This article is reproduced with the permission of New Scientist for exclusive use by Nova users.

Genetic road to super-immunity
24 February 2007
From New Scientist Print Edition.
Dan Jones

Enlarge The power of immunotherapy

Mark Origer of Watertown, Wisconsin, should be dead. In 2004 it became clear that conventional treatments for his skin cancer had failed. The disease had spread to his liver, and it wasn't clear whether he'd live long enough to make it to his daughter's wedding the following year. Yet he not only made it to the wedding, he is still alive today. His tumours are gone, and he appears to be free of cancer.

What saved Origer was genetic engineering of his immune system. Cells were taken from his body, given a gene that programmed them to attack melanoma cells and then re-implanted. The modified cells survived and thrived, and slowly destroyed his tumours.

Others were not so fortunate. Of the 17 patients in the pilot trial, only Origer and one other man responded to the treatment. But their survival proves the approach can sometimes work wonders - and soon it should help even more people with cancer. "We'll be able to improve on these results dramatically," says Steven Rosenberg, chief of surgery at the National Cancer Institute in Bethesda, Maryland, who led the trial.

What's more, this new form of gene therapy isn't limited to treating melanoma or other cancers. Genetic engineering of the immune system could also cure deadly infectious diseases such as HIV, tuberculosis and malaria, and reverse autoimmune disorders such as juvenile diabetes, multiple sclerosis and arthritis.

Our immune systems are incredibly effective, but they are not perfect. Sometimes they fail to recognise enemies such as tumour cells, or mistakenly attack healthy cells, while many pathogens have evolved cunning ways to dodge immune attack or replicate so fast they simply overwhelm unprepared defences.

So ever since Turkish or Arab physicians developed a form of inoculation for smallpox many centuries ago, doctors have been giving the immune system a helping hand. Vaccines have saved hundreds of millions of lives, and the latest generation promises to do everything from preventing cervical cancer to helping treat cocaine addiction. Yet conventional vaccination has its limits, as decades of failed attempts to develop vaccines against diseases such as malaria and HIV show all too clearly.

Enter the genetic engineers, who are taking the concept of helping the immune system to an entirely new level. This approach can be used to target an immune attack far more precisely than with a normal vaccine, or to tell the immune system when to call off an inappropriate attack. What's more, there is no need to rely on the natural powers of the immune system - cells can be given entirely new abilities and weapons for fighting disease.

Take cancer. As the immune system develops, it learns to ignore the natural components of the body - to distinguish between self and non-self. Because cancer cells derive from our own cells, they can grow and spread underneath the radar of immune surveillance. Our immune systems can mount a response to mutated or aberrantly expressed proteins on tumour cells and probably do kill off many cancers at an early stage, but all too many slip through the net.

Numerous groups around the world are working on cancer vaccines, but overcoming the body's tolerance of self and persuading the immune system to mount an all-out attack on established tumours is proving far from easy. The results of most trials have been disappointing.

Directly altering immune cells instead could overcome many of the problems. The approach being taken by Rosenberg's group and others, for instance, is to modify cytotoxic T-cells, those that seek out and destroy bacteria or cells infected with viruses. The key to their specificity is a kind of protein called a T-cell receptor that sticks out from their surface. In response to an attack, the immune system generates new sets of T-cells carrying receptors that latch onto the invaders. Add the right T-cell receptor to cytotoxic cells and you can make them target any cell type you want.

In the pilot trial, Rosenberg's team modified patients' cytotoxic T-cells to target MART-1, a protein common on the surface of melanoma cells. First, the researchers took the gene for a T-cell receptor that binds to MART-1 from the cytotoxic T-cells of a melanoma patient whose immune system was targeting the cancer. Next, they extracted cytotoxic T-cells from the blood of the 17 patients and used a retrovirus to add the gene to each patient's cells. Drugs that deplete the immune system were then given to the patients to "make room" for the modified cells before they were injected back into the body - without this step, too few of the cells survive.

In 15 patients the engineered cells were still circulating at high levels two months later, proving that you can remodel the immune system to some extent (Science, DOI: 10.1126/science.1129003). While it is disappointing that tumours regressed in only two patients, Rosenberg thinks he knows why: the T-cell receptor used in the study does not bind very strongly to MART-1. "We now have much more potent receptors," he says.

The results are nevertheless generating great excitement in the field. "This pilot study marks a milestone in tumour immunological research," says Rienk Offringa of the tumour immunology group at Leiden University in the Netherlands. "The clinical impact might be limited, but modifications to increase efficacy can readily be envisioned."

