UK HealthCare Publications »
Advances & Insights »
Colon cancer is the third most common cancer in the Western world, with about 105,000 new cases occurring each year in the United States. Inherited forms account for 5-10 percent of all diagnosed cases. The most common form of hereditary colon cancer is Lynch syndrome, or hereditary nonpolyposis colorectal cancer syndrome, which accounts for 2-5 percent of all colorectal cancers.
People with Lynch syndrome have an 80 percent chance of getting colon cancer by age 75.
People with Lynch syndrome have an 80 percent chance of getting colon cancer by age 75.
People with Lynch syndrome commonly have a mutation in one of two genes—the MLH1 and MSH2 genes—and have an 80 percent chance of getting colon cancer by age 75. This group is also at risk for other cancers, especially endometrial cancer in women.
Previously, diagnosis and referral for genetic counseling were based primarily on the Amsterdam Criteria, which looked at family history of colorectal cancer and other Lynch syndrome cancers such as uterine cancer. A significant number of people with genetic mutation in MLH1 or MSH2 were missed with these criteria. New prediction models that incorporate molecular diagnostic information with additional clinical and family history have a better chance of identifying people at risk for Lynch syndrome, according to two studies in the September 27 issue of JAMA, the Journal of the American Medical Association.
Researchers at Dana-Farber Cancer Institute in Boston used data from 1,914 patients undergoing genetic testing for Lynch syndrome to develop a tool called the PREMM1,2 (Prediction of Mutations in MLH1 and MSH2). The model uses history of Lynch syndrome cancers and adenomatous polyps in the patient and his or her first- and second-degree relatives to predict the likelihood of genetic mutation. Researchers validated the model in another group of 1,016 patients whose family history indicated a risk for the genetic mutations. The PREMM1,2 model missed fewer mutations (6 percent) than the Amsterdam Criteria (27 percent).
Another team at Johns Hopkins Bloomberg School of Public Health in Baltimore developed a model that predicts the likelihood of a person having a mutation in the MLH1, MSH2 or MSH6 genes and developing colorectal or endometrial cancer. They first looked at the patient’s age and gender, location of the colon tumor, presence of additional colon tumors and family history of colorectal and endometrial cancer. Next, they examined the results of molecular testing on the tumor.
The model, MMRpro, was validated on 279 individuals at risk for these genetic mutations. It showed a greater sensitivity than previously developed models, including the Amsterdam Criteria, in predicting mutations in MLH1 and MSH2.
The PREMMI1,2 model is available in a user friendly Web-based format for clinicians. MMRpro is part of a risk-prediction software package, CancerGene, which is available to the public.
“Because model-based prediction algorithms are increasingly used in genetic counseling and prevention activities, MMRpro is a timely tool for identifying and counseling families at risk for the Lynch syndrome and can improve current genetic counseling and early detection practice,” the authors of the Johns Hopkins study wrote.
An editorial accompanying these studies stated that both tools could be useful in helping clinicians decide whether genetic testing is needed. However, the editorial pointed out that further testing is needed to compare and validate the two models, using population-based rather than clinic-based data. In addition, more information is needed on the prevalence of Lynch syndrome among ethnic populations. The samples used to develop the models were composed primarily of Caucasians of European ancestry.
Lynch syndrome is named for Dr. Henry Lynch, who first described it. In a 2004 study published in JAMA, Dr. Lynch and his research team described an ongoing genealogical and historical study on the spread of a genetic mutation in MSH2 believed to be an American-founder mutation.
They traced the mutation to three families immigrating to the United States from Germany in the early 1700s. To date, 566 descendants of these families have been identified in 14 states, including Kentucky. The researchers continue to enroll carriers of this mutation to expand their genealogical studies.
For more information, see:
For more information, or to make an appointment with a UK HealthCare physician, please call UK Health Connection at 1-800-333-8874.
We’ve known for a long time that certain families appear to have a risk for colon cancer. The way we’ve evaluated the risk for Lynch syndrome-associated cancers so far has mostly depended upon a certain set of strict criteria: There have to be three cases of colon or other related cancers in two generations of the family and one of the cases has to be diagnosed before the patient is 50 years old.
“The [new] models are primarily tools to help physicians decide whether to order a genetic test based on the clinical data available to them.”
The problem is, not all Lynch syndrome families have that exact pattern, so a fair number of people are going to be missed. These new prediction models are designed to catch some of those other cases.
What the PREMM1,2 and the MMPro do is predict the likelihood that, if you do a genetic test on a person believed to be at risk, it will show that they have a mutation in the MLH1 or MSH2 genes. The models are primarily tools to help physicians decide whether to order a genetic test based on the clinical data available to them.
Any mutation prediction model is going to be complicated because in order to be accurate, you take into account lots of variables. The good thing about the original criteria is that they weren’t very complex.
These new prediction models are an attempt to formalize what genetic specialists have been doing for years. However, I don’t think most doctors will use them, although specialists such as gastrointestinal oncologists may. These models will probably be most useful for genetic counselors and for providing doctors with a better idea of what types of family history and clinical factors are useful in evaluating cancer risk.
When colon cancer is associated with Lynch syndrome, there is a 50 percent increase in the chances of a new colon cancer developing. People with this syndrome will need to be tested annually, as will members of their immediate family. Lynch syndrome also puts a person at greater risk for other cancers—including upper gastrointestinal, urinary, endometrial and ovarian cancers.
“Colonoscopy in families with Lynch syndrome reduces the incidence of colon cancer by 50 to 75 percent.”
Lynch syndrome-related colon cancers tend to occur as early as the age of 20 to 25, so screening should start at an earlier age. In the general population, screening with a colonoscopy or sigmoidoscopy is recommended beginning at age 50.
Information about genetic mutations is very important in making surgical decisions. The surgeon may decide, for example, to remove the entire colon because of the high risk of a second colon cancer developing later.
Understanding these hereditary cancers is just the first step in integrating genetics into medical care. As we do more research, we hope to get a better idea of how to prevent many types of cancer and other diseases based on a genetic profile.
We already know, for example, that colonoscopy in families with Lynch syndrome reduces the incidence of colon cancer by 50 to 75 percent. When the colonoscopy is performed, colorectal adenomas or polyps that turn into cancer can simply be removed. If cancers do develop, they tend not to be as advanced.
Estimates are that 5 to 10 percent of all cancers are hereditary. For some, like prostate or lung cancers, there are no genetic tests currently available. The most common tests are for colon, breast and ovarian cancers.
Genetic tests are most appropriate when there is a strong chance the patient has a mutation, the presence of a mutation will change the clinical management, and intervention is available that will impact the risk for cancer. For example, promising data shows that taking tamoxifen will reduce by 50 percent the chances for developing breast cancer even in women carrying a genetic mutation that puts them at risk.
Insurance often pays for all or part of these tests if the possibility of identifying a genetic mutation is high and the patient’s medical management will be significantly affected. The test itself involves taking a small blood sample. Results are usually available in four weeks.
If a person is trying to decide whether to pursue genetic counseling and testing, it’s a good idea to make a family medical history and show it to the family doctor. A good online tool for making your family tree can be found at
www.nsgc.org. Another option is to call a genetic counseling program such as the program at UK Markey Cancer Center for more information.
Heather Pierce is director of the Clinical Genetic Counseling Program at UK Markey Cancer Center and assistant professor of hematology and oncology in the UK College of Medicine.
Each issue of Advances & Insights summarizes an important piece of medical news, accompanied by commentary from a UK expert.
Subscribe here or call 859-257-1000 and provide your mailing address.