Pharmacokinetics of 2 Novel Formulations of Modified-Release Oral Testosterone Alone and With Finasteride in Normal Men With Experimental Hypogonadism

Authors

  • Christin N. Snyder,

    1. Center for Research in Reproduction and Contraception, Divisions of General Internal Medicine and Metabolism, Endocrinology and Nutrition, Department of Medicine, University of Washington Medical School, Seattle, Washington
    Search for more papers by this author
  • Richard V. Clark,

    1. GlaxoSmithKline Research and Development, Research Triangle Park, North Carolina
    Search for more papers by this author
  • Ralph B. Caricofe,

    1. GlaxoSmithKline Research and Development, Research Triangle Park, North Carolina
    Search for more papers by this author
  • Mark A. Bush,

    1. GlaxoSmithKline Research and Development, Research Triangle Park, North Carolina
    Search for more papers by this author
  • Mara Y. Roth,

    1. Center for Research in Reproduction and Contraception, Divisions of General Internal Medicine and Metabolism, Endocrinology and Nutrition, Department of Medicine, University of Washington Medical School, Seattle, Washington
    Search for more papers by this author
  • Stephanie T. Page,

    1. Center for Research in Reproduction and Contraception, Divisions of General Internal Medicine and Metabolism, Endocrinology and Nutrition, Department of Medicine, University of Washington Medical School, Seattle, Washington
    Search for more papers by this author
  • William J. Bremner,

    1. Center for Research in Reproduction and Contraception, Divisions of General Internal Medicine and Metabolism, Endocrinology and Nutrition, Department of Medicine, University of Washington Medical School, Seattle, Washington
    Search for more papers by this author
  • Dr John K. Amory

    Corresponding author
    1. Center for Research in Reproduction and Contraception, Divisions of General Internal Medicine and Metabolism, Endocrinology and Nutrition, Department of Medicine, University of Washington Medical School, Seattle, Washington
    Search for more papers by this author

University of Washington, Box 356429, 1959 NE Pacific Street, Seattle, WA 98195 (e-mail: jamory@u.washington.edu).

Abstract

ABSTRACT: Oral administration of testosterone might be useful for the treatment of testosterone deficiency. However, current “immediate-release” formulations of oral testosterone exhibit suboptimal pharmacokinetics, with supraphysiologic peaks of testosterone and its metabolite, dihydrotestosterone (DHT), immediately after dosing. To dampen these peaks, we have developed 2 novel modified-release formulations of oral testosterone designed to slow absorption from the gut and improve hormone delivery. We studied these testosterone formulations in 16 normal young men enrolled in a 2-arm, open-label clinical trial. Three hundred-mg and 600-mg doses of immediate-release and modified fast-release or slow-release formulations were administered sequentially to 8 normal men rendered hypogonadal by the administration of the gonadotropin-releasing hormone antagonist acyline. Blood for measurement of serum testosterone, DHT, and estradiol was obtained before and 0.5, 1, 2, 3, 4, 6, 8, 12, and 24 hours after each dose. A second group of 8 men was studied with the coadministration of 1 mg of the 5α-reductase inhibitor finasteride daily throughout the treatment period. Serum testosterone was increased with all formulations of oral testosterone. The modified slow-release formulation significantly delayed the postdose peaks of serum testosterone and reduced peak concentrations of serum DHT compared with the immediate-release formulation. The addition of finasteride further increased serum testosterone and decreased serum DHT. We conclude that the oral modified slow-release testosterone formulation exhibits superior pharmacokinetics compared with immediate-release oral testosterone both alone and in combination with finasteride. This formulation might have efficacy for the treatment of testosterone deficiency.

Testosterone is the most important male sex hormone and is crucial for male health and development. Approximately 6% to 10% of men, depending on age, have low testosterone concentrations and symptoms of testosterone deficiency, including low libido, erectile dysfunction, osteoporosis, sleep disturbance, depression, lethargy, and diminished physical performance (Araujo et al, 2007). These men benefit from testosterone replacement, which improves mood, energy, and sense of well-being; increases bone and muscle mass; and maintains sexual function (Katznelson et al, 1996; Wang et al, 1996; Snyder et al, 2000).

