肝细胞癌（HCC）是主要的原发性肝癌，全球新发案例中占5%、与癌症相关死亡中占8.4%。肝细胞癌显示出一系列的环境危险因素（病毒性慢性感染、黄曲霉毒素暴露、酒精和非酒精性脂肪肝疾病），导致分子复杂性和异质性增加，为流行病学负担的上升、预后不良和治疗选项不满意作出贡献。肝癌的出现（即肝癌发生过程）是一个多步骤和复杂的过程，涉及许多（表观）遗传改变和表型特征，即所谓的癌症标志物。"多态微生物群"、"表观遗传重组"、"细胞衰老"和"解锁表型可塑性"是癌症生物学中的趋势标志物/促进特征。由于肝细胞癌的主要分子驱动因子仍无法用药物靶向治疗，化学诱导的体内肝癌发生模型成为预临床研究中有用的工具。因此，本综述旨在回顾化学诱导的啮齿动物肝癌模型的基本特征，评估其在传统和新的肝癌标志物/促进特征方面的持久转化价值。我们收集了有关非肝硬化、炎症、酒精性肝病和非酒精性脂肪肝相关肝细胞癌模型的最新临床前证据，证明这些生物化学检测确实表达了最近新增的标志物，同时反映了经典和新的肿瘤特征之间的相互作用。我们的综述证明，这些方案对于转化前临床应用仍然具有价值，因为它们再现了癌症科学中的趋势特征。进一步的"组学"方法是必要的，同时鼓励进行多模型研究，以避免"模型偏差"的反应。版权所有 © 2023 Elsevier Inc. 保留所有权利。

Hepatocellular carcinoma (HCC), the main primary liver cancer, accounts for 5 % of all incident cases and 8.4 % of all cancer-related deaths worldwide. HCC displays a spectrum of environmental risk factors (viral chronic infections, aflatoxin exposure, alcoholic- and nonalcoholic fatty liver diseases) that result in molecular complexity and heterogeneity, contributing to a rising epidemiological burden, poor prognosis, and non-satisfactory treatment options. The emergence of HCC (i.e., hepatocarcinogenesis) is a multistep and complex process that addresses many (epi)genetic alterations and phenotypic traits, the so-called cancer hallmarks. "Polymorphic microbiomes", "epigenetic reprogramming", "senescent cells" and "unlocking phenotypic plasticity" are trending hallmarks/enabling features in cancer biology. As the main molecular drivers of HCC are still undruggable, chemically induced in vivo models of hepatocarcinogenesis are useful tools in preclinical research. Thus, this narrative review aimed at recapitulating the basic features of chemically induced rodent models of hepatocarcinogenesis, eliciting their permanent translational value regarding the "classic" and the "new" cancer hallmarks/enabling features. We gathered state-of-art preclinical evidence on non-cirrhotic, inflammation-, alcoholic liver disease- and nonalcoholic fatty liver-associated HCC models, demonstrating that these bioassays indeed express the recently added hallmarks, as well as reflect the interplay between classical and new cancer traits. Our review demonstrated that these protocols remain valuable for translational preclinical application, as they recapitulate trending features of cancer science. Further "omics-based" approaches are warranted while multimodel investigations are encouraged in order to avoid "model-biased" responses.Copyright © 2023 Elsevier Inc. All rights reserved.