Taxane Plus Doxorubicin Not Well Tolerated in Early Breast Cancer

PHILADELPHIA, PENNSYLVANIA-Adjuvant treatment withAdriamycin/Taxotere (AT) did not improveoutcome over standard treatmentwith doxorubicin/cyclophosphamide(AC) for women with early-stagebreast cancer, Lori J. Goldstein, MD,of Fox Chase Cancer Center reported(abstract 512). Both regimens testedin a phase III Eastern Cooperative OncologyGroup trial (E2197) producedbetter-than-expected results, but toxicitywas higher in the AT group.Neutropenia associated with feverand infection occurred in 28% of theAT group and in 10% of the AC group.There were 18 cases of congestive heartfailure in the AT group and 10 cases inthe AC group, but this was not consideredstatistically significant.Disease-free and overall survivalcurves for the two regimens overlapped.Subgroup AnalysisThe trial opened in July 1998 and"closed rapidly due to great accrual"in January 2000, Dr. Goldstein said.Patients were balanced for age, hormonereceptor status, menopausal status,nodes, surgery, as well as gradeand size of the tumor. The median agewas 51 years (range, 24-85 years). Sixty-four percent of patients were estrogenreceptor positive (ER+) and 65%of tumors were lymph node negative."The ER/PR subgroups were prespecifiedat randomization designedto balance the treatment arms," Dr.Goldstein noted.Patients were randomized to receiveAT (doxorubicin 60 mg/m² plusdocetaxel 60 mg/m²) with ciprofloxacin(Cipro) support or AC (doxorubicin60 mg/m² plus cyclophosphamide600 mg/m²) every 3 weeks for fourcycles. Tamoxifen was given for 5 yearspostchemotherapy to all patients withER+ and/or PR+ tumors.Subgroup analysis found no significanteffect for nodes, tumor size, age,menopausal status, grade, type of surgery,or race. Nor were there any significantinteractions between the treatmentand the subgroups.Data for disease-free survival "suggestthat the PR-negative tumors maypotentially benefit from AT, but thisstudy was not powered to statisticallydetect these differences," Dr. Goldsteinreported. The pattern was thesame for overall survival, she said, andsupports the hypothesis "that in ER+tumors, the large benefit presented bytamoxifen overwhelms the potentialbenefit of taxane therapy... The prognosisof these tumors is so good, it isdifficult to detect a difference betweenthese two chemotherapy arms. Takentogether, along with the genomichealth data, these data support thehypothesis that the biology of a tumormay predict the benefit to individualtherapies."The Case for Taxanes"Given their single-agent activity,relative non-cross-resistance, partiallynonoverlapping toxicities, and differentmechanisms of action, there wasa clear rationale for combining taxaneswith doxorubicin," Dr. Goldsteinexplained. The selection of AT was"based on the relative cardiac safety ofthis combination," and several ran-domized trials supported such investigation.In a discussion of the trial, GabrielN. Hortobagyi, MD, Director of theBreast Cancer Research Program atM. D. Anderson Cancer Center inHouston, referred to several trials thathave demonstrated a clear advantageto using taxanes in the adjuvant settingfor high-risk breast cancer. "Theaddition of a taxane clearly improvesevent-free survival, relapse-free survival,and probably overall survival inpatients with high-risk primary breastcancer," he observed. "Whether thiscan be readily translated to a lower riskpopulation awaits additional trials butis probably likely."Why was E2197 a negative study?Dr. Hortobagyi offered several possibilities,including insufficient durationof therapy and inadequate dose ofdocetaxel. He also speculated onwhether the sequential use of anthracyclinesand taxanes might be betterthan simultaneous use of these agents."AT x 4 remains unproven, moretoxic than AC x 4, unlikely to matchTAC (Taxotere [docetaxel], Adriamycin[doxorubicin], and cyclophosphamide)x 6 or any of the other regimens.Therefore, I would propose that itshould not be used outside of a clinicaltrial and probably not even in a clinicaltrial since there are more effectiveand probably better tolerated regimens,"he said."On the basis of our increased biologicalknowledge and our better understandingthat there are biologicmethods to select differences in subsetsof different patients, we shouldconcentrate on how to identify theoptimal regimens for individual tumorsand stop getting involved in theselarge trials in which one size is supposedto fit all patients."