Purpose: In the Anastrozole, Tamoxifen Alone or in Combination trial, the combination arm was inferior to anastrozole alone in terms of disease-free survival possibly due to an adverse pharmacokinetic interaction or a predominant estrogenic effect of tamoxifen under estrogen deprivation. We assessed whether the addition of a lower dose of tamoxifen influenced anastrozole bioavailability and favorably modulated biomarkers of bone fracture, breast cancer, cardiovascular disease, and endometrial cancer risk. The influence of CYP2D6 genotype on tamoxifen effects was also determined.

Experimental Design: Seventy-five postmenopausal women with breast intraepithelial neoplasia were randomly allocated to either 1 mg/d anastrozole or 10 mg/wk tamoxifen or their combination for 12 months. Study endpoints were plasma drug concentrations and changes of C-telopeptide, osteocalcin, estradiol/sex hormone binding globulin (SHBG) ratio, estrone sulfate, insulin-like growth factor-I (IGF-I)/insulin-like growth factor binding protein-3 (IGFBP-3), C-reactive protein, antithrombin-III, endometrial Ki-67 expression, and thickness.

Results: Anastrozole concentrations were not affected by the combination with low-dose tamoxifen, whereas endoxifen levels were lower in poor CYP2D6 metabolizers. C-telopeptide increased by 20 with anastrozole and decreased by 16 with tamoxifen and by 7 with their combination (P < 0.001); osteocalcin showed similar changes. Compared with anastrozole, the combination arm showed lower IGF-I/IGFBP-3 levels (17 versus 9; P = 0.004) and lower estradiol/SHBG and estrone sulfate reductions (15 versus 29 and 30 versus 38, respectively). However, IGF-I/IGFBP-3 and estradiol/SHBG did not decrease in poor CYP2D6 metabolizers. Endometrial thickness was not greater in the combination than in the anastrozole arm.

Conclusions: The addition of a weekly tamoxifen administration did not impair anastrozole bioavailability and modulated favorably its safety profile, providing the rationale for further studies. (Clin Cancer Res 2009;15(22):705360)

Translational Relevance

Primary and secondary breast cancer chemoprevention strategies are becoming a predominant issue for social health. The spread of the screening programs has increased diagnosis of intraepithelial lesions, and until today, there is no consensus on how to treat such lesions. In the cardiovascular setting, treatment of risk factors has been confirmed to be a very successful strategy, and recent protocols indicate that control of multiple pathways gives better results and attenuates adverse events. To further investigate the possible role of low-dose tamoxifen, anastrozole, and their combination in breast cancer prevention, we conducted this randomized trial. This study provides a rationale to further study the combination of the two drugs because, with these dosages, there is not a negative pharmacokinetic interaction. Furthermore, the combination maintains a reduction of breast cancer risk biomarkers and may have a safer toxicity profile on bone metabolisms and on the cardiocirculatory system.

Aromatase inhibitors are a standard treatment of estrogen receptorpositive breast cancer in postmenopausal patients (1,4) and are being tested as chemopreventive agents in at-risk women (5). However, these agents are associated with increased bone fracture rate, joint and tendon disorders, and possibly increased cardiovascular risk due to their profound estrogen-suppressive effect rising concern on their long-term safety, especially in the prevention setting. Moreover, prolonged aromatase inhibitor treatment may lead to the onset of endocrine resistance with the emergence of estrogen hypersensitive cell clones (6). One potential way to counteract these phenomena is to add tamoxifen to exploit its partial estrogenicity. In the Anastrozole, Tamoxifen Alone or in Combination study, a phase III adjuvant trial with three arms (1 mg anastrozole, 20 mg tamoxifen, and their combination), anastrozole significantly prolonged disease-free survival versus tamoxifen. Moreover, the combination of anastrozole and tamoxifen at the dose of 20 mg/d was inferior to anastrozole alone and slightly, albeit not significantly, worse than tamoxifen alone in terms of disease-free survival, thus leading to the early discontinuation of the combination arm (1, 7, 8). Different explanations have been advocated, including a documented adverse pharmacokinetic interaction leading to low anastrozole concentrations, although estradiol suppression was similar between anastrozole and combination arms (9) and a predominant estrogenic activity of tamoxifen under anastrozole-induced estrogen-deprived milieu (1).

Previous studies have shown that tamoxifen doses as low as 5 or 1 mg/d exert an equivalent antiproliferative effect on breast cancer cells compared with the standard dose of 20 mg/d (10). In contrast, a linear dose-response relationship was observed on the change in circulating biomarkers of breast cancer and cardiovascular disease known to reflect tamoxifen estrogenicity, including insulin-like growth factor-I (IGF-I), sex hormone binding globulin (SHBG), low-density lipoprotein-cholesterol, C-reactive protein (CRP), fibrinogen, and antithrombin-III, suggesting that a dose reduction of tamoxifen retains a significant antiproliferative effect while reducing its estrogenic effects. Moreover, the long half-life of tamoxifen and its most potent metabolites, endoxifen and 4-hydroxytamoxifen, provides the rationale for an alternative schedule using a single weekly dose in an attempt to reduce toxicity while retaining efficacy (11, 12).

In this study, we compared the biological and pharmacokinetic effect of the combination of 1 mg/d anastrozole and 10 mg/wk tamoxifen with either agent alone in postmenopausal patients with estrogen receptorpositive or progesterone receptorpositive breast intraepithelial neoplasia. The trial was designed to answer the following questions: (a) Does the addition of 10 mg/wk tamoxifen still impair anastrozole bioavailability as 20 mg/d? (b) Can the addition of low-dose tamoxifen counteract anastrozole bone resorption effect? (c) Is the addition of low-dose tamoxifen beneficial to breast carcinogenesis and cardiovascular disease biomarkers over anastrozole alone? (d) Does the addition of low-dose tamoxifen negatively influence anastrozole effect on endometrial tissue? (e) Does anastrozole negatively affect low-dose tamoxifen bioavailability? and (f) Does CYP2D6 genotype, which may influence tamoxifen efficacy in both adjuvant (13) and preventive (14) settings, affect tamoxifen metabolite concentrations and, in turn, the biological effects of tamoxifen-containing treatments?

Study participants

This randomized, open-label phase IIb trial was approved by the Institutional Review Board of the European Institute of Oncology, where it was conducted as a mono-institutional study (study S109). A total of 75 postmenopausal women with excised estrogen receptorpositive and/or progesterone receptorpositive or unknown ductal carcinoma in situ (DIN 2-3) or lobular carcinoma in situ (LIN 2-3) were enrolled. Postmenopausal status was defined as amenorrhea for 12 months and FSH levels >40 IU/L. Written informed consent was obtained from each subject. Women were excluded if they had previous malignancy, grade 2 or greater hematologic and biochemical tests (National Cancer Institute Common Toxicity Criteria version 2), retinal disorders, history of thromboembolic events, untreated osteoporosis, treatment with selective estrogen receptor modulators or aromatase inhibitors in the previous 6 months.

