© 2001 by Oxford University Press
Journal of the National Cancer Institute Monographs, No. 30, 56-61,
2001
© 2001 Oxford University Press
Duration of Adjuvant Tamoxifen Therapy
Affiliations of authors: J. Bryant, National Surgical Adjuvant Breast and Bowel Project (NSABP) Biostatistical Center, Pittsburgh, PA, and Department of Biostatistics, University of Pittsburgh; B. Fisher, NSABP and Department of Surgery, University of Pittsburgh; J. Dignam, NSABP Biostatistical Center and Department of Health Studies, University of Chicago, IL.
Correspondence to: John Bryant, Ph.D., NSABP Biostatistical Center, 1 Sterling Plaza, 230 N. Craig St., Suite 350, Pittsburgh, PA 15213 (e-mail: bryant{at}nsabp.pitt.edu).
 |
ABSTRACT |
|---|
 Top Abstract IntroductionDurations of up to...Treatment for 5 Years...DiscussionConclusions and RecommendationsReferences |
|---|
The benefit of using adjuvant tamoxifen to treat breast cancer has been firmly established for patients with estrogen receptor (ER)-positive tumors, regardless of age, lymph node status, or menopausal status. Uncertainty remains, however, regarding the optimal duration of tamoxifen therapy. We reviewed the findings of randomized clinical trials that directly compared alternative treatment durations. Trials comparing short-term adjuvant treatment with tamoxifen (i.e., 1–3 years) with treatments having durations of about 5 years consistently have demonstrated additional benefits stemming from the longer therapy. Trials testing 5 years of treatment with longer durations have, in the aggregate, suggested no additional benefit for the patient. Nevertheless, the number of recurrences reported to date in these trials is not large, and the results of the individual trials are heterogeneous. Furthermore, as a result of tamoxifen's "carryover" effect, duration trials require considerable follow-up before definitive results can be established. Until more definitive data become available, adjuvant treatment with tamoxifen should be limited to 5 years outside the clinical trials setting. Continued accrual of ER-positive patients to ongoing tamoxifen duration trials, including the Adjuvant Tamoxifen Treatment Offer More (aTTom) and Adjuvant Tamoxifen Longer Against Shorter (ATLAS) trials, is appropriate. Alternatively, patients who remain disease free after 5 years of tamoxifen therapy should be encouraged to participate in trials testing crossover to other hormonal interventions, including selective ER modulators or aromatase inhibitors.
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionDurations of up to...Treatment for 5 Years...DiscussionConclusions and RecommendationsReferences |
|---|
Clinical trials of adjuvant tamoxifen therapy have convincingly demonstrated benefit for patients with estrogen receptor (ER)-positive breast cancer, regardless of age, lymph node status, or menopausal status (1–3). Despite the large number of trials that have tested tamoxifen, however, uncertainty remains regarding the optimal duration of therapy. We will briefly review clinical trials comparing short-term administration of tamoxifen therapy (i.e., 1–3 years) with treatment durations of about 5 years, from which considerable evidence has accumulated in the last decade. We will be primarily concerned, however, with summarizing those trials that compare 5 years of treatment with longer durations. The data currently available from these trials are not extensive, and their interpretation is controversial.
|
DURATIONS OF UP TO 5 YEARS |
|---|
|
TopAbstractIntroductionDurations of up to...Treatment for 5 Years...DiscussionConclusions and RecommendationsReferences |
|---|
Four trials have been reported in the past decade that compare 5 years of tamoxifen treatment with shorter durations of therapy. Aspects of the trials' designs and patient populations are summarized in Table 1
. Two Eastern Cooperative Oncology Group (ECOG) trials, E4181 (4) and E5181 (5), compared 1 year of tamoxifen therapy with 5 years of the same therapy in postmenopausal and premenopausal lymph node-positive patients, respectively. Patients in both trials received adjuvant chemotherapy in addition to tamoxifen. The Swedish Breast Cancer Cooperative Group (Swedish BCCG) (6) trial was a larger study that compared 2 years of tamoxifen therapy with 5 years of the same therapy in 3545 postmenopausal patients, only 2.5% of whom received chemotherapy. The Cancer Research Campaign (CRC) (7) trial also compared 2 years of tamoxifen therapy with 5 years of tamoxifen therapy in 2937 patients aged 50 years or older. In addition to tamoxifen, patients in this trial were treated with adjuvant chemotherapy or not, according to local protocols. The results of a fifth study, TAM-01 (8), were first reported in 2000. This trial is designed to compare 2–3 years of tamoxifen therapy with 12–13 years of the same therapy, but the median duration of tamoxifen therapy at the time of the report was 30 months on the control arm and 70 months on the extended treatment arm, so the currently available data may be interpreted as comparing 2–3 years of treatment with approximately 6 years of treatment. About 30% of the 3793 patients included in this study also received adjuvant chemotherapy. Both ER-negative and ER-positive patients were included in all five trials, in the proportions indicated in Table 1.
