Calculation of outcome estimates for synchronous bilateral breast cancer patients using Adjuvant!

Abstract

561 Background: The recognition of synchronous breast cancers is increasing because of improving screening and pretreatment imaging. For synchronous breast cancers, many clinicians choose adjuvant therapies by considering only the high-risk cancer, but cohorts of synchronous bilateral cases have worse outcomes than cohorts of unilateral cases. Adjuvant! outcome estimates for unilateral breast cancers have been validated by the Breast Cancer Outcomes Unit (BCOU) (Olivotto et al. JCO 2005;23:2716–25). Synchronous bilateral cases from the BCOU were used to investigate our hypothesis that synchronous breast cancer outcomes can be calculated by treating events from the two cancers as independent. Methods: All women < 75 years treated in British Columbia from 1989 to 1995 with new synchronous (< 2 months) bilateral breast cancers with sufficient staging information for Adjuvant! outcome estimates (pT1–3, pN0–3, M0, ER, grade, size, number of positive nodes, LVI, and treatment) were eligible. Our sample was 50 women of median age 65 (range 42–75). Researchers blinded to the clinical outcomes entered clinical and treatment (tamoxifen, type of chemotherapy) variables into Adjuvant! Online Version 8.0 with default comorbidity (Minor Problems) to obtain outcome estimates for each cancer for each patient. Using the unilateral estimates, bilateral estimates of overall survival (OS), breast cancer-specific survival (BCSS) and relapse-free survival (RFS) were calculated. Bilateral OS = Low Risk OS * High Risk OS / (100 - High Risk Die of Cancer). Bilateral BCSS = (100 - High Risk Die of Cancer - Bilateral OS). The bilateral RFS calculation accounted for contralateral recurrences in Adjuvant!. The mean for all patients was compared to the 10-year cohort outcomes. The study is descriptive. Results: Adjuvant! Low Risk breast cancer estimates were OS 79.2%, BCSS 92.0%, and RFS 81.9%. Adjuvant! High Risk breast cancer estimates were OS 64.8%, BCSS 75.8%, and RFS 63.7%. Clinical outcomes of the bilateral patient cohort were OS 58.0%, BCSS 72.0%, and RFS 62.0%. BCOU synchronous bilateral estimates were OS 58.3%, BCSS 69.5%, and RFS 62.2%. Conclusions: Our calculations of bilateral outcomes were better estimates of clinical patient outcomes than the commonly used unilateral High Risk estimates. No significant financial relationships to disclose.