Table 2.

Types of in vitro, ex vivo and in vivo models to study brain cancer (images created with BioRender.com).

	Summary	Advantages	Disadvantages
Monolayer cell culture	2D culture system Can be either primary or immortalised cell lines	Readily available resource Minimal cost Easy to use Allow rapid studies of biological properties of tumours Allow screening of multiple drug compounds	Genetically homogeneous Genetic and epigenetic drift from the original tumour caused by the standard serum-based in vitro conditions Impossibility to test tumour-host interactions
Neuro/tumoursphere culture	3D culture system Generated from 2D cell lines or freshly resected tumour	Preserve the complex individual level of genetic and epigenetic heterogeneity of the original tumour sample Maintain genetic stability over time Allow screening of multiple drug compounds	Represent just the cancer stem cell population of the original tumour 3D nature means that media nutrients are not equally distributed Generation depends on tumour grade and genetics Lack a realistic brain microenvironment composed by vessels and immune cells
Cerebral organoids	Originate from hPSCs and/or hESC Form self-organising organ structures	Recapitulates the variety of cerebral cell types Retain anatomical features of a simplified mammalian brain Allow studies of biological properties of tumours such as invasion, in a more physiological system compared to monolayer cultures	Lack a realistic brain microenvironment composed by vessels and immune cells
Organotypic brain slice culture	Maintain mouse brain slices in culture The genetic alteration can be introduced in the germline (knock-out, knock-in, transgenic models) or in the somatic cells (via viral-mediated gene delivery)	Maintains the in vivo brain architecture Allow faster studies of invasion ex vivo Drug screens can take place in a more realistic brain microenvironment	Not suitable for long term studies
Genetically engineered mouse models	The genetic alteration can be introduced in the germline (knock-out, knock-in, transgenic models) or in the somatic cells (via viral-mediated gene delivery)	Genetic changes can be targeted to specific cell populations and induced at specific developmental stage of the animal Animals have an intact immune system Roles of individual mutations can be ascertained Tumour formation can be monitored from the very start	Only a select few genes can be altered and so does not reflect the true genetic heterogeneity Expensive and time consuming
Xenograft mouse models	Injection of individual tumour cells or en bloc biopsies Tumour sample can be injected heterotopically or orthotopically	Closely resembles the original features of the tumour Heterogeneity is maintained Tumours generate their own vasculature	Animals are immune compromised so tumours do not experience input from the immune system
“Humanised” xenograft mouse models	Addition of human immune cells to build a “human immune system”	Tumours experience a human immune response	Very expensive and time consuming Possibility of mice developing GVHD