|
Each of these five studies has reported statistically significant reductions in event rates in favor of the longer duration of treatment. The three studies that compared a treatment duration of about 2 years with treatment lasting at least 5 years [Swedish BCCG (6), CRC (7), and TAM-01 (8)] all estimated a proportional reduction in event rate of about 20%. Only one of the five studies [Swedish BCCG (6)] reported a statistically significant survival advantage for the longer duration (18% proportional reduction in mortality rate; 95% confidence interval [CI] = 1% to 31%; P =. 03). One additional trial (CRC) (7) also showed a modest but nonsignificant survival advantage (11% reduction in mortality rate; 95% CI = –15% to 31%). In light of the strong evidence for a reduced recurrence rate, however, it is reasonable to attribute this to the relatively short follow-up reported and to the well-known carryover effect of treatment with tamoxifen (1,3). A preliminary meta-analysis of direct comparisons of short-term tamoxifen therapy with those lasting about 5 years, which was presented for discussion at the September 21–23, 2000, Oxford meeting of the Early Breast Cancer Trialists' Collaborative Group (EBCTCG), did in fact indicate a modest but statistically significant mortality advantage for the longer duration. Published indirect comparisons of trials of about 1, 2, or 5 years' duration based on the 1998 EBCTCG overview led to a similar conclusion (1), as trends for both reduced recurrence rate and mortality over this range were statistically significant.
|
TREATMENT FOR 5 YEARS VERSUS LONGER DURATION |
|---|
|
TopAbstractIntroductionDurations of up to...Treatment for 5 Years...DiscussionConclusions and RecommendationsReferences |
|---|
Findings from three trials that compared 5 years of tamoxifen therapy with treatments of longer duration were first reported in 1996. Aspects of the trials' designs and patient populations are summarized in Table 2
. These trials included National Surgical Adjuvant Breast and Bowel Project (NSABP) B-14 (3), a double-blind comparison of 5 additional years of tamoxifen therapy versus placebo in 1172 lymph node-negative women who were disease free after completing 5 years of initial treatment with tamoxifen; the Scottish trial (9), in which 342 predominantly lymph node-negative women were assigned to receive indefinite tamoxifen therapy or to observation following the completion of 5 years of tamoxifen therapy disease-free; and ECOG E4181/E5181 (10), in which 87 postmenopausal lymph node-positive women from E4181 and 106 premenopausal lymph node-positive women from E5181 who had been treated with chemotherapy and 5 years of tamoxifen therapy and who were disease free at 5 years were randomly assigned to either indefinitely continued tamoxifen therapy or to observation. All patients in B-14 (3) were ER positive (
10 fmol/mg protein), compared with 73% in the ECOG trial (10). In the Scottish trial (9), 39% of patients had tumors with ER content of 20 fmol/mg or more, 22% had tumors with ER content of 19 fmol/mg or less, and 39% had tumors that were not assayed. Updated data from the NSABP study (3) and the Scottish trial (9) have been published recently (11,12).
|
National Surgical Adjuvant Breast and Bowel Project B-14
After a median follow-up of 6.75 years from random assignment to either placebo or continued tamoxifen, the B-14 data demonstrated no advantage for continued therapy, and in fact the trend moved in the opposite direction. Seven years following randomization, the overall survival rate was 94% for placebo-treated patients and 91% for tamoxifen-treated patients (39 versus 57 deaths, respectively; hazard ratio [HR] = 1.5 [95% CI = 1.0 to 2.2]; P = .07). Disease-free survival (DFS) rate was 82% for placebo patients and 78% for those who received more than 5 years of tamoxifen therapy (106 versus 137 events; HR = 1.3 [95% CI = 1.0 to 1.7]; P = .03).
