Skip to Content

Safer Gene Therapy

A gene therapy pioneer explains how new delivery mechanisms could help revive the once-promising technology.
April 26, 2006

In 1999, gene therapy – once touted as one of biomedicine’s most promising fields – suffered a major and highly publicized setback. Jesse Gelsinger, an 18-year-old participating in a clinical trial testing treatment for a rare immune disorder, suffered a fatal immune reaction to the virus used to deliver the gene. The field has been slow to recover from this early tragedy. But now clinical trials of gene therapy are taking place worldwide, mainly for cancer and diseases linked to a single known gene.

Gene therapy has enormous potential to treat disease – if scientists can find safe delivery technologies. In the therapy, a copy of a gene is delivered to a specific tissue to correct an abnormality – for example, the lungs in cystic fibrosis. The gene is often delivered with viruses, which have evolved the ability to deliver their genes into human cells. The danger is that the virus can trigger a vigorous immune response in the patient, as happened with Gelsinger.

James Wilson, a scientist at the University of Pennsylvania, who headed the trial in which Gelsinger died, has spent the last seven years searching for safer and more effective ways to deliver therapeutic genes. He’s taking advantage of the natural diversity of viruses to find candidates that have evolved the optimal characteristics for gene therapy. He tells Technology Review about some promising options.

Technology Review: Why do we need better delivery viruses, commonly known as vectors?

James Wilson: We recognized five or six years ago that the technology we had was wholly inadequate for what we wanted to achieve in the clinic. Back in the ’80s we used the vectors that were available off the shelf, not vectors that were perfected for use in gene therapy. But what we found was that they were not sufficiently efficient and/or caused an inflammatory or immune response. For many applications, that meant these vectors were either toxic or ineffective.

TR: What are the most important qualities in a delivery vector?

JW: First of all, when injected in vivo [into a live animal or person] they must target cells with high efficiency. Second, they must be easy to manufacture. And third, these vectors must evade the immune system.

TR: Do you have any vectors that possess these qualities?

JW: We initially worked a lot with adenoviruses [the virus behind the common cold], which were first brought into humans to treat cystic fibrosis. But they were found to be very immunogenic, meaning they caused severe inflammation.

For the last five years, we have studied adenovirus-associated viral (AAV) vectors. They seem to not cause inflammation – they evade immune detection. These vectors also seem to be very stable. For example, we injected monkeys with an AAV vector eight years ago – and gene expression has gone down only about three-fold since then. [This long-term expression is significant because gene expression of many types of vectors disappears over time.]

TR: Why can AAV vectors evade the immune system, while adenovirus vectors trigger immune reactions?

JW: AAV vectors do two things: they can get into a liver or heart cell without activating a type of T cell involved in the immune response, sort of like a stealth plane. These vectors can also activate a type of T cell that suppresses the immune response. We don’t know why AAV vectors do this or why they are in such contrast to the adenovirus vector. Maybe this is how the native virus evolved to escape immune detection.

TR: Do AAV vectors have a downside?

JW: We found that the prototype vector, known as AAV2, was not efficient enough. It could get into a few cells, but it was not efficient enough to have a therapeutic effect.

TR: So how do you make these vectors more efficient?

JW: There are two possibilities. One is to reengineer the vector, but it’s very difficult to reengineer a biological organism that has evolved over time. A virus is not like a small molecule drug such as aspirin. It’s a biological particle, so it’s much more difficult to tinker with.

Fortunately, evolution has generated a diversity of viruses. We screened monkeys and humans for lingering adenovirus-associated viral infection. Adenovirus-associated viruses infect humans and primates. No one knows what the virus does, what the infection looks like or whether it hurts or helps you. We discovered that 40 percent of human livers have persistent infections, and we identified over 100 new subtypes.

Now we are looking at the properties of the vectors and how well they can be transferred to different organs. We found that a variation of a vector called AAV9 can efficiently transfer genes to the heart.

TR: Have AAV vectors been tested in human trials? How safe are they?

JW: Yes. AAV2 has been tested for cystic fibrosis, muscular dystrophy, neurological disease, and hemophilia. Two patients in the hemophilia trial developed liver inflammation, although they did recover. Other than that, there have been no safety issues.

Since then, we’ve tried to determine if the new AAV vectors will have the same response. We don’t think they will – we think we’ve figured out what happened in those patients and how to get around it. [Wilson has a paper on this topic currently under review at a scientific journal.]

TR: Are the new vectors being tests in clinical trials?

JW: Penn [the University of Pennsylvania] licensed the vectors to GlaxoSmithKline [a pharmaceutical company headquartered in the United Kingdom.] They are pursuing various applications, such as treatments for lung, heart, and liver disease, although nothing has yet been tested in humans.

TR: How will scientists decide which vectors are the best to move forward into clinical trials?

JW: That is something we still need to decide. Will large animal studies be required? These studies are very resource intensive. We need to identify good animal models that will predict efficacy in humans.

Keep Reading

Most Popular

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

It’s time to retire the term “user”

The proliferation of AI means we need a new word.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Sam Altman says helpful agents are poised to become AI’s killer function

Open AI’s CEO says we won’t need new hardware or lots more training data to get there.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.