| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Clinical Pharmacology (R.C., M.T.), Pfizer Inc., La Jolla, California 92121; Queen Elizabeth Hospital (P.M.S.), University of Birmingham, Birmingham B15 2TT, United Kingdom; Clinical Research (M.-N.N.), Pfizer Inc., B-1050 Brussels, Belgium; and Cardiovascular and Metabolic Diseases (R.A.C., B.H.), Pfizer Inc., Groton, Connecticut 06340
Address all correspondence and requests for reprints to: Boaz Hirshberg, M.D., AstraZeneca, Room C4C-717 A, 1800 Concord Pike, P.O. Box 15437, Wilmington, Delaware 19850–5437. E-mail: Boaz.Hirshberg{at}AstraZeneca.com; or Rachel Courtney, Ph.D., Pfizer Global Research & Development, 10555 Science Center Drive (CB10; Room 2446), San Diego, California 92121. E-mail: Rachel.Courtney{at}pfizer.com.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: Demonstrating inhibition of 11βHSD1 is challenging because there is no accessible way to directly assess the enzyme activity in vivo. Thus, it was proposed to assess the enzyme activity, in an indirect fashion, using two biomarker methods: the prednisolone generation study (conversion of oral prednisone to prednisolone in plasma) and the ratio of cortisol and cortisone metabolites in urine.
Design: This was a phase 1, double-blind, placebo-controlled, randomized, multiple-dose study.
Setting: The study was conducted in a clinical research unit.
Participants: Sixty healthy adult volunteers participated in the study.
Intervention: Oral doses of PF-00915275 (0.3–15 mg) and prednisone (10 mg) were administered during the study.
Main Outcome Measures: Safety, tolerability, pharmacokinetics, and pharmacodynamics of PF-00915275, a selective 11βHSD1 inhibitor, were measured.
Results: Overall, multiple oral doses of PF-00915275 were safe and well tolerated. After oral administration, PF-00915275 was rapidly absorbed, slowly eliminated, and generally displayed dose-proportional increases in exposure. At the 15-mg dose, mean exposure to prednisolone was reduced by 37%, and there was a dose-dependent fall in the 5
-tetrahydrocortisol + 5β-tetrahydrocortisol to tetrahydrocortisone ratio with maximum inhibition of 26% after 14 d. The urinary free cortisol to urinary free cortisone ratio, an indicator of 11βHSD2 inhibition, did not change.
Conclusion: PF-00915275 was safe at all doses tested. The results of the prednisolone generation test and the urinary metabolite ratios confirm that PF-00915275 is a selective 11βHSD1 inhibitor.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In humans, elevated cortisol levels are associated with Cushing’s syndrome, with patients frequently presenting with hyperglycemia, visceral obesity, and hypertension as well as osteoporosis and depression (6). Normalizing the cortisol levels usually reverses these symptoms and signs (6). At a prereceptor level, 11β hydroxysteroid dehydrogenase type 1 (11βHSD1) catalyzes the intracrine conversion of inactive cortisone to the active glucocorticoid, cortisol. Because 11βHSD1 is highly expressed in the liver relative to other tissues, inhibition of 11βHSD1 activity provides an opportunity to reduce glucocorticoid levels specifically in the liver and splanchnic circulation (7). The 11βHSD1 knockout mouse is protected from hyperglycemia associated with stress or obesity through reduced hepatic expression of phosphoenol-pyruvate carboxy kinase, which controls the rate-limiting step of gluconeogenesis as well as other glucose mobilizing enzymes such as glucose-6-phosphatase (8, 9). Administration of a selective and potent 11βHSD1 inhibitor lowers body weight, insulin, fasting glucose, triglycerides, and cholesterol in diet-induced obese mice and lowers fasting glucose, insulin, glucagon, triglycerides, and free fatty acids, as well as improving glucose tolerance, in murine diet-induced obesity and high fat/streptozotocin models of type 2 diabetes (10). Finally, the nonselective 11βHSD inhibitor carbenoxolone increases hepatic insulin sensitivity in humans (11, 12). Carbenoxolone is the only 11βHSD inhibitor (published to date) that has been administered in the clinic; however, it was found not to be clinically viable due to adverse effects associated with the inhibition of 11βHSD2. Inhibition of 11βHSD2 results in cortisol-dependent mineralocorticoid excess with hypertension and hypokalemic alkalosis (11).
