The viruses that kill tumors

Jo Whelan, New Scientist, Nov 19, 2005

http://www.newscientist.com/channel/health/mg18825262.200-the-viruses-that-kill-tumours.html

Moira Brown hardly slept for four nights. Her team had just injected live herpes simplex virus directly into the brain of a 21-year-old man. The virus can cause fatal swelling of the brain, but it was worth a try: the man had an aggressive form of brain cancer and had been given four months to live. That was 1997. Remarkably, he is still alive today, his tumour gone.

The treatment was not quite as mad as it seems. Brown's team, at the University of Glasgow, UK, had mutated the herpes virus so that it could replicate only in tumour cells, leaving ordinary cells alone. Promising lab results helped convince the UK's medical authorities to give Brown the go-ahead for the risky attempt. "We knew that the treatment was safe in animals, but you cannot be sure it will be safe for people too. So we were very nervous when we treated the first patient," says Brown.

Another patient who took part in the trial is also still alive, eight years on. And some of the other 10 patients survived a few months longer than expected. Given that the average survival time after glioma is diagnosed is just one year, a figure that has not changed in 30 years, the results were encouraging enough to justify more small trials. Now the company founded by Brown, Crusade Laboratories, is about to begin final-stage trials involving patients with recurrent glioma. The mutant herpes strain might be the first virus to be approved for treating cancer.

It could be the first of many. Plenty of other cancer researchers around the world think they are also on the verge of turning viruses into potent cancer treatments. "When I look at viruses I see a whole pharmacopoeia of new drugs," says Stephen Russell of the Mayo Clinic in Rochester, Minnesota.

And the second generation of cancer-killing viruses are much more than the simple mutant that Brown created. Researchers are equipping them with an arsenal of weapons that can do anything from making cancer cells commit suicide to bringing down the wrath of the immune system upon them. "I am really excited about viruses because they are such a different kind of therapeutic," says John Bell of the Ottawa Regional Cancer Centre in Canada. "It's a very creative technology. There are lots of things you could do to the virus. You can use your imagination."

The idea of using viruses to kill cancers goes back nearly a century. In 1912, an Italian gynaecology journal reported the case of a woman with advanced cervical cancer who, after being bitten by a dog, was vaccinated with a live but weakened strain of the rabies virus. To the doctors' surprise, her tumour shrank.

After more reports of patients' tumours regressing after viral infections or vaccinations, doctors began to take the idea seriously. From the late 1940s onwards, several trials took place in which cancer patients were injected with live viruses. A few individuals showed striking improvements, but the results were mixed overall. Doctors pinned their hopes on chemotherapy and radiotherapy instead, and by the end of 1970s the approach had largely been abandoned.

The perfect bioweapon

While viral-therapy papers gathered dust on library shelves, a revolution was under way in biology. Armed with a burgeoning understanding of how viruses infect cells and a battery of techniques for manipulating their genes, researchers realised that they no longer had to rely on the natural tendency of some viruses to home in on cancer cells.

In 1991, Robert Martuza at Harvard Medical School created the first virus designed to target cancer cells. His team deleted the gene for an enzyme called thymidine kinase from the herpes virus. Without it, the virus cannot replicate. But human cells produce the enzyme when they are dividing, so the virus thrives in rapidly dividing cancer cells. Since then, researchers have created numerous tumour-targeting viruses, either by manipulating replication as Martuza did or by altering a virus's surface proteins so the only cells it can enter are cancerous ones (see Diagram).

Viruses seem to be the perfect biological weapon against cancer. The concept has a seductive elegance: the goal of any cancer therapy is to kill cancerous cells, and killing cells is what many viruses excel at. When they infect a host cell, they replicate and form thousands of new virus particles. Most then make the host cell burst open, killing it and releasing the daughter viruses to infect other cells. The bursting process is known as lysis, or oncolysis if it is happening to a cancer cell.

