Journal of the American Academy of Dermatology, November, 2013
Accepted 24 May 2013. published online 15 August 2013.
Background
A
high-fat meal is needed for optimal absorption of isotretinoin. A new
formulation of isotretinoin, which enhances absorption of isotretinoin
in the absence of dietary fat, has recently been approved by the Food
and Drug Administration (FDA).
Objective
We sought to
compare the pharmacokinetic profiles of a new formulation of
isotretinoin (isotretinoin-Lidose) with the innovator isotretinoin
formulation.
Methods
This study was an open-label,
single-dose, randomized, 4-treatment, crossover comparative trial
between a new and innovator formulation of isotretinoin in the fasting
and fed states.
Results
Both formulations were
bioequivalent under fed conditions. As expected in a fasting state,
absorption of both formulations was reduced. A considerable difference
between the 2 drugs occurred under fasted conditions–there was a marked
improvement in overall bioavailability of the isotretinoin-Lidose
formulation. Mean plasma levels of the isotretinoin-Lidose formulation
during fasting reached 66.8% of that observed with a fatty meal, and
those of the isotretinoin formulation only reached 39.6% of that
observed with a fatty meal.
Limitations
Only the FDA-stipulated standard high-fat, high-calorie meal of 50-g fat was studied in the fed state.
Conclusion
Isotretinoin-Lidose
formulation is bioequivalent to the innovator formulation
under fed conditions with regard to its pharmacokinetic
profile but delivers twice as much isotretinoin
and 4-oxo-isotretinoin when administered after an overnight fast.
Abbreviations used: AUCt, area
under plasma concentration versus time curve from time zero (0 hour) to
time of last measurable plasma concentration, calculated by linear
trapezoidal method, AUCi, area
under plasma concentration verus time curve from time zero (0 hour) to
infinity (AUCi calculated as sum of AUCt + measured plasma concentration
at time of last measurable plasma concentration/apparent first order
elimination rate constant), BA, bioavailability, BE, bioequivalent, Cmax, maximum measured plasma concentration over sampling time period, FDA, Food and Drug Administration, Tmax, time of maximum measured plasma concentration
Capsule Summary
Absorption of an orally administered drug is influenced by its solubility and intestinal permeability.∗
Food is a well-known agent that can have an impact in drug absorption.
For some drugs, food will have no effect and for other drugs, it can
either increase or decrease the amount of absorption. Isotretinoin is a
molecule with good permeability and poor solubility in the aqueous
environment of the intestine, and its absorption is greatly enhanced by
fatty foods. When taken without food, absorption (permeability) is low.
Fasted plasma levels of current isotretinoin formulations are nearly 60%
lower than levels in the fed state.1
To ensure optimal and consistent absorption and maintenance of
therapeutic blood concentrations, isotretinoin must be taken with food,1 preferably with a high-fat meal (50% caloric intake from fat) as defined by the Food and Drug Administration (FDA).2
A
technology (Lidose) now exists that enhances the intestinal absorption
of poorly soluble agents, such as isotretinoin. Lidose is a novel
formulation that uses lipid agents to encase lipophilic drugs and create
an optimal environment for absorption.3
The FDA has recently approved isotretinoin-Lidose, a new formulation of
isotretinoin, which can be taken without regard to meals.
In this
study, we describe the pharmacokinetic profiles of this new agent and
compare it with the innovator formulation of isotretinoin.
Methods
This
trial was an open-label, single-dose, randomized, 4-treatment,
crossover comparative bioavailability (BA) study. The ethics review
board at Optimum Clinical Research Inc approved the study protocol, its
amendments, and informed consent form. Full trial protocol was submitted
to the FDA during the approval process for isotretinoin-Lidose.
Subjects
All
60 individuals who were selected for the study met the inclusion and
exclusion criteria, and were judged by the investigator to be healthy.
Inclusion criteria included:
•Healthy, nonsmoking, men and nonpregnant, nonlactating women, 18 to 55 years of age, inclusive.
•Body weight within ±15% of the appropriate weight for the individual's height and frame.
•Women
were either unable to have children, willing to remain abstinent, used
an effective method of double-barrier birth control, or used a hormonal
contraceptive agent.
Exclusion criteria included:
•Known history or presence of any clinically significant disease.
•Known or suspected carcinoma.
Study drugs and administration
Study
drugs were administered in a clinical facility over 4 periods/intervals
with a 21-day washout period between successive administrations:
Isotretinoin-Lidose (Absorica) capsules manufactured for Cipher Pharmaceuticals Inc, Barbados, by Galephar PR Inc, lot 28A02.
