What does the Future Hold for Viral Gene Therapy?

‘The door is slowly swinging shut on viral gene therapy, and if substantial progress is not made within the next 5–10 years, it may be closed forever.’

Since the 1970s when researchers were first able to introduce genes into mammalian cells, consideration has been given to gene therapy for both inherited and acquired diseases of man. Over the last several decades we have observed many false starts, an incredible expansion in the number of investigators in the field, disinterest followed by interest turning back into disinterest from commercial entities, an evergrowing and perplexing array of vectors and approaches, a few high-profile medical misadventures, some small fortunes made and yet not a single US FDA-licensed product. The one clear success often pointed to is that of the X-linked severe combined immunodeficiency trial in France [1,2], which was notoriously qualified by several serious adverse events [3,4]. A clinical trial in Italy involving adenosine deaminase deficiency, has, however, not been plagued with similar problems [5]. Even staid adenoassociated virus, thought to be nonpathogenic in man, has been linked to the development of malignancy in mice [6].

Despite the pall cast over the gene therapy field, most investigators have pressed on, refining vectors, transgenes, delivery technologies and animal models. Although there has been some contraction (in part related to the flattening of the NIH budget, the withdrawal of pharmaceutical firms and the perceived lack of success), the American and European Societies of Gene Therapy are as active as before. Given advances in new technologies such as RNAi and siRNA, the pendulum will likely swing back, again favoring gene therapeutics for certain illnesses. But considering what we now know, what is the immediate and long-term future for gene therapeutics against viral diseases? For this editorial, I will not discuss approaches that are mainly cell-based (e.g., modified autologous cytotoxic T cells for acute or chronic herpes virus infections such as EBV-related post-transplant lymphoproliferative disease) or RNA-based (intravenous delivery of siRNA directed against HCV or respiratory syncytial virus genes). Nor will I mention vaccination/immune response strategies related to gene therapy.

I believe that gene therapy is still appropriate for three chronic viral illnesses; namely those caused by HBV, HCV and HIV. For HIV and HBV, the ‘window of opportunity‘ is closing, whereas for HCV it is still fairly wide open (and will likely remain so for the foreseeable future, until small-molecule therapeutics are developed).

Millions of individuals worldwide are chronic carriers of HBV and are unlikely to receive any therapeutics at all. Many of these will develop cirrhosis, hepatic failure or hepatoma, much of this being related to the chronic inflammatory changes in the liver caused by HBV infection. Over the last two decades, there has been great progress made against HBV, with the universal availability of a safe and efficacious recombinant protein vaccine and several small molecules that target HBV polymerase (not to forget interferon) [7]. Since there is no natural reservoir for HBV, if large-scale, worldwide vaccination efforts (ongoing in some countries for more than a decade) are successful, it may yet be possible to completely eliminate the virus from the world.

‘…a few high-profile medical misadventures, some small fortunes made and yet not a single US FDA-licensed product.’

Eradication of HBV, however, is unlikely to happen for several decades, especially given the number of individuals currently infected. In the Western world where therapeutics are often prescribed, is there any role for gene therapy? I would argue yes, but several barriers may make it difficult to conduct even a Phase I trial in the next few years. These include lack of a good tissue culture system and a convenient animal model, and a dearth of cellular and viral targets. This may change somewhat with the advent of the plasmid hydrodynamic mouse HBV replication system that allows measurement of viral particle production in whole animals [8]. The fact that HBV is relatively small with multiple overlapping reading frames, coupled with limited natural viral diversity, suggests that there are constraints on the viral DNA sequence. Such limitations might be advantageous for certain gene therapeutic approaches, especially shRNA [9–11], antisense RNA [12], RNA decoys, or ribozymes [13–15] directed against viral gene products or cis-acting sequences.

Ideally, any gene therapeutic should provide a selective advantage (compared with unmodified cells) in a cell-autonomous manner, in the setting of active viral replication. In other words, if the transgene does not prevent cell infection but simply limits the amount of virus produced or released from that cell, it may not be enough, especially since hepatocyte cell death is multifactorial. Thus, simply showing reduced viral replication in a population of gene-modified cells does not provide sufficient information and may even be misleading, especially if only short-term assays (duration of days to a week) are performed. To my knowledge there are no current transgenes that fulfill these criteria and given limitations in the in vitro and in vivo systems, none may be developed in the near future.

