¹Molecular Medicine Program, Guggenheim 18, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905, USA

²ICRF Molecular Oncology Unit, Department of Cancer Medicine, Imperial College School of Medicine, London, UK

Correspondence to: R Vile

Gene therapy for the treatment of cancer was initiated with high levels of optimism and enthusiasm. Recently, this perception has had to be tempered by the realisation that efficiency and accuracy of gene delivery remain the most significant barriers to its success. So far, there has been a disappointing inability to reach target cells with sufficient efficacy to generate high enough levels of direct killing and this has necessitated the invocation of bystander effects in order for any potential strategy to be convincing. At least in the foreseeable future, clinical advance will come from co-operation with other more established disciplines - such as chemotherapy, radiotherapy and immunotherapy. This is inevitable - and necessary - in order to prove that gene therapy can have efficacy as part of a combinatorial therapy, before hoping to move clinical mountains alone. In addition, there will have to be a thorough understanding of the clinical situations in which gene therapy will be used in order both to understand its own limitations, and to exploit its full potential. This will enable it to find the appropriate clinical niche in which its abilities will be optimally useful. Finally, anyone wishing to practise clinical cancer gene therapy will rapidly have to learn the ways of the free market and be able to juggle commercial necessities with ideological purity. Gene Therapy (2000) 7, 2-8.

cancer; gene therapy; immunotherapy; viral vector; targeting