Oncolytic viruses relish some of the very changes that cause cancer cells to run dangerously out of control in the first place. When healthy cells are infected by a virus, they try to commit suicide - a process known as apoptosis - before the virus can replicate. But cancer cells are resistant to apoptosis, making them ideal hosts. "Viruses prefer cells that have blocks to apoptosis," says David Kirn, a cancer researcher and founder of Jennerex Biotherapeutics of San Francisco. "That resistance is used against the cancer cell. It is a unique mechanism of action."

This is why many "wild" viruses cause cancers to regress. Engineer a virus to target only cancer cells, and you have a precision weapon. While chemotherapy drugs only destroy around six cancerous cells for every healthy cell they kill, oncolytic viruses can take out thousands per healthy cell.

Since Martuza's breakthrough, oncolytic virotherapy has again become a hot topic. The results from the lab look good: many modified viruses selectively kill cancer cells in cell culture and in animals. At least 14 have already made it to early-stage human trials.

But as so often, treating people is proving more complex. Few oncolytic viruses have showed any consistent action. There are tantalising glimpses of effectiveness in some individuals, but most viruses have failed to convince. One, called ONYX-015, raised hopes when it showed promising results against cancers of the head and neck, but it failed to make an impact on pancreatic, ovarian, lung or liver tumours. The company developing ONYX-015 decided to focus on other products instead, and in January licensed it to a Chinese company, Shanghai Sunway Biotech. No oncolytic virus has yet been approved.

The good news, as Brown discovered to her relief, is that the technique appears to be safe. In 1999, 18-year-old Jesse Gelsinger died from a severe immune reaction to a non-replicating virus used for gene therapy. Live viruses could in theory be even more dangerous. "These are replicating biologicals, and one can imagine something that we haven't anticipated could be a safety issue," says Bell.

So far, though, the only significant adverse effects seen in any of the trials are flu-like symptoms. In fact, some of the viruses might be too safe. In their concern to avoid dangerous infections, researchers may have weakened the viruses' replication abilities too much, preventing the viruses from reaching an effective concentration in a tumour. "We were erring on the side of safety to such an extent that the viruses we used were not as potent as they needed to be," says Bell.

However, the reasons for the failure of most trials run deeper. "We have got to the point of saying, 'They are safe but they don't work'," says Russell. "We know tumour destruction can occur, but it's a question of making it happen. People are now staring the big issues more directly in the face."
“The goal is to kill cancer cells, something that many viruses excel at”

The biggest of those issues is how to get around the immune system. Just as viruses have evolved to enter cells and replicate, our immune systems have evolved to stop them. And they are very good at it, as researchers in the early trials realised: "The most disappointing aspect is the fact that even when a virus is oncolytic and it punches a hole in a tumour, the immune response of the individual to the virus occurs so fast that the effects are quickly wiped out and the tumour continues to grow," said Albert Sabin, developer of the live oral polio vaccine, back in 1957.

Stealth viruses

Most of us carry an array of antibodies to common viruses, acquired either by natural exposure or by vaccination. As soon as one of these viruses enters our bloodstream, antibodies bind to it and neutralise it, while the immune system starts producing more antibodies. Even if you have never been exposed to a particular virus before, you will develop antibodies to it within days. "And once those get up to a certain count they will start to reduce the amount of virus you can get to a tumour," says Kirn.

So researchers are working on ways to get viruses past the antibody sentries. Some viruses are already adept at this, such as vaccinia, which is used as a vaccine against smallpox. Vaccinia can coat itself with proteins and move through the bloodstream undetected in a state called the "extracellular envelope" form. It might be possible to exploit this for virotherapy.

Another approach is to create "stealth" viruses. Len Seymour's group at the University of Oxford has coated adenoviruses with an inert polymer that makes them invisible to the immune system. This coat also covers the viral proteins that bind to receptors on the outside of cells and allow the virus to enter. Coated viruses are thus incapable of infecting their normal target cells. Add cancer-binding proteins to the polymer coat, however, and you have a virus that infects the tumour of your choice.

Seymour and his team have succeeded in infecting a range of cell types in this way. They aim to get stealth adenoviruses into clinical trials within three years, initially as delivery agents for gene therapy. The technique holds possibilities for oncolytic therapy too, although daughter viruses will lack the polymer coat and so be visible to the immune system.

