0090-9556/ 83/ 1105-0446$00.00/ 0
`DRUG METABOLISM AND DISPOSITIO N
`Copyright© 1983 by The American Society for Pharmacology and Experimental Therapeutics
`
`Vol.ll , No.5
`Printed in U.S.A.
`
`OXYMORPHONE METABOLISM AND URINARY EXCRETION IN HUMAN, RAT, GUINEA
`PIG, RABBIT, AND DOG
`
`EDWARD J. CONE, WILLIAM D. DARWIN, WILLIAM F. BUCHWALD, AND CHARLES W. GORODETZKY
`
`National Institute on Drug Abuse, Addiction Research Center, United States Department of Health and Human Services, Public Health Service,
`Alcohol, Drug Abuse, and Mental Health Administration
`
`(Received April 1, 1983; accepted May 23, 1983)
`
`ABSTRACT:
`
`Oxymorphone was extensively metabolized by human, rat, dog, and
`guinea pig and to a lesser extent by rabbit. The most abundant
`metabolite in urine for all species was conjugated oxymorphone
`(12. 7-81.7% administered dose) followed by 6/1- and 6a-carbinols
`produced by 6-keto reduction of oxymorphone. 6{1-0xymorphol
`(0.2-3.1 %) was found in the urine of all species, whereas 6a-oxy(cid:173)
`morphol (0.1-2.8%) was found only in human, rabbit, and guinea
`pig. Small amounts of free oxymorphone (~10%) were excreted by
`all species except rabbit, which excreted 31. 7%. Overall recoveries
`
`of oxymorphone and metabolites from urine ranged from 15-96%,
`of which >80% was excreted in the first 24 hr by all species except
`dog. Only 35% was excreted by dog during the first day. Stereose(cid:173)
`lectivity of 6-keto- reduction was observed for all species with the
`6{3-carbinol metabolite being most abundant in the urine of all but
`guinea pig. Considerable individual variability occurred in the excre(cid:173)
`tion of free and conjugated oxymorphone by six human subjects
`following oral dosing. Species trends in the metabolism of 6-keto(cid:173)
`opioids are discussed.
`
`OM 1 is a semisynthetic narcotic analgesic derived from the(cid:173)
`baine. The hydrochloride salt (Numorphan, Endo Pharmaceuti(cid:173)
`cals, Inc.) is available for parenteral administration and is approx(cid:173)
`imately 10 times more potent than morphine (1-4). The oral to
`parenteral potency ratio of OM in humans is about I: 10 which
`makes oral OM approximately equianalgesic to im administered
`morphine (1).
`Although OM has been used clinically for the relief of pain for
`many years, no reports of its metabolism and urinary excretion
`have appeared. This study of OM in humans and laboratory
`animal species was undertaken as part of an ongoing study of the
`metabolism of6-keto-opioids, i.e. narcotic agonists and antagonists
`with a morphinan-6-one skeletal structure. Thus far, the metabo(cid:173)
`lism of naltrexone (5, 6), naloxone (6), hydromorphone (7), and
`hydrocodone (8) in humans, dogs, rats, guinea pigs, and rabbits
`has been reported. This article describes the metabolic proftle and
`urinary excretion of OM in humans and laboratory animals
`following a single dose of OM.
`
`Materials and Methods
`
`Chemicals. Drugs and chemicals were obtained from the following
`sources: chloroform (preserved with 1% ethanol) and isopropanol, Burdick
`and Jackson Laboratories, Inc., Muskegon, MI; Tri-Sil Z, Pierce Chemical
`Co., Rockford, IL; OM, Endo Laboratories, Inc., Garden City, NY; 6{3-
`hydrocodol (IS), Drug Addiction Laboratory, University of Virginia.
`6aOM and 6{30M were synthesized by the selective reduction of OM as
`reported (9) (structure 1). The structure and purity of all drugs were
`checked by thin layer chromatography and GC/MS.
`
`Abbreviations used are: OM , oxymorphone (4,5a-epoxy-3, 14-dihydroxy-17-
`methylmorphinan-6-one): 6aOM, 6a-oxymorphol ; 6,80M, 6,8-oxymorphol; GC /
`MS, gas chromatography / mass spectrometry; Cl, chemical ionization; MF, mass
`fragmentography; TMS, trimethylsilyl; IS, internal standard.
