Adjuvant systemic therapy for hormone receptor positive early stage breast cancer INTRODUCTION — Adjuvant systemic therapy refers to the administration of chemotherapy, hormone therapy, or trastuzumab (a monoclonal antibody directed against HER-2/neu) following primary surgery for early breast cancer. The purpose is to eliminate or delay the subsequent appearance of clinically occult micrometastases, which are thought to account for distant treatment failures among women undergoing local therapy alone. The rationale for systemic adjuvant therapy is discussed elsewhere. A major determinant of the choice of adjuvant systemic therapy is whether an individual breast cancer expresses estrogen (ER) or progesterone (PR) receptors. Adjuvant hormone therapy benefits patients with hormone-responsive (ie, ER-positive) breast cancer, but not those with hormone-nonresponsive disease. Here we will review the data from adjuvant trials evaluating hormone therapy in all groups of women with early stage ER-positive (ER+) breast cancer, and those trials that specifically address whether chemotherapy adds benefit to hormone therapy in these women. BENEFIT OF HORMONE THERAPY — The goal of adjuvant hormone therapy is to prevent breast cancer cells from receiving stimulation from endogenous estrogen. Beatson's historic observations on the regression of advanced breast cancer following oophorectomy over 100 years ago provided the first insight into the estrogen-dependent nature of breast cancer. At about the same time, irradiation of the ovaries was suggested as a means of improving outcomes following primary breast cancer surgery in premenopausal women; this was perhaps the first demonstration of the benefit of adjuvant systemic therapy. Ovarian ablation (OA) became standard therapy in the 1950s and 1960s for women with metastatic breast cancer and in the adjuvant setting. It subsequently became clear that this approach was most effective for women whose tumors expressed ER and/or PR (see below). After the 1960s, surgical and radiotherapeutic OA was gradually replaced with pharmacologic hormone therapy, including blockade of the ER (eg, tamoxifen), suppression of estrogen synthesis by luteinizing hormone-releasing hormone (LHRH) agonists (eg, goserelin), and, in postmenopausal women, aromatase inhibitors (eg, anastrozole). Relative versus absolute benefit — When evaluating the benefit of adjuvant therapy for breast cancer, there is an important distinction in terminology between absolute reduction in the risk of recurrence or death from breast cancer and a relative or proportional reduction in these risks. The following example highlights the difference between absolute and relative risk reduction. Consider a woman with a 60 percent chance of a cancer recurrence based upon the stage of her primary tumor. A treatment that reduces the chance of recurrence by 33 percent (hazard ratio [HR] for recurrence 0.67) would decrease the likelihood of recurrence by one third, or 20 percent, because 20 is one-third of 60. In this example, the relative reduction in the risk of breast cancer recurrence is 33 percent, and the risk of breast cancer recurrence decreases from 60 to 40 percent (ie, the absolute magnitude of benefit is 20 percent). By contrast, a woman with a more favorable prognosis tumor, with only a 15 percent chance of recurrence based upon her stage of disease, would realize a 5 percent gain in absolute risk of recurrence for the same proportional hazard reduction. Predictive value of ER status — The importance of hormone receptor (ER) positivity as a predictive factor for response to adjuvant hormone therapy is best illustrated by data from the Early Breast Cancer Trialists Collaborative Group (EBCTCG) overview analyses This international group meets every five years to review and update the results of global trials evaluating the worth of adjuvant systemic therapy for early breast cancer. The most recent 2000 EBCTCG analysis (published in 2005) demonstrated that five years of tamoxifen (compared to no adjuvant therapy) was associated with a 41 percent reduction in the annual risk of relapse, and a 34 percent reduction in the annual risk of death for women with ER+ breast cancer. In contrast, tamoxifen provided no significant benefit for women with ER-negative disease, while for those whose tumors were ER-unknown, adjuvant tamoxifen was associated with a 31 percent reduction in the annual risk of relapse, and a 20 percent reduction in the annual death rate. TAMOXIFEN — Tamoxifen, a selective estrogen receptor modulator (SERM), inhibits the growth of breast cancer cells by competitive antagonism of estrogen at the ER. Its actions are complex, however, and it also has partial estrogen agonist activity. These agonist effects can be both beneficial (eg, they may help prevent bone demineralization), and detrimental, since they are associated with increased risk of uterine cancer and thromboembolic events. Benefits — The EBCTCG overviews have contributed greatly to our understanding of the benefit of adjuvant tamoxifen. In the most recent overview, five years of tamoxifen (compared to no adjuvant treatment) was associated with a 41 percent