文献速递肝病消化道肿瘤与早癌

基于机器学习的影像组学模型在预测肝细胞癌免疫治疗结局方面优于临床生物标志物

Machine learning based radiomic models outperform clinical biomarkers in predicting outcomes after immunotherapy for hepatocellular carcinoma.
PubMed 文献 · J Hepatol · 2025-04-17

本研究回顾性纳入两个国际中心共152例接受阿替利珠单抗联合贝伐珠单抗(A/B)治疗不可切除肝细胞癌(HCC)的患者,利用治疗前CT深度学习自动分割全肝并提取影像组学特征,结合临床变量构建预测12个月死亡的影像组学及影像组学-临床整合模型。整合模型在帝国理工学院伦敦(ICL)队列的AUC为0.89(95%CI 0.75-0.99),在巴黎公共医疗集团(AP-HP)独立队列为0.75(95%CI 0.64-0.85),均优于BCLC分期(AUC 0.61,p<0.001)和ALBI分级(AUC 0.48,p<0.001);整合模型分层的高危患者中位总生存期(OS)和无进展生存期(PFS)显著缩短,低危患者免疫检查点抑制剂应答率显著更高(35.6% vs 21.4%;p=0.038)。

要点

  • 整合影像组学-临床模型预测12个月死亡的AUC在ICL队列为0.89、AP-HP队列为0.75,优于BCLC分期和ALBI分级。
  • 整合模型分层高危患者中位OS在ICL为5.6个月 vs 28.2个月(p<0.001),AP-HP为5.8个月 vs 15.7个月(p<0.001)。
  • 低危患者免疫检查点抑制剂应答率35.6% vs 高危患者21.4%(p=0.038)。

核心结论

  • 基于影像组学的模型可预测晚期HCC患者接受免疫治疗的生存结局和应答。
  • 深度学习联合机器学习可对患者进行分层,支持精准治疗策略。

研究设计与样本

  • 回顾性纳入两个国际中心共152例接受阿替利珠单抗联合贝伐珠单抗治疗的不可切除HCC患者。
  • 利用治疗前CT深度学习自动分割生成全肝掩膜,提取影像组学特征并结合临床变量预测A/B治疗后12个月死亡。
  • 在帝国理工学院伦敦(ICL)队列中使用7种机器学习模型联合13种特征选择技术开发影像组学及影像组学-临床整合模型。
  • 采用K-means聚类识别高危和低危组,并在巴黎公共医疗集团(AP-HP)独立队列中评估模型性能。

主要发现

  • 整合影像组学-临床模型在ICL队列的AUC为0.89(95%CI 0.75-0.99),在AP-HP队列为0.75(95%CI 0.64-0.85)。
  • 整合模型优于BCLC分期(AUC 0.61,p<0.001)和ALBI分级(AUC 0.48,p<0.001)。
  • 整合模型分层高危患者中位OS在ICL为5.6个月 vs 低危28.2个月(p<0.001),在AP-HP为5.8个月 vs 15.7个月(p<0.001)。
  • 整合模型分层高危患者中位PFS在ICL为2.4个月 vs 14.6个月(p<0.001),在AP-HP为2.1个月 vs 6.1个月(p=0.046)。
  • 低危患者免疫检查点抑制剂应答率为35.6% vs 高危患者21.4%(p=0.038)。

关键数值与推荐

  • 多变量分析中影像组学分组是OS的最强预测因子(HR 3.22,95%CI 1.99-5.20;p<0.001)。
  • 多变量分析中影像组学分组是PFS的预测因子(HR 1.82,95%CI 1.18-2.80;p=0.010)。

临床怎么用

  • 原文未据此给出推荐。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND & AIMS: Atezolizumab plus bevacizumab (A/B) is a first-line therapy for unresectable hepatocellular carcinoma (HCC). Only a small proportion of patients respond to treatment. This study integrated radiomic and clinical data derived from routine pre-treatment imaging to predict outcomes after immunotherapy. METHODS: A total of 152 patients from two international centres receiving A/B were retrospectively reviewed. Deep learning autosegmentation generated whole liver masks from pre-treatment CTs. Radiomic features combined with clinical variables were used to predict 12-month mortality post A/B. Radiomic and integrated radiomic-clinical models were developed using seven machine learning models in combination with 13 feature selection techniques in the Imperial College London (ICL) cohort. K-means clustering identified high- and low-risk groups and predicted overall survival (OS), progression-free survival (PFS) and response. Model performance was assessed in the independent Assistance Publique-Hôpitaux de Paris (AP-HP) cohort. RESULTS: The integrated radiomic-clinical model outperformed BCLC stage (AUC 0.61, p <0.001) and ALBI grade (AUC 0.48, p <0.001) in the ICL (AUC 0.89, 95% CI 0.75-0.99) and AP-HP (AUC 0.75, 95% CI 0.64-0.85) cohorts. Integrated model-stratified high-risk patients had significantly shorter median OS (ICL: 5.6 months vs. 28.2 months; p <0.001; AP-HP: 5.8 months vs. 15.7 months; p <0.001) and PFS (ICL: 2.4 months vs. 14.6 months; p <0.001; AP-HP: 2.1 months vs. 6.1 months; p = 0.046). Low-risk patients had significantly higher immune checkpoint inhibitor response rates compared to high-risk patients (35.6% vs. 21.4%; p = 0.038). In multivariable analysis, radiomic group was the strongest predictor of OS (hazard ratio 3.22, 95% CI 1.99-5.20; p <0.001) and PFS (hazard ratio 1.82, 95% CI 1.18-2.80; p = 0.010). CONCLUSION: Radiomic-based models predict survival outcomes and response to immunotherapy in patients with advanced HCC. Deep learning in combination with machine learning can stratify patients and allows for precision treatment strategies. IMPACT AND IMPLICATIONS: Prognostic markers predicting survival and response to immunotherapy in hepatocellular carcinoma are lacking. This study used deep learning and machine learning to develop and validate an integrated radiomic-clinical model which can predict survival and response to atezolizumab plus bevacizumab from pre-treatment imaging. Radiomic-based machine learning models can risk-stratify patients with advanced HCC receiving atezolizumab plus bevacizumab.

来源:J Hepatol,查看原文。版权归原作者所有。

引用

  1. Vithayathil M, Koku D, Campani C, et al. Machine learning based radiomic models outperform clinical biomarkers in predicting outcomes after immunotherapy for hepatocellular carcinoma. J Hepatol. 2025. PMID: 40246150. doi:10.1016/j.jhep.2025.04.017

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。