Isotretinoin (Accutane) capsules manufactured by Roche Laboratories Inc, Nutley, NJ, lot U0622.
A
predetermined computer-generated randomization scheme for the 4
treatment periods (assigned to participants in 1 of 12 different
sequences):
Treatment A: Isotretinoin-Lidose (2 × 20 mg) administered after an overnight fast of at least 10 hours.
Treatment B: Isotretinoin-Lidose (2 × 20 mg) 30 minutes after consumption of a modified∗ high-fat, high-calorie breakfast with a reduced vitamin-A content.
Treatment C: Isotretinoin (1 × 40 mg) administered after an overnight fast of at least 10 hours.
Treatment D: Isotretinoin (1 × 40 mg) administered 30 minutes after consumption of a modified∗ high-fat, high-calorie breakfast with a reduced vitamin-A content.
Blood sample collection and handling
Blood
samples were collected within 1 minute of their scheduled time at
regular intervals from 10 hours before to 72 hours after each drug
administration.
Analyses for isotretinoin, 4-oxo-isotretinoin,
tretinoin, and 4-oxo-tretinoin in human plasma were performed using a
validated analytical method with a calibration range of 1.00 to 750
ng/mL (isotretinoin and 4-oxo-isotretinoin) and 1.00 to 250 ng/mL
(tretinoin and 4-oxo-tretinoin). All laboratory personnel were blinded
regarding the randomization scheme and treatment plan.
Clinical evaluations
Laboratory
tests included those for hematology, serum chemistry, urinalysis, HIV,
hepatitis B antigen, and hepatitis C for all individuals, and human
chorionic gonadotropin pregnancy tests on women who were not
postmenopausal and who did not have a hysterectomy. Blood pressure,
pulse rate, temperature, and respiration were also measured.
A physical examination was performed on all individuals upon study completion or upon removal from the study.
Data analysis
Pharmacokinetic analysis was conducted using a noncompartmental approach.4
The following parameters were estimated: area under plasma
concentration versus time curve from time zero (0 hour) to time of
last measurable plasma concentration, calculated by linear trapezoidal
method (AUCt); area under plasma concentration versus time curve from
time zero (0 hour) to infinity (AUCi) (AUCi calculated as sum of AUCt +
measured plasma concentration at time of last measurable plasma
concentration/apparent first order elimination rate constant); maximum
measured plasma concentration over sampling time period (Cmax); and time
of maximum measured plasma concentration (Tmax).
Statistical
analysis was performed to compare the BE of the isotretinoin-Lidose
formulation and the isotretinoin formulation under fasting and fed
conditions. The pharmacokinetic parameters were statistically analyzed
according to current FDA guidance.5
Arithmetic
means, SD, and coefficients of variation were calculated. Analyses of
variance were performed on the log-transformed AUCt, AUCi, and Cmax and
on the untransformed Tmax. A 5% level of significance was used for
within-subject comparisons (ie, period, treatment). Each analysis of
variance included a calculation of least-squares means, adjusted
differences between treatment means, and the SE associated with these
differences.
The 90% confidence intervals for the ratio of drug
treatment least-squares means were calculated for the AUCt, AUCi, and
Cmax parameters using log-transformed data. The confidence intervals are
expressed as a percentage relative to the least-squares mean of the
isotretinoin formulation.
Results
Patient demographics
A
total of 60 healthy individuals (40 male and 20 female) were dosed
in period 1. Three individuals did not complete period 2 of the study: 1
withdrew after period 2 check-in for personal reasons, and 2 were
dismissed because of nonadherence. The remaining individuals, 57, were
dosed in periods 3 and 4 and completed the study. They had a
mean ± SD (range) age of 36 ± 9 (21-55) years, height of 169.7
± 8.7 (150.0-188.0) cm, and weight of 71.5 ± 9.1 (49.8-86.4) kg.
Pharmacokinetic analyses
The
average of the measured predose concentration levels of isotretinoin,
4-oxo-istoretinoin, tretinoin, and 4-oxo-tretinoin at −10, −2, and 0
hours was used as baseline for each individual, within each period. The
subject- and period-specific baseline values were subtracted from the
subsequent postdose measured levels. The 0-hour levels of isotretinoin
and its metabolites and the negative levels for the baseline-adjusted
concentrations were set to 0. The pharmacokinetic analysis was performed
on the baseline-adjusted analyte levels.
