V,) computed to be 5.9 and 3.4 1 respectively, describe the distribution of MPA during the first part of the curve and correspond to the blood volume. All the radioactivity in the ether extract was found in the MPA fraction by TLC. Five days after the i.v. injection the concentration of ether extractable radioactivity was still 0.07% of the original dose per litre serum, which corresponds to a halflife of 2-3 days under the assumption
Phannacokinetics and metabolism of MPA
During
I
-I
.
0
I
.
100
I
I
I
200
300
400
Minutes Fig. 1. Radioactivity expressed as percent of total dose per litre serum as a function of time after iv. injection of 1OOpCi [‘HIMPA before and during treatment. Radioactivity was measured in untreated serum, ‘in an ether extract of serum, and in the residual water phase. The samples from 3 to 45 min represent the first, 45 to 120 the second, and 120 to 360 the third phase of the disappearance curve.
of a first order removal during this period. This significant amount of radioactivity in serum 5 days after administration indicates that MPA is accumulated in an outer compartment from which there is a slow re-entry into the blood. On the basis of the lipophilic nature of MPA, it is reasonable to assume that adipose tissue constitutes at least a part of the third compartment, as is the case for progesterone, a substance with similar polarity. loo
\
E 2 !
10-1
439
The radioactivity in the residual water phase increased rapidly, and within 15 min the concentration of radioactivity in the water phase was higher than in the ether extract. The ether extract of the residual water phase after hydrolysis with @-glucuronidase, also showed all the radioactivity in the MPA fraction by TLC. The fraction corresponding to MPA could not be acetylated and was crystallised to constant specific activity with authentic MPA (data not shown). This strongly suggests that a major metabolite of MPA following i.v. injection is a conjugated metabolite of the 3-enol form of MPA, most likely a glucuronide. This is in accordance with preliminary results from Pannuti et al. [7], indicating that conjugates, probably derived from the 3-enol form of MPA, are present in blood in appreciable concentrations. Only 25-50% of the total radioactive dose was excreted in urine during the first 3 days after the injection. This observation should be. seen in light of the finding of a significant concentration of radioactive MPA in serum 5 days after the i.v. administration, and it supports the conclusion that MPA is accumulated in the body. Oral administration After oral administration the total radioactivity in serum increased rapidly and reached a level of about 1.0% of the dose per litre serum after approx 2 h (Fig. 3). The total level of radioactivity did not decline during the first 48 h after administration. Less than 10% of the total radioactivity in serum after oral administration could be extracted by diethylether. As a result of this, the amount of radioactivity in the ether extract after administration of 1OOpCi [3H]MPA was too low to identify MPA or presumably free metabolites by TLC. To cope with this problem one patient was given 500 PCi [‘HIMPA and larger serum samples (20 ml) were extracted. A slightly more polar metabolite in addition to MPA was found in the extract by TLC. One can only speculate that the metabolite is an A-ring reduced form of MPA. Such a metabolite will probably crossreact in an RIA using an antiserum raised
I 8
0
100
200
300
400
Minutes Fig. 2. The disappearance curve for [‘HIMPA after i.v. injection before (0) and during (H) MPA treatment, as described by a function which is the sum of three exponentials. The solid lines are the calculated curves and the points represent the observed values.
IO”
4 0
-
IO
20
30
40
50
Hours
Fig. 3. Radioactivity expressed as percent of total dose per litre serum as a function of time after oral administration of I00j~Ci [)H]MPA. Radioactivity was measured in untreated serum and in the residual water phase after ether extraction of serum.
E.
