Glioblastoma is the most lethal form of primary brain tumour. Standard treatment combines surgery, radiotherapy and chemotherapy but is of limited efficacy, prompting strong interest in developing novel immune-based therapies. Fibroblast activation protein (FAP) is a cell surface protease with limited expression in healthy tissues. Here, we investigate its potential as a target antigen for CAR-T cell therapy of glioblastoma. Expression analyses at the gene and protein level revealed that, compared to normal brain, FAP was overexpressed in a large proportion of glioblastoma tissues, with the highest levels of expression associated with poorer prognosis. FAP was also commonly expressed by short-term patient-derived cultures of glioma neural stem (GNS) cells, but absent from stem cell-derived normal neurons and astrocytes. Within glioblastoma tissues, FAP was expressed not only by the tumour cells themselves, but also by blood vessels. In fact, almost every tumour vessel was highlighted by FAP expression, whereas normal tissue vessels and cultured endothelial cells (ECs) lacked expression. Single cell analyses of dissociated tumours (using high parameter flow cytometry and scRNAseq) facilitated a detailed characterisation of the main cellular components of the glioblastoma microenvironment, revealing that vessel-localised FAP is due to expression on both ECs and pericytes. In real time cytotoxicity assays, second generation CAR-T cells targeting FAP efficiently destroyed FAP-expressing GNS cells. Moreover, we used a model of mixed human glioblastoma cell lines subcutaneously implanted into mice to demonstrate significant tumour growth inhibition by FAP-targeting CAR-T cells, even where expression of FAP by tumour cells was heterogeneous. Together, these studies reveal FAP as a novel and clinically relevant CAR-T cell target antigen in glioblastoma, which may facilitate simultaneous targeting of both tumour cells and their supporting vascular network.