分析代謝圖譜鑑別B型肝炎帶原者逐步發展為肝細胞癌的代謝變化

Abstract

背景：肝細胞癌占全球癌症死因的第三位，B型肝炎病毒是肝細胞癌的主要危險因子。儘管有關病毒因子已有大量信息，但目前對於肝細胞癌多步驟病理生理學變化的理解仍然是有限的。本研究的目的是透過代謝組學方法鑑定潛在的生物標記，以反映了從健康的B型肝炎病度帶原狀態到肝硬化再到肝細胞癌進展期間的代謝改變。 材料與方法：629位B型肝炎帶原的男性是來自台灣公務員世代追蹤研究的受試者，在長達22年的追蹤後，確診了56位肝細胞癌患者。透過問卷調查的方式獲得人口統計學、生活習慣和疾病史的資訊，並使用核磁共振儀測量基線時血漿樣品的代謝圖譜。應用偏最小平方判別分析(PLS-DA)進行數據分析，以鑑別與健康的HBV帶原者和肝細胞癌患者有所區別的代謝小分子。進行Cox回歸估計風險比(HR)和95%信賴區間。對於潛在的生物標記繪製ROC曲線和曲線下的對應面積(AUC)。 結果：總共鑑別出32個代謝小分子。透過NMR光譜的分析顯示，在被診斷約3年之前，NMR代謝圖譜可以區分健康的HBV帶原者和肝細胞癌病患。其中有4個代謝小分子在健康帶原者到肝硬化再到肝細胞癌的過程中發展一致且呈現顯著的正相關，分別是麩胺酸、丙酮酸、檸檬酸、酪胺酸。以這4個代謝小分子組合所生成的AUC為0.7601(95% 信賴區間：0.6962-0.8241)，用於區分肝硬化和健康的HBV帶原狀態；並且AUC為0.7993(95% 信賴區間：0.6926-0.9060)用於區分發展為肝細胞癌的肝硬化患者和未發展為肝細胞癌的肝硬化患者 結論：針對無症狀的HBV帶原者，代謝平台或許可用於在早期診斷肝性化及肝細胞癌，甚至是臨床診斷前的風險評估。

Background: Hepatocellular carcinoma (HCC) is the third leading cause of cancer-related death worldwide. Hepatitis B virus (HBV) is the main cause of HCC. Despite of a wealth of information on viral factors, the understanding of the pathophysiological changes in the multistep hepatocellular carcinogenesis is limited. The purpose of this study was to identify potential biomarkers which reflect the metabolic alterations during progression from healthy HBV carrier state to liver cirrhosis to HCC by metabolomics approach. Materials and Methods: A subset of 629 men with chronic HBV infection were recruited from an established cohort study of Taiwanese civil servants. Fifty-six cases with incident HCC were identified after up to 22 years of follow-up. Information on demographics, lifestyle habits, and medical history was obtained from questionnaire. Metabolomics profiles of baseline plasma samples were determined using nuclear magnetic resonance (NMR) spectroscopy. Partial least squares discriminant analysis (PLS-DA) was applied to analyze the profiling data to identify the distinguishing metabolites of healthy HBV carrier and HCC. Cox regression was performed to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). ROC curves and corresponding area under the curve (AUC) were generated for potentail biomarkers. Results: A total of 32 metabolites were identified. NMR spectroscopic multivariate analysis revealed that NMR metabolic profile can discriminate, healthy HBV carriers and HCC cases prior to diagnosis of about 3 years. Four metabolites, including Glutamate, Pyruvate, Citrate, and Tyrosine, were consistently, positively associated with progression from healthy HBV carrier state to liver cirrhosis to HCC. The four metabolites combination yielded an AUC of 0.7601 (95% CI: 0.6962-0.8241) to discriminate liver cirrhosis from healthy HBV carrier state, and had an AUC of 0.7993 (95% CI: 0.6926-0.9060) for comparison of cirrhotic patients who developed HCC vs those who had liver cirrhosis and did not develop HCC. Conclusion: Metabolic patterns may be used to detect liver cirrhosis and HCC in early stages, even several years prior to clinical diagnosis, among asymptomatic HBV carriers.