Bioavailability of Sublingual
`Buprenorp hine
`John Mendelson, MD, Robert A. Upton, PhD, E. Thomas Everhart, PhD,
`Peyton Jacob III, PhD, and Reese T. Jones, MD
`
`Buprenorphine administered sublingually is a promising treatment for opiate depen-
`dence. Utilizing a new, sensitive, and specific gas chromatographic electron-capture
`detector assay, the absolute bioavailability of sublingual buprenorphine was determined
`in six healthy volunteers by comparing plasma concentrations after 3- and 5-minute
`exposures to 2 mg sublingual and 1 mg intravenous buprenorphine. The amount of
`unabsorbed buprenorphine in saliva was measured after 2-, 4-, and 10-minute exposures
`to 2 mg sublingual buprenorphine in 12 participants. Pharmacokinetic parameters were
`analyzed by analysis of variance; bioequivalence was evaluated by the Schuirmann
`two-sided test. The 3- and 5- minute sublingual exposures each allowed 29 t 10%
`bioavailability (area under the plasma concentration-time curve unextrapolated) and
`were bioequivalent. Buprenorphine recovered from saliva after 2-, 4-, and 10-minute
`exposures was, on average, 52% to 55% of dose. Increased saliva pH was correlated with
`decreased recovery from saliva. Study results indicate that bioavailability of sublingual
`buprenorphine is approximately 30%. Sublingual exposure times between 3 and 5
`minutes produce equivalent results. Buprenorphine remaining in saliva causes an al-
`most twofold overestimation of bioavailability.
`
`B (20-40 times greater analgesic potency than mor-
`
`uprenorphine is a synthetic, lipophilic, potent
`
`phine) oripavine opiate analgesic effective in the
`Low oral bio-
`treatment of opiate
`availability (approximately 14%),4 caused largely by
`hepatic first-pass metabolism, makes sublingual ad-
`ministration an attractive alternative for treatment.
`In clinical trials, buprenorphine was administered
`as a 30% ethanol solution with participants retaining
`the dose sublingually for up to 10 minutes. Assess-
`ment of the pharmacokinetics of buprenorphine has
`
`From the Drug Dependence Research Center, Langley Porter Psychiatric
`Institute (Drs. Mendelson, Upton, Everhart, Jacobs, and Jones), the De-
`partment of Biopharmaceutical Sciences (Dr. Upton), and the Division
`of Clinical Pharmacology and Experimental Therapeutics (Dr. Jacob),
`University of California, San Francisco, San Francisco, California. Sup
`ported in part by United States Public Health Service grants DA01696
`and DA00053 and contract No. 271-90-7307 from the National Institute
`on Drug Abuse, National Institutes of Health. Submitted for publication
`July 20, 1996; accepted in revised form October 16, 1996. Address
`for reprints: John Mendelson, MD, Drug Dependence Research Center,
`Langley Porter Psychiatric Institute, University of California, San Fran-
`cisco, 401 Parnassus Avenue, San Francisco, CA 94143-0984.
`
`been hampered by the difficulty in quantifying low
`plasma 1e~el.s.~ By measuring the amount of bupren-
`orphine remaining in saliva after 2.5 or 10 minutes
`of sublingual exposure, a prior study inferred a sub-
`lingual bioavailability of 25% to 50% . 6 Differences in
`the amount of buprenorphine recovered from saliva
`with increased exposure were not evident.
`Using a recently developed, sensitive, and specific
`gas chromatographic electron-capture detector (GC-
`ECD) assay for buprenorphine in plasma, absolute
`bioavailability of sublingual buprenorphine was esti-
`mated by comparing the plasma concentrations
`achieved with those from an intravenous dose. Sub-
`lingual exposure times of 3 and 5 minutes for bio-
`availability of buprenorphine were compared using
`the plasma-based method. In a separate study, those
`bioavailabilities were compared with estimates from
`a less direct method based upon the amount of bu-
`prenorphine recovered in saliva after exposures of 2,
`4, and 10 minutes.
`MATERIALS AND METHODS
`Subjects
`Six healthy volunteers (five men, one woman), 21 to
`38 years of age (mean -+ SD = 29 % 6 years), partici-
`
`J Clin Pharmacol 1997;37:31-37
`
`31
`
`from the SAGE Social Science Collections. All Rights Reserved.
