SABCS 2003 Adjuvant Therapy for Breast Cancer

December 3-6, 2003 San Antonio, TX Clifford Hudis, MD, Memorial Sloan Kettering Cancer Center


Clinically important data in the adjuvant setting presented at the 2003 San Antonio Breast Cancer Symposium related to the role of aromatase inhibitors, the utility of dose-density, the benefits of the taxanes, and the safety of Herceptin®.

Selective Aromatase Inhibitors

Following the initial reports of the ATAC trial demonstrating a statistically significant improvement in event-free survival among patients treated with Arimidex® in place of Nolvadex®, additional studies testing selective aromatase inhibitors are now maturing and several presentations at the SABCS focused on this issue.

An exploratory analysis presented by Dowsett, et. al suggests that the progesterone receptor may modulate the effectiveness of Arimidex®.1 Among ATAC subjects with known hormone receptor status, 71% were positive for both estrogen receptor (ER) and progesterone receptor (PR) and 19% were only positive for ER. The other subsets were too small for detailed analysis. A provocative but retrospective subset analysis suggests that the largest benefit is accrued to patients lacking PR. Although several hypotheses were offered to explain this observation, it is important to recognize that this data should not influence treatment decisions in the clinic, but rather should only be used to drive confirmatory research.

A relatively modestly sized Italian trial (ITA), at least in comparison to ATAC, was presented in which patients with node-positive hormone-receptor-positive breast cancer who had completed two to three years of Nolvadex® were randomly assigned to continue Nolvadex® for the planned five years or to switch toArimidex®.2 A significant reduction in distant recurrences was observed, although the lack of a survival difference and the small size of the trial make it premature to change routine practice.

Continuing the theme of aromatase inhibitor therapy following Nolvadex®, the MA17 trial was presented in a special session at the SABCS.3 In addition to the disease-free survival advantage benefit reported earlier in the New England Journal of Medicine , quality of life data was also provided. Consistent with ATAC, and with the ITA trial, the use of a non-steroidal selective aromatase inhibitor was again associated with improved disease-free survival and acceptable toxicities. In this study, the switch from Nolvadex® to the aromatase inhibitor (letrozole) was made after 4.5-6 years of therapy.

In context, the adjuvant aromatase inhibitor trials presented at SABCS confirm the activity seen for Arimidex® in the post-operative setting, but leave open the issue of timing. Whether post-menopausal patients with hormone receptor positive tumors should simply begin with an aromatase inhibitor (as per ATAC), switch to one after about 5 years (as per MA17), or switch at some intermediate time remains currently unresolved. Clinicians can use this data to support any of these strategies and to adjust their approach to suit the needs and treatment tolerance of individual patients.

Adjuvant chemotherapy

Two large randomized trials have demonstrated disease-free survival advantages for patients treated with paclitaxel following adjuvant doxorubicin and cyclophosphamide (AC) chemotherapy (CALGB 9344 and NSABP B28). In the neoadjuvant setting, Taxotere® has been shown to be effective after AC or a similar regimen (NSABP B27 and the Aberdeen trial), while several other studies have suggested, but failed to prove, benefits for the taxanes. A large number of randomized trials are unreported.

At an earlier ASCO meeting, the BCIRG presented the initial results of their 001 trial in which 5-FU, doxorubicin and cylcophosphamide (FAC) (using 500, 50, and 500 mg/m2 iv every 21 days x 6) was compared to the same regimen with Taxotere® 75 mg/m2 substituted for the 5-FU. Toxicity was somewhat increased, but disease-free survival was significantly improved. An interesting planned subset analysis suggested that the benefit was limited to patients with 1-3 involved nodes. At this year’s SABCS, this trial was updated with longer follow-up demonstrating that nodal status no longer mattered, but that overall survival was also significantly improved with the use of Taxotere®.4 This finding is important because it is consistent with the results of the earlier taxane-containing adjuvant trials and on principle, it confirms that early differences in disease-free survival in clinical trials can be considered an accurate surrogate for overall survival.

The next steps for adjuvant chemotherapy are not mutually exclusive, but are being tested separately. These include the use of novel schedules of administration and the addition of new agents, such as Herceptin®. Both of these efforts were addressed at this year’s meeting.

Adjuvant Herceptin®

The four large randomized trials testing Herceptin® as adjuvant therapy share several design traits. They all test one year of treatment (although the HERA trial also tests 2 years) against no therapy at all, and they mostly build on a backbone of AC followed by a taxane. The NCCTG 9831 trial uses low-dose weekly paclitaxel for 12 weeks; the paclitaxel is given alone, followed by, or concurrent with Herceptin®, which then continues for a full year of total antibody therapy. The HERA trial does not specify one chemotherapy regimen, but the Herceptin® is given for zero, one, or two years after chemotherapy is completed. The BCIG 006 trial uses AC followed by Taxotere® or the same regimen with Herceptin® concurrent with and following Taxotere®. The third arm of this trial omits the AC regimen entirely and instead uses only Paraplatin® and Taxotere® with Herceptin®. Finally, the most conventional of these trials is the NSABP B28, which tests Herceptin® with standard AC followed by standard paclitaxel. Patients on this trial receive 4 cycles of AC followed either by 4 cycles of paclitaxel alone or with Herceptin®, the latter continuing to complete a full year of therapy. In all of these trials, there is careful monitoring of cardiac function because congestive heart failure was observed in the earlier studies of Herceptin® in the metastatic setting.