`
`Send reprint requests to: Edward J. Cone, Ph.D., National Institute on Drug
`Abuse, Addiction Research Center, P.O. Box 12390, Lexington, KY 40583.
`
`STRUCTURE I
`
`Prior to use, chloroform was treated with calcium hydroxide (USP-FCC
`food grade, J . T. Baker Chemical Co., Phillipsburg, NJ) to eliminate
`phosgene-related impurities (10).
`Animals and Human Subjects. The animals used in this study consisted
`of one. male (9.6 kg) and one female (7.5 kg) mongrel beagle dogs
`(approximately 3 years old), six male Wistar albino rats (198 ± 5 g, 6-9
`months old), six male Hartley albino guinea pigs (322 ± 30 g, 6-9 months
`old), and four male New Zealand albino rabbits (2.5 ± 0.1 kg, 10-12
`months old). The six humans in this study were healthy, adult male federal
`prison volunteers from whom informed consent was obtained under NIH
`Guidelines; all were former narcotic addicts who were drug-free at the
`time of the study and had been so for at least I week. The age (years) and
`weight (kg), respectively, of each subject were as follows: subject I, 34,
`77.1; subject 2, 33, ll8.8; subject 3, 29, 74.8; subject 4, 43, 76.2; subject 5,
`38, 86.2; subject 6, 41 , 74.8.
`
`446
`
`

`

`OXYMORPHONE METABOLISM AND URINARY EXCRETION
`
`447
`
`Drug Administration and Urine Collection. Animals were housed in
`metal cages with stainless steel urine collection pans. Control urine was
`collected for 24 hr prior to drug administration. A single dose of OM.
`HCl (2.5 mg base/kg) was administered sc and urine was collected in 24-
`hr aliquots for 2 days. After each collection, the pans were rinsed with
`water, which was added to the urine. Feces were discarded.
`Control urine from the six human subjects was collected for 24 hr prior
`to drug administration. OM-HCl (10 mg) was administered orally in
`solution and urine collections were made at 2, 4, 8, 12, 24, 48, 72, 96, and
`120 hr following drug administration. During the intervals, subjects' urine
`specimens were combined for each interval.
`Urine specimens were pooled for each laboratory animal species but
`not for the six human subjects. Following collection, the samples were
`frozen until time of analysis.
`Gas Chromatography-Mass Spectrometry. Mass spectra were obtained
`on a Finnigan model 3300 quadrupole GC/ MS operating in the methane
`Cl mode. The system was interfaced to a Finnigan model 6000 interactive
`data system. Electron energy was 80 eV. The GC consisted of a 2 mm X
`1.52 m glass column packed with 3% OV-225 on Gas-Chrom Q (100/ 120
`mesh) coupled by a glass capillary tube to the mass spectrometer. Methane
`served as reagent and carrier gas at a flow rate of20 ml/min. Temperatures
`of the injector, column, and ion source were as follows: 250, 200, and
`100°C, respectively.
`Quantitative measurements were made by MF; ions specific for each
`compound at its respective retention time were as follows: OM, m/z 502,
`5.9 min; 6,BOM, m/ z 504, 4.7 min; 6aOM, m/z 504, 5.2 min; IS, m/ z 374,
`7.2 min. Plots Qf peak height ratios of compound/ IS vs. concentration (0-
`2.0 11g/ ml for OM; 0-1.0 !lg/ ml for metabolites) were linear throughout
`the concentration range. The lower limit of detection for each compound
`was approximately 20 ng/ ml. Daily standard curves were constructed from
`five standard urine samples which were processed in the same manner as
`drug specimens. Least squares regression of these data provided slopes and
`intercepts for calculation of drug content of samples.
`Urine Hydrolysis and Extraction. Urine aliquots (10 ml) were treated
`with IS (20 11g, 6,8-hydrocodol) and extracted with and without acid
`hydrolysis. For hydrolysis, samples were mixed with concentrated HCI in
`a 10: I ratio and heated in an autoclave at 1!5°C and 1.27 kg/ cm2 pressure
`for 20 m in a manner similar to that reported for the acid hydrolysis of
`morphine (I I). Following hydrolysis and adjustment to pH 10 with sodium
`hydroxide, the hydrolyzed samples were processed in the same manner as
`those without hydrolysis.
