Abstract P6-03-04: Central nervous system miliary metastasis in breast cancer patients

Abstract

Abstract Background: Little is known regarding central nervous system (CNS) miliary metastasis (MiM), which was first described as “carcinomatous encephalitis” by Madow and Alpers in 1951. The majority of reported cases arise from primary lung and gastrointestinal adenocarcinomas, with occasional melanoma primaries and one reported case in breast cancer. Moreover, clinicopathologic correlates, disease outcomes and prognostic factors in these patients are poorly understood. Although identified most frequently on neuroimaging, radiographic criteria to objectively diagnose MiM do not exist. In this analysis of patients with brain metastasis from primary breast cancer, we propose objective, stringent radiographic criteria for CNS MiM diagnosis and identify clinicopathologic factors contributing to disease outcomes. Methods: Using a prospectively maintained electronic database, 1,002 female patients diagnosed with brain metastasis from primary breast cancer between 2000 and 2015 were identified. Only patients with neuroimaging available for direct review (CT or MRI) were included. Our radiographic criteria for MiM diagnosis were: 1) ≥20 metastatic lesions per image slice on ≥2 noncontiguous image slices by MRI, or 2) ≥10 lesions per image slice on ≥2 noncontiguous image slices by CT, and 3) MiM lesions were required to be present bilaterally and in both the supra- and infratentorial compartments. These criteria were established upon direct review of all neuroimaging by a neuroradiologist. Number and anatomic distribution of metastatic lesions were the patterns best observed to identify cases of CNS MiM on case review; lesion size was not a reliable pattern for MiM identification. Log rank tests were used for statistical analyses. Results: Using stringent criteria, 486 patients were included in this analysis. Twenty patients with MiM were identified (4.1%). Ten patients were diagnosed with MiM at initial brain metastasis presentation; 10 developed MiM after known brain metastasis. Biomarker based subtype distribution was as follows: HR-/HER2- (TNBC) (n=8), HR+/HER2+ (n=3), HR+/HER2- (n=4), HR-/HER2+ (n=4), unknown (n=1). Table 1: Disease Outcomes Based on Biomarker SubtypeBiomarker SubtypeMedian Time to MiM (months) (p=0.104)Median Survival after MiM (months) (p=0.008)TNBC (n=8)32.3 (12.1-132.5)1.8 (0.5-4.0)HR+/HER2+ (n=3)44.2 (33.2-71.5)10.8 (10.2-13.3)HR+/HER2- (n=4)110.2 (23.0-156.0)4.8 (0.8-9.8)HR-/HER2+ (n=4)27.1 (3.7-39.4)4.0 (1.8-5.0)All* (n=20)37.4 (3.7-156.0)3.7 (0.4-12.3)Key: BM: Brain metastasis; * Includes 1 patient with unknown subtype. Conclusions: Reports of MiM consist overwhelmingly of lung and gastrointestinal adenocarcinoma primaries. This retrospective, observational study is the first to establish that CNS MiM occurs in breast cancer with an incidence of roughly 4%. Review of an additional 1,600 patient charts is underway, but this preliminary study is the first to identify clinicopathologic correlates and determine disease outcomes in patients with MiM; it is also the first to propose stringent radiographic criteria for the diagnosis of CNS MiM, and further updated data will be presented at the meeting. Citation Format: Bashour SI, Ibrahim NK, Schomer DF, Colen RR, Sawaya R, Suki D, Rao G, Murthy RK, Moulder SL, Abugabal Y, Hess KR, Fuller GN. Central nervous system miliary metastasis in breast cancer patients [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P6-03-04.