`Release Formulation of Mixed Amphetamine Salts:
`Bioavailability of Three Test Formulations and
`Comparison of Fasted, Fed, and Sprinkled Administration
`
`Simon J. Tulloch, M.D., Yuxin Zhang, Ph.D., Angus McLean, Ph.D., and
`Kathleen N. Wolf, B.A.
`
`Study Objectives. To assess the bioavailability of three test formulations of a
`single dose of extended-release Adderall 20-mg capsules compared with
`two doses of immediate-release Adderall 10-mg tablets, and to assess the
`bioequivalence of a single 30-mg dose of the chosen extended-release
`Adderall formulation (designated as SLI381) administered in applesauce
`(sprinkled) and the same dose administered as an intact capsule with or
`without food.
`Design. Randomized, open-label, crossover study.
`Setting. Clinical research unit.
`Patients. Forty-one healthy adults.
`Interventions. Study A had four treatment sequences: three test formulations
`(A, B, and C) of a single dose of extended-release Adderall 20 mg, and two
`10-mg doses of Adderall given 4 hours apart. Study B had three treatment
`sequences: a single dose of SLI381 30 mg as an intact capsule after
`overnight fast, an intact capsule after a high-fat breakfast, and the contents
`of a capsule sprinkled in 1 tablespoon of applesauce.
`Measurements and Main Results. The 20-mg test formulation A had
`comparable pharmacokinetic profiles and bioequivalence in rate and extent
`of drug absorption to Adderall 10 mg twice/day for both d- and l-
`amphetamine. Formulations B and C had statistically significant
`differences from the reference drug in some pharmacokinetic parameters.
`A 30-mg dose of SLI381 showed no significant differences in rate and
`extent of absorption of d- and l-amphetamine for fasted or sprinkled
`conditions compared with the high-fat meal condition.
`Conclusion. SLI381 20 mg/day is bioequivalent to Adderall 10 mg twice/day.
`SLI381 30 mg administered in applesauce is bioequivalent in terms of both
`rate and extent of absorption to the same dose administered as an intact
`capsule in both fasted and fed states.
`(Pharmacotherapy 2002;22(11):1405–1415)
`
`Attention-deficit–hyperactivity disorder
`(ADHD) is a neurobehavioral condition
`characterized by various degrees of develop-
`mentally inappropriate inattention, hyperactivity,
`and impulsivity.1 It is diagnosed most commonly
`in childhood, and prevalence rates vary from
`
`4–12% in school-age children.2
`In addition,
`50–65% of children with ADHD continue to
`display behavioral problems and symptoms into
`their adult lives. The disorder is associated with
`considerable disability, and the negative impact
`can be felt not only in academic and vocational
`
`Amerigen Ex. 1048, p. 1
`
`
`
`1406
`
`PHARMACOTHERAPY Volume 22, Number 11, 2002
`
`settings, but also in social situations and
`recreational activities.3–5 Psychostimulant agents
`(methylphenidate, amphetamine) are well
`tolerated and effective in treating core symptoms
`of ADHD.6–8 However, therapy can be problematic
`because of the need for several daily doses in most
`individuals, which can lead to poor compliance
`and decreased satisfaction with treatment. In-
`school dosing in children may lead to diversion
`of drug, ridicule by peers, and negative impact on
`self-esteem. Thus, a more effective once-daily
`dosage form of stimulant drugs that lasts throughout
`the school day and into the evening is necessary.
`Amphetamine compounds and other psycho-
`stimulants are first-line treatments for ADHD.
`Although the precise mechanism of action is not
`fully elucidated, the agents both accentuate
`release and block reuptake of neurotransmitters
`dopamine and norepinephrine in presynaptic
`neurons.9 The pharmacokinetic and pharmaco-
`dynamic effects of amphetamine are described in
`adults10, 11 and children.12, 13 Amphetamine’s
`absorption is rapid and complete from the gastro-
`intestinal tract, and maximum plasma concen-
`trations are reached in 3–4 hours. The agent
`undergoes hepatic metabolism by side-chain
`deamination and ring hydroxylation; most is
`excreted unchanged in urine. 14 Clinical
`behavioral effects are most apparent during the
`absorption phase and decrease after peak plasma
`concentrations are reached.10–13 Food has little effect
`on plasma amphetamine levels, although gastro-
`intestinal acidifying agents (e.g., ascorbic acid) may
`lower absorption and decrease bioavailability.
`Adderall (Shire US Inc., Florence, KY) is a
`mixture of neutral salts of dextroamphetamine
`sulfate, amphetamine sulfate, the dextro isomer
`of amphetamine saccharate, and d, l-ampheta-
`mine aspartate. For each Adderall tablet, the
`combination of salts and isomers results in a 3:1
`ratio of dextroamphetamine:levoamphetamine.