reduction in the annual risk of relapse, and a 34 percent reduction in the annual risk of death. This translated into a 12 percent absolute reduction in the likelihood of 15-year recurrence (33 versus 45 percent), and a 9 percent reduction in breast cancer-related mortality (26 versus 35 percent). The magnitude of the reduction in mortality was almost three times as great at 15 years as it was at five years (9 versus 3.6 percent), while the impact on recurrence was mainly seen in the first five years. Proportional risk reductions were similar for older as compared to younger women, and for node-negative (N-) and node-positive (N+) disease, although the absolute magnitude of benefit was greater for women with N+ breast cancer. Five-year recurrence rates for N+ breast cancer were 25 versus 41 percent with five years of tamoxifen versus no tamoxifen, an absolute difference of 16 percent; the corresponding values for N- disease were 11 versus 20 percent, an absolute difference of 9 percent. Benefits were independent of the use of chemotherapy. Compared to chemotherapy alone, the proportional reduction in the annual risk of recurrence and death from the addition of five years of tamoxifen to chemotherapy was 40 and 39 percent, respectively. This translated into a significant 10.6 percent reduction in the rate of recurrence at five years compared to the use of chemotherapy alone (17.5 versus 28.1 percent, respectively). There was no difference in efficacy between tamoxifen doses of 20 mg/day and 30 or 40 mg/day, although this represents an indirect comparison since patients were not randomly assigned to different tamoxifen doses within any one trial. The influence of treatment duration is discussed below. Risk of contralateral breast cancer — In addition to reducing rates of disease recurrence and death, tamoxifen also decreases the risk of developing a contralateral breast cancer. In the EBCTCG overview analysis, adjuvant tamoxifen for about five years was associated with a 39 percent reduction in the annual risk of contralateral breast cancer in women with ER+ tumors. The magnitude of reduction in risk of contralateral breast cancer using five years of tamoxifen was even higher (55 percent) in a contemporary randomized placebo-controlled trial Other data suggest that adjuvant treatment with the aromatase inhibitor anastrozole may result in an even greater reduction in risk of contralateral breast cancer compared to tamoxifen . The reduction in risk appears to be limited to ER+ and not ER-negative breast cancers Treatment duration — At least four trials, which were entirely or predominantly composed of postmenopausal women, have demonstrated significant improvements in outcome from five as compared to two years of tamoxifen therapy . As a result, five years has become the standard treatment duration. Five years versus longer — Three randomized trials have directly compared five years of tamoxifen to a longer treatment duration (10 years or indefinite) Two of the three, a National Surgical Adjuvant Breast and Bowel Project (NSABP) trial and a Scottish trial, suggest that five years of therapy provides optimal benefit, with a longer duration possibly associated with a worse outcome . In addition, the incidence of endometrial cancer appeared to be two-fold higher in women who continued tamoxifen for longer than five years (2.1 versus 1.1 percent in one trial . These data have led to the general recommendation that tamoxifen should not be continued beyond five years. However, this conclusion may not be valid for all subsets of women; all women in the NSABP trial and the majority in the Scottish trial had N- disease. In contrast, all of the women in the third trial, sponsored by the Eastern Cooperative Oncology Group (ECOG), had N+ breast cancer. The initial report of this trial suggested that outcomes with tamoxifen treatment beyond five years were not worse, and in fact there was a significantly longer relapse-free survival (RFS) and a trend toward longer overall survival (OS) in women with ER+ disease Longer follow-up of the ECOG trial and results of several ongoing trials comparing two to five years versus indefinite tamoxifen (the largest of which are the ATLAS and ATOM trials), will clarify these issues. Until then, five years has become the standard duration for adjuvant tamoxifen in women with ER+ early breast cancer. Treatment compliance may be a significant issue. In a series of women with early breast cancer enrolled in the New Jersey Medicaid or Pharmaceutical Assistance Program for the Elderly or Disabled over a six year period, 23 percent missed taking tamoxifen on more than one-fifth of the days studied, and overall adherence with therapy decreased to 50 percent by the fourth year of treatment The clinical impact of nonadherence is unknown; there are no data examining the efficacy of tamoxifen at lower doses or less frequent intervals than standard recommendations. Side effects — A number of potential adverse effects are associated with the administration of tamoxifen. These include hot flashes and vaginal discharge in the short-term, and a long-term increase in the risk of thromboembolic events as well as a