Arithmetic means and
geometric means for AUCt, AUCi, Cmax, and Tmax in the 4 dosing periods
were calculated for isotretinoin, 4-oxo-isotretinoin, tretinoin, and
4-oxo-isotretinoin, respectively.
Effect of food
The
high-fat, high-caloric meal produced the highest observed plasma
concentrations of isotretinoin and metabolites for both formulations
when compared with the fasting state. Food increased the systemic
exposure to isotretinoin to 6695.87 ng•h/mL with the isotretinoin-Lidose
formulation and to 6315.15 ng•h/mL with the innovative isotretinoin
formulation. The peak exposure to isotretinoin (Cmax) was 434.33 ng/mL
for the isotretinoin-Lidose formulation and 431.88 ng/mL for the
isotretinoin formulation. Food triggered a similar delay between the 2
formulations in the time to reach Cmax. Time of maximum plasma
concentrations reached with food was 6.85 hours for the
isotretinoin-Lidose formulation and 6.75 hours for the isotretinoin
formulation.
Table I. Menu for modified, high-fat, high-calorie breakfast
| Calories | Carbohydrates | Protein | Fat |
|---|
| 1 Regular bagel | 300 | 67 g | 12 g | 3 g |
| 2 T peanut butter | 250 | 9 g | 12 g | 22 g |
| 5 Slices bacon | 125 | 0.2 g | 10 g | 10 g |
| 6 Fluid oz apple juice | 80 | 22 g | 0.1 g | 0.1 g |
| 1 Dutchie donut (square donut containing raisins and coated with sugary glaze from Canadian Tim Hortons chain) | 250 | 33 g | 3 g | 15 g |
| Total | 1005 | 131.2 g | 37.1 g | 50.1 g |
With
respect to 4-oxo-isotretinoin, food increased systemic and peak
exposure of this metabolite with both formulations. Results for
tretinoin were similar. Although the 4-oxo-tretinoin data exhibited very
large variability, the trend in changes from fasting to fed conditions
was similar to the other analytes (data not shown).
Pharmacokinetic
data from the 2 formulations were BE under fed conditions. The 90%
confidence interval of the ratios of geometric means of the 2
formulations conducted under fed conditions for AUCt, AUCi, and Cmax
were within the 80% to 125% range for isotretinoin, 4-oxoisotretinoin,
tretinoin, and 4-oxo-tretinoin. Given these results, isotretinoin-Lidose
exhibited equivalent rate and extent of exposure to isotretinoin under
high-fat fed conditions in human beings after a single dose.
Effect of fasting
AUCt,
AUCi, and Cmax for both formulations were reduced in the absence of
food. Significant differences were observed between the 2 formulations
in the fasting state. After an overnight fast, the isotretinoin-Lidose
formulation delivered approximately twice as much isotretinoin (4470.45
vs 2500.01 ng•h/mL) and 4-oxo-isotretinoin (11,865.08 vs 6049.60
ng•h/mL) as did the isotretinoin formulation. Similar findings were
observed for tretinoin and 4-oxo-tretinoin (data not shown),
respectively. Mean AUCi plasma levels of isotretinoin from the
isotretinoin-Lidose formulation during fasting reached 66.8% of that
observed with food, and those of the isotretinoin formulation only
reached 39.6% of that observed with food.
Cmax for isotretinoin
was also twice as high with the isotretinoin-Lidose formulation
when compared with the isotretinoin formulation: 323.18 versus 161.43
ng•h, respectively. Similar to that observed with AUCi, Cmax for
isotretinoin with the isotretinoin-Lidose formulation during
fasting represented 74% of the Cmax achieved after a meal and with the
isotretinoin formulation, it only represented 37% of the Cmax achieved
after a meal. For 4-oxo-isotretinoin, Cmax was 163.73 ng•h after fasting
for the isotretinoin-Lidose formulation (46.5% of the Cmax after a
meal) and 70.55 ng•h for the isotretinoin formulation (22.7% of the Cmax
after a meal).
Mean plasma concentrations of isotretinoin over time
Although
both formulations share a similar Tmax for isotretinoin in the fasting
state (about half of the Tmax under fed conditions), mean peak plasma
concentration of isotretinoin is clearly higher with the
isotretinoin-Lidose formulation, reaching 289.49 ng/mL at 3 hours after
ingestion versus the isotretinoin formulation, which reached 133.75
ng/mL at 3.5 hours (Fig 1).
