Abstract
BACKGROUND: Although subsequent breast cancer risk after primary lobular carcinoma in situ (LCIS) has been studied intensively, whether the risk of second breast cancer after first LCIS varies with hormone receptor (HR) status of primary tumor remains unclear. METHODS: We identified 10,304 women with primary pure unilateral LCIS between 1998 and 2007 from the Surveillance, Epidemiology and End Results (SEER) 18 Registries. Kaplan-Meier estimates of 5 or 10-year probabilities of second ipsilateral breast cancers (IBCs) and contralateral breast cancers (CBCs) were calculated. Multivariable Cox proportional model was performed to identify impact of HR status of primary LCIS, and other demographic, clinicopathologic or treatment characteristics on risk of second IBCs or CBCs. RESULTS: Of the 10,304 women with primary LCIS included in this study, 9949 (96.5%) patients had HR+ tumors, and 355 (3.5%) had HR- tumors. Multivariable-adjusted analyses showed that although there was no difference in risk of total second IBCs between women with HR+ and HR- LCIS (P = 0.152), patients with HR+ LCIS had a statistically lower risk of second invasive IBCs compared to those with HR- LCIS (hazard ratio 0.356, 95% CI 0.141-0.899, P = 0.029). Women with primary HR+ LCIS had lower risks of both second total and invasive CBCs compared to those with HR- LCIS (total CBCs: hazard ratio 0.340, 95% CI 0.228-0.509, P<0.001; invasive CBCs: hazard ratio 0.172, 95% CI 0.108-0.274, P<0.001). Additionally, black women had a 2-fold risk of developing subsequent total IBCs than white women (P = 0.028). CONCLUSIONS: This population-based study demonstrated that the risk of second breast cancers was significantly increased in women with HR- first LCIS compared to those with HR+ LCIS. These findings warrant intensive surveillance for second breast cancers in HR- LCIS survivors.