生境感知放射组学与自适应2.5D深度学习预测接受新辅助化疗免疫治疗的食管鳞状细胞癌患者的治疗反应和长期生存
这项双中心回顾性研究纳入269例初治食管鳞状细胞癌(ESCC)患者,基于基线PET/CT构建生境放射组学与2.5D深度学习融合模型预测新辅助化疗免疫治疗(NACI)反应及生存。生境放射组学验证AUC为0.865(95%CI 0.778-0.953),优于传统放射组学(ΔAUC +3.6%,P<0.01)和临床模型(ΔAUC +6.4%,P<0.001);联合模型测试AUC为0.824、敏感度0.875、测试C指数0.809。
要点
- 生境放射组学验证AUC 0.865(95%CI 0.778-0.953),优于传统放射组学(ΔAUC +3.6%,P<0.01)和临床模型(ΔAUC +6.4%,P<0.001)。
- 2.5D多实例学习(MIL)模型验证AUC为0.861,显示较强泛化能力。
- 联合模型测试AUC 0.824、敏感度0.875、测试C指数0.809,校准良好(Hosmer-Lemeshow P>0.800)。
核心结论
- 整合生境放射组学与2.5D深度学习可实现ESCC可解释的双重诊断-预后分层,通过解码空间异质性推进精准肿瘤学。
研究设计与样本
- 双中心回顾性研究,分析269例初治ESCC患者的基线PET/CT,训练集n=144、验证集n=62、测试集n=63。
主要发现
- 生境放射组学验证AUC为0.865(95%CI 0.778-0.953),优于传统放射组学(ΔAUC +3.6%,P<0.01)和临床模型(ΔAUC +6.4%,P<0.001)。
- SHAP分析显示浸润前沿(H2)为主要预测区域(占前位特征的40%),wavelet_LHH_firstorder_Entropy影响最大(SHAP=+0.42)。
- 2.5D MIL模型验证AUC为0.861。
- 联合模型测试AUC为0.824、敏感度0.875,校准良好(Hosmer-Lemeshow P>0.800),测试C指数为0.809,决策曲线分析净获益提升23-41%。
关键数值与推荐
- 生境放射组学验证AUC 0.865(95%CI 0.778-0.953),较传统放射组学ΔAUC +3.6%(P<0.01),较临床模型ΔAUC +6.4%(P<0.001)。
- 联合模型测试AUC 0.824、敏感度0.875,Hosmer-Lemeshow P>0.800,测试C指数0.809,决策曲线分析净获益提升23-41%。
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
PURPOSE: Current radiomic approaches inadequately resolve spatial intratumoral heterogeneity (ITH) in esophageal squamous cell carcinoma (ESCC), limiting neoadjuvant chemoimmunotherapy (NACI) response prediction. We propose an interpretable multimodal framework to: (1) quantitatively map intra-/peritumoral heterogeneity via voxel-wise habitat radiomics; (2) model cross-sectional tumor biology using 2.5D deep learning; and (3) establish mechanism-driven biomarkers via SHAP interpretability to identify resistance-linked subregions. METHODS: This dual-center retrospective study analyzed 269 treatment-naïve ESCC patients with baseline PET/CT (training: n = 144; validation: n = 62; test: n = 63). Habitat radiomics delineated tumor subregions via K-means clustering (Calinski-Harabasz-optimized) on PET/CT, extracting 1,834 radiomic features per modality. A multi-stage pipeline (univariate filtering, mRMR, LASSO regression) selected 32 discriminative features. The 2.5D model aggregated ± 4 peri-tumoral slices, fusing PET/CT via MixUp channels using a fine-tuned ResNet50 (ImageNet-pretrained), with multi-instance learning (MIL) translating slice-level features to patient-level predictions. Habitat features, MIL signatures, and clinical variables were integrated via five-classifier ensemble (ExtraTrees/SVM/RandomForest) and Crossformer architecture (SMOTE-balanced). Validation included AUC, sensitivity, specificity, calibration curves, decision curve analysis (DCA), survival metrics (C-index, Kaplan-Meier), and interpretability (SHAP, Grad-CAM). RESULTS: Habitat radiomics achieved superior validation AUC (0.865, 95% CI: 0.778-0.953), outperforming conventional radiomics (ΔAUC + 3.6%, P < 0.01) and clinical models (ΔAUC + 6.4%, P < 0.001). SHAP identified the invasive front (H2) as dominant predictor (40% of top features), with wavelet_LHH_firstorder_Entropy showing highest impact (SHAP = + 0.42). The 2.5D MIL model demonstrated strong generalizability (validation AUC: 0.861). The combined model achieved state-of-the-art test performance (AUC = 0.824, sensitivity = 0.875) with superior calibration (Hosmer-Lemeshow P > 0.800), effective survival stratification (test C-index: 0.809), and 23-41% net benefit improvement in DCA. CONCLUSION: Integrating habitat radiomics and 2.5D deep learning enables interpretable dual diagnostic-prognostic stratification in ESCC, advancing precision oncology by decoding spatial heterogeneity.
引用
- Gao X, Yang L, She T, et al. Habitat-aware radiomics and adaptive 2.5D deep learning predict treatment response and long-term survival in ESCC patients undergoing neoadjuvant chemoimmunotherapy. Eur J Nucl Med Mol Imaging. 2025. PMID: 40960691. doi:10.1007/s00259-025-07522-6
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。