`
`Page 1
`
`RB Ex. 2012
`BDSI v. RB PHARMACEUTICALS LTD
`IPR2014-00325
`
`

`

`MENDELSON ET AL
`
`TABLE I
`Inclusion and Exclusion Criteria for Bioavailabili and
`Saliva Recovery Studies
`
`Inclusion Criteria
`Male or female and between 21 and 40 years of age
`Experienced in the use of illicit opiates
`In good health with normal physical examination and
`laboratory screening test results
`Within 215% of ideal body weight
`Without oral cavity pathologic conditions
`Women must have a negative urine pregnancy test result
`before each experimental session and must use barrier
`birth control methods until completion of the study
`Exclusion Criteria
`A history of clinically significant medical or psychiatric
`disorders
`Opiatedependence, as defined by the DSMIII-R criteria, or
`dependence on other psychoactive drugs other than
`nicotine or caffeine
`Known hypersensitivity to buprenorphine or other opiate-
`like analgesic agents
`Current treatment with any prescription medication
`DMSIII-R, Diagnostic and Statistical Manual, Third Edition (Revised).
`
`pated in the bioavailability study. Twelve similar
`volunteers (nine men, three women], 2 1 to 40 years
`of age (mean 2 SD = 32 ? 6 years), participated in
`the saliva recovery study. Inclusion and exclusion
`criteria for both studies are in Table 1. Written, in-
`formed consent was obtained. The protocols were
`approved by the Committee on Human Research,
`University of California, San Francisco.
`All participants were studied as outpatients. Each
`dose was administered after an overnight fast and
`a 12-hour requested abstinence from psychoactive
`drugs [including nicotine and caffeine). The partici-
`pants were not allowed to drink fluids or smoke ciga-
`rettes for 1 hour after drug administration. A low-fat
`lunch was provided 4 hours after administration.
`
`Study Design
`
`The data presented come from two separate studies.
`For the plasma-based bioavailability study, six parti-
`cipants were administered buprenorphine at approx-
`imate weekly intervals under the following experi-
`mental conditions: 2 mg buprenorphine in 30% etha-
`nol solution held sublingually for 3 minutes, 2 mg
`buprenorphine in 30% ethanol solution held sublin-
`gually for 5 minutes, and 1 mg buprenorphine by
`intravenous infusion. The sublingual conditions
`
`32 J Clin Pharmacol 1997;37:31-37
`
`were randomized in sequence. The intravenous infu-
`sion was always administered between the sublin-
`gual treatments.
`For the saliva recovery study, 1 2 participants were
`studied in a 3 X 3 balanced Latin Square design, with
`at least 3 days between sessions, under the following
`conditions: 2 mg buprenorphine in 30% ethanol so-
`lution held sublingually for 2 minutes, 2 mg bupren-
`orphine in 30% ethanol solution held sublingually
`for 4 minutes, and 2 mg buprenorphine in 30% etha-
`nol solution held sublingually for 10 minutes.
`Medications
`Buprenorphine hydrochloride was supplied by the Na-
`tional Institute on Drug Abuse, prepared as 2 mg/mL
`in 30% ethanol solution. A 3.3-mL aliquot of commer-
`cially available buprenorphine injection solution (Bu-
`prenex in 0.3 mg/mL ampules, Reckitt and Colman
`Products, Ltd.; United Kingdom] was diluted to 30 mL
`with 0.9% NaC1, and used for the intravenous dose.
`Dose Administration and Sample Collection
`The intravenous dose of buprenorphine was infused
`into a forearm vein at a rate of 1 mL/min under sy-
`ringe pump control over a 30-minute period. Sublin-
`gual doses were administered with a 1-mL tuberculin
`syringe. The buprenorphine solution was placed in
`the right posterior sublingual area at the base of the
`tongue. Participants did not swallow after adminis-
`tration until instructed to do so by an observer. In the
`bioavailability study, when instructed, participants
`swallowed once to terminate exposure and thereafter
`swallowed ad libitum. In the saliva recovery study,
`sublingual exposure was terminated by spitting the
`remaining solution and accumulated saliva into a
`preweighed 30-mL centrifuge tube. Two separate
`rinses of 35 mL of distilled water were then swirled
`around the oral cavity for 30 seconds each and col-
`lected in a preweighed jar. Saliva pH was measured
`before and after administration with a microprobe
`electronic pH meter (Lazar Research Lab; Los
`Angeles, CAI. Samples were frozen at -20°C until
`analyzed.
