Home: Educational Supplement: Appendix

Preoperative Chemotherapy: NSABP Protocols B-18 and B-27

Norman Wolmark, M.D.

If the sole purpose of this review were to determine whether preoperative chemotherapy should be used as part of the standard armamentarium in the treatment of breast cancer, the answer would be straightforward if not banal. Since the data have clearly demonstrated that there is no disadvantage to the use of preoperative chemotherapy with respect to disease-free survival and survival, its use as an intervention to downstage tumors, making them amenable to breast-preserving procedures, is not likely to generate intense controversy. To view preoperative chemotherapy in this narrow context, however, belies its compelling biologic significance.

The emergence of preoperative adjuvant therapy as an intervention in primary operable Stage I and II breast cancer is a direct consequence of data generated from a series of randomized prospective clinical trials. The rationale and justification for this intervention evolved on the heels of the retreat from radical mastectomy. Only with the acceptance of breast-preserving operations and the abandonment of en bloc resection could the primacy of the operation be challenged. The tentative and seemingly heretical question could finally be addressed: Should the operation now be regarded as adjuvant treatment? In order for this hypothesis to reach the point where it could be tested in randomized prospective clinical trials, it first had to be demonstrated that adjuvant therapy when used in the traditional postoperative setting could prolong disease-free survival and survival in both stage I and II carcinoma of the breast.

NSABP B-18 was conceptually provocative in that preoperative chemotherapy was given to women with tumors that were readily amenable to treatment by more traditional means. These were not women with locally advanced breast cancer whose tumor had reached the point where it was no longer “curable” with operative intervention and in whom chemotherapy was used as “salvage” treatment. But perhaps the greatest potential of preoperative adjuvant therapy is yet to be realized. This is a unique setting—the tumor is readily accessible to surgery while the patient is undergoing other treatment. A potentially powerful tool may become available whereby molecularly characterized tumor discriminants can be correlated with the efficacy of preoperative adjuvant treatment and, more importantly, with subsequent survival. Although it is premature to suggest that objective tumor regression during the course of adjuvant therapy is a definitive bioassay that presages efficacy, the data for NSABP B-18 suggest that this is a possibility.

NSABP Protocol B-18

Between October 1988 and April 1993, 1,495 eligible patients were randomized to receive either four cycles of Adriamycin (A) and cyclophosphamide (C) preoperatively followed by surgery, or to the same chemotherapeutic regimen administered in the traditional postoperative setting. Eligibility was restricted to women with clinical Stage I and II breast cancers whose tumors were palpable. Because minimal perturbation of the tumor in the preoperative setting was thought to be essential, open biopsy was not permitted; the diagnosis was established by either fine needle aspiration or core needle biopsy. In those women undergoing breast-preserving procedures, radiotherapy was administered within 4 weeks of the operation in the preoperative arm and within 4 weeks of the completion of chemotherapy in the postoperative arm. Over a quarter of the women in this trial had tumors whose diameter was >=2 cm in diameter.

The data at six-year followup demonstrate definitively that the four cycles of preoperative chemotherapy used in this trial were effective in downstaging the size of the primary tumor and of involved regional nodes. Of the 685 women undergoing preoperative chemotherapy, 36 percent demonstrated a complete clinical response with no evidence of palpable residual tumor at the completion of chemotherapy. An additional 43 percent demonstrated a partial response to treatment, providing an overall objective response of 79 percent. Of the 36 percent of women demonstrating a complete clinical response, approximately one-third had no evidence of invasive cancer by histologic exam of the operative specimen. There also was evidence of nodal downstaging attributable to preoperative chemotherapy; 58 percent of women randomized to traditional postoperative chemotherapy had histologically positive nodes, compared with 40 percent of those patients who received preoperative chemotherapy.

The overall rate of ipsilateral breast tumor recurrence (IBTR) in those women undergoing breast-preserving operations was 7.9 percent for those who received preoperative chemotherapy, compared to 5.8 percent in those who were treated with postoperative therapy. Since there were only 66 ipsilateral breast tumor recurrences in the 954 eligible patients with breast-preserving procedures, further subset analysis of IBTR relative to patient age, tumor size, and initial response to therapy is of questionable merit.

