B.

15 Absolute and Relative Bioavailability

Roland Wesch

PURPOSE AND RATIONALE For absolute BA studies, the AUC after intravascular
The assessment of a drug’s bioavailability (BA) is the most (IV or intra-arterial) administration is the reference and is
important information on its pharmacokinetics. Conse- set to 100% availability. Factors that reduce the availability
quently, numerous guidelines primarily focus on this issue of a drug prior to entering the systemic circulation may
as from the exposure efficacy as well as safety for the include poor absorption from the gastrointestinal tract
patient (Study Design et al. 2003; ICH E4: Dose–Response (Zhou 2003), an (entero-)hepatic recirculation (Ezzet
Information to Support Drug Registration March 1994; et al. 2001; Bergman et al. 2006), or a fast degradation
EU CPMP: Note for Guidance on Modified Release Oral prior to reaching the central compartment, the first pass
and Transdermal Dosage Forms: Section II (Pharmacoki- effect or first pass metabolization (Kharasch et al. 2005;
netic and Clinical Evaluation) July 1999; EU CPMP: Note Zahng and Benet 2001; Chan et al. 2004; Thummel and
for Guidance on the Investigation of Bioavailability and Wilkinson 1998).
Bioequivalence July 2001; US FDA Guidance for Industry: Bioavailability is defined for a formulation, not for
Food-Effect Bioavailability and Fed Bioequivalence Stud- a drug.
ies December 2002; US FDA Guidance for Industry: Bioavailability studies quantify rate and extent of
Bioavailability and Bioequivalence Studies for Orally absorption. They compare the efficiency of the disposition
Administered Drug Products – General Considerations of several drug formulations, for example, immediate-
March 2003; US FDA Guidance for Industry: Bioavailabil- release vs. modified-release solid formulation or capsule
ity and Bioequivalence Studies for Nasal Aerosols and vs. tablet or tablet A vs. tablet B, etc. or they compare
Nasal Sprays for Local Action April 2003; EU CPMP: the disposition of different routes of administration, for
Points to Consider on the Clinical Requirements of Mod- example, PO vs. SC or PO vs. IV. According to the defini-
ified Release Products to be Submitted as a Line Extension tion, a comparison to the intravenous bolus injection
of an Existing Marketing Authorization June 2003). yields the ‘‘absolute’’ bioavailability.
Bioavailability is defined as the rate and extent by Bioavailability figures should always be given for the
which the active moiety becomes available at the site of active moiety of a drug. If the parent drug is pharmacody-
action. Because neither concentrations nor amounts can namically inactive, details on relevant active metabolite(s)
generally be determined at the site of action, plasma/ should be given.
serum concentrations are used as a surrogate to determine The criterion of bioequivalence applies if there is
the rate and extent of bioavailability. Provided that the a similarity in bioavailability (statistically proven) that is
pharmacokinetics of the drug considered is linear and unlikely to result in clinically relevant differences in effi-
time invariant, the area under the curve (AUC) is cacy and/or safety.
a measure for the fraction of the dose available according The bioavailability of a drug formulation is best
to Dost’s law of corresponding areas. Absolute bioavail- described by the rate (Cmax/tmax) and the extent (area
ability is deduced from the comparison of an extravascular under the plasma concentration-time curve AUC).
and an intravascular administration, i.e., AUCPO/AUCIV. Details on the design of and the interpretation of data
Relative bioavailability compares the exposure following from bioavailability studies are given in guidelines and
two different extravascular application forms, i.e., AUCIM/ guidances from ICH, FDA, or CPMP.
AUCSC or AUCPOtest/AUCPOreference. Extravascular routes
of administration that require documentation of bioavail- PROCEDURE
ability and/or bioequivalence include the oral (PO), intra- The design for an absolute bioavailability study is presented
muscular (IM), or subcutaneous (SC) routes, and – in in > Example 1. The drug in question undergoes in-
most instances – vaginal, dermal, ocular, topic, rectal, tensive Phase II metabolism, leading to numerous conju-
nasal, or pulmonary administration. gates, the cysteine conjugate being predominant. For the