440
UTMKEiR et al.
against MPA3-CMSBSA. This substance might be the same as the immunoreactive material different from MPA, previously shown after high performance liquid chromatography of a petroleum ether extract of serum [8]. After treatment of the residual water phase with /?-glucuronidase, subsequent TLC demonstrated that the main part of the radioactivity appeared as metabolites more polar than MPA. Sturm and Schulz [8] have by means of isotope dilution gas chromatographic-mass fragmentographic technique, demonstrated several glucuronidated A-ring reduced metabolites of MPA after long-term MPA treatment. It has also been reported that the main fraction of MPA after intramuscular administration circulates as glucuronidated metabolites in blood [9]. The compound proposed to be the glucuronide of the enol form of MPA could not be detected. The early appearance of substantial amounts of conjugated metabolites more polar than MPA and the missing evidence for the formation of the 3-enol glucuronide of MPA after oral administration, highlights the significance of metabolism of MPA during first passage through the liver and suggests that the enol glucuronide may not be formed in the splanchnic area. Only 1%25% of the total radioactive dose was excreted in the urine during the first 3 days after oral administration. Radioimmunoassay
of MPA
MPA was measured on untreated serum (direct assay), on alcohol extracts, and on hexane extracts of serum from thirteen patients after 14 days of oral MPA treatment (1000 mg/day). The mean values found by direct assay and ethanol extraction were not different; 471.1 ng/ml and 490.2 ng/ml respectively, whereas hexane extraction yielded a mean value of 102.6 ng/ml (Fig. 4). Linear regression analysis of results found after hexane extraction compared to values obtained by direct assay showed a correlation coefficient of 0.68, while the correlation coefficient for values found after alcohol extraction as compared to direct assay was 0.97. RIA on the residual water phase after hexane extraction revealed the presence of crossreacting material, and $ of the overestimation could be accounted for by this crossreactivity. These results give additional support to our conclusion that the main fraction of orally administered MPA in serum are conjugated metabolites, which will crossreact in an RIA using an antiserum raised against
MPA-3-CMO-BSA. Because of this crossreactivity one may assume that the major fraction of metabolites are conjugated in the three-position. In conclusion, this study demonstrates that MPA is distributed very rapidly into three compartments after iv. injection. The major metabolite of MPA following i.v. injection is a glucuronide of MPA, presumably of the 3-enol form. After oral administration the main proportion of metabolites are glucuronidated substances more polar than MPA. The
Y
0
260
400
000
edo
I obo
ng/ml Dlmct
assay
Fig. 4. Linear regression analyses of MPA in thirteen plasma samples quantitated by different methods.
difference in metabolites seen after iv. and oral administration points to the importance of the first liver passage. Extraction of serum with a low polar solvent before RIA is essential in order to prevent great overestimation of MPA, as the glucuronidated polar metabolites most likely will crossreact in an assay with an antiserum raised against MPA-3CM@-BSA. Acknowledgements-The
authors wish to thank A. Eliassen for excellent technical assistance. The study was supported by the Norwegian Society for Fighting Cancer and the Norwegian Cancer Society.
REFERENCES
Salimtschik M., Mouridsen H. T., Lober J. and Johansson E.: Comparative pharmacokinetics of medroxyprogesterone acetate administered by oral and intramuscular routes. Cancer Chemother. Pharmac. 4 (1981) 267-269.
Camaggi C. M., Stroeehi E., Giovannini M., Angebelli B., Constanti B., Zebini E., Ferrari P. and Pannuti F.: Medroxyprogesterone acetate (MPA) plasma levels after multiple high-dose administration in advanced cancer patients. Cancer Chemother. Pharmuc. 11 (1983) 1922. Shrimanker K., Saxena B. N. and Fotherby K.: A radioimmunoassay for serum medroxyprogesterone acetate. J. steroid Eiochem. 9 (1978) 359-363. Royer M. E., Howard K., Campbell J. A., Murray H. C., Evans J. S. and Kaiser D. G.: Radioimmunoassay for medroxyprogesterone acetate (Provera@) using the 1la-hydroxy suceinyl conjugate. Steroids 23 (1974) 713-730.
