Unearthing Clues to Cancer Risk in the Aging Genome
Two new studies show that some people without cancer have large, structural chromosome abnormalities in a subset of their cells, including abnormalities that have been shown previously to occur primarily in blood cancers.
The frequency of this genetic mosaicism—a mixture of normal and mutated cells of the same type (for example, blood, hair, or skin cells) within an individual—appears to increase with age, especially after age 50. It may also be associated with an increased risk of hematologic (blood) cancers and possibly some solid tumor types, researchers found.
The studies, published online (here and here) May 6 in Nature Genetics, "provide fascinating new information and data on the frequency and distribution of chromosomal aberrations in healthy individuals," commented Drs. Fredrik Mertens and Bertil Johansson of the Department of Clinical Genetics at Lund University Hospital in Sweden in an e-mail message. Drs. Mertens and Johansson were not involved in either study.
The findings could lead to new insights on how and why the risk of cancer increases with age and might eventually help identify people at higher-than-normal risk of developing certain cancers, explained Dr. Stephen Chanock, chief of NCI's Laboratory of Translational Genomics, Division of Cancer Epidemiology and Genetics (DCEG), and a co-author of both studies.
The studies were conducted by two large groups, one led by NCI's DCEG and one comprising members of the NIH-funded Gene Environment Association Studies (GENEVA) consortium, which is directed by NIH's National Human Genome Research Institute (NHGRI).
Researchers first noticed the chromosome anomalies while doing quality control checks on data from genome-wide association studies (GWAS). These studies use single nucleotide polymorphism (SNP) microarray analysis to identify common genetic variants associated with cancer and other diseases. GWAS investigators found that DNA from cells assumed to be normal—mainly from blood samples, as well as buccal and saliva samples—contained unexpected structural abnormalities. The abnormalities occurred in a subset of analyzed cells from given individuals.
Many chromosome abnormalities originate as errors in eggs or sperm prior to fertilization. Therefore, the abnormalities are present in every cell of the body. Some abnormalities, however, can occur during an embryo's development or later in life. The result is acquired (noninherited) mosaicism, in which only a fraction of cells have the abnormality.
Chromosomal mosaicism has been shown to cause miscarriage, birth defects, developmental delay, and cancer, "but these are the first large-scale studies describing large [mosaic] chromosome abnormalities in people from the general population," said Dr. Cathy Laurie, a statistical geneticist at the University of Washington and one of the lead authors of the GENEVA study.
To determine the frequency of mosaicism in the general population, the NCI-led investigators looked for signs of large (greater than 2 million DNA base pairs) mosaic chromosome abnormalities in more than 31,000 case subjects with cancer and 26,000 control subjects without cancer from 13 GWAS of cancer. At the same time, GENEVA researchers looked for chromosome abnormalities larger than 50,000 base pairs in about 50,000 individuals from 12 GWAS, which looked mainly at conditions other than cancer.
In both studies, researchers found that mosaicism occurred in less than 0.5 percent of people younger than 50, but, after age 50, the percentage rose rapidly to about 2 to 3 percent in those older than 75, noted study author Dr. Anastasia Wise, an epidemiologist at NHGRI and program director for GENEVA.
These findings are intriguing because the incidence of many cancers also increases with age, Dr. Wise commented.
In the NCI-led studies, "we saw mosaicism in the cancer-free control subjects, as well as in the cancer case subjects. So we wanted to go to the next step and ask, 'What does it mean, and could it be a risk factor for cancer?'" explained Dr. Chanock.
The NCI-led study showed that the risk of leukemia was substantially higher among those with mosaic abnormalities. And GENEVA investigators estimated that the risk of all hematologic cancers (leukemia, lymphoma, and myeloma) was 10 times higher for mosaic individuals than for non-mosaic individuals.
When Dr. Chanock and his colleagues compared the subset of people who developed non-blood-based cancers and the corresponding cancer-free control subjects, they found that the risk of solid tumors, especially lung and kidney cancer, was also higher in mosaic individuals than in non-mosaic individuals. However, the correlation between mosaicism and solid tumors was not as strong as that for hematologic cancers.
Dr. Chanock emphasized that the presence of chromosome anomalies in a subset of blood cells may be an early marker of hematologic and possibly other cancers, as well as other diseases that are more common in older people. Dr. Laurie added, "The surprising thing is that the vast majority of people [in both studies] did not have a history of hematologic cancer, and we still found structural abnormalities in their blood cells."
Cataloging Chromosome Changes
The chromosome abnormalities that researchers detected included duplications and deletions of large sections of chromosomes or entire chromosomes. They also included abnormalities known as copy-neutral loss of heterozygosity, in which a person ends up with two identical copies of a partial or whole chromosome.
—Dr. Cathy Laurie
When the researchers compared the anomalies they detected with those in the Mitelman Database, a catalog of chromosome aberrations found in cancer (developed by Dr. Felix Mitelman in collaboration with Drs. Johansson and Mertens), they found significant overlaps. And some of the chromosome regions that were commonly deleted, or missing, in their samples contained genes previously associated with cancer, they discovered.
"It has been known for some time that acquired chromosome aberrations accumulate with increasing age, and that the overall level of chromosome aberrations is associated with cancer risk," Drs. Mertens and Johansson wrote. "What is new and exciting in the two studies is the use of SNP array analysis, which allows much more extensive analysis of these aberrations, both in terms of resolution and in terms of number of individuals analyzed," than is possible with classical cytogenetic analysis.
Although the findings for hematologic cancers (and for solid tumors in the NCI-led study) were statistically significant, they were based on a small number of cancer cases from studies not designed to look at clinical outcomes. Further studies, some of which are under way, are needed to understand how genetic mosaicism may influence the risk of these and other cancers.
The studies will need to follow large numbers of healthy people over time and examine repeated blood samples from them. This could reveal whether mosaic chromosome abnormalities are stable and whether people with these anomalies are more likely to develop cancer and other diseases, explained Dr. Laurie.
Future research will also explore the origins of mosaic events seen in older individuals. "They may occur at birth and sit at a very low and undetectable level, and then later in life the proverbial genie is let out of the bottle," explained Dr. Chanock. However, he said, "I think most people would put money on the second hypothesis, which is that as you grow older the mechanisms of DNA repair deteriorate" and the genome becomes less stable.