Events in the primary NSABP DFS analysis included recurrence, contralateral breast cancer (CBC), other second primary cancers, and death (11). The distribution of sites of first events is given in Table 3. Only about one half (118 of 243) of all first events were breast cancer related (i.e., recurrences or CBC). For comparison with other datasets, we performed an analysis of breast cancer first events only; the pattern is similar to that seen for DFS, but the suggestion of detriment is not as strong. Of the 118 recurrences or CBCs on both arms, 54 occurred among control patients, and 64 occurred among those who continued to take tamoxifen. This corresponds to a 21% proportional increase in event rate, which is not statistically significant (P = .31).
|
Table 3
also shows a twofold increase in the incidence of endometrial cancers (six versus 12) in women assigned to the longer therapy duration. This is consistent with the well-documented increase in risk for endometrial cancer in women using tamoxifen (1,13,14). Despite some (nonsignificant) excess seen in Table 3 for other second cancers and deaths before treatment failure among patients who received continued tamoxifen therapy, no other particular site of second primary cancer or cause of death was noticeably imbalanced between the two treatment arms (11).
It may be expected that the efficacy of continued tamoxifen therapy, relative to patients randomly assigned to receive placebo after an initial 5 years of tamoxifen therapy, will not remain constant over time. If the carryover effect of the initial tamoxifen on the outcome of patients assigned to receive placebo diminishes slowly over time, the relative benefit of continued treatment would be small in the initial period following randomization but greater later on as the effect of prerandomization therapy attenuates. Therefore, it is useful to look at the data after events have been categorized according to whether they occurred within years 5–10 from the initial surgical treatment (i.e., in the first 5 years after the divergence of treatments) or whether they occurred subsequent to that time. Table 4 summarizes the occurrence of first events that compose the DFS end point, while Table 5 shows the pattern for all-cause mortality. From Table 4 it is seen that, in the first 5 years after the divergence of treatments, the rate of occurrence of all DFS events among patients assigned to tamoxifen therapy was about 60% greater than that among patients who were randomly assigned to receive placebo (HR = 1.6 [95% CI = 1.1 to 2.2]; P = .007), but in the second 5 years, the relative risk was equal to 1.0. (A very similar pattern is seen for mortality in Table 5.) For breast cancer-related events (recurrences and CBCs), there is a statistically significant excess of events (48 versus 27 events; P = .02) in the first 5-year period among patients who received continued tamoxifen therapy. This pattern is not repeated in the second 5 years, and in fact, there is a nonsignificant net benefit of 11 events (16 versus 27 events; P = .13).
|
|
It is not clear how to interpret this pattern, which may be a result of chance. However, it is not easy to ascribe the suggestion of a "turnaround" in the data to a carryover effect, which would be more consistent with a pattern in which there was limited early benefit for continued treatment, coupled with greater benefit later on. The observed pattern could be consistent either with the phenomenon of tamoxifen withdrawal response or with the arrested growth of tumors that have become tamoxifen dependent, resulting in a temporary delay in recurrence following withdrawal of tamoxifen; but, of course, this is speculative.
Scottish Trial
The initial report of findings from the Scottish trial (9) concluded, after 6.2 years of follow-up, that a meaningful benefit from continuing adjuvant tamoxifen therapy beyond 5 years was unlikely. A recently published update of these data (12) (median follow-up, about 10 years) continues to show the same general pattern (i.e., those patients receiving >5 years of tamoxifen therapy had higher rates of recurrence and death than did those whose treatment stopped at 5 years), although the differences do not achieve statistical significance. Of the 169 patients randomly assigned to stop tamoxifen therapy after 5 years, 38 have had breast cancer events (recurrences or CBCs) subsequent to randomization, compared with 49 of the 173 patients assigned to continued treatment (Table 6). Similarly, there have been 16 more deaths from all causes among patients assigned to longer treatment (70 versus 54; HR=1.32 [95% CI = 0.93 to 1.88]; P = .12). Comparable results were obtained in analyses of deaths from breast cancer (HR=1.44 [95% CI = 0.87 to 2.38]; P = .15) and survival free of systemic disease, including CBCs (HR=1.36 [95% CI = 0.95 to 1.95]; P =.12). Even when the analysis was restricted to only those patients having strong ER expression (20 fmol/mg), the rates of breast cancer events, all-cause mortality, and breast cancer mortality did not differ statistically across study arms. As was the case in NSABP B-14, there were more endometrial cancers among patients randomly assigned to indefinite tamoxifen therapy than among those who stopped the drug after 5 years (six versus two, respectively).