Millions of individuals worldwide are also infected with HCV, and are accordingly also at risk for chronic hepatitis, cirrhosis and hepatoma. In the developed world, a combination of ribavirin and interferon can be curative [16,17]. Unfortunately, depending upon viral genotype and unknown host and viral factors, these therapeutics may not clear the virus. Although there is no natural host for HCV there is also no vaccine, and because of viral diversity it is unlikely that one will be developed in the near future. Taken together, this would suggest that HCV would be an excellent target for gene-transfer approaches. Until recently, however, there was only a very cumbersome animal model (chimpanzee) and no good cell-culture system. Now, however, there are a variety of HCV replicons to allow propagation of RNA along with the host hepatoma cells in vitro[18], and it is also possible to achieve full viral replication in cell culture [19]. Additionally, the xenotransplant models in which human liver tissue is implanted into immunodeficient mice afford some degree of viral replication [20–22]. These advances should allow testing of candidate transgenes [23], but it is important to note that the same caveats mentioned above still apply. I suspect that these systems will prove to be much more valuable for dissecting the replicative cycle of HCV [24] and also for the testing of small molecules that may have therapeutic activity against HCV [25], both of which will be much more important to explore in the short term than any envisioned gene therapeutic approaches.

‘Over the years, there has been no shortage of transgenes and vectors that have been shown to block HIV replication in cultured cells … researchers approached the problem with an almost irrational exuberance and keen naivety, and transpired time has tempered those traits.’

HIV has been the darling of gene therapists for almost two decades. Before the era of highly active antiretroviral therapy (HAART) began in 1996, licensed antiretrovirals did very little to stem viral replication or change the course of illness (and one could argue that clinicians were not even doing or measuring the right things). Now, at least in the developed world (and all too slowly for the developing world), HAART has changed the face of HIV. With over 20 medications currently approved, that cover several different aspects of the virus life cycle (more are on the way in 2007–2008), for the majority of patients, HIV replication is suppressed and immune function restored. HIV, more often than not, is considered a chronic illness today, akin to diabetes, hypertension and coronary artery disease. It is not curable but is treatable.

For a nontrivial percentage of patients, however, HIV can still be deadly, especially given poor medication adherence, drug toxicities and resistance to licensed antiretrovirals. The medications themselves are quite expensive (annual cost of Fuzeon or T-20 is nearly US$ 20,000). Few vaccines have made it as far as Phase III clinical trials (and if they have, they have fared poorly), although several are progressing through Phase I/II testing (and the preclinical pipeline is relatively full). Former President Clinton‘s decade deadline for a licensed HIV vaccine just passed; it is doubtful a safe and efficacious one will be available even 10 years from now.

Because of these reasons, I believe it is still reasonable to consider HIV gene therapy, even in the height of the HAART era, but the window is gradually closing. Even before David Baltimore‘s infamous 1988 Nature News and Views entitled ‘intracellular immunization‘ [26], investigators were already thinking about potential gene therapeutic targets for HIV. Over the years, there has been no shortage of transgenes and vectors that have been shown to block HIV replication in cultured cells. Few, however, have ever been rigorously tested or applied in a clinical setting (and none have demonstrated good efficacy). One might say that researchers approached the problem with an almost irrational exuberance and keen naivety, and transpired time has tempered those traits. Of course, we now know much more about HIV and requisite host factors; we also know that it is possible to cure immunodeficient individuals using retrovirally gene-modified hematopoietic stem cells. We now have much better (and presumably safer) vectors at hand.

‘…one would like to employ a transgene that protects a cell from ever getting infected. That cell should have a selective advantage compared with infected cells and be able to repopulate the individual over time.’

