How to Kill a Cancer

It has an exotic name, but the human papillomavirus is actually a very commonplace sexually transmitted disease. More than half of U.S. women will develop sex-related HPV infections at some point in their life, and in most cases the immune system knocks the virus out before it can do any damage. In some cases, however, the infection becomes chronic, with measurable levels of the virus remaining in the system. Perhaps 5% of these women will eventually develop precancerous lesions that could lead to cervical cancer. Or at least that's what would happen if they did not undergo routine Pap smears, which can detect abnormal cervical tissue before it turns malignant.

Right now, the fate of a woman with an abnormal cervical growth depends a lot on whether she lives in a developed country, how quickly her lesions are discovered and the type of care she receives. That may change soon. A study published in the New England Journal of Medicine last week showed for the first time that it's possible to vaccinate women against at least one strain of HPV--a key step in any campaign to eliminate cervical cancer, since practically all cervical tumors need an HPV infection to get started. The study was not designed to treat patients but rather to show that the vaccine approach works, notes Laura Koutsky, an epidemiologist from the University of Washington in Seattle, who led the study. In other words, she says, "it's a proof of principle."

There's reason to believe that the principle could bear real fruit. About 20 of the 100 or so strains of HPV have been linked to cervical cancer, but 70% of cervical tumors are caused by just two strains--HPV-16 and HPV-18. Koutsky and her colleagues focused on HPV-16. In their study, they inoculated 768 women with a prototype anti-HPV-16 vaccine developed by Merck Research Laboratories; an additional 765 women were injected with a placebo. Both groups carried on with their normal sexual activity. After an average of 17 months, not a single woman who received the vaccine developed an HPV-16 infection, while 41 of the women in the placebo group did.

So far, the most common side effect of the prototype vaccine has been a little bit of muscle pain at the injection site. The inoculum was designed so that it cannot cause an HPV infection, even accidentally. It contains only proteins from the virus' outer shell and no genetic material that can lead to disease.

Merck is moving forward with plans to test a more refined vaccine that may be available in five years. (Other groups are pursuing similar strategies.) Such a vaccine would protect against the two main cancer-causing strains of HPV as well as HPV-6 and HPV-11, which are not malignant but do trigger genital warts. It would be impractical to develop a vaccine against all 100 HPV strains, since so many are harmless. The types that cause garden-variety warts on the hands and feet are perhaps the most familiar to people and are not contracted sexually.