Lack of Genetic Diversity Linked to Cancer Risk

By Michael Smith
CLEVELAND, March 25 -- A lack of genetic diversity at "hot spots" throughout the genome may contribute to the onset of breast, prostate, or head and neck cancer, researchers said here.At several locations in their genomes, patients with these three types of solid cancers were significantly more likely (P<0.001) to be homozygous than were controls, according to Charis Eng, M.D., Ph.D., of the Cleveland Clinic, and colleagues.And when the patients had differences at one of the 16 locations, the DNA of their cancer tissue was likely to have lost one of the alleles, Dr. Eng and colleagues reported in the March 26 issue of the Journal of the American Medical Association.Currently, only a few genetic variations -- "high-penetrance" genes like BRCA1 in breast cancer -- are used in assessing cancer risk, Dr. Eng said.
But the new finding suggests a range of low-penetrance variations may also play a role, and opens the door to using that information in risk assessment, she said.
In the long run, the information may also be useful in managing cancer, she speculated.
Dr. Eng said the finding arose from the observation that patients with cancer appeared to have more regions where both alleles of a gene are identical than would be expected.
Most researchers who noticed that apparent lack of variation dismissed it as a spurious finding. "We chalked it up to bad luck," Dr. Eng said.
But as genetic technology has improved, she said, it became possible to test the observation. With her colleagues, she began examining a commercially available set of 345 autosomal microsatellite markers in 385 patients with invasive breast, prostrate, and head and neck carcinoma.
The researchers compared germline DNA with cancer DNA in the patients and in ancestry-matched controls. They found 16 regions -- common to all three cancer types -- where the frequency of homozygosity was significantly elevated compared with controls.
Some patients had different alleles at one of the 16 locations in germline DNA, Dr. Eng and colleagues found, but when that occurred, they were significantly more likely (P<0.001) to have lost the function of one of the alleles in cancer tissue.
In unpublished data, Dr. Eng and colleagues found an interesting link between the two types of patients. If a patient with different germline alleles at one of the 16 locations lost the function of one in cancer tissue, the remaining allele was almost always the same one seen in the homozygous patients.
To validate the findings, Dr. Eng and colleagues used a different technology and a different tumor type in a different set of patients.
They studied a publicly available set of paired germline and somatic DNA from 205 non-small cell lung carcinomas, which had been subjected to genome-wide single-nucleotide polymorphism (SNP) genotyping.
The two types of data did not overlap completely, Dr. Eng said, with the result that only 12 of the 16 regions were represented in the lung cancer samples.
Nevertheless, she and colleagues found that at least one SNP in each region was significantly more homozygous in lung cancer cases than in controls, with a significance ranging from P<0.05 to P<0.001.
Although the study included nearly 600 patients, it still needs to be validated in larger samples, Dr. Eng and colleagues said, and explored in other cancer types.
The research was supported by the National Cancer Institute, the Department of Defense Prostate Cancer Research Program, the Fondation de France, and the Fédération Nationale des Centres de Lutte contre le Cancer. Dr. Eng reported no conflicts.
Primary source: Journal of the American Medical AssociationSource reference:Assié G, et al "Frequency of germline genomic homozygosity associated with cancer cases" JAMA 2008; 299(12): 1437-45.