| 摘要: | 背景與目標：非酒精性脂肪肝疾病是世界最常見慢性肝疾病，其疾病定義是超過5%肝細胞有脂肪堆積，病人沒有過度飲酒及其他肝病或藥物使用。非酒精性脂肪肝在疾病組織學上進程包括單純性脂肪肝、非酒精性脂肪肝炎、肝纖維化到肝硬化。非酒精性脂肪肝疾病盛行率在男性約30-40%和女性15-20%，身體質量指數超過35 kg/m2病態性肥胖病人則高達90%。臨床上，病人肝臟切片是非酒精性脂肪肝疾病主要診斷的標準，但肝臟切片帶有併發症的風險，是否有取代性的方法來做診斷，仍是臨床努力的方向。現今已有數個非侵襲性血清標記被運用於評估非酒精性脂肪肝病人的肝纖維化狀態。此外，肝臟纖維掃描儀也是一種非侵襲性儀器來評估肝纖維化的情況。但是在病態性肥胖病人診斷上，仍未有適當的方式來診斷。根據先前的資料顯示，大約有25-30%非酒精性脂肪肝疾病患者會發展成非酒精性脂肪肝炎，而約有40-50%非酒精性脂肪肝炎患者會導致肝纖維化的產生。但是非酒精性脂肪肝疾病惡化的病理機轉仍不清楚，所以找出非酒精性脂肪肝炎診斷的預測性生物標記非常重要。最近非酒精性脂肪肝炎臨床研究網絡族群指出soluble interleukin-2 receptor alpha (IL2RA)增加肝纖維化嚴重性。但尚未有接受減重手術亞洲病態性肥胖病人肝臟檢體的研究。所以本論文主要是針對減重手術病人建立肝纖維化的評分系統及尋找評估非酒精性脂肪肝可能的預測因子為目標。材料與方法：此前瞻性族群於2016年10月到2020年12月在台北醫學大學附設醫院收錄200位病態性肥胖病人接受腹腔鏡袖狀胃切除術。手術前執行腹部超音波和肝臟纖維掃描儀。每位病人接受減重手術時由腹腔鏡接受楔形肝切片，而非酒精性脂肪性肝炎與肝纖維化則是利用組織染色方式做判定。在評估顯著肝纖維化和非酒精性脂肪肝炎之危險因子方面，是對於肝臟組織具有IL2RA陽性染色反應的淋巴細胞個數做計算。在探討分子生物標記預測因子，則是使用公開資料庫Gene Expression Omnibus (GEO)，GSE48452微陣列資料與臨床資料進行分析，找出可能的具診斷價值的生物標記。結果：我們總共收集了123位病人肝組織做分析，平均年齡35.5歲，平均身體質量指數40.6 kg/m2，87 (70.7%)位女性，其中11位 (8.9%)是沒有非酒精性脂肪肝疾病，46位 (37.4%)是單純性脂肪肝和66 位(53.7%)是非酒精性脂肪性肝炎。28 位 (22.8%)是第2期纖維化，14位 (11.4%)是第3期纖維化，2位 (1.6%)是第4期纖維化。將病人分成訓練組 (n=73)和驗證組 (n=50)，經由多變項分析發現肝硬度測量> 7 kPa 和 aspartate aminotransferase/platelet ratio index (APRI) >0.40是顯著肝纖維化的獨立因子。相較於其他非侵襲性指標，肝纖維掃描儀為基礎分數加權肝硬度測量> 7 kPa為2分和APRI >0.40為1分在訓練組和驗證組都顯示有較高曲線下的面積(0.854, P=0.0001; 0.785, P=0.0002)。非酒精性脂肪性肝炎組比沒有非酒精性脂肪性肝炎組有較大腰圍，較高IL2RA表現量，空腹血糖值，alanine aminotransferase (ALT), aspartate aminotransferase, gamma-glutamyltransferase，第2-4期纖維化，其他非侵襲性指標。經由多變項分析指出IL2RA表現量和ALT是非酒精性脂肪肝炎的獨立因子。IL2RA表現量預測非酒精性脂肪性肝炎之曲線下的面積是0.627 (臨界值：82，P = 0.0113)。在分子生物標記預測因子分析，GSE48452選擇單純性脂肪肝組和非酒精性脂肪肝炎組的不同基因表現量做比較發現差異基因 (FAS, RAD18, ME1, FANCI, CHST9, ITGAV, PIEZO2)可做為非酒精性脂肪肝炎候選標的基因。進一步分析30位病人顯示肝組織這些基因情況發現RAD18 (P=0.098), CHST9 (P=0.091), PIEZO2 (P=0.086), ME1 (P=0.072), ITGAV (P=0.085)在單純性脂肪肝F0-F1和非酒精性脂肪肝炎F0-F1分別有趨勢變化，沒有顯著差異。結論：在此論文中我們建立一個臨床分數系統結合肝硬度和APRI來預測病態性肥胖病人的肝纖維化狀態。這個系統可幫助臨床醫師在術前評估最適合接受減重手術之可能肝纖維化病態性肥胖病人。IL2RA與病態性肥胖病人非酒精性脂肪肝炎是顯著相關，可作為非酒精性脂肪肝炎診斷有用的單一生物標記。另外，生物資訊分析分子生物標記預測因子的方面，本研究中所鑑別出與非酒精性脂肪肝炎有關的生物標記皆有趨勢變化，可以進一步增加蒐集大型病態肥胖病人族群，並依據不同的基因體特徵(如基因甲基化與否)分群，探討非酒精性脂肪肝疾病的預測性生物標記和路徑。
Background and aims: Nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease in the world. NAFLD is defined as steatosis affecting > 5% of hepatocytes without excessive alcohol consumption, other liver disease or hepatotoxic drugs. The histological spectrum of NAFLD includes non-alcoholic fatty liver (NAFL; steatosis without hepatocellular injury), non-alcoholic steatohepatitis (NASH; steatosis with lobular inflammation and hepatocellular ballooning), fibrosis and cirrhosis. The prevalence rate of NAFLD has been reported to be 30–40% in men and 15–20% in women, and up to 90% of morbidly obese patients with a body mass index (BMI) higher than 35 kg/m2. Currently, liver biopsy is the gold standard for histological diagnosis of NAFLD. However, liver biopsy carries the risk of complications and another method for diagnosis is unmet need in clinical practice. Several non-invasive serum markers have been introduced to assess the degree of liver fibrosis in patients with NAFLD. Furthermore, transient elastography (FibroScan®) is a noninvasive, reproducible, and reliable method for predicting liver fibrosis. Nevertheless, the diagnostic accuracy of non-invasive serum markers and transient elastography in morbidly obese patients remain to be validated. Approximately 25-30% of people with NAFLD develop NASH and 40–50% of patients with NASH develop liver fibrosis. However, the pathogenesis of NAFLD progression is less well understood and predictive biomarkers for NASH diagnosis are crucial. Two recent studies in Nonalcoholic Steatohepatitis-Clinical Research Network cohorts showed soluble interleukin-2 receptor alpha (IL2RA) increased with fibrosis severity. Nevertheless, no hepatic study has been conducted in Asian morbidly obese patients underwent bariatric surgery. This study aimed to develop a scoring system to predict significant liver fibrosis and investigate the predictive factors of NASH in morbidly obese patients underwent bariatric surgery. Materials and Methods: This prospective cohort study