Aromatase inhibitors improve relapse-free survival in early breast cancer, but there is concern about possible detrimental effects on bone mineral density (BMD) and plasma lipids. This paper presents the results of a 2-year study evaluating the effects of exemestane versus placebo on BMD, bone markers, plasma lipids and coagulation factors, including a 1-year follow-up after termination of treatment in 147 patients. During treatment, the mean annual rate of loss of BMD in the lumbar spine was 2.17% in the exemestane group versus 1.84% in the placebo group (n.s.) and 2.72% versus 1.48%, respectively, in the femoral neck (P=0.024). A loss of BMD above that expected in both arms of this study could be due to low vitamin D status (88% of all patients had vitamin D levels <30 ng/ml). The changes observed with exemestane were partially reversed during a 1-year follow-up, with no significant difference between the two arms. Similarly, the moderate decrease in high-density lipoprotein (HDL)-cholesterol was reversed. The bone marker values decreased, although a difference at 6 months of follow-up was still recorded, in particular for the markers of bone synthesis.

554 Background: To evaluate potential detrimental effects of the aromatase inactivator exemestane on bone, 147 postmenopausal women with early breast cancer were randomised to receive either exemestane for 2 years or placebo (J. Clin. Oncol. 23 [22], 5126–5137, 2005). Exemestane increased the annual bone loss from the femoral neck (2.72%) compared to placebo (1.48%; P = 0.024) with a non-significant increase in the lumbar spine (exemestane 2.17% versus placebo 1.84%). The annual bone loss was higher than expected in the placebo arm. Methods: Various biomarkers involved in bone metabolism (25-hydroxyvitamin D, parathormone, calcium, estrogens, androgens) were analysed to elucidate their influence on bone status at baseline and BMD loss during treatment with exemestane compared to placebo. Results: Using a cut-off value of 30 ng/ml for 25-hydroxyvitamin D (J. Clin. Endocrinol. Metab. 90 [6], 3800–3801, 2005), the majority of study participants suffered from vitamin D deficiency (56 of 62 patients in the placebo group and 52 of 59 in the exemestane group). The mean levels (95% confidence interval) of vitamin D were 22.6 ng/ml (21.2 - 24.1) in the placebo group and 21.6 ng/ml (20.0 - 23.3) in the treatment group, revealing no differences between these groups. Low serum calcium levels at baseline were found to be significantly correlated to low BMD in the femoral neck in the exemestane group. However, individual levels of vitamin D, parathormone and estradiol at baseline were not correlated significantly to BMD. Conclusions: Considering an annual bone loss of 0.5% to be representative for postmenopausal women (Osteoporos. Int. 15, 881–886, 2004), our data indicate that vitamin D deficiency could be the most important factor elevating bone loss among patients treated with exemestane as well as in the placebo group. These findings, together with the observation of a moderate additional effect of exemestane on bone loss, underlines the need for proper vitamin D substitution of postmenopausal women in general and in breast cancer patients during treatment with aromatase inhibitors in particular. [Table: see text]

PURPOSETo evaluate potential detrimental effects of exemestane on bone and lipid metabolism.PATIENTS AND METHODSPostmenopausal women with early breast cancer were randomly assigned to exemestane 25 mg daily or placebo for 2 years in a double-blind setting. Primary objective was to evaluate the effect of exemestane on bone mineral density. Secondary objectives were effects on bone biomarkers, plasma lipids, coagulation factors, and homocysteine. Planned size was 128 patients.RESULTSOne hundred forty-seven patients were enrolled. All patients completed their 24-month visit except for those discontinuing treatment at an earlier stage. The mean annual rate of bone mineral density loss was 2.17% v 1.84% in the lumbar spine (P = .568) and 2.72% v 1.48% in the femoral neck (P = .024) in the exemestane and placebo arm, respectively. The mean change in T-score after 2 years was -0.21 for exemestane and -0.11 on placebo in the hip, and -0.30 and -0.21, respectively, in the lumbar spine. Exemestane significantly increased serum level and urinary excretion of bone resorption, but also bone formation markers. Except for a modest reduction in high-density lipoprotein cholesterol (P < .001) and apolipoprotein A1 (P = .004), exemestane had no major effect on lipid profile, homocysteine levels, or coagulation parameters.CONCLUSIONExemestane modestly enhanced bone loss from the femoral neck without significant influence on lumbar bone loss. Except for a 6% to 9% drop in plasma high-density lipoprotein cholesterol, no major effects on serum lipids, coagulation factors, or homocysteine were recorded. Bone mineral density should be assessed according to the US Preventive Services Task Force guidelines.