|
Published Findings From ECOG E4181/E5181
In contrast to both NSABP B-14 (3) and the Scottish trial (9), published results from ECOG E4181/E5181 are more suggestive of a possible benefit for continuing treatment with tamoxifen beyond 5 years (10). After a median follow-up of 5.6 years following randomization at 5 years, a nonsignificant advantage was seen for extended tamoxifen therapy in terms of time to first recurrence or CBC (23 versus 15 breast cancer events; recurrence-free survival at 5 years after randomization = 73% for those stopping therapy at 5 years versus 85% for those continuing tamoxifen therapy; P = .10). Similar results were obtained in an analysis of DFS (defined in this analysis as the time from the randomization to first recurrence, CBC, or death without a prior breast cancer event). Of note, in a secondary analysis restricted to those patients who were ER positive (10 fmol/mg), the comparison of times to recurrence or CBC became statistically significant (22 versus 12 events; P = .014). There was, however, no evidence of any survival benefit for continued tamoxifen therapy (10 deaths in patients who received 5 years of treatment versus 14 among patients who received extended tamoxifen therapy, P = .52).
Updated Data From ECOG E4181/E5181
Unpublished updated analyses of the ECOG E4181/E5181 data were provided by the ECOG statistical office for discussion in this review (Gray R: personal communication). These data, which provide about 4 years of extended follow-up beyond the published findings, are summarized in Table 6. They continue to show a trend toward benefit for extended treatment. In an analysis of time to recurrence or CBC, the difference in breast cancer event rates between the two arms was statistically significant: 29 of 93 patients randomly assigned to stop tamoxifen therapy at 5 years experienced breast cancer recurrences or CBCs, compared with 17 of 100 patients randomly assigned to receive indefinite tamoxifen therapy (HR = 0.53; P = .03). A comparison of DFS between the two study arms was not statistically significant but was suggestive of some benefit for continued treatment (HR = 0.68; P = .13). However, the updated data still failed to demonstrate evidence of benefit in terms of overall mortality (23 deaths among patients who received 5 years of tamoxifen therapy, compared with 22 deaths among those assigned to receive indefinite tamoxifen therapy). Three endometrial cancers have been reported in patients who received extended tamoxifen therapy, while one has been reported in the group that stopped treatment after 5 years.
Preliminary Data From Ongoing Duration Trials
Both the Adjuvant Tamoxifen Treatment Offer More (aTTom) (15) and the Adjuvant Tamoxifen Longer Against Shorter (ATLAS) (16) trials continue to accrue patients with primary breast cancer who have received initial adjuvant treatment with tamoxifen and who are randomly assigned to either continue receiving the drug or to observation. Current accrual is essentially limited to patients who have received about 5 years of initial treatment (Peto R: personal communication). These trials are considerably larger than previous comparisons of 5 years' duration to longer treatment duration and should, over the next decade, provide definitive answers to the duration question.
At the recent Oxford EBCTCG meeting (September 21–23, 2000), unpublished data from 6779 aTTom and ATLAS patients, all of whom had been randomly assigned after completing—disease-free—at least 4 years of tamoxifen therapy, were combined in a blinded fashion with data from NSABP B-14 (3), the Scottish trial (9), and ECOG E4181/E5181 (10) to provide a preliminary meta-analysis of the tamoxifen duration question. Available follow-up of the aTTom and ATLAS patients was very short, with a median follow-up of probably no more than 1 year and virtually no follow-up beyond 5 years after randomization. Nevertheless, these patients had experienced 195 breast cancer events and 128 deaths, thereby adding substantially to the number of events observed in the earlier trials. The combined data resulted in estimated HRs for both breast cancer events and mortality that were not statistically different from 1. The combined data also provided a (statistically nonsignificant) suggestion of benefit in the second 5-year period following the divergence of treatments. However, almost all of the data supporting this trend came from NSABP B-14 (3), as is illustrated in Table 7. As has been discussed previously, the time-dependent pattern of recurrence seen in the B-14 trial is not readily explained by the carryover phenomenon.
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionDurations of up to...Treatment for 5 Years...DiscussionConclusions and RecommendationsReferences |
|---|
Randomized clinical trials of adjuvant tamoxifen therapy have overwhelmingly demonstrated its efficacy for patients with ER-positive breast cancer, without regard to age, lymph node status, or menopausal status at random assignment (1–3). In addition, within the last decade, sufficient evidence has been obtained to firmly conclude that 5 years of treatment is superior to short-term treatment of about 2 years (4–8). The advisability of continuing treatment beyond 5 years has not been established (3,9,10–12).
