Center for Integrative Medicine, Beijing Ditan Hospital, Capital Medical University, Beijing, P.R. China.
Department of General Surgery, Huashan Hospital, Fudan University, Shanghai, P.R. China.
Cancer Immunol Res. 2021 Mar;9(3):337-347. doi: 10.1158/2326-6066.CIR-20-0616. Epub 2021 Jan 11.
A number of staging systems have been developed to predict clinical outcomes in hepatocellular carcinoma (HCC). However, no general consensus has been reached regarding the optimal model. New approaches such as machine learning (ML) strategies are powerful tools for incorporating risk factors from multiple platforms. We retrospectively reviewed the baseline information, including clinicopathologic characteristics, laboratory parameters, and peripheral immune features reflecting T-cell function, from three HCC cohorts. A gradient-boosting survival (GBS) classifier was trained with prognosis-related variables in the training dataset and validated in two independent cohorts. We constructed a 20-feature GBS model classifier incorporating one clinical feature, 14 laboratory parameters, and five T-cell function parameters obtained from peripheral blood mononuclear cells. The GBS model-derived risk scores demonstrated high concordance indexes (C-indexes): 0.844, 0.827, and 0.806 in the training set and validation sets 1 and 2, respectively. The GBS classifier could separate patients into high-, medium- and low-risk subgroups with respect to death in all datasets ( < 0.05 for all comparisons). A higher risk score was positively correlated with a higher clinical stage and the presence of portal vein tumor thrombus (PVTT). Subgroup analyses with respect to Child-Pugh class, Barcelona Clinic Liver Cancer stage, and PVTT status supported the prognostic relevance of the GBS-derived risk algorithm independent of the conventional tumor staging system. In summary, a multiparameter ML algorithm incorporating clinical characteristics, laboratory parameters, and peripheral immune signatures offers a different approach to identify patients with the greatest risk of HCC-related death.
已经开发出许多分期系统来预测肝细胞癌 (HCC) 的临床结局。然而，对于最佳模型尚未达成共识。新方法，如机器学习 (ML) 策略，是整合来自多个平台的风险因素的强大工具。我们回顾性地审查了三个 HCC 队列的基线信息，包括临床病理特征、实验室参数和反映 T 细胞功能的外周免疫特征。梯度提升生存 (GBS) 分类器在训练数据集中使用与预后相关的变量进行训练，并在两个独立的队列中进行验证。我们构建了一个包含一个临床特征、14 个实验室参数和五个外周血单个核细胞中 T 细胞功能参数的 20 个特征 GBS 模型分类器。GBS 模型衍生的风险评分在训练组和验证组 1 和 2 中表现出较高的一致性指数 (C 指数)：分别为 0.844、0.827 和 0.806。GBS 分类器可以根据所有数据集的死亡情况将患者分为高、中、低风险亚组（所有比较均 < 0.05）。较高的风险评分与较高的临床分期和门静脉癌栓 (PVTT) 的存在呈正相关。根据 Child-Pugh 分级、巴塞罗那临床肝癌分期和 PVTT 状态进行的亚组分析支持 GBS 衍生风险算法独立于传统肿瘤分期系统的预后相关性。总之，纳入临床特征、实验室参数和外周免疫特征的多参数 ML 算法为识别 HCC 相关死亡风险最大的患者提供了一种不同的方法。
