Although intrinsic tumours of the brain seldom metastasize to distant sites, their diffuse, infiltrative–invasive growth within the brain generally precludes successful surgical and adjuvant therapy. Hence, attention has now focused on novel therapeutic approaches to combat brain tumours that include the use of anti-invasive and anti-proliferative agents. The effect of four anti-invasive agents, swainsonine (a locoweed alkaloid), captopril (an anti-hypertensive drug), tangeretin and nobiletin (both citrus flavonoids), were investigated on various parameters of brain tumour invasion such as matrix metalloproteinase (MMP) secretion, migration, invasion and adhesion. A standard cytotoxicity assay was used to optimize working concentrations of the drugs on seven human brain tumour-derived cell lines of various histological type and grade of malignancy. A qualitative assessment by gelatin zymography revealed that the effect of these agents varied between the seven cell lines such that the low grade pilocytic astrocytoma was unaffected by three of the agents. In contrast, downregulation of the two gelatinases, MMP-2 and MMP-9 was seen in the grade 3 astrocytoma irrespective of which agent was used. Generally, swainsonine was the least effective whereas the citrus flavonoids, particularly nobiletin, showed the greatest downregulation of secretion of the MMPs. Furthermore, captopril and nobiletin were most efficient at inhibiting invasion, migration and adhesion in four representative cell lines (an ependymoma, a grade II oligoastrocytoma, an anaplastic astrocytoma and a glioblastoma multiforme). Yet again, the effects of the four agents varied between the four cell lines. Nobiletin was, nevertheless, the most effective agent used in these assays. In conclusion, the differential effects seen on the various parameters studied by these putative anti-invasive agents may be the result of interference with MMPs and other mechanisms underlying the invasive phenotype. From these pilot studies, it is possible that these agents, especially the citrus flavonoids, could be of future therapeutic value. However, further work is needed to validate this in a larger study.