Compared with current antidiabetic agents on the market, an 11βHSD1 inhibitor may ameliorate a number of comorbidities that occur in diabetic patients including hypertension (13), dyslipidemia, and visceral adiposity and therefore could be used to treat aspects of the metabolic syndrome (8, 9).
Measuring the inhibition of 11βHSD1 in vivo is a challenge because there is no easily accessible way to directly assess enzyme activity in the liver. Thus, it was proposed to assess the enzyme activity, in an indirect fashion, by administration of an exogenous synthetic steroid such as prednisone and assessing the conversion of prednisone to prednisolone, a step that is solely catalyzed by the oxoreductase activity of 11βHSD1. A second method of assessing 11βHSD1 is by measuring the ratio of the tetrahydrometabolites of cortisol and cortisone in the urine (13).
The aim of this phase 1b study was to evaluate the safety, tolerability, pharmacokinetics, and pharmacodynamics (target inhibition) of a selective 11βHSD1 inhibitor, PF-00915275, in normal healthy volunteers. PF-00915275 does not affect steroidogenesis and has no affinity for the glucocorticoid receptor. In a human kidney cell line (HEK293-LUC), into which the human 11βHSD1 gene was stably transfected, PF-00915275 was found to be an inhibitor of 11βHSD1 with an EC50 of 15 nM (5.25 ng/ml). In HEK293 cells stably transfected with 11βHSD2, PF-00915275 had no inhibitory effect on 11βHSD2 (only 1.5% inhibition when tested at 10 µM).
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Healthy men and women between the ages of 18 and 55 yr were eligible for inclusion into the study. Subjects were required to have a body mass index (BMI) of 18–30 kg/m2 and weight of 50 kg or more. Subjects with a history of febrile illness within 5 d before the first dose or any condition that could affect drug absorption and women of child-bearing potential were excluded from the study.
The study was approved by the Ethics Committee of the Erasme Hospital, and all subjects provided written informed consent to participate. The study was conducted at the Pfizer Clinical Research Unit (Brussels) in accordance with the protocol, International Conference on Harmonization Good Clinical Practice guidelines, and applicable local regulatory requirements and laws.
Study design
This was a randomized, placebo-controlled, double-blind (sponsor open), multiple-dose study in healthy volunteers (Fig. 1
). Subjects who met the entry criteria were confined to the clinical research unit at least 48 h before dosing (d –2) through d 17. Sixty subjects were enrolled into five parallel sequentially escalating dosing cohorts (12 subjects/cohort) evaluating the following doses: 0.3, 1, 3, 10, and 15 mg. Subjects were randomized to receive PF-00915275 or placebo (9:3) once daily (QD) for 14 d (d 1–14). In addition to study medication, subjects in all five cohorts were administered oral prednisone (10 mg tablet; Sanofi-Aventis, Paris, France) at 0800 h on d –1 and at the same time as PF-00915275 on d 11 to assess the ability of PF-00915275 to inhibit 11βHSD1.
|
Venous blood for pharmacokinetic sampling of PF-00915275 was collected into potassium EDTA-tubes on d 1 and 14 within 15 min before the dose (0 h) and at 1, 2, 4, 6, 8, 10, 12, 16, and 24 h after the dose. Supplementary pharmacokinetic samples were collected before the dose on d 4, 7, 9, 10, 11, 12, 13, and after d 14 dosing at 48, 72, 96, 120, and 144 h as well as at the follow-up visit (12–14 d after the last dose). Additional venous blood for pharmacokinetic sampling of prednisone and prednisolone was collected into heparin tubes on d –1 and 11 before the prednisone dose (0 h) and at 20, 40, 60, 80, and 100 min and 2, 2.5, 3, 3.5, and 4 h after the dose. All samples were centrifuged at a minimum of 1000 x g for approximately 10 min at 4 C within 1 h of collection and stored at –20 C or lower until assayed. The plasma samples for PF-00915275, prednisone, and prednisolone were analyzed using validated liquid chromatographic-tandem mass spectrometric methods with lower limit of quantitations of 0.1, 0.25, and 1.0 ng/ml, respectively.