two- to three-fold higher risk of endometrial cancer and uterine sarcomas. The magnitude of these risks is discussed in detail elsewhere. Summary — Tamoxifen 20 mg daily is a standard adjuvant treatment option for both premenopausal and postmenopausal women with ER+ early breast cancer. The optimal duration of therapy for women with N- disease is five years. The benefit of adding ovarian ablation to adjuvant tamoxifen in premenopausal women is discussed below, as is the benefit of sequential aromatase inhibitors in women who have completed two to five years of tamoxifen. AROMATASE INHIBITORS IN POSTMENOPAUSAL WOMEN — Aromatase inhibitors (AIs) markedly suppress plasma estrogen levels in postmenopausal women by inhibiting or inactivating aromatase, the enzyme responsible for synthesizing estrogens from androgenic substrates, In contrast to tamoxifen, these compounds lack partial agonist activity. Several generations of both steroidal and nonsteroidal AIs have been developed AIs are only used in postmenopausal women. In premenopausal women, the reduced feedback of estrogen to the hypothalamus and pituitary leads to an increase in gonadotropin secretion, which stimulates the ovary, leading to an increase in androgen substrate and aromatase Efficacy — The benefit of AIs was not addressed in the EBCTCG overview analysis. However, at least five randomized trials have explored AIs as adjuvant therapy:
AIs versus tamoxifen — In the ATAC trial, anastrozole (1 mg daily) was compared to tamoxifen (20 mg daily) or the combination for five years in 9366 postmenopausal women with ER+ (or unknown) breast cancer. In the latest update with median 68 month follow-up, compared to tamoxifen, anastrozole was associated with significantly longer disease-free survival (DFS, HR for the number of events [disease recurrence, contralateral breast cancer, death without recurrence] 0.87, 95% CI 0.78-0.97), a prolonged time to recurrence, reduced distant metastases (HR 0.86, 95% CI 0.74-0.99) and a 42 percent lower proportional risk of contralateral breast cancer compared to tamoxifen Since tamoxifen reduces the risk of contralateral breast cancer by about 50 percent, these findings suggest that anastrozole might prevent 70 to 80 percent of ER+ contralateral tumors. To date, there is no difference in OS, and no advantage to combination therapy compared to tamoxifen alone. Perhaps more importantly, there were differences in adverse effects. Compared to tamoxifen, the incidence of ischemic cerebrovascular events (2.0 versus 2.8 percent), endometrial cancer (0.2 versus 0.8 percent), hot flushes (40.9 versus 35.7 percent), venous thromboembolic events (2.8 versus 4.5 percent) and vaginal bleeding (5.4 versus 10.2 percent) were all less with anastrozole, but the incidence of bone fractures (11 versus 7.7 percent) and musculoskeletal pain were higher As a results of these data, anastrozole was approved in the United States for adjuvant therapy of postmenopausal women with ER+ breast cancer, and many clinicians prefer AIs over tamoxifen, particularly for ER+PR- breast cancers. Letrozole was compared to tamoxifen in the multicenter Breast International Group (BIG) 1-98 trial that randomly assigned 8,010 postmenopausal women with early breast cancer to tamoxifen or letrozole for five years, or tamoxifen or letrozole for two years, followed by the alternative agent for three years In an early report with 28.5 month follow-up, women assigned to initial letrozole (2.5 mg daily) rather than tamoxifen had significantly better event-free survival (HR for the number of events 0.81, p = 0.003), particularly the risk of a distant recurrence (HR 0.73). This translated into an absolute improvement in five year DFS of 84 versus 81.4 percent, respectively). Letrozole is approved in the United States for adjuvant therapy. Sequential tamoxifen and AIs — In addition to the BIG 1-98 trial, at least five other trials have compared tamoxifen alone to sequential administration of tamoxifen and an AI.
In the most recent report, with 30 month median follow-up, four-year DFS was significantly better with sequential letrozole (94 versus 90 percent, HR 0.58) as was distant DFS. While OS was similar in both groups at four years (95 percent each), a planned subgroup analysis showed a significant survival benefit in patients with N+ disease (HR 0.61, 95% CI 0.38-0.98). Longer-term survival data is unlikely to be useful because of unblinding and crossover in the placebo group. Low-grade hot flashes, anorexia, myalgias, and arthralgias were significantly more common with letrozole, while vaginal bleeding was less frequent. More patients receiving letrozole had a bone fracture, a new diagnosis of osteoporosis or cardiovascular disease, but only the incidence of osteoporosis was significantly different between the two groups. Despite these issues, the impact of switching to letrozole on quality of life was marginal Letrozole is approved in the United States for use in postmenopausal women after five years of tamoxifen. An extension of the MA 17 trial is ongoing, in which women receiving five years of letrozole are randomly assigned to discontinue or continue the drug for five more years.