Study procedures

Women were randomly assigned to one of the following three treatment arms: 1 mg/d anastrozole (n = 25), 10 mg/wk tamoxifen (n = 25), or their combination (n = 25) for 12 months. Blood tests and clinical examination were done at baseline and at 3-, 6-, and 12-month visits. At baseline and 12 months, women underwent transvaginal ultrasound (with endometrial biopsy at 12 months) and mammography.

Pathologists and laboratory researchers remained blinded to treatment arm until study completion. Study drugs were purchased by the Institute's Pharmacy using commercially available tamoxifen (10 mg tablets; Nomafen; Fidia) and anastrozole (1 mg tablets; Arimidex; AstraZeneca) specifically labeled and repackaged for the trial. Patients received 7-month drug supply at baseline and at 6-month visit and were instructed to take either 1 tablet/d anastrozole or 1 tablet/wk tamoxifen or both medications for 1 year. Compliance was evaluated by self-reporting calendar, pill count, and blood drug concentrations.

Study endpoints

The main study endpoint was the trough/peak plasma concentration of anastrozole, tamoxifen, and its metabolites at 3, 6, and 12 months. Secondary endpoints were the changes induced by treatment on biomarkers of bone fracture risk, such as C-telopeptide and osteocalcin (15); breast cancer risk, including SHBG, estradiol, estrone, estrone sulfate, IGF-I, and its ratio over insulin-like growth factor binding protein-3 (IGFBP-3; refs. 16,18); and cardiovascular risk, including lipids, coagulation, and ultrasensitive CRP (19, 20). Additionally, endometrial thickness and Ki-67 expression after a 12-month treatment were used as biomarkers of endometrial cancer risk (21).

Assay methods

Fasting blood samples were collected and stored at 80C until centrally assayed. Anastrozole concentrations were measured on sodium heparin-treated samples and analyzed using liquid chromatography with tandem mass spectrometric detection. The standard curve range is from 1.00 to 100 ng/mL for anastrozole using a plasma sample volume of 0.100 mL (Covance Laboratories). Tamoxifen serum concentrations and metabolites (N-desmethyltamoxifen, 4-hydroxytamoxifen, and endoxifen) were measured according to previously described methods (22).

Serum IGF-I, IGFBP-3, and high-sensitivity CRP were determined as described previously (21). Serum SHBG was measured by means of enzyme-labeled chemiluminescent immunometric assays (Diagnostic Products) designed for Immulite. Serum C-telopeptide and osteocalcin were measured with a chemiluminescent immunometric assay (Elecsys 1010 analyzer; Roche Diagnostics). Fibrinogen and antithrombin-III were measured on citrate plasma by ACL 10000 (Beckman Coulter). Estradiol and estrone sulfate serum concentrations were determined as described previously (23).

Expression of Ki-67 antigen (clone Mib-1; Immunotech) was detected in superficial glands and stromal cells scanning on 10 high-power field at 400 using Pipelle biopsies as described previously (21) based on the different proliferative capacity of these endometrial layers (24) and tamoxifen preferential effect on the stromal component (25).

Genomic DNA was extracted from whole-blood specimens by the use of QIAamp DNA blood kit (Qiagen). We employed the INFINITI analyzer (AutoGenomics) for CYP2D6T genotyping according to the protocol of the manufacturer. The test screens for the following CYP2D6 allele variants: *2 (2850C>T), normal activity; *3 (2549delA), *4 (1846G>A), *5 (CYP2D6 deleted), and *6 (1707 delT), no activity; and *9 (2613_2615delAGA), *29 (1659G>A), and *41A (2988G>A), reduced activity. We classified the resulting phenotype as poor metabolizers when there were nonfunctional variants on both alleles, whereas all other combinations were grouped in a single category and considered as extensive metabolizers.

Statistical methods

The sample size was calculated to have 90 power to assess equivalence in anastrozole plasma concentrations through 12 months between anastrozole and combination arms, assuming one-tailed 5 significance and normal distribution of the log anastrozole [c], with SD = 0.41. Drug concentrations and quantitative biomarker measurements were analyzed by means of the repeated-measures ANOVA model at 3, 6, and 12 months, adjusted for baseline values. Treatment and time effects, with their interaction taken as fixed factors and patient effect as random factor, were evaluated in a mixed-effect model (PROC MIXED, SAS). As exploratory investigation, we evaluated the effect of CYP2D6 phenotype as a factor in the mixed-effect models. The analysis on endometrial thickness and Ki-67 expression was carried out with simple ANOVA models at 12 months using available data. Log and square-root transformations were used when necessary to achieve normality.

Two orthogonal contrasts were used to compare biomarker changes among treatment groups: anastrozole versus combination and tamoxifen versus anastrozole or combination. These contrasts were specified a priori and are consistent with the aims of the study. Least square means for the three treatment groups, with P values for contrasts and for time effect, are presented. P values for all the biomarkers were compared with P values adjusted for multiple comparisons (26), with significant P values in bold. All statistical tests were two-sided. Analyses were done using SAS statistical software version 9.0 (SAS Institute).

Of 158 subjects assessed for eligibility, 41 (26) refused to participate and 42 (26.5) were not eligible. Of the 75 randomized participants, 68 completed the 12-month treatment and 7 dropped out. A total of 74 patients were included in the final analysis. One subject was excluded from the analysis for protocol deviation (chronic hepatitis C at baseline). The participant flow diagram is illustrated in Appendix A.

The main subject characteristics and biomarker levels at baseline by allocated treatment group are summarized in Table 1. There were no statistically significant differences among arms, including steroid and HER-2 receptor expression level.

Table 1.

