
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cancer Therapy: Clinical |
Authors' Affiliations: 1 School of Pharmacy, 2 Comprehensive Cancer Center, and 3 Biostatistics and Medical Informatics, University of Wisconsin, Madison, Wisconsin and 4 Comprehensive Cancer Center, The Ohio State University, Columbus, Ohio
Requests for reprints: Paul R. Hutson, School of Pharmacy, University of Wisconsin, 777 Highland Avenue, Madison, WI 53705-2222. Phone: 608-263-2496; Fax: 608-265-5421; E-mail: prhutson{at}pharmacy.wisc.edu.
| Abstract |
|---|
|
|
|---|
Design: Thirty-two postmenopausal women who were clinically disease-free following primary treatments for breast cancer receiving tamoxifen for at least 3 months were studied. Blood was collected for pharmacokinetic analysis after at least 4 months of receiving 20 mg tamoxifen daily. Subjects then began 8 weeks of oral exemestane (25 mg daily), followed by another set of blood samples.
Results: There were no serious toxicities noted when the two drugs were combined. There was no significant effect of exemestane on the area under the plasma concentration versus time curve (AUC) of tamoxifen at steady state before [3.04 mg h/L; 90% confidence interval (90% CI), 2.71-3.44] and during exemestane treatment (3.05 mg h/L; 90% CI, 2.72-3.41). There were no significant changes in the formation of primary tamoxifen metabolites. Oral clearance of exemestane averaged 602 L/h based on an average plasma exemestane AUC of 41.5 µg h/L (90% CI, 36.7-62.6). Plasma concentrations of estradiol, estrone, and estrone sulfate decreased when exemestane was begun; estradiol concentrations consistently decreased below the limit of quantitation.
Conclusions: There is no pharmacokinetic interaction between tamoxifen and exemestane. No modification in the standard regimen of either drug seems to be indicated if they are used in combination. The combination of the two drugs was well tolerated during the 8-week evaluation period.
An increasingly favored alternative is the suppression of estrogen synthesis using the aromatase inhibitors. The peripheral formation of estrogen from androgens such as androstenedione in postmenopausal women is dependent on the enzyme aromatase, also known as CYP2C19. This enzyme can be inhibited by multiple drugs, termed aromatase inhibitors, which are broadly classified as either steroidal in structure (type I) or nonsteroidal (type II; refs. 68). Exemestane (Aromasin, Pfizer, New York, NY) is a type I steroidal aromatase inhibitor that has shown single-agent activity against hormone-sensitive breast cancer (914). Unlike other type I competitive aromatase inhibitors such as letrozole or anastrazole, exemestane covalently binds to the enzyme, leading to irreversible inactivation of the aromatase molecule. In addition, exemestane is novel in that it also possesses mild androgen-like effects, which may yield additive or synergistic effects with the inhibition of estrogen synthesis in women with breast cancer (15).
The sequential use of adjuvant tamoxifen for 2 to 3 years, followed by exemestane in postmenopausal women, was recently reported to provide significantly increased disease-free survival when compared with the standard 5 years of tamoxifen (16). More recently, aromatase inhibition has been reported as an alternative or even preferred first line hormonal adjuvant treatment for breast cancer (1, 14). Not yet clinically tested in a controlled manner is the benefit of combining tamoxifen with exemestane. There are theoretical advantages in combining the estrogen receptor antagonism of tamoxifen with the suppression of estrogen synthesis by an aromatase inhibitor. This is analogous to the two-drug androgen blockade regimens commonly used in advanced prostatic neoplasms (17, 18). Murine mammary tumor models suggest that the combination of tamoxifen and exemestane has greater activity than does either drug alone (19). In contrast, the addition of the type I aromatase inhibitors letrozole or anastrazole did not add to the benefit of tamoxifen in this 7,12-dimethylbenanthracene-induced tumor model.