An EORTC trial randomly assigned 4742 women with ER+ (81 percent) or unknown breast cancer (51 percent N-) to two to three years of tamoxifen, followed by a randomization to exemestane (25 mg daily) or further tamoxifen to complete five years of hormone therapy. With median 31 month follow-up, women who switched to exemestane had a significant 32 percent reduction in the risk of recurrence, which corresponded to a 5 percent absolute improvement in DFS at three years (the primary endpoint, 92 versus 87 percent, respectively). OS was similar in both groups. The exemestane group also had a significantly fewer contralateral breast cancers (9 versus 20 cases, respectively). Although arthralgias and diarrhea were more common with exemestane, gynecologic symptoms and thromboembolic events were significantly less common. Similar results were shown in three trials using anastrozole. In the larger of the two (which was a combined analysis of two trials conducted by the German Adjuvant Breast Cancer Group [the ARNO 95 trial] and the Austrian Breast and Colorectal Cancer Study Group [ABCSG trial 8]), 3224 postmenopausal women with ER-positive early breast cancer completing two years of tamoxifen were randomly assigned to anastrozole (1 mg daily) or to continue tamoxifen for a total of 5 years. With 28 month median follow, women receiving anastrozole had a significant 40 percent reduction in the risk of recurrence (three-year event-free survival 95.8 versus 92.7). There were significantly more fractures and significantly fewer thromboses with anastrozole. Thus, compared to five years of tamoxifen, sequencing of tamoxifen and AIs reduces disease relapse and the incidence of contralateral second primary breast cancers, but an impact on survival has not yet been shown. Letrozole and exemestane are both approved in the United States for this indication. Side effects — All AIs cause bone loss by lowering the levels of endogenous estrogen, while tamoxifen is associated with estrogenic (ie, protective) effects in the bones of postmenopausal women. AIs are also associated with an increased incidence of musculoskeletal complaints (5 to 10 percent, ranging from mild to severe). Unlike tamoxifen, they carry no increased risk of endometrial cancer, thromboembolism Summary and ASCO recommendation — An expert panel convened by ASCO in 2004, which incorporated data from all four randomized trials, came to the following conclusions regarding the role of AIs in the adjuvant therapy for ER-positive (ER+) breast cancer
We agree with these recommendations. It is unclear whether an AI alone is more effective than a combination of tamoxifen for two, three or five years followed by an AI. Furthermore, the role of additional tamoxifen in women who have received AIs for five years is unknown. The role of AIs in adjuvant therapy for postmenopausal women will be better defined in the near future with the mature results of the BIG 1-98 trial, which directly compares five years of letrozole, five years of tamoxifen, and sequential letrozole plus tamoxifen. DOES CHEMOTHERAPY ADD BENEFIT TO TAMOXIFEN IN POSTMENOPAUSAL WOMEN? — The benefit of adding systemic chemotherapy to tamoxifen in postmenopausal women has been separately studied in N+ and N- breast cancer. The most common chemotherapy regimens are outlined in Table 2 Node-positive — In women with N+ ER+ breast cancer, randomized trials comparing tamoxifen alone with tamoxifen plus oral (but not IV CMF, anthracycline-based combination regimens, and epirubicin alone have all demonstrated significant benefits for combined therapy in terms of improved DFS. An OS benefit for adding chemotherapy to tamoxifen has been shown in two trials, both of which used a doxorubicin-containing regimen. These trials are briefly summarized below: NSABP B16 — In
NSABP trial B16, in which AC plus tamoxifen was compared to tamoxifen
alone in 1124 women
INT 0100 — Intergroup trial 0100 compared tamoxifen with or without CAF in 1558 premenopausal or postmenopausal women with ER+, N+ breast cancer; a further randomization was to concurrent or sequential administration With median seven year follow-up, the addition of CAF to tamoxifen was associated with a significant benefit in DFS (76 versus 67 percent) and OS (84 versus 79 percent). The latest update of this trial also confirmed the superiority of sequential rather than concurrent chemotherapy (see below) Node-negative — For postmenopausal women with ER+ N- early breast cancer, the benefit of adding chemotherapy to tamoxifen is less clear: NSABP B-20 — NSABP B-20 randomly assigned 2036 premenopausal or postmenopausal women with ER+, N- breast cancer to tamoxifen alone, tamoxifen plus methotrexate and 5-fluorouracil, or tamoxifen plus six cycles of oral CMF. The combination of either chemotherapy regimen with tamoxifen was associated with significantly better DFS and OS compared to tamoxifen alone. The investigators concluded that both premenopausal and postmenopausal women with breast cancer who met NSABP entry criteria regardless of other factors, were "candidates for chemotherapy." In a later analysis with 12-year follow-up, the benefit of combined
therapy was statistically significant for overall RFS (89 versus 79
percent) but not OS (87 versus 83 percent, p = 0.063). In subset analysis,
a significant survival advantage was evident in women
