Method of detection of local recurrence in patients following breast conserving surgery and its utility for surveillance
611 Background: It is estimated that approximately 10% of all breast cancer patients will develop local recurrence (LR) at 10 years. Routine surveillance for detection of early breast cancer recurrence is widely performed despite lack of conclusive evidence for an improvement in patient quality of life or potential for cure. A number of historical studies evaluating the effectiveness of routine screening for LR following treatment for early-stage breast cancer have suggested that the diagnosis of LR is more frequent during a routine visit and occurring in asymptomatic patients. However, differentiating the method of detection is not often elucidated. In this study we evaluated the manner in which patients presented with an isolated LR in clinical practice. Methods: Our routine patient follow-up consists of physical exam and mammogram every 6 mos for the first 2 years following breast conserving surgery (BCS) and yearly thereafter. We queried our prospectively collected breast cancer database (1632 patients from July 1986 - July 2004) for patients with an isolated LR following BCS (n=59 (3.6%); two patients had bilateral LRs). Medical records were not available for three patients. Results: At a median follow-up of 45 mos (range: 5–122 mos) there were 58 evaluable LRs: 15 DCIS, 31 infiltrating ductal carcinoma (IDC), 6 infiltrating lobular (ILC), 2 mixed IDC/ILC, 3 invasive cancers NOS and 1 unknown. Patient presentation was as follows: 25 were diagnosed by self-exam, 28 on screening mammogram, 2 were diagnosed by physician (includes one referral), and 3 unknown. Mammogram detected recurrences were more frequent among patients with DCIS whereas self-detected recurrences were more common in patients with IDC (79% vs 33% and 21% vs 67%, respectively; P<0.2). Mean tumor size was larger in self-presentation (2.1 cm; range: 0.8–4.5 cm) than in mammogram detected group (1.6 cm; range 0.4–6 cm). Conclusions: These findings demonstrate the value of mammography as compared to patient detected LRs. Whether a survival advantage exists remains uncertain. Nevertheless routine physician examination in this setting is highly insensitive and its further utility must be considered when devising cost-effective strategies for surveillance of breast cancer patients. No significant financial relationships to disclose.