Protocol B-27, conducted by the National Surgical Adjuvant Breast and Bowel Project (NSABP), is a phase III, randomized trial designed to evaluate whether sequencing docetaxel (Taxotere) to neoadjuvant doxorubicin/cyclophosphamide (Cytoxan, Neosar) prolongs disease-free and overall survival in patients with operable breast cancer.
ABSTRACT: Protocol B-27, conducted by the National Surgical Adjuvant Breast and Bowel Project (NSABP), is a phase III, randomized trial designed to evaluate whether sequencing docetaxel (Taxotere) to neoadjuvant doxorubicin/cyclophosphamide (Cytoxan, Neosar) prolongs disease-free and overall survival in patients with operable breast cancer. Patients are being randomized into three groups. The control group receives four 21-day courses of doxorubicin/cyclophosphamide chemotherapy with tamoxifen (Nolvadex), followed by breast surgery (and postoperative radiation for patients receiving breast-conserving surgery). Two experimental groups receive the same doxorubicin/cyclophosphamide chemotherapy and tamoxifen, followed by docetaxel--either before (preoperative group) or after (postoperative group) surgery. In the first 11 months of the study, 283 patients--of a projected 1,606 patients over a 5-year period--have been entered. Slightly more than half of the patients are younger than 50 years of age. Nearly half of the patients presented with tumors that were more than 4.0 cm in greatest diameter. Biopsy was performed by fine-needle aspiration slightly more than half of the time. Slightly more than two-thirds of the patients had clinically negative nodes. Lumpectomy was the proposed surgery at entry in 40% to 43% of the patients. As of November 1996, toxicity information is available on 29 patients in the preoperative docetaxel group and 23 patients in the postoperative docetaxel group. So far, there have been no unexpected toxicities, but the data are too preliminary to report in detail. [ONCOLOGY 11(Suppl 6):37-40, 1997]
During the last quarter of the 20th century, significant changes haveoccurred in our understanding of the biology and management of operablebreast cancer. Results from well-designed, well-conducted clinical trialshave convincingly demonstrated that the extent of surgical resection isnot paramount for patient outcome. Furthermore, the administration ofsystemic therapy following surgery has been shown to significantly improveboth disease-free and overall survival in such patients.
There has been a shift in emphasis away from surgery as the sole treatmentfor operable breast cancer, and the use of systemic therapy has been adoptedas an integral part of treatment in the majority of cases. With these developments,the question soon arose as to whether variations in the timing of systemictherapy administration might further influence the outcome of patientswith breast cancer. Several biologic and clinical observations contributedsupport for the notion of evaluating such variations in timing of systemictherapy.
In 1988, the National Surgical Adjuvant Breast and Bowel Project (NSABP)initiated a randomized trial (B-18) in patients with operable breast cancerto compare preoperative vs postoperative administration of adjuvant chemotherapy.Following diagnosis of breast cancer by fine-needle aspiration or tru-cutbiopsy, patients were randomized to receive either surgery (lumpectomyand axillary node dissection or modified radical mastectomy), followedby four courses of doxorubicin/cyclophosphamide (Cytoxan, Neosar) chemotherapyevery 21 days, or the same chemotherapy first followed by surgery.
All patients 50 years of age or older were also given tamoxifen (Nolvadex)10 mg, twice daily, for 5 years, starting after the completion of adjuvantchemotherapy. Patients undergoing lumpectomy also received postoperativeradiotherapy. Preoperative chemotherapy administration resulted in a highrate of clinical response (79%) and a low rate of complete pathologic tumorresponse (about 11%). In addition, there was evidence of axillary nodaldownstaging and an increase in the rate of lumpectomy.[3,4]
The development of taxoids and the demonstration of their significantantitumor activity in patients with advanced breast cancer provide therationale for investigating docetaxel in the neoadjuvant setting. In 1995,the NSABP implemented protocol B-27, a randomized trial designed to evaluatethe safety and efficacy of docetaxel administered in the preoperative orthe postoperative setting, following preoperative doxorubicin/cyclophosphamidechemotherapy.
The primary objective of the study is to determine whether the additionof four courses of preoperative or postoperative docetaxel, following fourcourses of preoperative doxorubicin/cyclophosphamide, can more effectivelyprolong disease-free and overall survival in patients with operable breastcancer than four courses of preoperative doxorubicin/cyclophosphamide alone.
Secondary objectives are to determine whether the additional preoperativedocetaxel following preoperative doxorubicin/cyclophosphamide: (1) increasesthe rates of locoregional clinical response and pathologic complete response,(2) contributes to further pathologic nodal downstaging, and (3) increasesthe number of patients who receive lumpectomy.
Additional secondary objectives are to determine whether the additionof postoperative docetaxel following preoperative doxorubicin/cyclophosphamideincreases disease-free and overall survival in specific subgroups of patients;ie, as defined by nodal status.