`Plasma samples (10 mL) were obtained through an
`intravenous catheter in the forearm of the nondomi-
`nant hand. Samples were collected before buprenor-
`phine and at 5,20, 30, and 40 minutes, and 1, 1.5, 2,
`3,4,5,6,8,10,12, and 24 hours after administration.
`Determination of Buprenorphine Concentration in
`Plasma
`The assay involved a three-step extraction of analyte
`from 1.0 mL of plasma sample spiked with 50 ng
`
`Page 2
`
`

`

`BZOA VAILABZLZT Y OF S UBLINGUAL B UPRENORPHINE
`
`of internal standard (N-n-propylnorbuprenorphine).
`Buprenorphine and internal standard in the extract
`were converted to the heptafluorobutyryl derivatives
`using heptafluorobutyric anhydride, excess reagent
`was removed under vacuum, and the residues were
`reconstituted in 20 pL of n-butyl acetate. Samples (3
`pL) were analyzed by gas chromatograph using a 25
`m x 0.2 mm (internal diameter) fused silica capillary
`column, splitless injection, and electron-capture de-
`tection. Results were quantitated by measuring chro-
`matographic responses of a series of calibration stan-
`dard samples prepared with each run.
`Calibration curves were linear hom 0.1 ng/mL to 20
`ng/mL. The limit of quantitation was 0.1 ng/mL for all
`except two of the runs (which were 0.3 ng/mL and 0.2
`ng/mL). Accuracy and precision of the method were
`such that replicate assays of spiked control samples at
`0.1 ng/mL and 0.2 ng/mL, or 0.15, 0.5, 2.0, 5.0, 10, or
`15 ng/mL, assayed concurrently with study samples,
`had coefficients of variation ranging from 3.97% to
`18.47%, and a bias of only -8.6% to +9.4% (for each
`control, n = 1-3 per run).
`
`Determination of Buprenorphine Concentration in
`Saliva
`
`Saliva samples were diluted one hundredfold and
`mouth-rinse samples tenfold with 0.01 mol/L sulfu-
`ric acid. Aqueous standards were used neat. One mil-
`liliter of sample, 100 pL of 30 pg/mL N-n-pentylnor-
`buprenorphine, 0.5 mL of 1 mol/L NaOH, and 2 mL
`of ethyl acetate/heptane (4:l vol/vol) were com-
`bined, vortexed 5 minutes, centrifuged at 5,000 g for
`10 minutes, and the aqueous phase frozen by immer-
`sion in a dry ice and acetone bath. The organic phase
`was then decanted, the aqueous phase reextracted
`with another 2-mL aliquot of ethyl acetate/heptane
`(4:l vol/vol) as described above, and the second or-
`ganic extract added to the first. The combined or-
`ganic extracts were evaporated to dryness, reconsti-
`tuted in 0.5 mL of high-performance liquid chroma-
`tography mobile phase, and 25-pL aliquots were
`injected via autosampler into the high-performance
`liquid chromatograph.
`The high-performance liquid chromatography sys-
`tem consisted of an autosampler (Model WISP 7100,
`Waters Associates; Milford, MA), Shimadzu pump
`(Model LC-6A, Shimadzu Corp.; Kyoto, Japan), ultra-
`sphere ODS column (&, average particle diameter
`5 p, 4.6 x 25 cm) (Model 235329, Beckman Instru-
`ments; Fullerton, CA), fluorescence detector (Shi-
`madzu model RF-5351, and Hewlett-Packard integ-
`rator (Model 3397A, Hewlett-Packard; Wilmington,
`DE). The mobile phase was acetonitrile/O.l% phos-
`phoric acid in a 40:60 ratio, with pH adjusted to 3.0
`
`with aqueous NaOH. The flow rate was 1 mL/min.
`The detector excitation and emission wavelengths
`were 285 nm and 355 nm, respectively. Retention
`times for buprenorphine and N-n-pentylnorbupren-
`orphine were -4.8 and -9.8 minutes, respectively.
`Norbuprenorphine, buprenorphine’s major metabo-
`lite, had a retention time of -3.1 minutes and was
`not detected in saliva or mouth-rinse samples.
`Buprenorphine standard samples were prepared
`by diluting buprenorphine HC1 in 0.01 mol/L of sul-
`furic acid. Ten or more standards, spanning the range
`from 0 pg/mL to 10 pg/mL, were included with each
`batch of samples and used to construct a standard
`curve based on peak area ratios of buprenorphine
`and N-n-pentylnorbuprenorphine. Standard curves
`were linear in the range of 0 pg/mL to 10 pg/mL.
`Control samples, prepared by spiking blank saliva
`and mouth rinse, at concentrations spanning the ex-
`pected concentration ranges, were included in each
`batch of samples.
`N-n-propylnorbuprenorphine was synthesized in
`the laboratory by reductive alkylation of norbupren-
`orphine with propionaldehyde and sodium borohy-
`dride and was converted to hydrochloride salt and
`crystallized from ethanol solution by addition of
`ether. Thin layer chromatography (TLC) revealed
`complete conversion of norbuprenorphine to the
`propyl derivative. N-n-pentylnorbuprenorphine was
`prepared in an analogous manner from norbuprenor-
`phine and n-pentanal, and was used as the free base.
`Calibration curves were linear from 0.2 pg/mL to
`30 pg/mL. The limit of quantitation was 0.2 pg/mL.
`Accuracy and precision of the method were such that
`replicate assays of spiked control samples at 0.2, 2,
`5, and 30 pg/mL had coefficients of variation ranging
`from 2.63% to 6.92% and a bias of 2% to 4% (for
`each control, n = 5 per run).
`
`Pharmacokinetic Analysis
`
`The area under the plasma concentration-time
`curve (AUC) was estimated from the time of adminis-
`tration (t = 0) to the time of the last assayed sample
`(t = t,), using the trapezoidal equation for periods of
`increasing or stationary concentration, and the loga-
`rithmic-trapezoidal equation for periods of decreas-
`ing concentration.’ This area was extrapolated from
`t, to infinity (AUCo-,) by dividing the last concentra-
`tion measured by an estimate of the terminal log-
`linear slope. The terminal slope was estimated, for
`purposes of extrapolating AUC, from a least-squares
`linear fit (unweighted) to the last three time points
`of the plasma concentration-time curve (semiloga-
`rithmic). The peak plasma concentration (Cmax) was
`taken as the concentration in the plasma sample hav-
`
`PHARMACOKINETICS
`
`33
`
`Page 3
`
`

`

`MENDELSON ET AL
`
`hours after administration in the bioavailability
`study and at 1, 2 , 3 , 4 , 5, 6, and 7 hours in the saliva
`recovery study.
`
`STATISTICAL ANALYSIS
`
`Statistical analyses were performed on AUCs, C,,
`,
`and peak time (tmaX) using the Statistical Analysis
`System (SAS; version 6.10) program (SAS Institute,
`Cary, NC). The t,,, was analyzed untransformed, and
`AUC and C,,,
`as their logarithmic (natural) trans-
`forms, divided by the varying dose sizes. For AUC
`from all three doses, dose was analyzed as an in-
`tersubject effect and sequence as an intrasubject ef-
`fect, with the dose-by-sequence interaction being
`evaluated as a surrogate for period effect (because
`of the limited number of degrees of freedom). Peak
`concentrations and tmas were compared by analysis
`of variance (ANOVA) between the two sublingual
`treatments only, allowing dose and period each to
`be analyzed as intersubject effects and sequence as
`an intrasubject effect.
`Bioequivalence between the two sublingual treat-
`ments, measured by AUC, C,,,
`, and t,,,
`, was evalu-
`
`
`
`-
`
`0
`
`3
`5
`Sublingual Exposure (min)
`
`Figure 2. Individual areas under the concentration-time curve
`[AffCJfor 3- and 5-minute sublingual exposures 10-0, one indi-
`vidual).
`
`0.01 I
`
`0
`
`0.5
`
`I
`
`1.5
`
`I
`2.5
`2
`Time (hours)
`
`I
`
`3
`
`3.5
`
`4
`
`Figure I . Buprenorphine in plasma after 1 -mg intravenous dose
`(solid line, mean; 0, individual values], and 2-mg sublingual
`doses: 3-minute exposure (dotted line, mean; 0, individual val-
`ues]; 5-minute exposure (dashed line, mean; A, individual val-
`ues); n = 6 for all doses.
`
`ing the highest concentration. No attempt was made
`to interpolate concentrations between sampling
`times. The total amount of buprenorphine base re-
`maining in the saliva and in each of the two rinses
`was determined. The values for saliva plus first and
`second rinses were summed.
`
`Drug Effect Measures
`
`Heart rate and systolic and diastolic blood pressure
`were measured with a cardiovascular monitor
`(Model VSM-2, Physio-Control Corp.; Redmond,
`WA). Respiratory rate was measured by counting the
`number of inhalations per minute. Verbal ratings of
`global intoxication on a 0 to 100 scale, with 0 repre-
`senting no effect and 100 the maximum effect experi-
`enced after opiate drugs were obtained in both stud-
`ies. Additional measures of subjective drug effects
`were obtained in the bioavailability study using self-
`ratings of symptom intensity on three subscales mea-
`suring euphoria, sedation, and dysphoria (MBG,
`PCAG, LSD) from the Addiction Research Center In-
`ventory? a 31-item, adjective-rating checklist con-
`sisting of opiate agonist and antagonist symptoms,
`visual analog scales (range, 0- 100) measuring
`“good” drug effect, “bad” drug effect, “high,” drunk-
`enness, sickness, and hangover, and the Profile of
`Mood Scale.g The self-ratings were obtained before
`drug administration and at 1, 2, 4, 6, 8, 10, and 1 2
`
`34
`
`J Clin Pharrnacol 1997;37:31-37
`
`Page 4
`
`

`

`BIOA VAILA BILITY OF S UBLINGUAL B UPRENORPHINE
`
`Pharmacokinetic
`Parameters
`
`3-minute
`SL Dose A,
`n = 6
`(2 mg)
`
`5-minute
`SL Dose B,
`n = 5
`(2 mgl
`
`TABLE II
`Pharmacokinetics of Buprenorphine Measured in Plasma
`IV Infusion
`Dose C,
`n = 6
`(1 mgl
`
`8.75 t 4.75 8.89 2 5.22
`
`14.3 2 8.7
`13.2 2 8.8
`1.60 -t 0.66 1.72 2 0.87
`1.25 2 0.42 1.62 -c 0.55
`
`14.7 -C 3.5
`
`Statistical P Value for
`Treatment Effect
`(Contrasts)
`0.0001
`(A = B,A # C, B # C)
`0.0002
`18.4 -t 6.5
`(A = B, A # C, B # C)
`14.3 -+ 3.0
`0.114'
`0.44 t 0.09 0.474'
`62.5 2 21.8
`16.2 -t 20.1
`
`Ratio
`3-minute
`SLIV
`
`Ratio
`5-minute
`SL:IV
`
`Ratio
`5-minute SL:
`3-minute SL
`
`0.28 t 0.10 0.29 t 0.10 1.11 z 0.12
`
`0.36 2 0.13 0.33 t 0.13 0.95 z 0.18
`1.13 z 0.12
`0.32 z 0.84t
`
`AUC unextrapolated
`(hr . ng/mL)
`AUC extrapolated
`(hr . ng/mL)
`C,,,
`(ng/mL)
`(hr)
`t,,,
`CI (Lhr)
`(MI
`* Analysis of variance for peak concentration and peak time, performed on the data from the sublingual treatments only.
`t Difference, 5-minute sublingual minus 3-minute sublingual.
`t Approximate estimate limited by assay sensitivity considerations.
`Values are presented as the mean I+_ standard deviation. SL, sublingual; IV, intravenous; AUC, area under the concentra-
`time to C,,,; CI, clearance; tin, elimination half-life.
`, peak concentration; t,,,,
`tion-time curve; C,,,
`
`ated using the two one-sided (a = 0.05) confidence
`intervals (Schuirmann) tests. The mean square error
`term for all three variables in this analysis derived
`from a two-sample ANOVA that compared the sub-
`lingual treatments only and included a period effect.
`Subjective drug effects were analyzed by ANOVA.
`Saliva concentration data were also analyzed by
`ANOVA, with sublingual exposure time as the in-
`tersubject factor and exposure sequence as the intra-
`subject factor. The relationship between saliva pH
`and recovery of buprenorphine in saliva was investi-
`gated by linear correlation analysis of data pooled
`across all three exposure times.
`
`RESULTS
`
`Pharmacokinetics in Plasma
`
`Mean plasma concentrations of buprenorphine are
`shown in Figure 1; unextrapolated AUCs for each
`participant are shown in Figure 2; pharmacokinetic
`parameter estimates are shown in Table 11.
`The two sublingual doses each had smaller AUCs
`(after adjustment of dose) than did the intravenous
`dose ( P < 0.0002), whether AUC was estimated un-
`or t,,,
`extrapolated or extrapolated. No AUC, C,,,,
`differences were evident between the two sublingual
`exposure times. With each AUC and peak concentra-
`tion, bioequivalence between the two sublingual
`treatments was confirmed by two one-sided confi-
`dence intervals (Schuirmann) tests. Bioequivalence
`for the 3-minute treatment was within 80% to 125%
`of that for the 5-minute treatment (P 5 0.05). No
`
`significant sequence or period effect (dose-by-se-
`quence interaction) was evident.
`
`Pharmacokinetics in Saliva
`
`The amount of buprenorphine remaining in saliva
`and mouth rinses was not significantly different after
`the 2-, 4-, and 10-minute exposures (1.01 ? 0.5 mg,
`0.97 ? 0.5 mg, and 0.98 2 0.3 mg, respectively, corre-
`sponding to 55 k 26%, 52 2 25%, and 53 2 15% of
`the dose; Table 111). No significant sequence effect
`or period effect (dose-by-sequence interaction) was
`seen. Saliva pH was significantly but not closely cor-
`related with recovery of buprenorphine (Figure 3).
`With increasing saliva pH, less buprenorphine was
`recovered (r = -0.33, P = 0.05 saliva alone; r =
`-0.40, P = 0.02 saliva plus mouth rinses).
`
`Drug Effect Measures
`
`No significant differences were evident in any drug
`effect measures in either experiment.
`
`DISCUSSION
`
`The absolute bioavailability (Cl/F) of buprenorphine
`from a sublingual solution dose in ethanol was 28%
`to 36%. Sublingual holding times between 3 and 5
`minutes were bioequivalent in the extent of bioavail-
`ability that resulted. Although differences in t,,,
`could not be established statistically between 3- and
`5-minute treatments, bioequivalence between the
`two treatments could not be established either. How-
`
`PHARMACOKINETICS
`
`35
`
`Page 5
`
`

`

`MENDELSON ET AL
`
`Sublingual
`Exposure
`Saliva Accumulated
`Duration
`mL
`PH
`(min)
`mL/min
`2
`3.7
`7.42 t 0.47
`1.8
`1.1
`4.5
`4
`7.10 +- 0.44
`6.74 ? 0.42
`0.8
`8.3
`10
`Values are presented as the mean -c standard deviation; n = 12.
`
`TABLE 111
`Buprenorphine Recovered in Saliva
`Buprenorphine
`in Saliva and
`Rinses
`(mg base)
`1.01 ? 0.5
`0.97 t 0.5
`0.98 t 0.3
`
`Buprenorphine
`in Saliva
`(mg base)
`0.87 t 0.5
`0.80 ? 0.4
`0.82 t 0.3
`
`Dose Recovered in Saliva
`and Rinses
`
`%
`55 t 26
`52 t 25
`5 3 t 15
`
`Range (%)
`4-98
`20-83
`24-81
`
`ever, with an observed difference of only 19 minutes
`and with bioequivalence having been established us-
`ing AUC and C,,,, any difference in time points of
`Cmaxs would probably not be of clinical importance.
`Indeed, exposure times between 2 and 10 minutes
`did not have a significant impact on subjective ef-
`fects.
`The amount of buprenorphine recovered from sa-
`liva after 2-, 4-, and 10-minute exposures was 52%
`to 55%. Results of both the saliva study reported
`here and a prior study,6 using saliva recovery to esti-
`mate the upper limits of sublingual bioavailability,
`overestimate true bioavailability almost twofold.
`This discrepancy between the plasma and saliva
`methods could be attributable to incomplete absorp-
`tion into the buccal tissues before spitting or swal-
`lowing, slow reequilibration back from the buccal
`
`0
`
`I
`
`I
`
`I
`
`I
`
`0
`
`Figure 3. Regression plot showing the correlation between saliva
`pH and the amount of buprenorphine recovered in saliva and
`mouth rinses (r = -0.40, P = 0.02).
`
`36
`
`J Clin Pharmacol 1997;37:31-37
`
`tissues into the saliva, or some presystemic metabo-
`lism after buccal absorption.
`Results of a previously reported plasma-based bio-
`availability study estimated bioavailability to be ap-
`proximately 55%.*O3"
`In that study the intravenous
`dose was given to patients under general anesthesia.
`Plasma samples were taken for only 3 hours after
`administration, a time when most of our study's par-
`ticipants (whose samples were drawn up to 24 hours
`after dosage) would not have been close to achieving
`the terminal log-linear phase. After the oral doses,
`samples were taken for only 13 hours after adminis-
`tration. Unfortunately, oral doses were always given
`3 hours after the intravenous dose. Inadequate char-
`acterization of the terminal phase after intravenous
`doses and potential carryover from intravenous to
`oral doses (but never vice versa) could account for
`an overestimation of bioavailability.
`Saliva pH was negatively correlated with recovery
`of buprenorphine. Saliva pH may alter the absorp-
`tion of buprenorphine (a weak base with a pKa of
`8.24) by dictating the degree of its ionization, as has
`been observed for other narcotic6.l2 and nonnarcotic
`basic drugs.13 Ensuring a more basic saliva pH during
`sublingual administration might thus be a strategy
`worthy of investigation to increase the extent, and
`potentially to decrease the variability, of absorption.
`The effect of saliva pH on plasma-based indices of
`bioavailability was not examined.
`Absorption of drugs from the sublingual space is a
`function of mucosal biology and drug characteristics.
`Differences in mucosal surface area, buccal perme-
`ability or the degree of epithelial keratinization can
`affect systemic delivery of a drug.14 The low intersub-
`ject variability between 3- and 5-minute absorption
`(ratio 5 min:3 min unextrapolated AUC = 1.11 5
`0.12, see Table I1 and Figure 2) suggests the fourfold
`range in AUCs is probably because of interindividual
`differences in drug uptake or systemic clearance. Pre-
`
`Page 6
`
`

`

`BIOAVAILABILITY OF SUBLINGUAL BUPRENORPHINE
`
`dictable intraindividual absorption, even allowing
`for the small sample size, is suggested by this study.
`Sublingual drug absorption is greatest for small,
`nonpolar, lipophilic m01ecules.'~~'~ Buprenorphine
`is a moderate-sized (molecular weight, 467) lipo-
`philic weak base. Despite buprenorphine's lipophil-
`icity, there was a lag time [tlag) of 1 hour, probably
`related to the molecular size and ionization state of
`buprenorphine. Similar tl,,s for butorphanol (molec-
`ular weight, 327) of approximately 1.7 hrsl' and clon-
`idine (molecular weight, 230) of 2.3 ? 1.9 hrsI7 have
`been reported.
`In conclusion, buprenorphine is approximately
`30% bioavailable from a sublingual solution in etha-
`nol. Sublingual holding times between 3 and 5 min-
`utes are bioequivalent. In opiate-dependence treat-
`ment programs, therefore, sublingual exposure times
`greater than 3 minutes probably offer no advantage.
`
`The authors thank Tina Melby, James Brown, Peter Shwonek,
`Isahella Fernandez, David Naimie, Catherine Klumpp, and Mai
`Chow for assistance in conducting the study; Dr. Emil Lin, Stan-
`ford Jones, Polly Cheung, and Peter Shwonek for carrying out the
`buprenorphine analyses; Susette Welm and Robert Jimison for
`data analysis: and Kaye Welch for editorial assistance.
