Adjuvant Chemotherapy: Taxanes—the "Pro" Position

I. Craig Henderson, M.D., Donald A. Berry, Ph.D., George D. Demetri, M.D., Constance T. Cirrincione, M.S., Lori J. Goldstein, M.D., Silvana Martino, D.O., James N. Ingle, M.D., M. Robert Cooper, M.D., Daniel F. Hayes, M.D., Katherine Tkaczuk, M.D., Gini Fleming, M.D., James F. Holland, M.D., David B. Duggan, M.D., John T. Carpenter, M.D., Emil Frei III, M.D., Richard L. Schilsky, M.D., William C. Wood, M.D., Hyman B. Muss, M.D., and Larry Norton, M.D.

Following publication of the first paclitaxel trials for patients with metastatic breast cancer in 1993, the taxanes were quickly incorporated into standard practice because they are less cross-resistant with anthracyclines than most other drugs (Sledge, Neuberg, Ingle, et al., 1997). The Cancer and Leukemia Group B (CALGB) and the Intergroup had planned a randomized trial evaluating the importance of doxorubicin dose in an adjuvant chemotherapy regimen, but because of the compelling nature of the paclitaxel data this trial was enlarged, and a 3x2 design was employed so that two questions could be addressed. Between May 1, 1994, and April 15, 1997, some 3,170 women with operable breast cancer and involved lymph nodes were treated with cyclophosphamide (C) (600 mg/M), and randomized to one of three doses of doxorubicin (A) (60, 75, or 90 mg/M2). Four cycles of CA were given to all patients at 3-week intervals. G-CSF was routinely given to the patients receiving the 90 mg/M2 dose. On the basis of their initial randomization, patients either discontinued all adjuvant chemotherapy after completion of CA or received 4 cycles of paclitaxel (T) (175 mg/M2).

Tamoxifen, 20 mg daily for 5 years, was offered to all patients with hormone receptor-positive (ER and/or PgR) tumors. All patients treated with breast-conserving surgery were also treated with radiotherapy to the breast. Adjuvant radiotherapy was given to patients treated with mastectomy at the discretion of the primary physician. Radiotherapy was given after all chemotherapy was completed (after 3 months for those randomized to CA alone, and after 6 months for those randomized to CA+T).

The first analysis of this trial was completed after 453 events (recurrence or death) and a median followup of 20 months (Henderson, Berry, Demetri, et al., 1998). The second analysis was completed after 624 events and a median followup of 30 months. There were no differences in the main endpoints between the first and second analysis. A third analysis will be conducted, as originally planned in the protocol design, after 900 events. An independent data safety monitoring board performed group sequential monitoring and determined the date of the first presentation of these data.

The entry characteristics of the patients were balanced among the three study arms; 62 percent of the patients were premenopausal; 66 percent had receptor-positive (ER and/or PgR) tumors; 46 percent had 1 to 3 involved axillary nodes, 42 percent had 4 to 9 involved axillary nodes, and 12 percent had 10 involved axillary nodes. Thirty percent of the patients chose breast-conserving surgery as their primary treatment.

There were no differences in outcome related to doxorubicin dose. However, the hazard reductions from adding paclitaxel to CA were 22 percent for recurrence (p = 0.0022) and 26 percent for death (p = 0.0065). At 3 years the disease-free survival rate was 73 percent after CA alone and 77 percent after CA plus paclitaxel; overall survival was 84 percent after CA alone and 87 percent after CA plus paclitaxel. There was no evidence of an interaction between doxorubicin and paclitaxel, and adding paclitaxel was equally beneficial among those initially given 60, 75, or 90 mg/M2 of doxorubicin. The benefits from adding paclitaxel were not significantly different in patient subsets defined by tumor size, number of positive lymph nodes, or menopausal status. These were the only subset analyses anticipated in the original protocol.

The most recent Oxford overview of adjuvant therapy trials suggested that chemotherapy was generally less effective among women who had hormone receptor-positive tumors and/or had been treated with adjuvant tamoxifen. Reductions in annual odds of recurrence or death from the use of chemotherapy were substantially (but not always significantly) smaller among receptor-positive or tamoxifen-treated patients regardless of patient age (EBCTCG, 1998). Although a subset analysis based on either receptor status or the use of tamoxifen was not planned in the original protocol for our trial, a post hoc analysis demonstrated a trend similar to that seen in the overview. The reduction in the hazard of recurrence was 8 percent and 32 percent, respectively, for those with receptor-positive and receptor-negative tumors. Almost all of the patients with receptor-positive tumors received adjuvant tamoxifen, and the differences in the effects of adding paclitaxel among those who did and did not receive tamoxifen is similar to the analysis of hormone receptor subsets. These differences in the effect of paclitaxel are not statistically significant, and when each subgroup is analyzed separately the effect of adding paclitaxel is not statistically significant in the receptor-positive or tamoxifen-treated groups.

