Cytomegalovirus (CMV), a form of herpes, is the most commonly transmitted infection between pregnant women and their babies, affecting 30,000 U.S. newborns a year and leaving some 8,000 of them with permanent damage ranging from mental retardation to vision and hearing loss. Scientists have had a tough time trying to come up with a vaccine against the virus, which is harmless in healthy adults, though it can cause eye inflammation and changes in blood cell count and the gastrointestinal tract in people with AIDS or other immune system conditions.

But one experimental vaccine is showing promise in preventing women from becoming infected with CMV. (As Scientific American noted in its January issue, CMV vaccines are also being tested in brain tumor patients, because many of the malignancies are found to contain the virus.)

In a phase II trial, women who received the vaccine (developed by Sanofi Pasteur that uses an adjuvant, or molecule that boosts the effect of the vaccine, developed by Novartis) caught CMV at about half the rate as women who got a placebo shot, according to results set to be published tomorrow in the New England Journal of Medicine. (Phase I trials test the safety of a treatment on a small group of people and determine proper dosage and Phase II trials test the safety and effectiveness of a therapy in larger groups. The Food and Drug Administration requires Phase III trials before it will approve a med for use. )

This trial involved 441 women who were followed for three years. Half of them received the three-dose vaccine and the other half received saline (salt water) injections. Eight percent (or 19) of those who received the vaccine became infected with CMV, compared to 14 percent (or 32) in the placebo group. The findings were first reported in 2007 at the annual Infectious Disease Society of America meeting.

The trial was too small to determine whether the vaccine protected developing fetuses (conceived after women received the shots) from becoming infected with CMV, according to study co-author Robert Pass, a professor of pediatrics at the University of Alabama at Birmingham. One child was born with congenital CMV infection to a woman who had received the vaccine and three infected babies were born to women who had received the placebo. "If the numbers were exactly the same, that would give me pause. But it's a hint in the right direction," Pass tells He adds that if the vaccine were found to alter a woman's immune system to reduce the amount of virus transferred to her placenta or shortened the amount of time it were circulating in the bloodstream, "there is still a possibility that the vaccine's ability to prevent congenital infection may exceed its ability to protect against maternal infection."

A previous experimental immunization known as the Towne CMV vaccine did not prevent immune-compromised kidney-transplant patients from becoming infected with the virus, but it did keep them from developing symptoms, according to a 1984 study published in The Lancet. CMV subverts the machinery humans use to mount an immune response against viruses. It also uses some of that machinery, such as infection-fighting white blood cells, to spread itself throughout the body. Once a person has a virus, he or she typically develops antibodies against it and will not become infected again. But it appears that having CMV does not make a person immune to future infections — another reason skeptics doubt a vaccine could accomplish such a feat, either

Other companies, including Novartis and Vical, have their own vaccine candidates in development, and GSK completed a phase 1 safety trial of another vaccine, but has yet to release the results.

Pascal Barollier, a Sanofi-Pasteur spokesperson, says the company is reworking the vaccine, possibly with a different antigen (a substance that stimulates the body's immune system to make disease-fighting antibodies against a germ) or adjuvant, and will need to test the new formulation in the lab before trying it out on people. "The study is suggestive that prevention of maternal CMV infection is an achievable goal," Barollier said. "Sanofi Pasteur is committed to bringing such a vaccine to market as soon as possible to respond to an unmet medical need."

Antiviral drugs can slow down replication of the virus, but cannot cure it. Other treatments may be given for its effects, such as eye inflammation and pneumonia. CMV does not always damage developing fetuses, but when it does, its effects can be crippling.

"People who most resoundingly say, 'yes, that would be of value' are families of children who have congenital CMV infection," study co-author Pass says. "To people who are at risk, 50 percent looks a whole lot better than nothing."

Histopathology of cytomegalovirus lung infection/CDC/Haraszti