Minichmayr Iris K, Plan Elodie L, Weber Benjamin, Ueckert Sebastian
Department of Clinical Pharmacology, Medical University of Vienna, Vienna, Austria.
Department of Pharmacy, Uppsala University, Uppsala, Sweden.
Pharm Res. 2025 Jan;42(1):123-135. doi: 10.1007/s11095-024-03791-2. Epub 2024 Dec 19.
Nonalcoholic fatty liver disease (NAFLD) comprises multiple heterogeneous pathophysiological conditions commonly evaluated by suboptimal liver biopsies. This study aimed to elucidate the role of 13 diverse histological liver scores in assessing NAFLD disease activity using an in silico pharmacometric model-based approach. We further sought to investigate various noninvasive patient characteristics for their ability to reflect all 13 histological scores and the NAFLD activity score (NAS).
A histological liver score model was built upon 13 biopsy-based pathological features (binary and categorical scores) from the extensive NASH-CRN (Nonalcoholic Steatohepatitis-Clinical Research Network) observational NAFLD Database study (n = 914 adults) using the concept of item response theory. The impact of 69 noninvasive biomarkers potentially reflecting NAFLD activity was quantitatively described across the entire spectrum of all 13 histological scores.
The model suggested that four different disease facets underlie the cardinal NAFLD features (steatosis, inflammation, hepatocellular ballooning (= NAS); fibrosis; highest correlations: corr = 0.69/corr = 0.62/corr = 0.60). The 13 histological liver scores were best described by contrasting noninvasive biomarkers: Age and platelets best reflected the fibrosis score, while alanine and aspartate aminotransferase best described the NAS, with diverging contributions of the three individual NAS components to the results of the overall NAS.
An in silico histological liver score model allowed to simultaneously quantitatively analyze 13 features beyond NAS and fibrosis, characterizing different disease facets underlying NAFLD and revealing the contrasting ability of 69 noninvasive biomarkers to reflect the diverse histological (sub-)scores.
非酒精性脂肪性肝病（NAFLD）包含多种异质性病理生理状况，通常通过不太理想的肝脏活检进行评估。本研究旨在使用基于计算机药理学模型的方法，阐明13种不同的肝脏组织学评分在评估NAFLD疾病活动中的作用。我们还进一步研究了各种非侵入性患者特征反映所有13种组织学评分和NAFLD活动评分（NAS）的能力。
基于广泛的非酒精性脂肪性肝炎临床研究网络（NASH-CRN）观察性NAFLD数据库研究（n = 914名成年人）中的13种基于活检的病理特征（二元和分类评分），利用项目反应理论的概念建立了肝脏组织学评分模型。在所有13种组织学评分的整个范围内，定量描述了69种可能反映NAFLD活动的非侵入性生物标志物的影响。
该模型表明，NAFLD的主要特征（脂肪变性、炎症、肝细胞气球样变（=NAS）；纤维化）基于四个不同的疾病方面；相关性最高：corr = 0.69/corr = 0.62/corr = 0.60）。通过对比非侵入性生物标志物可以最好地描述这13种肝脏组织学评分：年龄和血小板最能反映纤维化评分，而丙氨酸和天冬氨酸转氨酶最能描述NAS，NAS的三个个体组成部分对总体NAS结果有不同的贡献。
基于计算机的肝脏组织学评分模型能够同时对NAS和纤维化以外的13个特征进行定量分析，表征NAFLD潜在的不同疾病方面，并揭示69种非侵入性生物标志物反映不同组织学（亚）评分的对比能力。
