利用前瞻型研究探討血液DNA甲基化標記用於B型肝炎相關之肝細胞癌風險評估的可用性

Abstract

研究背景與目的： 異常的DNA甲基化會在癌症生成過程中在腫瘤組織以及正常組織中累積。這種表觀遺傳變化的累積可以反映過去暴露於感染因子的影響，且顯示與癌症風險具有相關。本篇研究的目的欲透過前瞻性的研究，證明B型肝炎相關的肝細胞癌可以經由表觀遺傳這種新型態的標記來進行預測。 材料與方法： 研究對象選自先前具有病毒因子相關標記的巢氏病例對照研究。利用焦磷酸定序(pyrosequencing)定量先前選取的五支探針之白血球DNA甲基化程度，共有137名肝細胞癌病患以及258名健康對照。使用羅吉斯迴歸建構甲基化分數。進一步，使用Odds ratios、ROC以及曲線下面積評估甲基化分數對於預測肝細胞癌的潛在有用性。 結果： 在137名肝細胞癌病例中，採血時間距離肝細胞癌確診之中位數為7.08年。比較病例與對照組之Q4以及Q1組別，發現位於PRF1基因的cg22900360_2甲基化程度在單變項分析中顯示與肝細胞癌有顯著性的危險(OR=2.058, 95% CI=1.13-3.75)，同樣的在多變項分析中也有達到顯著(OR=2.469, 95% CI=1.18-5.17)。雖然位於IFI44L基因的cg05696877_1在多變項分析中並沒有顯著的結果，但在單變項分析中顯示出與肝細胞癌有顯著的負相關。利用這兩支探針建構出甲基化分數，可區分出病例組以及對照組(AUC=0.6200)。將本研究的甲基化分數和REACH-B Score結合，並比較單一使用REACH-B Score的AUC，發現結合後預測能力有些微提升並且兩者之間有達到顯著差異(AUC: 0.7083 vs 0.6542; p=0.0126)。 結論： 血液當中的甲基化標記被評估為可能有助於鑑定高風險群的慢性B型肝炎帶原者是否發展成肝細胞癌。

Background and Aims Aberrant DNA methylation accumulates in tumorous tissues and also in normal tissues during carcinogenesis. Accumulation of such epigenetic changes can reflect past exposure to infectious agents, and was shown to be associated with cancer risk.The aim of this study was to demonstrate, by a prospective analysis, that the risk of hepatitis B virus (HBV)-related hepatocellular carcinoma (HCC) can be predicted by such a new type of cancer risk marker.

Materials and Methods Study subjects were recruited from a previous nested case-control study on the role of HBV DNA and genotype in the etiology of HCC. Methylation levels of five preselected genes were quantified with white blood cell DNA by pyrosequencing for 137 cases with HCC and 258 control subjects. We used logistic regression to construct a weighted composite methylation score. Odds ratios, receiver operating characteristic curve and area under the curve (AUC) were used to evaluate the potential usefulness of the score for HCC prediction.

Results Among the 137 HCC cases, the median of the time interval between blood draw and HCC diagnosis was 7.08 years. The highest (vs lowest) quartile of the (PRF1) methylation levels (cg22900360_2) showed a significant univariate odds ratio (OR) (95% CI) (2.058 (1.13-3.75)) and a multivariate-adjusted OR (2.469 (1.18-5.17)) of HCC. Although not significant in multivariate analysis, a significant inverse association with HCC was seen for (IFI44L) methylation levels (cg05696877_1). A methylation score, constructed using these two probes, could distinguish between HCC cases and controls (AUC=0.6200). Combining this methylation score with REACH-B score resulted in a slight but significant improvement in the discriminatory power, compared with REACH-B score alone (AUC: 0.7083 vs 0.6542; p=0.0126).

Conclusions Assessment of blood-based methylation maker is potentially useful for the identification of HBV carriers at high risk for HCC development.