Rosenberg already has his sights set on other cancers. The team is now targeting a protein called p53. This normally protects us against cancer by triggering cell suicide, but mutations can render it ineffective. Around half of all tumours produce high levels of mutant p53, so in theory getting T-cells to target p53 should work against a wide range of cancers. "We've already treated our first patients, and we expect to get results within a year," says Rosenberg. Other groups are tackling cancers such as leukaemia, and T-cell receptor gene therapy could also be used to target infectious agents, such as the Epstein-Barr virus.

Besides cytotoxic T-cells, antibodies are the other main weapons of the immune system. These free-floating proteins bind to specific viral or bacterial proteins, disabling the invaders or labelling them for destruction. David Baltimore's team at the California Institute of Technology, Pasadena, is planning to take antibody-producing B-cells from people with HIV, genetically engineer them to produce potent anti-HIV antibodies and return the modified cells to the body. "With conventional vaccines, you're limited to what the immune system naturally produces," says Pamela Bjorkman, a protein expert at Caltech who is working with Baltimore, "but we can redesign antibodies so that they work more effectively."

With the help of a $14 million grant from the Bill & Melinda Gates Foundation, Baltimore's lab is now refining the methods for adding antibody genes to B-cells. Meanwhile, Bjorkman's lab is grappling with the formidable problem of designing new anti-HIV antibodies, as well as antibody-like proteins that are smaller and better able to access the binding sites on HIV. "Right now we're running tests to see which work best against different HIV strains," says Bjorkman.

She is confident she can create proteins that will remain effective against HIV, despite its high mutation rate: "The virus has evolved to thwart an immune system it knows, but it doesn't know about the protein design tricks that we have up our sleeve to tackle it." Whereas natural antibodies have two identical binding sites for latching onto their target, artificial antibodies can have four different binding sites, minimising the likelihood that HIV will able to outwit them. Antibodies also interact with other parts of the immune system and can be modified to send out a stronger "alarm call".

Although the technology is some way short of being applied to humans, animal studies have been promising. And the same approach could help to tackle other hard-to-treat infections. "It could be applied to other diseases where there is enough time to intervene this way and there is a need for a new style of therapeutic. Malaria and tuberculosis come to mind, as does hepatitis. There may even be opportunities for treating animal diseases," says Baltimore.

Tackling cancer and infectious diseases involves boosting the firepower of the immune system. Other groups are instead working on ways to prevent friendly fire - the autoimmune diseases caused when the immune system attacks our own tissues. One of the diseases researchers are focusing on is juvenile (type 1) diabetes, caused by the immune system destroying insulin-producing beta cells in the pancreas.

Potential treatments for diabetes often hit the headlines, but almost all involve ways of replacing the insulin-producing cells - which is treating the symptom rather than the cause. Halting the immune attack would ensure the replacement cells survive and might even allow the beta cells to regenerate naturally if people are treated early enough.

Stopping friendly fire

One strategy is to modify cells to produce immune signalling molecules that damp down autoimmune reactions. Studies of twins suggest that the lack of a signalling molecule called interleukin-4 contributes to diabetes, and mice with no IL-4 get diabetes. So C. Garrison Fathman, an immunologist at Stanford School of Medicine, California, took immune-regulating cells called dendritic cells from these mice and engineered them to express IL-4 before injecting them back into mice. The hope was that the engineered cells would deliver IL-4 to the pancreatic lymph node, the site from which most attacks on beta cells originate. It worked: for reasons no one understands, the modified cells actually homed in on the pancreatic lymph nodes, preventing diabetes in most of the mice. Now Fathman's team plans to find out if the approach will work for people too.

Other groups are working on more specific ways of producing immunosuppressing molecules where they are needed. One technique is to exploit T-cell receptors: add a gene for an appropriate T-cell receptor along with the gene for the necessary signalling molecule, and the modified cell will home in on a specific tissue, such as cartilage. Animal studies at the University of Tokyo show this approach could be used to treat arthritis.

Perhaps the most promising strategy for tackling autoimmune diseases, though, is to intervene higher up the chain, to get the generals to call off the attack rather than disarming the foot soldiers. Several groups have managed to induce tolerance to a specific protein in animals using a variety of approaches, such as engineering B-cells to produce this protein. Such methods could one day be used to induce tolerance to organ transplants as well as for curing autoimmune diseases.