In the aggregate, data from published clinical trials comparing 5 years of tamoxifen therapy with more than 5 years of tamoxifen therapy are disappointing and do not demonstrate any benefit from extended treatment (Table 6). However, there are good reasons to withhold judgment on the duration question until considerably more data are available. First, the number of breast cancer events (recurrences and CBCs) and deaths observed to date in the published trials [NSABP B-14 (3), Scottish trial (9), and ECOG E4181/E5181 (10)] is considerably smaller (251 breast cancer events and 265 deaths) than is required to achieve consensus. This remains the case even if currently available preliminary data are included from the ongoing aTTom (15) and ATLAS (16) trials. Second, it has been firmly established that treatment with tamoxifen conveys a considerable carryover benefit to the patient that lasts some years after the termination of treatment. For this reason, one might expect the effectiveness of extended tamoxifen therapy (compared with therapy of just 5 years' duration) to be partially attenuated in the initial period following the divergence of treatments; indeed, it may not become apparent until extended follow-up has been obtained. Finally, the results of published trials comparing 5 years of tamoxifen therapy with treatments of longer duration are heterogeneous: Both the NSABP B-14 trial (3) and the Scottish trial (9) show no benefit and, in fact, are weakly suggestive of some detriment. The ECOG E4181/E5181 (10) trial, however, while smaller than the other two trials, is supportive of the hypothesis of continued benefit.
The apparent difference in the results of the ECOG trial (10) and the other two studies could be the result of chance, although a heterogeneity test for between-trial differences based on breast cancer events is marginally significant (P = .025). Alternatively, it has been suggested that the optimal duration of treatment might differ for lymph node-negative and lymph node-positive patients (10). The proportional reduction in risk afforded by extended tamoxifen therapy would probably not be expected a priori to be strongly dependent on lymph node status, since extensive overview data indicate that the proportional reduction in risk resulting from treatment with tamoxifen, in contrast to no hormonal treatment, does not differ greatly between lymph node-negative and lymph node-positive patients (1). Nevertheless, even if the proportional risk reduction associated with extended treatment relative to treatment of only 5 years were independent of lymph node status, this would not be the case for absolute risk, since the recurrence rate in node-negative patients is considerably less than that of node-positive patients, even after 5 disease-free years (17). Treatment with tamoxifen is associated with certain serious risks, most notably endometrial cancer, thromboembolic disease, and possibly stroke (1,2,11,13,14). Therefore, even if it could be demonstrated that continued tamoxifen therapy were efficacious, its risk/benefit ratio for lymph node-negative women would be higher than for lymph node-positive women (and higher for older women compared with younger women).
|
CONCLUSIONS AND RECOMMENDATIONS |
|---|
|
TopAbstractIntroductionDurations of up to...Treatment for 5 Years...DiscussionConclusions and RecommendationsReferences |
|---|
Strong evidence exists that 5 years of tamoxifen therapy is superior to 2–3 years of therapy, but as yet, there is no convincing evidence that extending treatment beyond 5 years provides additional benefit. Outside the clinical trials setting, adjuvant treatment with tamoxifen should, therefore, be limited to 5 years' duration until such time as additional data become available to definitively resolve the issue.
Because currently available evidence is insufficient to achieve consensus regarding the benefit of treatment with tamoxifen beyond 5 years, continued accrual of patients with hormone-responsive tumors from whom proper consent has been obtained to long-term tamoxifen trials, including aTTom and ATLAS, is appropriate. Alternatively, patients who have remained disease free for extended periods of time on tamoxifen therapy should strongly consider participation in clinical trials of other hormonal interventions, including aromatase inhibitors or selective ER modulators. Currently, crossover of hormone-responsive postmenopausal patients to aromatase inhibitors following 5 disease-free years of tamoxifen therapy is being studied in at least three trials, including National Cancer Institute of Canada Study MA 17 (letrozole) (18), NSABP B-33 (exemestane) (17), and Austrian Breast Cancer Study Group Study 6A (anastrozole) (18).
|
REFERENCES |
|---|
|
TopAbstractIntroductionDurations of up to...Treatment for 5 Years...DiscussionConclusions and RecommendationsReferences |
|---|
1 Early Breast Cancer Trialists? Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomised trials. Lancet 1998;351:1451–67.[CrossRef][Web of Science][Medline]
2 Fisher B, Costantino J, Redmond C, Poisson R, Bowman D, Couture J, et al. A randomized clinical trial evaluating tamoxifen in the treatment of patients with node-negative breast cancer who have estrogen-receptor-positive tumors. N Engl J Med 1989;320:479–84.[Abstract]
3
Fisher B, Dignam J, Bryant J, DeCillis A, Wickerham DL, Wolmark N, et al. Five versus more than five years of tamoxifen therapy for breast cancer patients with negative lymph nodes and estrogen-receptor-positive tumors. J Natl Cancer Inst 1996;88:1529–42.
