Important Effect of Food on the Bioavailability of Oral Testosterone Undecanoate

Transcription

1 Important Effect of Food on the Bioavailability of Oral Testosterone Undecanoate Wilma M. Bagchus, Ph.D., Rita Hust, M.D., Frans Maris, Ph.D., Peter G. Schnabel, M.S., and Natalie S. Houwing, M.S. Study Objective. To assess the effects of food on the bioavailability of testosterone undecanoate, testosterone, and 5 -dihydrotestosterone (DHT) after administration of a new oral testosterone undecanoate formulation, Andriol Testocaps. Design. Randomized, open-label, crossover study with a 1-week washout period. Setting. Clinical pharmacology unit. Subjects. Sixteen healthy postmenopausal women. Intervention. Single oral doses of testosterone undecanoate 80 mg were administered either during a fasting period or after consumption of a standardized continental breakfast. Measurements and Main Results. Serum concentrations of testosterone undecanoate were assayed by liquid chromatography with mass spectrometry detection; testosterone and DHT were assayed by gas chromatography with mass spectrometry detection. Serum concentrations of testosterone, testosterone undecanoate, and DHT were low to negligible when testosterone undecanoate was administered to subjects in a fasting state; these values were significantly higher when the test drug was coadministered with food. For testosterone, the maximum serum concentration and area under the plasma concentration time curve were 0.67 ng/ml and 5.37 ng hr/ml, respectively, in the fasting state, versus 10.7 ng/ml and 56.4 ng hr/ml, respectively, in the fed state. The same parameters were also significantly higher for testosterone undecanoate and DHT in the fed versus fasting subjects. Conclusion. Food increases the bioavailability of testosterone undecanoate, testosterone, and DHT. For proper absorption, Andriol Testocaps must be taken with meals. (Pharmacotherapy 2003;23(3): ) Testosterone replacement therapy is administered in a wide range of circumstances. It restores natural androgen levels in patients with primary or secondary hypogonadal disorders, whether congenital or acquired, as occurs after castration or with eunuchoidism, hypopituitarism, endocrine impotence, and certain types of infertility. It is used to treat symptoms of partial androgen deficiency in aging men (andropause); these symptoms include depressed mood, fatigue, loss of energy, sexual problems, and problems with physical agility. Testosterone is given to femaleto-male transsexuals to induce masculinization, and it is used to treat osteoporosis caused by androgen deficiency. 1 Oral administration of testosterone results in low bioavailability due to presystemic clearance by the liver. To circumvent this first-pass effect in the liver, testosterone sometimes is administered in patches, gels, injections, or implants. Another option is Andriol (Organon, Oss, The Netherlands) the only oral testosterone formulation, designed

2 320 PHARMACOTHERAPY Volume 23, Number 3, 2003 to deliver testosterone to the systemic circulation by the intestinal lymphatic route, thereby circumventing first-pass inactivation in the liver. Andriol, which is available in more than 80 countries, contains testosterone undecanoate dissolved in oleic acid inside a soft gelatin capsule. The esterification of testosterone with undecanoate renders testosterone undecanoate sufficiently lipophilic to be incorporated in chylomicrons formed during lipid digestion in the intestine. These chylomicrons then are transported by the intestinal lymphatic system. Another factor that is relevant to the extent of lymphatic absorption is the nature of the lipophilic solvent in the capsule. 2, 3 During absorption, testosterone undecanoate is partly reduced to 5 -dihydrotestosterone undecanoate (DHTU), which also is absorbed by the lymphatic system. 4 From the lymphatic system, testosterone undecanoate and DHTU are released into the circulation by the thoracic duct. Both testosterone undecanoate and DHTU are hydrolyzed to yield the natural male androgens, testosterone and 5 -dihydrotestosterone (DHT). 