`
`From the Departments of Clinical Research (Dr. Tulloch
`and Ms. Wolf), Biostatistics (Dr. Zhang), and
`Biopharmaceutics (Dr. McLean), Shire Pharmaceutical
`Development Inc., Rockville, Maryland.
`Supported by Shire Pharmaceutical Development Inc.,
`Rockville, Maryland.
`Presented in part at the National Institutes of Mental
`Health New Clinical Drug Evaluation Unit program,
`Phoenix, Arizona, May 28, 2001.
`Manuscript received October 15, 2001. Accepted
`pending revisions December 24, 2001. Accepted for
`publication in final form August 14, 2002.
`Address reprint requests to David A. Mays, Pharm.D.,
`Shire Pharmaceutical Development Inc., 1901 Research
`Boulevard, Suite 500, Rockville, MD 20850-1801; e-mail:
`dmays@us.shire.com.
`
`The efficacy and tolerability of this product in
`treating children and adults with ADHD were
`proven in clinical trials.15–21
`Historically, the efficacy of Adderall was
`attributed to the chemical composition of d- and
`l-amphetamine salts. Several small studies in the
`1970s explored different effects of d- versus l-
`amphetamine in children with ADHD.22–24 The
`results indicated that both isomers are
`pharmacologically active and efficacious, with
`most children responding well to either isomer.
`However, some children responded only to the d-
`and some only to the l- isomer. No further
`investigations have evaluated the different
`pharmacodynamic activity of the isomers.
`SLI381 (Adderall XR; Shire US Inc.) is a new
`extended-release capsule for treatment of ADHD
`designed to produce a therapeutic effect that lasts
`throughout the day, with one morning dose. The
`capsule contains the same active ingredients as
`immediate-release Adderall and is composed of
`two types of beads combined in a 50:50 ratio.
`Immediate-release beads are designed to release
`drug content in a time course similar to Adderall.
`Delayed-release beads are designed to release
`drug content approximately 4 hours after
`administration. With the delayed-release
`component, the capsule, taken once/day, is
`expected to produce similar pharmacokinetic and
`pharmacodynamic effects to immediate-release
`Adderall taken twice/day.
`We conducted two studies to address several
`issues. The primary objective of the first trial
`(study A) was to assess the bioavailability of a
`single dose of three different test formulations of
`extended-release Adderall 20-mg capsules
`compared with two Adderall 10-mg immediate-
`release tablets administered 4 hours apart to
`determine the optimal formulation to take into
`final development. The purpose of the second
`trial (study B) was to assess whether the contents
`of a single 30-mg dose of the chosen extended-
`release Adderall capsule formulation (SLI381)
`administered in applesauce is bioequivalent to
`the same dose administered as an intact capsule
`with or without food, and to determine the effect,
`if any, on bioavailability of a single dose of a
`SLI381 30-mg capsule administered with a high-
`fat breakfast compared with the same dose
`administered in the fasted state.
`
`Methods
`
`Subjects
`All subjects were screened within 21 days of
`
`Amerigen Ex. 1048, p. 2
`
`
`
`BIOAVAILABILITY OF EXTENDED-RELEASE ADDERALL Tulloch et al
`
`1407
`
`enrollment into either study. Men and women
`between 18 and 55 years of age with no clinically
`significant abnormal findings on physical
`examination, medical history, and clinical
`laboratory tests during screening were enrolled.
`Body weight was not to be more than 10% above
`or below ideal weight for height and estimated
`frame adapted from 1983 Metropolitan Life
`Insurance tables.
`Major exclusion criteria were treatment with
`any known cytochrome P450 enzyme–altering
`agents (e.g., barbiturates, phenothiazines,
`cimetidine) within 30 days before or during the
`study; use of any prescription drug within 14
`days before or during the study (hormonal
`contraceptive and hormone replacement therapy
`for women were allowed); use of any over-the-
`counter agent within 7 days before or during the
`study; pregnancy or lactation; positive urine
`screen for alcohol or drugs of abuse; history of
`allergic or adverse response to amphetamine or
`any related drug; history of drug or alcohol
`abuse; history of clinically significant gastro-
`intestinal tract, renal, hepatic, neurologic,
`hematologic, endocrine, oncologic, pulmonary,
`immunologic, psychiatric, or cardiovascular
`disease; and any other condition that, in the
`opinion of the investigator, would jeopardize the
`safety of the subject or affect the validity of study
`results.