`The extraction procedure was identical to that used for the extraction
`of naltrexone (5) with the exception that the pH of the aqueous phases
`was adjusted to 10.0 ± 0.1 rather than 9.5 and the solvent, chloroform/
`isopropanol (9: I, v / v), was substituted for chloroform. Extracts were
`evaporated to dryness under a stream of nitrogen and transferred with
`methanol to acylation tubes (Regis Chemical Co., Morton Grove, IL). The
`methanol was evaporated and the residue was treated with Tri-Sil Z (100
`!!1). The sample tube was sealed and heated for 3 hr at 90°C. Samples
`were maintained at 90°C until just prior to analysis by GC/ MS.
`Extraction Efficiency. Preliminary studies were performed to determine
`the optimum conditions for extraction of OM, 6,BOM, and 6aOM from
`urine with chloroform/isopropanol. Extraction efficiencies were highest in
`the pH range 9.5-10.5 with chloroform/isopropanol in a 9:1 (v/v) ratio.
`Recoveries ± SE at pH 10.0 for the entire extraction procedure were as
`follows: OM, 73.4 ± 1.2%; 6,BOM, 67.9 ± 0.4%; 6aOM, 72.6 ± 0.4%.
`
`Results
`
`Identification of OM and Metabolites in Urine. OM, conjugated
`OM, and 6-keto reduction metabolites (6f30M and conjugated
`6,80M) appeared immediately in human urine (0-2-hr collection)
`following the administration of a single 10-mg oral dose of OM·
`HCL These substances also appeared in the 0-24-hr urines of rat,
`dog, guinea pig, and rabbit following a single sc dose of OM·
`HCl (2.8 mg/kg). Small amounts of free and conjugated 6aOM
`
`were found in human, guinea pig, and rabbit urine. The identities
`of OM and the metabolites were established by comparison of GC
`retention times and methane GC/MS-CI spectra of the silyl
`derivatives obtained from drug urine extracts with equivalent
`extracts of authentic standards (fig. 1). Conjugated metabolites
`were cleaved by an acid hydrolysis procedure (II) prior to anal(cid:173)
`yses. Initial studies on the stability of OM, 6f30M, and 6aOM
`during acid hydrolysis ruled out the possibility of artifact forma(cid:173)
`tion or loss from hydrolysis. No other metabolites were detected
`by these methods.
`Assay Development. A reliable quantitative assay was devel(cid:173)
`oped for OM, 6f30M, and 6aOM in urine. The method consisted
`of extraction of untreated or acid-hydrolyzed urine by an estab(cid:173)
`lished procedure (5) followed by measurement with methane CI(cid:173)
`MF. Various GC liquid phases (OV-225, OV-17, Silar-5CP, SE-
`30) were tested for separation of drug components. GC columns
`packed with OV-225 provided the best resolution of all phases
`tested. The (M - 15t ions at m/z 502 and 504 for OM and 6-
`hydroxy metabolites were selected for monitoring by MF on the
`basis of their relative abundance (fig. 1). Confirmatory molecular
`ions at m/ z 517 and 519 were monitored on an occasional basis to
`ensure identification. Typical MF scans are shown in fig. 2 for
`extracts of standards, and drug urine of human and animals.
`Quantitative Studies. The amount of unchanged OM excreted
`in urine varied from about 2% (administered dose) for human and
`rat to 31.7% for rabbit (table 1). Dog and guinea pig excreted 5.3
`and 10%, respectively. Conjugated OM was present in substantial
`quantities for all species. Rabbit (11.7%) and rat (12.7%) excreted
`almost equivalent amounts, whereas human (44.1%), dog (56.4%),
`and guinea pig (81.7%) excreted considerably larger amounts.
`Metabolic reduction of the 6-keto group of OM occurred to a
`small extent in all species. Guinea pig and rabbit produced the
`largest amount of reduced metabolites (about 4% total free and
`conjugated 6aOM and 6/30M) followed by human (3.0%), dog
`(1.6%), and rat (0.2%). Stereoselectivity of reduction was found
`for all species with 6f30M being the favored isomer by all animals
`except guinea pig.
