Christina Bennett Freelance Writer GEN
A Large Portion of Patients Have Pre-Existing Antibodies Against AAV
Gene therapies offer the promise of a one-time cure for genetic diseases, but they have to make it past the immune system first. For gene therapies that are infused directly into the body, that is, in vivo gene therapies, a large fraction of people have immune systems that are primed with pre-existing neutralizing antibodies to gene therapies in clinical development. The exact proportion of people depends on the design and material makeup of the gene therapy, but if present, these antibodies neutralize the therapy and any transduced cells, rendering the treatment ineffective.
With the latest finding that many people could have pre-existing antibodies to popular gene-snipping tool CRISPR/Cas9, the pool of treatable patients shrinks further, putting more pressure on researchers and biotech companies to find solutions. (The study was published on the preprint server bioRxiv and has not been peer-reviewed.)
The reason people commonly have pre-existing antibodies to a gene therapy is because the vector used to ferry the therapeutic cargo is often of viral origin. Viral vectors include oncoretroviruses, lentiviruses, adenoviruses, adeno-associated viruses (AAV), and herpes viruses. For genetic diseases, one vector has taken the lead.
Right now, AAV is “by far” the leading vector for in vivo gene therapies designed to correct a genetic mutation, says Thomas Weber, Ph.D., associate professor at Icahn School of Medicine at Mount Sinai.
AAV is a nonpathogenic virus, making it an attractive vector for clinical use, and has many serotypes, such as AAV2, AAV5, and AAV9. As of November 2017, 204 clinical trials worldwide have used AAV vectors, according to data from Gene Therapy Clinical Trials Worldwide. The first in vivo gene therapy approved by the FDA (called Luxturna [(voretigene neparvovec-rzyl] and made by Spark Therapeutics) uses AAV as the delivery vehicle. It is administered into the eye to treat a rare inherited form of blindness.
“Most of us at some point get infected by AAV because AAV is a nonpathogenic virus,” Dr. Weber explains to GEN. He notes that pre-existing neutralizing antibodies to AAV vector “is a considerable problem in the field.”
“The issue is longstanding,” says Kenneth Berns, M.D., Ph.D., Distinguished Professor Emeritus of Molecular Genetics and Microbiology at the University of Florida. “We’ve known for who knows how long, that when you get a viral infection, you develop immunity and develop antibodies against the virus. The problem from the point-of-view of gene therapy is if you’re going to use a viral vector to bring in your transgene, then it’s important that the patient that you’re treating not have significant antibodies against the viral vector.”
He continues, for instance, if you give somebody a drug made wit] an AAV9 vector and the person has pre-existing antibodies, “the antibodies will simply neutralize the vector, and so [the therapy] won’t work at all.”
Shrinking Patient Pool
Clinical trials are being designed with this in mind, with investigators excluding patients for whom the therapy won’t work. For example, patients were excluded from the AveXis Phase I spinal muscular atrophy clinical trial if they had pre-existing antibody titer levels to AAV9 above a certain threshold. Similarly, for the BioMarin Phase I/II clinical trial for hemophilia A (that showed promise at the American Society of Hematology conference in December 2017), patients were excluded from the trial if they had pre-existing antibodies to the AAV capsid. Results for the hemophilia A study and spinal muscular atrophy study were published in the New England Journal of Medicine in December 2017.
“Across serotypes, if you average it out, about half the people have neutralizing antibodies against at least some of the [AAV] serotypes,” says Dr. Weber.
Margaret Ragni, M.D., M.P.H., professor of medicine in the division hematology/oncology at the University of Pittsburgh, agreed. She also cited about half of adults as having antibodies to AAV and not being eligible for AAV-based gene therapy.
“And that, of course, is a big medical problem. If you can only treat some people, and not others, it’s good for the people you can treat, but kind of leaving others out on a limb,” says Dr. Berns.
Upon exposure to AAV, whether naturally or via gene therapy administration, antibodies are generated against the specific serotype. However, as Dr. Ragni explains, “If you develop an antibody to one [serotype], it may also have cross-reactivity with other serotypes. They’re very similar.” Therefore, patients may be ineligible for any AAV-based therapy if that have antibodies against any AAV vector serotype, even if it’s not the serotype used in the therapy.
Dr. Ragni notes that the practice of excluding patients with pre-existing antibodies to AAV in the Phase I/II hemophilia A trial would also be done in a physician’s office, should the therapy be approved by the FDA. “At this time, we would have to,” she says.
She also explains another reason the pool of treatable patients with hemophilia A patients shrinks: Patients can develop alloantibodies against factor VIII as a result of factor VIII infusions. Hemophilia A patients are deficient in factor VIII, so factor VIII infusions are given as treatment.
“Presumably, if that patient received a vector that expressed factor VIII, they would have that same neutralizing effect on the expression of that gene therapy. That’s another big question that’s under scientific scrutiny,” says Dr. Ragni. “Thirty percent of patients with severe hemophilia A develop these neutralizing antibodies that we call inhibitors, so for that thirty percent, they can’t get gene therapy either, but that’s something I truly believe in the future will resolve.”
Similarly, for certain gene therapies, the transgene that is carried in the vector can also trigger a humoral immune response, rendering a gene therapy ineffective.