Urine samples were collected for urine pharmacokinetic analysis and to evaluate urinary free cortisol [UFF], urinary free cortisone [UFE], and the A-ring reduced metabolites of cortisol and cortisone [5-tetrahydrocortisol (5THF), 5β-tetrahydrocortisol (5βTHF), and tetrahydrocortisone (THE)]. Urine samples were collected over the 24-h interval before d 1 dosing for pharmacokinetic and pharmacodynamic (PD) analysis. On d 1 and 14, urine was collected from 0–12 and 12–24 h after the dose (0–24 h urine was pooled for PD analysis). On d 7, urine was collected from 0–24 h after the dose for PD analysis. All urine collected was refrigerated (2–8 C) throughout the collection intervals and the aliquots stored frozen at –70 C or lower until analyzed. The urine samples for PF-00915275 and UFF, UFE, 5THF, 5βTHF, and THE were analyzed using validated liquid chromatographic-tandem mass spectrometric methods with lower limit of quantitations of 0.1 and 2, 2, 20, 20, and 20 ng/ml, respectively. Blood samples for the measurement of testosterone, dehydroepiandrosterone sulfate (DHEA-s) and 4-androstenedione were collected on d 1, 4, 9, 14, and 17 and for ACTH at 0600, 0800, 1000, and 1200 h on d 0 and 14 all as part of the safety panel.
Pharmacokinetic analyses
Pharmacokinetic parameter estimates were determined from plasma concentration vs. time data using noncompartmental analyses (WinNonLin, version 4.1; Pharsight Inc., Mountain View, CA). For the PF-00915275 pharmacokinetic parameters, the terminal phase elimination half-life (t
) was calculated using the equation ln (2)/
z, where z is the slope of the plasma terminal elimination phase. The area under the concentration-time curve from time 0 to 24 h [AUC(0–24 h)] was calculated using the linear trapezoidal rule from 0 to 24 h after the dose. The apparent oral clearance and volume of distribution during the terminal phase were calculated as dose/AUC(0–24 h) and [(dose/AUC[0–24 h])]·[t/ln2], respectively. The accumulation (Rac) was calculated as the ratio of AUC(0–24 h) on d 14 relative d 1. The following additional pharmacokinetic parameters were observed: peak concentration (Cmax) and time to first occurrence of observed peak concentration (Tmax). The pharmacokinetic analysis of PF-00915275 urinary excretion was also performed with the calculation of the cumulative amount and percentage of the dose excreted in the urine obtained over the first 24-h dosing interval, respectively. The renal clearance (CLR) was calculated using the following equation: CLR = cumulative amount excreted in urine over 24 h/AUC (0–24 h).
The area under the concentration-time curve from time 0 to 4 h [AUC(0–4 h)] was calculated for prednisone and prednisolone. In addition, Cmax and Tmax were observed values. Predose plasma concentrations of PF-00915275 were used to determine the attainment of steady state and assess whether there was a pronounced trend for time-dependent changes in PF-00915275 concentrations.