`Total drug and metabolite present in the 0- 120-hr samples for
`humans and the 0-48-hr samples for laboratory animals ranged
`from a low of 14.9% for rat to a high of 95.8% for guinea pig.
`Intermediate recoveries were found for rabbit (47.3%), human
`(49.0%), and dog (63.3%). Of these totals, greater than 80% was
`excreted in the first 24-hr period following drug administration
`for all species except dog which showed a greater excretion of OM
`and metabolites during the second day.
`Considerable intersubject variability was present for the six
`human subjects in their excretion pattern of OM and metabolites
`(table 2). Free OM excretion over a 5-day period ranged from 0.3
`to 2.5%. Excretion of conjugated OM ranged from 27.2 to 61.3%.
`Free 6f30M ranged from 0.1 to 0.7%. Excretion of conjugated
`6f30M was less variable with a range of2.1 to 3.0%. Small amounts
`of conjugated 6aOM were detected in three subjects.
`OM, conjugated OM, and conjugated 6f30M were detectable
`through 5 days and reached cumulative mean excretions of 1.9,
`44.1, and 2.6%, respectively (fig. 3). Free 6f30M was detectable
`only through 48 hr at a mean cumulative excretion of 0.3%. The
`half-life (h;2) of free OM was estimated from urinary excretion
`data of three subjects. With the Sigma-Minus method for calcu(cid:173)
`lation of t112, (12) the h;2 ± SE was 8.9 ± 0.4 hr. The mean t1;2
`± SE from urinary excretion rate plots for the same three subjects
`was 7.6 ± 2.0 hr.
`
`

`

`448
`
`CONE ET AL.
`
`TH3
`
`200
`
`250
`
`300
`
`350
`M/Z
`
`400
`
`100
`
`A.
`
`50
`
`169
`
`0
`150
`
`100
`
`B.
`
`50
`
`169
`
`_1
`
`0
`150
`
`>-
`!::=
`(f)
`z
`w
`f-
`~
`w
`>
`i=
`~
`_J
`w
`a::
`
`N ®
`~TMS
`
`0
`
`TMSO
`
`0
`
`OXYMO::H.o~1~- ITMSJ3
`2["
`
`!M-1~1+
`
`302
`
`M•
`011
`
`MH+
`
`500
`(W·I~)+
`504
`
`...
`
`(M+291+
`
`~
`550
`
`IM-891•
`
`T
`
`...
`
`450
`
`TH3
`N
`
`®
`
`OTMS
`0
`TMSO
`6,8-0XYMORPHOL- ITMSJ3
`MW•519
`
`200
`
`250
`
`300
`
`IM-891+
`430
`
`I
`
`4i;
`
`408
`
`w•
`019
`
`MH+
`
`Ill
`
`IM+291+
`
`100
`
`C.
`
`50
`
`169
`
`TH3
`
`©fib
`
`OTMS
`TMSOc=) 0
`6a- OX YMORPHOL- I TMS 1
`3
`MW•519
`
`350
`M/Z
`
`400
`
`450
`
`(M -89)+
`414 430
`
`•••
`550
`
`I
`500
`IM-Ia)•
`504
`
`w•
`019
`
`0
`150
`
`200
`
`250
`
`300
`
`350
`M/Z
`FIG. 1. Methane chemical ionization spectra of the silyl derivatives of oxymorphone, 6{3-oxymorphol, and 6a-oxymorphol.
`
`400
`
`450
`
`500
`
`550
`
`TABLE I
`Recovery of drug and metabolites from urine following a single dose of oxymorphone hydrochloride
`These data represent the means of triplicate determinations and are expressed as percentage of administered dose. Conjugated (Conj) drug and
`metabolites were determined by subtraction of free from total concentration after acid hydrolysis.
