Abstract P6-06-43: Clinicopathological features of breast cancer patients with late recurrences

Abstract

Abstract [Background] With advances in early detection and improvements in treatments, long-term breast cancer survivors who had risk of late recurrence are increasing. Recent trials suggested 10 years of endocrine treatment produces a further reduction in recurrence and mortality rather than 5 years treatment. However, it remains to be elucidated which patients would have clinical benefit from longer endocrine treatment. We investigated clinicopathological features of breast cancer patients with late recurrences. [Patients and Methods] From 1989 to 2003, 1051 patients with primary breast cancer were treated at Keio University Hospital and classified into three groups: early recurrences (within 5 years), late recurrences (after 5 years) and no recurrences. Clinicopathological features of breast cancer patients with early and late recurrences were compared. Differences in frequency distribution were compared by the chi-squared test. Smoothed hazards estimates were used for estimation of recurrence rates depending on time. [Results] Among 1051 patients, 193 (18.4%) had early recurrences and 113 (10.8%) had late recurrences. Of all 306 recurrent patients, primary recurrent site was lung (24.2%), liver (8.8%), bone (20.0%), brain (1.3%), loco-regional (34.6%), distant lymph node (5.6%) and others (5.6%). Compared to patients with early recurrences, patients with late recurrences showed higher rate of positive hormone receptor (HR) status (58.0% vs. 81.4%, p<0.001), lower rate of pathological nodal involvement (68.4% vs. 54.0%, p = 0.016), lower rate of lymphovascular invasion (LVI) (81.3% vs. 65.5%, p = 0.003) and smaller tumor size (2.99 cm vs. 2.46 cm, p = 0.005). For node-positive patients, the hazard of estimates was similar between the HR-positive and HR-negative groups beyond 5 years. For node-negative patients, however, the hazard of HR-positive patients increased consistently, crossed with the hazard of HR-negative patients between 5 and 6 years and was higher than the hazard of HR-negative patients beyond 6 years. Among 204 HR-positive recurrent patients, 112 (54.9%) had early recurrences and 92 (45.1%) had late recurrences. Compared to HR-positive patients with early recurrences, HR-positive patients with late recurrences had lower rate of pathological nodal involvement (71.4% vs. 55.4%, p = 0.026) and lower rate of LVI (81.3% vs. 66.3%, p = 0.023), however, tumor size was not significantly correlated with late recurrences of HR-positive patients. On the other hand, no clinicopathological factors including pathological nodal status, LVI and tumor size were significantly correlated with late recurrences among 102 HR-negative recurrent patients. Among 193 patients with early recurrences, HR-positive patients had significantly longer post-recurrence survival than HR-negative patients (99.1 months vs. 60.0 months, p<0.001). However, among 113 patients with late recurrences, post-recurrence survival of HR-positive patients was not significantly different from HR-negative patients (233.1 months vs. 231.2 months, p = 0.797). [Conclusions] These data suggested breast cancer patients with positive HR and negative nodal status had a high likelihood of developing late recurrences and should be treated with longer endocrine treatment. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P6-06-43.