Gene therapy originally carried the idea of the cure of genetic diseases through replacement of dysfunctional genes or, more realistically, through augmentation of an inadequate genome by addition of correctly functioning genes. Currently, the term may be used to cover any treatment where a therapeutic effect is mediated by the transfer and subsequent expression of exogenous nucleic acid. The therapeutic effect may be due to the production of a beneficial protein, for example the CFTR protein in cystic fibrosis patients, or it may be due to the production of directly or indirectly toxic proteins, such as prodrugs for the elimination of cancer cells. Strictly speaking, gene therapy does not include the use of nucleic acids to mediate the production of antigenic proteins for the purposes of vaccination, although it has to be said that the technology overlap between vaccination and gene therapy may be extensive.

Gene therapy has been described both as 'the ultimate application of genomics' and as pure 'hype'. The purpose of this article is to briefly review gene therapy technologies and to suggest which of the views encapsulated in the title is closer to the truth.

Major gene transfer systems in use

A wide variety of gene transfer systems has been developed and tested. Most success has been achieved with gene transfer systems based on modified viruses. Such systems include vectors derived from retroviruses, adenoviruses, adeno-associated viruses, and herpes viruses. These are used because they have evolved precisely to bring about expression of foreign genes in human cells.

Retroviruses were among the first gene vectors. They have the advantages of being well understood, and easy to produce and to manipulate. In tissue culture they are capable of mediating long term gene expression due to the ability of the vector to integrate into the host DNA. However, in vivo the retroviral vectors often mediate only short-term expression of exogenous DNA due to 'vector shutdown' by mechanisms which may involve host cell methyltransferases (enzymes which act on chromosomal DNA). Furthermore, integration carries a theoretical disadvantage in that if random it could be dangerously disruptive. For example, if a vector were to integrate upstream of a gene involved in cell cycle regulation, then the strong viral promoter might interfere with expression of the cell cycle gene in such a way as to cause uncontrolled (malignant) cell growth. However it has to be said that there is an increasing body of evidence to suggest that the risk of insertional mutagenesis from retroviral integration is very small. Other disadvantages of retroviruses include the relatively small amount of DNA that they can accommodate (up to about 9 kilobases (kb - a measure of the length of a DNA sequence) depending on the type of vector), the relative fragility of the retroviral particle (which makes it difficult to routinely concentrate these vectors to high titres), and the general requirement for host cell division for successful infection.

The perceived disadvantages of retroviruses stimulated the development of alternative gene transfer systems, such as adenoviral vectors. These have the advantages of a good theoretical safety profile (long history of use in humans, no integration, only mildly pathogenic in the natural state), the ability to mediate expression in non-dividing cells, the stability of the particle allowing concentration to very high titres, and the ability to carry significantly more DNA (~30kb) than retroviruses. Disadvantages include transient expression and the immunogenic nature of the adenoviral particle. This immunogenicity results in host cells infected with the adenoviral vector being eliminated by the cellular arm of the immune system, and in the generation of circulating antibodies which directly sequester viral particles. It is also worth noting that the single death attributable to a gene therapy protocol, that of Jesse Gelsinger in September 1999, was associated with an adenoviral vector, albeit at very high doses.

Adeno-associated viral vectors (AAVs) have the interesting ability to integrate into host genomes in a relatively site-specific manner, raising the possibility of mediation of long-term expression at a greatly reduced risk of insertional mutagenesis. However, the real question is whether AAV vectors can mediate expression without shutdown or immune attack. Some data suggest that these vectors can mediate expression for over two years in vivo, perhaps due to a relatively limited cellular immune response to these vectors. Other advantages of AAV include the ability to infect non-dividing cells. The great disadvantages of AAV vectors are the minute amount of DNA that they can package - only about 4.5 kb - and the low titres achievable.

Herpes virus vectors are a newer class of vector which appear to be able to mediate long term expression through their ability to persist in a latent state without interfering with the host cell's functions and without inducing cellular immunity. A particular advantage of this class of vector is the DNA carrying capacity - up to 50kb. Disadvantages include the toxicity and immunogenicity of the wild-type virus; however some herpetic vector systems may largely have overcome these issues.

It seems clear that there will never be one universally applicable gene transfer system; vectors will have to be carefully designed and customised according to the specific requirements of each therapeutic application. There are various novel vectors currently under development, and the likely future development of gene therapy technology, and the key technical and commercial issues it faces, may be explored in detail if required (please contact the author).

Current sentiment towards gene therapy technology

In the US, subsequent to the furore following Jesse Gelsinger's death, sentiment towards gene therapy has been bolstered by improvements in the clinical gene transfer regulatory system, and by recent promising results from gene therapy trials. These include the apparent cure of two infants with severe combined immunodeficiency disease (SCID) through ex vivo transduction with a retroviral vector, and expression of Factor IX at near therapeutic levels in haemophilia B patients through injection of an AAV vector. In terms of clinical trials activity, as of May 2000, 206 investigational new drugs were still active.

Perhaps as a result of the trials results, there is increasing interest in gene therapy investments among certain providers of capital, and several gene therapy companies have taken advantage of this financing window. In June 2001 Transkaryotic Therapies Inc raised nearly $90m, and in July 2001 Intronn received a $2.5m SBIR grant to develop its genetic therapy for cystic fibrosis. Oxford Biomedica, a company developing retroviral vectors, raised about £27m in April 2001, and Biovex, which is focusing on herpes simplex vectors, raised £10m in July 2001. This general sentiment is also reflected in the recent (Nov 2001) award of $2.7m by the NIH to 5 US academic centres to establish new National Gene Vector Laboratories, and the recent FDA decision to grant orphan drug status to Coagulin B, Avigen's AAV-Factor IX gene therapy for haemophilia B.

Gene therapy's future

There may still be a degree of hype surrounding gene therapy. The SCID cures are slightly less impressive when one considers that the transduced cells probably enjoyed a selective advantage over their non-transduced brethren - thus a very inefficient initial infection event could have been amplified by outgrowth of the treated cells. There will be very few other diseases that are so amenable to amelioration by transduction of a minority of cells.

However, excitement over gene therapy clinical trials results is only natural given the transforming nature of the technology; the birth of a new therapeutic modality unavoidably generates enthusiasm and sometimes hyperbole, as well as the usual complement of carpetbaggers. The fact remains that gene therapy is the only approach capable of actually curing (as opposed to treating) human genetic disease, and that it has clear potential as an efficient method of delivering therapeutic proteins in a correctly regulated manner. Furthermore, the success of the genomics project means that there is a growing number of known genes which could mediate therapeutic effects. Although many claim that a gene is something that exists in nature and therefore cannot be patented, nevertheless Incyte, Celera, Human Genome Sciences and Millennium Pharmaceuticals have been and are patenting gene sequences. To date, the USPTO has issued patents on more than 6000 genes, 1000 of which are human, and although the US requirements for gene patenting have been tightened as of Jan 7th 2001, efforts to claim ownership of genes are likely to continue for the foreseeable future. All these factors, together with technical advances and promising clinical trials results, have led to a resurgence of interest in gene therapy, and the capital markets (at least prior to Sep 11th) appear to be favourably disposed towards gene therapy companies

However, it is also true that the time to market for many gene therapy products will certainly be no shorter than for many other biotech products, indicating that a strong cash position and the ability to decrease burn rate through revenue-generating corporate collaborations will be very important factors for investors in gene therapy companies. It is also clear that gene therapy remains high risk - only three such procedures have advanced to Phase III trials. To conclude: is gene therapy the ultimate application of genomics, or has it been hyped? The answer is yes, to both. Probably.