Main subject characteristics and biomarker levels (median and interquartile range) at baseline according to treatment arm

Ana (n = 25)Tam (n = 25)Ana + Tam (n = 25)
Age (y; mean SD) 60.6 5.7 61.1 6.7 59.1 5.7 
Age at menopause (y; mean SD) 50.0 4.6 51.5 2.9 50.2 4.1 
Body mass index (kg/m2; mean SD) 24.9 5.0 26.3 5.1 24.7 4.4 
Ever use of menopause hormone therapy () 36 44 48 
Lobular/ductal intraepithelial neoplasia (n1/24 2/23 2/23 
Radiotherapy () 16 20 28 
Estrogen receptor expression level* 90 (70-90) 90 (90-95) 90 (80-90) 
Progesterone receptor expression level* 28 (1-75) 20 (10-80) 30 (1-70) 
HER-2 overexpression (2/3+; )* 46 39 48 
C-telopeptide (ng/mL) 0.51 (0.39-0.67) 047 (0.27-0.64) 0.46 (0.39-0.57) 
Osteocalcin (ng/mL) 29 (24-39) 26 (19-35) 28 (23-32) 
IGF-I/IGFBP-3 80 (61-93) 63 (57-94) 80 (67-94) 
Estrone sulfate (ng/mL) 0.85 (0.7-1.03) 0.89 (0.7-1.01) 0.86 (0.7-0.97) 
Estradiol/SHBG 0.21 (0.16-0.34) 0.24 (0.16-0.34) 0.18 (0.13-0.23) 
CRP (mg/dL) 0.19 (0.10-0.37) 0.28 (0.10-0.52) 0.11 (0.07-0.29) 
Fibrinogen (mg/dL) 293 (262-318) 287 (249-324) 295 (266-328) 
Antithrombin-III () 96 (89-101) 98 (91-105) 97 (92-103) 
Ana (n = 25)Tam (n = 25)Ana + Tam (n = 25)
Age (y; mean SD) 60.6 5.7 61.1 6.7 59.1 5.7 
Age at menopause (y; mean SD) 50.0 4.6 51.5 2.9 50.2 4.1 
Body mass index (kg/m2; mean SD) 24.9 5.0 26.3 5.1 24.7 4.4 
Ever use of menopause hormone therapy () 36 44 48 
Lobular/ductal intraepithelial neoplasia (n1/24 2/23 2/23 
Radiotherapy () 16 20 28 
Estrogen receptor expression level* 90 (70-90) 90 (90-95) 90 (80-90) 
Progesterone receptor expression level* 28 (1-75) 20 (10-80) 30 (1-70) 
HER-2 overexpression (2/3+; )* 46 39 48 
C-telopeptide (ng/mL) 0.51 (0.39-0.67) 047 (0.27-0.64) 0.46 (0.39-0.57) 
Osteocalcin (ng/mL) 29 (24-39) 26 (19-35) 28 (23-32) 
IGF-I/IGFBP-3 80 (61-93) 63 (57-94) 80 (67-94) 
Estrone sulfate (ng/mL) 0.85 (0.7-1.03) 0.89 (0.7-1.01) 0.86 (0.7-0.97) 
Estradiol/SHBG 0.21 (0.16-0.34) 0.24 (0.16-0.34) 0.18 (0.13-0.23) 
CRP (mg/dL) 0.19 (0.10-0.37) 0.28 (0.10-0.52) 0.11 (0.07-0.29) 
Fibrinogen (mg/dL) 293 (262-318) 287 (249-324) 295 (266-328) 
Antithrombin-III () 96 (89-101) 98 (91-105) 97 (92-103) 

Abbreviations: Ana, anastrozole; Tam, tamoxifen. Biomarker levels are expressed as median and interquartile range.

*Values referred to ductal intraepithelial neoplasia patients only.

Compliance was very high, with all the subjects on actual treatment taking 90 expected drug medication. Most adverse events were grade 1, with no significant difference among arms (data not shown). There were three serious adverse events, one breast cancer and one colorectal cancer in the combination arm and one grade 3 sideropenic anemia in the anastrozole arm.

The results of repeated-measures analysis of circulating concentrations of anastrozole, tamoxifen, and its metabolites at 6 and 12 months are illustrated in Table 2. Mean circulating levels of anastrozole and endoxifen at 3, 6, and 12 months by treatment arm are illustrated in Fig. 1. There was no difference in anastrozole concentration in the combination arm compared with the anastrozole arm through month 12. Conversely, a trend to lower concentrations of endoxifen and N-desmethyltamoxifen was observed at 12 months in the combination arm compared with the tamoxifen arm (P = 0.049 and 0.075, respectively). In a further analysis based on the CYP2D6 genotype, the differences for endoxifen were no longer significant after excluding poor metabolizers [median (ng/mL; interquartile range), 4.65 (2.07-6.49) and 4.6 (3.98-6.85) for the combination and tamoxifen arm, respectively; P = 0.19, Wilcoxon test].

Table 2.

Repeated-measures analysis of drug and metabolite concentrations (ng/mL) at 6 and 12 months by allocated arm (mean and 95 confidence interval)

DrugMonthArmLeast square mean (95 confidence interval)ContrastP
Anastrozole Ana 31 (25-37) Ana vs Ana + Tam 0.794 
Ana + Tam 32 (26-39) 
12 Ana 27 (22-32) Ana vs Ana + Tam 0.224 
Ana + Tam 32 (26-40) 
Tamoxifen Tam 9.4 (6.75-12.34) Tam vs Ana + Tam 0.301 
Ana + Tam 7.35 (5.08-10.04) 
12 Tam 9.03 (6.49-11.99) Tam vs Ana + Tam 0.293 
Ana + Tam 6.98 (4.75-9.65) 
4-Hydroxytamoxifen Tam 0.46 (0.34-0.60) Tam vs Ana + Tam 0.290 
Ana + Tam 0.36 (0.25-0.49) 
12 Tam 0.43 (0.31-0.57) Tam vs Ana + Tam 0.295 
Ana + Tam 0.34 (0.23-0.47) 
N-desmethyltamoxifen Tam 20.86 (15.62-26.86) Tam vs Ana + Tam 0.224 
Ana + Tam 16.25 (11.67-21.59) 
12 Tam 22.25 (16.83-28.44) Tam vs Ana + Tam 0.075 
Ana + Tam 15.40 (10.90-20.66) 
Endoxifen Tam 4.69 (3.52-6.02) Tam vs Ana + Tam 0.281 
Ana + Tam 3.77 (2.73-4.97) 
12 Tam 4.83 (3.65-6.19) Tam vs Ana + Tam 0.049 
Ana + Tam 3.18 (2.22-4.32) 
DrugMonthArmLeast square mean (95 confidence interval)ContrastP
Anastrozole Ana 31 (25-37) Ana vs Ana + Tam 0.794 
Ana + Tam 32 (26-39) 
12 Ana 27 (22-32) Ana vs Ana + Tam 0.224 
Ana + Tam 32 (26-40) 
Tamoxifen Tam 9.4 (6.75-12.34) Tam vs Ana + Tam 0.301 
Ana + Tam 7.35 (5.08-10.04) 
12 Tam 9.03 (6.49-11.99) Tam vs Ana + Tam 0.293 
Ana + Tam 6.98 (4.75-9.65) 
4-Hydroxytamoxifen Tam 0.46 (0.34-0.60) Tam vs Ana + Tam 0.290 
Ana + Tam 0.36 (0.25-0.49) 
12 Tam 0.43 (0.31-0.57) Tam vs Ana + Tam 0.295 
Ana + Tam 0.34 (0.23-0.47) 
N-desmethyltamoxifen Tam 20.86 (15.62-26.86) Tam vs Ana + Tam 0.224 
Ana + Tam 16.25 (11.67-21.59) 
12 Tam 22.25 (16.83-28.44) Tam vs Ana + Tam 0.075 
Ana + Tam 15.40 (10.90-20.66) 
Endoxifen Tam 4.69 (3.52-6.02) Tam vs Ana + Tam 0.281 
Ana + Tam 3.77 (2.73-4.97) 
12 Tam 4.83 (3.65-6.19) Tam vs Ana + Tam 0.049 
Ana + Tam 3.18 (2.22-4.32) 
Fig. 1.