Current testosterone therapies approved for use in the United States include intramuscular injections, transdermal patches and gels, subdermal pellets, buccal tablets, and oral alkylated testosterone derivatives (eg, methyltestosterone and oxandrolone). Each of these formulations has drawbacks. Injections must be given intramuscularly every 1 to 3 weeks and can be painful (Fossa et al, 1999). Patches cause skin reactions in more than one-half of patients using them (Dobs et al, 1999). Testosterone gels are safe and effective (Wang et al, 2000) but are expensive and must be applied to a fairly large area of skin. In addition, gels have a risk of inadvertently exposing women and children to testosterone (Brachet et al, 2005; de Ronde, 2009), which has recently resulted in a “black box” warning regarding their use from the U.S. Food and Drug Administration.

Oral administration of testosterone might be preferable to currently available options. Unfortunately, the currently available oral testosterone formulations, all of which are alkylated at the 17-carbon position, are associated with an unacceptably high rate of liver toxicity, including cholestatic jaundice, peliosis hepatis, and even liver tumors in one-third to one-half of long-term users (Westaby et al, 1977; Turani et al, 1983; Lowdell and Murray-Lyon, 1985; Pavlatos et al, 2001). As a result, these oral androgens are not considered safe for the long-term treatment of male hypogonadism.

Oral administration of nonalkylated testosterone was previously thought to be ineffective because of rapid hepatic metabolism (Foss, 1939; Daggett et al, 1978). Nevertheless, we recently reported that when nonalkylated testosterone was administered at a sufficient dose, serum testosterone levels within the normal range were achieved without adverse effects on liver function (Amory and Bremner, 2005; Amory et al, 2006). In addition, we demonstrated that when oral testosterone was combined with a 5α-reductase inhibitor, which inhibits the conversion of testosterone to dihydrotestosterone (DHT), the resulting serum testosterone concentrations were roughly doubled and serum concentrations of DHT were reduced. Combining 5α-reductase inhibition with testosterone therapy is potentially attractive because DHT concentrations are elevated above the normal range after administration of oral testosterone formulations such as oral testosterone undecanoate (Neischlag et al, 1975; Franchimont et al, 1978), and DHT has been implicated in the pathophysiology of androgenic alopecia, acne, benign prostatic hyperplasia, and possibly prostate cancer. Therefore, because current “immediate-release” formulations of oral testosterone exhibit suboptimal pharmacokinetics, with supraphysiologic peaks of testosterone and DHT immediately after dosing (Page et al, 2008), we developed 2 novel modified-release formulations of oral testosterone designed to slow absorption from the gut and improve hormone delivery. In this study, we sought to determine the pharmacokinetics of these novel formulations of oral testosterone alone and in combination with the 5α-reductase inhibitor finasteride in normal men with experimentally induced hypogonadism.

Materials and Methods

Subjects

Sixteen men, 18 to 52 years of age, in good health, were recruited through local newspapers and campus flyers. After informed consent was obtained, subjects underwent screening procedures consisting of a medical history, a physical examination, measurements of serum hormone levels, and routine laboratory tests, including complete blood counts, serum chemistries, liver function tests, and prostate-specific antigen test. Specific exclusion criteria included current use of testosterone, infertility, poor general health, clinically significant abnormal laboratory results, history of testicular disease or severe testicular trauma, major psychiatric disorders, use of illicit drugs or use of more than 3 alcoholic beverages daily, participation in a hormonal drug study within the last month, history of bleeding disorders or use of anticoagulants, or use of medications known to interfere with testosterone metabolism, including finasteride and dutasteride. Participants were compensated $750 for study participation.