`Subjects were restricted from food or beverages
`containing alcohol, caffeine, or any xanthine-
`containing product 48 hours before and during
`each period of confinement, fruit juices
`(including grapefruit juice) containing ascorbic
`acid during confinement, strenuous exercise
`during confinement, and lying down for the first
`4 hours after drug administration to ensure
`proper stomach emptying.
`All subjects gave written informed consent,
`and the studies were approved by the institu-
`tional review board of MDS Harris, Lincoln,
`Nebraska. All study drugs were supplied by
`Shire Pharmaceutical Development Inc.
`
`Determination of Sample Size
`
`Findings from previous studies of Adderall and
`SLI381 delayed-release pellets25 indicated that the
`estimate of area under the curve (AUC)
`test:reference ratio was within 0.90–1.10 for d-
`amphetamine, and the estimated within-
`subject–between-formulation (log scale) was
`less than 0.10. Given that the true AUC mean for
`a test formulation is within the 90% region of the
`
`reference, for a sample of 16 subjects, the
`proposed crossover design would have at least
`80% power to reject the null hypothesis of
`bioinequivalence at the 0.05 level. Based on the
`assumptions, we planned to enroll 20 and 21
`subjects, without replacement, in studies A and
`B, respectively.
`
`Study A Design
`The prototype formulation assessment was a
`four-way, open-label, crossover design in 20
`healthy subjects with 5 subjects/sequence. A
`standard 4 x 4 Latin square was used to assign
`subjects to treatments. In each sequence,
`subjects were given a single 20-mg dose of one of
`the test products (extended-release Adderall
`formulation A, B, or C) or two 10-mg doses of
`the reference drug (Adderall) administered 4
`hours apart . Subjects received the other dosing
`conditions in subsequent study periods according
`to the randomization scheme. A 7-day washout
`period separated each treatment.
`A 20-mg dose was selected to enable
`quantification of anticipated blood levels of d-
`and l-amphetamine over the 48-hour time period
`analyzed. Experience suggests this dose is often
`used in clinical practice and would be well
`tolerated by healthy subjects.
`
`Drug Administration
`Subjects were admitted to the clinic in the
`evening, approximately 12 hours before the
`scheduled dose. At each treatment period check-
`in, they completed a brief written questionnaire
`to affirm that exclusion criteria and restrictions
`had not been violated since the screening or
`previous confinement period. In addition, a
`urine sample was collected to test for alcohol and
`drugs of abuse, and a blood sample was collected
`from women for a serum pregnancy test.
`Subjects remained at the clinic until completion
`of the 24-hour postdose blood collection and
`returned to the clinic for 36- and 48-hour
`postdose specimen collections. After check-in,
`each subject received an evening snack. On the
`next day, they consumed a standard high-fat
`breakfast approximately 20 minutes before drug
`administration. The breakfast consisted of one
`English muffin with butter, one fried egg, one
`slice of American cheese, one slice of Canadian
`bacon, one 2-oz serving of hash-brown potatoes,
`and 8 fluid oz whole milk. Water was allowed ad
`libitum during the study, except for 1 hour before
`and 2 hours after dosing.
`
`Amerigen Ex. 1048, p. 3
`
`
`
`1408
`
`PHARMACOTHERAPY Volume 22, Number 11, 2002
`
`The study drug (a single dose of extended-
`release Adderall 20 mg as test capsule
`formulations A, B, or C, or a single dose of
`Adderall 10 mg) was administered within 5
`minutes of meal completion with 8 fluid oz
`room-temperature tap water. A mouth check was
`performed after dosing to ensure that the capsule
`was swallowed. A second single dose of Adderall
`10 mg was administered 4 hours later, during
`lunch, for subjects who were assigned to Adderall
`treatment. A standard meal schedule was begun
`with lunch, dinner, and an evening snack. The
`same menu and meal schedule were administered
`uniformly for all subjects and for all treatment
`periods.
`
`Blood Collection
`Beginning on each dosing day, 17 blood
`samples (7 ml/sample) were collected through
`the 48-hour postdose interval during each study
`period to determine plasma concentrations of d-
`and l-amphetamine. Samples were collected by
`venipuncture into tubes containing ethylene-
`diaminetetraacetic acid and stored on ice before
`plasma was separated by centrifugation
`(approximately 2500 rpm x 15 min at 4°C).
`Plasma samples were frozen and stored at -20°C
`until assayed. Blood was collected 5 minutes
`before dosing and at 1, 2, 3, 4, 5, 6, 7, 8, 9, 10,
`12, 14, 16, 24, 36, and 48 hours. In addition, 15
`ml of blood was collected for the screening
`clinical laboratory evaluation. For women,
`another 5 ml/check-in period was collected to
`test for pregnancy.
