An overview of the pharmacology and pharmacokinetics of the newer generation aromatase inhibitors anastrozole, letrozole, and exemestane

Aman U. Buzdar, M.D   
Cancer 2002;95:2006-16

It has long been recognized that approximately one-third of all breast carcinomas are estrogen dependent and will regress after estrogen deprivation.Thus, reducing the level of estrogen remains a valuable target for breast carcinoma treatment in both premenopausal and postmenopausal women. Reducing the effects of estrogen can be mediated by agents that block estrogen at the receptor level, such as tamoxifen, or by inhibitors of estrogen biosynthesis, such as the aromatase inhibitors (AIs). AIs act through inhibition of the cytochrome P450 enzyme, aromatase, which catalyses the conversion of androgens to estrogens. The AIs have been developed primarily for use in women in whom ovarian function has ceased either in naturally or surgically postmenopausal patients. It should be noted that, in premenopausal women, the ovaries are the primary site of estrogen production, and, because AIs are not capable of blocking ovarian estrogen synthesis completely in premenopausal women, they cannot be used as the sole endocrine treatment. In this patient group, there are data showing that anastrozole can be used in combination with ovarian ablation therapy with such drugs as goserelin (ZoladexTM; a gonadotrophin-releasing hormone agonist) to ensure blockade of estrogen synthesis. Preliminary pharmacokinetic data for this combination are promising and show that 15 premenopausal patients with metastatic breast carcinoma who received anastrozole (1 mg once daily) with goserelin (3.6 mg monthly) for up to 41 months had significantly lower estradiol (E2) levels (P < 0.05; measured at 6 months) compared with baseline levels.[6] In addition, 11 of those patients (73%) achieved either an objective response (OR) or stable disease after 6 months of treatment.

The first-generation AI, aminoglutethimide, became available in the late 1970s. However, despite its proven efficacy as second-line therapy after tamoxifen in postmenopausal women with advanced hormone-responsive breast carcinoma, its widespread use was limited by both toxicity and lack of selectivity for the aromatase enzyme, necessitating concomitant corticosteroid supplementation. Formestane, which is a steroidal AI based on the androgen androstenedione that also has been shown as an effective treatment for this patient population and is a more selective AI, became available in 1993. Because of its improved selectivity, formestane has fewer side effects compared with aminoglutethimide; however, as a result of extensive first-pass metabolism, it has to be administered by twice-monthly intramuscular injection, and local reactions have been reported in up to 17% of patients.More recently, the so-called newer generation AIs, which include anastrozole, letrozole,  and exemestane, have become available for use in postmenopausal women with advanced, hormone-responsive breast carcinoma.

Exemestane and formestane are classified as Type I AIs on the basis of their steroidal nature and irreversible binding to the aromatase enzyme, causing permanent inactivation even after the drug is cleared from the circulation. By contrast, anastrozole and letrozole, which are nonsteroidal, are classified as Type II AIs, because they competitively inhibit the conversion of androgens to estrogens.

Androgenic Effects

Anastrozole and letrozole have no androgenic, progestrogenic, or estrogenic effects, such as weight gain, acne, and hypertrichosis; although they generally are associated with supratherapeutic doses, these effects have been reported in patients receiving the approved clinical doses of exemestane. These side effects may lead to tolerability issues, particularly if prolonged treatment is envisaged, for example, in the adjuvant setting.

Lipid Profile

Four decades of epidemiologic research in the Framingham Study have provided data to support the finding that an elevated total cholesterol (comprised of low-density lipoprotein [LDL] and high-density lipoprotein [HDL]) to HDL cholesterol ratio is an important risk factor for the development of cardiovascular disease. Thus, drugs that alter lipid profiles may increase the risk of developing cardiovascular disease. A recent report from a large data set indicates that treatment with anastrozole does not markedly alter lipid profiles (compared with baseline measures); data derived from patients (n = 952 patients at study entry) who were recruited into two large advanced breast carcinoma trials comparing anastrozole and tamoxifen showed that anastrozole (1 mg once daily) did not affect nonfasting lipids in women with advanced breast carcinoma (measurements were taken every 12 weeks for > 2 years). Although no formal statistics were carried out, the mean baseline total cholesterol level in the anastrozole group was 5.8 mmol/L, and, after 84 weeks of treatment, there was a slight increase of + 0.3 mmol/L. For the same parameter in the tamoxifen group, the baseline level was 5.9 mmol/L at baseline, and the level decreased by 0.6 mmol/L after 84 weeks of treatment. The baseline HDL cholesterol level in the anastrozole group was 2.4 mmol/L, and the level decreased by 2.1 mmol/L after 84 weeks of treatment. In the tamoxifen group for the same parameter, the mean baseline level was 3.7 mmol/L, and the level decreased by 2.2 mmol/L after 84 weeks of treatment. The LDL level in the anastrozole group was 3.7 mmol/L at baseline, and this level increased slightly by 0.2 mmol/L after 84 weeks of treatment. In the tamoxifen group for the same parameter, the baseline LDL cholesterol level was 3.8 mmol/L, and the level decreased by 0.9 mmol/L after 84 weeks of treatment.These data are supported by another smaller study in postmenopausal women (n = 44 patients) with breast carcinoma in which, compared with baseline measures, up to 32 weeks of treatment with anastrozole did not alter any lipid parameters significantly (total cholesterol, LDL and HDL cholesterol, and triglycerides; no raw values were available in the report).