The possibility that gene expression analysis on formalin-fixed, paraffin-embedded tissue (the Oncotype DX assay) could identify those women with N- ER+ disease who have a low enough risk of recurrence to justify eliminating chemotherapy was suggested in a separate analysis of data from this trial. However, whether the Oncotype DX assay should be used for clinical decision making at present is controversial. These data, and clinical use of this assay are discussed in detail elsewhere. section on Markers to predict benefit from chemotherapy). IBCSG Trial IX — Contrasting results were noted in International Breast Cancer Study Group (IBCSG) Trial IX, in which 1669 postmenopausal women with N- breast cancer were randomly assigned to oral CMF (only three courses compared to the six courses administered in NSABP B-20) followed by tamoxifen, versus tamoxifen alone In the ER+ group (n = 1227), there was no benefit to the addition of chemotherapy compared to tamoxifen alone (five year DFS 84 versus 85 percent, OS 95 versus 93 percent, respectively). Not surprisingly, the addition of CMF to tamoxifen was of significant benefit for the 382 women with ER-negative tumors (five-year DFS 84 versus 69 percent, OS 89 versus 81 percent). Older women — Women over 70 represent a small minority in reported trials involving postmenopausal women. They often have comorbid diseases and are more likely to tolerate treatment poorly. However, age alone should not exclude women from receiving adjuvant chemotherapy if she is in good health. The potential risks and benefits of chemotherapy for each individual patient should be discussed thoroughly; good clinical judgment is paramount. Timing of chemohormonal therapy — As noted above, the results of INT 0101 suggested a significant benefit for sequential chemotherapy followed by tamoxifen as compared to concurrent chemohormonal therapy at 10 years in terms of both DFS (60 versus 53 percent), and OS (68 versus 62 percent). As a result of these data, sequential rather than concurrent therapy has been adopted as standard of care when both chemotherapy and tamoxifen are administered.
For postmenopausal woman with N- disease and larger tumor size or other adverse features as defined by the International Breast Cancer Consensus Group, we suggest the use of oral CMF chemotherapy in addition to hormone therapy. Although the benefit of adding oral CMF to tamoxifen may be mainly limited to younger postmenopausal women,it is not clear what age should be used to define the "cutoff." The benefit of anthracyclines in this setting is incompletely studied.
OVARIAN ABLATION IN PREMENOPAUSAL WOMEN — A resurgence of interest in ovarian ablation (OA) followed the publication of data from the Early Breast Cancer Trialists Collaborative Group (EBCTCG) suggesting that OA either by surgery or irradiation was significantly better than no adjuvant treatment Benefit of ovarian ablation — The most recent EBCTCG overview analysis summarized results from 4900 women included in 15 randomized trials assessing OA or ovarian suppression. Only about 1300 of these women were in trials of OA in the absence of chemotherapy, while more than 3500 were in trials of OA in the presence of chemotherapy. Because menopausal status was not uniformly reported, the main analysis was limited to women under age 50 when randomized, most of whom would have been premenopausal at diagnosis. There was a clear separation between trials that studied OA versus no treatment in the absence of chemotherapy. and those that studied OA plus chemotherapy versus the same chemotherapy. For women with ER-positive or unknown breast cancer, there were large and significant positive effects of OA in the absence of chemotherapy in terms of 15-year recurrence (59 versus 46 percent, an absolute benefit of 13 percent), breast cancer deaths (59 versus 49 percent, an absolute difference of 10 percent), and all deaths (57 versus 46 percent, an absolute difference of 11 percent). In contrast, the trials of OA plus chemotherapy versus the same chemotherapy found no significant difference in terms of recurrence (53 versus 56 percent), breast cancer deaths (47 versus 52 percent) or all deaths (47 versus 52 percent). This might reflect the interference of concurrent hormonal therapy on the efficacy of chemotherapy, or the effect of chemotherapy-induced ovarian failure on responsiveness to other ovarian treatments. LHRH analogs — Oophorectomy and radiotherapeutic OA have been largely replaced by temporary ovarian suppression using luteinizing hormone releasing hormone (LHRH) agonists. At least in the setting of metastatic breast cancer, goserelin is as effective as oophorectomy The benefit of adjuvant goserelin was shown in the ZIPP (Zoladex in Premenopausal Patients) trial, in which 2648 premenopausal women with early breast cancer (70 percent ER-positive) were randomly assigned, using a 2x2 factorial design, to tamoxifen for two years (an inferior duration by modern standards), goserelin for 26 months, goserelin plus tamoxifen (GT), or no endocrine therapy In a preliminary report with median 4.3 year follow-up, goserelin (but not two years of tamoxifen) was associated with a significant 23 percent reduction in the risk of a first event (20 versus 25 percent with no endocrine therapy) that was most pronounced in women with ER-positive disease, and those over age 40. Benefit appeared to be somewhat less among women who received concurrent tamoxifen or chemotherapy, but the difference compared to those who did not receive such concurrent treatments was not statistically significant. There