Study Design

We conducted an open-label, 2-arm pharmacokinetic study of 2 novel formulations of oral testosterone, compared with an “immediate-release” formulation described previously (Page et al, 2008). The study was conducted in normal men whose endogenous testosterone production was suppressed by the administration of 300 μg/kg of the potent gonadotropin-releasing hormone antagonist acyline, shown previously to suppress testosterone to castrate levels within 24 hours, an effect lasting for 2 weeks in normal healthy men (Herbst et al, 2004). The first 8 men were enrolled in the testosterone-only arm of the study, and an additional 8 men were enrolled in the testosterone plus finasteride arm. All doses of oral testosterone were given after fasting for 8 hours, and no food was given for 4 hours after dosing. The immediate-release oral testosterone and 2 novel oral testosterone formulations, modified fast release and slow release, were prepared by GlaxoSmithKline (Research Triangle Park, North Carolina). For the 300-mg testosterone dose, 1 tablet was administered orally; for the 600-mg dose, 2 tablets were administered. Finasteride tablets (1 mg; Propecia, Merck & Co Inc, Whitehouse Station, New Jersey) were purchased commercially. Acyline was obtained from NeoMPS (San Diego, California). After administration of acyline on day 1, subjects were given 300-mg immediate-release, modified fast-release, and modified slow-release oral testosterone on days 2, 3, and 4, respectively. Participants were then given 600 mg of these same formulations on days 8, 9, and 11, respectively. A 1-day break was taken between 600 mg of the modified fast- and slow-release formulations to allow for washout of residual testosterone from the previous day (Figure 1). On each day of drug dosing, subjects had blood drawn prior to dosing and 0.5, 1, 2, 3, 4, 6, 8, 12, and 24 hours after dosing for measurement of serum testosterone, DHT, and estradiol. Participants in the testosterone plus finasteride arm self-administered 1 mg of finasteride by mouth daily beginning 3 days prior to the acyline injection and continued finasteride through the last day of oral testosterone administration. Finasteride and oral testosterone were taken at the same time. For safety monitoring, blood counts, serum chemistries, and liver function tests were measured before and after each dosing day. Subjects were queried by study personnel about the presence of adverse events daily during testosterone exposure. Any complaint was documented in the study chart. The University of Washington Investigational Review Board approved all aspects of this study before study initiation. This study was registered in advance at ClinicalTrials.gov as study NCT00663793.

Figure 1.

. Study design.

Measurements

Serum total testosterone was measured by a radioimmunoassay (SiemensUSA, Los Angeles, California). The assay has a sensitivity of 0.35 nmol/L; interassay variations for low, mid, and high pools of 9.9%, 8.8%, and 9.5%; and intra-assay variations of 4.4%, 5.1%, and 6%, respectively. The normal range was 6.8 to 29 nmol/L. Serum estradiol was measured using a radioimmunoassay (SiemensUSA). The assay has a sensitivity of 18 pmol/L. The interassay variations for the low- and medium-range values were 17% and 5.4%, respectively. The normal range for estradiol in men is 27 to 132 pmol/L. DHT was measured using high-performance liquid chromatography–mass spectroscopy as described previously (Kalhorn et al, 2007). Interassay coefficients of variation for low-, mid-, and high-range samples were 13.2%, 2.7%, and 5.4% for DHT. The normal range for serum DHT is 1.0 to 3.1 nmol/L. Blood counts, serum chemistries, and liver function tests were measured by the University of Washington Hospital clinical laboratory.

Statistical Analysis

Baseline characteristics were compared between groups using a 2-sample t test. Because of nonnormality, hormone concentrations among the 3 formulations and between the testosterone-only and the testosterone plus finasteride groups were compared using a Wilcoxon sign-rank and Wilcoxon rank-sum test, respectively. For the pharmacokinetic analyses, maximum concentration after dosing (Cmax), time to maximum concentration (Tmax), and area under the curve (AUC) were calculated for each subject using a computer program (PK Solutions, Golden, Colorado). Pharmacokinetic parameters among the 3 formulations and between the testosterone-only and testosterone plus finasteride groups were compared using the Wilcoxon sign-rank and Wilcoxon rank-sum test. No corrections were made for multiple comparisons. Statistical analyses were performed using STATA (College Park, Texas). For all comparisons, α = .05 was considered significant.

Results

Subjects

Sixteen men were enrolled, completed all aspects of the study, and were included in the analysis. The baseline characteristics of the study participants are displayed in Table 1. Baseline serum estradiol was significantly greater in the testosterone plus finasteride group compared with the testosterone-only group (P = .048), and the participants in the testosterone plus finasteride group were significantly younger than those in the testosterone-only group (P = .02). One participant missed the day 11 blood draws owing to an arm fracture he sustained while bicycling to the study visit. Seven participants experienced transient headaches, and 5 men complained of mild hot flashes. Four of these participants were given a single injection of 200 mg of testosterone enanthate after the last blood draw, with marked improvement in their symptoms. One subject had a mildly increased aspartate aminotransferase level to 2 times the upper limit of normal on study day 8 after an episode of increased alcohol intake. However, this value did not meet the prespecified criteria of 3 times the upper limit of normal for study discontinuation. Therefore, the participant continued in the study. His aspartate aminotransferase level returned to normal over the next 2 days despite continued oral testosterone administration. There were no other significant laboratory abnormalities in any participant. Specifically, the aspartate aminotransferase level was 28.2 ± 14 U/L on day 1 and 34.3 ± 15 U/L the day following the last dose of oral testosterone (P = .26), and the alanine aminotransferase level was 24.6 ± 5 U/L on day 1 and 27.9 ± 6 U/L the day following the last dose of oral testosterone (P = .09). Additionally, there were no adverse gastrointestinal symptoms associated with the oral testosterone formulations. At the study completion visit, 1 participant in the testosterone plus finasteride group disclosed that he had human immunodeficiency virus and was taking highly active antiretroviral therapy (HAART), which would have excluded him from the study. However, the inclusion of this participant in the final analysis did not impact interpretation of study results; therefore, his data were included in the final analysis.

