Comparisons of screening strategies for identifying Lynch syndrome among patients with MLH1-deficient colorectal cancer

Binyi Xiao 1, Jun Luo 2, E Xie 3, Lingheng Kong 1, Jinghua Tang 1, Dingxin Liu 1, Linlin Mao 4, Qiaoqi Sui 1, Weirong Li 1, Zhigang Hong 1, Zhizhong Pan 5, Wu Jiang 6, Pei-Rong Ding 7

Affiliations

PMID: 32661327
DOI: 10.1038/s41431-020-0687-8

Abstract

BRAF and MLH1 promoter methylation testings have been proven effective prescreens for Lynch Syndrome. We aimed to compare different screening strategies for Lynch Syndrome in patients with MLH1(-) CRC. Patients with MLH1(-) CRC who had been tested for BRAF mutation and germline variants of DNA mismatch repair genes were included. We compared the sensitivities and specificities for identifying Lynch Syndrome and the cost-effectiveness of four screening approaches that used the following tests as prescreens: BRAF testing, MLH1 methylation testing, MLH1 methylation & BRAF testing, and MLH1 methylation testing & Revised Bethesda Criteria. Of 109 patients included, 23 (21.1%) were Lynch Syndrome patients. BRAF mutation and MLH1 methylation occurred in 6 (5.5%) and 40 (36.7%) patients, respectively. The sensitivity for identifying Lynch syndrome of BRAF testing was 100%, but the specificity was only 7%. MLH1 methylation testing had a lower sensitivity than BRAF testing (97.5% vs 100%), but had a markedly higher specificity (45.3% vs 7%). The combination of the two testings had a slightly higher specificity than MLH1 methylation testing alone (47.7% vs 45.3%). The MLH1 methylation testing approach had a 10% lower cost of identifying MLH1(-) Lynch syndrome carriers per case than universal genetic testing, but it missed 4.5% of patients. BRAF and MLH1 promoter methylation testings as prescreens for Lynch syndrome are less effective in Chinese patients with MLH1(-) CRC than in their Western counterparts. Universal genetic testing could be considered an up-front option for this population.