Ìý
Brain tumours are among the most fatal and rapidly progressing cancers, and survival beyond two years is rare. In a recent collaborative study, GCI researchers profiled the cellular composition and spatial organization of the tumour immune microenvironment in brain cancer patient tumours using highly multiplex imaging technology.
.
Meet the co-first authors
Miranda Yu, GCI trainee, is a PhD student in the Department of Physiology in the lab of Prof. Daniela Quail. Follow her on Twitter @Miranda_WYu
Sarah Maritan, GCI trainee, is an MD-PhD candidate in the Division of Experimental Medicine in the lab of Prof. Peter Siegel.ÌýHer training in the MD-PhD program was instrumental to her success.ÌýFollow her on Twitter @SarahMaritan
Elham Karimi, Research Associate in the lab of Prof. Logan Walsh at the GCI.
Ìý
Ìý
Ìý
Ìý
Ìý
What’s new in this paper?
Sarah – Our paper uses a relatively new technology called Imaging Mass Cytometry, or IMC. IMC lets us look at many different cell types at once in a tumor sample, far above the limitations of traditional technologies. In using IMC, we get a global look at what types of cells are in and around the tumor, and how exactly these cells interact with neighbouring cells. In our paper, we use IMC to look at more than 1.1 million cells in both primary and metastatic patient brain tumours, and generate one of the first and largest datasets that maps how immune cells spatially interact within brain tumours.
Miranda - Thanks to the many clinicians, collaborators and most importantly, patients, who contributed to this study, we have assembled one of the largest brain tumour data sets published to-date including samples from both glioblastoma and brain metastasis patients. While survival beyond 2 years is rare in glioblastoma, we were also very fortunate to have access to many samples from long-term survivors (>3 years), which allowed us to hunt for tumour characteristics that could be associated with better patient outcomes.
Elham – The IMC brain tumor dataset that we have access to in this paper is a novel dataset of images, for which, we needed to design and implement new pipelines/analysis frameworks that uniquely fit this data. Although there are several existing pipelines in the literature, these frameworks were not able to handle the complexity of cell structures that we were looking after and did not necessarily lead to the interesting results that we expected. Therefore, our proposed novel pipelines which are designed from scratch played an essential role in finding these new findings
How does it add to what was already known?
Sarah –The brain has previously been thought of as an immune-privileged sanctuary for cancer cells, somewhere the immune system has limited access and the tumor can freely grow. Our paper adds to emerging evidence that this isn’t necessarily the case, and many types of immune cells are present in and around brain tumours, where they directly interact with cancer cells. Our work delves into these cellular interactions and looks at different groups of cells, called neighbourhoods, that are distinct to different types of brain tumours. In doing so, this work adds to the growing number of datasets that use spatial information to get a better understanding of how tumours interact with their environment.
Miranda – In addition to providing a spatial map of primary and metastatic brain tumors, we also shed light on how an immune cell called macrophages play a critical role in glioblastoma. Across many solid cancers, macrophages are well-known for their tumor-promoting and immune-suppressing properties. That’s why we were so intrigued and excited to find a subset of macrophages containing myeloperoxidase (MPO) protein, that was associated with long-term survival. Immunotherapies that broadly target macrophages have generally been unsuccessful in glioblastoma clinical trials, and we are beginning to appreciate the complexity of these cells and that not all tumour macrophages are ‘bad’.
How did you discover it? What led you to ask the question and explore it this way?
Elham – The discovery approach in this paper was an incremental and iterative process. We first implemented the needed tools/pipelines to extract cellular spatial and phenotypic information, and then we tried to tie that information to biological hypotheses. Deep dive into how this structural information that we extract from our images could be related to each other, led us to try many different scenarios some led us to interesting clues and discoveries that are presented in this paper.
Miranda – As cancer immunologists, we often study how specific immune cells influence tumour biology. However, these cells almost never act alone – they are constantly in communication with one another and it’s through these coordinated interactions within the tumour microenvironment that tumours can persist or perish. IMC gives us the power to take a deep dive into these networks of cells and ask important questions: which cells tend to interact with each other? Do cells function differently when they are communicating with different parts of the tumor? And how does this all relate to patient outcomes?
Why is this finding important?
Sarah – Part of the beauty of working with data generated using IMC is just how many different scientific questions can be asked. This paper lays the foundation for how our dataset can be used to identify clinically relevant features of the brain tumour environment. We followed one thread of the data to identify a unique cell type, MPO+ macrophages, in one specific type of brain tumours, glioblastoma. However, having collected spatial information of 20 different cell types in tumors from 185 patients, there is no shortage of other threads to follow. We are hopeful that this work will serve as a resource for the research community, where our dataset can be leveraged to ask more questions, and ultimately uncover other potential therapeutic targets for the treatment of brain tumours.
Miranda - Glioblastoma is a notoriously aggressive disease, and it is extremely difficult to treat. Taking advantage of the patient’s immune system to fight glioblastoma (such as with T cell-targeted immunotherapies) has been historically challenging, but here, we have identified a new cell subset, MPO+ macrophages, as critical players that have yet to be explored therapeutically.
What are the practical implications?
Elham – Our paper is a deep study that suggests the use of spatial information alongside other information that we extract from our cohorts of data can ultimately help us analyze primary brain tumors and metastases. Specifically, our analyses unveiled cellular neighborhoods that showed the association of a specific population of a cell type (obtained using our pipelines) with long-term survival patients. In the future, findings like this enable research/clinical works to integrate spatial resolution into single-cell datasets where we can probably better dissect the microenvironmental context of cancer in the brain.
Ìý