Primary tumours of the central nervous system (CNS) are an important cause of cancer-related deaths in adults and children. CNS tumours are mostly glial cell in origin and are predominantly astrocytomas. Conventional therapy of high-grade gliomas includes maximal resection followed by radiation treatment. The addition of adjuvant chemotherapy provides little improvement in survival time and hence assessment of novel therapies is imperative. We have evaluated the potential therapeutic use of the herpes simplex virus (HSV) mutant 1716 in the treatment of primary brain tumours. The mutant is deleted in the RL1 gene and fails to produce the virulence factor ICP34.5. 1716 replication was analysed in both established human glioma cell lines and in primary cell cultures derived from human tumour biopsy material. In the majority of cultures, virus replication occurred and consequential cell death resulted. In the minority of tumour cell lines which are non-permissive for mutant replication, premature shut-off of host cell protein synthesis was induced in response to lack of expression of ICP34.5. Hence RL1-negative mutants have the distinct advantage of providing a double hit phenomenon whereby cell death could occur by either pathway. Moreover, 1716, by virtue of its ability to replicate selectively within a tumour cell, has the potential to deliver a 'suicide' gene product to the required site immediately. It is our opinion that HSV which fails to express ICP34.5 could provide an effective tumour therapy.