Patterns and predictors of early recurrence in postmenopausal women with estrogen receptor-positive early breast cancer.

Previous studies suggest that disease recurrence peaks at around 2 years in patients with early stage breast cancer (EBC), but provide no data regarding recurrence type. This retrospective analysis aimed to identify early recurrence types and risk factors in estrogen receptor-positive (ER+) EBC patients treated with adjuvant tamoxifen following breast cancer surgery. Postmenopausal women diagnosed with ER+ EBC from 1995 to 2004 were evaluated. Annual hazard ratios (HR) for recurrence at different sites were calculated. Time-dependent Cox regression analysis was used to identify predictors of recurrence within 2.5 years of diagnosis, including factors that were more strongly predictive of early than later recurrence. Of 3,614 patients evaluated, 476 developed recurrence during the 5-year median follow-up. Cumulative recurrence rates at 2.5 years (95% confidence interval) were: overall 6.3% (5.5-7.1), locoregional 1.1% (0.7-1.5), contralateral 0.5% (0.3-0.7), and distant 4.8% (4.0-5.6). The annual HR of overall recurrence peaked at 2 years (4.3% per annum). The majority of this peak represented distant recurrence (3.4% per annum). In Cox regression analysis, tumor size and grade, lymph node involvement, lymphovascular invasion, and symptomatic presentation were significant independent predictors of early recurrence. Age at diagnosis was independently predictive of recurrence within 2.5 years of diagnosis but not later recurrence. This study identified an early recurrence peak at 2 years, most of which were distant recurrences. Implementing an aromatase inhibitor after an initial 2-3 years of tamoxifen fails to address this early peak of distant recurrence and the potential breast cancer-associated mortality.