Unfortunately, the results of the few Phase I clinical HIV gene-therapy trials have not been very impressive. In several, the transdominant Rev M10 protein was the transgene (in one case introduced as naked DNA into peripheral T cells; in another introduced as part of a retroviral vector). After several months, the level of ‘marking‘ was low and there was no clear effect on CD4⁺ T-cell numbers or viral loads [27,28]. More recently, an HIV vector with all cis-acting sequences carrying antisense envelope was used to transduce peripheral T cells of five advanced patients [29]. These T cells were expanded ex vivo prior to reinfusion. Again, level of marking was low at 1 year (generally less than one in 10,000 peripheral blood mononuclear cells), but for one patient there was a 2.0 log₁₀ drop in viral load associated with an increase in CD4⁺ T-cell count of 100. At the time, this patient was on a complex antiretroviral regimen that included both a protease inhibitor and a non-nucleoside reverse transcriptase inhibitor, and his viral load had decreased by at least 0.5 log₁₀ prior to infusion; thus, the further decrease in viral load and increase in CD4⁺ count may be simply coincident with T-cell reinfusion. No adverse events were reported, and privately financed Phase II studies are planned.

Neither Rev M10 or antisense-envelope transgene is truly protective in so much as expressing cells may still get infected with HIV. In both cases, one would expect reduced virus production from such cells, but that may not prevent that particular cell from dying. Perhaps a saving grace of the HIV vector encoding antisense envelope is that it should get packaged and co-spread with HIV (although that was not evident from the data presented) [29]. Once enough cells get transduced by this mobilization process, overall viral replication, as measured in plasma, may be curtailed. Depending upon the transgene and host-range of the vector, mobilization may be dangerous to both the individual and others, and avoidance is preferred.

More typically, one would like to employ a transgene that protects a cell from ever getting infected. That cell should have a selective advantage compared with infected cells and be able to repopulate the individual over time. This is critical since unless myeloablative chemotherapy is being used, the gene-modified cells will be in the distinct minority (less than 1% of the total), even if bone marrow-derived CD34⁺ cells are transduced, expanded and reinfused. Such a transgene might be an shRNA against CCR5.

CCR5 is an essential coreceptor for HIV replication. After gp120 binds CD4 and undergoes a conformational change, it recognizes as co-receptor CXCR4 or CCR5, and then fuses with the cell membrane. CCR5 is an attractive target since it is mainly nonessential in man (individuals who have the Δ32 version appear phenotypically and immunologically normal) [30]. We now know that small molecule noncompetitive antagonists of coreceptor function are capable of reducing viral loads and increasing CD4⁺ T-cell counts [31–33]. Maraviroc has only recently been approved by the FDA (and vicriviroc will not be far behind). shRNAs against CCR5 have shown promise in vitro in reducing surface expression of CCR5 and protecting transduced cells against lentiviral infection [34] (even after transduction of rhesus macaque hematopoietic stem cells [35]).

‘CCR5 is an essential coreceptor for HIV replication.’

Perhaps buoyed by the success of maraviroc, John Rossi and colleagues at City of Hope Medical Center (Duarte, CA, USA) are proceeding with a Phase I trial that involves patients with non-Hodgkin‘s lymphoma who are also receiving autologous bone marrow transplants after myeloablation [36,101,102]. CD34⁺ cells will be transduced with an HIV-based vector that encodes an anti-CCR5 ribozyme, an siRNA directed against tat and rev and a TAR decoy. Because of reduced levels of surface expression of CCR5, these cells should be at a selective advantage compared with nontransduced cells. Although this is just a safety trial, success will be partially based upon percentage marking, reduction in viral loads and increases in peripheral CD4⁺ T-cell counts.

It is true that the short history of human gene therapy has been a series of ups and downs, perhaps even more so than that found for other fields because of enhanced public interest and scientific scrutiny. Naturally, we are all intrigued with the idea that we might be able to gene-modify cells to make them resistant to viruses. In the past, most of the work has been performed by individual academic researchers or small teams, each of whom has his own favorite gene, vector and approach, with peer reviewed but relatively limited funding from the NIH. Perhaps it is time to bring this to a larger and more open arena, folding gene therapy into one of the HIV treatment/evaluation networks. That way an investigator (or several investigators) may bring a concept to an experienced committee or group of individuals, which may then debate the merits of the transgene, the vector, the target cells, the intended patient population and the end points of the trial (with some back and forth discussion). I would argue at this juncture that this approach will be far superior to individual or small teams of investigators hammering away, in relative isolation, on this very important and potentially soluble problem. The door is slowly swinging shut on viral gene therapy, and if substantial progress is not made within the next 5–10 years, it may be closed forever.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.