Validation of a Multiplex Hypermethylation-based Blood Test to Detect Hepatocellular Carcinoma: A Prospective Case-control Study

Abstract

Background/Aims

This study aimed to evaluate the performance of an investigational multiplex hypermethylation-based HCC screening test (mhsH) in the detection of hepatocellular carcinoma (HCC) using blood samples.

Methods

Adult patients with chronic liver disease (CLD) were enrolled in this prospective case-control study. For the mhsH test, blood samples were collected, and the cell-free DNA obtained from the samples was analyzed for methylation patterns using multiplex droplet digital polymerase chain reaction. The performance of the mhsH test for the detection of HCC was evaluated according to sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the receiver operating characteristic curve and a comparative analysis with alpha-fetoprotein (AFP) was performed. Radiological imaging was used as the clinical reference standard.

Results

A total of 649 participants were screened for recruitment, and the mhsH test performance assessment was carried out for 588 patients with complete local data, comprising 142 patients in the HCC group and 446 in the non-HCC CLD control group. The test demonstrated robust HCC signal detection, achieving an AUC of 0.91 (95% confidence interval [CI]: 0.88-0.94). The sensitivity and specificity were 0.91 (0.85-0.95) and 0.88 (0.85-0.91), with PPV and NPV of 0.71 (0.66-0.76) and 0.97 (0.95-0.98), respectively. The sensitivity of early-stage, intermediate-stage, and late-stage (Barcelona Clinic Liver Cancer) was 0.87 (73-0.96), 0.91 (0.77-0.98), and 0.94 (0.82-0.99), respectively. The test correctly identified 57 of 62 (92%) patients with AFP-negative HCC. Combining AFP with the mhsH test increased sensitivity to 0.96 (0.92-0.99).

Conclusion

The mhsH test demonstrated high accuracy for detecting HCC signals. Furthermore, when combined with AFP, the test performance for detecting HCC in patients with CLD has significantly improved.