4 Falkson HC, Gray R, Wolberg WH, Gillchrist KW, Harris JE, Tormey DC, et al. Adjuvant trial of 12 cycles of CMFPT followed by observation or continuous tamoxifen versus four cycles of CMFPT in postmenopausal women with breast cancer: an Eastern Cooperative Oncology Group phase III study. J Clin Oncol 1990;8:599–607.[Abstract]
5 Tormey DC, Gray R, Abeloff MD, Roseman DL, Gilchrist KW, Barylak EJ, et al. Adjuvant therapy with a doxorubicin regimen and long-term tamoxifen in premenopausal breast cancer patients: an Eastern Cooperative Oncology Group trial. J Clin Oncol 1992;10:1848–56.[Abstract]
6
Swedish Breast Cancer Cooperative Group. Randomized trial of two versus five years of adjuvant tamoxifen for postmenopausal early stage breast cancer. Natl Cancer Inst 1996;88:1543–9.
7
Current Trials Working Party of the Cancer Research Campaign Breast Cancer Trials Group. Preliminary results from the Cancer Research Campaign trial evaluating tamoxifen duration in women aged fifty years or older with breast cancer. J Natl Cancer Inst 1996;88:1834–9.
8
Delozier T, Spielmann M, Mace-Lesec'h J, Janvier M, Hill C, Asselain B, et al. for the Federation Nationale des Centres de Lutte Contre le Cancer Breast Group. Tamoxifen adjuvant treatment duration in early breast cancer: initial results of a randomized study comparing short-term treatment with long-term treatment. J Clin Oncol 2000;18:3507–12.
9 Stewart JH, Forrest AP, Everington D, McDonald CC, Dewar JA, Hawkins RA for the Scottish Cancer Trials Breast Group. Randomised comparison of 5 years of adjuvant tamoxifen with continuous therapy for operable breast cancer. Br J Cancer 1996;74:297–9.[Web of Science][Medline]
10
Tormey DC, Gray R, Falkson HC. Postchemotherapy adjuvant tamoxifen therapy beyond five years in patients with lymph node-positive breast cancer. J Natl Cancer Inst 1996;88:1828–33.
11
Fisher B, Dignam J, Bryant J, Wolmark N. Five versus more than five years of tamoxifen for node-negative breast cancer: updated findings. J Natl Cancer Inst 2001;93:684–90.
12
Stewart JH, Prescott RJ, Forrest AP. Scottish adjuvant tamoxifen trial: a randomized study updated to 15 years. J Natl Cancer Inst 2001;93:456–62.
13
Fisher B, Costantino JP, Redmond CK, Fisher ER, Wickerham DL, Cronin WM. Endometrial cancer in tamoxifen-treated breast cancer patients: findings from the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-14. J Natl Cancer Inst 1994;86:527–37.
14
Fisher B, Costantino JP, Wickerham DL, Redmond CK, Kavanah M, Cronin WM, 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.
15 Cancer Research Campaign Trials Unit. Adjuvant tamoxifen treatment offer more? (aTTom). Protocol. CRC Trials unit, Clinical Research Block, Queen Elizabeth Hospital, Birmingham, England, UK.
16 Clinical Trial Service Unit. Adjuvant tamoxifen longer against shorter (ATLAS). Protocol. April, 1995. ATLAS Office, Clinical Trials Service Unit. Radcliffe Infirmary, Oxford, England, UK.
17 National Surgical Adjuvant Breast and Bowel Project (NSABP B-33): A randomized, placebo-controlled, double-blind trial evaluating the effect of exemestane in clinical stage cT1–3 cN0–1 M0 postmenopausal breast cancer patients completing at least five years of tamoxifen therapy. Protocol. Pittsburgh (PA): NSABP; September, 2000.
18 Piccart M, Goldhirsh A, on behalf of the Breast International Group. An overview of recent and ongoing adjuvant clinical trials for breast cancer, 2nd ed. In: Biganzoli L, and Straehle C, editors. Mollem (Belgium): Moreau PCE; 2000.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. L. Kahn Moving Research From Bench to Bedside to Community: There Is Still More to Do J. Clin. Oncol., February 1, 2008; 26(4): 523 - 526. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||