Stealth gene therapy

In this issue of Blood, Martino et al¹  report on a novel adeno-associated viral (AAV) vector, which overcomes one of the last remaining impediments for liver gene therapy, the anticapsid immune response.

Just over 7 years ago, the first successful gene therapy for hemophilia appeared to be at hand, until an unexpected immune response against the vector capsid led to clearance of corrected liver cells and a loss of the replacement factor IX gene.²  Since that trial, much has been learned, and in 2011, it was reported that intravenous injection of an AAV8 vector encoding factor IX was able to achieve sustained factor IX expression in 6 individuals with hemophilia B.³  This stunning achievement, long-lasting correction of a genetic disease from a single drug injection, is nearly unprecedented in medicine and will likely change the course of hemophilia therapy.⁴  However, although the trial was largely successful, the anticapsid immune response did occur at the most corrective vector dose, and transient immunosuppression with prednisolone was used to prevent the immune system from eliminating hepatocytes harboring the replacement factor IX gene. Thus, improved strategies are still needed to circumvent the anticapsid immune response before AAV can become an off-the-shelf drug for hemophilia B. In Martino et al, High, Herzog, Mingozzi, and colleagues, who pioneered the use of AAV for hemophilia gene therapy,^2,5,6  describe an innovative modification to the AAV capsid that demonstrates the potential to avoid immune-mediated clearance.

One of the challenges of studying the anti-AAV capsid response has been the lack of experimental systems that model the outcome in humans. More than a decade of AAV research in rodents and dogs failed to elicit the anticapsid response that was ultimately observed in humans. Thus, before testing whether they could generate an AAV that would evade the anticapsid response, Martino et al developed a model that would mimic the response in humans. To do this, they immunized mice with a known immune epitope from the AAV2 capsid, isolated CD8+ T cells, and expanded them ex vivo by repeat stimulation with the antigen. This established a pool of CD8+ T cells with specificity against the AAV capsid. When these anticapsid T cells were transferred into mice that were injected 24 hours prior with an AAV2 vector expressing factor IX, they killed the hepatocytes harboring the vector, and this resulted in diminished factor IX expression and an elevation in transaminases in the serum. This was similar to the outcome observed in the earlier human trial of AAV2 and in 1 of the patients in the recent AAV8 trial.^2,3 

With a suitable model established for studying the anticapsid response, the authors evaluated the immune evasive potential of a novel AAV2 vector variant they previously generated,⁷  which harbors mutations in 3 different tyrosine residues that are normally exposed on the vector’s surface. In contrast to what was observed with the wild-type AAV2 vector, when they transferred anticapsid CD8+ T cells into mice injected with the AAV2 variant vector, there was no transaminitis, and factor IX expression was similar to the levels in mice that did not receive the anticapsid T cells. This is a promising achievement because it was performed in a system that models some of the hallmarks of the patients’ response to therapy. It is important to note that the anticapsid response that occurred in patients was subdued by transient immunosuppression.³  A therapy that does not require any immune modulation would obviously be preferable, but clinical adoption of their AAV2 variant will initially be warranted mostly in patients where immunosuppression is contraindicated. It also remains to be determined whether the tyrosine mutations can be introduced into the capsids of other AAV vector serotypes and improve immune evasion without impacting the efficiency of gene transfer.

The success of Nathwani et al in using gene therapy to establish long-term expression of factor IX in patients with hemophilia B is among a string of recent successes in human gene therapy, which have restored vision to the sightless and released immunodeficient patients from isolation.^8-10  There are still obstacles for gene therapy before it will become a routine treatment, but studies such as that by Martino et al are an excellent example of how bench to bedside and back to bench can help to overcome these obstacles.