EDINBURGH - A new set of clinical criteria-embodied in a Web-based tool-improves the ability of oncologists to diagnose patients with Lynch syndrome, according to Scottish researchers.
EDINBURGH, June 29 - A new set of clinical criteria-embodied in a Web-based tool-improves the ability of oncologists to diagnose patients with Lynch syndrome, according to Scottish researchers.
Lynch syndrome-also known as hereditary nonpolyposis colorectal cancer (HNPCC)-is caused by germ-line mutations in DNA mismatch-repair genes, but testing for those alterations in all cases of colorectal cancer would be costly and impractical, according to Rebecca Barnetson, Ph.D., of the colon cancer genetics group at the University of Edinburgh.
Instead, physicians use established sets of diagnostic rules-dubbed the Amsterdam and Bethesda criteria-to determine which patients should be sent for detailed genetic analysis.
In the June 29 issue of the New England Journal of Medicine, Dr. Barnetson and colleagues reported on a new algorithm that they say is better than either the Amsterdam or Bethesda criteria. It has been encoded into a Web-based tool that uses six questions to establish the probability of a patient carrying mutations in DNA mismatch-repair genes.
The Web site is located at www1.hgu.mrc.ac.uk/Softdata/MMRpredict.php.
The algorithm is based on a prospective, population-based survey of 870 patients, 55 and younger, enrolled soon after they were diagnosed with colorectal cancer; they were not preselected and family history was not considered, Dr. Barnetson and colleagues reported.
The patients were studied for germ-line mutations in the DNA mismatch-repair genes MLH1, MSH2, and MSH6 and the researchers used multivariate logistic regression techniques to develop a model to predict the mutations. Finally, the model was validated in a retrospective cohort of 155 patients younger than 45.
The final algorithm, Dr. Barnetson and colleagues reported, includes six criteria:
•Location of the tumor?proximal or distal.
•Presence or absence of a synchronous/metachronous tumor.
•Colorectal cancer family history.
•Endometrial cancer family history.
For example, a 50-year-old man with a proximal tumor, but no synchronous/metachronous tumor and no family history of either colorectal or endometrial cancer, would have a probability of DNA mismatch-repair mutations of 3%. By contrast, the same man with a synchronous/metachronous tumor and a relative whose colorectal cancer occurred before the age of 50 would have a probability of 93%.
Better tools to diagnose Lynch syndrome are important, because colorectal cancer appears to progress more rapidly in these patients than in patients with sporadic disease, according to Richard Boland, M.D., of Baylor University Medical Center in Dallas.
The syndrome is neither common nor rare, occurring in one or two of every 1,000 people, Dr. Boland wrote in an accompanying editorial. But because the disease progresses rapidly, those with the syndrome (and their first-degree relatives) need more frequent screening colonoscopies.
Dr. Boland said the best estimate is that between 3,300 to 6,000 people with Lynch syndrome are among the 150,000 people diagnosed with colorectal cancer every year in the U.S., and each carrier of the germ-line mutations has a family.
"Our job is to find them," he concluded.