Home: Educational Supplement: Appendix

Duration of Adjuvant Tamoxifen Therapy

John L. Bryant, Ph.D.

Clinical trials of adjuvant tamoxifen have convincingly demonstrated benefits for patients with estrogen receptor (ER)-positive breast cancer, regardless of nodal status, age, or menopausal status (EBCTCG, 1998; Fisher, Dignam, Bryant, et al., 1996). But despite the large number of trials, there is still uncertainty regarding the optimal duration of tamoxifen therapy. This review will address comparisons of duration of tamoxifen treatment up to about 5 years (about which considerable evidence has been reported in the past decade), and conclude with a summary of trials which have compared 5 years of treatment with longer durations (for which the evidence is not extensive).

Durations Up To 5 Years

Four trials which have been reported in the past decade present direct comparisons of the worth of about 5 years of tamoxifen to shorter durations. Two Eastern Cooperative Oncology Group (ECOG) trials (E4181 and E5181) compared 1 to 5 years of tamoxifen in post- and premenopausal node-positive patients, respectively (Falkson, Gray, Wolberg, et al., 1990; Tormey, Gray, Abeloff, et al., 1992). The Swedish Breast Cancer Cooperative Group (SBCCG) and Cancer Research Campaign (CRC) trials were larger studies comparing 2 to 5 years of treatment (SBCCG, 1996; CRC, 1996). A fifth trial (TAM-01), which was reported in abstract form in 1997, is designed to compare 2 to 3 years duration to 12 to 13 years, but the mean followup available as of the report was 4.5 years, so that currently available data from this trial might best be interpreted as comparing 2 to 3 years with 6 to 7 years (Delozier, Spielmann, Mace-Lesec’h, et al., 1997). All five studies have reported statistically significant reductions in event rates in favor of the longer duration. The three studies comparing about 2 to at least 5 years (SBCCG, CRC, TAM-01) all estimated a reduction in the event rate of about 20 percent. Only one (SBCCG) demonstrated a statistically significant survival advantage for the longer duration (18 percent reduction in mortality rate, SD 7 percent), with one other trial (CRC) also showing a modest but nonsignificant survival advantage (11 percent reduction, SD 13 percent). In light of the evidence for a reduced recurrence rate, it may be reasonable to attribute this to the relatively short followup reported to date, and to the well-known carryover effect of treatment with tamoxifen (EBCTCG, 1998; Fisher, Dignam, Bryant, et al., 1996). Indirect comparison trials of about 1, 2 or 5 years duration based on the 1998 EBCTCG overview lead to a similar conclusion, since the trends of a reduced recurrence rate and reduced mortality over this time range were strongly significant.

Five Years Treatment Versus Longer Duration

Three trials comparing 5 years of tamoxifen to longer periods of time were first reported in 1996. These included NSABP B-14, a double-blind comparison of 5 additional years of tamoxifen versus placebo in 1,172 node-negative women who had completed 5 years of initial treatment disease-free; the Scottish trial, in which 342 predominately node-negative women were assigned to indefinite tamoxifen or observation following the completion of 5 years of tamoxifen disease-free; and ECOG E4181/E5181, in which 87 postmenopausal node-positive women from E4181 and 107 premenopausal node-positive women from E5181 who were disease-free at 5 years were randomized to either indefinite tamoxifen or observation. All patients in B-14 were ER-positive >10 fmol/mg protein), compared to 73 percent in the ECOG trial. In the Scottish trial, 39 percent of the patients had tumors with > 20 fmol/mg, 22 percent had < 19 fmol/mg, and 39 percent were not assayed. The NSABP data have recently been updated; the Scottish and ECOG data will be updated at the EBCTCG meeting in 2000.

After a median followup from re-randomization of 6.75 years, the B-14 data indicate no advantage for continued tamoxifen, and in fact trend in the opposite direction. Disease-free survival (DFS) 7 years following re-randomization was 82 percent for placebo patients and 78 percent for those receiving more than 5 years of tamoxifen (HR=1.3, p=0.03). Overall survival was 94 percent for placebo patients and 91 percent for tamoxifen patients (39 versus 57 deaths, HR=1.5, p=0.07). The distribution of first events is shown in table 1. Events in the primary NSABP DFS analysis included recurrence, contralateral breast cancer (CBC), second primary cancer, and death. Table 1 shows that only about one-half (118 of 243) of all first events were “breast cancer related” (i.e., recurrences or CBC). To facilitate comparison with results of the ECOG trial described below, the comparison of treatment arms was restricted to only first recurrences and CBC. In terms of this endpoint the treatment difference was nonsignificant (HR=1.21, p=0.31), although the trend was still suggestive of a disadvantage for continued tamoxifen.

The NSABP data are compared to the Scottish and ECOG data in table 2. Findings to date from the Scottish trial are similar to those of B-14. A nonsignificant detriment was reported in the analysis of time to first recurrence, CBC, or death (HR=1.27, 95 percent CI 0.87–1.85) as well as overall survival (40 deaths versus 48). As in the NSABP trial, there were more endometrial cancers in those patients receiving tamoxifen. The published results from ECOG 4181/5181 are somewhat more positive. A nonsignificant advantage was seen for extended tamoxifen in terms of time to first recurrence or CBC (23 versus 15 events; recurrence-free survival at 5 years after re-randomization was 85 percent for those continuing tamoxifen versus 73 percent for those stopping at 5 years, p=0.10). In a secondary analysis restricted to patients who were ER-positive, this comparison became significant (22 versus 12 events, p=0.014). There was, however, no evidence of any survival benefit for continued tamoxifen (10 deaths in patients receiving 5 years of treatment versus 14 among patients receiving extended tamoxifen, p=0.52).

