Exploring methylation signatures for high de novo recurrence risk in hepatocellular carcinoma

Abstract

Background/aims: Hepatocellular carcinoma (HCC) exhibits high de novo recurrence rates post-resection. Current post-surgery recurrence prediction methods are limited, emphasizing the need for reliable biomarkers to assess recurrence risk. We aimed to develop methylation-based markers for classifying HCC patients and predicting their risk of de novo recurrence post-surgery.

Methods: In this retrospective cohort study, we analyzed data from HCC patients who underwent surgical resection in Korea, excluding those with recurrence within one year post-surgery. Using the Infinium Methylation EPIC array on 140 samples in the discovery cohort, we classified patients into low- and high-risk groups based on methylation profiles. Distinctive markers were identified through random forest analysis. These markers were validated in the cancer genome atlas (n=217), Validation cohort 1 (n=63) and experimental Validation using a methylation-sensitive high-resolution melting (MS-HRM) assay in Validation cohort 1 and Validation cohort 2 (n=63).

Results: The low-risk recurrence group (methylation group 1; MG1) showed a methylation average of 0.73 (95% confidence interval [CI] 0.69-0.77) with a 23.5% recurrence rate, while the high-risk group (MG2) had an average of 0.17 (95% CI 0.14-0.20) with a 44.1% recurrence rate (P<0.03). Validation confirmed the applicability of methylation markers across diverse populations, showing high accuracy in predicting the probability of HCC recurrence risk (area under the curve 96.8%). The MS-HRM assay confirmed its effectiveness in predicting de novo recurrence with 95.5% sensitivity, 89.7% specificity, and 92.2% accuracy.

Conclusion: Methylation markers effectively classified HCC patients by de novo recurrence risk, enhancing prediction accuracy and potentially offering personalized management strategies.

Keywords: Biomarker; DNA methylation; Hepatocellular carcinoma; Machine learning; Recurrence.

Figure 1.

HCC classification based on DNA methylation using consensus clustering. (A) Co-classification matrix for K=2 featuring the most variable 3,000 probes in tumor samples. The pink bar represents methylation group 1 (MG1), and the grey bar indicates methylation group 2 (MG2). In this heatmap of the clustered consensus matrix, both rows and columns represent tumor samples, with colors indicating the frequency of co-clustering across multiple iterations of k-means clustering. Dark blue represents samples that consistently cluster, while white indicates samples that rarely cluster. The intensity of the color reflects the frequency of co-clustering, with darker shades indicating higher clustering consistency. The color bar on top shows sample groupings within each methylation group. (B) Kaplan–Meier plots of RFS between the two HCC subgroups, highlighting the significance of P-values. (C) Box plot illustrating the methylation β-value of four candidate markers across normal liver (N), MG1, and MG2 samples. P-values are determined using an unpaired t-test. ^****Represents P-value <0.0001. (D) T-distributed Stochastic Neighbor Embedding (t-SNE) analysis using four candidate markers. DMP, differentially methylated probe; HCC, hepatocellular carcinoma.

Figure 2.

Validation of four HCC prognostic candidates in two independent HCC cohorts. (A, B) Methylation heatmap of prognostic candidates in the Validation cohort 1 (Pre-MG1: 31 samples, PreMG2: 32 samples) (A) and the TCGA cohort (Pre-MG1: 82 samples, PreMG2: 135 samples) (B). X-axis is liver tumor patients and y-axis is methylation probe. In the TCGA cohort (B), only one CpG site (cg10544510) was used due to the absence of two methylation panel members in the TCGA dataset, which utilized a different version of the methylation chip (TCGA: 450K) compared to Validation cohort 1 (850K). (C, D) Kaplan–Meier plots for recurrence-free survival in the Validation cohort 1 (C) and TCGA cohort (D), with P-value significance indicated. HCC, hepatocellular carcinoma; TCGA, the cancer genome atlas.

Figure 3.

Analysis of methylation in Validation cohort 2. (A) Kaplan–Meier plot for recurrence-free survival in the Validation cohort 2. (B) Bar graph showing recurrence rates in the Validation cohort 2, with black indicating recurrence and grey representing non-recurrence, comparing the outcomes between MG1 and MG2. MG, methylation group.