Both tamoxifen and exemestane are extensively metabolized by various mixed-function oxygenase (P450) enzymes and the potential exists for pharmacokinetic interactions with their concurrent use. The plasma concentrations of letrozole and anastrazole are known to be lower when combined with tamoxifen (2022). However, Rivera et al. (23) found no significant effect of the addition of tamoxifen on the pharmacokinetics of exemestane. Reports from the Anastrozole or Tamoxifen Alone or in Combination trial and from the North Central Cancer Treatment Group indicate no effect of anastrazole and letrozole on the pharmacokinetics of tamoxifen (21, 24). We report the final results of a clinical trial that examined the effect of adding exemestane on the pharmacokinetics of tamoxifen.
| Materials and Methods |
|---|
|
|
|---|
4 years previously with stage I/II invasive breast cancer and had been receiving a daily oral dose of 20 mg tamoxifen as a single-agent adjuvant therapy for a minimum of 3 months and a maximum of 10 years. "Postmenopausal" was defined as having had (a) a prior bilateral oophorectomy; (b) a previous hysterectomy with one or both ovaries intact, but either
60 years of age or with an follicle-stimulating hormone plasma concentration in the postmenopausal range; or (c)
1 year since last menstrual period and no prior oophorectomy or hysterectomy. Participants had to have had a normal physical examination without evidence of signs or symptoms of metastatic cancer, a normal chest roentgenogram
14 days before entry on study, and an Eastern Cooperative Oncology Group performance status of 0, 1, or 2. Patients using estrogenic or progestrogenic therapies other than or in addition to tamoxifen were not eligible.
Patients had to have adequate organ and hematologic function within 14 days of beginning the trial as indicated by serum creatinine
1.5 mg/dL, alkaline phosphatase and total bilirubin
1.5 times institutional upper limit of normal, and absolute neutrophil and platelet counts
1,500/µL and
100,000/µL, respectively. All subjects provided written informed consent.
After study entry, participants continued their tamoxifen treatment at a dose of 20 mg once daily. Four weeks after study registration, baseline blood samples were obtained to measure plasma concentrations of tamoxifen, N-desmethyl-tamoxifen, and 4-hydroxy-tamoxifen. Blood samples were obtained before the daily dose and at 0.5, 1, 2, 4, 6, 24, and 48 hours after the dose. The 24- and 48-hour samples were drawn before the daily dose.
After blood samples were drawn at week 4, exemestane was begun in all subjects at a daily oral dose of 25 mg for 8 weeks. At the end of this second treatment period (week 12), the blood samples for determining the pharmacokinetics of tamoxifen and its metabolites were repeated. Plasma aliquots from these blood samples were also collected to measure the plasma concentrations of exemestane and its active 17-hydroxylated metabolite. A portion of the plasma aliquots from the week 4 and week 12 blood samples was separated and assayed for estrone, estrone sulfate, and estradiol concentrations, as well as for osteocalcin, total and HDL cholesterol, triglycerides, and alkaline phosphatase. The results of the enzyme, lipid, and bone resorption assays are reported elsewhere (25). After the collection of blood samples at week 12, the subjects discontinued the exemestane and left the study on tamoxifen.
The plasma concentrations of tamoxifen, exemestane, and their primary metabolites were measured using validated high-performance liquid chromatography methods. The high-performance liquid chromatography assay for tamoxifen, 4-hydroxy-tamoxifen, and N-desmethyl-tamoxifen used fluorescence detection with a lower limit of quantitation of 5 µg/L for each compound. The concentrations of N-desmethyl-4-hydroxy tamoxifen (endoxifen) were not assayed (26). Concentrations of exemestane and the 17-hydroxylated metabolite were measured with high-performance liquid chromatography-tandem mass spectrometry with a lower limit of quantitation of 0.1 µg/L (27). Plasma concentrations of estradiol, estrone, and estrone sulfate were determined using solid-phase extraction and RIA (28). The lower limit of quantitation for estradiol, estrone, and estrone sulfate was 0.7, 1.8, and 6 ng/L, respectively. All assays were done by a contract laboratory (Pharma BioResearch, Kuidlaren, the Netherlands).
Pharmacokinetic methods. The pharmacokinetics of tamoxifen, exemestane, and their metabolites were evaluated by a combination of noncompartmental and compartmental methods. The area under the plasma concentration versus time curve (AUC024) was determined over a 24-hour dosing interval using the trapezoidal method. The fractional clearance of each metabolite was determined by dividing the AUC024 of each metabolite by that of the parent drug (29). Because all sampling was done at a steady state, the oral clearance of the parent compounds was determined by dividing the administered dose by the AUC024. Oral clearance is the ratio of plasma clearance to bioavailability. Because the bioavailability of an orally administered drug cannot be known in an individual without giving an i.v. dose as well, it was the oral clearance of tamoxifen and exemestane in this study that was determined rather than the plasma or systemic clearance. The mean tamoxifen, exemestane, and metabolite concentrations over a 24-hour dosing interval were obtained from the respective ratios of AUC024/24 hours.
