The adapted virus that immunized hundreds of millions of people against smallpox has now been enlisted in the war on cancer. Vaccinia poxvirus joins a herpesvirus and a host of other pathogens on a growing list of engineered viruses entering late-stage human testing against cancer.

After a decade of development of so-called oncolytic viruses, the newest strains hold the most promise yet, researchers say. This new generation of viruses has been genetically "targeted and armed," says Winald Gerritsen of the VU University Medical Center in Amsterdam, who is involved in an early human trial of an engineered adeno-associated virus that attacks glioblastoma, an aggressive form of brain cancer.

In a two-pronged attack, these viruses specifically target tumor cells while delivering a cargo of immune-boosting genes. In contrast, viruses that cause cancer, such as the human papillomavirus that is responsible for most cases of cervical cancer, disrupt a cell's genome, thereby triggering out-of-control growth.

When the engineered viruses recognize and infect cancer cells, they replicate and sometimes destroy their hosts. Several of the viruses also release the gene for granulocyte-macrophage colony-stimulating factor (GM-CSF) an immune system protein. The GM-CSF attracts a swarm of white blood cells and other immune system operatives that mount a further attack on the tumor.

The vaccinia virus has been developed by the biotechnology company Jennerex—named after Edward Jenner, who in the 18th century discovered that a cowpox virus could inoculate against smallpox. It showed effectiveness against liver cancer in a phase II clinical trial and will move into a phase III trial later this year, David Kirn, an oncologist and the company's president and chief executive officer, said at a recent meeting of the American Society for Gene & Cell Therapy in Washington, D.C. In the phase II study, 18 of 24 patients survived at least 12 months; with standard treatment, only about half of patients survive one year. The company also tested the virus in a 23-person, early-stage trial against colorectal, lung, ovarian and skin cancers.

The virus cannot infect noncancerous cells, Kirn explained, because researchers deleted its thymidine kinase gene, which it needs to replicate in the body. However, some 80 percent of solid tumors churn out extra thymidine kinase, which is thought to prevent cancer cell death. The result is a "viral factory" inside cancer cells, Kirn said. "Within 24 hours we see really impressive replication and spread within tumors." Replication of vaccinia is the first step to killing cancer cells and shrinking tumors with this approach.

Later this year, the company plans to launch a phase III clinical trial in advanced liver cancer patients, in which the virus will be added to standard antibody treatment.

Patients undergoing this viral therapy typically experience mild flulike symptoms, owing to immune response to the virus, Kirn said, but the company has reported no other side effects. Most Americans older than 40 received vaccinia as children in smallpox vaccines, but the company found no evidence that having had this vaccination hindered the effectiveness of the new virus.

Meanwhile, an engineered herpesvirus targeting advanced melanoma, developed by the company BioVex, has entered phase III clinical trials to test for effectiveness. Some 200 patients have received the virus as part of the trial, and the company expects to announce early results from the test by the end of the year. The group is launching a phase III trial of the engineered virus, called OncoVex, for head and neck cancer. Like the Jennerex product, the herpesvirus delivers GM-CSF to tumors, creating a localized immune response. A similar virus, called H101, is already marketed in China to treat head and neck cancers.

Finally, a Calgary, Alberta–based company, Oncolytics Biotech, is testing a reovirus (an RNA virus often found in human lungs but thought to be nonpathogenic) against several types of cancer, including that of the lung and skin as well as head and neck malignancies. The company says the reovirus selectively infects cancer cells over healthy ones because once a cell turns malignant it stops making an antiviral factor called protein kinase R. The reovirus takes advantage of this deficit to replicate inside cancer cells.

Researchers have been experimenting with oncolytic viruses for decades, but early attempts were quite cautious and the early viruses showed limited effectiveness, says Michael Lairmore, associate director of basic sciences at The Ohio State University Comprehensive Cancer Center. The targeted viruses "have the potential to add a new tool to our arsenal," he says, because they home in on cancer cells more aggressively.

Both Lairmore and Gerritsen cautioned, however, that oncolytic viruses will still need to be paired with chemo or radiation therapies to achieve the best results. "The response rates we're seeing [in early human trials] are very similar to what we see with all new cancer drugs," Gerritsen says. Of patients who have received only the virus, without other treatment, "about 5 to 10 percent of patients respond really well," he added. "So it's only when we combine oncolytic viruses with standard treatment that we can expect to see some very good effects."