基于临床与影像数据的头颈癌放疗毒性动态预测
本研究回顾性分析1012例头颈癌放疗患者,开发多分支深度学习模型整合临床数据与每日锥形束CT(CBCT)预测鼻胃管置入、住院和放射性坏死。结果显示仅用临床特征预测准确性最高,早期体重下降是最强预测因子,早期影像特征未提升性能;鼻胃管预测准确性随放疗分次增加而提高。
要点
- 临床特征联合体重下降是头颈癌放疗毒性最可靠的早期预测因子。
- 早期锥形束CT影像特征或放射组学未改善毒性预测性能。
- 鼻胃管预测准确性随放疗分次增加而提高,第25分次达75%。
核心结论
- 临床数据联合体重下降仍是头颈癌放疗毒性最可靠的早期预测因子,影像数据未带来额外获益。
研究设计与样本
- 回顾性分析2017至2022年接受放疗的1012例头颈癌患者,采用5折交叉验证训练多分支3D ResNet50和多层感知机模型。
主要发现
- 队列中男性占78%,中位年龄63岁(范围35-84),主要原发部位为口咽47%、喉19%、口腔16%。
- 鼻胃管置入、住院和放射性坏死的发生率分别为16.6%、4.2%和4.6%。
- 仅用临床特征预测鼻胃管、住院和放射性坏死的最高准确性分别为70%、67.3%和74.2%。
- 早期体重下降是最强预测因子。
- 早期Jf或放射组学特征未改善模型性能。
临床怎么用
- 在头颈癌放疗早期可依据临床特征尤其是体重下降动态评估鼻胃管置入、住院和放射性坏死风险。
- 每日CBCT影像特征在早期毒性预测中未显示额外价值,临床实践中应优先关注临床指标。
局限性
- 本研究为回顾性分析,结论需前瞻性验证。
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 AND PURPOSE: Head and neck cancer (HNC) radiotherapy (RT) is effective but causes significant toxicity. We aimed to develop a dynamic deep learning model to predict three major HNC RT toxicities-nasogastric (NG) tube placement, hospitalization, and radionecrosis-by integrating clinical data and daily cone-beam computed tomography (CBCTs), assessing whether serial imaging or dosimetry features improve early prediction. MATERIALS AND METHODS: We retrospectively analyzed 1,012 HNC patients treated with RT between 2017 and 2022. A multibranch 3D ResNet50 and multilayer perceptron model was trained using 5-fold cross-validation. Inputs included anatomical deformations from daily CBCTs (converted to Jacobian determinant matrices, Jf), radiomics, and clinical variables (demographics, tumor and treatment details, early weight loss). Each toxicity was modeled using weighted binary cross-entropy loss to address class imbalance. Prediction at the 10th RT fraction was compared with and without Jf integration. RESULTS: The cohort was 78% male, median age was 63 years (range 35-84). Primary sites were mainly oropharynx (47%), larynx (19%), and oral cavity (16%). Concurrent chemoradiation was given to 57%, induction chemotherapy to 7%, and postoperative RT to 18% of patients. Incidences of NG tube, hospitalization, and radionecrosis were 16.6%, 4.2%, and 4.6%, respectively. Clinical features alone yielded highest predictive accuracy: 70% for NG tube, 67.3% for hospitalization, and 74.2% for radionecrosis. Early weight loss was the strongest predictor. Early Jf or radiomics did not improve performance. NG tube prediction accuracy improved with later RT fractions (up to 75% at fraction 25). CONCLUSION: Clinical data combined with weight loss remains the most reliable early predictor of toxicity without added benefit from imaging data.
引用
- Bang C, Gautier H, Le WT, et al. Dynamic prediction of Radiotherapy toxicities in Head and neck cancer using clinical and imaging data. Radiother Oncol. 2025. PMID: 41314396. doi:10.1016/j.radonc.2025.111312
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。