531 Background: Estrogens (Es) reduction in menopause increases osteoporosis and fractures risk. Aromatase inhibitors are being introduced for treating postmenopausal early breast cancer (EBC). Exemestane (E, Aromasin®), a steroidal aromatase inactivator, inhibits peripheral aromatase by 98%, markedly reducing Es. We report the effect of E on Es and bone metabolism in postmenopausal EBC patients (pts). Methods: We conducted a randomized, double blind study of E vs. placebo (P) on bone metabolism, endocrinological and metabolic parameters in postmenopausal pT1N0 or T2N0 (with no other risk factors) or ductal carcinoma in situ (DCIS) pts. ER and/or PgR+ve or unknown pts with bone mineral density (BMD) within 2 SD of the mean for 65-year (yr) women, were randomized to E 25 mg or P p.o. daily for 2 yrs, with a 1-yr follow-up. Primary objective was non-inferiority on BMD. Planned sample size was 128 BMD-evaluable pts. Estradiol (E2), estrone (E1), bone alkaline phosphatase (ALP), PINP, osteocalcin (OC), serum CTX (s-CTX), urinary CTX (u-CTX) and NTX (u-NTX) were evaluated at baseline (BS), 6, 12 and 24 months (mos). Results: 147 pts have been enrolled, 73 E and 74 P. Median age: 61 yrs (E) and 60 yrs (P). Median time from menopause: 9.7 yrs E, 8.0 yrs P. ER and/or PgR: +ve 130 pts, unknown 17 (15 with DCIS). E was well tolerated, 70% of pts had drug-related events on E, 73% on P. Nine pts withdrew on E and 3 on P for adverse events. Three withdrew consent on E and 2 on P. The % change from baseline on Es and bone metabolism markers are described in the Table. Conclusions: E induces a marked Es suppression coupled with an increase in bone resorption (CTX, NTX), and even more in bone formation (ALP, PINP, OC). This indicates an increase of bone remodeling, although no major effect is observed on bone mineral density (Lonning ASCO 2004, submitted). Correlation analyses are ongoing. Author Disclosure Employment or Leadership Consultant or Advisory Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Pfizer Italia S.p.A. Pfizer Co. Pfizer Co.

650 Background: Estrogens (Es) favorably modify the lipid profile. Aromatase inhibitors reduce Es; LDL and triglycerides (tri) increase have been observed (Elisaf, EJC 2001). We report the effect of exemestane (E, Aromasin), a steroidal aromatase inactivator, on lipid profile and coagulation in postmenopausal early breast cancer (EBC) patients (pts). Methods: We conducted a randomized, double blind study on the effect of E on bone metabolism, endocrinological and metabolic parameters in postmenopausal pT1N0 or T2N0 (with no other risk factors) or ductal carcinoma in situ (DCIS) pts. ER and/or PgR+ve or unknown pts with bone mineral density (BMD) within 2 SD of the mean for 65-years (yrs) women were randomized to E 25 mg day or placebo (P) p.o. for 2 yrs, with a 1-yr follow-up. Primary objective was non-inferiority on BMD. Planned sample size was 128 BMD-evaluable pts. Cholesterol (Chol), HDL-cholesterol (HDL-c), tri, Apo-AI, Lipoprotein A (Lipo-AI), Apo B, PT, APTT, fibrinogen (Fib) and homocysteine (hoc) were evaluated at baseline (BS), 6, 12, and 24 months (mos). Results: 147 pts have been enrolled, 73 E and 74 P. Median age: 61 yrs (E) and 60 yrs (P). Median time from menopause: 9.7 yrs E and 8.0 yrs P. ER and/or PgR: +ve 130 pts, unknown 17 (15 with DCIS). E was well tolerated, 70% of pts had drug-related events on E and 73% on P. Nine pts withdrew on E and 3 on P for adverse events (1 E for DVT and 2 P for increased Fib or apoplexia). No PT changes from BS were observed on both arms. At 24 mos median Fib was reduced by 1.2% on E and 4.8% on P, and median APTT by 3% on both. Lipid profile evolution is described in the Table. Out of 64 E pts, 4 developed Gr. 2 Chol, and 1 Gr. 1 Tri elevation. Out of 62 P pts, 7 developed Gr. 2 and 1 Gr. 1 Chol, 3 Gr. 1 Tri elevations. Conclusions: After 2 yrs of E, lipid profile and homocysteine levels are similar in E and P pts, except for a modest reduction in HDL-c on E. Author Disclosure Employment or Leadership Consultant or Advisory Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Pfizer Italia S.p.A. Pfizer Co. Pfizer Co.