Circulating concentrations of anastrozole and endoxifen at 3, 6, and 12 mo according to treatment group.

Fig. 1.

Circulating concentrations of anastrozole and endoxifen at 3, 6, and 12 mo according to treatment group.

Close modal

The results of repeated-measures analysis of circulating biomarkers at 12 months are summarized in Table 3 with P values for the two contrasts: anastrozole alone versus the combination and tamoxifen versus anastrozole alone or their combination. Six-month data are not shown because they are similar to the 12 month results. C-telopeptide levels increased by 20 in the anastrozole arm relative to baseline, whereas they decreased by 16 in the tamoxifen arm and by 7 in the combination arm (P < 0.001 for both contrasts). Likewise, the addition of low-dose tamoxifen blunted the 22 increase of osteocalcin associated with anastrozole treatment (P < 0.001 for both; Fig. 2).

Table 3.

Repeated-measures analysis of biomarkers at 12 months by allocated arm (mean and 95 confidence interval) and median percentage change from baseline

BiomarkerTreatment armLeast square mean (95 confidence interval) ChangeContrastsP
C-telopeptide* (ng/mL) Ana 0.67 (0.62-0.71) 20 Ana vs Ana + Tam <0.001 
Tam 0.43 (0.38-0.48) 16 Tam vs (Ana or Ana + Tam) <0.001 
Ana + Tam 0.5 (0.45-0.55)   
Osteocalcin* (ng/mL) Ana 36 (34-38) 22 Ana vs Ana + Tam <0.001 
Tam 25 (23-28) 22 Tam vs (Ana or Ana + Tam) <0.001 
Ana + Tam 27 (25-30) 10   
IGF-I/IGFBP-3 Ana 76 (71-81) Ana vs Ana + Tam 0.004 
Tam 64 (59-69) 17 Tam vs (Ana or Ana + Tam) 0.037 
Ana + Tam 66 (61-71) 17   
Estradiol/SHBG Ana 0.18 (0.14-0.21) 29 Ana vs Ana + Tam 0.44 
Tam 0.24 (0.2-0.27) 14 Tam vs (Ana or Ana + Tam) 0.029 
Ana + Tam 0.2 (0.16-0.23) 15   
Estrone sulfate (ng/mL) Ana 0.64 (0.57-0.71) 38 Ana vs Ana + Tam 0.942 
Tam 0.89 (0.8-0.99) Tam vs (Ana or Ana + Tam) <0.001 
Ana + Tam 0.64 (0.57-0.71) 30   
CRP (mg/dL) Ana 0.14 (0.11-0.2) 13 Ana vs Ana + Tam 0.674 
Tam 0.18 (0.13-0.24) 25 Tam vs (Ana or Ana + Tam) 0.429 
Ana + Tam 0.16 (0.12-0.22) 18   
Fibrinogen (mg/dL) Ana 299 (281-316) Ana vs Ana + Tam 0.383 
Tam 276 (259-293) Tam vs (Ana or Ana + Tam) 0.111 
Ana + Tam 288 (270-305)   
Antithrombin-III () Ana 96 (92, 99) Ana vs Ana + Tam 0.757 
Tam 92 (89-95) Tam vs (Ana or Ana + Tam) 0.089 
Ana + Tam 95 (92-98)   
BiomarkerTreatment armLeast square mean (95 confidence interval) ChangeContrastsP
C-telopeptide* (ng/mL) Ana 0.67 (0.62-0.71) 20 Ana vs Ana + Tam <0.001 
Tam 0.43 (0.38-0.48) 16 Tam vs (Ana or Ana + Tam) <0.001 
Ana + Tam 0.5 (0.45-0.55)   
Osteocalcin* (ng/mL) Ana 36 (34-38) 22 Ana vs Ana + Tam <0.001 
Tam 25 (23-28) 22 Tam vs (Ana or Ana + Tam) <0.001 
Ana + Tam 27 (25-30) 10   
IGF-I/IGFBP-3 Ana 76 (71-81) Ana vs Ana + Tam 0.004 
Tam 64 (59-69) 17 Tam vs (Ana or Ana + Tam) 0.037 
Ana + Tam 66 (61-71) 17   
Estradiol/SHBG Ana 0.18 (0.14-0.21) 29 Ana vs Ana + Tam 0.44 
Tam 0.24 (0.2-0.27) 14 Tam vs (Ana or Ana + Tam) 0.029 
Ana + Tam 0.2 (0.16-0.23) 15   
Estrone sulfate (ng/mL) Ana 0.64 (0.57-0.71) 38 Ana vs Ana + Tam 0.942 
Tam 0.89 (0.8-0.99) Tam vs (Ana or Ana + Tam) <0.001 
Ana + Tam 0.64 (0.57-0.71) 30   
CRP (mg/dL) Ana 0.14 (0.11-0.2) 13 Ana vs Ana + Tam 0.674 
Tam 0.18 (0.13-0.24) 25 Tam vs (Ana or Ana + Tam) 0.429 
Ana + Tam 0.16 (0.12-0.22) 18   
Fibrinogen (mg/dL) Ana 299 (281-316) Ana vs Ana + Tam 0.383 
Tam 276 (259-293) Tam vs (Ana or Ana + Tam) 0.111 
Ana + Tam 288 (270-305)   
Antithrombin-III () Ana 96 (92, 99) Ana vs Ana + Tam 0.757 
Tam 92 (89-95) Tam vs (Ana or Ana + Tam) 0.089 
Ana + Tam 95 (92-98)   

NOTE: Bold P values are statistically significant after adjustment for multiple comparisons.

*P < 0.01 for the time by treatment interaction.

Fig. 2.

Serum levels of C-telopeptide (CTX; A), osteocalcin (B), IGF-I/IGFBP-3 (C), and estradiol/SHBG (D) according to treatment group. P values for contrasts between arms at 12 mo.

