Colorectal-Tumor Screening for Lynch Syndrome Has Low Yield in Elderly

By Will Boggs MD

June 14, 2019

NEW YORK (Reuters Health) - The diagnostic yield of universal screening of colorectal tumors for Lynch syndrome decreases substantially after ages 70 to 75, a retrospective study suggests.

"Prior studies have shown the rapid decline of Lynch syndrome prevalence with increasing age, but few studies have formally studied the screening efficiency of reflex testing of colorectal cancer after age 70 years," Dr. Dan X. Li from Kaiser Permanente Northern California, in Santa Clara, told Reuters Health by email.

Several societies in the U.S. and Europe have endorsed universal screening of tumors from all patients with newly diagnosed colorectal cancer for Lynch syndrome with immunohistochemistry followed by genetic testing if indicated. The diagnostic yield and cost-effectiveness of this approach among elderly individuals has not been well investigated.

Dr. Li's team investigated the performance of colorectal tumor-based Lynch syndrome screening among different age groups in their retrospective study of 3,891 colorectal tumors.

Among all patients, 85.6% had normal mismatch repair (MMR) immunohistochemistry results. The most common MMR deficiencies were MLH1 and PMS2 deficiency (12.2%), followed by MSH2 and MSH6 deficiency (0.9%) and PMS2-only deficiency (0.9%).

Overall, 63 patients with Lynch syndrome were identified by universal screening. The diagnostic yield rose from 0.72% with 50 years as the upper age limit for screening to 1.54% with 75 years as the upper limit and 1.59% with 80 years as the upper limit.

The incremental diagnostic yield decreased substantially after age 70 to 75, with minimal incremental gain after age 80 years, the team reports in Annals of Internal Medicine, online June 11.

Using 70 years as the upper age limit missed five of the 63 cases (7.9%) but resulted in 1,499 (38.5%) fewer cases requiring MMR immunohistochemistry.

Using 75 years as the upper limit missed three cases (4.8%) and resulted in 27.1% fewer cases requiring tumor MMR immunohistochemistry, and using 80 years as the upper limit missed only one case and resulted in 17.2% fewer cases requiring MMR immunohistochemistry.

The number of tumors needed to screen to identify one Lynch syndrome case increased from 20 among patients diagnosed by age 50 to 208 among those diagnosed at ages 71 to 80 and to 668 among those diagnosed after age 80.

"Based on our data, it is likely not cost-effective to reflex screen patients over age 75-80 years at this point," Dr. Li said. "However, our data call for formal cost-effective analyses in this age population. Because the cost-effectiveness of Lynch syndrome screening is also related to the success of engaging the at-risk family members to take the genetic testing (which was not addressed in this study), studies assessing the genetic-testing uptake rate among relatives of Lynch syndrome patients in the advanced-age group will be important to address this question. Additionally, as the costs of gene sequencing are decreasing, the cost-effectiveness of screening elderly individuals may change in the future."

"If the healthcare resources are limited, it is reasonable to stop reflex colorectal cancer tumor screening for Lynch syndrome after age 80 years," he said.

Dr. Li added, "Colorectal cancers that require chemotherapy would still benefit from checking the mismatch-repair status to guide chemotherapy, regardless of age. This is a separate indication for checking mismatch-repair status that is beyond the scope of this study."

Dayna R. Cenin from Erasmus Medical Center, in Rotterdam, the Netherlands, who recently evaluated the costs and outcomes of Lynch syndrome screening in the Australian colorectal-cancer population, told Reuters Health by email, "There is significant benefit to be gained from screening for Lynch syndrome in those with colorectal cancer under the age of 70 years. Based on these results, it would be considered reasonable to not screen for Lynch syndrome in colorectal cancer cases over the age of 80 due to low yield."

"The message between age 70 and 80 is less clear - although screening will yield additional Lynch syndrome cases, a substantial number of additional tests are required," she said. "Further research into this age group may help to provide clearer direction."

"Universal screening for Lynch syndrome is an important and topical issue," Cenin said. "A diagnosis of Lynch syndrome can assist clinical decision-making and positively impact patient outcomes. In addition, it allows cascade testing to identify at-risk family members. This, in turn, enables the commencement of intensive surveillance, which has been shown to reduce Lynch syndrome related cancer incidence and mortality."

"However, questions remain about the cost-effectiveness of unrestricted universal screening versus age-restricted screening," she said. "Our investigation found that unrestricted universal screening would not be considered cost effective compared to stopping screening at age 70. However, to really determine the appropriate cut-off, a formal cost-effectiveness analysis, which also considers the downstream costs and effects of cascade testing, is required."


Ann Intern Med 2019.