Hepatocellular carcinoma (HCC) is a leading global cause of cancer-related mortality. While curative resection is pivotal, high postoperative recurrence rates remain a major challenge. Adjuvant immune checkpoint inhibitors (ICIs) show promise in improving outcomes, but biomarkers to identify patients who will benefit are lacking. Current clinicopathological risk factors for minimal residual disease (MRD) are suboptimal in sensitivity and specificity.

Circulating tumor DNA (ctDNA) analysis, reflecting real-time tumor dynamics, offers a promising approach for MRD detection. This study focuses on the methylation status of GNB4 and Riplet-genes located within HCC-associated CpG islands-using a bespoke bisulfite-conversion and qPCR assay to sensitively detect methylated alleles, thereby enabling MRD monitoring.

To clinically validate this approach, we will conduct a prospective, multicenter cohort study assessing the predictive value of serial *GNB4/Riplet* methylation testing for recurrence and adjuvant therapy benefit.

Hepatocellular carcinoma (HCC) is a leading global malignancy and a primary cause of mortality in patients with chronic liver disease. While curative resection offers a critical treatment strategy, postoperative recurrence remains a major obstacle. Emerging evidence suggests adjuvant therapy with immune checkpoint inhibitors (ICIs), such as PD-1/PD-L1 inhibitors, may improve disease-free survival; however, identifying the patient subgroup most likely to benefit remains challenging. Theoretically, patients at high risk for minimal residual disease (MRD) post-surgery represent the ideal target for adjuvant treatment. Current clinical practice relies on indirect pathological surrogates of MRD-such as microvascular invasion, poor differentiation, or satellite nodules-which lack sufficient sensitivity and specificity.

Cell-free DNA (cfDNA), with its short half-life, dynamically mirrors tumor evolution, making it a promising tool for monitoring recurrence. The genes GNB4 and Riplet are located within CpG islands strongly associated with hepatocarcinogenesis. We developed specific primers and fluorescent probes targeting their bisulfite-converted sequences to enable selective PCR-based detection of methylated alleles. Monitoring MRD via *GNB4/Riplet* methylation status thus offers a potential method for dynamic assessment of tumor progression and recurrence, optimizing patient risk stratification and guiding therapeutic decisions.

To evaluate this approach, we will conduct a prospective, multi-center cohort study to investigate the predictive value of MRD for recurrence and adjuvant therapy benefit. The study comprises two cohorts: 1) an Active Surveillance Cohort, undergoing a pre-operative MRD assessment followed by regular post-operative surveillance and serial MRD testing; and 2) an Adjuvant Therapy Cohort, receiving pre-operative MRD assessment followed by standard adjuvant PD-1 inhibitor therapy alongside serial MRD monitoring. This study aims to validate a more effective tool for predicting recurrence and to identify patients most likely to benefit from adjuvant immunotherapy.

Inclusion Criteria:

• Age between 18 and 75 years, inclusive, regardless of gender.
• Newly diagnosed, treatment-naïve patients with HCC.
• Received radical treatments, such as liver resection or microwave ablation.
• Combine at least one of the risk factors for tumor recurrence, such as microvascular/macrovascular invasion, poor differentiation, satellite nodules, multiple tumors, and tumor diameter greater than 5 cm.
• Child-Pugh liver function score ≤ 7.
• Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1.
• Absence of severe organic diseases affecting the heart, lungs, brain, or other major organs.

Exclusion Criteria:

• History of other malignancies.
• Recurrent HCC.
• Prior systemic therapy for HCC.
• Unable to complete the follow-up and dynamic MRD monitoring.
• Having an immune deficiency disorder.
• Allergic to PD-1 inhibitors or unable to tolerate related treatments.