Fig. 2.

Serum levels of C-telopeptide (CTX; A), osteocalcin (B), IGF-I/IGFBP-3 (C), and estradiol/SHBG (D) according to treatment group. P values for contrasts between arms at 12 mo.

Close modal

As for breast cancer risk biomarkers, the addition of tamoxifen to anastrozole significantly lowered IGF-I/IGFBP-3 levels compared with anastrozole alone (17 versus 9; P = 0.004), whereas the reduction of estradiol/SHBG and estrone sulfate was only lowered, albeit not significantly, in the combination arm relative to the anastrozole arm (15 versus 29 and 30 versus 38, respectively; Fig. 2; Table 3). As expected, estrogen levels were not decreased in the tamoxifen arm.

Cardiovascular disease biomarkers, fibrinogen, antithrombin-III, and CRP, were nonsignificantly lowered in the combination arm relative to the anastrozole arm, especially at 6 months. Total cholesterol/high-density lipoprotein ratio and triglycerides did not significantly change during any treatment (not shown).

The endometrial effects are shown in Appendix B. Endometrial thickness was evaluated by transvaginal ultrasound in 60 patients, the addition of low-dose tamoxifen to anastrozole was not associated with an increased thickness, whereas tamoxifen alone was associated with a greater thickness (least square mean, 2.19, 2.30, and 2.9 mm for anastrozole, combination, and tamoxifen respectively; contrast anastrozole versus combination P = 0.728 and contrast tamoxifen versus anastrozole + combination P = 0.025). The number of evaluable subjects for Ki-67 expression was smaller given the inability to obtain sufficient tissue due to atrophic endometrium in several subjects. However, the combination arm induced a higher proliferation of both the stroma and the superficial gland compartments compared with anastrozole alone (Ki-67 1.21 versus 0.21 in the stroma compartment and 3.85 versus 0.91 in the superficial gland P contrast 0.034 and 0.017, respectively). The contrast tamoxifen versus anastrozole plus the combination was not significant.

Based on CYP2D6 phenotype, there were 5 poor metabolizers and 20 extensive metabolizers in the combination arm and 2 poor metabolizers and 23 extensive metabolizers in the tamoxifen arm (data not shown). Table 4 shows median and interquartile range of endoxifen and biomarkers of tamoxifen activity in a combined analysis of the subjects on tamoxifen (tamoxifen alone and the combination arm) according to CYP2D6 phenotype. Endoxifen levels were lower in poor metabolizers relative to extensive metabolizers (median concentration at 12 months, 4.63 ng/mL in extensive metabolizers versus 3.67 ng/mL in poor metabolizers). Notably, in tamoxifen users, modulation of IGF-I/IGFBP-3 and estradiol/SHBG ratio showed a different trend based on the genotype of the subjects, with no evidence of reduction of IGF-I/IGFBP-3 and estradiol/SHBG ratio in poor metabolizers at 12 months. No difference in endometrial thickness between the two phenotypes was noted (Table 4).

Table 4.

Endoxifen and biomarker levels (median and interquartile range) according to CYP2D6 phenotypes in all tamoxifen users

MonthsPhenotypenMedianQ1Q3
Endoxifen (ng/mL) Extensive metabolizers 43 4.62 3.52 5.62 
12 Extensive metabolizers 42 4.63 2.98 6.62 
Poor metabolizers 5.00 4.18 6.79 
12 Poor metabolizers 3.67 3.03 5.13 
IGF-I/IGFBP-3 Extensive metabolizers 43 72.11 58.37 98.69 
12 Extensive metabolizers 42 58.80 47.47 79.47 
Poor metabolizers 77.91 52.20 94.53 
12 Poor metabolizers 71.53 50.38 80.06 
Estradiol/SHBG Extensive metabolizers 43 0.19 0.13 0.28 
12 Extensive metabolizers 42 0.15 0.11 0.24 
Poor metabolizers 0.14 0.11 0.21 
12 Poor metabolizers 0.15 0.12 0.16 
Endometrial thickness (mm) Extensive metabolizers 37 2.3 1.6 
12 Extensive metabolizers 35 2.4 1.8 3.7 
Poor metabolizers 1.9 3.6 
12 Poor metabolizers 3.1 1.8 4.1 
MonthsPhenotypenMedianQ1Q3
Endoxifen (ng/mL) Extensive metabolizers 43 4.62 3.52 5.62 
12 Extensive metabolizers 42 4.63 2.98 6.62 
Poor metabolizers 5.00 4.18 6.79 
12 Poor metabolizers 3.67 3.03 5.13 
IGF-I/IGFBP-3 Extensive metabolizers 43 72.11 58.37 98.69 
12 Extensive metabolizers 42 58.80 47.47 79.47 
Poor metabolizers 77.91 52.20 94.53 
12 Poor metabolizers 71.53 50.38 80.06 
Estradiol/SHBG Extensive metabolizers 43 0.19 0.13 0.28 
12 Extensive metabolizers 42 0.15 0.11 0.24 
Poor metabolizers 0.14 0.11 0.21 
12 Poor metabolizers 0.15 0.12 0.16 
Endometrial thickness (mm) Extensive metabolizers 37 2.3 1.6 
12 Extensive metabolizers 35 2.4 1.8 3.7 
Poor metabolizers 1.9 3.6 
12 Poor metabolizers 3.1 1.8 4.1 

Aromatase inhibitors have become a standard endocrine treatment of postmenopausal women with breast cancer (2,4) and are being assessed in at-risk individuals, but there is concern on their long-term safety profile, particularly at bone level. Our proof-of-principle trial aimed at determining whether, in contrast to the standard dose of 20 mg/d, the addition of a weekly low dose of tamoxifen might not impair anastrozole availability but, in fact, improve its safety profile.

Our results indicate that the addition of 10 mg/wk tamoxifen does not affect anastrozole bioavailability as measured by serial plasma concentrations up to 12 months. Conversely, endoxifen levels were lower in the combination arm, but the difference at 12 months was no more significant after excluding CYP2D6 poor metabolizers. Although the frequency of poor metabolizers in our study was comparable with the general population and therefore the power of the analysis is low, these results indicate that the reduction of endoxifen levels may be related to the patient phenotype and not to the drug combination.

Importantly, the addition of low-dose tamoxifen to anastrozole provided a significant anti-resorption effect on bone metabolism as shown by the blunting effect on C-telopeptide and osteocalcin increase induced by anastrozole alone. As observed previously (27), an increase in C-telopeptide levels by 20 after 1 year of anastrozole treatment was associated with 50 higher bone fracture risk in the Anastrozole, Tamoxifen Alone or in Combination trial, where the annual bone fracture rate in the intervention period was 2.93 in the anastrozole arm versus 1.90 in the tamoxifen arm (28). Also, a decrease in osteocalcin during different drug interventions is associated with a bone formation effect in several prospective studies (29,31).