Pharmacokinetics
and metabolism of MPA
5. Ortiz A., Hiroi M., Stanczyk U., Goebelsmann U. and Mishell D. R.: Serum medroxyprogesterone acetate (MPA) concentrations and ovarian function following intramuscular injection of depo-MPA. J. clin. Endow. Metab. 44 (1977) 32-38. 6. Gupta C., Osterman J., Santen R. and Bardin C. W.: In uiuo metabolism of progestins V. The effect of protocol design on the estimate metabolic clearance rate and volume of distribution of medroxyprogesterone acetate in women. J. c/in. Endocr. Metab. 48 (1979) 816820. 7. Pannuti F., Camaggi C. M., Strocchi E., Martoni A., Beghelli P., Biondi S., Constanti B. and Grieco A.: Medroxyprogesterone acetate pharmacokinetics. In
441
Role of Medroxyprogesterone in Endocrine-Related Tumors (Edited by A. Pellegrini, G. Robustelli Della
Cuna, F. Pannuti, P. Pouillart and W. Jonat). Raven Press, New York, Vol. 3 (1984) p. 43, p. 46. Sturm G. and Schulz K.-D.: MPA assays: measurement of plasma MPA levels in high-dose MPA-treated patients. In Role of Medroxyprogesterone in EndocrineRelated Tumors (Edited by A. Pellegrini, G. Robustelli Della Cuna. F. Pannuti. P. Pouillart and W. Jonat). Raven Press, New York; Vol. 3 (1984) p. 23, p. 37. ’ Mathrubutham M. and Fotherby K.: Medroxyprogesterone acetate in human serum. J. steroid Eiochem. 14 (1981) 783-786.
J. steroid Biochem.
0022-4731/88 $3.00+ 0.00 Copyright0 1988PergamooPressplc
PHARMACOKINETICS AND METABOLISM OF MEDROXYPROGESTERONE ACETATE IN PATIENTS WITH ADVANCED BREAST CANCER E. UTAAEEE*$B,S. LLJNDGRENt$,S. KVINNSLAND~and A. AAKVAAG* *Departments of Biochemical Endocrinology and toncology, University of Bergen, Bergen, Norway (Received 11 August 1986; receivedfor publication22 April 1988)
Summary--The pha~a~okinetics and metabolism of medroxyprogesterone acetate (MPA) were studied in patients with advanced breast cancer after i.v. injection and oral administration of [‘HIMPA. MPA was distributed very rapidly into three compartments after i.v. injections, revealing half-lives of 47 h. Using a nonlinear model fitting metabolic clearance rates (MCR) were found to be 652 I/day before and 601 l/day during MPA treatment, and dist~bution volumes (V,) 5.9 and 3.4 1 respectively. The major metaboiite of MPA following iv. injection was a glucuronide of MPA, presumably of the 3-enol form. After oral administration the radioactivity in serum increased rapidly and reached a plateau of about I% of the dose per litre serum after approx 2 h. About W-90% of the radioactivity was found in the water phase after hexane extraction, persumably as glucuronides of metabolites more polar than MPA. Radioimmunoassay (RIA) of MPA in untreated serum samples showed 3-8-fold greater MPA values as compared to measurements in hexane extracts of serum. Ethanol extraction did not remove these interfe~ng substances. Extraction of serum with a low polar solvent before RIA of MPA is essential in order to prevent great overestimation, as the glucuronidated polar metabolites most likely will crossreact in an assay with an antiserum raised against a MPA-3-0-carboximethyloxime coupled to bovine serum albumin.
EXPERIMENTAL
INTRODUCTION Medroxyprogesterone acetate (MPA; 6a-methyl-17a hydroxyprogesterone acetate), is a synthetic progestin used in the treatment of advanced breast cancer. Differences in the pharmacokinetics and metabolism of MPA in individual patients might be reasons for the interpatient variations in serum levels after administration of the same dosages fl, 25 The method used for MPA determinations may be crucial. In radioimmunoassays (RIA) the type of antigen used to raise the antiserum and the extraction procedure will influence the validity of the result obtained ]3,41. The present study was initiated to investigate the pharmacokinetics and metabolism of MPA in patients with advanced breast cancer. We also wanted to study if any potential metabolites of MPA might crossreact with the antiserum raised against MPA-3CMO-BSA, used in our RIA after hexane extraction of serum.
fiAddress for correspondence: Bdle Utaaker, Laboratory of Bi~h~i~l Endocrinoloav, Haukeland Sykehus, N5021 Bergen, Norway. -_’ SResearch fellow of the Norwegian Society for Fighting Cancer. PResearch fellow of the Norwegian Cancer society.