4, 5 Because of this lipid-associated route of absorption, one would expect that administration of oral testosterone undecanoate with a meal would enhance absorption and thus increase the bioavailability of testosterone undecanoate, testosterone, and DHT. The enhancing effects of food on the lymphatic absorption of lipophilic compounds have been confirmed in a dog model using halofantrine 6 and in humans with an experimental oral testosterone undecanoate formulation. 7 Although the relevance of food to the bioavailability of testosterone undecanoate is acknowledged in the product labeling, no detailed clinical study had yet been conducted. Andriol must be stored in a refrigerator (2 8 C) in the pharmacy for stability reasons, whereas patients must store it at room temperature to ensure optimal absorption. Shelflife at room temperature is only 3 months. These From the Clinical Pharmacology Department (Dr. Bagchus) and the Department of Drug Metabolism and Kinetics (Dr. Maris, Mr. Schnabel, and Ms. Houwing), Organon, Oss, The Netherlands; and Focus Clinical Drug Development GmbH, Neuss, Germany (Dr. Hust). Supported by Organon, Oss, The Netherlands. Presented in part at the 7th International Congress on Andrology, Montreal, Canada, June 15 19, 2001, and the 3rd World Congress on the Aging Male, Berlin, Germany, February 7 10, Address reprint requests to Wilma M. Bagchus, Ph.D., Organon, Clinical Pharmacology Department, Post Office Box 20, 5340 BH Oss, The Netherlands; wilma. bagchus@organon.com. complicated storage requirements carry the risk of nonadherence, which might result in administration of inactive compound. Therefore, a new, more stable pharmaceutical formulation of Andriol Andriol Testocaps (Organon) was developed in which the oleic acid solvent is replaced by castor oil and propylene glycol laurate. The improved formulation can be stored at room temperature (15 30 C) for 3 years. This study evaluated serum concentrations of testosterone undecanoate, testosterone, and DHT, with the goal of assessing the effects of a standardized meal on the pharmacokinetic properties of a single oral dose of testosterone undecanoate 80 mg in the form of Andriol Testocaps. Materials and Methods Pharmaceutical Formulation We evaluated Andriol Testocaps, which were oral oval soft-gelatin capsules containing testosterone undecanoate 40 mg with propylene glycol laurate (lipophilic surfactant) and castor oil. Study Center The clinical part of the study was performed at Focus Clinical Drug Development GmbH, Neuss, Germany. The study protocol was approved by the Physicians Chamber North-Rhine Ethics Committee. The study was conducted in compliance with the Declaration of Helsinki and with Good Clinical Practice guidelines. Subjects Sixteen women participated in this study; all 16 completed the study in accordance with the protocol. To be considered for inclusion, subjects had to provide written informed consent before screening evaluations, be postmenopausal, be years old, and have pretrial screening total testosterone levels of 1.00 ng/ml or below. In addition, participants were required to smoke fewer than 10 cigarettes/day, have a body mass index of kg/m 2, be in good age-appropriate physical and mental health, have normal cervical smears performed within the last 12 months, and have normal mammograms performed within the last 24 months. Subjects were excluded from the trial if they had a history of sensitivity to testosterone undecanoate or chemically related substances or used drugs known to alter estrogen metabolism