Tamoxifen added to anastrozole induced a significant decline of free IGF-I and did not significantly affect the estrogen drop induced by anastrozole alone, whereas cardiovascular risk biomarkers were not differently modulated by the addition of low-dose tamoxifen. The 8 decline of IGF-I in the combination arm may be clinically relevant, because several prospective studies have shown an association between higher IGF-I and IGF-I/IGFBP-3 levels and breast cancer risk (18, 32). Interestingly, we found a clear association between biomarker modulation by tamoxifen and patient CYP2D6 phenotype, with no evidence of a decrease of IGF-I/IGFBP-3 and estradiol/SHBG ratio in poor metabolizers. In a recent model applied to an adjuvant endocrine trial (33), it was shown that tamoxifen could be as effective as letrozole if poor metabolizers were excluded.

Notably, no increase in endometrial thickness with the addition of tamoxifen to anastrozole alone was detected, whereas tamoxifen alone was associated with a higher thickness. In a small subset of women with available endometrial tissue, however, tamoxifen addition induced a higher proliferation of both glandular and stromal compartments relative to anastrozole alone. Yet, the expression of Ki-67 in the endometrial gland was as low as 2 to 3, much lower than that observed in normal endometrium adjacent to endometrial cancer (34), or under treatment with 20 mg/d tamoxifen in hyperplastic or premalignant tissue (35, 36), suggesting that the addition of low-dose tamoxifen should not lead to an increased risk of endometrial cancer. CYP2D6 genotype did not influence endometrial thickness changes under tamoxifen, but the power of this observation is too limited to draw reliable conclusions.

Our findings confirm the ability of a weekly administration of 10 mg tamoxifen to exert significant biological effects. In a previous dose-ranging trial in healthy hormone replacement therapy users (21), 10 mg/wk tamoxifen mainly exerted antiestrogenic effects on biomarkers of breast carcinogenesis and no increase in endometrial proliferation and menopausal symptoms. Conversely, in the present study, tamoxifen exhibited mild estrogenic effects in an estrogen-deprived environment such as that attained by anastrozole alone. Likewise, at bone level, the drug has some antiestrogenic effects in premenopausal women and estrogenic effects in postmenopausal women, including a significant bone fracture risk reduction (37, 38). Moreover, endometrial changes during tamoxifen differ significantly in premenopausal versus postmenopausal women, with an increased endometrial cancer risk observed predominantly after the menopause (39, 40). Another factor influencing tamoxifen estrogenicity is the dosage, where a daily dose reduction from 20 to 1 mg tends to attenuate its estrogenic effects in postmenopausal women without impairing its antiproliferative effects on breast cancer cells (10).

The results of the present study may help to interpret the complex results of the Anastrozole, Tamoxifen Alone or in Combination trial, where the combination arm was inferior to anastrozole alone and possibly slightly worse than the tamoxifen arm alone in terms of disease-free survival, underlying the clinical relevance of the hormonal environment and the dose in determining tamoxifen efficacy, an issue that has not sufficiently been addressed thus far. At a 20 mg/d dose, tamoxifen was shown to lower anastrozole concentrations by 30, although this effect did not impair the estrogenic drop induced by anastrozole (9) and has not been considered important to explain the lower efficacy of the combination arm. A second, most likely, explanation encompasses the shift to a predominant estrogenic effect of tamoxifen when administered at the dose of 20 mg/d in an estrogen-deprived environment, thus explaining why the combination arm (weak estrogen) was not only inferior to the anastrozole arm (pure antiestrogen) but also to the tamoxifen arm (partial antiestrogen) as regards disease-free survival. Our findings indicate that a dose reduction of tamoxifen by 15 times still retains some desirable estrogenic properties when added to anastrozole alone, such as decreased bone turnover and increase SHBG levels, suggesting that the combination may have a more favorable pharmacodynamic profile over anastrozole alone.

The right balance between estrogenic and antiestrogenic effects in both pharmacologic treatment and prevention of breast cancer is the subject of intense debate. In the advanced setting, prolonged tamoxifen use leads to resistant tumors in animal models and in humans mostly because of the emergence of tamoxifen-dependent clones, known to respond to aromatase inhibitor treatment (41). However, a prolonged estrogen deprivation, such as during anastrozole treatment, may induce resistance for the emergence of estrogen hypersensitive clones (42), which may paradoxically respond to estrogen restoration (43). In the adjuvant setting, the results of a recent trial indicate that the bone loss induced by anastrozole can be prevented by the association with biphosphonates and this may lead to a benefit also in terms of breast cancer outcome (44, 45). In the prevention setting, the use of biphosphonates to compensate bone loss by aromatase inhibitor is also being investigated with initial promising results (46). Our results provide an alternative strategy, the addition of a very low dose of tamoxifen that, based on surrogate endpoints, counteracts the accelerated bone resorption induced by anastrozole and maintains a favorable profile for breast cancer risk biomarkers. In line with this approach, a further option may contemplate the use of a lower aromatase inhibitor dose or the addition of low-dose estrogen to aromatase inhibitors (47). In the prevention setting, where safety is as important as efficacy, a drug combination approach may be more reasonable as largely proven in cardiovascular medicine where control on multiple pathways reaches better results and attenuates adverse events (48). The lesson from selective COX-2 inhibitors has shown that a superselective approach, albeit very effective, may even become harmful, especially in cancer prevention where long-term intervention is required (49).

Our study has several limitations, including lack of blinding and of more definitive breast cancer biomarkers such as mammographic density or tissue samples. Moreover, we were unable to measure bone mineral density changes pretreatment and post-treatment. Despite these limitations, our proof-of-principle study shows that the combination of a standard dose of anastrozole with a low, weekly dose of tamoxifen does not affect anastrozole blood concentrations, is safe and well tolerated, and can diminish anastrozole adverse effects at bone level in postmenopausal women. These findings provide the rationale for larger clinical studies in both the prevention and the adjuvant setting.

No potential conflicts of interest were disclosed.

We thank Astra Zeneca for providing at no cost anastrozole tablets and its plasma level measurements; Medical Systems for providing instruments for CYP analysis; Lega Italiana per la Lotta contro i Tumori, Italian Ministry of Health (Ricerca Finalizzata); and Fondazione Italiana per la Ricerca sul Cancro for financial support; and Giorgia Bollani for excellent contribution as data manager.