Materials [I ,2-” H(N)]MPA, 60.0 Ci/mmol was purchased from New England Nuclear (Boston, MA), crystalline nonradioactive MPA, from Sigma Chem. Co (St Louis, MO), and tablets of 100 mg MPA, Proverag, from Upjohn Co. (Kalamazoo, MI). Antiserum against MPA3-CMG-BSA, No. 15698FAK-53, was a gift from the Upjohn Co. (Kalamazoo, MI). Precoated plates of Silica gel 60 F254 from Merck were used in thin layer chromatography (TLC). /3-Glucuronidase 2000 U/mg (Patella vulgaris) was obtained from David Love Co. (Musselburgh, Scotland), and all solvents from Merck were of analytical grade. Opti-Fluor from Packard instruments was used as liquid scintillation cocktail. Patients and patient protocols
Three women with advanced breast cancer were given a single iv. injection of 13H]MPA (100 &i), in 10 ml 0.9% saline. Serum samples were drawn after 3, 5, 7, 10, 15, 30, 45, 60, 90, 120, 180 and 360min. After 5 days 13H]MPA (lOO@), was administered orally on 5 tablets of 100mg MPA, to the same patients. Serum samples were obtained 2 min before, and 0.5, 1, 2, 3, 4, 6, 8, 12, 24, 36 and 48 h after the 431
438
E. UTAAKER et al.
administration. In two of the patients the i.v. injection &as repeated after 1 month of oral MPA treatment (lOOOmg/day). Serum samples were drawn as described for the first i.v. injection, One woman with advanced breast cancer was given [3H]MPA (500 PCi), on five tablets of 100 mg MPA. Urine was collected for 3 days after each administration of [3HIMPA. Steroid extraction and fractionation Total radioactivity was measured in aliquots of serum. 6ml serum was then extracted 3 times with 2 vol of diethylether after addition of authentic MPA (5Ogg). The extracts were dried with Na,SO, and evaporated. The residues were dissolved in 0.5 ml dichlormethane and 1 ml methanol. Aliquots were obtained for scintillation counting. The remaining extracts were evaporated and defatted by partitioning between 70% aqueous metanol and hexane. Following evaporation the extracts were subjected to TLC, using ethylacetate-cyclohexane (4: 1, v/v) as mobile phase. The chromatographic lanes were cut in pieces of 1 cm, the silica gel scraped off, and the radioactivity eluted with methanol and aliquots were obtained for scintillation counting. The serum proteins in the residual water phase were precipitated with ethanol (80%), and removed by centrifugation. The ethanol extracts were evaporated, dissolved in 10 ml 0.15 M acetate buffer with a pH of 4.5, and incubated with 10,000 U fi-glucuronidase at 37°C for 20 h. After addition of authentic MPA (5Opg), the hydrolysed steroids were extracted twice wtih 1 vol of diethylether. The combined ether extracts were washed once with 0.1 N NaOH and twice with distilled water, dried with NazSO1, evaporated and dissolved in methanol. Aliquots were obtained for scintillation counting and the extracts fractionated by TLC as described previously. The radioactivity corresponding to MPA on TLC were acetylated over night at room temperature in 0.5 ml pyndine-acetic anhydride (1: 1, v/v>, and subjected to TLC as described above. The radioactivity corresponding to MPA after acetylation, was combined with 25 mg authentic MPA and crystallised to constant specific activity. The crystals were obtained from methanol-water, acetone-water, ethylacetate-hexane, and ethanolwater. Radioimmunoassay RIA of MPA was carried out as described by Ortiz et al. [5], with some modifications. A phosphate-BSA buffer (PBSA), at pH 7.5 (0.05 M potassium phosphate, 0.1 M sodium chloride, 0.2% BSA and 0.05% sodium azide) was used. Approximately 1000 cpm of [‘HIMPA was added in 300 ~1 buffer to 300 ~1 serum and extracted once with 5 ml hexane. The extracts were evaporated and the residues dissolved in 1 ml PBSA. One aliquot
