Modelling, characterization, and therapeutic targeting of human brain metastasis

Abstract

Brain metastases (BM) are the most frequently diagnosed neoplasm to affect the adult central nervous system (CNS), occurring in 20-40% of all cancer patients throughout the course of the disease. The significant advancements to the treatment and control of primary cancers have unfortunately resulted in an increased incidence of BM, however, this complication of cancer progression continues to be met with a dismal outcome and limited therapeutic options. There remains a poor understanding of the several cellular hallmarks of BM, encompassing various molecular, genetic and epigenetic changes that underlay the stages of metastasis, which requires the development of clinically relevant models of metastasis. Our group has previously established the existence of a cancer stem cell/tumor initiating population with patient samples of BM, which established the foundation of this thesis. Thus, I postulate that there exists a subpopulation of cancer stem-like cells, termed brain metastasis initiating cells (BMICs), that is responsible for the initiation of BM and is identifiable by an exclusive subset of genes that regulate self-renewal and metastasis.

To support this hypothesis, I established novel experimental models of BM by inoculation of BMICs derived from patient samples of lung-to-brain metastases into intracranial (ICr), intracardiac (ICa), and intrathoracic (IT) routes into NOD/SCID mice. ICr injections validated the presence of a tumor initiating cell (TIC) capacity of BMICs in the secondary environment (brain). From ICa injections I was able to recapitulate macro-metastatic growth, whereas with IT injections I was able to capture the complete metastatic process, from primary lung tumor formation to micro-metastasis growth. Utilizing these models, I determined that the STAT3 pathway and genes SPOCK1 and TWIST2 all contribute to the regulation of BM development, where SPOCK1 may pose as a potential BMIC marker. Further interrogation of the metastatic process utilizing the IT model of BM led to the characterization of “pre-metastasis”, a stage where BMIC cells have crossed the blood-brain barrier and employ mechanisms to invade and seed the neural environment, while simultaneously repressing mechanisms of proliferation and cell growth that would indicate tissue colonization.

In summation, I propose a shift in the cancer research paradigm to target the metastatic process itself, to prevent the dissemination of primary tumor cells to the brain. I present models of clinically relevant models of human BM that have proved to be reliable as platforms to interrogate the process of BM, providing insight into the stage of pre-metastasis as a novel therapeutic window into BM prevention and possible extension of patient survival.