3 FOOD AND BIOAVAILABILITY OF ORAL TESTOSTERONE UNDECANOATE Bagchus et al 321 or affect cytochrome P450 enzymes. Other exclusion criteria were oral or transdermal estrogen-progestin hormone replacement therapy within 4 weeks of the start of the study; sex hormone depot injections within 6 months of the study; sex hormone implants at any time; history of significant allergic or other serious diseases, including malignancies; history of or current abuse of drugs, alcohol, or solvents; and blood donation or participation in an investigative drug trial within 90 days before the start of this study or during the study. Subjects were not allowed to use any drugs, whether prescribed by a physician or acquired over the counter, from 7 days before the first dose until the end of the study. An exception was the sporadic use of acetaminophen during days when the subject was not hospitalized (i.e., 24 hours after dosing). Clinical Study Design and Blood Sampling This was an open-label, single-center, singledose, randomized, two-way crossover pharmacokinetic study with a 7-day washout between treatment periods. A single dose of Andriol Testocaps 80 mg was administered to either fed or fasting participants. Volunteers were randomized to receive one of two sequences of therapy. The first sequence was a single oral dose of Andriol Testocaps 80 mg (2 x 40-mg capsules) in the fasting state. This was followed, after a 1-week washout period, by administration of Andriol Testocaps 80 mg after a standardized breakfast. In the second sequence, the order of fasting versus fed was reversed. For logistical reasons, the trial was performed in two cohorts of eight people. Volunteers received a single oral dose of Andriol Testocaps 80 mg with 200 ml of water after an overnight fasting period. Thereafter, the subjects fasted for another 4 hours. Volunteers under nonfasting conditions received the capsules 5 minutes after a standardized continental breakfast. This meal consisted of two rolls, two slices of cheese, two slices of ham, jam (20 g), butter (20 g), and two cups of caffeine-free coffee. This breakfast provided approximately 460 kcal (1945 kj), 23 g of fat, 48 g of carbohydrates, and 14 g of protein. Four hours after dosing, all volunteers received a standardized hot meal containing approximately 530 kcal (2219 kj), 16.6 g of fat, 42.8 g of carbohydrates, and 48.3 g of protein. Seven hours after dosing, participants received a light snack, and 10 hours after dosing, a standardized cold meal was served. This meal provided approximately 1043 kcal (4367 kj), with 73.7 g of fat, 38.4 g of carbohydrates, and 48.7 g of protein. Thirteen hours after dosing participants received another light snack. Subjects were required to refrain from consuming grapefruit juice, caffeine, and other methyl-xanthines (e.g., coffee, tea, cola, or chocolate) from 48 hours before dosing until after the last pharmacokinetic blood sampling, which was taken 24 hours after dosing. Serial blood samples were taken before dosing and at 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 16, 18, 20, and 24 hours after dosing for determination of testosterone undecanoate, testosterone, and DHT. Immediately after collection, blood was processed to serum and stored in a freezer. Analytical Assays Testosterone undecanoate concentrations were determined using a validated liquid chromatographic assay with mass spectrometry detection after solid-phase extraction. Testosterone and DHT concentrations in serum were assayed using a validated gas chromatographic assay with mass spectrometry detection. Each analytical series of testosterone, DHT, or testosterone undecanoate consisted of patient samples, eight calibration samples (in duplicate), three or four quality control samples (in triplicate), two blank internal standard samples (in duplicate) and at least one blank serum (in the testosterone undecanoate analysis only) and one blank water sample. The lower limit of quantification was 0.2 ng/ml for testosterone undecanoate and 0.1 ng/ml for both testosterone and DHT. During bioanalysis, interassay coefficients of variation for the independent quality control samples were % for testosterone undecanoate, % for testosterone, and % for DHT. Accuracy of the quality control samples was % for testosterone undecanoate, % for testosterone, and % for DHT. Bioanalysis was performed at Organon Development GmbH, Waltrop, Germany (Department of Drug Metabolism and Kinetics, Section Bioanalytics). The analytical methods were fully validated, and the analyses were conducted in compliance with the Good Laboratory Practice Principles of the Organisation of Economic Cooperation and Development.