enrolled 200 morbidly obese patients underwent laparoscopic sleeve gastrectomy at Taipei Medical University Hospital from October 2016 to December 2020. Fibroscan and ultrasonography were performed before surgery. A wedge liver biopsy was performed during surgery. The diagnosis of NASH and liver fibrosis was made histologically. We assessed the risk factors to be associated with significant liver fibrosis (fibrosis stage≧2) and NASH. Immunohistochemistry (IHC) of IL2RA was counted the number of lymphocytes with IL2RA immunoreactivity in 5 high power fields (400 x, total 1.19 mm2). On the other hand, predictors of molecular biomarkers were investigated by public database, Gene Expression Omnibus (GEO). We used the keywords “bariatric surgery”, “nonalcoholic steatohepatitis” to search the GEO repository for materials for our study. We selected Microarray results with accession numbers GSE48452 and the related clinical data. Results: Of the 123 patients with detailed liver histological data, mean age was 35.5 years, 87 (70.7%) were female, mean BMI was 40.6 kg/m2, 11 (8.9%), 46 (37.4%) and 66 (53.7%) had non-NAFLD, NAFL and NASH, 28 (22.8%), 14 (11.4%) and 2 (1.6%) patients were in fibrosis stage 2, 3 and 4 groups, respectively. Patients were divided into 2 groups: a derivation cohort (n=73) and a validation cohort (n=50). Multivariate analysis disclosed liver stiffness measurement (LSM) > 7 kPa and aspartate aminotransferase/platelet ratio index (APRI) >0.40 were associated with significant liver fibrosis in the derivation cohort. A weighted sum of Fibroscan-base score was: (2 for presence of LSM >7 kPa) + (1 for presence of APRI >0.40). For the prediction of significant liver fibrosis by image modalities and non-invasive serum markers, this newly established Fibroscan-base score yielded the highest area under receiver operating curve (0.854, P=0.0001; 0.785, P=0.0002) compared to LSM, ultrasonographic (US) fibrosis score, splenic arterial pulsatility index, Fibrosis-4 score (FIB-4 score), nonalcoholic fatty liver disease fibrosis score and APRI in the derivation and validation cohorts, respectively. Patients in the NASH group had larger waist circumference, higher IHC of IL2RA, fasting glucose, alanine aminotransferase (ALT), aspartate aminotransferase, gamma-glutamyltransferase, Fibrosis stage 2-4, APRI, FIB-4 score, fatty liver index, LSM, controlled attenuation parameter, US fatty and fibrosis score than non-NASH group. Multivariate analysis disclosed IHC of IL2RA and ALT were the independent factor associated with NASH. The area under receiver operating curve of IL2RA IHC for NASH was 0.627 at the cut-off value of 82 (P = 0.0113). In the analysis of predictors of molecular biomarkers, GSE48452 dataset contained 73 samples of human liver grouped into control (n=14), healthy obese (n=27), steatosis (n=14) and NASH (n=18). The differentially expressed genes between the steatosis and NASH groups were selected. We selected candidate target genes of NASH: FAS, RAD18, ME1, FANCI, CHST9, ITGAV, PIEZO2. In our cohort (n= 30), gene expression of liver tissue of RAD18 (P=0.098), CHST9 (P=0.091), PIEZO2 (P=0.086), ME1 (P=0.072), ITGAV (P=0.085) revealed a trend of different between NAFL with F0-F1 and NASH with F0-F1 groups, respectively. Conclusions: In this dissertation, we developed a simple clinical scoring system incorporating APRI and Fibroscan to predict significant liver fibrosis in morbidly obese patients. This scoring system can help physicians identify morbidly obesity patients with possible significant liver fibrosis most likely to benefit from bariatric surgery. IL2RA is significantly associated with NASH in morbidly obese patients and would be a useful single biomarker for NASH diagnosis. Furthermore, regarding predictors of molecular biomarkers on bioinformatics analysis, this study identified biomarkers had a trend of associated with NASH. Further large cohort studies and the clustering of genomic feature (such as DNA methylation) are warranted to investigate the predictive biomarkers and pathways of NAFLD in morbidly obese patients. |