`OM
`6,BOM
`
`Species (N)
`
`Dose/ Route
`
`Time
`
`Free
`
`Conj
`
`Free
`
`Conj
`
`Free
`
`Conj
`
`6aOM
`
`Total
`
`Total First 24 hr
`
`Human (6)
`Rat (6)
`Dog (2)
`Guinea pig (6)
`Rabbit (6)
`
`10 mg (oral)
`2.5 mg/kg (sc)
`2.5 mg/kg (sc)
`2.5 mg/kg (sc)
`2.5 mg/kg (sc)
`
`hr
`0-120
`0-48
`0-48
`0-48
`0-48
`
`1.9
`2.0
`5.3
`10.0
`31.7
`
`44.1
`12.7
`56.4
`81.7
`11.7
`
`%
`
`0.3
`0.1
`0.4
`0.3
`2.0
`
`2.6
`0.1
`1.2
`1.0
`1.1
`
`0
`0
`0
`0.5
`0.6
`
`0.1
`0
`0
`2.3
`0.2
`
`%
`49.0
`14.9
`63.3
`95 .8
`47.3
`
`%
`82
`97
`35
`94
`99
`
`Discussion
`
`The metabolite profile of OM in humans and laboratory animals
`is similar in a number of respects to that observed for other 6-
`keto-opioids. The high degree of conjugation of OM for all species
`except rabbit is found with only minor exception for naltrexone
`(6), naloxone (6), and hydromorphone (7). The free phenolic
`group common to each of these substances is highly accessible for
`conjugation (presumably as the glucuronide) by liver microsomal
`enzymes similar to that found for morphine (13). The low ratio of
`conjugated material to intact OM in the rabbit appears to be an
`exceptional case since more conjugate than free parent was found
`in urine following naltrexone (6), naloxone (6), hydromorphone
`(7), and hydrocodone (8) administration. Saturation of enzyme
`systems does not appear to be the reason for the low conjugate/
`
`parent excretion ratios of OM in rabbit since the dosage (2.5 mg/
`kg) was intermediate in the dosage range of the other 6-keto(cid:173)
`opioids, all with high conjugate/parent ratios. The dosage in the
`other studies varied from approximately 1-5 mg/kg (6- 8). It seems
`more likely that differences in liver glucuronyltransferase activity
`toward OM vs. the other 6-keto-opioids account for the observed
`differences in glucuronidation. Separate enzymes have been iso(cid:173)
`lated from rabbit liver microsomes for the glucuronidation of
`morphine and p-nitrophenol ( 13) indicating narrow substrate spec(cid:173)
`ificities within this family of conjugating enzymes.
`The presence of the free phenolic group of OM also appears to
`preclude significant N-dealkylation, similar to that found for
`naltrexone (6), naloxone (6), and hydromorphone (7). Methylation
`of the phenolic group substantially increases N-demethylation as
`evidenced by the N-demethylation of hydrocodone (8) and oxy-
`
`

`

`OXYMORPHONE METABOLISM AND URINARY EXCRETION
`
`449
`
`A. STANDARDS
`
`B. HUMAN(8-12HR)
`
`OM
`
`6/IOM
`\
`
`C. DOG (0-24HR)
`OM
`
`6/IOM
`I
`
`504X4
`
`502
`"-Is
`
`502
`
`IS
`
`374
`
`374
`
`374XIO
`
`100
`
`200
`
`300
`
`100
`
`200
`
`300
`
`100
`
`200
`
`300
`
`D. RAT (0-24HR)
`
`E. GUINEA PIG(0-24HR) f RABBIT(0-24HR)
`OM
`
`OM
`
`OM
`
`~
`
`6aOM
`6/IOM, \
`
`502X4
`...J----'1 ' - (cid:173)
`'-IS
`
`374
`\.__
`
`502
`
`-
`
`IS
`
`374 X2
`
`502
`
`t
`
`w
`en
`:z:
`0 a..
`en
`w
`a::
`
`a::
`0
`1-
`u w
`1-
`w
`0
`
`t
`
`w
`en
`:z:
`0 a..
`en
`w
`a::
`a::
`0
`1-u
`w
`1-
`w
`0
`
`200
`100
`SCAN NUMBER
`FIG. 2. Mass fragmentography recordings of extracts of standards and human and animal urine following oxymorphone administration.