4 322 PHARMACOTHERAPY Volume 23, Number 3, 2003 Pharmacokinetic Parameters Pharmacokinetic parameters were calculated by subject and treatment from serum concentrations of testosterone undecanoate, testosterone, and DHT. Maximum concentrations (C max ) in serum and time to C max (T max ) were taken from the measured serum concentration data. In case all concentrations for a subject throughout the sampling period were below the limit of quantification, C max was set to one half the limit of quantification for calculation of descriptive statistics. The area under the plasma concentration time curve from zero to the sampling time of the last measurable concentration within a subject (AUC 0 tlast ) was calculated using the linear trapezoidal rule. In case all concentrations for a subject throughout the sampling period were below the limit of quantification,, the AUC 0 tlast was set to the value of one half the limit of quantification during 1 hour. Testosterone, testosterone undecanoate, and DHT concentrations were generally back to baseline at the end of the sampling period. Consequently, AUC 0 essentially would be equal to AUC 0 tlast and therefore was not determined. In this study AUC has been used to denote AUC 0 tlast. Testosterone Undecanoate (ng/ml) Testosterone (ng/ml) Time (hour) Statistical Analysis The food-effect testing comparison was based on C max and AUC. Values achieved during fasting were treated as references, whereas values obtained during fed treatment were considered the test treatment. The C max and AUC parametric point estimates of test:reference ratios with 90% confidence intervals (CIs) (multiplicative model) were derived from analysis of variance (ANOVA) on log-transformed values. These point estimates were obtained by calculating the ratio of the geometric least-squares means of the test and reference treatments. The following model was used for the ANOVA: between subjects: sequence; subject (sequence) = error (between); within subjects: treatment (fasting or fed); period; residual = error (within). Effects were considered statistically significant when the p value (2-tailed) was less than or equal to The acceptance range was A food effect was considered absent if the 90% CI for C max and AUC for testosterone were fully contained within the acceptance range. Data pertaining to testosterone undecanoate and DHT were considered to be supportive. In case all concentrations for a subject DHT (ng/ml) Time (hour) Time (hour) Figure 1. Mean plasma concentration time curves for testosterone undecanoate, testosterone, and 5 -dihydrotestosterone (DHT) after administration of single oral dose of Andriol Testocaps 80 mg after overnight fasting ( ) and with a standardized continental breakfast ( ) in 16 healthy postmenopausal women. If more than one third of the data for a given time point were less than the limit of quantification, the mean concentration was set to one half the limit of quantification for graphical presentation.

5 FOOD AND BIOAVAILABILITY OF ORAL TESTOSTERONE UNDECANOATE Bagchus et al 323 Table 1. Statistical Analysis of the Pharmacokinetic Parameters of Testosterone, Testosterone Undecanoate, and DHT After Single Oral Doses of Andriol Testocaps in 16 Healthy Postmenopausal Women Compound, Results a Food Effect Parameter Fasting Fed Point Estimate 90% CI or p Value Present b Testosterone C max (ng/ml) ± ± Yes AUC (ng hr/ml) 5.37 ± ± Yes T max (hrs) 3.53 ( ) 5.00 ( ) p= No Testosterone undecanoate C max (ng/ml) ± ± Yes AUC (ng hr/ml) 2.09 ± ± Yes T max (hrs) 5.00 ( ) 5.00 ( ) p= No DHT C max (ng/ml) ± ± Yes AUC (ng hr/ml) 1.86 ± ± Yes T max (hrs) 5.00 ( ) 6.00 ( ) p= No CI = confidence interval; DHT = 5 -dihydrotestosterone; C max = maximum concentration; AUC = area under the plasma concentration time curve from time zero to last serum measurement; T max = time to maximum concentration. a Data are mean ± SD for C max and AUC; median (range) for T max. b For C max, and AUC, food effect was determined by calculating parametric point estimates of the true test:reference ratio (µ[fed]:µ[fasted]) with their 90% CIs (multiplicative model), as obtained from analysis of variance on log-transformed values; and for T max by classic hypothesis testing using an analysis of variance on ranks. Food effect was considered to be present for C max and AUC when the 90% CI was fully outside the acceptance range ( ) and for T max when the p value was less than throughout the sampling period were below the limit of quantification, a value of one half the limit of quantification was substituted for C max, and a value representing the AUC resulting from a concentration of one half the limit of quantification during 1 hour was substituted for