(400 ~1)~ was obtained for scintillation counting to correct for extraction loss, and aliquots of 50-200 ~1 were used in the RIA. RIA measurements were also done on serum directly and on 80% ethanol extracts of serum. In both cases a 50 ~1 aliquot of a 1:30 dilution of serum in buffer was used for analysis. The aliquots were adjusted to 20011, and [3H]MPA (5000 cpm) in 100 ~1 buffer, and 100 ~1 antiserum with a final dilution 1: 16,000 in buffer was added to the incubation tubes. Samples and standards (62.5 to 4000 pg of MPA) were run in duplicates and incubated overnight at 4°C. Ice-cold dextran coated charcoal suspension (750 ~1) [0.25% charcoal Norit A and 0.025% dextran] was added to separate free and antibody-bound MPA. The antibody-bound [‘HIMPA in the supernatant was obtained for liquid scintillation counting. A nonlinear fit was applied to construct standard curves and calculate MPA concentrations in unknown samples. Scintillation counting The radioactivity was counted in a Mark III /Icounter from Searle Analytic Inc., U.S.A. All samples were corrected for quenching based on the “Sample Channels Ratio” (SCR). RESULTS AND DISCUSSION
Metabohim
of [3H]MPA
i.v. injection. The concentration of radioactivity in serum, in the ether extract of serum, and in the residual water phase from the patients before and during MPA treatment, showed only minor differences between individual patients. In the initial phase the total radioactivity declined rapidly, thereafter it increased slightly before it again declined slowly (Fig. 1). Using a computer program which applies the Marquardt algorithm [6] and successive approximations to find the least square best fit multiexponential function describing a curve, we found that the disappearance of ether extractable radioactivity (mean fraction of dose per litre serum) was most accurately described by a function which was the sum of three exponentials both before and during MPA treatment (Fig. 2). The coefficients found from the functions describing the disappearance curves were applied in the formulas given by Gupta et al. [S], and the metabolic clearance rates (MCR) calculated to be 652 l/day before and 601 l/day during MPA treatment. The distribution volumes (V,) computed to be 5.9 and 3.4 1 respectively, describe the distribution of MPA during the first part of the curve and correspond to the blood volume. All the radioactivity in the ether extract was found in the MPA fraction by TLC. Five days after the i.v. injection the concentration of ether extractable radioactivity was still 0.07% of the original dose per litre serum, which corresponds to a halflife of 2-3 days under the assumption
Phannacokinetics and metabolism of MPA
During
I
-I
.
0
I
.
100
I
I
I
200
300
400
Minutes Fig. 1. Radioactivity expressed as percent of total dose per litre serum as a function of time after iv. injection of 1OOpCi [‘HIMPA before and during treatment. Radioactivity was measured in untreated serum, ‘in an ether extract of serum, and in the residual water phase. The samples from 3 to 45 min represent the first, 45 to 120 the second, and 120 to 360 the third phase of the disappearance curve.
of a first order removal during this period. This significant amount of radioactivity in serum 5 days after administration indicates that MPA is accumulated in an outer compartment from which there is a slow re-entry into the blood. On the basis of the lipophilic nature of MPA, it is reasonable to assume that adipose tissue constitutes at least a part of the third compartment, as is the case for progesterone, a substance with similar polarity. loo
\
E 2 !
