More and more in recent years, cancer biologists are pointing their fingers at viruses. Human papillomavirus, they found, causes cervical cancer; hepatitis B induces liver cancer; and Epstein-Barr virus has been implicated in lymphoma. Most recently, scientists discovered that malignant brain tumors called glioblastoma multiforme, the late-stage version of the cancer that has afflicted Senator Edward Kennedy of Massachusetts, are almost always teeming with cytomegalovirus (CMV), a common, typically harmless herpesvirus. Although the nature of the association is still a mystery, researchers are already taking advantage of the link to find new cancer treatments.

The saga began in the late 1990s, when Charles Cobbs, a neurosurgeon then at the University of California, San Francisco, started pondering the link between inflammation and brain cancer. Malignant tumors are often associated with abnormal immune activity, and he wanted to know why. “Is it just something that happens out of the blue, or is it possible that there’s something maybe driving that inflammatory cascade?” he recalls wondering.

Because they elicit immune responses, infections immediately sprang to mind as possible candidates. Cobbs and his colleagues analyzed glioblastoma samples from 22 patients and found that all harbored CMV. Four out of five people have this virus, which remains in the body for life. Usually a person’s immune system keeps CMV in a latent state in which it does not replicate, but Cobbs found the virus actively reproducing in these tumor cells—and not in healthy cells nearby. “It was stunningly obvious that these tumors were infected,” says Cobbs, whose findings, published in Cancer Research in 2002, were confirmed in 2007 by Duke University neuro-oncologist Duane Mitchell.

What was not obvious was why, exactly, the infection was there. Did CMV cause the cancer, or did it simply proliferate in tumor cells? “It’s a chicken-and-egg question: What came first, the virus or the tumor?” Mitchell points out. Glioblastoma patients have compromised immune systems, which might enable a latent CMV infection to reactivate, Mitchell says. And CMV might be frequently found in brain tumor cells because these cells are easy to infiltrate. A 2008 study Cobbs published in Nature revealed that a cell-surface receptor responsible for letting CMV inside is more frequently found on brain tumor cells than other cell types.

Cobbs, now at San Francisco’s California Pacific Medical Center Research Institute, believes that CMV plays a more active role in generating tumors. He points to a study published in May in Science showing that CMV makes proteins that “turn off” human genes important for preventing unwanted cell growth, a prerequisite to tumor development. It is as if CMV is “clipping the brake line,” remarks study co-author Robert Kalejta, a molecular virologist at the University of Wisconsin–Madison. Other studies have shown that CMV can interrupt a cell’s ability to commit suicide when the cell growth has gone awry. Still, no one has shown that CMV can turn healthy cells into cancer cells, Kalejta notes. So although the virus has some of the tools necessary to cause cancer, there is no proof that it does.

The good news is that when it comes to formulating cancer treatments, understanding the details of CMV’s link to brain cancer is less important than the link itself. “For our purposes, it doesn’t really matter,” says Mitchell, whose lab focuses on new cancer treatments. “We see the presence of the virus as a unique opportunity to go after it as a target in tumor cells.” His lab has “trained” immune system cells to recognize CMV proteins and has used those cells to identify and kill CMV-infected tumor cells.

Mitchell and his colleagues are currently testing their vaccine—and a second version using a different immune cell—in clinical trials, and although they have not yet published their results, he says that outcomes look promising. Cobbs, for one, is hopeful. “I’m holding my breath,” he remarks. “It looks like this may be a radi­cally new way to consider treating these tumors.”

When Cleanliness Is Next to Malignancy
Cytomegalovirus (CMV) infects about 80 percent of the population. So if CMV causes glioblastoma multiforme, as Charles Cobbs of San Francisco’s California Pacific Medical Center Research Institute hypothesizes, why do only a small number of people develop brain tumors? Cobbs argues that the same question could be asked for known cancer-causing pathogens such as human papillomavirus: “That’s actually the dogma—that you have widespread infection and in only a small percentage of cases there is cancer.” In glioblastoma, he has noticed that the majority of patients are affluent, and he speculates that people infected with latent CMV might be more likely to get tumors if they grow up in hygienic environments.

The idea stems from the “hygiene hypothesis,” used to explain the rising incidence of allergies in developed countries. It posits that childhood exposure to pathogens primes the immune system to respond appropriately; however, when people grow up in “superclean” environments, their immune system does not mature properly. When infected with CMV, these patients might then be at a heightened risk for developing glioblastoma, Cobbs says—but he admits his idea is based on little more than a hunch.

Note: This article was originally printed with the title, "Virus in the Brain".