H. G. Vogel, J. Maas, A. Gebauer (eds.), Drug Discovery and Evaluation: Methods in Clinical Pharmacology, DOI 10.1007/978-3-540-89891-7_B.15,
# Springer-Verlag Berlin Heidelberg 2011
174 B.15 Absolute and Relative Bioavailability

purposes of simplicity, the description is limited to the B.15.1.1.4 Treatments

collection, handling, and interpretation of pharmacokinetic
data although safety parameters were also in the focus. Regimen A (Reference Treatment):
The design for a relative bioavailability study is presented Intravenous (IV) administration of XYZ1234 (10 mg,
in > Example 2. The drug in question had exhibited a administered over 30 min)
marked positive food effect when being administered as Regimen B (Test Treatment):
film tablet. The purpose of this study was to compare Oral (PO) administration of XYZ1234 (25 mg, as a
a newly developed capsule to a pilot capsule formulation capsule formulation)
and to include an oral solution as the reference.

B.15.1.1.5 Pharmacokinetic Data

B.15.1 Example 1
Concentrations of unconjugated XYZ1234 and Cystein
B.15.1.1 Protocol Outline (CYS)-conjugated XYZ1234 in plasma were measured
pre-dose and at predetermined times up to 48 h post-dose.
A phase I, open-label, randomized, crossover study to The primary analysis examined pharmacokinetic
investigate the bioavailability, safety, tolerability, and parameters calculated from plasma concentrations of
pharmacodynamics following single oral administration CYS-conjugated XYZ1234 using non-compartmental
of 25 mg XYZ1234 as capsule and single intravenous techniques. The secondary analysis examined the pharma-
administration of 10 mg XYZ1234 in healthy men. cokinetic parameters of unconjugated XYZ1234.

EVALUATION
The primary analyses consisted of characterizing the bio-
B.15.1.1.1 Primary Objective availability of oral XYZ1234 using intravenous XYZ1234 as
the reference. Determination of bioavailability was based on
To characterize the bioavailability of XYZ1234 drug sub- the plasma concentrations of CYS-conjugated XYZ1234.
stance (25 mg) as a capsule formulation following a single Descriptive statistics and formal statistical analysis were
oral administration in fasting conditions in healthy male used to summarize and analyze the pharmacokinetic
adult volunteers, using 10 mg of intravenously adminis- parameters of unconjugated XYZ1234 and CYS-conjugated
tered XYZ1234 as the reference formulation. XYZ1234 in all evaluable subjects.
The secondary analyses consisted of assessing the
safety, tolerability, and pharmacodynamic responses after
B.15.1.1.2 Study Design administration of XYZ1234 and XYZ1234 in plasma and
urine using descriptive statistics.
This was an open-label, single-dose, randomized, two-
period crossover study with a minimum washout period CRITICAL ASSESSMENT OF THE METHOD
of 7 days. Each treatment group received treatment A For the oral route of administration, the dose of 25 mg was
(10 mg XYZ1234, intravenously administered) and treat- selected according to the experience from the first-in-man
ment B (25 mg XYZ1234 as capsule, orally administered), study, where this dose was safe and well tolerated and was
once each under fasting conditions. at the higher end of the dose-proportional range. The dose
for the intravenous route of administration was adjusted
according to the results from animal bioavailability studies
B.15.1.1.3 Inclusion Criteria where the absolute bioavailability was in the range of 50%.
Bioavailability – or, in a stricter sense, bioequivalence
Healthy male subjects, aged 18–45 years (inclusive), with – studies are usually conducted in healthy subjects.
a Body Mass Index between 18 and 27 kg/m2 (inclusive), Although the inclusion of women is now being encour-
normal or clinically irrelevant abnormal findings (in the aged, we enrolled only men. This study was the second
opinion of the investigator) in the medical history and clinical trial in the project.
physical examination, laboratory values, ECG, blood pres- As the bioequivalence rules are clearly defined, the
sure and pulse rate, negative serology (HIV antibody, study population must ensure a high level of standardiza-
hepatitis B surface antigen, hepatitis C antibody), and tion, making it sometimes difficult to extrapolate to
urine screen for drugs of abuse. patient settings. It must be ensured that inter-occasion
 Absolute and Relative Bioavailability B.15 175