were also fewer deaths in the women allocated to goserelin, although the difference was not statistically significant (10.7 versus 12.4 percent, hazard ratio [HR] 0.84, p = 0.12). Several issues related to study design are pertinent. Women were allowed to electively receive tamoxifen while being randomized just for goserelin. Furthermore, adjuvant chemotherapy was permitted at the discretion of the treating physician (and given to 43 percent of enrolled women). Only 70 percent of the 2106 patients with known hormone receptor status were ER-positive. Does ovarian ablation add benefit to tamoxifen? — At least in the setting of metastatic disease, four small studies suggest a survival benefit for tamoxifen plus an LHRH analog versus an LHRH analog alone. These data suggest that combining an LHRH analog and tamoxifen may confer additional benefit over either drug alone in the adjuvant setting. However, the available data are conflicting:
Ovarian ablation and chemotherapy — The recognition of the similar magnitude of benefit for OA and chemotherapy in premenopausal women undergoing adjuvant systemic therapy led to several trials directly comparing both interventions, Ovarian ablation versus CMF — At least six studies have compared OA with CMF , five of which have been criticized because they used IV CMF , which is less effective than standard oral CMF The largest (ZEBRA) trial compared CMF x 6 (with 83 percent of patients receiving IV CMF) to two years of goserelin in 1640 node-positive (N+) premenopausal women. There was a strong interaction between treatment and ER status. With six-year median follow-up, goserelin was equivalent to CMF in women with ER-positive disease (75 percent of those in the study) in terms of disease-free survival (DFS) and OS, while goserelin was significantly inferior to CMF in ER-negative women (HR for DFS, 1.76; for OS, 1.77). Within the group randomly assigned to CMF, five year DFS was significantly better among women who were amenorrheic at 36 weeks compared to those who were not (HR 0.65); older women were more likely to become amenorrheic. A later report compared QOL among women assigned to CMF for 6 months or two years of goserelin, with the following conclusions
These tradeoffs between the early (three to six months) relative improvement with goserelin, intermediate (six to 24 months) benefits of chemotherapy, and late (beyond two years) benefit of goserelin may be of interest to patients when considering the two treatment options. The late benefits of goserelin occur because side effects disappear after the drug is discontinued, and patients are no longer menopausal. In a preliminary report of a survey in which patients were asked whether they would prefer a type of therapy with the general side effects of CMF versus one with the side effects of two to three years of goserelin, goserelin was almost uniformly preferred The only trial to compare goserelin alone to oral CMF was the IBCSG trial VIII, in which 1063 pre- or perimenopausal women (70 percent ER-positive) with N- breast cancer were randomly assigned to two years of monthly goserelin, oral CMF x 6, or oral CMF x 6 followed by 18 months of goserelin . With median seven year follow-up, outcomes were similar in all three groups among women with ER-positive disease; those with ER-negative disease fared better with CMF Combined hormonal therapy versus CMF — At least two completed trials suggest that tamoxifen plus OA provides similar or better efficacy compared to CMF In an Italian study, oral CMF x 6 was compared to tamoxifen plus OA (oophorectomy, ovarian irradiation, or monthly goserelin for two years) in 244 premenopausal or perimenopausal women with early breast cancer (85 percent N+). At a median follow-up of 76 months, there were no significant differences in either OS or DFS between the groups. A slightly better outcome for combined hormone therapy compared to IV CMF was noted in an Austrian trial, in which 1034 premenopausal women with early stage ER-positive breast cancer (one-half N+) were randomly assigned to IV CMF x 6 or three years of monthly goserelin injections plus tamoxifen (GT) for five years. With 60 month median follow-up, combined hormonal therapy was associated with significantly better five year DFS (81 versus 76 percent) and local relapse-free survival (RFS, 95 versus 92 percent), and a statistically insignificant trend towards improved OS (p = 0.20). The use of IV CMF may have contributed to the better results with hormonal therapy. Hormone therapy versus anthracyclines — Given the relative superiority of anthracycline-containing chemotherapy regimens in the adjuvant setting, a relevant question is whether the same result could be obtained if anthracycline-based chemotherapy were compared with OA. This issue has not been well studied. Two trials published in abstract form by the same investigators suggest similar outcomes with FEC/FAC versus tamoxifen plus OA. In the larger trial, three years of tamoxifen plus triptorelin was compared to IV FEC x 6 in 333 premenopausal women with ER-positive, N+ disease. In a preliminary report with 54 month follow-up, OS was similar, and there was only an insignificant trend towards higher DFS with hormone therapy (92 versus 81 percent). Summary — Taken together, these results support the view that ovarian ablation (OA) is at least as effective as oral CMF and possibly FEC/FAC chemotherapy in premenopausal women with ER-positive, N+ breast cancer However, this remains a controversial issue for the following reasons:
As a result, in North America, premenopausal women with N+ breast cancer are more likely to be offered chemotherapy plus hormone therapy as compared to hormone therapy alone. For women who choose medical hormone therapy, the optimal regimen is also unclear; we recommend two to five years of an LHRH analog or five years of tamoxifen . We do not recommend combined hormonal therapy. Is adding ovarian ablation to chemotherapy beneficial? — A more important question in premenopausal women is whether the addition of OA to chemotherapy provides added benefit over chemotherapy alone. EBCTCG analysis — In the EBCTCG analyses, there was no significant advantage of adding OA to chemotherapy versus the same chemotherapy alone in women with ER-positive breast cancer. This finding was attributed, at least in part, to some degree of medical ovarian ablation from the chemotherapy itself. This hypothesis is supported by several (but not all) studies, in which amenorrheic premenopausal women have better RFS and OS compared to those who remain menstrual after chemotherapy In keeping with this conclusion, at least three contemporary trials have failed to show a benefit for adding OA by itself (ie, without tamoxifen) to chemotherapy in premenopausal women INT 0101 trial — Unites States Intergroup trial INT 0101 randomly assigned 1504 premenopausal women (70 percent over the age of 40) with N+, ER-positive disease to oral CAF x 6 alone, or followed by five years of goserelin, or five years of goserelin plus tamoxifen (GT). The addition of goserelin to CAF did not improve DFS or OS compared to CAF alone. In an unplanned retrospective analysis, the combination seemed to be more effective for women under the age of 40 (who are less likely to become postmenopausal as a result of chemotherapy). However, this hypothesis is speculative, and requires further study in prospective randomized controlled trials. Furthermore, interpretation of the trial results are complicated by the fact that although many women became amenorrheic and postmenopausal from the chemotherapy, they were included in the trial because they started as premenopausal women. IBCSG trial VIII — In the previously described IBCSG trial VIII, 1063 women (70 percent ER-positive) with N- breast cancer were randomly assigned to two years of goserelin, oral CMF x 6, or oral CMF x 6 followed by 18 months of goserelin). With seven-year median follow-up, there was no significant benefit for goserelin plus CMF versus CMF alone in women with ER-positive tumors (five-year DFS, 86 versus 81 percent). However, when the subset of women under the age of 39 were considered, the five year DFS rate was significantly greater with CMF plus goserelin as compared to either of the two other groups (83 versus 62 and 63 percent, respectively). These data may reflect the importance of prolonged ovarian suppression in women at highest risk of resuming menses after cytotoxic chemotherapy. French trial — A similar lack of benefit from adding OA to chemotherapy was seen in a multicenter randomized trial comparing adjuvant chemotherapy with and without OA (the LHRH analog triptorelin or ovarian irradiation) in 926 premenopausal women with N+ disease (76 percent ER-positive). Anthracycline-based regimens were given to 75 percent, while the remainder received a CMF-like regimen. There was no improvement in ten-year DFS (49 percent in both arms) or OS (66 versus 68 percent) from the addition of OA. One reason for the lack of benefit may have been that 73 percent of control patients became menopausal in the first three years of follow-up. Summary — Compared to chemotherapy alone, the available data does not support a benefit for adding OA to adjuvant chemotherapy in premenopausal women. However, whether this approach is beneficial for women who remain menstrual after chemotherapy is unclear (see below). Is adding tamoxifen plus OA to chemotherapy beneficial? — At least two trials have addressed the question of whether the combination of ovarian suppression plus tamoxifen adds benefit to chemotherapy in premenopausal women: INT 0101 trial — The previously described INT 0101 trial randomly assigned 1504 women with N+, ER-positive disease to oral CAF x 6 alone or followed by five years of goserelin, or CAF x 6 followed by five years of goserelin plus tamoxifen (GT) [71]. Compared to both CAF alone and CAF plus goserelin, CAF plus GT was associated with significant better nine-year DFS (68 versus 57 and 60 percent, respectively) but the difference in OS did not reach the level of statistical significance (76 versus 70 and 73 percent, respectively). Mam-1 trial — Similar results were seen in the Italian Mam-1 trial in which 466 premenopausal women with N+ breast cancer were randomly assigned to one of four groups: CMF, CMF followed by GT for two years, doxorubicin followed by CMF, or doxorubicin plus CMF followed by GT for two years. With median 6-year follow-up, there was a significant DFS advantage for the GT-containing arms compared to chemotherapy alone (HR 0.71), but only a trend toward better OS (HR 0.84, 95% CI 0.54 to 1.32). Results were not stratified according to ER status. These results are consistent with the overall body of literature (including the EBCTCG overview that some sort of endocrine therapy is beneficial for women with ER-positive disease, even if they have receive chemotherapy. The results do not provide insight as to whether tamoxifen, OA, or both represent superior treatment. Three international trials are ongoing to address the issue of hormone therapy in young women:
