LONDON ¿ Women who have a sister with cervical cancer have double the risk of developing cervical cancer themselves, Swedish researchers have found. Their study shows for the first time there is a genetic component influencing the risk of cervical cancer. The team now is embarking on a search for the genes responsible.

Ulf Gyllensten, professor of medical molecular genetics at the University of Uppsala in Sweden, told BioWorld International, ¿We predict that it may one day be possible to identify women who are at high risk genetically of developing this disease, and tailor their treatment accordingly.¿

The results of the study are reported in a letter to the July 1 Nature titled, ¿Genetic link to cervical tumours.¿

Gyllensten said, ¿We wanted to embark on a study of the genes responsible for cervical cancer, but first we needed some basic support for the idea that there was a gene or several genes to be found. These results now give us a very good reason for looking for those genes.¿

The researchers were aided by two resources unique to Sweden: the Swedish cancer registry, which contains a record of all cancers diagnosed in Sweden since 1958, and the Swedish family registry, which is a record of the parents of everybody born after 1940. That made it possible to identify the relatives of women who had been diagnosed with cervical cancer, and find out whether they also had been diagnosed with the same disease.

Cervical cancer is associated with certain varieties of the human papillomavirus (HPV). Many studies have examined the relationship between the two. Most recently, for example, a paper in the July 3 issue of The Lancet by Chris Meijer and colleagues, titled ¿Relation of human papillomavirus status to cervical lesions and consequences for cervical cancer screening: a prospective study,¿ concluded that persistent infection with this virus is needed for cervical cancer to develop. Meijer¿s group recommended women who have mild to moderate precancerous changes in the cervix should be referred to a gynecologist only if they still test positive for HPV six months later.

Some women can therefore clear the infection themselves, and may not need further treatment. Gyllensten added, ¿In some populations, the frequency of infection with cancer-causing types of HPV may be as high as 20 percent. But fewer than 1 percent of those women actually go on to develop cancer, so there must be other factors at work, which we now believe are genetic.¿

The Swedish study showed that the relative risk for biological full sisters of women with cervical cancer was 1.93. In other words, this group had double the normal risk of developing the disease. The for adoptive sisters of cases was only 1.15.

The relative risk for half sisters of cases was 1.45. ¿This is exactly the risk you would expect due to the amount of genes that they are sharing,¿ Gyllensten said.

One theory proposed to explain familial clustering of cervical cancer is that HPV may be transmitted vertically, from infected mothers to their daughters. If that was the case, the relative risk for half sisters would be expected to be increased, but only when the mother is the parent in common. But Gyllensten and his colleagues found the relative risk of half sisters was similar, whether they had a mother in common (1.49) or a father in common (1.41).

¿Thus the familial clustering cannot be explained by transmission of HPV from mother to child,¿ the researchers wrote.

¿These results support the idea that there is a genetic predisposition for this type of cancer, as for many other forms of cancer,¿ Gyllensten said. ¿The effect is of the same order as that seen in prostate cancer, where fewer than 5 percent of cases are familial.¿

The finding may explain, he added, why some women are able to deal with HPV infection and do not go on to develop cancer. It is possible, in the future, that doctors may be able to identify women who are at high risk of the disease because their immune systems are unable to clear HPV infection. ¿Treatment could be directed at this group,¿ Gyllensten said, ¿while most women who are infected would be able to clear the infection, and may not need any other treatment at all.¿

A search for the genes responsible for family clusters of cervical cancer is next on the agenda. Gyllensten and his colleagues are searching for those genes using both candidate gene analysis and genetic linkage analysis.

¿When we find these genes, they will have an important role in determining how to design each patient¿s treatment,¿ Gyllensten concluded.