In aggregate, there have been more reported recurrences and CBCs (117 versus 105) and deaths (119 versus 89) in patients randomized to continue tamoxifen than to those who stopped treatment. The apparent difference in the results of the ECOG trial and the other two studies may be due to chance (a crude heterogeneity test for between-trial differences based on recurrence counts is suggestive but not significant, p=0.09). Alternatively, it has been suggested that the optimal duration of treatment may differ for node-negative and node-positive patients (Tormey, Gray, Falkson, 1996). This may be so even if the risk reduction associated with breast cancer was independent of nodal status (as suggested by the EBCTCG overview data), since the recurrence rate in node-negative patients is considerably less than that in node-positive patients, even after 5 disease-free years (Tormey, Gray, Falkson, 1996). Long-term treatment with tamoxifen is associated with certain serious risks, most notably thromboembolic disease, endometrial cancer, and, possibly, stroke (EBCTCG, 1998; Fisher, Costantino, Redmond, et al., 1994; Fisher, Costantino, Wickerham, et al., 1998). Therefore, even if it could be demonstrated that continued tamoxifen were efficacious, its risk/benefit ratio for node-negative women would be higher than for node-positive women (and higher for older women compared to younger).

Active Trials Testing Durations Beyond 5 Years

Both the aTTom (CRC Trials Unit, no date) and the ATLAS (Clinical Trials Service Unit, 1995) trials are now randomizing patients after 5 years of successful treatment with tamoxifen, either to stop the treatment or to continue it for an additional 5 years (Peto, personal communication). There are also three currently active trials in which hormone-responsive postmenopausal patients who have been successfully treated with tamoxifen for about 5 years are randomized to receive an additional 2 to 5 years of treatment with an aromatase inhibitor or a placebo. These include NCIC/EORTC/IBCSG/NCCTG/ECOG/ SWOG/CALGB MA 17 (Piccart, Goldhirsch, no date) (letrozole); NSABP B-33 (NSABP, 2000) (exemestane); and ABCSG Study 6A (Piccart, Goldhirsch, no date) (anastrozole).

Conclusions

There is strong evidence that 5 years of tamoxifen is superior to 2 to 3 years, at least in terms of delaying recurrence. Results up to now from trials comparing 5 to more than 5 years fail to indicate an additional advantage for continued tamoxifen, but the evidence is insufficient to achieve a satisfactory level of consensus, and continued enrollment of properly consented patients in long-term tamoxifen trials is appropriate. Alternatively, patients who have remained disease-free for extended periods of time on tamoxifen should be strongly considered for participation in clinical trials of other hormonal interventions, including aromatase inhibitors or selective estrogen receptor modulators (SERMs). Outside a clinical trials setting, adjuvant treatment with tamoxifen should be limited to 5 years duration until such time as additional data become available to definitively resolve the issue of tamoxifen duration.

References

Clinical Trial Service Unit, Radcliffe Infirmary, Oxford. Adjuvant tamoxifen longer against shorter (ATLAS). Protocol. April, 1995. ATLAS Office, Clinical Trials Service Unit. Radcliffe Infirmary, Oxford OX2 6HE, U.K. Abstract.

CRC Trials Unit, Birmingham. Adjuvant tamoxifen treatment offer more? (aTTom). Protocol. CRC Trials unit, Clinical Research Block, Queen Elizabeth Hospital, Birmingham B15 2TH, U.K. Abstract.

Current Trials Working Party of the Cancer Research Campaign (CRC) 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. Abstract.

Delozier T, Spielmann M, Mace-Lesec’h J, Janvier M, Luboinski M, Asselain B, et al. Short-term versus lifelong adjuvant tamoxifen in early breast cancer (EBC): a randomized trial (TAM-01). [abstract]. Proc Am Soc Clin Oncol 1997;16. Abstract.

Early Breast Cancer Trialists’ Collaborative Group (EBCTCG). Tamoxifen for early breast cancer: an overview of the randomized trials. Lancet 1998;351:1451-67. Abstract.

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 [published erratum appears in J Clin Oncol 1990;8:1603]. J Clin Oncol 1990;8:599-607. Abstract.

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. Abstract.

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. Abstract.

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. Abstract.

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 [submitted]. Abstract.

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. 2000. Pittsburgh: NSABP. Peto R, personal communication. Abstract.

Piccart M, Goldhirsch A, on behalf of the International Breast Cancer Study Group. Biganzoli L and Straehle C, editors. An overview of recent and ongoing adjuvant clinical trials for breast cancer, second ed. No Abstract Available.

Stewart HJ, Forrest AP, Everington D, McDonald CC, Dewar JA, Hawkins RA, et al. Randomised comparison of 5 years of adjuvant tamoxifen with continuous therapy for operable breast cancer. The Scottish Cancer Trials Breast Group. Br J Cancer 1996;74:297-9. Abstract.

Swedish Breast Cancer Cooperative Group (SBCCG). Randomized trial of two versus five years of adjuvant tamoxifen for postmenopausal early stage breast cancer. J Natl Cancer Inst 1996;88:1543-9. Abstract.

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.

Tormey DC, Gray R, Falkson HC. Postchemotherapy adjuvant tamoxifen therapy beyond five years in patients with lymph node-positive breast cancer. Eastern Cooperative Oncology Group. J Natl Cancer Inst 1996;88:1828-33. Abstract.