`
`300
`
`100
`
`200
`
`300
`
`100
`
`I
`200
`
`300
`
`TABLE 2
`Urinary excretion of oxymorphone and metabolites in six human subjects
`These data are expressed as percentage of dose and represent the cumulative means of triplicate determinations of samples collected over a 120-hr
`period following a single 10-mg oral dose of oxymorphone hydrochloride.
`OM
`
`Subject
`
`Free
`
`Conj
`
`Free
`
`Conj
`
`Free
`
`Conj
`
`6,BOM
`
`6crOM
`
`Total
`
`2
`3
`4
`5
`6
`Mean± SE
`
`2.5
`5.0
`0.6
`0.7
`0.3
`2.4
`1.9 ± 0.7
`
`28.5
`27.2
`63.1
`48.8
`42.0
`54.9
`44.1 ± 5.9
`
`%
`
`0.6
`0.2
`0.1
`0.2
`0.7
`0.2
`0.3 ± 0.1
`
`2.8
`2.5
`2.1
`2.4
`2.8
`3.0
`2.6 ± 0.1
`
`0
`0
`0
`0
`0
`0
`0
`
`0.1
`0.2
`0
`0.3
`0
`0
`0.1 ± 0.1
`
`%
`
`34.5
`35.1
`65.9
`52.4
`45.8
`60.5
`49.0 ± 5.3
`
`codone (14), the 0-methyl ether of OM. Whether these effects of
`the phenolic group on N-dealkylation are a result of differences in
`enzyme specificities or differences in substrate accessibility due to
`the higher lipid solubilities of the ethers vs. free phenols remains
`unclear.
`
`Reduction of the ketone group of OM to 6a- and 6/3-carbinols
`also occurred for naltrexone (6), naloxone (6), and hydromor(cid:173)
`phone (7) for all species. Reduction of hydrocodone also was
`evident for all species except dog (8). The degree of reduction of
`OM, naltrexone, and naloxone varied from species to species, but
`
`

`

`450
`
`50
`
`CONE ET AL.
`
`--·-------·-------·
`
`• - - • CONJUGATED OM
`•----• OM
`o -o CONJUGATED 6,80M
`o-----os,eoM
`
`Emerging species trends from these and previous studies on the
`metabolism of 6-keto-opioids indicate that of the five mammalian
`species: a) conjugation (glucuronidation) activity is greatest in the
`dog, but is high in human, guinea pig, and rabbit; b) 6-ketore(cid:173)
`ductase activity is high in human and guinea pig followed by
`rabbit and is almost nonexistent in rat and dog; c) 6-keto reduction
`of opioids with a free phenolic group is stereoselective; reduction
`generally favors formation of the 6-,8-hydroxymetabolite for all
`species except guinea pig; d) N-dealkylation of 6-keto-opioids with
`a free phenolic group is not a significant metabolic pathway but
`becomes important with 0-methyl ethers such as hydrocodone for
`human and dog. Perhaps further metabolic studies of the opioids
`will add to these generalizations of species trends in metabolism.
`
`References
`
`l. W. T. Beaver, S. L. Wallenstein, R . W. Houde, and A. Rogers,
`Comparison of the analgesic effects of oral and intramuscular
`oxymorphone and of intramuscular oxymorphone and morphine in
`patients with cancer. J. Clin. Pharmacal. 17, 186-198 (1977).
`2. N. B. Eddy and L. E. Lee, The analgesic equivalence to morphine and
`relative side action liability of oxymorphone (14-hydroxydihydro(cid:173)
`morphinone). J. Pharmacal. Exp. Ther. 125, 116-121 (1959).
`3. T. J. De Kornfeld, A clinical and laboratory study of hydroxydihy(cid:173)
`dromorphinone (Numorphan HCl). Fed. Proc. 20, 309 (1961).
`4. A. S. Keats and J. Telford, Studies of analgesic drugs. V. The
`comparative subjective effects of oxymorphone and morphine. Clin.
`Pharmacal. Ther. I, 703-707 (1960).
`5. E. J. Cone, C. W. Gorodetzky, and S. Y. Yeh: The urinary excretion
`profile of naltrexone and metabolites in man. Drug Metab. Dispos.
`2, 506-512 (1974).
`6. E. J. Cone, in "Critical Concerns in the Field of Drug Abuse" (A.
