MODL-25. INTERNAL CAROTID INJECTION MODEL OF BRAIN METASTASIS DESCRIBES LEPTOMENINGEAL DISEASE

Abstract

Abstract Leptomeningeal disease (LMD), the most aggressive form of central nervous system (CNS) metastasis in which cancer cells infiltrate leptomeninges, is universally and rapidly fatal due to poor detection and no effective treatment. Treatment of LMD is challenged by the diffuse nature of the disease, low specificity of detection, poor drug penetration into CNS and high neurotoxicity. LMD is a heterogeneous disease: co-existence of leptomeningeal spread and parenchymal brain metastases in not uncommon, and LMD tumors vary in size and may be found at various locations throughout entire neuroaxis. Radiographic detection (i.e., MRI) works well for big “bulky” metastases but may produce normal results in patients with active disease identified by cerebrospinal fluid (CSF) tap. This spatial and morphological heterogeneity of LMD may ultimately be responsible for a varied and limited therapy response. Mechanistic and translational studies of LMD are hampered by the lack of adequate in vivo models that recapitulate the consecutive steps of LMD progression separate from systemic disease. In this study, we show that the internal carotid (ICA) injection of tumor cells, which is widely used to model brain metastasis, produces leptomeningeal spread. The ICA injection of various tumor cell lines faithfully recapitulated the stages, heterogeneity, and spatial distribution of human LMD. While some cell lines demonstrated clear preference for growth in brain parenchyma, they still had a measurable population of cells within the leptomeninges. Most bulky LMD metastases formed and grew near fenestrated endothelium, where the direct access to blood might support tumor growth and survival. A population not associated with the fenestrated endothelium consisted mostly of single non-dividing non-hypoxic cells. This diffuse cancer spread limited the efficacy of standard whole-brain radiation therapy. Thus, leptomeningeal spread should be taken into account when interpreting the data from the ICA model that is widely used to study brain metastasis.