Abstract P6-11-02: Patient characteristics and outcomes for patients with HER2-overexpressing breast cancer with brain metastases undergoing radiation therapy in the pre- and post adjuvant trastuzumab era

Abstract

Abstract Background: Trastuzumab (H) has improved outcomes for patients with HER2 overexpressing (HER2+) breast cancer (BC) and thus increased the at-risk period for development of brain metastases (BM). In this study we describe the characteristics and outcomes of patients with HER2+BCBM who underwent stereotactic radiosurgery (SRS) or whole brain radiation (WBRT) as initial BM treatment (tx) in the pre- and the post adjuvant (adj) H era at MSKCC. Methods: 100 consecutive pts with HER2+ BCBM who received SRS or WBRT from January 2001 to December 2011 were identified. Clinical, pathologic and tx information were obtained by retrospective review. Pt characteristics at the time of BM diagnosis (dx) and their associations with time from BM to death were evaluated by Kaplan-Meier (KM) curves, log-rank tests, and Cox proportional hazard models. Results: The median age at BM dx was 54 yrs (range 26-79). 31% and 69% of pts received SRS and WBRT respectively. After the BM dx, 97% continued to receive systemic tx consisting of chemotherapy and/or anti-HER2 tx. 97% of pts received H for MBC and 17% received H in the adj setting (all after 2005). Patient characteristics were compared between pre-adj H (2001-2005) and post-adj H (2006-2011) cohorts. The only significant differences noted between the cohorts in univariate analysis were extra-CNS disease control and use of anti-HER2 therapy after BM dx: pre-adj H era pts had a higher likelihood of extra-CNS disease control (79.5% vs. 52%, p = 0.004) at BM dx and less use of anti-HER2 tx after BM dx (70% vs. 87.5%. p = 0.05). For all pts, the median follow-up for survivors was 33.5 mos (range 18-103). There were 79 deaths. The median survival from BM dx was 19.4 mos (95%CI: 15.5, 26.6). KM curves and log-rank tests showed significantly better survival from BM for pts with higher KPS, single BM, extra-CNS disease control, lack of neurologic sx at BM dx, initial presentation without LMD, use of lapatinib (ever), SRS as initial RT, and use of any anti-HER2 tx after BM dx. Multivariate analysis showed that higher KPS [HR 0.21 (0.09,0.53)], extra-CNS disease control [HR 2.89 (1.67, 5.00)], single BM [HR 4.73 (2.11,10.60)], use of anti-HER2 tx after BM dx [HR 0.30 (0.17,0.53)], asymptomatic status at BM dx [HR 3.69 (1.69,8.07)] were associated with improved survival from BM. Lack of neurologic sx at BM dx was significantly associated with longer survival from BM even after adjustment for other potentially confounding prognostic factors. Conclusion: These data are mostly consistent with prior reports regarding prognostic factors and inform contemporary clinical trial design. In pts with HER2+BCBM where the majority were subsequently exposed to anti-HER2 and other systemic tx, presence of neurologic sx at BM dx was significantly associated with worse survival after BM dx. Prospective evaluation of screening brain MRI allowing for earlier detection, leading to more careful monitoring or pre-emptive tx of asymptomatic BM may be warranted. The observation of better extra-CNS disease control at BM dx among pts in the pre- adj H era was unexpected, is hypothesis generating, and requires independent confirmation. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P6-11-02.