Pharmacodynamic analyses
The effect of PF-00915275 on prednisone conversion to prednisolone after a single oral 10-mg prednisone dose was evaluated by assessing individual and mean AUC(0–4 h) values of prednisolone on d 11 relative to those on d –1 by dose group. Secondary analyses assessed the individual and mean AUC(0–4 h) values of prednisone on d 11 relative to those on d –1, also by dose group. The ability of PF-00915275 to inhibit the conversion of cortisone to cortisol was assessed by comparing predose (d 0) and postdose (d 1, 7, and 14) ratios of urinary cortisol metabolites (5THF + 5βTHF) to THE and UFF to UFE, by dose group, using descriptive statistics. In addition, the effect of PF-00915275 on total corticosteroid production (5THF + 5βTHF + THE + UFF + UFE) was compared on each postdose day to baseline (d 0) using descriptive statistics by dose group. For each ratio, the group means difference between each dose of PF-00915275 and placebo with the corresponding 90% confidence interval (CI) were calculated. For ACTH, DHEA-s, 4-androstenedione, and testosterone, the groups mean differences on d 14 between each dose of PF-00915275 and placebo with the corresponding 90% CI were computed. For all the statistical analyses (urinary cortisol, androgen, and prednisone/prednisolone), the placebo data were pooled (n = 15) to increase the power of the analysis.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Sixty-two healthy adult men (six blacks, 56 whites) were randomized into the study. Forty-five subjects received PF-00915275 and 15 subjects received placebo. Two subjects who enrolled in the study (1 mg dosing group) elected to discontinue early for reasons not related to study drug. Subjects were similar across dosing groups with respect to age, weight, and BMI distribution. Subjects ranged in age from 20 to 55 yr (mean 34 yr), body weight from 55 to 104 kg (mean 72.3 kg), and BMI from 20.7 to 29.9 kg/m2 (mean 24.8 kg/m2).
Pharmacokinetic parameter estimates
PF-00915275 was detectable in the plasma of all subjects at the first time point (1 h) after oral administration. The median Tmax was 1 h across all doses on d 1 and for the majority of doses on d 14 (Table 1), indicating that PF-00915275 is rapidly absorbed. AUC(0–24 h) and Cmax increased in a generally dose-proportional manner on both d 1 and 14 (Table 1). On d 14 the dose increased in increments of 1:, 3.3:, 10:, 33.3:, and 50, whereas Cmax increased in increments of 1:, 3.2:,10.7:, 25.5:, and 50.9, and AUC(0–24 h) increased in increments of 1:, 3.3:, 10.6:, 23.2:, and 51.3. The only exception on d 14 was for the 10-mg dose group, which was slightly less than dose proportional (23.2/25.5 vs. 33.3); the reason for this difference is unknown. The volume of distribution was unchanged as the dose increased and was approximately the same as total body water (
42 liters), indicating that PF-00915275 does not significantly penetrate into the peripheral tissues. PF-00915275 was slowly eliminated with a mean t across doses of 30 h (range 23.1–32.3 h). The mean Rac ranged from 1.9 to 2.7 (Table 1) after 14 d of dosing, which is consistent for a compound with a 30-h half-life. Overall, the variability in the pharmacokinetic parameters was moderate; the percent coefficient of variation (CV) for AUC(0–24 h) and Cmax ranged from 14 to 65% and 10 to 49%, respectively. The variability in AUC(0–24 h) and Cmax did not appear to change with dose; however, it did appear to increase on d 14 relative to d 1. Based on a visual inspection of the trough concentrations, PF-00915275 attained steady state by d 7.
|
Pharmacodynamics
Prednisone/prednisolone
The mean prednisolone exposure on d 11 relative to d –1 was reduced in a dose-dependent manner after PF-00915275 administration (range 0.3–37% from 0.3 to 15 mg; Table 2). This decrease in exposure can also be seen in the prednisolone plasma concentration-time profile shown in Fig. 2, top. A minimal increase (range 4–7%) in the mean prednisone exposure on d 11 relative to d –1 was observed up to the 10-mg dose (Table 2 and Fig. 2, bottom). At the 15-mg dose level of PF-00915275, the mean prednisone exposure was reduced by 13%. The relationship between PF-00915275 AUC(0–24 h) and percent decrease in prednisolone was not highly correlated, showing a correlation ratio of 0.57 (data not shown).
|
|
Activation of the hypothalamus-pituitary-adrenal (HPA) axis
Based on no obvious trend in the placebo-corrected total urinary corticosteroids (5THF + 5βTHF + THE + UFF + UFE) with increasing dose (Table 3) and the lack of increase in ACTH and androgens (DHEA-s, 4-androstenedione, testosterone) (Table 4), the data are consistent with lack of activation of the HPA axis.
|
|
THF + 5βTHF to THE) by dose group using descriptive statistics. A maximum decrease in the 5THF + 5βTHF to THE ratio, an indicator of 11βHSD1 inhibition, of 25–26% was observed on d 7 and 14 at the 15-mg dose level (Table 3
).