variability is limited to the formulations used. Typical B.15.1.2 Example 1

enrolment criteria are:
To illustrate the type of data that can be obtained using the
● Non-smoking subjects between 18 and 45 years
discussed study, a high level summary of the pharmacoki-
● Normal for weight and BMI
netic results obtained from the study described above
● (Clinically) healthy
under ‘‘PROCEDURE’’ is presented below. Due to the
● Not using any medication
anticipated mode of action of the drug (blood pressure
● Massive dietary and general restrictions (‘‘life style’’)
lowering) in this example, instead of an intravenous bolus
● No hypersensitivities
injection an intravenous infusion over 30 min was chosen.
● No recent history or presence of any condition that
might interfere with the absorption, distribution,
metabolism, or elimination of the drug under B.15.1.2.1 Results – Pharmacokinetics
investigation
In the context with the last criterion, it is important to The calculated bioavailability on the basis of the AUClast of
document any adverse event during the study, especially conjugated XYZ1234 was 38%. However, this could be
close to the PK profiling day(s) that might affect the a slight underestimation of the bioavailability since this
disposition of an orally administered drug under investi- AUClast could only be determined until 6 h post-dose.
gation, for example, nausea (delaying the gastric emptying A calculation of the bioavailability on the basis of the
time, increasing the intestinal residence times), vomiting AUClast or AUC0–inf of unconjugated XYZ1234 yielded
(erasing drug still being in the stomach, reducing the a slightly higher bioavailability of 45–47%.
absorption from the intestine), or diarrhea (decreasing A summary of the pharmacokinetic parameters in
absorption from the intestine). plasma is presented in > Table B.15-1.
For unconjugated and conjugated XYZ1234, Cmax was
MODIFICATIONS OF THE METHOD reached on average 1–1.5 h after oral treatment, after which
In this example, an oral formulation has been compared to a rapid initial elimination phase and a slow terminal elim-
an intravenous one, aiming at ‘‘absolute’’ bioavailability. ination phase was observed, with a terminal elimination
More often, the relative bioavailabilities of different oral half-life of 3.5–4 days (unconjugated XYZ1234). This long
formulations are assessed in BA studies (see > Example 2). half-life in combination of a washout period of a minimum
Depending on the primary purpose of these investiga- of 7 days resulted in a small carry-over effect in Period 2 for
tions, the reference formulation can be a marketed unconjugated XYZ1234. Correction for pre-dose concen-
(solid) drug product, an early clinical phase ‘‘pilot’’ drug trations was not needed, as none of the individual values
product, or an oral solution/suspension. exceeded a 5% threshold of Cmax. Due to a relatively high
If the drug under investigation has a toxic potential lower limit of quantitation of 10 ng/mL of the assay for
(e.g., drugs directed against cancer), BA studies have to be conjugated XYZ1234, the terminal elimination phase for
conducted in the patient setting the drug is intended for use. this analyte could only be reliably determined for one
Deviations from the high level of standardization subject. The concentrations of conjugated XYZ1234 in
might become necessary depending on the properties of plasma were five- to tenfold higher than of unconjugated
the compound. XYZ1234. The AUClast (both analytes) and AUC0–inf
Crossover study designs allowing for intra-individual (unconjugated XYZ1234 only) were similar after treat-
comparisons are preferred for the purpose of bioavailability ment with XYZ1234 25 mg PO and XYZ1234 10 mg IV.
testing. Exceptions, i.e., parallel group design, might become
necessary, for example, if the terminal half-life of the drug Results of bioavailability analysis
exceeds 7 days. Such a long half-life would translate into Treatment ratio
Analyte Parameter (PO/IV) 90% CI
a washout period of five times the elimination half-life, or
more than 35 days, in order to avoid trough concentrations XYZ1234 AUClast 0.38 0.33–0.43
for the second trial period of more than 5% the individual (conjugated)

maximum concentration (Cmax) in this trial period. XYZ1234 AUC0–inf 0.47 0.38–0.59
(unconjugated)
Almost all clinical study types described in the PK
section of this book deal in any way with bioavailability XYZ1234 AUClast 0.45 0.43–0.47
(unconjugated)
and/or bioequivalence questions. Specifics – if there are
any – are mentioned there. Note: Data were dose corrected
176 B.15 Absolute and Relative Bioavailability