Any change in the tamoxifen AUC caused by exemestane was identified by a change in the ratio of the AUC024 tamoxifen at weeks 4 and 12. A ratio of 1 would indicate no effect of exemestane on tamoxifen AUC and 90% confidence boundaries that did not fall outside the limits of 0.8 and 1.25 were considered not clinically different from unity and thus meeting accepted standards for bioequivalence (30). First, the logarithm was taken of the 0- to 24-hour trapezoidal AUC for each subject before and during exemestane treatment. The arithmetic mean was determined for the week 12/week 4 log-AUC ratios, as was the 90% confidence interval (90% CI) around this mean: the antilog was then taken of this mean and confidence interval.
Changes in the log-transformed plasma concentrations of estrone, estrone sulfate, and estradiol before and after starting exemestane were evaluated using the paired t test. If concentrations were below quantifiable limits, the value of the lower limit of quantitation was used for the statistical comparison. Confidence intervals for the observed data were determined using EXCEL (Microsoft, Redmond, WA) and paired t tests were done using SigmaStat 3.01 (Systat, Port Richmond, CA).
Nonlinear compartmental modeling of the pharmacokinetics of the two drugs was done with the NONMEM program (31). Differential equation-based control streams were prepared to model the concentration data for each drug and the respective metabolites in all subjects rather than for one subject at a time. The model was based on first-order absorption and elimination, with two "compartments" defined for the gut and sampled compartments. Model development led to the addition of a tissue compartment for exemestane but not for tamoxifen or its metabolites. The mean tamoxifen elimination rate constant was allowed to differ with and without the presence of exemestane and intersubject variability was allowed to vary at weeks 4 and 12. These variables reflecting intersubject variability (
) were added in stepwise manner to the model when they provided a statistically significant improvement in the objective function. Confidence intervals for the pharmacokinetic variables of half-life and distribution volume were obtained from the log-transformed output of 100 iterations of the bootstrap algorithm within the Wings for NONMEM software package (32).
| Results |
|---|
|
|
|---|
|
|
|
|
As anticipated, the addition of exemestane caused a dramatic and statistically significant decline in the plasma concentrations of estradiol, estrone, and estrone sulfate (Fig. 2; Table 4). Estradiol concentrations in all women decreased below measurable concentrations after 8 weeks of exemestane treatment. Substantial or complete suppression of estrone and estrone sulfate concentration was also consistently observed.
|
|
| Discussion |
|---|
|
|
|---|
We found no significant pharmacokinetic interaction of exemestane on tamoxifen or its metabolites. Furthermore, the combined regimen was well tolerated and yielded the expected and desired suppression of plasma estrogen concentrations. A complimentary study was recently reported by Rivera et al. (23) who studied the effect of adding tamoxifen to exemestane treatment. Tamoxifen was found to have no significant effect on the pharmacokinetics of exemestane, arguing that no dose modification is needed for either drug when used in this combination.
Few pharmacokinetic reports exist describing the concentrations of tamoxifen and its 4-hydroxyl and N-desmethyl metabolites at steady state on an oral regimen of 20 mg/d. The mean tamoxifen plasma concentration of 127 µg/L in the present study is similar to the values of 93 and 133 µg/L reported by Herrlinger et al. (35) and Wilkinson et al. (36) but less than the value of 310 µg/L reported by Kemp et al. (37). Assay cross-interference by metabolites in the earlier Kemp study may be responsible for this difference. As expected, the AUC and average concentration of 4-hydroxy-tamoxifen were
3.5% of the values for the parent drug whereas the values for N-desmethyl-tamoxifen were 1.8-fold higher than tamoxifen. This is in agreement with other descriptions of single dose and steady-state dosing of tamoxifen (3739).