The data are equally noteworthy for what was not demonstrated. There was no advantage for preoperative chemotherapy relative to disease-free survival and overall survival. The life table curves for the two treatments were virtually superimposable. These findings are significant when related to the specific aims of the trial. Prior to the initiation of NSABP B-18, it was postulated—on the basis of a number of kinetic models—that preoperative chemotherapy would be superior to chemotherapy given in the postoperative setting. One popular kinetic model suggested that as a result of spontaneous mutations a progressively increasing subset of drug-resistant cells developed with time, thus favoring the preoperative setting. The results of NSABP B-18 are not supportive of this hypothesis.

Perhaps the most interesting finding of NSABP B-18 is the correlation between local tumor response and subsequent outcome. Women who demonstrated a complete clinical tumor response had a better outcome than those who did not. Moreover, the most favorable outcome occurred in women who had no evidence of residual histologic invasive cancer. In a multivariate analysis in which other baseline prognostic variables were included, breast tumor response to preoperative chemotherapy continued to be a significant independent predictor of patient outcome (Fisher, Brown, Mamounas, et al., 1997; Fisher, Bryant, Wolmark, et al., 1998).

NSABP Protocol B-27

This later study was a natural extension of B-18. Patients with similar characteristics to those in B-18 are randomized to one of three treatment options, all utilizing preoperative chemotherapy. The control arm consists of the same regimen that was used in the preoperative setting of B-18, namely, four cycles of AC. Arm 2 will determine whether adding four additional preoperative cycles of sequential Taxotere following AC will result in a higher proportion of women with clinical and pathologic tumor response. Patients randomized to the third arm receive the same preoperative four cycles of AC, with the four cycles of Taxotere delayed until after the operation has been completed. The trial was initiated in December of 1995, and the required sample size of 2,400 patients is expected to be achieved by the end of this year.

B-27 will assess whether the addition of the taxane regimen will prolong disease-free survival and survival, and whether these endpoints can be correlated with the response of the primary tumor. The preoperative setting provides the opportunity to evaluate serial changes in serum and tumor biomarkers, and to correlate such changes to tumor response and patient outcome. As part of this study, formalin-fixed and paraffin-embedded core biopsy or fine needle aspiration have been collected for biomarker studies. It will be possible, using the collected materials, to evaluate the prognostic and predictive value of several biomarkers, including HER2, p53, p-glycoprotein, Ki67, and array-based CGH (Mamounas, 1997).

Conclusions

If the correlation between tumor response and outcome observed in B-18 can be verified and extended in B-27, a rapid evaluation of candidates for therapeutic intervention could become available. Although it is tempting to speculate on the potential biologic contributions of the preoperative setting, the practical charge of the consensus committee must be addressed: Should preoperative chemotherapy be part of the therapeutic standard? The data support the use of adjuvant preoperative chemotherapy in settings where the downstaging of the primary tumor is desirable, and in this context it should be an accepted standard of care. Updated results of B-18 and results from other preoperative chemotherapy trials will be reviewed during this conference. Although the information generated so far is unlikely to represent a threat to the domain of the surgeon, the primacy of the operative procedure has been significantly challenged.

References

Fisher B, Brown A, Mamounas E, Wieand S, Robidoux A, Margolese RG, et al. Effect of preoperative chemotherapy on local-regional disease in women with operable breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-18. J Clin Oncol 1997; 15:2483–93. Abstract.

Fisher B, Bryant J, Wolmark N, Mamounas E, Brown A, Fisher ER, et al. Effect of preoperative chemotherapy on the outcome of women with operable breast cancer. J Clin Oncol 1998; 16:2672–85. Abstract.

Mamounas EP. NSABP Protocol B-27. Preoperative doxorubicin plus cyclophosphamide followed by preoperative or postoperative docetaxel. Oncology 1997; 11:37–40. Abstract.