Abstract P4-11-12: MammaPrint accurately predicts long-term survival (25 years) and adjuvant tamoxifen therapy benefit in lymph node negative patients

Abstract

Abstract Background: Adjuvant endocrine therapy is a standard of care for early stage estrogen receptor positive (ER+) patients, but analyses show the majority of benefit accrues after 10 years of treatment. Recent changes to the American Society of Clinical Oncology recommendations now include the option to extend tamoxifen treatment duration to 10 years. In contrast, screening trial data with 25 years of follow-up suggest that 50% of screen-detected cancers may never come to clinical attention (Miller et al BMJ 2014). Better tools are needed to identify patients who benefit from any, 5 or 10 years of adjuvant endocrine intervention. MammaPrint (MP) is an FDA-cleared gene expression signature that categorizes untreated patients as Low (LR) or High risk (HR) for distant recurrence (DR) at 10 years, independent of ER and HER2 status. Herein, we examined the performance of MP to predict long-term (25 years) survival benefit from adjuvant endocrine therapy in early stage breast cancer patients randomized after surgery to receive 2 or 5 years tamoxifen vs observation only (control arm). Methods: The Stockholm Tamoxifen Trial (STO) is a uniquely annotated trial with long-term follow-up. 733 samples were made available from the 1774 lymph node negative breast cancer patients, age <71, with a tumor size <30 mm, enrolled between November 1976 and May 1990. This group was randomized to adjuvant tamoxifen versus no adjuvant therapy (surgery alone). Long-term breast cancer specific survival was determined according to MP status (LR or HR as categorized by Agendia; or UltraLow, an additional threshold proposed in 2011 as predicting exceptionally good distant disease free survival >10 years) for all patients as well as for each of the randomized treatment arms. MP status was successfully assessed for 656 patients; 3 patients were excluded due to missing randomization data. Multivariate proportional hazards analyses (Cox) was performed, with the multivariate model adjusted for ER, PR and HER2 status (original assessment), tumor grade and size, and treatment cohort. Results: A statistically significant difference in survival by MP risk categories was seen for all patients (P–log rank=0.0013), as well as for the tamoxifen treatment arm (P–log rank=0.011) and the control arm (P–log rank=0.047). By multivariate proportional hazards (Cox) analyses and after adjustment, women with MP–HR had a statistically significant two–fold increased risk of dying from breast cancer by 25 years (Hazards ratio, HR = 1.68, CI 1.13–2.48), compared to women with MP–LR. In this STO trial, the UltraLow threshold (MP index of > 0.355) was associated with a statistically significant survival advantage across all patient groups with a 95% breast cancer specific survival at 15 years. Conclusions: MP accurately predicts for a statistically significant long–term survival benefit in MP–LR patients across all STO patients and, specifically, for those who received adjuvant tamoxifen. Separation of 'UltraLow' risk patients from the MP-LR group may facilitate clinical decisions for the duration of hormonal therapy. Additional analyses are underway to evaluate the timing of risk with and without endocrine treatment stratified by MP risk categories. Citation Format: Linda S Lindstrom, Christopher C Benz, Christina Yau, Laura J van't Veer, Carlie K Thompson, Laura J Esserman. MammaPrint accurately predicts long-term survival (25 years) and adjuvant tamoxifen therapy benefit in lymph node negative patients [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P4-11-12.