With considerable evidence linking elevated blood levels of the insulin-like growth factor I (IGF-I) to several common cancers, researchers have begun to investigate the gene for this protein.

Three new studies independently test variants of the IGF1 gene for associations with prostate, breast, or colon cancers. The results suggest that more research in this area is needed and warranted.

In the prostate cancer study, researchers identified a region of the IGF1 gene that may be related to a modest increased risk of this cancer.

"We have a 'signal' coming from the region but have not identified the actual variant that may explain the risk," says Dr. Matthew Freedman of the Broad Institute of Harvard University and the Massachusetts Institute of Technology, who led the project.

"The goal was to undertake a comprehensive analysis of the gene," explains Dr. Freedman, "and to see whether inherited variation conferred an increased risk of prostate cancer in a multiethnic population."

His team tested a panel of 29 variants that represented the common variation of the IGF1 gene in their study population. The variants are single nucleotide polymorphisms, or SNPs (pronounced "snips") - places in the gene where a DNA letter varies from one person to the next.

The SNPs were analyzed using DNA from 2,320 prostate cancer patients and a similar number of healthy individuals. The study is the largest of its kind and included African Americans, Caucasians, Hawaiians, Japanese, and Latinos.

The analysis revealed that two SNPs were significantly overrepresented in the cancer patient group compared with the other group, according to findings in the January 18 Journal of the National Cancer Institute (JNCI).

The association was consistent and of a similar magnitude across the ethnic groups, and therefore does not help explain the disproportionately high incidence rates of prostate cancer among African American men, notes Dr. Freedman.

"This study is the most extensive investigation on this gene so far and does a nice job of assessing the association across ethnic groups," says Dr. Jing Ma of Harvard Medical School, who investigates IGF1 and was not involved in the study.

The researchers conclude that compared with the breast cancer susceptibility genes BRCA1 and BRCA2, which confer a high degree of risk but are relatively rare in the general population, IGF1 variants may confer a low degree of risk but are relatively common.

"One way to think about the statistical evidence," says Dr. Freedman, "is that if you could eliminate this particular variant from the population you would reduce the prevalence of prostate cancer by approximately 10 percent."

But at this early stage, the researchers acknowledge that the results need to be replicated, and they have begun to test the SNPs in a much larger population.

In the second new study, this one involving breast cancer, researchers in Europe found a "weak but nominally significant" association between a block of SNPs in IGF1 and breast cancer risk, particularly among women younger than 55 years old.

The researchers tested 5 SNPs in 800 breast cancer patients and 1,500 others from the European Prospective Investigation into Cancer and Nutrition. The study involved several IGF1-related genes and appears in the British Journal of Cancer.

Taken together, the American and European studies provide further support for previous findings on IGF-I protein and cancer risk, according to Dr. Ma, who co-authored the 1998 study in Science that first linked the blood levels and prostate cancer.

A challenge for the field now, she says, is to determine whether IGF1 gene variants alter the function of the IGF-I protein in ways that lead to elevated blood levels.

But even the current findings raise the question: Can anything be done for people who might carry adverse variants?

One answer is that testing for these variants in high-risk populations could identify carriers, who could be monitored and who could potentially reduce their risk by making changes to their lifestyles (IGF-I blood levels are influenced by diet and other nongenetic factors).

Researchers at the University of Texas M.D. Anderson Cancer Center, who published the third new study, are trying to identify gene variants associated with an increased risk of colon cancer among individuals with hereditary nonpolyposis colon cancer (HNPCC).

These variants could be used along with information about lifestyle and family history to identify HNPCC patients who might develop colon cancer at a relatively young age, noted Dr. Marsha Frazier, who led the research.

Her team tested an IGF1 variant consisting of two genetic letters that are repeated for various lengths. The researchers found that shorter length "CA" repeats (fewer than 18) were associated with an earlier onset of colon cancer in their group.

Reporting their findings in the January 18 JNCI, the team writes that more research is needed because their study was small (121 patients with HNPCC) and because the association they detected might have been caused by another variant in the gene, such as a SNP.

The researchers may test the 29 SNPs reported by Dr. Freedman on their patients with HNPCC. "A joint study of SNPs along with the CA repeat would help to evaluate which genetic variant is most strongly related to the risk of disease," says co-author Dr. Christopher Amos.

By Edward R. Winstead