. Table B.15-1
Summary of the pharmacokinetic parameters in plasma

Treatment Cmax (ng/mL) Tmax a (h) AUClast (ng.h/mL) AUC0–inf (ng.h/mL) t½ (h)
Unconjugated XYZ1234
10 mg IV 49.5 (30.4–96.7) 0.50b (0.50–0.58) 81.8 (42.7–149.2) 207.4 (85.1–602.4) 85 (36–165)
25 mg PO 15.1 (8.8–24.3) 1.00 (0.50–4.00) 85.0 (60.5–136.4) 215.9 (127.8–337.8) 94 (61–192)
Conjugated XYZ1234
10 mg IV 280.8 (198.0–420.6) 0.50b (0.50–0.75) 367.1 (256.5–1344.5) nd nd
25 mg PO 98.1 (45.9–209.9) 1.50 (1.00–4.00) 317.1 (139.0–704.7) nd nd
nd not determined
a
For Tmax, the median (range) is given instead of the geometric mean (range)
b
The time point 0.50 h is identical with the end of infusion of 30 min/0.5 h duration

In summary, the calculated bioavailability on basis of B.15.2.1.3 Inclusion Criteria

the AUClast of conjugated XYZ1234 was 38%. However,
this could be a slight underestimation of the bioavailabil- Healthy Caucasian males, aged 18–55 years (inclusive),
ity since this AUClast could only be determined until 6 h with a Body Mass Index between 18 and 28 kg/m2 (inclu-
post-dose. A calculation of the bioavailability on basis of sive), normal or clinically irrelevant abnormal findings
the AUClast or AUC0-inf of unconjugated XYZ1234 yielded (in the opinion of the investigator) in the medical history
a slightly higher bioavailability of 45–47%. and physical examination, laboratory values, ECG, blood
pressure and pulse rate, negative serology (HIV antibody,
B.15.2 Example 2 hepatitis B surface antigen, hepatitis C antibody), and
urine screen for drugs of abuse.
B.15.2.1 Protocol Outline
B.15.2.1.4 Treatments
An open-label, crossover study to compare bioavailability,
pharmacokinetics, safety, and tolerability of three different Treatment A (reference):
oral formulations of 50 mg HMR123 in healthy men. Single oral dose of 50 mg solution PEG400/Water/
HMR123
Treatment B (test 1):
B.15.2.1.1 Primary Objective Single oral dose of two 25 mg capsules (filled with
granules) HMR123
To assess the relative bioavailability and pharmacokinetics Treatment C (test 2):
of three oral formulations (one liquid and two capsule Single oral dose of two 25 mg capsules (liquid-filled)
formulations) containing 50 mg HMR123. HMR123

B.15.2.1.5 Pharmacokinetic Data

B.15.2.1.2 Study Design
Concentrations of HMR123 in plasma were measured
This was an open-label, single-dose, randomized, three- pre-dose and at predetermined times up to 96 h post-dose.
way, three sequences (ABC, BCA, CAB), three treatments, The primary analysis examined pharmacokinetic
three periods crossover study with a minimum washout parameters calculated from plasma concentrations of
period between treatments of 7 days. Each subject received HMR123 using non-compartmental techniques.
a single dose of each of the three oral formulations, each of
which containing 50 mg HMR123 under fasting condi- EVALUATION
tions. The sequence of administration (treatments A, B, The primary analyses consisted of characterizing the
and C) was determined according to a randomization bioavailability of three different oral formulations of
schedule. HMR123. Determination of bioavailability was based on
 Absolute and Relative Bioavailability B.15 177