Table 1. . Screening characteristics of study participants (n = 8/group)a
 Testosterone OnlyTestosterone Plus Finasteride
  1. Abbreviation: BMI, body mass index.

  2. a All values are means ± SD.

  3. b P < .05 compared with testosterone-only group.

Age, y44 ± 3.334 ± 10b
Weight, kg83 ± 1889 ± 22
Height, cm178 ± 5179 ± 7
BMI, kg/m226 ± 428 ± 8
Testosterone, nmol/L14.5 ± 2.516.8 ± 8.9
Sex hormone–binding globulin, nmol/L35.5 ± 12.832.5 ± 19.6
Estradiol, pmol/L70.1 ± 16.1114 ± 54b

Serum Hormones

Twenty-four hours after acyline administration, serum testosterone concentrations in all participants were markedly suppressed (testosterone only, 2.0 ± 1.3 nmol/L; testosterone plus finasteride, 2.3 ± 1.4 nmol/L). In the testosterone-only group, administration of 300 mg of all 3 oral testosterone formulations increased the mean serum testosterone concentrations to the normal range within 30 minutes of dosing (Figure 2A). At 30 minutes, the mean serum testosterone concentration of the immediate-release formulation was significantly greater than that of the slow-release formulation (immediate-release testosterone, 22 ± 10 nmol/L; slow-release testosterone, 12 ± 7.2 nmol/L; P = .03). Mean serum testosterone concentrations were below the lower limit of the normal range by 10 hours after dosing with all 3 formulations. Similarly, in the testosterone plus finasteride group, all 3 formulations increased mean serum testosterone concentrations above the lower limit of the normal range by 30 minutes after dosing (Figure 2B). The mean serum testosterone concentrations with the immediate-release testosterone formulation were significantly greater than those for the slow-release formulation 30 minutes (immediate release, 32 ± 27 nmol/L vs 12 ± 7.9 nmol/L; P = .01) and 60 minutes after dosing (immediate release, 29 ± 23 nmol/L vs 16 ± 12 nmol/L; P = .03). Mean serum testosterone concentrations were below the lower limit of the normal range by 12 hours after dosing with all 3 formulations given in combination with finasteride.

Figure 2.

. Serum testosterone 300 mg (A and B) and 600 mg (C and D) after dosing with the immediate-release, modified fast-release, and modified slow-release formulations of oral testosterone in normal men rendered experimentally hypogonadal with acyline in the testosterone-only (A and C) and testosterone plus finasteride (B and D) groups (n = 8/group). The dotted lines represent the upper and lower limits of the normal range. All values are means ± SEM. * indicates P < .05 compared with both modified fast release and modified slow release; Φ, P < .05 compared with modified fast release; Ψ, P < .05 compared with modified slow release.

With the 600-mg dose of testosterone alone, the immediate-release and fast-release formulations both resulted in supraphysiologic serum testosterone peaks (Figure 2C), as did all 3 formulations when combined with finasteride (Figure 2D). In the testosterone plus finasteride group, the 600-mg doses for all 3 formulations resulted in supraphysiologic serum testosterone concentrations for up to 6 hours after administration.

Serum DHT concentrations for all participants in both groups were also below the lower limit of the normal range 24 hours after the acyline injection (testosterone only, 0.6 ± 0.32 nmol/L; testosterone plus finasteride, 0.54 ± 0.32 nmol/L). In the testosterone-only group at 300 mg, the mean serum DHT level 2 hours after dosing for the fast-release formulation was significantly greater than that of the slow-release formulation (6.1 ± 5.2 nmol/L vs 3.0 ± 1.4 nmol/L, respectively; P = .03), and the mean serum DHT level was greater than the upper limit of the normal range for the immediate-release and fast-release formulations (Figure 3A). In contrast, in the testosterone plus finasteride group, the mean serum DHT concentrations for all 3 formulations stayed within the normal range for the first 8 hours after dosing and fell slightly below the lower limit of the normal range after 12 hours (Figure 3B).