1
Baum
M
,
Buzdar
AU
,
Cuzick
J
, et al
. 
Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomised trial
.
Lancet
2002
;
359
:
2131
9
.
2
Thurlimann
B
,
Keshaviah
A
,
Coates
AS
, et al
. 
A comparison of letrozole and tamoxifen in postmenopausal women with early breast cancer
.
N Engl J Med
2005
;
353
:
2747
57
.
3
Goss
PE
,
Ingle
JN
,
Martino
S
, et al
. 
A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer
.
N Engl J Med
2003
;
349
:
1793
802
.
4
Jakesz
R
,
Jonat
W
,
Gnant
M
, et al
. 
Switching of postmenopausal women with endocrine-responsive early breast cancer to anastrozole after 2 years' adjuvant tamoxifen: combined results of ABCSG trial 8 and ARNO 95 trial
.
Lancet
2005
;
366
:
455
62
.
5
Cuzick
J
. 
Chemoprevention of breast cancer
.
Breast Cancer
2008
;
15
:
10
6
.
6
Dowsett
M
,
Martin
LA
,
Smith
I
,
Johnston
S
. 
Mechanisms of resistance to aromatase inhibitors
.
J Steroid Biochem Mol Biol
2005
;
95
:
167
72
.
7
Nabholtz
JM
,
Buzdar
A
,
Pollak
M
, et al
. 
Anastrozole is superior to tamoxifen as first-line therapy for advanced breast cancer in postmenopausal women: results of a North American multicenter randomized trial. Arimidex Study Group
.
J Clin Oncol
2000
;
18
:
3758
67
.
8
Dowsett
M
,
Tobias
JS
,
Howell
A
, et al
. 
The effect of anastrozole on the pharmacokinetics of tamoxifen in post-menopausal women with early breast cancer
.
Br J Cancer
1999
;
79
:
311
5
.
9
Dowsett
M
,
Cuzick
J
,
Howell
A
,
Jackson
I
. 
Pharmacokinetics of anastrozole and tamoxifen alone, and in combination, during adjuvant endocrine therapy for early breast cancer in postmenopausal women: a sub-protocol of the Arimidex and tamoxifen alone or in combination (ATAC) trial
.
Br J Cancer
2001
;
85
:
317
24
.
10
Decensi
A
,
Robertson
C
,
Viale
G
, et al
. 
A randomized trial of low-dose tamoxifen on breast cancer proliferation and blood estrogenic biomarkers
.
J Natl Cancer Inst
2003
;
95
:
779
90
.
11
Guerrieri-Gonzaga
A
,
Baglietto
L
,
Johansson
H
, et al
. 
Correlation between tamoxifen elimination and biomarker recovery in a primary prevention trial
.
Cancer Epidemiol Biomarkers Prev
2001
;
10
:
967
70
.
12
Lien
EA
,
Solheim
E
,
Ueland
PM
. 
Distribution of tamoxifen and its metabolites in rat and human tissues during steady-state treatment
.
Cancer Res
1991
;
51
:
4837
44
.
13
Goetz
MP
,
Rae
JM
,
Suman
VJ
, et al
. 
Pharmacogenetics of tamoxifen biotransformation is associated with clinical outcomes of efficacy and hot flashes
.
J Clin Oncol
2005
;
23
:
9312
8
.
14
Bonanni
B
,
Macis
D
,
Maisonneuve
P
, et al
. 
Polymorphism in the CYP2D6 tamoxifen-metabolizing gene influences clinical effect but not hot flashes: data from the Italian Tamoxifen Trial
.
J Clin Oncol
2006
;
24
:
3708
9
.
15
Gerdhem
P
,
Ivaska
KK
,
Alatalo
SL
, et al
. 
Biochemical markers of bone metabolism and prediction of fracture in elderly women
.
J Bone Miner Res
2004
;
19
:
386
93
.
16
Key
T
,
Appleby
P
,
Barnes
I
,
Reeves
G
. 
Endogenous sex hormones and breast cancer in postmenopausal women: reanalysis of nine prospective studies
.
J Natl Cancer Inst
2002
;
94
:
606
16
.
17
Renehan
AG
,
Egger
M
,
Minder
C
,
O'Dwyer
ST
,
Shalet
SM
,
Zwahlen
M
. 
IGF-I, IGF binding protein-3 and breast cancer risk: comparison of 3 meta-analyses
.
Int J Cancer
2005
;
115
:
1006
7
.
18
Kaaks
R
,
Lundin
E
,
Rinaldi
S
, et al
. 
Prospective study of IGF-I, IGF-binding proteins, and breast cancer risk, in northern and southern Sweden
.
Cancer Causes Control
2002
;
13
:
307
16
.
19
Ridker
PM
,
Hennekens
CH
,
Buring
JE
,
Rifai
N
. 
C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women
.
N Engl J Med
2000
;
342
:
836
43
.
20
Vogel
VG
. 
Can statin therapy reduce the risk of breast cancer?
J Clin Oncol
2005
;
23
:
8553
5
.
21
Decensi
A
,
Gandini
S
,
Serrano
D
, et al
. 
Randomized dose-ranging trial of tamoxifen at low doses in hormone replacement therapy users
.
J Clin Oncol
2007
;
25
:
4201
9
.
22
Gjerde
J
,
Kisanga
ER
,
Hauglid
M
,
Holm
PI
,
Mellgren
G
,
Lien
EA
. 
Identification and quantification of tamoxifen and four metabolites in serum by liquid chromatography-tandem mass spectrometry
.
J Chromatogr A
2005
;
1082
:
6
14
.
23
Johansson
H
,
Bonanni
B
,
Mariette
F
, et al
. 
Effects of raloxifene on sex steroid hormones and C-telopeptide in postmenopausal women with primary breast cancer
.
Breast Cancer Res Treat
2006
;
98
:
167
72
.
24
Shiozawa
T
,
Li
SF
,
Nakayama
K
,
Nikaido
T
,
Fujii
S
. 
Relationship between the expression of cyclins/cyclin-dependent kinases and sex-steroid receptors/Ki67 in normal human endometrial glands and stroma during the menstrual cycle
.
Mol Hum Reprod
1996
;
2
:
745
52
.
25
Decensi
A
,
Fontana
V
,
Bruno
S
,
Gustavino
C
,
Gatteschi
B
,
Costa
A
. 
Effect of tamoxifen on endometrial proliferation
.
J Clin Oncol
1996
;
14
:
434
40
.
26
Benjamini
Y
,
Hochberg
Y
. 
Controlling the false discovery rates: a practical and powerful approach to multiple testing
.
J R Stat Soc B
1995
;
35
:
289
300
.
27
Eastell
R
,
Hannon
RA
,
Cuzick
J
,
Dowsett
M
,
Clack
G
,
Adams
JE
. 