`
`Safety Evaluations
`Adverse event data were obtained by
`observation and by unsolicited reporting before,
`during, and after each dosing and collection
`phase. Blood pressure and pulse were measured
`at screening and four other times (immediately
`before the dose, and 2, 4, and 24 hrs after the
`dose) on the dosing day of each treatment period.
`
`Analytical Methods
`Plasma samples were analyzed by validated
`procedures. 26 Amphetamine isomers and
`deuterated analogs as internal standards were
`extracted from plasma under alkaline conditions
`into organic solvent. Analytes were back-
`extracted into acid, made alkaline again,
`derivatized with benzoyl chloride, and
`reextracted into organic solvent. After aqueous
`wash to remove excess reagent, the organic
`
`extract was evaporated to dryness and
`reconstituted in mobile phase. Analysis was
`performed by chiral high-performance liquid
`chromatography with turbo-ionspray tandem
`mass spectrometry detection. A weighted [(1/x)
`where x = concentration of the compound] linear
`regression was used to determine slopes,
`intercepts, and correlation coefficients for d- and
`l-amphetamine concentrations in study samples
`and internal standards. For d- and l-amphetamine,
`concentrations were linear over 0.5–50 ng/ml
`with a limit of quantification of 0.5 ng/ml.
`Coefficients of variation were less than or equal
`to 5.28% for l- and 4.71% for d-amphetamine.
`
`Pharmacokinetic Analysis
`
`Pharmacokinetic parameters were determined
`for bioavailability and bioequivalence evaluations
`for each type of dosing for d- and l-amphetamine
`by standard noncompartmental methods. The
`primary pharmacokinetic parameters were area
`under the drug concentration–time curve from
`time zero to t hour (AUC0–t), with t the last time
`point over the time interval with a measurable
`drug concentration; area under the drug
`concentration–time curve from time zero to
`infinity (AUC0–∞); elimination half-life;
`maximum observed drug concentration (Cmax);
`and time to Cmax (Tmax). For both isomers,
`AUC0–t was calculated by the linear trapezoidal
`rule. The residual AUC between the last time
`point measured and infinity (AUC t–∞) was
`determined and added to AUC 0–t to obtain
`AUC0–∞. The AUCt–∞ = Ct/ke, where Ct was the
`last measurable plasma concentration and ke was
`the terminal elimination rate constant determined
`by linear regression of the terminal log linear
`phase of the plasma drug concentration-time
`curve. The half-life for each isomer equalled
`0.693/ke.
`
`Statistical Analysis
`
`Descriptive statistics (N, mean, SD) of d- and l-
`amphetamine were obtained for all pharmaco-
`kinetic parameters based on the intent-to-treat
`population. Standard analysis of variance
`(ANOVA) model of a 4-way crossover design
`with a general linear approach was applied to
`AUC, Cmax, and Tmax to determine differences
`among the formulations. The model included
`sequence, patient-within-sequence, period, and
`formulation. The sequence effect was tested
`using the patient-within-sequence effect, and all
`other effects were tested using the residual error
`
`Amerigen Ex. 1048, p. 4
`
`
`
`BIOAVAILABILITY OF EXTENDED-RELEASE ADDERALL Tulloch et al
`
`1409
`
`of the model. For each parameter, mean values
`of the formulations were compared with the
`reference formulation (Adderall 10 mg twice/day)
`using Dunnett’s test with the type I error rate of
`0.05. The AUC and Cmax were further analyzed
`on a log scale to assess bioequivalence of each
`test formulation and the reference formulation.
`The recommended two one-sided t
`test
`hypotheses for average bioequivalence were
`tested at the 0.05 level by constructing the 90%
`confidence interval (CI) of the ratio of the
`test:reference means.27 Adverse events, blood
`pressure, and pulse were tabulated descriptively
`and compared with a paired t test.
`
`Study B Design
`This trial had a three-way, open-label,
`crossover design. Twenty-one healthy men and
`women were randomized to one of the three dose
`administration sequences with seven subjects/
`sequence. They were given a single 30-mg dose
`of the chosen extended-release Adderall
`formulation (test formulation A from study A)—
`designated as SLI381—under one of three drug
`dosing conditions: an intact 30-mg capsule after
`an overnight fast, an intact 30-mg capsule after a
`high-fat breakfast, or the contents of a 30-mg
`capsule sprinkled in 1 tablespoon of applesauce.
`Subjects received alternate dosing conditions in
`subsequent periods according to the randomization
`scheme. A 7-day washout period separated each
`treatment.