the plasma concentrations. Descriptive statistics and for- optimization efforts from the Galenics department. In
mal statistical analysis were used to summarize and ana- this context please refer also to > Chap. B.14, Specific
lyze the pharmacokinetic parameters of HMR123. Studies for Formulation Development.
Analysis of variance (ANOVA) was performed on the
log-transformed pharmacokinetic parameters Cmax, AUC0–t
and AUC0–inf, with sequence, subject nested within B.15.2.2 Example 2
sequence (subject (sequence), period and treatment effects
as main effects. The sequence effect was tested using the To illustrate the type of data that can be obtained using the
subject (sequence) mean square from the ANOVA as an discussed study, a high-level summary of the pharmacoki-
error term. All other main effects were tested against the netic results obtained from the study described above under
residual error (error mean square) from the ANOVA. ‘‘PROCEDURE’’ as > Example 2 is presented below.
The ANOVA was performed on ln-transformed data.
The mean square error was used to construct 90% confi- B.15.2.2.1 Results – Study Accounting
dence intervals for treatment ratios. The point estimates
were calculated as ratio of the antilogs of the least square In total, 13 male subjects were enrolled in the study and
means and were expressed as percentages. received the investigational product according to the ran-
In order to compare the relative bioavailability of the domization schedule. Twelve subjects completed the trial.
three oral formulations, the following ratios (point esti- One subject (subject no. 1xxx) was withdrawn due to
mates and corresponding 90% confidence intervals) were adverse events on day 5 of trial period I (treatment A).
calculated using adequate contrasts: A:B, A:C, and B:C for He was replaced by subject no. 6xxx with the same treat-
both AUC and Cmax. ment sequence, i.e., ABC.
Bioequivalence was concluded if the 90% confidence
interval for the treatment ratios were fully contained
within the (80–125%)-acceptance range. B.15.2.2.2 Results – Pharmacokinetics
The secondary analyses consisted of assessing the safety
and tolerability of single oral 50 mg doses of HMR123. All subjects who completed the study and for whom the
concentrations of HMR123 were considered sufficient and
CRITICAL ASSESSMENT OF THE METHOD interpretable by the sponsor were included in the phar-
An oral solution consisting of water, PEG400, and macokinetic analyses. The 12 subjects who met these
HMR123 was selected as the reference formulation. Oral criteria were included in the PK analysis according to
solutions are widely accepted as the ‘‘gold standard,’’ the analysis procedures described in the study-specific
because they are devoid of any limitations concerning Statistical Analysis Plan.
dissolution. In order to maintain a high level of standard- In > Figs. B.15-1 and > B.15-2, the plasma concentra-
ization single units of solid formulations are preferred, for tion versus time profiles are given in linear and log-linear
example, one tablet or one capsule. We used two units presentation. For all subjects who were included in the PK
instead, as at the time of implementation of this study, analysis, the individually latest time point of quantifiable
the maximal dose strength was 25 mg, and a minimum concentration Tlast was 72 h post-dose.
effective dose of 50 mg was anticipated. In preceding Cmax and Tmax were obtained from the highest con-
studies, single doses of up to 100 mg had proven to be centration of the measured data. The apparent terminal
safe and well tolerated by healthy male subjects. elimination rate constants (lz) were determined using
If a subject prematurely terminates the study – this was nonlinear regression analysis on those concentration–
the case in our study – the replacer subject has to be time pairs visually assessed to be in the terminal phase.
administered the same treatment sequence as the subject The terminal phase half-life (t1/2,z) was calculated as the
he or she replaces. ratio of ln 2 to lz. Areas under the curve were determined
This study was the third clinical trial in the project. In using the log/linear trapezoidal rule (linear up to the
the First-in-Man study, a film-coated tablet has been used, maximum concentration and log thereafter). The area
which was dropped due to high interindividual variability up to infinity (AUC0–inf ) was determined by extrapolation
in PK parameters. The second one used the test formula- from the last observed data point: the extrapolated area
tion 1, compared to an oral solution. Test formulation 1 was calculated as Clast/lz.
was to be dropped also due to a marked negative food Arithmetic and geometric means for the pharmacoki-
effect. The test formulation 2 was the result of netics parameters are displayed in the table below.
178 B.15 Absolute and Relative Bioavailability

µg/mL
1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0
0 6 12 18 24 36 48 72
Relative time (h)
Treatment:
A B C

Treatment A (reference): oral solution

Treatment B (test 1): capsule filled with granules
Treatment C (test 2): capsule (liquid-filled)

. Figure B.15-1
HMR123 plasma concentration, linear scale

Following treatment with the solution formulation, no (Tmax) and for the last quantified plasma concentration
blood samples were available for subject 1zzz for the 15, (Tlast).
24, and 48 h sampling times. Consequently, a number of Tmax was observed most frequently at 4 h after treat-
pharmacokinetic parameters could not be determined and ment A (4 out of 12 subjects), at 6 h after treatment B
this subject had to be excluded from the population used (4 out of 12 subjects) and at 8 h after treatment C (5 out
for the sensitivity analysis. of 12 subjects). Tlast was most frequently observed at 48 h
after dosing for all three treatments (for 5 of 12 subjects
Descriptive statistics for pharmacokinetic parameters each).
Arithmetic mean (geometric mean)
PK Parameter Treatment A Treatment B Treatment C B.15.2.3 Comparison of Treatments
Cmax (mg/mL) 1.37 (1.35) 1.21 (1.15) 1.25 (1.17)
AUC0–t 26.53 (25.65) 23.97 (22.67) 24.80 (23.04)
B.15.2.3.1 Treatment A vs. Treatment B
(mg*h/mL)
The 90% confidence intervals are not completely within
AUC0–inf 30.06 (29.24) 26.28 (25.01) 27.35 (25.87)
(mg*h/mL)
the 0.8–1.25 range of bioequivalence and therefore no
equivalence could be concluded.
t1/2,z (h) 21.39 (20.90) 20.61 (19.99) 20.03 (19.33)
Rate constant 0.034 (0.033) 0.036 (0.035) 0.037 (0.036)
(1/h) B.15.2.3.2 Treatment A vs. Treatment C
Vz (L) 52.16 (51.57) 60.09 (57.75) 59.33 (53.82)
The 90% confidence intervals are not completely within
Frequency tables were prepared for the blood sam- the 0.8–1.25 range of bioequivalence and therefore no
pling times of maximum observed plasma concentration equivalence could be concluded.
 Absolute and Relative Bioavailability B.15 179

µg/mL
10.00

1.00

0.10

0.01
0 6 12 18 24 36 48 72
Relative time (h)

Treatment:
A B C

Treatment A (reference): oral solution

Treatment B (test 1): capsule 1 (filled with granules)
Treatment C (test 2): capsule 2 (liquid-filled)

. Figure B.15-2
HMR123 plasma concentration, log-linear scale

B.15.2.3.3 Treatment B vs. Treatment C only be concluded for the comparison of treatment B vs.
treatment C, i.e., for both capsule formulations. The lack
The 90% confidence intervals are completely within the of equivalence between the capsule formulations and the
0.8–1.25 range of bioequivalence and therefore equiva- reference oral solution probably results from an increased
lence could be concluded. ratio of the treatments and not from increased variability.

90% CI for treatment effect

Parameter Treatment LS-mean Reference LS-mean Ratio Lower Upper

Cmax (mg/mL) A 1.347 B 1.154 1.168 0.988 1.380

A 1.347 C 1.167 1.154 0.977 1.364
B 1.154 C 1.167 0.988 0.837 1.168
AUC0–t (mg*h/mL) A 25.649 B 22.674 1.131 0.978 1.309
A 25.649 C 23.042 1.113 0.962 1.288
B 22.674 C 23.042 0.984 0.851 1.138
AUC0–inf (mg*h/mL) A 29.967 B 25.948 1.155 1.003 1.330
A 29.967 C 27.048 1.108 0.962 1.276
B 25.948 C 27.048 0.959 0.833 1.105

The results of the sensitivity analysis for all treatment Generally, slightly lower Cmax and later tmax values
comparisons are presented below. Bioequivalence could were observed for HMR123 following administration of
180 B.15 Absolute and Relative Bioavailability

the capsule formulations in comparison to the solution EU CPMP: Note for Guidance on Modified Release Oral and Transdermal
Dosage Forms: Section II (Pharmacokinetic and Clinical Evalua-
formulation. This suggests a slightly lower rate of bioavail-
tion). July 1999
ability for the capsules, which would be anticipated owing EU CPMP: Note for Guidance on the Investigation of Bioavailability and
to the time required for breakdown of the capsules to Bioequivalence. July 2001
occur. The extent of absorption was also a little higher EU CPMP: Points to Consider on the Clinical Requirements of Modified
for the solution formulation, as evidenced by the AUC0–t Release Products to be Submitted as a Line Extension of an Existing
Marketing Authorization. June 2003
and AUC0–inf data. After 8 h post-dose, however, the
Ezzet F, Krishna G, Wexler DB, Statkevich P et al (2001) A population
median plasma concentration versus time curves were pharmacokinetic model that describes multiple peaks due to
almost identical for all the three formulations investigated enterohepatic recirculation of ezetimibe. Clin Ther 23(6):871–885
in this study. Statistical comparison of Cmax, AUC0–t and ICH E4: Dose–Response Information to Support Drug Registration.
AUC0–inf between the capsule formulations and the oral March 1994
Kharasch ED, Walker A, Hoffer C, Sheffels P (2005) Evaluation of first-
solution indicated that the confidence intervals for the
pass cytochrome P4503A (CYP3A) and P-glycoprotein activities
parameter ratios were not fully contained within the using alfentanil and fexofenadine in combination. J Clin Pharmacol
0.80–1.25 range: a slightly higher bioavailability was thus 45:79–88
confirmed for the solution formulation. A statistical com- Thummel KE, Wilkinson GR (1998) In vitro and in vivo drug interactions
parison of the two-capsule administrations on the other involving CYP3A. Annu Rev Pharmacol Toxicol 38:389–430
US FDA Guidance for Industry: Bioavailability and Bioequivalence Stud-
hand showed the corresponding confidence intervals to be
ies for Nasal Aerosols and Nasal Sprays for Local Action. April 2003
fully within the 0.80–1.25 range; i.e., the criteria for bio- US FDA Guidance for Industry: Bioavailability and Bioequivalence Stud-
equivalence were met for these two capsule formulations. ies for Orally Administered Drug Products – General Considerations.
Coefficients of variation on the primary pharmacoki- March 2003
netic parameters (reflecting inter-subject variability) were US FDA Guidance for Industry: Exposure–Response Relationships –
Study Design, Data Analysis, and Regulatory Applications. April
generally about 30% for the two-capsule formulations and
2003
about 25% for the oral solution. Slightly lower values were US FDA Guidance for Industry: Food-Effect Bioavailability and Fed
seen for the capsule filled with wet granules in comparison Bioequivalence Studies. December 2002
to the liquid-filled capsules. Zahng Y, Benet LZ (2001) The gut as a barrier to drug absorption:
combined role of cytochrome P450 3A and P-glycoprotein. Clin
Pharmacokinet 40(3):159–168
REFERENCES AND FURTHER READING
Zhou H (2003) Pharmacokinetic strategies in deciphering atypical drug
Bergman AJ, Burke J, Larson P, Johnson-Levonas AO et al (2006) Interac-
absorption profiles. J Clin Pharmacol 43:211–227
tion of single-dose ezetimibe and steady-state cyclosporine in renal
transplant patients. J Clin Pharmacol 46:328–336
Chan LM, Lowes S, Hirst BH (2004) The ABCs of drug transport in
intestine and liver: efflux proteins limiting drug absorption and
bioavailability. Eur J Pharm Sci 21(1):25–51