AUC. For the pharmacokinetic parameter T max, classic hypothesis testing was performed on ranks using the ANOVA model described above. Statistical analyses and calculations of pharmacokinetic parameters were performed using SAS version 6.12 (SAS Institute, Cary, NC). Results Subjects All randomized subjects completed the study. Mean (range) age, weight, height, and body mass index were 57.3 (47 64) years, 69.6 (55 87) kg, ( ) cm, and 25.2 ( ) kg/m 2, respectively. All subjects were Caucasian women. Twelve subjects (6 in each group) were nonsmokers. There were no obvious differences between the two groups. One subject used Hydergine drops for an ear problem 3 days after the first dose of the study drug, but this was not considered to have affected the outcome of the study. Two subjects did not eat the light snack 13 hours after dosing. No major adverse events or other tolerability concerns were noted. Only headache, reported by two subjects, was judged to be possibly drug related. No findings of clinical relevance were indicated by electrocardiography, physical examination, vital signs, or laboratory investigations (assessed before dosing and 8 16 days after the last dose). Serum Concentrations and Pharmacokinetic Parameters When subjects received Andriol Testocaps during the fasting state, they had low serum concentrations of testosterone undecanoate, testosterone, and DHT. In many instances, these concentrations were below or close to the limit of quantification. This was particularly true for testosterone undecanoate and DHT. As shown in Figure 1, serum concentrations of all three compounds were comparatively high when Andriol Testocaps was administered with a standardized breakfast. Table 1 summarizes the calculated pharmacokinetic parameters and the results of our statistical analyses. Values of C max for testosterone averaged 0.66 ng/ml under fasting conditions versus 10.7 ng/ml under fed conditions. Similarly, testosterone s AUC was 5.37 ng hour/ml for fasting state versus 56.4 ng hour/ml for fed state. Similar food effects were achieved for testosterone undecanoate and DHT. For testosterone undecanoate, C max averaged ± ng/ml under fasting conditions and 291 ± 144 ng/ml under fed conditions, whereas AUC averaged 2.09 ± 2.79

6 324 PHARMACOTHERAPY Volume 23, Number 3, 2003 ng hour/ml under fasting conditions versus 584 ± 146 ng hour/ml under fed conditions. For DHT, the fasting state was associated with C max of ± ng/ml and AUC of 1.86 ± 1.75 ng hour/ml, while the fed state yielded a C max of 3.78 ± 2.90 ng/ml and an AUC of 25.0 ± 16.5 ng hour/ml. No relevant food effects were observed with regard to T max. The sequence of fasting-fed versus fed-fasting had no effect. The results of statistical analyses incorporate the parametric point estimates µ(fed):µ(fasting) with 90% CIs (multiplicative model) derived from ANOVA on log-transformed values (acceptance range ). On the basis of AUC and C max, food had a statistically significant effect on the bioavailability of testosterone, testosterone undecanoate, and DHT. In some subjects, all concentrations throughout the sampling period in the fasting state were below the limit of quantification. In one subject, this occurred for testosterone undecanoate and DHT, and in another subject it occurred for testosterone undecanoate only. Because exclusion of these data would introduce a bias to the data, in that the subjects with the largest effects would be omitted, we substituted one half the limit of quantification for C max and AUC in these subjects (see Materials and Methods). A separate analysis without these substitutions (i.e., without data from these subjects) yielded results that were identical (data not included) to those shown in Table 1. The T max for all three compounds was not affected by food (classic hypothesis testing on ranks using ANOVA). Discussion We investigated the effect of food on the bioavailability of a new formulation of testosterone undecanoate, Andriol Testocaps. When this formulation was taken after an overnight fasting period, serum concentrations of testosterone, testosterone undecanoate, and DHT were low, and in many cases, below the limit of quantification. Thus, little absorption occurred when Andriol Testocaps was administered in the fasting state. Administration of Andriol Testocaps after a standardized breakfast significantly increased serum concentrations of testosterone, testosterone undecanoate, and DHT. Andriol Testocaps were designed to deliver testosterone undecanoate into the systemic circulation by the intestinal lymphatic route, thereby circumventing first-pass inactivation in the liver. 