`A 30-mg dose was selected to enable
`quantification of anticipated blood levels of d-
`and l-amphetamine over the 60-hour period. It
`also was the highest strength marketed for
`immediate-release Adderall tablets. Experience
`suggested this dose would be well tolerated by
`healthy subjects.
`
`Drug Administration
`Subjects were admitted to the clinic in the
`evening, at least 10 hours before the scheduled
`dose and followed a protocol identical to that for
`study A.
`On the next day, subjects in the high-fat fed
`condition received a standard high-fat breakfast
`approximately 30 minutes before drug
`administration as described for study A and
`completed the meal 5 minutes before dosing. For
`this condition and the fasted condition, the study
`drug, a single intact capsule of SLI381 30 mg,
`was administered with 8 fluid oz room-
`temperature tap water. For subjects receiving the
`
`study drug sprinkled on applesauce, a single
`capsule of SLI381 30 mg was opened and
`sprinkled into 1 tablespoon of applesauce. A
`mouth check was performed after dosing to
`ensure that the dose was swallowed. After
`administration, subjects were required to fast for
`4 hours. Water was allowed ad libitum during
`the study, except for 1 hour before and 2 hours
`after dosing. A standard meal schedule was
`begun with lunch, dinner, and evening snack.
`The same menu and meal schedule were
`administered uniformly for all subjects and for all
`treatment periods. Subjects were monitored for
`adverse events for the entire study as described
`for study A.
`
`Blood Collection
`Samples (7 ml) of venous blood were collected
`and processed as in study A, with the addition of
`11-hour and 60-hour postdose samples. Analytic
`methods and pharmacokinetic parameters were
`determined as described for study A.
`
`Statistical Analysis
`Descriptive statistics of d- and l-amphetamine
`were obtained for all pharmacokinetic parameters
`based on the intent-to-treat population. An
`ANOVA model of a 3-way crossover design with
`a general linear approach was applied to AUC,
`Cmax, and Tmax to determine differences among
`the conditions. The model included sequence,
`patient-within-sequence, period, and condition.
`The sequence effect was tested using the patient-
`within-sequence effect, and all other effects were
`tested using the residual error of the model. For
`each parameter, mean values of fasted and
`sprinkled conditions were compared with the fed
`condition using Dunnett’s test with the type I
`error rate of 0.05. The AUC and Cmax were
`further analyzed on a log scale to assess
`bioequivalence between each pair of dosing
`conditions.27 Adverse events, blood pressure,
`and pulse were tabulated descriptively and
`compared by a paired t test.
`
`Results
`
`Study A
`Twenty subjects (mean age 40.4 yrs) were
`enrolled and randomized to treatment (Table 1).
`Nineteen subjects completed the study; one was
`withdrawn before dosing in the fourth dosing
`period as a result of a positive drug test for
`opiates at check-in. This subject received the
`
`Amerigen Ex. 1048, p. 5
`
`
`
`1410
`
`PHARMACOTHERAPY Volume 22, Number 11, 2002
`
`Table 1. Subject Demographics and Baseline Characteristics
`Study A
`(n=20)
`13/7 (65/35)
`40.4 (23–55)
`
`Study B
`(n=21)
`11/10 (52/48)
`35 (20–53)
`
`Variable
`M/F, no. (%)
`Age, yrs, mean (range)
`Race, no. (%)
`Caucasian
`Black
`Asian
`Hispanic
`Native American
`Height, in., mean (range)
`Weight, lbs, mean (range)
`
`17 (85)
`2 (10)
`1 (5)
`0
`0
`69.8 (63.5–76.0)
`167.8 (122.0–226.0)
`
`17 (81)
`2 (9)
`0
`1 (5)
`1 (5)
`67.9 (60.0–74.0)
`162.0 (110.0–203.0)
`
`Table 2. Pharmacokinetic Parameters for d- and l-Amphetamine (Study A)
`Pharmacokinetic Parameter
`AUC0–t
`Cmax
`(ng(cid:129)hr/ml)
`(ng/ml)
`
`AUC0–∞
`(ng(cid:129)hr/ml)
`
`Tmax
`(hrs)
`
`Half-life
`(hrs)
`
`Formulation
`d-Amphetamine
`Test A 20 mg q.d. (SLI381)
`Test B 20 mg q.d.
`Test C 20 mg q.d.
`Adderall 10 mg b.i.d. (reference)
`l-Amphetamine
`Test A 20 mg q.d. (SLI381)
`Test B 20 mg q.d.
`Test C 20 mg q.d.
`Adderall 10 mg b.i.d. (reference)
`Data are mean ± SD.