While most immune gene therapies merely redirect the immune system, a few researchers plan to enhance it. Some of the nastiest bugs around have evolved sophisticated ways of evading immune attack. HIV, for instance, not only changes its coat, it also attacks the immune system itself, slowly destroying the body's ability to defend itself.

John Rossi of the Beckman Research Institute of the City of Hope in Duarte, California, is planning to make the immune system immune to HIV. He wants to take the blood stem cells that give rise to the cells of the immune system from patients, add three genes to them to protect them from HIV and then replace the cells.

The first gene codes for an RNA molecule known as a ribozyme that blocks the expression of the human gene for a cell surface receptor called CCR5, which HIV uses to invade cells. The second produces a short interfering RNA molecule that should trigger the destruction of any viral RNA that still manages to enter the cell. The final gene expresses a "decoy" RNA molecule that should prevent HIV replicating and infecting other cells if it manages to outflank the first two defences. "We're at the point of filing an Investigational New Drug Application with the FDA, which will allow us start clinical trials," says Rossi.

It's not just bacteria and viruses that evolve ways to outsmart the immune system. Some cancers do too. Certain tumours survive by pumping out an immunosuppressant called TGF-beta, for instance, blocking any attack by immune cells. So Chung Lee's team at Northwestern University in Chicago has developed a way to modify immune cells to ignore TGF-beta. The idea is to isolate cytotoxic T-cells, persuade them to target tumour cells and also to modify them so they are insensitive to TGF-beta. Mouse studies show that these modified cells launch a far more effective attack on prostate tumours than those that respond to TGF-beta.

While the potential benefits of this new kind of immunotherapy are huge, there are of course some big stumbling blocks. "Safety will be a challenge because there have been issues with gene therapy," says Baltimore. In a French trial, gene therapy triggered cancer in three children (New Scientist, 15 March 2003, p 6). This risk might be acceptable to those with a life-threatening illness such as melanoma, but could slow the development of treatments for less serious conditions.

Mucking around with the immune system is also inherently risky. Cancer immunotherapy in particular can trigger a dangerous autoimmunity. "You want a vigorous immune response and maximal efficacy, but you want it to be safe - that is the big challenge," says Offringa.

And while genetic engineering of the immune system could potentially cure diseases that plague poor countries, such as HIV and malaria, gene therapy-based treatments will not come cheap. Baltimore is keenly aware of this problem. "We're trying to design [gene-delivery systems] that will be cheap to make, and we're designing methodologies that will be cheap," he says. "We want to develop these therapies for countries that can't afford expensive drugs, a point the Gates Foundation insists on, and one which I wholeheartedly support."

Rossi points out that conventional HIV therapies cost around $25,000 a year in the US, or $250,000 a decade. "The costs of our therapy for HIV are likely to be lower than current drug regimes because this will be a one-shot deal," he says. "Our approach could cost under $50,000."

The same logic applies to cancer. For instance, treatments for colorectal cancer - which only extend life by up to 12 months - cost $180,000 annually. Rosenberg predicts that cell-based immunotherapy will cost much less.

There is little point in worrying about the price at this stage, he says. "Right now we need to find things that work, and then we'll work out how to make it economically feasible."

Dan Jones is a freelance science writer based in Brighton, UK. See http://psom.blogspot.com

From issue 2592 of New Scientist magazine, 24 February 2007, page 42-45

Send in the nanoparticles While genetically engineering immune cells has huge promise, it is also risky. Some groups are exploring ways to "talk to" immune cells without necessarily altering their genes. Darrell Irvine at the Massachusetts Institute of Technology hopes to achieve this with sophisticated nanoparticles that could travel through the body and alter immune responses in a far more sophisticated way than a conventional vaccine. His nanoparticles are tiny synthetic capsules that can carry conventional drugs or more sophisticated molecules that would not survive in the bloodstream. The key is ensuring that nanoparticles home in on specific cells. Irvine plans to add antibodies to their surface that bind to the target cell. The nanoparticles will be absorbed into the cell, where they release their payload. The cargo that the nanoparticles ferry to their destination will differ from one disease to the next. In some cases, it might be a protein that draws in other immune cells to fight a tumour. In others, small interfering RNAs could be used to temporarily block the expression of genes that promote autoimmunity. By combining different payloads it could be possible to give cells complex instructions such as "go here and attack this".

For the latest from New Scientiist visit www.newscientist.com