A preliminary study showed that, compared with baseline values, letrozole (2.5 mg once daily) had an unfavorable effect on fasting lipids in 20 women with advanced breast carcinoma after 8 weeks and 16 weeks of treatment, with letrozole treatment resulting in significant increases in total cholesterol (P = 0.05; mean ± standard deviation [SD] at baseline vs. 16 weeks: 239 mg/dL ± 56 mg/dL vs. 258 mg/dL ± 53 mg/dL, respectively), LDL cholesterol (P < 0.01; mean ± SD at baseline vs. 16 weeks: 148 mg/dL ± 50 mg/dL vs. 170 mg/dL ± 53 mg/dL, respectively), and apolipoprotein B (P = 0.05; mean ± SD at baseline vs. 16 weeks: 109 mg/dL ± 36 mg/dL vs. 117 mg/dL ± 32 mg/dL, respectively) levels. In addition, 9 weeks of treatment with exemestane in postmenopausal women with advanced breast carcinoma resulted in a significant decrease in total cholesterol (P < 0.01; mean ± SD at baseline vs. 12 weeks: 6.58 mmol/L ± 0.98 mmol/L vs. 5.74 mmol/L ± 0.79 mmol/L, respectively), HDL cholesterol (P < 0.01; mean ± SD at baseline vs. 12 weeks: 1.39 mmol/L ± 0.23 mmol/L vs. 0.95 mmol/L ± 0.20 mmol/L, respectively), and total triglycerides (P = 0.023; mean ± SD at baseline vs. 12 weeks: 1.64 mmol/L ± 0.45 mmol/L vs. 1.38 mmol/L ± 0.38 mmol/L, respectively), which decreases are in a favorable direction for total cholesterol and triglycerides but in an unfavorable direction for HDL cholesterol. However, the results of a European Organization of Research and Treatment of Cancer Trial (n = 122 patients) showed that 24 weeks of exemestane had no impact on serum lipids.[53] Although it is difficult to extrapolate these data into the adjuvant setting, particularly because it is not known whether there will be any long-term clinical relevance in terms of cardiovascular morbidity, the results of these four studies clearly show that anastrozole, letrozole, and exemestane appear to have different effects on plasma lipids.

Adrenal Steroidogenesis

The mineralocorticoid aldosterone is involved in the regulation of blood volume and serum sodium and potassium levels; therefore, macroscopic changes in aldosterone levels ultimately can affect blood pressure control and result in disturbances of electrolyte balance. Cortisol is the major stress hormone in humans, with plasma levels rising dramatically under conditions of physical or mental stress, and it is believed generally that this rise enables humans to cope and adapt to stress. Reducing cortisol levels can result in a reduced ability to withstand the impact of stress. In terms of adrenal steroidogenesis, data indicate that anastrozole is the most selective nonsteroidal AI. It lacks an impact on adrenal steroidogenesis at doses up to 10 times the recommended clinical doses  thus suggesting that anastrozole has little activity on other cytochrome P450 enzymes. The cytochrome enzyme 11-hydroxylase mediates the formation of cortisol in humans and the formation of 11-deoxycorticosterone (11-DOC) in monkeys and dogs. In preclinical studies in monkeys at a dose of 3 mg/kg, twice-daily treatment with anastrozole did not alter 11-DOC, a result that the authors suggest indicates a margin of safety for anastrozole of 30-fold. In the clinical setting, basal and adrenocorticotrophic hormone (ACTH)-challenged cortisol levels of healthy postmenopausal women did not differ from baseline levels after treatment for 14 days with anastrozole (5 mg and 10 mg once daily) at doses 5-fold and 10-fold higher than the clinically administered dose (raw data not available). After long-term treatment with the same doses of anastrozole, as an extension to the short-term study, there was no effect on basal (baseline levels vs. 115 days: 378 nmol/L vs. 398 nmol/L, respectively) and ACTH-stimulated cortisol (60 minutes poststimulation baseline levels vs. 115 days: 1099 nmol/L vs. 1008 nmol/L, respectively)at approximately 3 months, thus further indicating a wide margin of selectivity for anastrozole.

