1097-0142/asset/cover.gif?v=1&s=a7299bc18f075294c232ade468773cd0672bd470 "Cancer")
The following individuals are members of the ATAC Trial Steering Committee (asterisks denote members of the writing group for the current article): *Professor M. Baum (founding chairman and principal investigator for the main ATAC trial), University College London, London, United Kingdom; Professor A.R. Bianco, Universita degli Studi di Napoli Federico II, Naples, Italy; *Dr. A. Buzdar (vice chairperson elect for the main ATAC trial), The University of Texas M.D. Anderson Cancer Center, Houston, Texas; Dr. M. Coibion, Institut Bordet, Brussels, Belgium; Professor R Coleman, Cancer Research Centre, Weston Park Hospital, Sheffield, United Kingdom; Dr. M. Constenla, Hospital Montecelo, Pontevedra, Spain; *Professor J. Cuzick (independent statistician), Cancer Research UK, London, United Kingdom; Professor Dr. W. Distler, Carl Gustav Carus Universitætsklinkan Dresden, Dresden, Germany; Professor M. Dowsett, The Royal Marsden Hospital, London, United Kingdom; *Professor J. Forbes, Newcastle Mater Misericordiae Hospital, Newcastle, Australia; Professor W.D. George, Beatson Oncology Centre, Western Infirmary, Glasgow, United Kingdom; Sr. J. Gray, Belfast City Hospital, Belfast, United Kingdom; Dr. J.P. Guastalla, Centre Leon Berard, Lyon, France; *Mrs. J. Houghton and Dr. N. Williams, Clinical Trials Group of the Department of Surgery, University College London, London, United Kingdom; *Professor A. Howell (current chairman for the main ATAC trial), Christie Hospital, Manchester, United Kingdom; Professor Dr. J.G.M. Klijn, Dr. Daniel den Hoed Kliniek, Rotterdam, The Netherlands, and University Hospital Rotterdam, Rotterdam, The Netherlands; Dr. G.Y. Locker, Evanston Hospital, Kellogg Cancer Care Center, Evanston Illinois; Dr. J. Mackey, Cross Cancer Institute, Edmonton, Alberta, Canada; Professor R.E. Mansel, University of Wales College of Medicine, Cardiff, United Kingdom; Professor J.M. Nabholtz, University of California–Los Angeles, Los Angeles, California; Dr. T. Nagykalnai, Uzsoki University Hospital, Budapest, Hungary; Dr. A. Nicolucci, GIVIO Co-ordinating Centre, Consorzio Mario Negri Sud, Centro Di Ricerche Farmacologiche e Biomediche, Chieti, Italy; Dr. U. Nylen, Radiumhemmet, Karolinska Sjukhuset, Stockholm, Sweden; *Dr. T. Sahmoud and Dr. R. Hellmund, AstraZeneca Pharmaceuticals, Wilmington, Delaware; Mr. R. Sainsbury, University College London, London, United Kingdom; Dr. N. Griffiths and Dr. G. Hoctin-Boes, AstraZeneca Pharmaceuticals, Macclesfield, United Kingdom; Professor J.S. Tobias, The Meyerstein Institute of Clinical Oncology, Middlesex Hospital, London, United Kingdom.
1097-0142/asset/olbannerleft.gif?v=1&s=ca681f5719430b26e1bc15e9ea4c9fc0a7110104)
1097-0142/asset/olbannerright.gif?v=1&s=8142566facf7e76aef9be6c51162a2e920b3b9f9)
Original Article
Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early-stage breast cancer
Results of the ATAC (Arimidex, Tamoxifen Alone or in Combination) trial efficacy and safety update analyses
Article first published online: 29 SEP 2003
DOI: 10.1002/cncr.11745
Copyright © 2003 American Cancer Society
Additional Information
How to Cite
The ATAC (Arimidex, Tamoxifen Alone or in Combination) Trialists' Group (2003), Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early-stage breast cancer. Cancer, 98: 1802–1810. doi: 10.1002/cncr.11745
Publication History
- Issue published online: 22 OCT 2003
- Article first published online: 29 SEP 2003
- Manuscript Accepted: 30 JUN 2003
- Manuscript Revised: 23 JUN 2003
- Manuscript Received: 13 MAY 2003
- Abstract
- Article
- References
- Cited By
Keywords:
- anastrozole;
- tamoxifen;
- aromatase inhibitor;
- early-stage breast cancer;
- postmenopausal women;
- ATAC;
- randomized clinical trial;
- adjuvant
Abstract
BACKGROUND
The first analysis of the ATAC (Arimidex, Tamoxifen Alone or in Combination) trial (median follow-up, 33 months) demonstrated that in adjuvant endocrine therapy for postmenopausal patients with early-stage breast cancer, anastrozole was superior to tamoxifen in terms of disease-free survival (DFS), time to recurrence (TTR), and incidence of contralateral breast cancer (CLBC). In the current article, the results of the first efficacy update, based on a median follow-up period of 47 months, are reported along with the results of an updated safety analysis, performed 7 months after the first analysis (median duration of treatment, 36.9 months).
METHODS
DFS, TTR, CLBC incidence, and safety were assessed in the same patient group as in the first analysis of the ATAC trial.
RESULTS
DFS estimates at 4 years remained significantly more favorable (86.9% vs. 84.5%, respectively) for patients receiving anastrozole compared with those receiving tamoxifen (hazard ratio [HR], 0.86; 95% confidence interval [CI], 0.76–0.99; P = 0.03). The benefit generated by anastrozole in terms of DFS was even greater in patients with hormone receptor–positive tumors (HR, 0.82; 95% CI, 0.70–0.96; P = 0.014). The HR for TTR also indicated a significant benefit for patients receiving anastrozole compared with those receiving tamoxifen (HR, 0.83; 95% CI, 0.71–0.96; P = 0.015), with additional benefit for patients with hormone receptor–positive tumors (HR, 0.78; 95% CI, 0.65–0.93; P = 0.007). CLBC incidence data also continued to favor anastrozole (odds ratio [OR], 0.62; 95% CI, 0.38–1.02; P = 0.062), and statistical significance was achieved in the hormone receptor–positive subgroup (OR, 0.56; 95% CI, 0.32–0.98; P = 0.042). The updated safety analysis also confirmed the findings of the first analysis, in that endometrial cancer (P = 0.007), vaginal bleeding and discharge (P < 0.001 for both), cerebrovascular events (P < 0.001), venous thromboembolic events (P < 0.001), and hot flashes (P < 0.001) all occurred less frequently in the anastrozole group, whereas musculoskeletal disorders and fractures (P < 0.001 for both) continued to occur less frequently in the tamoxifen group. These results indicated that the safety profile of anastrozole remained consistent.
CONCLUSIONS
After an additional follow-up period, anastrozole continues to show superior efficacy, which is most apparent in the clinically relevant hormone receptor–positive population. Furthermore, anastrozole has numerous noteworthy advantages in terms of tolerability compared with tamoxifen. These findings suggest that the benefits of anastrozole are likely to be maintained in the long term and provide further support for the status of anastrozole as a valid treatment option for postmenopausal women with hormone-sensitive early-stage breast cancer. Cancer 2003. © 2003 American Cancer Society.
The results of the first analysis of the ATAC (Arimidex, Tamoxifen Alone or in Combination) trial indicated that disease-free survival (DFS) and time to recurrence (TTR) were significantly prolonged for patients who received anastrozole (Arimidex; AstraZeneca, Wilmington, DE) alone compared with those who received tamoxifen alone or anastrozole and tamoxifen together. Survival data were based on 1079 first events, and the median follow-up time for the first DFS analysis was 33 months.1
At the time of that analysis, concerns were raised over the immaturity of the data and the fact that the benefits of tamoxifen persist beyond the 5 years of active treatment.2 Therefore, it is important to determine whether the initial superiority of anastrozole compared with tamoxifen would continue to be observed after additional follow-up.
In the current article, we report the results of the first updated efficacy analysis, based on a median follow-up period of 47 months (this analysis hereafter will be referred to as the 47-month update analysis), and the first updated safety analysis, based on an additional 7 months of follow-up beyond the first analysis and a median therapy duration of 36.9 months. The updated safety analysis was performed in accordance with regulatory requirements at a prespecified time. The data cutoff for the updated efficacy analysis was approximately 12 months after the first analysis. The first analysis was scheduled to take place after 1056 events (all breast cancer events plus deaths before recurrence) had occurred. At the time of the data cutoff, this number had been exceeded, the actual number of first events being 1079; 850 of these events were breast cancer events, and the remaining 229 were deaths without breast cancer recurrence and therefore were not directly applicable to the efficacy evaluation. For the 47-month update analysis, the total number of events observed was 1373, of which 1055 were breast cancer events (where deaths before recurrence had been censored); therefore, the clinically important endpoint of breast cancer events also reached the number of occurrences on which the first analysis was based. The updated efficacy results were first presented at the San Antonio Breast Cancer Symposium (December 11–14, 2002, San Antonio, TX).3 Updated safety data also were presented at the 2002 San Antonio Breast Cancer Symposium.4
Based on the findings of the first analysis, it was decided that combination therapy (anastrozole plus tamoxifen) should be discontinued, because it demonstrated no benefit compared with tamoxifen monotherapy in terms of either efficacy or tolerability. For completeness, however, the results from the combination therapy arm are presented as supplementary data in the current update.
MATERIALS AND METHODS
Full details of the trial design, methodology, primary objectives, and endpoints are provided in the report on the initial efficacy analysis.1
In the time between the first efficacy analysis and the updated analysis, there was a change in the censoring methodology such that a data cutoff 12 months after the cutoff for the first analysis caused the median follow-up period for DFS to increase from 33 months to 47 months. In the first analysis, the approach was to censor at the time of the most recent trial visit before June 29, 2001, whereas in the 47-month update, for patients who had not experienced an event and who still were being followed, censoring occurred on June 28, 2002. This change was introduced to give a more accurate representation of the extent of follow-up, which was understated in the first analysis by approximately 2–3 months.
RESULTS
At the time of the updated analysis, 46% (4310 of 9366) of patients had been followed for at least 4 years.
Efficacy Endpoints
A total of 1373 first events were recorded (Table 1), of which 1055 (77%) were recurrences or new contralateral breast tumors and 318 (23%) were deaths without recurrence.
| Anastrozole (n = 3125) | Tamoxifen (n = 3116) | Combination therapy (n = 3125) | Total (n = 9366) | |
|---|---|---|---|---|
| ||||
| All first events | 413 | 472 | 488 | 1373 |
| Locoregional events | 84 | 101 | 107 | 292 |
| Distant eventsa | 195 | 222 | 246 | 663 |
| CLBC (invasive) | 20 | 35 | 30 | 85 |
| CLBC (DCIS) | 5 | 5 | 5 | 15 |
| Deaths without recurrence | 109 | 109 | 100 | 318 |
DFS
The primary endpoint, DFS, was significantly longer for patients receiving anastrozole alone compared with patients receiving tamoxifen alone (hazard ratio [HR], 0.86; 95% confidence interval [CI], 0.76–0.99; P = 0.03). Similarly, among patients with hormone receptor–positive breast cancer, the anastrozole arm fared significantly better than the tamoxifen arm (HR, 0.82; 95% CI, 0.70–0.96; P = 0.014). There were no significant differences in DFS between the tamoxifen arm and the combination arm, either in the overall population (HR, 1.04; 95% CI, 0.92–1.19; P = 0.5) or among patients with hormone receptor–positive disease (HR, 1.03; 95% CI, 0.89–1.20; P = 0.7). The absolute difference in DFS in favor of the anastrozole arm increased over time, from 1.5% in the overall intention-to-treat (ITT) population and 1.7% in the hormone receptor–positive subgroup at 3 years to 2.4% in the overall ITT population and 2.9% in the hormone receptor–positive subgroup at 4 years.
Breast cancer events (TTR)
As in the first analysis, the HR for TTR indicated an advantage in the anastrozole arm compared with the tamoxifen arm (HR, 0.83; 95% CI, 0.71–0.96; P = 0.015) (Fig. 1). Among patients with hormone receptor–positive tumors, patients receiving anastrozole showed even more improvement in terms of TTR compared with patients receiving tamoxifen (HR, 0.78; 95% CI, 0.65–0.93; P = 0.007) (Fig. 2). There were no significant differences between the tamoxifen arm and the combination arm, either in the overall population (HR, 1.08; 95% CI, 0.94–1.24; P = 0.3) or in the hormone receptor–positive subgroup (HR, 1.10; 95% CI, 0.93–1.30; P = 0.3). The absolute difference in breast cancer event rates between the anastrozole and tamoxifen arms increased over time such that by Year 4, the absolute difference was 2.3% for the overall ITT population and 2.6% for patients with hormone receptor–positive tumors.
Figure 1. Probability of recurrence in the intention-to-treat population. For anastrozole arm compared with tamoxifen arm, hazard ratio = 0.83 (95% confidence interval, 0.71–0.96; P = 0.015). Solid line: anastrozole (A); dashed line: tamoxifen (T); dotted line: combination therapy (C).
Figure 2. Probability of recurrence in the hormone receptor–positive subgroup. For anastrozole arm compared with tamoxifen arm, hazard ratio = 0.78 (95% confidence interval, 0.65–0.93; P = 0.007). Solid line: anastrozole (A); dashed line: tamoxifen (T); dotted line: combination therapy (C).
Contralateral breast cancer (CLBC)
The data on reduction in the incidence of primary CLBC continued to favor anastrozole over tamoxifen, both in the overall population (n = 25 vs. n = 40; odds ratio, 0.62; 95% CI, 0.38–1.02; P = 0.06) and in the hormone receptor–positive subgroup (n = 20 vs. n = 35; odds ratio, 0.56; 95% CI, 0.32–0.98; P = 0.04). When only invasive malignancies were considered (exploratory analysis), contralateral tumors were noted in 20 cases in the anastrozole arm and in 35 cases in the tamoxifen arm (odds ratio, 0.57; 95% CI, 0.33–0.98; P = 0.044).
Subgroups
In the initial efficacy analysis, an exploratory subgroup study revealed a potential difference in treatment outcome between patients who previously had received chemotherapy and those who had not. In the subgroup of patients who had received chemotherapy, the benefit produced by anastrozole in the overall population was not observed. In the updated analysis, this potential interaction was not evident, with the HR indicating a marginal benefit due to anastrozole (HR, 0.98; 95% CI, 0.76–1.28). The HR for patients with 4 or more positive lymph nodes also was close to 1 (HR, 0.95; 95% CI, 0.72–1.25). Currently, the number of events in these subgroups remains low, and it is possible that the differences between the overall analysis and the subgroup analyses may disappear over time. The interaction between treatment and hormone receptor status that was reported in the first analysis persists. In terms of all other predictive factors, no significant changes from the first analysis were observed in the update analysis, and the effects of anastrozole in all subgroups were found to be consistent with the overall results.
Safety
The updated safety analysis also confirmed the findings of the first analysis, demonstrating that the critical differences between the safety profiles of anastrozole and tamoxifen were maintained (Table 2). For all of the predefined adverse events, the results were consistent between the first analysis and the 7-month safety update. In all cases, there was no discernible difference between the tamoxifen arm and the combination therapy arm.
| Adverse event | No. of patients (%) | P valuea | ||
|---|---|---|---|---|
| Anastrozole (n = 3092) | Tamoxifen (n = 3093) | Combination therapy (n = 3097) | ||
| ||||
| Median duration of therapy (mos) | 37.3 | 36.9 | 36.5 | |
| Hot flashes | 1082 (35.0) | 1246 (40.3) | 1261 (40.7) | < 0.001 |
| Nausea and emesis | 346 (11.2) | 339 (11.0) | 379 (12.2) | 0.777 |
| Fatigue/tiredness (asthenia) | 512 (16.6) | 491 (15.9) | 468 (15.1) | 0.469 |
| Mood disturbances | 519 (16.8) | 508 (16.4) | 506 (16.3) | 0.707 |
| Musculoskeletal disorders | 936 (30.3) | 732 (23.7) | 765 (24.7) | < 0.001c |
| Vaginal bleeding | 147 (4.8) | 270 (8.7) | 265 (8.6) | < 0.001 |
| Vaginal discharge | 94 (3.0) | 378 (12.2) | 368 (11.9) | < 0.001 |
| Endometrial malignanciesb | 3 (0.1) | 15 (0.7) | 9 (0.4) | 0.007 |
| Fractures | 219 (7.1) | 137 (4.4) | 178 (5.7) | < 0.001c |
| Ischemic cardiovascular disease | 86 (2.8) | 67 (2.2) | 78 (2.5) | 0.121 |
| Ischemic cerebrovascular events | 34 (1.1) | 70 (2.3) | 60 (1.9) | < 0.001 |
| All venous thromboembolic events | 68 (2.2) | 116 (3.8) | 136 (4.4) | < 0.001 |
| Deep venous thromboembolic events | 35 (1.1) | 57 (1.8) | 70 (2.3) | 0.027 |
| Cataracts | 124 (4.0) | 139 (4.5) | 126 (4.1) | 0.378 |
All three treatment regimens were tolerated well by most patients. The proportion of patients who withdrew from treatment continued to be lower in the anastrozole group (24.1%) than in the tamoxifen group (28.3%) or the combination therapy group (29.4%); 5.6%, 8.1%, and 8.1% of patients withdrew from the anastrozole, tamoxifen, and combination treatment arms, respectively, because of treatment-related adverse events.