Following stratification according to age, clinical tumor size, andclinical nodal status, patients are randomized to receive one of threeregimens: (1) four courses of preoperative doxorubicin/cyclophosphamidechemotherapy followed by surgery (group 1); (2) four courses of preoperativedoxorubicin/cyclophosphamide, followed by four courses of preoperativedocetaxel, followed by surgery (group 2); or (3) four courses of preoperativedoxorubicin/cyclophosphamide, followed by surgery, followed by four coursesof postoperative docetaxel (group 3). All patients receive tamoxifen dailybeginning on day 1 of the first doxorubicin/cyclophosphamide course andcontinuing for 5 years (Figure 1).
The eligibility criteria for this study are as follows:
Patients with no clinically palpable nodes are eligible only when thegreatest diameter of the tumor exceeds 1 cm. Patients with clinically palpablenodes are eligible regardless of tumor size.
All three treatment groups receive four 21-day courses of 60 mg/m²of doxorubicin plus 600 mg/m² of cyclophosphamide and oral tamoxifen,10 mg, twice daily, beginning on day 1 of the first course for a planned5-year treatment duration. The type of surgery is left to the discretionof the surgeon, but surgeons are asked to state prior to randomizationtheir intended procedure if a patient was not to receive preoperative chemotherapy.
Patients who undergo a lumpectomy also receive postoperative breastirradiation. Patients in group 2 receive, in addition, four 21-day coursesof 100 mg/m² of docetaxel given intravenously after four courses ofthe doxorubicin/cyclophosphamide regimen and before surgery. Patients ingroup 3 receive, in addition, four 21-day courses of 100 mg/m²of intravenousdocetaxel after surgery.
All patients in groups 2 and 3 receive the following premedicationsduring each docetaxel course: 20 mg of oral dexamethasone, 12 and 6 hoursbefore treatment; 50 mg of diphenhydramine intravenously, 1 hour beforetreatment; and either 300 mg of cimetidine or 50 of mg ranitidine intravenously,also 1 hour before treatment.
Assessments of Tumor Response and Statistical Considerations
The criteria for assessing tumor response in this trial are those usedin the previous NSABP protocol (B-18). Tumor measurements are taken aftereach cycle of chemotherapy and before surgery. Response is assessed whenchemotherapy (preoperative doxorubicin/cyclophosphamide alone or followedby preoperative docetaxel) has been completed, prior to surgery.
This study is scheduled to accrue an average of 318 patients per yearover 5 years, for a total of 1,606 patients. Survival and disease-freesurvival are the primary end points. With this number of patients, at 1.5years of additional follow-up, a 33% improvement in disease-free survivalcan be detected, with a power of 0.81. After 3 years of additional follow-up,a 40% reduction in mortality can be detected, with a power of 0.80.
Patient Accrual and Patient Characteristics
The trial opened in December 1995. As of November 1996, approximately11 months into the trial, 283 patients were entered. Patient and clinicalcharacteristics at study entry for patients accrued to date are shown inTable 1. Slightly more than half the patientsare younger than 50 years of age. Nearly half (47% to 48%) of the tumorsare more than 4.0 cm in greatest diameter. Biopsy was performed by fine-needleaspiration in a little more than half of the patients. Slightly more thantwo-thirds of the patients had clinically negative nodes. Lumpectomy wasthe proposed surgery at entry for 40% to 43% of patients. The distributionof all these characteristics was similar across treatment groups.
The results for toxicity are preliminary: as of November 1996, toxicityinformation is available on 167 patients who received doxorubicin/cyclophosphamidepreoperatively (all three groups combined), 29 patients who received docetaxelpreoperatively, and 23 patients who received docetaxel postoperatively.So far there have been no unexpected toxicities, but the data are too preliminaryto report in detail. Toxicity information is likely to change as more patientscomplete all of the courses of their assigned chemotherapy.
NSABP B-27 is designed to assess the effect of adding docetaxel to theneoadjuvant chemotherapy regimen of doxorubicin/cyclophosphamide. In thefirst 11 months of the study, 283 patients--of a projected 1,606 patientsover a 5-year period--have been entered. Average accrual during the past6 months has exceeded the projected average accrual. This preliminary reportshows no unexpected toxicities. Thus far, the incorporation of docetaxelin the manner described with doxorubicin/cyclophosphamide in the neoadjuvant/adjuvantsetting appears to be safe and feasible.
In contrast to the common practice of 5 days or less of premedicationwith docetaxel, this study mandates only 1 day of corticosteroids andantihistamines. The 1-day regimen usually recommended with paclitaxel (Taxol)was chosen for this study because the number of courses of docetaxel inthis protocol is limited to four. The development of clinically significantedema requiring treatment discontinuation is considered unlikely. For thisreason, and because this study may serve as a model for subsequent adjuvantstudies, a shorter course of steroid premedication is used.
In the absence of the recommended steroid premedication regimen, andassuming a maximum cumulative docetaxel dose of 400 mg/m², some degreeof edema can be expected in about half of the patients. These patientsare carefully monitored. Should edema become a problem during the courseof the trial, the premedication regimen may be amended as necessary.