10-1
439
The radioactivity in the residual water phase increased rapidly, and within 15 min the concentration of radioactivity in the water phase was higher than in the ether extract. The ether extract of the residual water phase after hydrolysis with @-glucuronidase, also showed all the radioactivity in the MPA fraction by TLC. The fraction corresponding to MPA could not be acetylated and was crystallised to constant specific activity with authentic MPA (data not shown). This strongly suggests that a major metabolite of MPA following i.v. injection is a conjugated metabolite of the 3-enol form of MPA, most likely a glucuronide. This is in accordance with preliminary results from Pannuti et al. [7], indicating that conjugates, probably derived from the 3-enol form of MPA, are present in blood in appreciable concentrations. Only 25-50% of the total radioactive dose was excreted in urine during the first 3 days after the injection. This observation should be. seen in light of the finding of a significant concentration of radioactive MPA in serum 5 days after the i.v. administration, and it supports the conclusion that MPA is accumulated in the body. Oral administration After oral administration the total radioactivity in serum increased rapidly and reached a level of about 1.0% of the dose per litre serum after approx 2 h (Fig. 3). The total level of radioactivity did not decline during the first 48 h after administration. Less than 10% of the total radioactivity in serum after oral administration could be extracted by diethylether. As a result of this, the amount of radioactivity in the ether extract after administration of 1OOpCi [3H]MPA was too low to identify MPA or presumably free metabolites by TLC. To cope with this problem one patient was given 500 PCi [‘HIMPA and larger serum samples (20 ml) were extracted. A slightly more polar metabolite in addition to MPA was found in the extract by TLC. One can only speculate that the metabolite is an A-ring reduced form of MPA. Such a metabolite will probably crossreact in an RIA using an antiserum raised
I 8
0
100
200
300
400
Minutes Fig. 2. The disappearance curve for [‘HIMPA after i.v. injection before (0) and during (H) MPA treatment, as described by a function which is the sum of three exponentials. The solid lines are the calculated curves and the points represent the observed values.
IO”
4 0
-
IO
20
30
40
50
Hours
Fig. 3. Radioactivity expressed as percent of total dose per litre serum as a function of time after oral administration of I00j~Ci [)H]MPA. Radioactivity was measured in untreated serum and in the residual water phase after ether extraction of serum.
E.
440
UTMKEiR et al.
against MPA3-CMSBSA. This substance might be the same as the immunoreactive material different from MPA, previously shown after high performance liquid chromatography of a petroleum ether extract of serum [8]. After treatment of the residual water phase with /?-glucuronidase, subsequent TLC demonstrated that the main part of the radioactivity appeared as metabolites more polar than MPA. Sturm and Schulz [8] have by means of isotope dilution gas chromatographic-mass fragmentographic technique, demonstrated several glucuronidated A-ring reduced metabolites of MPA after long-term MPA treatment. It has also been reported that the main fraction of MPA after intramuscular administration circulates as glucuronidated metabolites in blood [9]. The compound proposed to be the glucuronide of the enol form of MPA could not be detected. The early appearance of substantial amounts of conjugated metabolites more polar than MPA and the missing evidence for the formation of the 3-enol glucuronide of MPA after oral administration, highlights the significance of metabolism of MPA during first passage through the liver and suggests that the enol glucuronide may not be formed in the splanchnic area. Only 1%25% of the total radioactive dose was excreted in the urine during the first 3 days after oral administration. Radioimmunoassay
of MPA
MPA was measured on untreated serum (direct assay), on alcohol extracts, and on hexane extracts of serum from thirteen patients after 14 days of oral MPA treatment (1000 mg/day). The mean values found by direct assay and ethanol extraction were not different; 471.1 ng/ml and 490.2 ng/ml respectively, whereas hexane extraction yielded a mean value of 102.6 ng/ml (Fig. 4). Linear regression analysis of results found after hexane extraction compared to values obtained by direct assay showed a correlation coefficient of 0.68, while the correlation coefficient for values found after alcohol extraction as compared to direct assay was 0.97. RIA on the residual water phase after hexane extraction revealed the presence of crossreacting material, and $ of the overestimation could be accounted for by this crossreactivity. These results give additional support to our conclusion that the main fraction of orally administered MPA in serum are conjugated metabolites, which will crossreact in an RIA using an antiserum raised against
MPA-3-CMO-BSA. Because of this crossreactivity one may assume that the major fraction of metabolites are conjugated in the three-position. In conclusion, this study demonstrates that MPA is distributed very rapidly into three compartments after iv. injection. The major metabolite of MPA following i.v. injection is a glucuronide of MPA, presumably of the 3-enol form. After oral administration the main proportion of metabolites are glucuronidated substances more polar than MPA. The
Y
0
260
400
000
edo
I obo
ng/ml Dlmct
assay
Fig. 4. Linear regression analyses of MPA in thirteen plasma samples quantitated by different methods.