When breast radiotherapy was given, it was initiated 3 months after surgery among those randomized to CA alone and 6 months after surgery among those on the CA+T arm of the trial. No significant differences in local control of disease were seen between the two arms of the study. In fact, there was a small trend towards better local control following CA+T, but the total number of local recurrences is still too small to draw firm conclusions about the effect, if any, of this delay.

The improvements in disease-free and overall survival among patients randomized to receive paclitaxel might be due to the longer duration of treatment for patients on the paclitaxel arm, the addition of a taxane that is noncross-resistant with cyclophosphamide and doxorubicin, or a combination of these two factors. Several studies that are each too small to draw independent conclusions provide support for the conclusion that the addition of taxane is an important element. In one of these trials, 524 patients were randomized to either four cycles of CAF (CA plus 5-fluorouracil) and four cycles of paclitaxel or eight cycles of CAF (Thomas, Buzdar, et al., 2000). About half of these patients received chemotherapy before surgery; 15 percent had T3 lesions, but 28 percent were node-negative. At 4 years there were 75 events. Disease-free survival was improved, and there was a 26 percent reduction in recurrences on the paclitaxel arm. However, these differences were not statistically significant.

The other study was designed for patients with large or locally advanced tumors, and all patients received primary chemotherapy with CVAP (cyclophosphamide 1000 mg/M2,vincristine 1.5 mg/M2, doxorubicin 50 mg/M2, prednisolone 40 mg daily x5) at 3-week intervals (Hutcheon, Ogston, Heys, et al., 2000). The 104 patients who had a complete or partial response after four cycles of CVAP were randomized to either another four cycles of CVAP or to four cycles of docetaxel (100 mg/M2 at 3-week intervals) before undergoing surgery. After all eight cycles of chemotherapy had been administered, there was a highly significant improvement in both the clinical and pathological response rates for those on the docetaxel treatment arm.

Conclusions

The addition of four cycles of paclitaxel after the completion of a standard course of CA substantially improves disease-free and overall survival of patients with early breast cancer. In light of more recent smaller but nonconfounded studies that demonstrated an advantage from adding four cycles of single agent taxane to four cycles of a doxorubicin-containing regimen, it is highly unlikely that all of the benefit for patients on the paclitaxel arm of this study is due simply to the longer duration of treatment in that arm of the trial. The decision to use any form of adjuvant chemotherapy should be taken thoughtfully in a patient with a receptor-positive tumor scheduled to receive adjuvant tamoxifen. However, the evidence from this trial is not now sufficient to determine which chemotherapy regimen should be selected for these patients because this was an unplanned subset analysis, and the large trend towards a greater effect in receptor-negative patients was not statistically significant. When patients receive breast radiotherapy in addition to chemotherapy, it is reasonable to wait until completion of 6 months of CA plus paclitaxel before starting the radiotherapy

References

Early Breast Cancer Trialists' Collaborative Group (EBCTCG). Polychemotherapy for early breast cancer: an overview of the randomised trials. Lancet 1998;352;930-42.

Henderson IC, Berry D, Demetri G, Cirrincione C, Goldstein L, Martino S, et al. Improved disease-free (DFS) and overall survival (OS) from the addition of sequential paclitaxel (T) but not from the escalation of doxorubicin (A) dose level in the adjuvant chemotherapy of patients (pts) with node-positive positive primary breast cancer (BC). Proc Amer Soc Clin Oncol 1998;16:390A.

Hutcheon AW, Ogston KN, Heys SD, Smith IC, Whitaker T, Miller ID, et al. Primary chemotherapy in the treatment of breast cancer: significantly enhanced clinical and pathological response with docetaxel. Proc Amer Soc Clin Oncol 2000;19:83a.

Sledge GW, Neuberg D, Ingle J, Martino S, Wood W. Phase III trial of doxorubicin (A) vs. paclitaxel (T) vs. doxorubicin + paclitaxel (A+T) as first-line therapy for metastatic breast cancer (MBC): an Intergroup trial. Proc Amer Soc Clin Oncol 1997;16:1a.

Thomas E, Buzdar A, Theriault R, Singletary S, Booser D, Valero V, et al. Role of paclitaxel in adjuvant therapy of operable breast cancer: preliminary results of prospective randomized clinical trial. Proc Amer Soc Clin Oncol 2000;19:74a.

Home · Search