Effect of an aromatase inhibitor on BMD and bone turnover markers: 2-year results of the Anastrozole, Tamoxifen, Alone or in Combination (ATAC) trial (18233230)
.
J Bone Miner Res
2006
;
21
:
1215
23
.
28
Forbes
JF
,
Cuzick
J
,
Buzdar
A
,
Howell
A
,
Tobias
JS
,
Baum
M
. 
Effect of anastrozole and tamoxifen as adjuvant treatment for early-stage breast cancer: 100-month analysis of the ATAC trial
.
Lancet Oncol
2008
;
9
:
45
53
.
29
Lenora
J
,
Ivaska
KK
,
Obrant
KJ
,
Gerdhem
P
. 
Prediction of bone loss using biochemical markers of bone turnover
.
Osteoporos Int
2007
;
18
:
1297
305
.
30
Confavreux
CB
,
Fontana
A
,
Guastalla
JP
,
Munoz
F
,
Brun
J
,
Delmas
PD
. 
Estrogen-dependent increase in bone turnover and bone loss in postmenopausal women with breast cancer treated with anastrozole. Prevention with bisphosphonates
.
Bone
2007
;
41
:
346
52
.
31
Zidan
J
,
Keidar
Z
,
Basher
W
,
Israel
O
. 
Effects of tamoxifen on bone mineral density and metabolism in postmenopausal women with early-stage breast cancer
.
Med Oncol
2004
;
21
:
117
21
.
32
Hankinson
SE
,
Willett
WC
,
Colditz
GA
, et al
. 
Circulating concentrations of insulin-like growth factor-I and risk of breast cancer
.
Lancet
1998
;
351
:
1393
6
.
33
Punglia
RS
,
Burstein
HJ
,
Winer
EP
,
Weeks
JC
. 
Pharmacogenomic variation of CYP2D6 and the choice of optimal adjuvant endocrine therapy for postmenopausal breast cancer: a modeling analysis
.
J Natl Cancer Inst
2008
;
100
:
642
8
.
34
Sivridis
E
,
Giatromanolaki
A
. 
Proliferative activity in postmenopausal endometrium: the lurking potential for giving rise to an endometrial adenocarcinoma
.
J Clin Pathol
2004
;
57
:
840
4
.
35
Elkas
J
,
Armstrong
A
,
Pohl
J
,
Cuttitta
F
,
Martinez
A
,
Gray
K
. 
Modulation of endometrial steroid receptors and growth regulatory genes by tamoxifen
.
Obstet Gynecol
2000
;
95
:
697
703
.
36
Mourits
MJ
,
ten Hoor
KA
,
van der Zee
AG
,
Willemse
PH
,
de Vries
EG
,
Hollema
H
. 
The effects of tamoxifen on proliferation and steroid receptor expression in postmenopausal endometrium
.
J Clin Pathol
2002
;
55
:
514
9
.
37
Powles
TJ
,
Hickish
T
,
Kanis
JA
,
Tidy
A
,
Ashley
S
. 
Effect of tamoxifen on bone mineral density measured by dual-energy X-ray absorptiometry in healthy premenopausal and postmenopausal women
.
J Clin Oncol
1996
;
14
:
78
84
.
38
Vogel
VG
,
Costantino
JP
,
Wickerham
DL
, et al
. 
Effects of tamoxifen vs raloxifene on the risk of developing invasive breast cancer and other disease outcomes: the NSABP Study of Tamoxifen and Raloxifene (STAR) P-2 trial
.
JAMA
2006
;
295
:
2727
41
.
39
Guerrieri-Gonzaga
A
,
Robertson
C
,
Bonanni
B
, et al
. 
Preliminary results on safety and activity of a randomized, double-blind, 2 2 trial of low-dose tamoxifen and fenretinide for breast cancer prevention in premenopausal women
.
J Clin Oncol
2006
;
24
:
129
35
.
40
Fisher
B
,
Costantino
JP
,
Wickerham
DL
, et al
. 
Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study
.
J Natl Cancer Inst
1998
;
90
:
1371
88
.
41
Buzdar
AU
,
Jonat
W
,
Howell
A
, et al
. 
Anastrozole versus megestrol acetate in the treatment of postmenopausal women with advanced breast carcinoma: results of a survival update based on a combined analysis of data from two mature phase III trials. Arimidex Study Group
.
Cancer
1998
;
83
:
1142
52
.
42
Chen
S
,
Masri
S
,
Wang
X
,
Phung
S
,
Yuan
YC
,
Wu
X
. 
What do we know about the mechanisms of aromatase inhibitor resistance?
J Steroid Biochem Mol Biol
2006
;
102
:
232
40
.
43
Osipo
C
,
Gajdos
C
,
Cheng
D
,
Jordan
VC
. 
Reversal of tamoxifen resistant breast cancer by low dose estrogen therapy
.
J Steroid Biochem Mol Biol
2005
;
93
:
249
56
.
44
Gnant
M
,
Mlineritsch
B
,
Schippinger
W
, et al
. 
Endocrine therapy plus zoledronic acid in premenopausal breast cancer
.
N Engl J Med
2009
;
360
:
679
91
.
45
Brufsky
A
,
Bundred
N
,
Coleman
R
, et al
. 
Integrated analysis of zoledronic acid for prevention of aromatase inhibitor-associated bone loss in postmenopausal women with early breast cancer receiving adjuvant letrozole
.
Oncologist
2008
;
13
:
503
14
.
46
Cuzick
J
. 
IBIS II: a breast cancer prevention trial in postmenopausal women using the aromatase inhibitor anastrozole
.
Expert Rev Anticancer Ther
2008
;
8
:
1377
85
.
47
Frank
DH
,
Kimler
BF
,
Fabian
CJ
, et al
. 
Digital image analysis of breast epithelial cells collected by random periareolar fine-needle aspirates (RPFNA) from women at high risk for breast cancer taking hormone replacement and the aromatase inhibitor, letrozole, for six months
.
Breast Cancer Res Treat
2009
;
115
:
661
8
.
48
Yusuf
S
,
Pais
P
,
Afzal
R
, et al
. 
Effects of a polypill (Polycap) on risk factors in middle-aged individuals without cardiovascular disease (TIPS): a phase II, double-blind, randomised trial
.
Lancet
2009
;
373
:
1341
51
.
49
Solomon
SD
,
Pfeffer
MA
,
McMurray
JJ
, et al
. 
Effect of celecoxib on cardiovascular events and blood pressure in two trials for the prevention of colorectal adenomas
.
Circulation
2006
;
114
:
1028
35
.

Competing Interests

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.