difference in metabolites seen after iv. and oral administration points to the importance of the first liver passage. Extraction of serum with a low polar solvent before RIA is essential in order to prevent great overestimation of MPA, as the glucuronidated polar metabolites most likely will crossreact in an assay with an antiserum raised against MPA-3CM@-BSA. Acknowledgements-The
authors wish to thank A. Eliassen for excellent technical assistance. The study was supported by the Norwegian Society for Fighting Cancer and the Norwegian Cancer Society.
REFERENCES
Salimtschik M., Mouridsen H. T., Lober J. and Johansson E.: Comparative pharmacokinetics of medroxyprogesterone acetate administered by oral and intramuscular routes. Cancer Chemother. Pharmac. 4 (1981) 267-269.
Camaggi C. M., Stroeehi E., Giovannini M., Angebelli B., Constanti B., Zebini E., Ferrari P. and Pannuti F.: Medroxyprogesterone acetate (MPA) plasma levels after multiple high-dose administration in advanced cancer patients. Cancer Chemother. Pharmuc. 11 (1983) 1922. Shrimanker K., Saxena B. N. and Fotherby K.: A radioimmunoassay for serum medroxyprogesterone acetate. J. steroid Eiochem. 9 (1978) 359-363. Royer M. E., Howard K., Campbell J. A., Murray H. C., Evans J. S. and Kaiser D. G.: Radioimmunoassay for medroxyprogesterone acetate (Provera@) using the 1la-hydroxy suceinyl conjugate. Steroids 23 (1974) 713-730.
Pharmacokinetics
and metabolism of MPA
5. Ortiz A., Hiroi M., Stanczyk U., Goebelsmann U. and Mishell D. R.: Serum medroxyprogesterone acetate (MPA) concentrations and ovarian function following intramuscular injection of depo-MPA. J. clin. Endow. Metab. 44 (1977) 32-38. 6. Gupta C., Osterman J., Santen R. and Bardin C. W.: In uiuo metabolism of progestins V. The effect of protocol design on the estimate metabolic clearance rate and volume of distribution of medroxyprogesterone acetate in women. J. c/in. Endocr. Metab. 48 (1979) 816820. 7. Pannuti F., Camaggi C. M., Strocchi E., Martoni A., Beghelli P., Biondi S., Constanti B. and Grieco A.: Medroxyprogesterone acetate pharmacokinetics. In
441
Role of Medroxyprogesterone in Endocrine-Related Tumors (Edited by A. Pellegrini, G. Robustelli Della
Cuna, F. Pannuti, P. Pouillart and W. Jonat). Raven Press, New York, Vol. 3 (1984) p. 43, p. 46. Sturm G. and Schulz K.-D.: MPA assays: measurement of plasma MPA levels in high-dose MPA-treated patients. In Role of Medroxyprogesterone in EndocrineRelated Tumors (Edited by A. Pellegrini, G. Robustelli Della Cuna. F. Pannuti. P. Pouillart and W. Jonat). Raven Press, New York; Vol. 3 (1984) p. 23, p. 37. ’ Mathrubutham M. and Fotherby K.: Medroxyprogesterone acetate in human serum. J. steroid Eiochem. 14 (1981) 783-786.