Dr. Weber explains that this can happen for null genotypes because the protein that’s being made by the transduced cells has never been seen before by the immune system. Even though it’s an endogenous protein, that particular person’s immune system never learned that the protein is “self,” so it attacks and destroys any transduced cells. However, if the genetic disease is the result of a mutation that produces a nonfunctional protein, immune response is usually not an issue, says Weber.
Gene Editing: Getting Ahead of the Problem
CRISPR is slated to enter clinical trials in the United States this year, and unlike gene-replacement therapies, which have been in clinical development for decades, gene-editing therapies are early in clinical development. The first person to receive a gene-editing therapy received it in 2017, and no one in the United States has received any therapies modified with CRISPR yet. CRISPR and other gene editing therapies may be able to get ahead of the problem thanks to a group at Stanford University who published a study on the preprint server bioRxiv on January 5, 2017. The study suggests a preexisting humoral immune response to Cas9 is common.
“The possibility that Cas9 would elicit an immune response has been discussed and has been shown in animal models,” says senior author Matthew Porteus, M.D., Ph.D., associate professor of pediatrics at Stanford University. Dr. Porteus is one of the founders of CRISPR Therapeutics. He says “people weren’t really following up on” the possibility of an immune response to Cas9. “At some point I said, ‘well, it doesn’t appear that anyone else is going to test it, so I better put my money where my mouth is,’ and test it myself in my own lab.”
Dr. Porteus says that the research for this study was all done in his academic lab and CRISPR Therapeutics had no input. He says the paper was submitted to a peer-reviewed journal a few hours before posting on bioRxiv, and he plans to have the paper published in a peer-reviewed journal. He explains that the decision to publish a preprint was because the “peer-review process can take months or longer” and they “wanted to make sure that this finding was out there so people could discuss, consider, study further, and evaluate.”
In line with suspicions, the Stanford researchers found antibodies against Cas9 protein derived from Staphylococcus aureus and Streptococcus pyogenes, which are the two most widely used homologues for Cas9 protein. The main finding was that in a group of 12 healthy adults, 79% had antibodies against Cas9 from S. aureus and 65% had antibodies against S. pyogenes.
“What we as a field do not want to have happen is to go through what the gene therapy field went through, which was to find that the human immune system was a real barrier to translating safe and effective gene therapies, and it led to an important setback,” says Dr. Porteus. “It’s been great that the gene therapy field has learned from those [setbacks], and now we’re seeing a lot of successes, but hopefully, the genome editing field won’t have to go through that painful process.”
Not commenting specifically on the study, Dr. Weber explains why CRISPR/Cas9 would be expected to elicit an immune response.
“It’s the same as if it’s a virus. So, any protein that is expressed in a cell, including self-proteins, they’re degraded to some extent and they’re presented on the surface of the cell on major histocompatibility complex (MHC) class I proteins,” he says. T cells and antigen-presenting cells, but “especially” T cells, will recognize this MHC peptide complex, Dr. Weber asserts. If MHC class I displays self-peptides, the immune system ignores it, but if the peptide is foreign, like a Cas9 peptide, then the immune system will attack the cell.
“There will be Cas9 peptide presented on class I MHC. There’s no question about it,” says Dr. Weber. He says this logic also applies to other gene-editing technologies of bacterial origin, noting that zinc finger nucleases have a bacterial component.
Only One Shot
Another caveat to gene therapies is a person can receive a particular gene therapy only once. The bacterial or viral makeup of the therapy primes the immune system to attack next time it sees that same therapy, or even a different therapy that uses the same bacterial or viral component.
People who may need to have more than one infusion with a gene therapy include clinical trial participants who received an experimental therapy that didn’t work and people who have a genetic disease that requires more than one infusion. Genetic diseases that may require multiple infusions are those that involve cells that repopulate more often.
“I’m skeptical that we will be able to really efficiently do that,” says Dr. Weber, referring to the ability to administer a gene therapy more than once. He notes that this issue will depend on a number of factors, like where the therapy is administered in the body and how quickly the target cells turn over.
“For instance, cystic fibrosis is obviously a very important target, but there you have lung epithelial cells that turn over much more rapidly than hepatocytes,” says Dr. Weber.
Although pre-existing antibodies are a problem, a variety of solutions have been proposed.
“There are a number of approaches that people are trying,” says Dr. Weber. He stresses that multiple approaches will likely be needed and adds, “I don’t know if you can ever completely solve the problem.”
Possible solutions to overcoming pre-existing antibodies that are under investigation include plasmapheresis to remove antibodies from the blood, increasing therapy dosage, suppressing the immune system, and administering earlier in life.
Also, it may be possible to use synthetic vectors. It also may be useful to switch or modify the bacterial or viral component in an existing vector. For example, Dr. Porteus says it may be possible to design a Cas9 system from a bacteria that doesn’t infect humans, making it unlikely that humans would have immunity to it.
One solution that circumvents the immune system is removing the target stem cells from the body (such as liver stem cells). The stem cells can be transduced with the therapy ex vivo and then administered back into the body.
Despite the challenges of current gene therapies, Dr. Weber is optimistic.
“Gene therapy in general, but especially AAV[-based] gene therapy, has come of age,” he says. “If you look at Big Pharma, they are rushing back into the field. I think [the field is] exploding and it’s exploding for a reason.”