探索新型生物標記對於改善粥狀動脈硬化心血管疾病臨床風險評估之效用

Abstract

背景與目標：粥狀動脈硬化心血管疾病目前仍然是全世界非傳染性疾病造成的主要死因之一。利用心血管疾病風險因子建立預測模型來評估心血管疾病發生風險是目前在臨床照護和預防心血管疾病扮演很重要的醫療決策角色。但目前傳統的心血管風險因子卻仍無法準確估計病患的心血管疾病風險。利用尋找新型生物標記來目前改善傳統模型預測能力為具有潛力發展的方向。近年來，因為技術和設備上的進步，高通量分析實驗方式已運用在有效率地探索新型生物標記。此計畫的目的主要是尋找、驗證、且建立嶄新型生物指標運用在心血管疾病風險評估的效用。 方法：此研究分成兩個部分。第一部分是利用金山社區心血管疾病追蹤世代建立巢式病例對照研究方式，探討新型生物標記和發生中風的相關性。選取131位在追蹤期間發生中風病患和131位年齡和性別相符的對照組，利用液相層析串聯式質譜儀 (liquid chromatography and tandem mass spectrometry [LC-MS/MS]) 測量尿液中氧化壓力(oxidative stress)生物標記。第二部分是利用34位高膽固醇患者，於分別使用2 mg pitastatin及10 mg atorvastatin十二週前後，利用微陣列 (microarray) 方式測量免疫發炎細胞內微核醣核酸(microRNAs, miRNA)表現量差異性，藉以探索史他汀(statin)治療和膽固醇的變化對於發炎細胞產生的細胞內分子調控的作用和變化。 結果：在尿液氧化壓力生物指標研究中，發現尿液中生物指標前列腺素F2α(8-iso-PGF2α) 在中風患者尿中度顯著高於控制組 [median (interquartile range), 1.13 (2.23 – 4.36) ug/g creatinine versus 0.71 (1.34 – 3.02) ug/g creatinine, p=0.004]。尿液中8-iso-PGF2α 濃度每增加一個標準差濃度，中風風險會增加40% [per 1 standard deviation increase in log-transformed value, adjusted odds ratio, 1.40; 95% confidence interval (CI), 1.06 – 1.85; p=0.005] with a significant increasing trend across its quartiles (p for trend=0.016)]。進一步更發現將尿液8-iso-PGF2α 作為生物標記加入傳統中風預測模型中，可以有效增進目前傳統中風風險模型預測能力 (integrated discrimination improvement, 0.025; 95% CI, 0.006 – 0.045; p=0.005; net reclassification improvement, 19.8%; 95% CI, 4.6 – 35.1 %; p=0.011)。 在史他汀治療高膽固醇病患研究中，在12週治療後，我們發現miR-483-5p, miR-4667-5p, miR-1244, and miR-3609在循環血液單核細胞中的表現量有顯著變化 [qPCR-validated fold changes, 1.74 (95% confidence interval, 1.33 to 2.15), 1.61 (1.25 to 1.98), 1.61 (1.01 to 2.21), and 1.68 (1.19 to 2.17), respectively)。這四個微核醣核酸和低密度脂蛋白膽固醇治療後的變化並無相關性，亦和CRP無關聯性。根據miRSystem及miTALOS分析結果，這四個微核醣核酸，和RhoA相關的focal adhesion和TGF-β 訊號途徑有相關。而這兩個訊號途徑皆是發炎細胞進行發炎反應中重要的訊號途徑。這些新發現的miRNAs不僅可能可以做為發炎反應程度評估的生物標記，並有可能發展成對於粥狀動脈硬化心血管疾病病患的抗發炎治療方式之一。 結論：運用新型omic生物標記於心血管風險評估，可以讓心血管疾病風險評估和預防，在個人化醫療與公共衛生方面進入嶄新的階段。大型樣本的族群流行病學研究將是未來驗證並運用將新型生物標記於重要的工作。

Background and objectives: To date, knowledge gaps have, however, existed in comprehensive assessment for cardiovascular risk. The discovery of novel biomarkers could be a key to improving the clinical risk prediction of atherosclerotic cardiovascular diseases with the help of recent advances in high-throughput laboratory technique. Hence, the purpose of this research was aimed at the identification and application of novel biomarkers using and high-throughput experiment platforms. Project 1: Urinary Biomarkers of Oxidative and Nitrosative Stress and the Risk for Incident Stroke: A Nested Case-Control Study from a Community-Based Cohort Methods: In this nested case-control study, cases consisted of 131 participants who were free of stroke at screening and experienced incident stroke during the follow-up period. Controls were 1:1 time-matched for age and sex. Baseline levels of urinary creatinine-indexed biomarkers were measured using liquid chromatography-tandem mass spectrometry, including 8-iso-prostaglandin F2α (8-iso-PGF2α), 4-hydroxynonenal conjugate with mercapturic acid, 8-hydroxydeoxyguanosine and 8-nitroguanine. Results: The levels of urinary 8-iso-PGF2α in stroke cases were higher than in controls [median (interquartile range), 1.13 (2.23 – 4.36) ug/g creatinine versus 0.71 (1.34 – 3.02) ug/g creatinine, p=0.004]. After adjusting cardiovascular risk factors, the association remained that higher level of urinary 8-iso-PGF2α entailed the greater risk for incident stroke [per 1 standard deviation increase in log-transformed value, adjusted odds ratio, 1.40; 95% confidence interval (CI), 1.06 – 1.85; p=0.005] with a significant increasing trend across its quartiles (p for trend=0.016). After adding urinary 8-iso-PGF2α, the prediction model not only improved discrimination between participants with or without incident stroke (integrated discrimination improvement, 0.025; 95% CI, 0.006 – 0.045; p=0.005), but enhanced stroke risk stratification (net reclassification improvement, 19.8%; 95% CI, 4.6 – 35.1 %; p=0.011). Project 2: miRNAome profiling of immune cells in search of novel inflammatory biomarkers for atherosclerotic cardiovascular diseases Methods: We recruited 19 hypercholesterolemic patients initiating 2 mg/day pitavastatin and 15 initiating 10 mg/day atorvastatin (mean age 63.7 years, 79% male, 56% with coronary artery disease), and performed microarray-based profiling of 1,733 mature miRNAs in peripheral blood mononuclear cells (PBMCs) before and after 12 weeks of statin treatment. MiRNAs were defined as differentially expressed after statin treatment if their fold changes were >1.50 or <0.67, after validation using quantitative polymerase chain reaction (qPCR). The miRSystem and miTALOS platforms were utilized for pathway analysis. Results: We found that statins induced differential expression of miR-483-5p, miR-4667-5p, miR-1244, and miR-3609, with qPCR-validated fold changes of 1.74 (95% confidence interval, 1.33 to 2.15), 1.61 (1.25 to 1.98), 1.61 (1.01 to 2.21), and 1.68 (1.19 to 2.17), respectively. The fold changes of the four miRNAs were not correlated with changes in low-density lipoprotein cholesterol or CRP, after sex, age, and statin type were adjusted. Pathway analysis showed that RhoA and transforming growth factor-β signaling pathways might be regulated by these four miRNAs. Conclusions: The identification and application of novel biomarkers remain the unmet need for comprehensive cardiovascular risk assessment. With accumulating relevant evidence of scientific, legal, and ethical research, multi-omic approaches would revolutionize precision medicine and public health for management and prevention of cardiovascular disease in general population.