Benefit from exemestane as extended adjuvant therapy after 5 years of adjuvant tamoxifen: Intention-to-treat analysis of the national surgical adjuvant breast and bowel project B-33 trial

Eleftherios P. Mamounas, Jong Hyeon Jeong, D. Lawrence Wickerham, Roy E. Smith, Patricia A. Ganz, Stephanie R. Land, Andrea Eisen, Louis Fehrenbacher, William B. Farrar, James N. Atkins, Eduardo R. Pajon, Victor G. Vogel, Joan F. Kroener, Laura F. Hutchins, André Robidoux, James L. Hoehn, James N. Ingle, Charles E. Geyer, Joseph P. Costantino, Norman Wolmark

Research output: Contribution to journalArticle

224 Scopus citations


Purpose: Patients with early-stage, hormone receptor-positive breast cancer have considerable residual risk for recurrence after completing 5 years of adjuvant tamoxifen. In May 2001, the National Surgical Adjuvant Breast and Bowel Project (NSABP) initiated accrual to a randomized, placebo-controlled, double-blind clinical trial to evaluate the steroidal aromatase inhibitor exemestane as extended adjuvant therapy in this setting. Patients and Methods: Postmenopausal patients with clinical T1-3N1M0 breast cancer who were disease free after 5 years of tamoxifen were randomly assigned to 5 years of exemestane (25 mg/d orally) or 5 years of placebo. Our primary aim was to test whether exemestane prolongs disease-free survival (DFS). In October 2003, results of National Cancer Institute of Canada (NCIC) MA.17 showing benefit from adjuvant letrozole in this setting necessitated termination of accrual to B-33, unblinding, and offering of exemestane to patients in the placebo group. Results At the time of unblinding, 1,598 patients had been randomly assigned; 72% in the exemestane group continued on exemestane and 44% in the placebo group elected to receive exemestane. With 30 months of median follow-up, original exemestane assignment resulted in a borderline statistically significant improvement in 4-year DFS (91% v 89%; relative risk [RR] = 0.68; P = .07) and in a statistically significant improvement in 4-year relapse-free survival (RFS; 96% v 94%; RR = 0.44; P = .004). Toxicity, assessed up to time of unblinding, was acceptable for the adjuvant setting. Conclusion: Despite premature closure and crossover to exemestane by a substantial proportion of patients, original exemestane assignment resulted in non-statistically significant improvement in DFS and in statistically significant improvement in RFS.

Original languageEnglish (US)
Pages (from-to)1965-1971
Number of pages7
JournalJournal of Clinical Oncology
Issue number12
Publication statusPublished - 2008


ASJC Scopus subject areas

  • Cancer Research
  • Oncology
  • Medicine(all)

Cite this