A one-stop approach to diagnosing hereditary colorectal cancer in the Chinese population.

Abstract

The current procedure for identifying hereditary colorectal cancer (HCRC) is time consuming in clinical practice. This study aimed to develop a time-saving approach to diagnosing HCRC. A total of 100 suspected HCRC patients were prospectively enrolled (cohort 1) and 116 colorectal cancer patients with DNA mismatch repair-deficient were retrospectively included (cohort 2). Next-generation sequencing (NGS) tests were performed on tumors and matched white blood cells (WBCs) or normal tissues. Using the conventional method upon WBC/normal tissue-based NGS data as a reference, the performance of the ColonCore method using tumor-only-based NGS data in predicting germline variants was explored in cohort 1 and validated in cohort 2. In cohort 1, the ColonCore method diagnosed 17 Lynch syndrome (LS) and 14 familial adenomatous polyposis (FAP); and by the conventional method, the cases were 16 and 10, respectively. The ColonCore method showed sensitivities of 100% in diagnosing LS (positive predictive value [PPV] 94.1%) and FAP (PPV 71.4%). Moreover, two of seven patients with multiple adenomas/polyps who did not meet existing clinical criteria for HCRC were predicted to harbor germline variants in APC and MUTYH. Additionally, the sensitivity of the ColonCore method in identifying LS patients from cohort 2 reached 85.7% with a PPV of 85.7%. The ColonCore method might be an acceptable tool for predicting germline variants associated with HCRC. Our work indicates the essentiality of NGS tests in CRC patients for precision diagnosis and treatment.