Until these studies are completed, optimal adjuvant hormone therapy for menstruating women will remain uncertain. Eligible women should be encouraged to enroll on these trials Timing of chemohormonal therapy — For women who receive both chemotherapy and hormone therapy, sequential is preferable to concurrent administration. This issue was directly tested in the INT 0100 trial, which randomly assigned postmenopausal women to tamoxifen versus CAF plus either sequential or concurrent tamoxifen. In a preliminary report, there was a significant DFS advantage for women receiving sequential tamoxifen and CAF compared to concurrent therapy. Although a significant OS benefit is not yet evident, further maturation of these data is needed. Summary — The addition of hormone therapy to adjuvant chemotherapy is indicated in women with higher risk, ER-positive disease (eg, N+ or advanced primary tumor stage) The optimal hormonal regimen is unclear. If young women remain menstrual after chemotherapy, we recommend three years of monthly goserelin, and for others, five years of tamoxifen. We encourage participation in on of the three ongoing trials described above When tamoxifen is added to systemic chemotherapy, it should be administered after chemotherapy is completed rather than concurrently. TIMING OF HORMONE THERAPY AND RT — A related and equally important question is whether there is any adverse impact of administering concurrent tamoxifen and radiation therapy (RT). There are two concerns: cultured breast cancer cells exposed to tamoxifen have diminished radiosensitivity, perhaps because of cell cycle arrest in the relatively radioresistant G0/G1 phases; and some clinical studies suggest the possibility of increased breast and pulmonary fibrosis with concurrent treatment At least three retrospective reports have addressed issues of efficacy and toxicity with combined tamoxifen and RT
Two cohorts were identified retrospectively because RT was allowed either before adjuvant therapy (sequential group, n = 107) or after chemotherapy but concurrent with tamoxifen (n = 202). With a median follow-up of over 10 years, DFS and OS rates were similar in both the sequential and concurrent groups, as were in-breast recurrence rates. Limited toxicity data were collected. Taken together, these data support the view that sequencing of tamoxifen and RT does not substantially affect outcomes. However, the only way this question can truly be answered is with a randomized controlled trial. ADJUNCTIVE THERAPY: DIET, EXERCISE, AND BISPHOSPHONATES — Emerging data support a benefit for adopting a low-fat diet and pursuing an active exercise program after a diagnosis of breast cancer. These issues as well as data on the benefit of bisphosphonates in the adjuvant setting, are discussed in detail elsewhere.
An acceptable alternative approach is for postmenopausal women to crossover to an AI (anastrozole, letrozole or exemestane) after completing two to three years of tamoxifen, to complete a total of five years of adjuvant endocrine therapy. It is not known if a longer duration of therapy with an AI provides further benefit. Whether tamoxifen for two or three years followed by an AI is more effective than five years of anastrozole alone is also not known.
When considering whether to add chemotherapy to hormone therapy for women with node-negative breast cancer, an important component of the decision making process is an accurate assessment of the likelihood of breast cancer relapse and mortality, and magnitude of benefit that can be expected from chemotherapy. Web-based tools (eg, Adjuvant! and Numeracy are available to help guide this discussion. Gene expression analysis on formalin-fixed, paraffin-embedded tissue (ie, the Oncotype DX assay) may be useful to stratify women with ER-positive, node-negative breast cancer into prognostic groups in order to more effectively tailor their adjuvant therapy program. However, whether the Oncotype DX assay should be used for clinical decision making at this time is controversial. These topic are discussed in detail elsewhere. Premenopausal women — The best adjuvant treatment regimen for premenopausal women with ER-positive early stage invasive breast cancer is uncertain.
Consensus recommendations — Contemporary recommendations for adjuvant treatment of early breast cancer have been guided by the International Consensus Panel on the Treatment of Primary Breast Cancer, which last updated their guidelines at the 2003 St Gallen meeting . They recommended that risk categories for women with node-negative disease be defined based upon tumor size, histologic grade, ER status and age However since these guidelines appeared, significant new information has emerged, particularly the benefit of adjuvant aromatase inhibitors and trastuzumab. Updated recommendations from the International Consensus Panel are expected in early 2006 |
50
years old with ER-positive, N+ breast cancer, combined therapy was
associated with a significant improvement in both DFS and three-year OS
(93 versus 85 percent)
49
years of age (HR 0.61, 95% CI 0.40-0.95), but not those 50 to 59 years old
(HR 0.57, 95% CI 0.32-1.01), or over the age of 60 (HR 1.21, 95% CI
0.79-1.85). These age cutoffs were chosen arbitrarily, and precise
recommendations based upon these results are difficult to make.