Markers of 11βHSD2 inhibition
The UFF to UFE ratio has been proposed as an indicator of 11βHSD2 inhibition (14). The UFF to UFE ratio trended slightly higher with increasing dose. This increase in placebo-corrected UFF to UFE ratio ranged from 14 to 17% on d 1, 7, and 14 at the 15-mg dose level (Table 3); however, apart from on d 1, the 90% CIs all encompassed zero, indicating that this increase is not statistically significant.
Safety and tolerability
Multiple oral doses of PF-00915275 up to 15 mg were safe and well tolerated by the healthy adult subjects. No subjects reported serious adverse events or discontinued the study prematurely due to adverse events. Thirty-nine of 62 subjects (63%) experienced treatment-related adverse events (AEs): 31 PF-00915275-treated subjects and eight placebo-treated subjects. The distribution of AEs was similar across dosing groups (including placebo). The greatest incidence of AEs (all causalities) were reported in the following treatment groups: placebo (33 events by 10 subjects), 0.3 mg (20 events by seven subjects), 1 mg (14 events by eight subjects), 3 mg (26 events by eight subjects), 10 mg (12 events by five subjects), and 15 mg (28 events by seven subjects). The reported AEs were mild to moderate in severity with the most common AEs related to gastrointestinal abnormalities (24 events) and nervous system disorders (22 events). The most frequently reported AEs across all treatment groups were headache, diarrhea, and conjunctivitis. The pattern in the AEs was similar across dosing groups for all three of the most common AEs. Thirteen of the reported 18 incidences of headache were mild in intensity and five were moderate.
Seven subjects had creatine kinase (CK) values that exceeded 2.0 times the upper limit of normal (ULN) at their most abnormal value, and three of these subjects had CK levels that also exceeded 1.5 times the ULN at their baseline value. These subjects CK levels returned to their baseline values by the follow-up visit (12–14 d after the last dose). These elevations were not considered clinically significant by the investigator. Two subjects had alanine aminotransferase and aspartate aminotransferase values that exceeded 3 times the ULN (both subjects baseline values were in the normal range). No changes in 
-glutamyltransferase or bilirubin were observed in either subject. One subject was a placebo subject and one subject was in the 3-mg treatment group. These increases were not considered clinically significant by the investigator. Both subjects’ alanine aminotransferase and aspartate aminotransferase values returned to the normal range by the follow-up visit. No clinically relevant changes in physical exams, electrocardiograms, or vital signs were detected in the study.
Taking into account the high variability, there were no clinically significant postdose differences from baseline in ACTH, free testosterone, DHEA-s, or 4-androstenedione at any dose level (Table 4), consistent with a lack of over activation of the HPA axis.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
30 h), displayed a generally dose proportional increase in exposure, and attained steady state by d 7. PF-00915275 did not significantly distribute into the peripheral tissues and did not undergo significant renal elimination (<1% of the administered dose was excreted unchanged in the urine).