`AUC0–∞ = area under the drug concentration-time curve from time zero to infinity; AUC0–t = AUC from time zero to t hour; Cmax = maximum
`observed drug concentration; Tmax = time to Cmax.
`ap<0.05 compared with reference by Dunnett’s test.
`
`566.62 ± 114.30
`473.83a ± 114.46
`546.76 ± 126.02
`529.92 ± 114.44
`
`522.47 ± 100.72
`426.24a ± 95.37
`496.68 ± 101.75
`494.63 ± 103.10
`
`28.13 ± 8.84
`18.51a ± 4.76
`22.86a ± 5.85
`28.33 ± 7.13
`
`6.95 ± 2.35
`5.60 ± 2.56
`9.37a ± 3.02
`6.90 ± 1.25
`
`11.83 ± 2.74
`13.87a ± 3.29
`12.21 ± 2.97
`10.90 ± 2.04
`
`203.12 ± 46.04
`169.34a ± 46.56
`197.47 ± 49.85
`202.67 ± 49.05
`
`178.28 ± 40.44
`144.70a ± 36.41
`168.97 ± 39.86
`180.83 ± 41.96
`
`8.67 ± 2.80
`5.75a ± 1.56
`7.16a ± 1.95
`9.25 ± 2.41
`
`8.15 ± 4.44
`5.70 ± 2.62
`9.74a ± 3.21
`7.10 ± 1.37
`
`13.72 ± 2.83
`15.78a ± 3.61
`14.70 ± 3.63
`13.19 ± 2.69
`
`assigned treatments of test formulations A, B, and
`Adderall in periods 1, 2, and 3, respectively. All
`information collected from this subject was
`
`Figure 1. Mean plasma d- and l-amphetamine
`concentration versus time profiles for single 20-mg doses of
`test formulations A, B, and C, and immediate-release
`Adderall 10 mg twice/day with a 4-hour interval.
`
`included in the analyses.
`
`Pharmacokinetic Parameters
`Mean plasma concentrations versus time
`profiles of d- and l-amphetamine after drug
`administration are shown in Figure 1. Table 2
`gives descriptive statistics of pharmacokinetic
`parameters for each formulation. The ANOVA
`results of the 4-way crossover indicate
`statistically significant differences among the four
`formulations in AUC0–∞, AUC0–t, Cmax, and Tmax.
`In reference to Adderall, multiple means
`comparisons by Dunnett’s test disclosed
`statistically significant (p<0.05) differences in
`AUC0–∞, AUC0–t, and Cmax for test formulation B;
`statistically significant (p<0.05) differences in
`Cmax and Tmax for test formulation C; and no
`statistically significant differences in these
`parameters for test formulation A. These
`observations held for both isomers.
`Table 3 shows bioequivalence results on
`logarithmic transformations of pharmacokinetic
`
`Amerigen Ex. 1048, p. 6
`
`
`
`BIOAVAILABILITY OF EXTENDED-RELEASE ADDERALL Tulloch et al
`
`1411
`
`Table 3. Bioequivalence of Pharmacokinetic Parameters (Study A)
`Test:Reference Ratioa (90% confidence interval)
`AUC0–∞
`AUC0–t
`Cmax
`
`Formulation
`d-Amphetamine
`Test A (SLI381)
`Test B
`Test C
`l-Amphetamine
`0.92 (0.86–0.98)b
`0.99 (0.94–1.04)b
`1.01 (0.95–1.07)b
`Test A (SLI381)
`0.62 (0.58–0.66)
`0.80 (0.76–0.84)
`0.84 (0.79–0.89)
`Test B
`0.95 (0.90–1.00)b
`1.00 (0.94–1.06)b
`Test C
`0.79 (0.74–0.84)
`AUC0–∞ = area under the drug concentration-time curve from time zero to infinity; AUC0–t = AUC from time
`zero to t hour; Cmax = maximum observed drug concentration.
`aImmediate-release Adderall administered twice/day.
`bThe 90% confidence interval falls within the 0.80–1.25 limits of bioequivalence when analyzed on a
`logarithmic scale.
`
`1.07 (1.02–1.13)b
`0.90 (0.86–0.95)b
`1.06 (1.01–1.11)b
`
`1.06 (1.01–1.11)b
`0.86 (0.82–0.90)b
`1.02 (0.97–1.06)b
`
`0.97 (0.92–1.04)b
`0.65 (0.61–0.70)
`0.82 (0.77–0.88)
`
`data. The 90% CIs of the test:reference ratio fell
`within the 0.80–1.25 limits of average
`bioequivalence for AUC0–∞, AUC0–t, and Cmax for
`test formulation A for d- and l-amphetamine.