At the SABCS, the NSABP presented a preliminary cardiac safety evaluation from their study.5 In their study, they obtained cardiac ejection fractions at baseline, after AC, after paclitaxel, and at 9 and 18 months after initiation of treatment. Using their definition of cardiac events, they predetermined that a 4% increase in the rate of cardiac events would represent a significant safety concern for patients treated with Herceptin®. A 3.5% increase was observed, leading them to conclude that B-28 could continue to accrue. On the other hand, this result should not eliminate concern among clinicians for the use of this agent and underscores the need to wait for guidance and assurance of safety from the ongoing trials before considering Herceptin® for routine use outside of the research setting.


Another initiative with applicability to the majority of patients with HER2 non-overexpressing breast cancer focuses on optimization of scheduling of chemotherapy. Simply stated, the hypothesis being tested is that more frequent applications of therapy, all else being equal, should be more effective than less frequent applications. An earlier trial from Milan using doxorubicin alone and a regimen of cyclophosphamide, methotrexate and 5-FU (CMF), as well as a recent study (CALGB 9741), support this hypothesis, but the safety and efficacy of the approach is not assured with all regimens.

At the SABCS, the GONO investigators presented a study using 6 cycles of FEC-60.6 Over 1,200 patients with node-positive disease were randomly assigned to standard treatment given every third week or dose-dense therapy given every second week and supported by prophylactic G-CSF. There was a non-significant trend in favor of the more frequent dosing, but the trial has to be interpreted as negative. On the other hand, the disease-free survival rate on the control arm was significantly better than anticipated and the number of events therefore reduced, thereby diminishing the power of the study. This fortunate trend in clinical trials – towards fewer than anticipated events – is an important issue as we attempt to design studies to test newer and presumably better treatments. As patients have better outcomes, clinical trials require larger numbers of patients and longer follow-up to reliably detect smaller differences in treatment effects. This is in fact a global challenge that may be best addressed through the use of neoadjuvant testing of regimens that will subsequently be applied in the post-operative setting.


The main clinical questions confronting clinicians in the adjuvant setting were addressed at the 2003 SABCS. The evolving role of the selective aromatase inhibitors was bolstered by the positive results of the MA-17 and ITA trials. The possible contribution of PR status was addressed in a retrospective study of the ATAC data, providing some leads for further study. The beneficial effects of adjuvant taxanes were confirmed again by the updated results of the BCIRG 001 study, now demonstrating a survival advantage. The need for caution in applying Herceptin® off-study in the adjuvant setting was confirmed by the NSABP B-28 results, which clearly showed an increase in cardiac events. In addition, the possible benefits and safety of accelerated dosing (dose-dense treatment) were seen in the GONO trial, but the need for adequate size was underscored by the inability to confirm a benefit in this trial. Finally, the support of erythropoietic agents in the adjuvant setting for both conventional and dose-dense regimens appear to markedly improve Hb levels and improve quality of life for patients, as well as allow optimal scheduling of chemotherapy to be delivered.


1.Dowsett M, on Behalf of the ATAC Trialists Group Royal Marsden Hospital, London, United Kingdom. Analysis of time to recurrence in the ATAC (arimidex, tamoxifen, alone or in combination) trial according to estrogen receptor and progesterone receptor status. Proceedings from the 26th Annual San Antonio Breast Cancer Symposium. December 2003. Abstract 4.

2.Boccardo F, Rubagotti A, Amoroso D, et al. Anastrozole appears to be superior to tamoxifen in women already receiving adjuvant tamoxifen treatment. Proceedings from the 26th Annual San Antonio Breast Cancer Symposium. December 2003. Abstract 3.

3.Goss PE, Ingle JN, Martino S, et al. Randomized placebo-controlled trial of letrozole in postmenopausal women with early breast cancer completing five years of tamoxifen. Proceedings from the 26th Annual San Antonio Breast Cancer Symposium. December 2003. Abstract 42.

4.Martin M, Pienkowski T, Mackey J, et al. TAC improves disease free survival and overall survival over FAC in node positive early breast cancer patients, BCIRG 001: 55 months follow-up. Proceedings from the 26th Annual San Antonio Breast Cancer Symposium. December 2003. Abstract # 43.

5.Geyer, Jr. CE, Bryant J, Romond E, et al. Cardiac safety analysis of the first stage of NSABP B-31, a randomized trial comparing the safety and efficacy of Adriamycin and cyclophosphamide (AC) followed by Taxol to that of AC followed by Taxol plus Herceptin in patients (Pts) with operable, node-positive (N+), HER-2 overexpressing breast cancer (HER2+BC). Proceedings from the 26th Annual San Antonio Breast Cancer Symposium. December 2003. Abstract 23.

6.Venturini M, Aitini E, Del Mastro L, et al. Phase III adjuvant trial comparing standard versus accelerated FEC regimen in early breast cancer patients. Results from GONO MIG1 study. Proceedings from the 26th Annual San Antonio Breast Cancer Symposium. December 2003. Abstract 12.