With an elimination half-life of
86 hours, any changes in tamoxifen dose or clearance would require
2.5 weeks to reachieve steady state. Subjects on our trial were already on tamoxifen on study entry and the additional 4-week run-in and 8-week delay after starting exemestane treatment support the premise that tamoxifen concentrations were at steady state when sampled. Exemestane, with an elimination half-life of 1.6 hours determined from the NONMEM modeling, was also at steady state at the second assessment period. The 9,400-hour half-life of N-desmethyl-tamoxifen estimated from the NONMEM computer analysis is also suggested by inspection of Fig. 1 and by the earlier reports of Soininen et al. (39) and Fuchs et al. (40). Only
10% of any change in steady-state concentrations in N-desmethyl-tamoxifen caused by the 8-week coadministration with exemestane would have been apparent given this half-life of
1 year. Although the week 12 plot of N-desmethyl-tamoxifen concentrations tends to be higher than the week 4 plot in Fig. 1, there is not enough of a change to support computer models that include a change in the fraction of tamoxifen metabolized to N-desmethyl-tamoxifen or a change in N-desmethyl-tamoxifen elimination rate. The consistent findings of a very long half-life for N-desmethyl-tamoxifen argue that any studies of this metabolite must anticipate the need for very prolonged sampling schedules.
Tamoxifen is known to be metabolized to its 4-hydroxyl and N-desmethyl metabolites by the P450 enzymes CYP3A4, CYP2C9, CYP2C19, and CYP2D6 (41). Although the N-desmethyl-tamoxifen is the predominant metabolite, its affinity for the estrogen receptor is low compared with tamoxifen and the 4-hydroxy metabolite (37). In contrast, although the 4-hydroxy metabolite has a higher affinity for the receptor than tamoxifen itself, it is present in low concentrations. Even so, both metabolites are considered to contribute to the effect of tamoxifen (42).
The compound termed endoxifen was not assayed in this trial because the seminal reports of its importance were published after the completion of this trial (26, 43). A hundredfold more potent than tamoxifen, endoxifen was found to have estrogen receptor binding activity similar to that of 4-hydroxy-tamoxifen and estradiol. Prior work has shown that endoxifen concentrations are
10-fold lower than tamoxifen but 10-fold higher than 4-hydroxy-tamoxifen, indicating that it is likely to serve as a more active contributor to the antagonism of the estrogen receptor than 4-hydroxy-tamoxifen or even tamoxifen itself (26). It is unlikely that endoxifen concentrations were significantly affected by exemestane because no significant changes were seen in the AUC of tamoxifen or of the two metabolites intermediate to the formation of endoxifen.
Exemestane is also hydroxylated by CYP3A4 (44) but has not been reported to interfere with the metabolism of other drugs by competitive or noncompetitive processes. Our data support this experience. Although the inhibition of aromatase (CYP2C19) by exemestane might be expected to inhibit the clearance of tamoxifen, the availability of alternate CYP enzyme pathways acting on tamoxifen apparently minimizes any such effect of exemestane.
The prescribing information for Aromasin (44) asserts that the exemestane plasma AUC following a 25-mg oral dose in postmenopausal women treated for breast cancer is significantly higher than in normal postmenopausal women (75.4 versus 41.4 µg h/L). Our data and those of other groups refute this statement. The exemestane AUC of 41.5 µg h/L determined at week 12 in the present study is similar not only to the value for normal women in the prescribing information (41.4 µg h/L) but also to the value of 41.7 µg h/L reported by Jannuzzo et al. (45) in healthy women and the value for normal males (36.4 ± 8.8 µg h/L) given the same 25-mg daily dose (46). A history of breast cancer does not seem to have an effect on the AUC or oral clearance of exemestane and no adjustments in dose for this reason seem to be warranted.
Our data, combined with those of Rivera et al. (23), indicate a lack of significant pharmacokinetic interaction between tamoxifen and exemestane. No dose modifications seem necessary when the two drugs are administered concurrently. The pharmacokinetics of exemestane do not seem to be different between normal volunteers and women treated adjuvantly for breast cancer.
| Footnotes |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 4/28/05; revised 9/12/05; accepted 9/27/05.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. U Buzdar and J. F. Robertson Fulvestrant: Pharmacologic Profile Versus Existing Endocrine Agents for the Treatment of Breast Cancer Ann. Pharmacother., September 1, 2006; 40(9): 1572 - 1582. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Cell Growth & Differentiation |