In this paper we show for the first time the effect of a selective 11βHSD1 inhibitor, PF-00915275, on biomarkers of 11βHSD activity. Inhibition can be measured in urine (a measure of global 11βHSD activity) and plasma (largely reflecting hepatic activity). PF-00915275 inhibits the generation of bioactive prednisolone through inhibition of 11βHSD1 as well as reduces the urinary 5THF + 5βTHF to THE ratio. Both the urinary and plasma biomarkers for 11βHSD1 inhibition showed a dose-dependent trend with increasing PF-00915275 administration, resulting in 26 and 37% inhibition of 11βHSD1, respectively, at the 15-mg dose. In contrast to the only previously tested clinical compound, the nonselective carbenoxolone (14), our compound did not result in noteworthy changes in the UFF to UFE ratio. In addition, even at the highest doses, we did not observe hypertension or hypokalemia, clinical signs that could have indicated 11βHSD2 inhibition. This confirms the selectivity of this compound toward 11βHSD1 and the lack of activity toward 11βHSD2. Thus, for the first time, it is possible to validate the utility of these biomarkers as mechanistic noninvasive tools in the assessment of 11βHSD1 activity/inhibition. In previous studies, carbenoxolone increased the UFF to UFE ratio by 45–102%, reflecting inhibition of renal 11βHSD2 (14). In the current study, the UFF to UFE ratio trended slightly higher with increasing dose. The increase in placebo-corrected UFF to UFE ratio ranged from 14 to 17% on d 1, 7, and 14 at the 15-mg dose level; however, apart from d 1, the 90% CIs all encompassed zero. Whereas a minor effect on 11βHSD2 cannot be excluded, this is unlikely, and we believe there is no clinical significance over this short time frame. There was no change in the prednisone AUC(0–4 h) ratio on d 11 relative to d –1 at PF-00915275 doses ranging from 0.3 mg to 10 mg QD. Yet at the 15-mg dose, the mean ratio decreased 13% on d 11 relative to d –1; however, this decrease is unlikely real when the intersubject variability is considered as shown in Fig. 2, bottom, by the error bars.
One of the major concerns surrounding this target is that administration of an 11βHSD1 inhibitor to persons with an intact HPA axis could potentially reduce the local levels of glucocorticoids in target tissues, including regulatory feedback centers of the HPA axis in the brain and pituitary gland. This may lead to a compensatory activation of the HPA axis that would try to correct for the perceived glucocorticoid deficit throughout the body (15). The HPA over activation may result, in addition to the necessary extra cortisol, in excess mineralocorticoid precursors, such as corticosterone and deoxycorticosterone, and excess adrenal androgens, such as DHEA-s and 4-androstenedione. To assess possible over activation of the HPA axis, the concentrations of ACTH (between 0600 and 1200 h), free testosterone, DHEA-s, and 4-androstenedione were prospectively tested. Based on no obvious trend in the placebo corrected total urinary corticosteroids (5THF + 5βTHF + THE + UFF + UFE) with increasing dose and lack of changes in the levels of ACTH, DHEA-s, 4-androstenedione, and testosterone, it can be concluded that PF-00915275 administration did not result in overactivation of the HPA axis, at least at the doses used in this study. One limitation to the observations described is that prednisone was administered 24 h before the baseline measurements (d 0) of the urinary metabolites and HPA axis biomarkers. Whereas the half-life of prednisone is approximately 3 h (16), the biological half-life of prednisone is longer (up to 36 h) (17). Thus, as the administered prednisone dose was relatively low, we cannot exclude an effect on the baseline measurements.
In summary, PF-00915275 was well tolerated across all doses studied when administered for up to 2 wk in normal healthy volunteers. The prednisolone generation tests as well as ratios of urinary corticosteroid metabolites are useful biomarkers of 11βHSD1 inhibition.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online November 6, 2007
Abbreviations: AE, Adverse event; AUC(0–4 h), area under the concentration-time curve from time 0 to 4 h; BMI, body mass index; CI, confidence interval; CK, creatine kinase; CLR, renal clearance; Cmax, peak concentration; CV, coefficient of variation; DHEA-s, dehydroepiandrosterone sulfate; HPA, hypothalamus-pituitary-adrenal; 11βHSD1, 11β-hydroxysteroid dehydrogenase type 1; PD, pharmacodynamic; PK, pharmacokinetics; QD, once daily; Rac, accumulation ratio; Tmax, peak concentration; t, terminal phase elimination half-life; THE, tetrahydrocortisone; 5THF, 5-tetrahydrocortisol; 5βTHF, 5β-tetrahydrocortisol; UFE, urinary free cortisone; UFF, urinary free cortisol; ULN, upper limit of normal.
Received August 27, 2007.
Accepted October 30, 2007.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
