• Paul E Goss, James N Ingle, Silvana Martino, Nicholas J Robert, Hyman B Muss, Martine J Piccart, Monica Castiglione, Dongsheng Tu, Lois E Shepherd, Kathleen I Pritchard, Robert B Livingston, Nancy E Davidson, Larry Norton, Edith A Perez, Jeffrey S Abrams, David A Cameron, Michael J Palmer, Joseph L Pater, and National Cancer Institute of Canada Clinical Trials Group MA.17.
• Division of Hematology and Oncology, Massachusetts General Hospital, Boston, MA, USA. pgoss@partners.org
• J. Clin. Oncol. 2007 May 20; 25 (15): 2006-11.

PurposeControversy exists regarding estrogen (ER) and progesterone (PgR) receptor expression on efficacy of adjuvant endocrine therapy. In the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial, the benefit of anastrozole over tamoxifen was substantially greater in ER+/PgR-than ER+/PgR+ tumors. In BIG 1-98 (Breast International Group), the benefits of letrozole over tamoxifen were the same in ER+ tumors irrespective of PgR. MA.17 randomized postmenopausal women after 5 years of tamoxifen, to letrozole or placebo. We present outcomes according to tumor receptor status.Patients And MethodsDisease-free survival (DFS) and other outcomes were assessed in subgroups by ER and PgR status using Cox's proportional hazards model, adjusting for nodal status and prior adjuvant chemotherapy.ResultsThe DFS hazard ratio (HR) for letrozole versus placebo in ER+/PgR+ tumors (N = 3,809) was 0.49 (95% CI, 0.36 to 0.67) versus 1.21 (95% CI, 0.63 to 2.34) in ER+/PgR-tumors (n = 636). ER+/PgR+ letrozole patients experienced significant benefit in distant DFS (DDFS; HR = 0.53; 95% CI, 0.35 to 0.80) and overall survival (OS; HR = 0.58; 95% CI, 0.37 to 0.90). A statistically significant difference in treatment effect between ER+/PgR+ and ER+/PgR-subgroups for DFS was observed (P = .02), but not for DDFS (P = .06) or OS (P = .09).ConclusionThese results suggest greater benefit for letrozole in DFS, DDFS, and OS in patients with ER+/PgR+ tumors, implying greater activity of letrozole in tumors with a functional ER. However, because this is a subset analysis and receptors were not measured centrally, we caution against using these results for clinical decision making.

31 keywords...

hide…