2 Due to its lipophilic nature, the active ingredient (testosterone undecanoate), like many other lipophilic drugs, becomes incorporated into chylomicrons with the lipophilic solvent and dietary lipids. It is transported by the intestinal lymphatic system into the peripheral circulation. 2 4 The improved bioavailability of Andriol Testocaps when administered with food may reflect heightened formation of chylomicrons in response to the increased lipid load of the coadministered food and/or increased solubilization of the compound within the intestinal tract due to digestion of the ingested meal. Prolongation of gastric emptying also may play a role. 2, 3 The enhancing effect of food on the bioavailability of lipophilic drugs that are absorbed by the lymphatic system is well known. This effect has been demonstrated with halofantrine in a dog model. 6 A study with human subjects demonstrated that food consumption increases the absorption of lipophilic drugs, including an experimental preparation of testosterone undecanoate. 7 That study identified peak testosterone levels of about 7 9 ng/ml in subjects who received the drug after a standardized breakfast. Data from that study pertaining to subjects in the fed state are comparable to those observed in our study. However, the study showed peak levels of testosterone in the fasting state that were much higher than those found in our study. We have no explanation for that disparity. Most studies of the effects of food on the bioavailability of testosterone have limited their measurements to serum levels of testosterone. A small number of studies have analyzed serum levels of DHT. Our study expands on this research by showing that coadministration with food increases serum levels of testosterone undecanoate itself. The T max after a meal is about 5 6 hours, a value that is common among drugs absorbed by the lymphatic pathway. Our results show the importance of food to the bioavailability of testosterone undecanoate. We identified significantly increased serum levels of testosterone undecanoate, testosterone, and DHT in subjects who took Andriol Testocaps with a meal, as opposed to those who received the drug in a fasting state. The most common indication for testosterone undecanoate is male hypogonadism. For patients with this condition, the therapeutic goal is to restore serum testosterone to the normal range of approximately 3 10 ng/ml. In this study, a single dose of Andriol Testocaps 80 mg administered to women elevated

7 FOOD AND BIOAVAILABILITY OF ORAL TESTOSTERONE UNDECANOATE Bagchus et al 325 testosterone levels to this normal range, but only if taken with food. Thus, testosterone undecanoate appears to be successful in restoring physiologic testosterone levels if taken with food. The large individual variations and low testosterone levels that occasionally have been observed after administration of testosterone undecanoate might reflect that some subjects have taken this agent in a fasting state. 8 In our study, only one type of food (a standardized continental breakfast) was evaluated; further studies are needed to assess the effects of foods with different compositions and contents on the bioavailability of testosterone undecanoate. Conclusion Our results show that food increases the bioavailability of testosterone undecanoate, testosterone, and DHT. For proper absorption, Andriol Testocaps must be taken with meals. Acknowledgments Elisabeth Dingler, Anja Grüning, and Markus Groß are acknowledged for performing the bioanalysis. The authors thank Anneke J. Jellema and Herman A. M. Verheul for editorial support. References 1. Geurts TBP, Coelingh Bennink HJT. Testosterone replacement therapy: testosterone undecanoate (Andriol). J Urol Urogenakol 2000: special edition; Charman WN, Porter CJH. Lipophilic prodrugs designed for intestinal lymphatic transport. Adv Drug Deliv Rev 1996;19: Noguchi T, Charman WN, Stella VJ. The effect of drug lipophilicity and lipid vehicles on the lymphatic absorption of various testosterone esters. Int J Pharm 1985;24: Horst HJ, Höltje WJ, Dennis M, Coert A, Geelen J, Voigt KD. Lymphatic absorption and metabolism of orally administered testosterone undecanoate in man. Klin Wschr 1976;54: Hirschhäuser C, Hopkinson CRN, Sturm G, Coert A. Testosterone undecanoate: a new orally active androgen. Acta Endocrinol 1975;80: Khoo S-M, Edwards GA, Porter CJH, Charman WN. A conscious dog model for assessing the absorption, enterocytebased metabolism and intestinal lymphatic transport of halofantrine. J Pharm Sci 2001;90: Frey H, Aakvaag A, Saanum D, Falch J. Bioavailability of oral testosterone in males. Eur J Clin Pharmacol 1979;16: Cantril JA, Dewis P, Large DM, Newman M, Anderson DC. Which testosterone replacement therapy? Clin Endocrinol 1984;21: