- © 2000 by American Society of Clinical Oncology
Anastrozole Versus Tamoxifen as First-Line Therapy for Advanced Breast Cancer in 668 Postmenopausal Women: Results of the Tamoxifen or Arimidex Randomized Group Efficacy and Tolerability Study
- J. Bonneterre,
- B. Thürlimann,
- J.F. R. Robertson,
- M. Krzakowski,
- L. Mauriac,
- P. Koralewski,
- I. Vergote,
- A. Webster,
- M. Steinberg,
- M. von Euler and
- for the Arimidex Study Group
- From the Centre Oscar Lambret, Lille; Institut Bérgonié, Bordeaux, France; Medizinische Klinik C Kantonsspital, St Gallen, Switzerland; City Hospital, Nottingham; AstraZeneca, Alderley Park, Macclesfield, United Kingdom; Maria Sklodowska-Curie Memorial Centre of Oncology, Warsaw; Rydygier Memorial Hospital, Cracow, Poland; Department of Gynaecological Oncology, Leuven, Belgium; and AstraZeneca, Wilmington, DE.
- Address reprint requests to Jacques Bonneterre, MD, PhD, Centre Oscar Lambret, 3 Rue Frédéric Combemale, BP 307 Lille, France; email j-bonneterre{at}o-lambret.fr
Abstract
PURPOSE: To compare the efficacy and tolerability of anastrozole (Arimidex; AstraZeneca, Wilmington, DE, and Macclesfield, United Kingdom) with that of tamoxifen as first-line therapy for advanced breast cancer (ABC) in postmenopausal women.
PATIENTS AND METHODS: This randomized, double-blind, multicenter study evaluated the efficacy of anastrozole 1 mg once daily relative to tamoxifen 20 mg once daily in patients with tumors that were hormone receptor–positive or of unknown receptor status who were eligible for endocrine therapy. The primary end points were time to progression (TTP), objective response (OR), and tolerability.
RESULTS: A total of 668 patients (340 in the anastrozole arm and 328 in the tamoxifen arm) were randomized to treatment and followed-up for a median of 19 months. Median TTP was similar for both treatments (8.2 months in patients who received anastrozole and 8.3 months in patients who received tamoxifen). The tamoxifen:anastrozole hazards ratio was 0.99 (lower one-sided 95% confidence limit, 0.86), demonstrating that anastrozole was at least equivalent to tamoxifen. Anastrozole was also as effective as tamoxifen in terms of OR (32.9% of anastrozole and 32.6% of tamoxifen patients achieved a complete response [CR] or partial response [PR]). Clinical benefit (CR + PR + stabilization of ≥ 24 weeks) rates were 56.2% and 55.5% for patients receiving anastrozole and tamoxifen, respectively. Both treatments were well tolerated. However, incidences of thromboembolic events and vaginal bleeding were reported in fewer patients treated with anastrozole than with tamoxifen (4.8% v 7.3% [thromboembolic events] and 1.2% v 2.4% [vaginal bleeding], respectively).
CONCLUSION: Anastrozole satisfied the predefined criteria for equivalence to tamoxifen. Together with the lower observed incidence of thromboembolic events and vaginal bleeding, these findings indicate that anastrozole should be considered as first-line therapy for postmenopausal women with ABC.
BREAST CANCER IS the most common form of cancer affecting women in Europe, North America, Australia and New Zealand, and South America.1-3 Advanced breast cancer in postmenopausal women remains an incurable disease, and so treatment is aimed at palliation and improved quality of life, inhibition of disease progression, and improvement in survival time where possible.
Endocrine therapy to treat breast cancer is now a widely accepted treatment modality and is primarily directed at reducing the synthesis of estrogen or alternatively blocking estrogen receptors in those tumors that are hormone-sensitive. Since the late 1970s, tamoxifen has been the accepted gold-standard first-line treatment for advanced breast cancer in postmenopausal women, as studies showed that it was as effective as other therapies used at that time but was associated with fewer side effects.4,5 Tamoxifen acts by blocking the binding of estrogen to its receptor and has an overall response rate of 30% to 35% when used as first-line therapy for advanced breast cancer.6 Other benefits of tamoxifen use include favorable effects on bone mineral density7 and a reduction in serum cholesterol levels,8 with a possible concomitant benefit on cardiovascular health.9,10
Adverse effects that have been associated with tamoxifen can be classified as either owing to its antiestrogenic actions (eg, hot flashes, vaginal bleeding, discharge, or dryness) or as more general effects (eg, nausea, vomiting, tumor flare, and skin rash). In adjuvant studies involving patients who took either placebo or tamoxifen, thrombotic events were rare overall but were more common with tamoxifen than with placebo.9 A two- to three-fold increase in the risk of developing endometrial cancer has been documented in association with long-term tamoxifen use.11,12 Postmenopausal women with advanced breast cancer whose disease has progressed after treatment with tamoxifen are frequently prescribed a second endocrine agent, such as the new-generation nonsteroidal aromatase inhibitors anastrozole13 or letrozole.14
Anastrozole (Arimidex, AstraZeneca, Wilmington, DE, and Macclesfield, United Kingdom) is both potent and selective and can be given orally. It is well tolerated, while providing near maximal estrogen suppression.15,16 A combined analysis of two large randomized trials in postmenopausal women with advanced breast cancer who had experienced treatment failure with tamoxifen has shown that 1 mg of anastrozole significantly increases survival time compared with megestrol acetate at a mean follow-up of 31 months.13
It has been suggested that potent estrogen suppression via aromatase inhibition might result in a greater tumor response than with tamoxifen, particularly because tamoxifen is also known to be a weak or partial estrogen agonist in addition to being an estrogen antagonist.17 This article describes the results of a double-blind, multicenter randomized study involving 668 patients and comparing the efficacy and tolerability of anastrozole and tamoxifen as first-line therapy for advanced breast cancer in postmenopausal women (the Tamoxifen or Arimidex Randomized Group Efficacy and Tolerability [TARGET] study). TARGET was one of two studies designed to stand alone but also to allow for combined analysis, the second being the North American study.18
PATIENTS AND METHODS
Study Design
This was a randomized, double-blind, multicenter study comparing the efficacy and tolerability of anastrozole 1 mg once daily with tamoxifen 20 mg once daily as first-line therapy for advanced breast cancer in postmenopausal women. The study was conducted at 83 sites in Europe, Australia, New Zealand, South America, and South Africa.
The primary objectives of the trial were to compare the two drugs with respect to time to disease progression (TTP), objective response (OR) rate, and tolerability. The secondary objectives were to compare treatment groups with respect to time to treatment failure (TTF), duration of response, duration of clinical benefit, and survival. All patients were followed-up until objective progression and death, irrespective of treatment received.
Patient Population
All patients were required to be postmenopausal and to have a diagnosis of locally advanced or metastatic breast cancer. Postmenopausal women were defined as (1) those ≥ 50 years of age who had not menstruated during the preceding 12 months or who had castrate follicle-stimulating hormone levels (> 40 IU/L), (2) those younger than 50 years of age who had castrate follicle-stimulating hormone levels (> 40 IU/L), or (3) those who had undergone a bilateral oophorectomy. Patients had to be suitable for endocrine therapy as first-line treatment; prior adjuvant chemotherapy or endocrine therapy for early breast cancer was permissible in enrolled patients, but they were not permitted to have received tamoxifen within 12 months before entry onto the trial. In addition, only patients with hormone receptor–positive (estrogen receptor–positive [ER+] and/or progesterone receptor–positive [PgR+]) tumors or tumors of unknown hormone receptor status could be included in the study.
Exclusion criteria included previous systemic therapy for advanced breast cancer, extensive visceral disease (significant hepatic involvement, brain metastases, and pulmonary lymphangitic spread of tumor; serum liver enzymes could be no greater than five times the upper limit of the reference range), any concurrent medical illness or laboratory abnormalities that would compromise safety or prevent interpretation of results, or an estimated survival of less than 3 months from the start of trial therapy (based on clinical judgment). At the beginning of the study, patients receiving bisphosphonates were excluded; however, because of the increasing numbers of women with advanced breast cancer being treated with bisphosphonates, the protocol was subsequently amended (at which stage 540 patients had been enrolled) and these patients could then be included. In these patients, bone metastases were considered nonassessable. All patients gave their informed consent, and appropriate ethics committee approval was obtained at each site before initiation of the study.
Treatment Program
Patients were randomized to receive a daily dose of two tablets (one active and one placebo tablet). The oral treatment regimen was the combination of either anastrozole 1 mg once daily and tamoxifen placebo, or tamoxifen 20 mg once daily and anastrozole placebo. Patients were instructed to take the two tablets together at approximately the same time each day. Trial treatment continued until objective disease progression was observed, at which time it was stopped. Further treatment was left to the discretion of the investigator, and follow-up was performed until death.
Baseline screening assessments were completed within 4 weeks before randomization. These assessments included demographic information, complete history, and clinical examination to document the sites of disease. Radiology studies included chest x-ray, liver scan by ultrasound, computed tomography scan or magnetic resonance imaging, bone scan and bone survey, or x-rays of areas positive on the bone scan. History of symptoms related to disease were also documented. Blood samples were collected for hematology and blood chemistry. On day 1 (the date of randomization), eligible patients underwent a complete physical examination. Each patient’s disease was then assessed clinically every 4 weeks for the first 24 weeks of treatment and then every 12 weeks until objective evidence of disease progression was obtained. Where possible, all assessments were repeated at the end of trial therapy.
Patients could be withdrawn from active treatment for the following reasons: objective disease progression, the occurrence of a serious adverse event, noncompliance with protocol procedures, unwillingness or inability to continue the trial, or withdrawal at the investigator’s discretion. All patients who were withdrawn for reasons other than objective disease progression were monitored until progression was observed. After objective disease progression, patients were followed up at 6-month intervals for survival information.
Efficacy Assessments
The primary efficacy measures were TTP and OR rate. The secondary efficacy measures were TTF, duration of response, duration of clinical benefit, and survival. Measurable disease was defined as the presence of bidimensionally or unidimensionally measurable lesions as determined by physical examination, ultrasound, or radiographic scan. Osteolytic bone lesions were considered measurable. Single metastatic lesions smaller than 0.5 cm, malignant pleural effusions or ascites, positive bone scans, and purely osteoblastic or intratrabecular bone lesions were not classified as measurable disease. Lesions not classified as measurable constituted assessable (but nonmeasurable) disease.
All randomized patients were assessed based on International Union Against Cancer criteria for tumor response 4 weeks after the initial administration of trial medication and at all subsequent visits, up to and including the visit at which disease progression was observed. ORs were classified as complete response (CR), partial response (PR), stable disease (SD), or progressive disease for both measurable and nonmeasurable disease.19 For a best response of CR, the patient had to have two consecutive assessments of CR at least 4 weeks apart, and for a best response of PR, the patient had to have two consecutive assessments of PR or better at least 4 weeks apart. The assessment criteria used were stricter than the International Union Against Cancer criteria in that patients having only nonmeasurable disease could not qualify for a PR, and a best response of SD was only assigned when responses of SD or better were observed for at least 24 weeks. If such responses had been observed for less than 24 weeks because a patient did not have measurements for 24 weeks at the time of data cutoff, then a best response of SD for less than 24 weeks was recorded. This criteria was based on data showing that a response of SD for at least 24 weeks is equivalent to CR and PR in terms of overall survival.20-23 Responders were those patients with a best objective response of CR or PR. Patients with clinical benefit were defined as those responding (CR + PR) plus those with SD for at least 24 weeks.
TTP, TTF, duration of response, and duration of clinical benefit were calculated from the date of randomization. TTP represented the time to objective disease progression or death, whichever occurred first. TTF was the time to the earliest occurrence of progression, death, or withdrawal from randomized treatment. Duration of response, which was recorded for those with either a CR or PR, was the time from randomization to the first observation of objective progression or death. Duration of clinical benefit in patients who achieved CR or PR or who had stable disease for 24 weeks or more was also defined as the time from randomization to the first observation of objective progression or death. Survival was measured from the date of randomization to the date of death.
Tolerability Assessments
Adverse events were recorded on a treatment-received basis. An adverse event was defined as any detrimental change in a patient’s condition after randomization and during any follow-up period, unless considered by the investigator to be related to disease progression. All adverse events were recorded, irrespective of whether the event was considered to be related to the trial therapy. Adverse events that might be expected to occur on the basis of the pharmacology of anastrozole and tamoxifen were also specifically identified (predefined events). The predefined events were depression, tumor flare, thromboembolic disease, gastrointestinal disturbance, hot flashes, vaginal dryness, lethargy, vaginal bleeding, and weight gain.
In addition to monitoring for adverse events, routine laboratory tests were performed at baseline, at selected times during therapy, and at withdrawal. The results of clinical laboratory tests were reviewed for clinically relevant changes. Physical examinations were performed and body weight, blood pressure, and pulse rate were recorded at baseline, at selected times during therapy, and at withdrawal.
Statistical Analysis
The trial was designed to compare anastrozole with tamoxifen, using TTP and OR rate as the two primary efficacy end points, and was designed to demonstrate equivalence in each of these end points. For time to progression, the comparison between anastrozole and tamoxifen was expressed in terms of the hazards ratio (tamoxifen:anastrozole), which estimates the chance of progression on tamoxifen in a given time period in relation to the chance of progression on anastrozole in the same time period. A hazards ratio of more than 1 indicates a superiority for anastrozole.
The estimated hazards ratio for TTP from this study provides the best estimate of the true value of the hazards ratio in the overall population of patients. The lower one-sided 95% confidence limit indicates the range of values for the true hazards ratio, which may be considered to be consistent with the result from this study. The prespecified criterion for equivalence would be met if the lower one-sided 95% confidence limit was ≥ 0.80; ie, equivalence would be concluded if a 20% or greater advantage for tamoxifen could be ruled out with 95% confidence.
For response rate (CR + PR), the comparison between treatments was expressed in terms of the difference in response rates (anastrozole − tamoxifen). A difference greater than zero would indicate a higher response rate for anastrozole, whereas a difference less than zero would indicate a higher response rate for tamoxifen.
The estimated difference in response rates from this study provides the best estimate of the true differences in response rates in the overall population of patients. The lower one-sided 95% confidence limit indicates the range of values for the true difference in response rates, which may be considered to be consistent with the result from this study. The prespecified criterion for equivalence would be met if the lower one-sided 95% confidence limit was ≥ −10%; ie, equivalence would be concluded if a difference in response rates of greater than 10% in favor of tamoxifen could be ruled out with 95% confidence.
A population of 660 patients (330 in each treatment group) was estimated to be sufficient to demonstrate treatment equivalence with 80% power based on a one-sided 5% significance level. The statistical analyses were performed on an intention-to-treat basis. To test the robustness of the equivalence assessment, a per-protocol analysis was also performed.
The Cox proportional hazards regression model was used to assess treatment equivalence for TTP and TTF. The OR rate was compared between the treatment groups using logistic regression. These efficacy analyses were adjusted for the covariates of age (≤ 65 years, > 65 years), previous endocrine therapy (yes or no), extent of disease at randomization, and hormonal receptor status at diagnosis. In addition to the prospectively identified statistical analyses (designed to demonstrate equivalence), a retrospective analysis assessed whether anastrozole showed any superiority over tamoxifen for the TTP and OR end points.
Duration of response and duration of clinical benefit were summarized by randomized trial treatment using the Kaplan-Meier method. For survival, no formal statistical comparisons were performed because of the immaturity of the data.
On completion of the trial and initial prospectively identified data analyses, it was noted that more than 50% of the patients recruited to the trial had tumors of unknown receptor status. To determine whether this observation could have impacted on the TTP results, TTP was summarized retrospectively for those patients with tumors confirmed as ER+ and/or PgR+ and for those whose tumor receptor status was unknown.
RESULTS
Patient Characteristics
At the time of analysis, the median duration of follow-up was 19 months, and 496 patients (74%) had experienced disease progression. A total of 668 patients from 83 centers in Europe, Australia, New Zealand, South America, and South Africa, recruited between August 21, 1995, and July 1, 1998, were randomized to double-blind treatment (anastrozole, n = 340; tamoxifen, n = 328). The groups were well balanced with respect to demographic and pretreatment characteristics (Table 1). One hundred fifty-four (45.3%) of 340 patients in the anastrozole group and 144 (43.9%) of 328 patients in the tamoxifen group had an ER+ and/or PgR+ tumor. Of these, 80 (23.5%) of 340 patients in the anastrozole group and 85 (25.9%) of 328 in the tamoxifen group had tumors that were both ER+ and PgR+. A total of 185 (54.4%) of 340 patients in the anastrozole group and 183 (55.8%) of 328 patients in the tamoxifen group included in the study had a tumor of unknown ER/PgR status.
Demographic and Pretreatment Characteristics
Efficacy
OR rate.
Anastrozole was at least as effective as tamoxifen in terms of best OR rates (Table 2), with 32.9% of patients in the anastrozole group and 32.6% in the tamoxifen group achieving a CR or a PR (estimated difference in response rates for anastrozole − tamoxifen, −1.0%; lower one-sided 95% confidence limit, −6.7%; two-sided P = .787). A total of 56.2% of patients in the anastrozole group and 55.5% of patients in the tamoxifen group gained clinical benefit from therapy (Table 2).
Best Objective Tumor Responses
The covariates of age, receptor status, and extent of disease were significant or near-significant variables in the model for OR rates (P = .048, P = .051, and P < .01, respectively) with patients older than 65 years, patients with known positive receptor status, and patients with only soft tissue and/or lung disease achieving better OR.
The median duration of response for the 112 patients in the anastrozole group who had a CR or PR was 16.4 months from randomization (range, 3.6 to 39.2 months). For the 107 patients in the tamoxifen group who had a CR or PR, the median duration of response was 17.0 months (range, 2.7 to 36.9 months).
The median duration of clinical benefit for the 191 patients in the anastrozole group who had a CR or PR or had SD for at least 24 weeks was 15.2 months from randomization (range, 3.6 to 39.2 months). For the 182 patients in the tamoxifen group who had a CR or PR or had SD for at least 24 weeks, the median duration of clinical benefit was 14.7 months (range, 2.7 to 41.4 months).
TTP.
Anastrozole was at least as effective as tamoxifen in terms of TTP, with the median TTP being 8.2 months in the anastrozole group and 8.3 months in the tamoxifen group (tamoxifen:anastrozole hazards ratio, 0.99; lower one-sided 95% confidence limit, 0.86; two-sided P = .941). A Kaplan-Meier plot of TTP is presented in Fig 1. The covariates of age, previous endocrine therapy, and extent of disease were significant or near-significant variables in the model for TTP (P = .017, P = .059, and P = .042, respectively), with patients older than 65 years, patients with no previous hormonal therapy, and patients with only soft tissue and/or lung disease achieving greater TTP.
Fig 1. Kaplan-Meier probability of TTP in patients receiving anastrozole 1 mg or tamoxifen 20 mg once daily.
The Kaplan-Meier probability of TTP for the subgroup of patients with receptor-positive tumors is shown in Fig 2; the Kaplan-Meier probability of TTP for the subgroup of patients with ER/PgR unknown disease is shown in Fig 3. No statistical analysis of subgroups was performed for this trial, because the larger combined data set from this trial and the North American trial provides the most appropriate setting for statistical exploration of subgroups; however, in the receptor-positive subgroup there was a separation of the curves suggesting a benefit in favor of anastrozole, as shown in Fig 2.
Fig 2. Kaplan-Meier probability of TTP in patients receiving anastrozole 1 mg or tamoxifen 20 mg once daily (subgroup of patients with ER+ and/or PgR+ tumors).
Fig 3. Kaplan-Meier probability of TTP in patients receiving anastrozole 1 mg or tamoxifen 20 mg once daily (subgroup of patients with receptor-unknown tumors).
TTF.
For the majority of patients who experienced treatment failure in each treatment group, the reason for treatment failure was disease progression (236 [69.4%] of 340 patients in the anastrozole group and 227 [69.2%] of 328 patients in the tamoxifen group). Overall, treatment failure had occurred in 267 (78.5%) of 340 patients in the anastrozole group and 266 (81.1%) of 328 patients in the tamoxifen group. In the anastrozole group, treatment failure occurred in 13 patients (3.8%) because of adverse events and in 18 patients (5.3%) for other reasons. In the tamoxifen group, treatment failure occurred in 15 patients (4.6%) because of adverse events and in 24 patients (7.3%) for other reasons. Figure 4 shows a Kaplan-Meier plot of TTF. The estimated hazards ratio for TTF for tamoxifen versus anastrozole was 1.03 (lower 95% confidence limit, 0.89).
Fig 4. Kaplan-Meier probability of TTF in patients receiving anastrozole 1 mg or tamoxifen 20 mg once daily.
Survival.
Overall, 165 deaths (24.7% of patients) were reported up until data cutoff. Evaluation of potential differences in survival was not carried out at this time, because the data were considered to be immature. An analysis based on these data may therefore be potentially misleading at this stage. Per-protocol analyses of the efficacy data provided similar results to those of the intention-to-treat analyses.
Tolerability.
Both treatments were generally well tolerated. The five most frequently reported adverse events in the anastrozole and tamoxifen groups are listed in Table 3. Hot flashes and nausea were the two most frequently reported adverse events in both groups. Overall, 15 patients (4.5%) in the anastrozole group and 19 (5.8%) in the tamoxifen group had adverse events that led to withdrawal. Of these patients, the adverse events were considered to be drug-related in seven patients in the anastrozole (2.1%) and eight patients in the tamoxifen group (2.4%).
Most Frequently Reported Adverse Events in the Anastrozole Group and the Tamoxifen Group
The incidences of predefined adverse events are listed in Table 4. The incidence of events in the category of thromboembolic disease occurred at a higher observed frequency in the tamoxifen group (7.3%) compared with the anastrozole group (4.8%). With respect to the venous thromboembolic complications experienced by some patients, the study showed that they were superficial in both of the patients in the anastrozole group. In the tamoxifen group, six of these events were superficial and four were deep (one patient experienced venous retinal thrombosis, one patient experienced both pulmonary embolism and a cerebrovascular accident, another patient experienced a deep venous thrombosis in the left leg, and the fourth patient experienced a pulmonary embolism). Additionally, vaginal bleeding was observed at a two-fold higher incidence in the tamoxifen group (1.2% in the anastrozole group v 2.4% in the tamoxifen group).
Incidence of Predefined Adverse Events, Irrespective of Causality, Reported in Each Treatment Group
There were 13 deaths during the treatment period of this study (seven in the anastrozole group and six in the tamoxifen group) that were not considered to be related to breast cancer. None were related to study treatment. Deaths among patients receiving anastrozole were caused by shock in two of seven patients (one died of multiorgan failure that was linked to deterioration of general status; the other died of hypovolemic shock related to sepsis due to an infected diabetic ulcer), cardiovascular events in four patients, and pneumonia in one patient. Deaths among patients receiving tamoxifen were caused by cardiovascular events in four patients, pneumonia in one patient, and hypoglycemia in one patient.
DISCUSSION
The data from this study show that anastrozole is at least as effective as tamoxifen as first-line therapy of advanced breast cancer in postmenopausal women. A total of 56.2% of patients in the anastrozole group and 55.5% of patients in the tamoxifen group gained clinical benefit from treatment. The results in terms of the efficacy of tamoxifen were similar to those of previously reported studies of tamoxifen in this patient population, such as that reported by Rose and Mouridsen.6 This study is one of two to first report on the comparative efficacy of anastrozole and tamoxifen as first-line therapy of advanced breast cancer in postmenopausal women, the other study also appears in this issue of the Journal of Clinical Oncology (pp 3758-3767).18
In this trial, only approximately 45% of patients had receptor-positive tumors. Therefore, to establish whether the high proportion of patients with unknown receptor status in the TARGET trial masked any benefit of anastrozole over tamoxifen, subgroups were retrospectively examined, comparing TTP in patients whose tumors were ER+ and/or PgR+ with those whose tumors were of unknown or negative receptor status (Figs 2 and 3⇑). No statistical analysis of these subgroups was performed for this trial, because the larger combined data set from this trial and the North American trial provides the most appropriate setting for statistical exploration of subgroups; however, in the subgroup of patients whose tumors were receptor-positive, there was a separation of the curves in favor of anastrozole. This is suggestive of a differential effect of anastrozole compared with tamoxifen in patients with tumors known to be receptor-positive, a finding that is supported by the data from the North American trial.18
With respect to formal tolerability comparisons, no statistical analyses were carried out on the data generated from the individual trials. In both trials, predefined adverse events were prospectively identified on the basis of the known pharmacology and expected side effects of both treatments. In the present study, both tamoxifen and anastrozole were well tolerated, and there was a similar overall incidence of adverse events (including those leading to withdrawal) in both groups. There were numerical differences in the incidence of two important side effects; thromboembolic events and vaginal bleeding both increased in the tamoxifen arm. These results support those observations seen in another study where cardiovascular events (including three fatalities) were only observed in patients taking tamoxifen.24 If clinically relevant differences regarding thromboembolic (or cardiovascular) events in the advanced disease setting also occur in the adjuvant setting, differences in thromboembolic and vaginal bleeding events will become apparent in the Arimidex or Tamoxifen, Alone or in Combination (ATAC) trial, which compares the adjuvant use of tamoxifen with anastrozole and the combination of these two endocrine agents in postmenopausal women with early breast cancer.25
There was no difference between the two treatment groups with respect to tumor flare (2.4%). Although tumor flare associated with tamoxifen treatment has been ascribed to estrogen agonist activity,26,27 clearly the same explanation cannot account for tumor flare associated with anastrozole. A total of three of the 16 patients with tumor flare had an OR; allowing for the limitation of small numbers, these patients did not seem more likely to respond to treatment.
Previous studies of the new-generation aromatase inhibitors fadrozole, vorozole, letrozole, and anastrozole have indicated that nausea, vomiting, and diarrhea are probably a class effect of these agents.13,14,25,28 Reassuringly, in this study there were no notable differences in the incidence of these side effects between anastrozole and tamoxifen.
Anastrozole is currently used as second-line therapy after tamoxifen treatment in patients whose cancers have recurred or progressed despite treatment.13 The results of the present study indicate that anastrozole is at least as effective as the current treatment of choice, tamoxifen, when used as first-line treatment for advanced breast cancer. In patients with tumors known to be receptor-positive, there was a suggestion of an efficacy difference in favor of anastrozole. There will be an opportunity to further study the relevance of receptor status in the ongoing ATAC trial. Combined with good tolerability, these data confirm that anastrozole should now be considered as first-line treatment for postmenopausal women with advanced breast cancer.
APPENDIX
The Arimidex Study Group includes the following investigators: Prof Dr F. Gago, Francisco CEAM (Centro de Estudio de Afecciones Mamarias), Mendoza; Dr C. Maañon, Claudio Hospital San Bernardo, Salta; Prof Dr A. Nuñez de Pierro, Anibal Hospital J. Fernandez, Buenos Aires; Prof Dr R. Tozzini, Roberto Hospital del Centenario-Ginecología, Rosario, Santa Fe, Argentina; Dr J. Beith, Royal Prince Alfred Hospital, Camperdown New South Wales; Dr M. Byrne, Sir Charles Gairdner Hospital, Nedlands; Dr M. Green, Royal Melbourne Hospital, Parkville, Australia; Dr J. Ph. Janssens, Salvatorziekenhuis, Hasselt; Prof Dr S. Van Belle, Dienst Oncologie, Gent; Dr G. De Wasch, Henri Serruys Ziekenhuis, Oostende, Belgium; Dr A.L. Frasson, Catholique University (PUC), Porto Alegre; Dr A. Malzyner, Av. Nove de Julo 4644; Dr S. Moraes, Womans Reference Centre; and Dr C. Tosello de Oliveira, Brazilian Institute for Combat to Cancer, Sao Paulo, Brazil; Dr O. Bednarik, Istav, Brno; Dr J. Fischer, Nemocnice versus Ceskych Budejovicich, Ceske Budejovice; Dr B. Konopasek, Onkologicka klinika, University Karlovy, Onkologicke oddeleni, Prague; Dr J. Machacek, Onkologicka klinika, Fakultni nemocnice Olomouc; Dr P. Vodvarka, Radioterapeuticka klinika, Ostrava-Poruba, Czech Republic; Prof F. Given, University College Hospital, Galway, Eire; Dr P. Chollet, Centre Jean Perrin, Clermont Ferrard; Dr D. Langlois, Centre St Michel, La Rochelle; Dr M. Rios, Centre Alexis Vautrin, Vandoeuvre les Nancy; Prof M. Tubiana-Hulin, Centre Rene Higuenin, Saint Cloud; Dr Veyret, Centre Henri Becquerel, Rouen; Dr E. Vuillemin, Centre St Yves, Vannes, France; Prof Dr W. Eiermann, Frauenklinik vom Roten Kreuz, Munich; Prof J. Hilfrich, Henriettenstiftung, Hannover; Dr E. Kettner, Staedtishes klinikum, Madgeburg; Dr Med K. Kuhndel, Univ Frauenklinik Leipzig, Leipzig; Dr E. Merkle, Olgohospital, Frauenklinik Berg, Stuttgart, Germany; Dr J. Erfan, Josa A County Hospital, Nyiregyhaza; Dr T. Nagykalnai, Uzsoki u Hospital, and Dr I. Szakolczai, National Institute of Oncology, Budapest; Dr T. Pinter, Petz Aladar County Hospital, Gyor; Dr J. Szanto, BAZ County Hospital, Miskolc, Hungary; Prof F. Boccardo, Instituto Nazionale per la Ricerca sul Cancro, Genoa; Dr G. Brignone, Ospedale M Ascoli, Palermo; Prof C. Bumma, Ospedale S Giovanni, Turin; Dr Alberto Scanni, Ospedale Fatebenefratelli, Milan, Italy; Dr J. De La Garza, Instituto Nacional de Cancerologia, Mexico City; Dr J. L Gonzales-Vela, Centro Medico Darox, Monterrey; Dr G. Morgan-Villela, Centro Medico Nacional de Occidente IMMS, Guadalajara, Mexico; Dr S. Allan, Palmerston North Hospital, Palmerston North, New Zealand; Dr B. Karczmarek-Borowska, Regional Oncology Centre, Lublin; Dr J. Zaluski, Maria Sklodowska-Curie Memorial Institute, Poznan, Poland; Dr B. da Costa, Hospital de Santa Maria, Dr J. Gouveia, Hospital de Santo Antonio dos Capuchos, Dr A. Monteiro, Hospital de Santo Antonio dos Capuchos, and Dr R. da Luz, Hospital de Santa Maria, Klinika Radioterapie a Onkologie Fakultna Nemocnica L, Lisbon, Portugal; Dr L. Bohunicky, National Cancer Institute, Bratislava; Dr M. Wagnerova, Radiotherapie a Onkologie Fakultna Nemocnica L Pasteura, Kosice, Slovakia; Dr G. Landers, Medical Mews 7, Overport, South Africa; Prof P.E. Jonsson, Kirurgiska kliniken, Lasarettet, Helsingborg; Prof G. Westman, Onkologiska kliniken Regionsjukhuset, Orebro, Sweden; PD Dr M. Castiglione, Onkologische Abteilung Inselspital, Bern; Dr D. Hess, Medizinische Klinik C Kantonsspital, St Gallen; Dr L. Perey, Centre Pluridisciplinaire d’Oncologie, Lausanne; PD Dr Ch Rochlitz, Onkologie Kantonsspital, Basel; Dr C. Sessa, Ospedale San Giovanni, Bellinzona, Switzerland; Dr N. Güler, Hacettepe University, Ankara; Prof Dr N. Molinas Mandel, Istanbul University, Cerrahpasa Medical School, Prof Dr A. Ober, Istanbul University, Prof Dr S. Okkan, Istanbul University, and Prof Dr S. Serdengecti, Istanbul University, Cerrahpasa Medical School, Istanbul; Prof Dr D. Firat, Hacettepe University, Ankara, Turkey; Dr W.P. Abram, The Belvior Park Hospital, Belfast; L.C. Barr, Withington Hospital, and Dr A. Stewart, Christie Hospital & Holt Radium Institute, Manchester; Prof R.W. Blamey, City Hospital, Nottingham; Dr D.J. Dodwell, Cookridge Hospital, Leeds; S.R. Ebbs, Mayday University Hospital, Surrey; C.D.M. Griffith, Royal Victoria Hospital, and Dr T.W.J. Lennard, Department of Surgery, University of Newcastle, Newcastle-On-Tyne; R. Hall, York District Hospital, York; C. Humphrey, Birch Hill Hospital, Littleborough; Dr C. Keen, Maidstone Hospital, Maidstone; M. Kissin, Royal Surrey County Hospital, Guildford; Prof S.J. Leinster, Royal Liverpool University Hospital, Liverpool; Dr R.C.F. Leonard, Western General Hospital, Edinburgh; Prof M.B. Mclllmurray, Royal Lancaster infirmary, Lancaster; and Dr J. Waxman, Hammersmith Hospital, and A. Wilson, Whittington Hospital, London, United Kingdom.
Acknowledgments
Supported by a grant from AstraZeneca, Macclesfield, United Kingdom.
- Received February 17, 2000.
- Accepted September 7, 2000.