Figure 3.

. Serum dihydrotestosterone (DHT) after dosing with 300 mg of the immediate-release, modified fast-release, and modified slow-release formulations of oral testosterone in normal men rendered experimentally hypogonadal with acyline in the testosterone-only (A) and testosterone plus finasteride (B) groups (n = 8/group). The dotted lines represent the upper and lower limits of the normal range. All values are means ± SEM. Ψ indicates P < .05 compared with modified slow release.

In contrast to serum testosterone and serum DHT, the mean serum estradiol concentrations were not suppressed below the lower limit of the normal range 24 hours after acyline administration (Figure 4A and B). After dosing with either 300-mg or 600-mg oral testosterone, the mean serum estradiol concentration transiently exceeded the normal range for all 3 formulations, with or without finasteride. This effect appeared more pronounced for the testosterone plus finasteride group, with estradiol levels above the normal range for 6 to 8 hours.

Figure 4.

. Serum estradiol levels after dosing with 300 mg of the immediate-release, modified fast-release, and modified slow-release formulations of oral testosterone in normal men rendered experimentally hypogonadal with acyline in the testosterone-only (A) and testosterone plus finasteride (B) groups (n = 8/group). The dotted lines represent the upper and lower limits of the normal range. All values are means ± SEM.

Hormone Pharmacokinetics

Because of the supraphysiologic concentrations of serum hormones observed with the 600-mg doses of oral testosterone, the pharmacokinetic analyses focused on comparison of the 300-mg doses for both the testosterone-only and testosterone plus finasteride groups. In the testosterone-only group, there were no statistically significant differences in the Cmax of serum testosterone or the AUC0–24 among the 3 formulations (Table 2). The Tmax of the slow-release formulation was significantly increased compared with that of the immediate-release testosterone formulation (2.5 ± 0.8 hours vs 1.2 ± 1.1 hours; P = .02). In the testosterone plus finasteride group, the high Cmax for the fast-release formulation was due to 2 participants who had markedly elevated serum testosterone concentrations of 349 nmol/L and 284 nmol/L 1 hour after dosing. The Cmax for the remaining 6 participants ranged from 21 to 67 nmol/L. As in the testosterone-only group, the Tmax for the slow-release formulation was significantly increased compared with that of the immediate-release testosterone formulation (2.9 ± 1.0 hours vs 1.3 ± 1.1 hours; P = .008). The AUC0–24 of the modified fast-release formulation was significantly greater in the testosterone plus finasteride group compared with that of the testosterone-only group (P < .05). There were no statistically significant differences in DHT pharmacokinetic parameters among the 3 formulations within a group (Table 2). When the testosterone-only and testosterone plus finasteride groups were compared, the fast-release testosterone plus finasteride group had a significantly lower DHT Cmax compared with that of the fast-release testosterone-only group (2.1 ± 1.2 nmol/L vs 6.2 ± 4.8 nmol/L; P = .04). The estradiol pharmacokinetic analyses are presented in Table 2. There were no significant differences between the testosterone-only and testosterone plus finasteride groups except for a delayed Tmax for slow-release testosterone plus finasteride compared with immediate-release testosterone (3.1 ± 1.6 hours vs 1.6 ± 0.9 hours; P ≤ .05).

Table 2. . Hormone pharmacokinetics with 300-mg oral testosterone administration alone or with 1-mg finasteride daily (n = 8/group)a
 Testosterone OnlyTestosterone Plus Finasteride
 Cmax, nmol/LbTmax, hAUC0–24, nmol•h/LbCmax, nmol/LbTmax, hAUC0–24, nmol•h/Lb
  1. Abbreviations: Cmax, maximum concentration; Tmax, time of maximum concentration; AUC0–24, area under the curve from baseline through 24 hours postdosing.

  2. a All values are means ± SD. The normal range of serum testosterone was 6.8–29 nmol/L.

  3. b The Cmax and AUC0–24 for estradiol were measured in pmol/L and pmol•h/L, respectively.

  4. c P < .05 compared with testosterone only.

