Elucidating Primary CNS Germ Cell Tumors via Single Cell and Spatial Multi-omics
Brain tumors that develop from germ cells — the cells that normally give rise to eggs and sperm — occur in children and young adults. These tumors, called central nervous system (CNS) germ cell tumors, manifest as several distinct types. Some, like germinomas, respond well to treatment, while others, like yolk sac tumors and choriocarcinomas, are aggressive and challenging to cure. Collectively, they affect hundreds of children each year in the United States alone. Today, treatment relies on radiation and chemotherapy, which can cause lasting harm to a child’s developing brain, growth, and hormonal systems and even lead to the development of other types of brain cancers. What makes these tumors particularly puzzling is that despite behaving so differently, they largely share the same type of cancer-causing defects in the DNA, or genetic mutations. This tells us that mutations alone do not determine how a tumor behaves. No targeted drugs exist, largely because we do not understand what makes each subtype biologically unique. Previously, it has been challenging to study these tumors with the latest technologies for studying cancers, because these tools were applicable only to fresh or frozen tissue. Now, we have developed similarly powerful tools that can be applied to pathology specimens that have been chemically preserved, allowing us to study rare tumor types, such as CNS germ cell tumors, that would be nearly impossible to collect fresh in large enough numbers.
Project Goals
Our goal is to elucidate the biology behind the specification of CNS germ cell tumors into distinct subtypes. We will achieve this by studying thousands of individual cancer cells through new technologies developed in our laboratory. We will examine both the DNA mutations these cancer cells harbor and other factors that may influence the cancer behavior, specifically, the type of cell the tumor resembles and the surrounding tumor environment. In the first part of the project, we will read the genes inside individual tumor cells to understand what type of cell each tumor most closely resembles, and how specific mutations change behavior depending on that identity. In the second part, we will build spatial maps of tumors to determine how cell types are arranged and how they interact with surrounding non-cancerous cells and normal brain tissue. Together, these approaches will reveal the molecular fingerprints that make each subtype unique. We will then validate and prioritize the most promising therapeutic targets in mouse models, in which patient-derived CNS germ cell tumor cells are grown, to confirm which vulnerabilities can be exploited to stop tumor growth. The long-term goal is to reduce or replace radiation and chemotherapy with targeted therapies in children with these tumors, improving survival and quality of life.