Even if an oncolytic virus manages to evade marauding antibodies and enter a tumour cell, it is still under threat. The infected cell quickly displays viral proteins on its surface, marking it out for destruction by immune cells. But many researchers see this as an advantage. As well as killing infected tumour cells, they think this response might encourage the immune system to target tumour cells directly, by exposing them to the tumour surface proteins alongside viral ones. The mere presence of the virus also causes a general ramping-up of the immune system, again encouraging it to attack cancer cells. "You really don't want to suppress the immune system if you can avoid it, because there is so much upside to having a robust immune response against the tumour," says Bell.

Not everyone agrees. Russell thinks infected tumour cells are often killed before the virus has had time to replicate. This might account for the difficulty in getting viruses to spread efficiently within tumours. "It's a question of how far the virus gets in the time allowed to it to propagate, and how much tumour remains for the immune system to have to mop up," he says. "The immune system will only be effective if there is minimal residual disease. There isn't much evidence that it can get rid of a big established tumour." His team plans to use immunosuppressant drugs to temporarily knock out the immune system.
Hunting down tumours

In the end, there is unlikely to be a single universally effective strategy for evading the immune system. "This is a complicated subject and varies from one virus to another," says Brown. And it's not the only challenge. Oncolytic virotherapy has so far worked best when the virus is injected directly into the tumour, but most cancer deaths are caused by cancers that spread from their original site. What patients really need is a treatment that can kill tumours wherever they are in the body.

Much early work was with adenoviruses, which target the mucous membranes of the nose and throat to cause the all-too-familiar symptoms of the common cold, or herpes simplex, which targets skin and nerve cells. Some researchers are trying to adapt these viruses to spread throughout the body via the bloodstream. Earlier this year, a team at the Memorial Sloan-Kettering Cancer Center in New York treated 12 patients with skin cancer that had spread to the liver by injecting a herpes strain into the liver artery. The results are encouraging; it is the first time that injecting an oncolytic virus into the bloodstream has produced an anti-tumour response.

"But the best way to approach systemic delivery is to use viruses that have evolved to spread through the bloodstream, instead of re-engineering viruses to do something they don't naturally do," says Kirn. Several groups around the world, including Kirn's company Jennerex, say they have perfected systemic delivery of vaccinia in animals. Human trials should start before the end of the year.

Meanwhile, Russell's group has high hopes for the weakened measles strain that is routinely used in vaccinations. The team has genetically engineered the virus so that the proteins it uses to bind to and enter cells can be replaced by antibodies targeting various cancer types.

For virotherapy researchers, dodging the immune system and hunting down cancers around the body is just the beginning. They have a grander vision. "For the first time in cancer treatment history, we have the opportunity to kill by multiple mechanisms in a single product," says Kirn. By arming different viruses with different cargoes, such as a drug, a radioisotope, an antibody or a gene that codes for a cancer-fighting protein, researchers could in theory produce a limitless number of new treatment agents.
“The grand vision is to arm viruses with multiple weapons”

Part of the aim is to kill tumour cells that are not actively dividing. Oncolytic viruses are not very good at replicating inside and killing resting cells, but they can infect them and release a payload. The first "armed" oncolytic viruses are already in early trials. Most carry a gene for GM-CSF, a protein intended to stimulate the immune system to attack tumours.

Despite the progress being made, some doubt that oncolytic viruses will prove to be a magic bullet. "It is unlikely that the oncolytic viruses currently available will be able to fully eradicate tumours on their own," says Henk van der Poel of the Netherlands Cancer Institute in Amsterdam. "Their greatest potential will be as part of a multi-modal treatment regime."

Those in the field agree that virotherapy is not about to revolutionise cancer treatment just yet, but they are quietly optimistic about its prospects. "It's a novel platform that's going through some of the same issues as monoclonal antibodies did," says Kirn. "Ten or 15 years ago, some people said monoclonal antibodies were dead, but a few individuals stuck at it and now they are a very important part of our anti-cancer armamentarium."