  5. d P < .05 compared with immediate-release testosterone.

Testosterone      
    Immediate release23.6 ± 9.11.2 ± 1.1143 ± 4737.6 ± 24.71.3 ± 1.1198 ± 86
    Modified fast release29.9 ± 14.61.9 ± 0.6144 ± 44109 ± 1302.2 ± 1.5384 ± 273c
    Modified slow release20.8 ± 4.42.5 ± 0.8d162 ± 6037 ± 322.9 ± 1.0d237 ± 141
Dihydrotestosterone      
    Immediate release4.0 ± 2.12.2 ± 0.736 ± 132.2 ± 1.32.5 ± 0.923 ± 10
    Modified fast release6.2 ± 4.82.2 ± 0.742 ± 212.1 ± 1.2c2.5 ± 0.926 ± 12
    Modified slow release3.3 ± 1.62.5 ± 0.939 ± 161.9 ± 13.5 ± 0.926 ± 11
Estradiol      
    Immediate release171 ± 632.4 ± 0.71812 ± 405278 ± 711.6 ± 0.92241 ± 783
    Modified fast release196 ± 722.3 ± 0.71961 ± 528262 ± 1072 ± 0.82002 ± 861
    Modified slow release149 ± 463.3 ± 2.11944 ± 652210 ± 783.1 ± 1.6d3129 ± 1574

Discussion

In this work, we demonstrated that the pharmacokinetics of orally dosed testosterone can be improved by the modification of the testosterone tablet. Administration of 2 types of these modified-release formulations, fast and slow release, of oral testosterone without finasteride increased the concentrations of serum testosterone for 8 to 10 hours in healthy men with experimentally induced hypogonadism. Importantly, in contrast to the 300-mg doses of the immediate-release and fast-release formulations, administration of the slow-release formulation exhibited peak concentrations of hormones that did not exceed the upper limit of the normal range. In addition, the time of the peak hormone concentrations observed with this slow-release formulation was significantly delayed compared with the other formulations, peaking between 2 to 4 hours, rather than 30 to 60 minutes after dosing. Furthermore, the combination of slow-release oral testosterone with finasteride blunted the Cmax for serum testosterone and maintained the mean serum testosterone in the normal range for 10 to 12 hours. This pharmacokinetic profile is very encouraging for the development of a twice-daily oral testosterone preparation for the treatment of men with testosterone deficiency. As a result, this oral testosterone formulation will be the subject of future studies of oral testosterone therapy in hypogonadal men.

A second issue in terms of the development of a safe, effective form of oral testosterone therapy for testosterone-deficient men is the elevations in serum DHT observed with other forms of oral testosterone therapy such as oral testosterone undecanoate (Nieschlag et al, 1975; Franchimont et al, 1978). Indeed, both the immediate-release and fast-release formulations in the absence of finasteride had supraphysiologic peaks in DHT levels 2 hours after dosing, although the maximum mean DHT level observed with the slow-release formulation remained in the normal range. All DHT values were dramatically reduced by the coadministration of the 5α-reductase inhibitor finasteride. However, in contrast to previous studies in which the addition of dutasteride, a more potent 5α-reductase inhibitor, resulted in DHT concentrations below the normal range (Page et al, 2008), the addition of 1 mg of finasteride daily maintained serum DHT concentrations within the normal range during most of the dosing interval. This is relevant in that supraphysiologic levels of DHT may have adverse effects on the prostate, potentially increasing the risk of prostatic hyperplasia or prostate cancer. A combination of oral testosterone with a 5α-reductase inhibitor may also result in lower serum DHT concentrations than other forms of testosterone therapy such as patches and gels, which are associated with slightly increased concentrations of DHT presumably because of the presence of 5α-reductase in the skin (Swerdloff et al, 2000). Slightly reduced concentrations of DHT might have beneficial effects on DHT-dependent conditions, such as benign prostatic hyperplasia and androgenic alopecia. Indeed, there may be benefit to the use of 5α-reductase inhibitors in preventing prostate cancer (Thompson et al, 2003).

All formulations appeared to increase the serum estradiol concentrations above the upper limit of the normal range after administration. Estradiol concentrations in the testosterone plus finasteride group appeared to be somewhat greater than those in the testosterone-only group. However, this is likely due to the significantly increased baseline estradiol levels observed in the former group, rather than any effect of finasteride on estradiol synthesis, because chronic administration of finasteride to men has no apparent effect on serum estradiol concentrations (Amory et al, 2007). In any case, future studies of this approach to testosterone therapy will need to carefully measure serum estradiol levels to ensure that they are not inordinately elevated after oral administration. If elevations in serum estradiol are present, participants will need to be closely followed for potential estrogen-related side effects such as gynecomastia.