DISCUSSION
Analysis of the ATAC trial at a median follow-up time of 47 months demonstrated that after an additional follow-up period, anastrozole continued to exhibit superior efficacy (in terms of DFS, TTR, and reduction in the incidence of CLBC) compared with tamoxifen. The absolute reduction in risk as measured by DFS and TTR was found to be greater in the anastrozole arm than in the tamoxifen arm, with the time-to-event curves continuing to diverge for the two monotherapy arms. Based on the year-to-year gains for breast cancer events in favor of anastrozole, it is likely that these differences will continue to be observed after a longer follow-up period.
In the time between the first efficacy analysis and the updated analysis, new first events were distributed relatively evenly across the three arms of the trial, although, of course, the number of patients at risk for a first event now differs across treatment arms. The estimated relative risk for all efficacy endpoints has shifted marginally toward unity in the updated analysis compared with the first analysis. Nonetheless, these fluctuations are within the range associated with statistical variation, and the critical finding is that overall, the absolute benefits associated with anastrozole continue to increase with increasing follow-up duration. Currently, the number of events remains insufficient for formal analysis of the secondary endpoints, distant recurrence and survival. Additional follow-up is necessary for the accumulation of the 704 events required (among patients in the monotherapy arms) for formal analysis of these endpoints.
The overall favorableness of the safety profile of anastrozole compared with tamoxifen, as noted in the first analysis, was confirmed in the updated safety analysis; the updated findings provided reassurance regarding the safety of anastrozole after extended treatment. The decreased incidence of endometrial malignancies, vaginal bleeding and discharge, cerebrovascular events, thromboembolic events, and hot flashes in the anastrozole arm compared with the tamoxifen arm in the first analysis persisted after additional follow-up, as did the differences in arthralgia and fractures, which were less common in the tamoxifen arm than in the anastrozole arm.
After the first analysis, it was concluded that anastrozole represented an alternative option to tamoxifen for adjuvant treatment of postmenopausal women with hormone receptor–responsive tumors but that further follow-up and maturer data were required before a final risk-benefit assessment could be made. Based on the updated efficacy and safety data presented in the current report, the results continue to favor anastrozole. Furthermore, because the gap in absolute risk reduction (as measured by both DFS and TTR) continues to widen over time, it is likely that the differences in efficacy between anastrozole and tamoxifen will persist after further follow-up. These updates confirm the value of anastrozole in this setting and provide additional support for the use of anastrozole as adjuvant therapy for postmenopausal women with hormone-sensitive early-stage breast cancer.
Acknowledgements
The ATAC (Arimidex, Tamoxifen Alone or in Combination) Trialists' Group thanks the British Association of Surgical Oncology Breast Group, Cancer Research UK (London, United Kingdom); the trial investigators, monitors, nurses, data managers, and other support staff; and most importantly, the patients participating in the trial.
REFERENCES
- 1ATAC Trialist Group. 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: 2131–2139.
- 2. Aromatase inhibitors for the endocrine adjuvant treatment of breast cancer. Lancet. 2002; 359: 2126–2127.
- 3
- 4, on behalf of the ATAC Trialists' Group. Beneficial side-effect profile of anastrozole compared with tamoxifen confirmed by additional 7 months of exposure data: a safety update from the ‘Arimidex’, Tamoxifen, Alone or in Combination (ATAC) trial. Breast Cancer Res Treat. 2002; 76 (Suppl 1): S156.