`
`REFERENCES
`
`1. Ling \V, Rawson RA, Compton MA: Substitution pharmaco-
`therapies for opioid addiction: From methadone to LAAM and
`buprenorphine. J Psychoactive Drugs 1994; 26:119-128.
`2. Johnson RE, Cone EJ, Henningfield JE, Fudala PJ: Use of bu-
`prenorphine in the treatment of opiate addiction. I. Physiologic
`and behavioral effects during a rapid dose induction. Clin Phar-
`rnacol ThPr 1989;46:335-343.
`3. Seow SS, Quigley AJ, Ilett KF, Dusci LJ, Swensen G, Harrison-
`Stewart A, et al: Buprenorphine: a new maintenance opiate? Med
`I Australia 1986:144:407-411.
`4. McQuay HJ, Moore RA: Buprenorphine kinetics in humans. In,
`Cowan A, Lewis JW (eds.): Buprenorphine: Combatting Drug
`
`Abuse with a Unique Opioid. New York: Wiley-Liss, Inc., 1995;
`137-147.
`5. Moore RA: Analysis. In, Cowan A, Lewis JW (eds.): Buprenor-
`phine: Combatting Drug Abuse with a Unique Opioid. New York:
`Wiley-Liss, Inc., 1995; 105-112.
`6. Weinberg DS, Inturrisi CE, Reidenberg B, Moulin DE, Nip TJ,
`Wallenstein S, et al: Sublingual absorption of selected opioid anal-
`gesics. Clin Pharmacol Ther 1988;44:335-342.
`7. Gibaldi M, Perrier D: Pharmacokinetics. Second edition. New
`York: Marcel Dekker, 1982;27-109, 409-416.
`8. Haertzen CA: An Overview of Addiction Research Center In-
`ventory Scales (ARCI): A n Appendix and Manual of Scales. Dept.
`of Health, Education, and Welfare Publication #(ADM)79. Wash-
`ington, DC: Department of Health, Education, and Welfare, 1974.
`9. McNair DM, Lorr M, Doppleman LF: Profile of Mood States.
`San Diego: Educational and Industrial Testing Service, 1971.
`10. Bullingham RES, McQuay HJ, Dwyer D, Allen MC, Moore RA:
`Sublingual buprenorphine used postoperatively: Clinical observa-
`tions and preliminary pharmacokinetic analysis. Br J Clin Pharma-
`C O ~ 1981 ; 12: 11 7 - 122.
`11. Bullingham RES, McQuay HJ, Porter EJB, Allen MC. Moore
`RA: Sublingual buprenorphine used postoperatively: ten-hour
`plasma drug concentration analysis. Br 1 Clin Pharmocol 1982:
`13:665-673.
`12. Kaufman JJ, Semo NM, Koski WS: Microelectrometric titra-
`tion measurement of the pKa's and partition and drug distribution
`coefficients of narcotics and narcotic antagonists and their pH and
`temperature dependence. J Med Chem 1975;18:647-655.
`13. McElnay JC, Al-Furaih TA, Hughes CM, Scott MG, Elborn
`JS, Nicholls DP: The effect of pH on the buccal and sublingual
`absorption of captopril. Eur J Clin Pharmacol 1995:48:373-379.
`14. Harris D, Robinson JR: Drug delivery via the mucous mem-
`branes of the oral cavity. JPharmaceut Sci 1992:81:1-10.
`15. Hardman JG, Gilman AF, Limbird LE: Goodman 6 Gilnian's
`The Pharmacological Basis of Therapeutics. Ninth Edition. New
`York: McGraw-Hill, 1996.
`16. Shyu WC, Mayol RF, Pfeffer M, Pittman KA, Gammans RE,
`Barbhaiya RH: Biopharmaceutical evaluation of transnasal, sub-
`lingual, and buccal disk dosage forms of butorphanol. Biophorma-
`ceut Drug Dispos 1993;14:371-379.
`17. Cunningham FE, Baughman VL, Peters J, Laurito CE: Compar-
`ative pharmacokinetics of oral versus sublingual clonidine. J Clin
`Anesth 1994;6:430-433.
`
`PHARMACOKINETICS
`
`37
`
`Page 7
`
`