There was substantial intrasubject and intersubject variation in the measured serum testosterone concentrations. Therefore, future studies of the multidose pharmacokinetics of the modified slow-release formulation of oral testosterone will require larger sample sizes to fully understand the range of response to this method of androgen therapy. It seems likely that individualized dose titration may be required with oral testosterone administration to ensure that therapeutic serum concentrations are achieved in a given patient. In addition, the participants fasted prior to dosing; therefore, the effect of food on the absorption of this formulation is uncertain. However, prior work in this area has not shown a significant effect of food on absorption of oral testosterone (Amory et al, 2006), in contrast to oral testosterone undecanoate for which absorption is dependent on concurrent ingestion of a fatty meal and is minimal in the fasting state (Bagchus et al, 2003; Schnabel et al, 2007). Lastly, 5 of the participants were taking concomitant medications for common conditions, although none of these medications were inducers or inhibitors of the cytochrome P450 isoform (3A4) involved in testosterone metabolism. It is possible that certain commonly prescribed medications, such as calcium channel blockers and most 3-hydroxy-3-methyl-glutaryl-CoA reductase inhibitors, might induce or inhibit the metabolism of oral testosterone. Such potential drug interactions will be the subject of future research.

All formulations were well tolerated and free of gastrointestinal side effects. Moreover, it is important to note that there were no significant alterations in liver or kidney function in this study requiring discontinuation. One participant had an elevated aspartate aminotransferase level, which was likely the result of mild alcoholic hepatitis after an episode of heavy alcohol intake. The fact that his hepatocellular injury marker, elevated level of aspartate aminotransferase, resolved despite continued testosterone exposure implies that it was not likely secondary to the oral testosterone. The observation that these formulations of nonalkylated oral testosterone were not associated with liver toxicity is consistent with prior reports using oral testosterone for extended periods of time without liver inflammation (Johnsen et al, 1974; Johnsen, 1978). Indeed, we are unaware of any instance of liver toxicity following the administration of nonalkylated forms of oral testosterone. The majority of the other adverse events were related to the profound hypogonadism induced by the acyline. Because the serum testosterone concentrations were below normal by 10 hours after testosterone administration and testosterone was only administered once daily on 6 days of 12 study days, these participants were hypogonadal for a significant amount of time. In future studies, participants will receive twice-daily dosing of the oral testosterone formulation to maintain adequate concentrations of testosterone throughout the day.

This study had some limitations. It would have been preferable if participants had been randomized to the different groups; however, the decision to proceed with the second group was made only after the results from the first group were reviewed. This resulted in significant differences in age and baseline estradiol levels between the groups, which may lead to significant differences in steroid metabolism. Nevertheless, the study did demonstrate the superior pharmacokinetics of the modified slow-release testosterone formulation and the ability of the combination of finasteride and oral testosterone to normalize serum testosterone and DHT concentrations.

In conclusion, we have shown that oral administration of 2 novel formulations of modified-release testosterone results in significantly elevated testosterone concentrations in healthy men with medically induced hypogonadism. Furthermore, the modified slow-release formulation appears to prevent the transient postdose supraphysiologic peaks in serum hormone levels observed with the other formulations by slowing hormone absorption. This formulation appears to be attractive for further development as an oral form of testosterone and will be the focus of future studies investigating the safety and efficacy of this approach to androgen replacement therapy in hypogonadal men.

Acknowledgment

The authors would like to thank Ms Marilyn Busher, Ms Kathy Winter, Ms Iris Nielsen, Ms Kathryn Torrez Duncan, Ms Constance Pete, Ms Dorothy McGuinness, and Mr Robert Bale for assistance with the clinical aspects of this study and Dr David W. Amory for critical review of the manuscript.

Footnotes

  1. This work was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development, a division of the National Institutes of Health (NIH), through cooperative agreement U54 HD42454 as part of the Cooperative Contraceptive Research Centers Program and GlaxoSmithKline. A portion of this work was conducted through the Clinical Research Center facility at the University of Washington and supported by NIH grant UL1-RR-025014.

Ancillary