After reaching peak concentrations, plasma concentrations of
isotretinoin declined at approximately the same rate for both
formulations; however, overall, isotretinoin levels from the Lidose
formulation always remained higher during the clearance phase.
The
mean peak concentration of isotretinoin for the isotretinoin-Lidose
formulation under fasting conditions was nearly as high as the peak
concentration under fed conditions (289.49 vs 317.50 ng/mL). Mean plasma
concentrations of the 2 formulations under fed conditions appeared to
follow the same kinetics: rapid increase after ingestion to a peak at
4.5 hours followed by a decline and then a second peak at 10 hours.
Plasma concentrations disappearance from 12 hours to the end of the
study at 72 hours were virtually superimposable (Fig 1).
Mean
measured plasma concentrations of 4-oxo-isotretinoin, tretinoin, and
4-oxo-tretinoin over time followed similar kinetics as mean plasma
concentrations of isotretinoin (data not shown).
Adverse events
There
were 55 adverse events involving 25 individuals during the course of
the 4 periods of the study. Headache was the most commonly reported
adverse event for all 4 periods. There were no serious adverse events
recorded, and none of the adverse events had a significant impact on the
safety of the participants or on the integrity of the study results.
Discussion
Results
from this crossover comparative study demonstrate a BE of the
isotretinoin-Lidose formulation and the isotretinoin formulation under
fed conditions: AUCt, AUCi, Cmax, and Tmax for the 4 analytes were
comparable. As anticipated, a high-fat, high-calorie meal increased the
relative BA of both formulations compared with the fasting state and is
in agreement with results reported by Colburn et al.1
In
contrast, striking differences in the BA of isotretinoin were observed
between the 2 formulations in the fasting state. Systemic exposure of
isotretinoin was 1.8-fold higher with isotretinoin-Lidose than with the
isotretinoin formulation. In addition, mean Cmax of isotretinoin in the
fasting state was 66.8% of the mean Cmax observed in the fed state with
the isotretinoin-Lidose formulation versus 39.6% with the isotretinoin
formulation.
More than 75% of individuals in this study absorbed
less than 50% isotretinoin when it was taken in a fasting state compared
with a high-fat meal (Fig 2).
In the case of isotretinoin-Lidose, 75% of individuals absorbed 60% or
more isotretinoin in the fasting state compared with taking it with a
fatty meal. The box plot analysis (Fig 2)
shows the variability of absorption of these formulations. Only 1
individual had a higher ratio of fasting to fed area under the curve
(AUC) for isotretinoin and only in a handful of individuals was the
isotretinoin-Lidose not more effective than isotretinoin alone.
Fig 2.
Effect of food on absorption. Top and bottom of the box are the 75th and 25th percentiles, respectively, dash line is the median, solid line is the mean, bars are the highest and lowest values of not more than 1.5 times the length of the box, circles are outliers, and x
is an extreme outlier (greater than 3 times the intrabox distance).
More than 75% of individuals in the fasting state achieved plasma levels
less than 50% of that when isotretinoin was taken with food (mean
difference 39.6%). In the same individuals, plasma levels with the new
formulation reached 66.8% of that found with food. AUC, Area under the curve.
From
a clinical perspective, a formulation of isotretinoin that can be taken
without regard to meal and provides more consistent absorption may be
advantageous for both the clinician and patient. It is known that total
exposure of isotretinoin is an important attribute in preventing
relapse. Dose-ranging trials for isotretinoin show no difference in
efficacy for 0.1, 0.5, and 1.0 mg/kg/d of isotretinoin; yet, there is a
clear dose response for the rate of relapse.6
In addition, numerous studies have reported that cumulative doses of
less than 120 mg/kg/body weight are associated with relapse and repeated
treatment.6, 7, 8, 9
Young adults and teenagers, who comprise a large segment of
isotretinoin use, are known for their poor and irregular eating habits.10, 11 Skipping meals or limiting intake of fat and cholesterol, as recommended by the FDA and nutrition experts,12 would lower the level of isotretinoin that is absorbed and possibly might affect outcomes.
In
summary, isotretinoin-Lidose is a novel formulation of isotretinoin
that is BE to innovator isotretinoin under fed conditions with regard to
its pharmacokinetic profile but delivers twice as
much isotretinoin and 4-oxo-isotretinoin when administered after an
overnight fast. The clinical consequences of the decreased variability
in absorption between the fed and fasted states with isotretinoin-Lidose
will need to be established in future studies.