Advantages of Preoperative Chemotherapy
The role of preoperative chemotherapy in the treatment of operable breastcancer is currently evolving. Until data from randomized clinical trialsdemonstrate that preoperative chemotherapy results in disease-free andoverall survival rates that are superior, or equivalent, to those achievedwith postoperative chemotherapy, preoperative chemotherapy may be consideredonly in women with tumors too large for breast-conserving surgery.
There may be an additional advantage to preoperative chemotherapy overpostoperative chemotherapy, even if these two treatments result in thesame disease-free and overall survival rates. This advantage is providedthat a correlation between clinical and/or pathologic response to chemotherapyand outcome, can be convincingly demonstrated. If this is the case, clinicaland pathologic tumor response to preoperative chemotherapy can be usedas a prognostic factor for outcome and as a guide for further locoregionaland systemic therapy.
From a research standpoint, there is ample rationale for continuingto evaluate the role of preoperative chemotherapy in patients with operablebreast cancer. If response to preoperative chemotherapy correlates withpatient outcome, response to chemotherapy can then be used as an intermediateend point in testing new chemotherapeutic regimens or new drugs administeredafter well-established regimens.
This approach offers an additional clinical advantage with strong biologicimplications. For example, it is conceivable that subgroups of patientswith a high likelihood of pathologic complete tumor response could be identifiedthrough the evaluation of proven and putative prognostic tumor markers(eg, estrogen receptors, progesterone receptors, ploidy and S-phase, erb-B2and p53 oncogenes, and other tumor oncogenes and growth factors). The potentialcorrelation of such markers, individually or in combination with tumorresponse to preoperative chemotherapy and outcome, could ultimately sparesome patients from radiation or surgical resection.
Furthermore, serially monitoring tumor marker changes while a patientis undergoing preoperative chemotherapy may provide biologic insight intothe nature and function of these tumor markers. Knowledge may also be obtainedregarding the mechanisms of action of new chemotherapeutic agents or newtreatment modalities.
Two ancillary trials to the B-27 protocol were recently implementedto study serum and tumor biomarkers as they relate to outcome and responseto preoperative doxorubicin/cyclophosphamide or docetaxel chemotherapy.The first (NSABP Protocol B-27.1) is designed to evaluate the usefulnessof serum erb-B2 extracellular domain and serum erb-B2 antibodies for predictingresponse to preoperative chemotherapy and long-term outcome. In addition,by obtaining serum at specified times (before administration and aftercompletion of preoperative chemotherapy, after surgery, 1 year after randomization,and at the time of recurrence), the study will determine whether potentialchanges in the levels of erb-B2 extracellular domain and erb-B2 antibodiesare induced by chemotherapy or are associated with breast cancer recurrence.
The second trial (NSABP Protocol B-27.2) is designed to evaluate theusefulness of tumor biomarkers obtained by fine-needle aspiration or corebiopsy in predicting response to preoperative chemotherapy and long-termoutcome in B-27 patients. The study will also assess whether preoperativechemotherapy results in changes in tumor biomarker expression, and whetherthese changes can be correlated with tumor response and long-term outcome.
Finally, if there is prolonged disease-free or overall survival withthe addition of preoperative or postoperative docetaxel after preoperativedoxorubicin/cyclophosphamide, the study will examine whether this prolongationis associated with the expression of certain biomarkers. The followingbiomarkers will be evaluated both in material obtained before preoperativechemotherapy and in material obtained at the time of surgery: nuclear grade,estrogen and progesterone receptors, proliferation markers, p53 oncogenemutations, erb-B2 overexpression, P-glycoprotein, and apoptosis markers(bcl-2).
1. Early Breast Cancer Trialists' Collaborative Group: Effects of radiotherapyand surgery in early breast cancer: An overview of the randomized trials.N Engl J Med 333:1444-1455,1995.
2. Early Breast Cancer Trialists' Collaborative Group: Systemic treatmentof early breast cancer by hormonal, cytotoxic, or immune therapy: 33 randomisedtrials involving 31,000 recurrences and 24,000 deaths among 75,000 women.Lancet. 339:1-15, 71-85, 1992.
3. Fisher B, Rockette H, Robidoux A, et al: Effect of preoperative therapyfor breast cancer (BC) on local-regional disease: First report of NSABPB-18 (abstract 57). Proc Am Soc Clin Oncol 13:64, 1994.
4. Mamounas E, Fisher B, Rockette H, et al: Clinical and pathologiclocal-regional response of operable breast cancer (BC) to preoperativechemotherapy: Results from NSABP B-18 (abstract). Society of Surgical OncologyMeeting; Atlanta, GA, March 20-24, 1996.
5. Data on file. Rhone-Poulenc Rorer Pharmaceuticals, Inc. Collegeville,PA.
6. Bristol-Myers Squibb Oncology. Taxol (paclitaxel) for injection concentrateprescribing information, in Physicians' Desk Reference, 50th ed, pp 714-717.Montvale, New Jersey, Medical Economics, 1996.
7. Fisher B, Mamounas EP: Preoperative chemotherapy: A model for studyingthe biology and therapy of primary breast cancer. J Clin Oncol 13:537-540,1995.