`Schecter, H. Alksne, and E. Kaufman, eds.), p. 1228. Marcel Dekker,
`New York, 1978.
`7. E. J. Cone, B. A. Phelps, and C. W. Gorodetzky, Urinary excretion of
`hydromorphone and metabolites in humans, rats, dogs, guinea pigs
`and rabbits. J. Pharm. Sci. 66, 1709-1713 (1977).
`8. E. J. Cone, W. D. Darwin, C. W. Gorodetzky, and T. Tan, Compar(cid:173)
`ative metabolism of hydrocodone in man, rat, guinea pig, rabbit,
`and dog. Drug Metab. Dispos. 6, 488-493 (1978).
`9. E. J. Cone, General procedure for the isolation and identification of
`6o:- and 6{3-hydroxy metabolites of narcotic agonists-and antagonists
`with a hydromorphone structure. J. Chromatogr. 129, 355-361
`(1976).
`10. E. J. Cone, W. F. Buchwald, and W. D. Darwin, Analytical controls
`in drug metabolic studies. II. Artifact formation during chloroform
`extraction of drugs and metabolites with amine substituents. Drug
`Metab. Dispos. 10,561-567 (1982).
`II. S. Y. Yeh and L. A. Woods, Determination of radioactive-labeled
`codeine, morphine, dihydromorphine, and their metabolites in bio(cid:173)
`logical materials. J. Pharm. Sci. 59,380-384 (1970).
`12. M. Gilbaldi and D. Perrier, in "Pharmacokinetics" (J. Swarbrick, ed.),
`vol. I, p. 287. Marcel Dekker, Inc., New York, 1975.
`13. E. D. Villar, E. Sanchez, and T. R. Tephly, Morphine metabolism. V.
`Isolation of separate glucuronyltransferase activities for morphine
`and p-nitrophenol from rabbit liver microsomes. Drug Metab. Dis(cid:173)
`pos. 5, 273-278 (1977).
`14. S. H. Weinstein and J. C. Gaylord, Determination of oxycodone in
`plasma and identification of a major metabolite. J. Pharm. Sci. 68,
`527-528 (1979).
`
`--c-------o-------c-------o
`-·-------·---------------
`
`- - -
`
`0"
`r:l
`·---
`1 ......... ..
`-o·c··--o----- ---o
`72
`48
`24
`Tl ME I HOURS
`FIG. 3. Mean cululative urinary excretion of free and conjugated
`oxymorphone and 6{3-oxymorphol for six human subjects following a 10-
`mg oral dose of oxymorphone hydrochloride.
`
`--0--
`
`,o
`
`.
`
`96
`
`120
`
`was consistently low (<10% of administered dose) for rat, dog,
`and rabbit. A highly variable rate of reduction has been found for
`humans and guinea pigs with up to 40% of the dose of naltrexone
`being excreted by humans as reduced metabolites (5). The selec(cid:173)
`tivity of reduction (6o:- vs. 6,8-epimer formation) also was variable
`with the 6,8-carbinol being highly favored by most species except
`guinea pig.
`Overall urinary recoveries of OM and metabolites were quite
`comparable with those found for other 6-keto-opioids. Recoveries
`for the 6-keto-opioids for rat range from 3-17% (6, 8), whereas
`recoveries for the remaining species generally range from 30-60%
`(6-8) of administered dose. It is likely that the remainder of the
`dose was present as polar unidentified metabolites as well as being
`partially eliminated in feces. The moderate recovery of OM
`(-50%) and other 6-keto-opioids from human urine following oral
`dosing indicates that these compounds are well absorbed from the
`gastrointestinal tract. Excretion of free OM in urine was low for
`all species (2-10%) in this study except rabbit. This also appears
`to be an exception for this species since the rabbit generally has
`excreted significantly more conjugated metabolite than free parent
`drug in the 6-keto-opioid series (6-8).
`Individual variation in human excretion of OM and metabolites
`was quite similar to that found for hydromorphone (7), the
`chemical derivative of OM without a 14-hydroxy group. Like
`hydromorphone, excretion of free OM varied over a 10-fold range
`within six subjects, whereas conjugated OM varied only over an
`approximate 2-fold range. Also, mean keto reduction for these
`two compounds was less than 3%.
`
`