Second-Generation Vector Technology
Unfortunately, the first-generation vector technologies deployed in these clinical successes still had the limitations described in this article and will be of limited value beyond these test cases. The good news is that the field has been quietly developing the technology of the future, which is now ready for prime time.
One of the most important limitations of vectors based on murine retroviruses was the need for the target cell to be dividing at the time of exposure to the vector for transduction to occur. This complicates ex vivo applications of stem cells, which often persist in a quiescent state. Furthermore, attempts to induce stem cell division to render them susceptible to retrovirus transduction may irreversibly alter their phenotype essentially losing the capacity for self renewal.
Development of vectors based on the family of retroviruses called lentiviruses represented a major advance in overcoming this limitation. Lentiviruses are pathogens capable of infecting non-dividing cells; the most notorious is HIV.
Investigators at the Salk Institute developed replication-defective vectors based on HIV, which demonstrated efficient stable transduction in nondividing cell targets.15
The first clinical application of lentiviral vectors for a genetic disease deployed an ex vivo bone marrow strategy in a lethal pediatric neurologic disease called adrenoleukodystrophy (ALD). The difference between ALD and the previous work with SCID is that there is no selective advantage of genetically corrected HSCs in ALD as was the case with SCID.
Successful reconstitution of the ALD patient with corrected cells would require very high transduction efficiency ex vivo and some form of ablation of the recipient’s marrow to make space for the corrected cells. The pilot human study was a dramatic scientific and clinical success.16 Patients were repopulated with a high frequency of transduced cells in virtually all hematopoietic lineages that stabilized their disease.
The other technological advance was in the development of second-generation AAV vectors for in vivo gene therapy. We hypothesized that natural variation in the AAV capsid that would occur during natural infections may translate into improved vector performance, albeit in a stochastic manner. We sought to identify AAV capsids from viruses that circulate in primates as infectious agents.
The problem was identifying primates with active AAV infections to obtain samples for virus recovery since the known AAV serotypes were isolated from laboratory stocks of adenoviruses; these AAVs could have been passengers from the initial adenovirus isolations or come from tissue culture components such as bovine serum.
Serologic studies in primates, however, showed antibodies against some of the known AAVs suggesting they circulate in these populations. Based on previous in vitro work we speculated that AAV may remain as a latent genome following resolution of the infection as is the case with the Herpes simplex virus.
We tested this hypothesis by analyzing DNA from primate sources using PCR with primers to conserved regions of the known AAV capsids. The results were startling—latent AAV genomes are widely disseminated through many tissues from a wide array of primates, including macaques, great apes, and humans.17,18,19 Sequencing of these genomes demonstrated a marvelous diversity of structure.
Vectors were created from many of these novel capsids and shown in animal models to be substantially improved over vectors based on the previously known AAVs 1–6. For example, AAV8 shows substantially higher transduction in a number of highly relevant target cells such as liver, muscle, and photoreceptors of the retina.
A clinical trial of liver-directed gene transfer with AAV8 in subjects with hemophilia showed partial correction of the clotting defect that has replaced the need for some protein replacement and has been stable.20
AAV9 is a capsid that is capable of transferring its genome across vascular and blood brain barriers and has been shown in preclinical models to treat a variety of cardiac and neurological disorders.21
Another advantage of the novel AAV isolates we recently discovered, such as AAV8 and AAV9 noted above, is that they present diminished problems of host immunity. They show much lower levels of pre-existing immunity in humans which, if present, can diminish efficacy and are less likely to activate destructive T-cell responses.