Li Kang, Zang Chaoran, Zhao Yanan, Guo Dandan, Shi Wanting, Mei Tingting, Li Ang, Zhang Yonghong
Biomedical Information Center, Beijing You'An Hospital, Capital Medical University; Beijing 100069, People's Republic of China.
Beijing Key Laboratory (BZ0373), Beijing You'An Hospital, Capital Medical University, Beijing 100069, People's Republic of China.
Oncologist. 2025 Aug 4;30(8). doi: 10.1093/oncolo/oyae292.
Early screening of hepatocellular carcinoma (HCC) is strongly recommended for hepatitis B virus (HBV)-infected patients. We aimed to develop and validate a predictive nomogram based on HCC occurrence trajectory for screening precancerous patients with HCC.
Peripheral blood mononuclear cells (PBMC) samples from 22 patients with HCC with their precancerous stage (n = 55) and 18 healthy controls were measured using HumanMethylation EPIC BeadChip assay. HCC trajectory was assessed by pseudotime based on TimeAx algorithm and chronological time. The 43 candidate CpG sites were selected from the methylation signature and measured using multiplex bisulfite sequencing in a retrospective cohort of HBV-infected patients (n = 604). A 5-CpG-classifier was built using the LASSO Cox regression model, based on the association between the methylation level of every CpG and the duration from enrollment to HCC occurrence of individual patient. We validated the risk stratification and predictive accuracy of this classifier in both the primary cohort (n = 300) and independent validation cohort (n = 304).
Pseudotime and chronological time of HCC trajectory analysis revealed that the PD-1/PD-L1 pathway underwent changes in the precancerous stage. Based on the trajectory of methylation signature, we built a 5-CpG-classifier which remained powerful and independent predictive efficiency after stratified analysis by clinicopathological risk factors in both primary cohort and independent validation cohort. A predicting nomogram including the 5-CpG-classifier was constructed after multivariate analysis. One-year cumulative hazard of HCC in low- and high-risk groups of HBV-infected patients was 3.0% (0.1%-5.8%) and 17.90% (11.00%-24.3%) (P < .0001) in primary cohort, 4.5% (1.20%-7.80%) and 27.3 (18.90-34.90) (P < .0001) in the independent validation cohort.
One-year before HCC was a critical period of transitional time when parts of the methylation profile underwent shifting toward HCC like. The nomogram could identify precancerous stage patients with HCC who should be screened for early diagnosis and intervention.
强烈建议对乙型肝炎病毒（HBV）感染患者进行肝细胞癌（HCC）的早期筛查。我们旨在开发并验证一种基于HCC发生轨迹的预测列线图，用于筛查HCC癌前患者。
使用HumanMethylation EPIC BeadChip检测法对22例HCC患者及其癌前阶段（n = 55）的外周血单个核细胞（PBMC）样本以及18例健康对照进行检测。基于TimeAx算法和时间顺序，通过伪时间评估HCC轨迹。从甲基化特征中选择43个候选CpG位点，并在一组回顾性HBV感染患者（n = 604）中使用多重亚硫酸氢盐测序进行检测。基于每个CpG的甲基化水平与个体患者从入组到发生HCC的持续时间之间的关联，使用LASSO Cox回归模型构建了一个5-CpG分类器。我们在主要队列（n = 300）和独立验证队列（n = 304）中验证了该分类器的风险分层和预测准确性。
HCC轨迹分析的伪时间和时间顺序显示，PD-1/PD-L1通路在癌前阶段发生了变化。基于甲基化特征轨迹，我们构建了一个5-CpG分类器，在主要队列和独立验证队列中，经临床病理危险因素分层分析后，该分类器仍具有强大且独立的预测效能。多因素分析后构建了一个包含5-CpG分类器的预测列线图。在主要队列中，HBV感染患者低风险组和高风险组的HCC一年累积风险分别为3.0%（0.1%-5.8%）和17.90%（11.00%-24.3%）（P <.0001），在独立验证队列中分别为4.5%（1.20%-7.80%）和27.3（18.90-34.90）（P <.0001）。
HCC发生前一年是一个关键的过渡期，此时部分甲基化谱向HCC样状态转变。该列线图可识别HCC癌前阶段患者，应进行早期诊断和干预筛查。