`Test formulation B did not fall within the limits
`on any parameter for l-amphetamine, and for d-
`amphetamine, fell outside the limits for Cmax.
`For d- and l-amphetamine levels, test
`formulation C did not fall within the limits for
`C max. Thus, a single 20-mg dose of test
`formulation A was bioequivalent to Adderall 10
`mg twice/day for the two isomers in terms of rate
`(Cmax) and extent (AUC) of absorption.
`
`Safety
`
`All formulations were well tolerated. Eight
`subjects reported a total of 10 adverse events
`after starting study drugs: headache (4), insomnia
`(2), pharyngitis (1), rash (1), somnolence (1),
`and abnormal vision (1). All events were mild
`and resolved. Four of the 10 events were
`attributed as related or possibly related to study
`drug: headache (1), insomnia (2), and abnormal
`vision (1). The event rate was similar among the
`four dosing conditions. No subjects withdrew as
`a result of adverse events and no deaths or other
`serious events occurred during the study.
`Compared with baseline, consistent increases
`in pulse 2–24 hours after the dose and slight
`increases in blood pressures 2–4 hours after dose
`were seen for all treatment conditions. None of
`these changes was deemed by investigators to be
`clinically significant.
`
`Study B
`
`Twenty-one subjects (mean age 35 yrs) were
`
`enrolled (Table 1). They all received one oral 30-
`mg dose of test formulation A (SLI381) as an
`intact capsule in the fed state. Twenty subjects
`received one 30-mg dose as an intact capsule in
`the fasting state and received the contents of one
`30-mg capsule sprinkled over applesauce. One
`subject after receiving a single dose withdrew as a
`result of necessary drug treatment for gout
`(colchicine, indomethacin). All information
`collected from this subject was included in the
`analyses.
`
`Pharmacokinetic Parameters
`Mean plasma levels of d- and l-amphetamine
`are shown in Table 4 and Figure 2. The ANOVA
`results of the 3-way crossover indicate no
`statistically significant differences in AUC0–∞,
`AUC0–t, or half-life for l- or d-amphetamine in
`the fasted and sprinkled conditions compared
`with the fed condition. However, quantitatively
`small but statistically significant differences were
`noted for Tmax and Cmax.
`The results of bioequivalence in the fasted and
`sprinkled conditions compared with the fed
`condition using logarithmic transformations of
`pharmacokinetic data (Table 5) indicate that the
`90% CIs of the test:reference ratio fell within the
`0.80–1.25 limits of average bioequivalence for all
`three pharmacokinetic parameters (AUC0–∞,
`AUC0–t, Cmax) for both d- and l-amphetamine.
`Also, 90% CIs of the test:reference ratio for the
`sprinkled versus fasted condition fell within
`these limits. Thus, according to criteria of
`average bioequivalence, the extent and rate of
`drug absorption for a single 30-mg dose of
`SLI381 were bioequivalent under the three
`dosing conditions. The Tmax was approximately
`
`Amerigen Ex. 1048, p. 7
`
`
`
`Tmax
`(hrs)
`
`Half-life
`(hrs)
`
`Test Condition
`d-Amphetamine
`Fasted
`Sprinkled
`Fed
`l-Amphetamine
`Fasted
`Sprinkled
`Fed
`Data are mean ± SD.
`AUC0–∞ = area under the drug concentration-time curve from time zero to infinity; AUC0–t = AUC from time zero to t hour; Cmax =
`maximum observed drug concentration; Tmax = time to Cmax.
`ap<0.05 compared with fed condition by Dunnett’s test.
