Validation of Liquid Biopsy Using a Methylated SEPT9 Assay and Integrated Index for Detecting Hepatocellular Carcinoma

ABSTRACT Aim The incidence of nonviral hepatocellular carcinoma (N‐HCC) is increasing due to improvements in hepatitis control. Patients are often diagnosed at an advanced stage owing to a lack of surveillance systems. We previously reported the usefulness of a liquid biopsy test based on a sensitive methylated septin 9 (m‐ SEPT9 ) assay for detecting HCC. In this study, we validated the diagnostic ability of m‐ SEPT9 and developed a novel integrated index to improve HCC prediction. Methods A total of 1450 samples (464 healthy individuals, 202 patients with chronic liver disease without HCC, and 784 patients with HCC) were analyzed using the combined restriction digital polymerase chain reaction assay for m‐ SEPT9 . Results The sensitivity and specificity of m‐ SEPT9 were 68.1% (95% confidence interval [CI], 0.649–0.714) and 84.9% (95% CI: 0.817–0.882), respectively, with an area under the curve (AUC) of 0.831; however, the lower limit of the 95% CI for specificity did not exceed the threshold of our setting (> 85%). Multivariable analysis identified age, sex, des‐gamma carboxy prothrombin, alpha‐fetoprotein, and m‐ SEPT9 as independent diagnostic markers. Based on these markers, the ASDAm‐S9 index was developed, demonstrating a sensitivity of 89.9% and specificity of 90.9% (healthy individuals vs. HCC, AUC: 0.961), regardless of HCC etiology or stage. Conclusions The diagnostic performance of m‐ SEPT9 alone could not be fully verified; however, the ASDAm‐S9 index demonstrated excellent accuracy for detecting HCC, including its early stages. This index shows promise as a blood‐based cancer screening tool for early‐stage HCC in the general population.