The enzyme 18-hydroxylase, which mediates the formation of aldosterone, also does not appear to be effected by anastrozole: In the clinical setting, basal and ACTH-stimulated aldosterone levels (raw data not available) in healthy postmenopausal women did not differ from baseline levels after following 14 days of treatment with anastrozole (5 mg and 10 mg once daily) at doses 5-fold and 10-fold higher than the clinically administered dose.[56] In addition, long-term treatment with the same doses of anastrozole for up to 115 days in the same patients had no effect on basal (baseline levels vs. 115 days: 372 nmol/L vs. 383 nmol/L, respectively) and ACTH-stimulated (60-minute poststimulation baseline levels vs. 115 days: 901 nmol/L vs. 860 nmol/L, respectively) aldosterone.

With respect to letrozole treatment, one report by Bisagni and colleagues showed that basal cortisol levels and ACTH-stimulated cortisol synthesis were reduced significantly by letrozole (0.5 mg once daily) after treatment for 56 days (P = 0.0029; mean baseline value vs. 56 days: 188 ng/mL vs. 149 ng/mL, respectively) and treatment for 84 days (P = 0.0006; mean baseline value vs. 84 days: 188 ng/mL vs. 138 ng/mL, respectively) in 14 postmenopausal women with progressive, metastatic breast carcinoma. However, in the same study, it was found that letrozole had no effect on basal aldosterone levels after 84 days of treatment (at baseline vs. 84 days: 74 pg/mL vs. 58 pg/mL, respectively).[31] By contrast, more recently, Bajetta and colleagues found a significant increase (P = 0.025) in basal aldosterone levels (raw data not available) in 46 postmenopausal women with advanced breast carcinoma after 3 months of treatment with clinical doses of letrozole of either 0.5 mg (n = 22 patients) or 2.5 mg (n = 24 patients) once daily and found a significant decrease in ACTH-stimulated aldosterone levels after treatment with the clinically administered dose of letrozole (2.5 mg; P = 0.04).

There has been only one study published to date with respect to the impact of treatment with exemestane on adrenal steroidogenesis. Evans and colleaguesreported that exemestane administered acutely to 29 healthy postmenopausal women (n = 3-4 patients per group) at doses ranging from 0.5 mg to 800 mg resulted in no change in cortisol or aldosterone levels for up to 7 days after treatment (raw data not available).

Results to date show that anastrozole treatment for up to 115 days has little impact on cortisol and aldosterone levels at up to 10 times the clinical dose, and one acute study with exemestane also reported no changes in these levels. Treatment for up to 84 days with letrozole alters cortisol and aldosterone levels however, the clinical relevance of these differences has yet to be elucidated.

Bone Metabolism

Epidemiologic data have suggested a correlation between low serum estrogen levels and bone resorption in late-postmenopausal women, although a direct cause-and-effect relation has not been established. This phenomenon raises the possibility that the newer generation AIs may have adverse effects on bone by increasing resorption, and this may give rise to diseases, such as osteoporosis. To date, exemestane has been assessed only in the preclinical setting.One study found that six men (age > 65 years) who were treated with anastrozole (1 mg once daily) for 3 weeks had a nonsignificant increase in bone resorption (measured by N-telopeptide and C-terminal), although there was little change in bone formation (as measured by osteocalcin and bone specific alkaline phosphatase). Treatment with letrozole (2.5 mg once daily) for 6 months in healthy postmenopausal women resulted in a significant increase in bone resorption, as measured by urinary pyridinoline and deoxypyridinoline, compared with baseline (P < 0.005 for both markers). More recently, a study published by Harper-Wynne and colleagues showed that 3 months of treatment with letrozole resulted in increases in bone resorption (as measured by C-terminal). In the absence of head-to-head studies with anastrozole and letrozole, it is not possible to determine whether they have any different effects on bone resorption, because the duration of treatment was different as well as the bone markers and the gender of the volunteers.