`
`1412
`
`PHARMACOTHERAPY Volume 22, Number 11, 2002
`
`Table 4. Mean Pharmacokinetics for d- and l-Amphetamine After Administration of SLI381 30 mg (Study B)
`Pharmacokinetic Parameter
`Cmax
`(ng/ml)
`
`AUC0–∞
`(ng(cid:129)hr/ml)
`
`AUC0–t
`(ng(cid:129)hr/ml)
`
`851.17 ± 213.51
`855.98 ± 179.68
`822.56 ± 200.18
`
`827.99 ± 201.96
`834.49 ± 175.14
`799.28 ± 190.50
`
`44.33a ± 11.10
`43.51a ± 9.61
`39.70 ± 8.84
`
`288.59 ± 79.17
`290.38 ± 64.49
`273.56 ± 68.98
`
`271.72 ± 72.23
`274.65 ± 61.30
`258.31 ± 64.36
`
`13.32a ± 3.66
`13.04a ± 3.20
`11.98 ± 2.89
`
`5.20a ± 1.96
`5.50a ± 1.76
`7.67 ± 2.31
`
`5.55a ± 2.09
`5.60a ± 1.73
`8.33 ± 2.89
`
`10.40 ± 2.31
`10.39 ± 2.05
`10.34 ± 1.98
`
`12.71 ± 3.30
`12.73 ± 2.83
`12.50 ± 2.56
`
`Measures
`
`Test Condition
`
`Fasted
`
`Sprinkled
`
`Ratio of test:fed condition (90% CI)
`Ratio of test:fed condition (90% CI)
`Ratio of test:fed condition (90% CI)
`Ratio of test:fasted condition (90% CI)
`Ratio of test:fasted condition (90% CI)
`Ratio of test:fasted condition (90% CI)
`
`1.04 (0.98–1.10)a
`1.04 (0.99–1.10)a
`1.12 (1.05–1.18)a
`—
`—
`—
`
`1.05 (0.99–1.11)a
`1.05 (1.00–1.11)a
`1.10 (1.04–1.16)a
`1.01 (0.96–1.07)a
`1.01 (0.96–1.07)a
`0.99 (0.93–1.04)a
`
`Table 5. Bioequivalence of Pharmacokinetics for d- and l-Amphetamine (Study B)
`Pharmacokinetic
`Parameter
`d-Amphetamine
`AUC0–∞ (ng(cid:129)hr/ml)
`AUC0–t (ng(cid:129)hr/ml)
`Cmax (ng/ml)
`AUC0–∞ (ng(cid:129)hr/ml)
`AUC0–t (ng(cid:129)hr/ml)
`Cmax (ng/ml)
`l-Amphetamine
`1.07 (1.00–1.14)a
`1.05 (0.99–1.13)a
`AUC0–∞ (ng(cid:129)hr/ml)
`Ratio of test:fed condition (90% CI)
`1.07 (1.01–1.14)a
`1.05 (0.99–1.12)a
`Ratio of test:fed condition (90% CI)
`AUC0–t (ng(cid:129)hr/ml)
`1.09 (1.03–1.16)a
`1.11 (1.05–1.18)a
`Ratio of test:fed condition (90% CI)
`Cmax (ng/ml)
`1.01 (0.95–1.09)a
`—
`Ratio of test:fasted condition (90% CI)
`AUC0–∞ (ng(cid:129)hr/ml)
`1.02 (0.96–1.08)a
`—
`Ratio of test:fasted condition (90% CI)
`AUC0–t (ng(cid:129)hr/ml)
`0.98 (0.93–1.04)a
`—
`Ratio of test:fasted condition (90% CI)
`Cmax (ng/ml)
`aThe 90% confidence interval (CI) fell within the 0.80–1.25 limits of bioequivalence when analyzed on a logarithmic scale.
`
`2 hours longer for the d-isomer and 3 hours
`longer for the l-isomer in the presence of a high-
`
`Figure 2. Mean plasma d- and l-amphetamine
`concentration versus time profiles for a single 30-mg dose of
`SLI381 administered under three dosing conditions: fasted,
`after a high-fat meal, and sprinkled in applesauce.
`
`fat meal.
`
`Safety
`All formulations were well tolerated. Eleven
`subjects reported a total of 54 adverse events
`after the start of dosing, with one subject
`reporting 18. The events were mild (51) or
`moderate (3) in severity and resolved or
`improved. Of the 54 events, 25 were unrelated to
`study drug and 29 were assessed as related or
`possibly related to study drug. Most frequently
`reported were insomnia (7), headache (6),
`nausea (5), and dizziness (4). The adverse event
`rate was similar among the three dosing
`conditions (fasted 14, fed 22, sprinkled 18). No
`subjects were withdrawn as a result of adverse
`events, and no deaths or other serious events
`occurred during the study.
`Compared with baseline, a significant increase
`
`Amerigen Ex. 1048, p. 8
`
`
`
`BIOAVAILABILITY OF EXTENDED-RELEASE ADDERALL Tulloch et al
`
`1413
`
`in pulse (p<0.01) 24 hours after dosing was
`noted for all three dosing conditions (average
`change 10–11 beats/min). A significant increase
`(p<0.01) in systolic blood pressure was noted 2
`and 4 hours after the dose for the sprinkled
`condition only. By 24 hours after dosing, mean
`systolic blood pressures returned to baseline. No
`significant increases in diastolic blood pressure
`were noted. No changes in blood pressure or
`pulse were considered clinically significant by
`investigators.
`
`Discussion
`
`Other pharmacokinetic studies of d-
`amphetamine in adults reported time to