Yin Liangyu, Li Na, Lin Xin, Zhang Ling, Fan Yang, Liu Jie, Lu Zongliang, Li Wei, Cui Jiuwei, Guo Zengqing, Yao Qinghua, Zhou Fuxiang, Liu Ming, Chen Zhikang, Yu Huiqing, Li Tao, Li Zengning, Jia Pingping, Song Chunhua, Shi Hanping, Xu Hongxia
Department of Nephrology, Chongqing Key Laboratory of Prevention and Treatment of Kidney Disease, Chongqing Clinical Research Center of Kidney and Urology Diseases, Xinqiao Hospital, Army Medical University (Third Military Medical University), Chongqing, China.
Department of Clinical Nutrition, Daping Hospital, Army Medical University (Third Military Medical University), Chongqing, China.
Am J Clin Nutr. 2025 Mar;121(3):535-547. doi: 10.1016/j.ajcnut.2025.01.006. Epub 2025 Jan 8.
Cachexia is associated with multiple adverse outcomes in cancer. However, clinical decision-making for oncology patients at the cachexia stage presents significant challenges.
This study aims to develop a machine learning (ML) model to identify potentially reversible cancer cachexia (PRCC).
This was a multicenter cohort study. Cachexia was retrospectively diagnosed using Fearon's framework. PRCC was defined as a diagnosis of cancer cachexia at baseline that turned negative 1 mo later. Body weight dynamics accessible upon patient admission were screened and modeled to predict PRCC. Multiple ML models were trained and cross-validated using 70% of the data to predict PRCC, with the remaining 30% reserved for model evaluation. The interpretability and clinical usefulness of the optimal model were assessed, and external validation was performed in an independent cohort of 238 patients.
The study enrolled 1983 men and 1784 women (median age = 58 y). PRCC was identified in 1983 patients (52.6%). Breast cancer exhibited the highest rate of PRCC (72.1%), whereas cachexia associated with various gastrointestinal cancers was less likely to be reversed. Weight change (WC) from 6 mo ago to 1 mo ago, WC from 1 mo ago to baseline (-1 to 0), and baseline body mass index were selected for modeling. A multilayer perceptron model showed good performance to predict PRCC in the holdout test set [area under the curve (95% confidence interval): 0.887 (0.866, 0.907); accuracy: 0.836; sensitivity: 0.859; specificity: 0.812] and the external validation set [area under the curve (95% confidence interval): 0.863 (0.778, 0.948)]. The WC -1 to 0 showed the highest impact on model output. The model was demonstrated to be clinically useful and statistically relevant.
This study presents an explainable ML model for the early identification of PRCC that utilizes simple body weight dynamics. The findings showcase the potential of this approach in improving the management of cancer cachexia to optimize patient outcomes.
恶病质与癌症的多种不良结局相关。然而，处于恶病质阶段的肿瘤患者的临床决策面临重大挑战。
本研究旨在开发一种机器学习（ML）模型，以识别潜在可逆转的癌症恶病质（PRCC）。
这是一项多中心队列研究。使用费伦框架对恶病质进行回顾性诊断。PRCC定义为基线时诊断为癌症恶病质，但1个月后转为阴性。筛选患者入院时可获取的体重动态数据并进行建模，以预测PRCC。使用70%的数据训练多个ML模型并进行交叉验证，以预测PRCC，其余30%的数据留作模型评估。评估最佳模型的可解释性和临床实用性，并在一个由238名患者组成的独立队列中进行外部验证。
该研究纳入了1983名男性和1784名女性（中位年龄 = 58岁）。1983名患者被识别为PRCC（52.6%）。乳腺癌的PRCC发生率最高（72.1%），而与各种胃肠道癌症相关的恶病质逆转的可能性较小。选择从6个月前到1个月前的体重变化（WC）、从1个月前到基线（-1至0）的WC以及基线体重指数进行建模。多层感知器模型在保留测试集[曲线下面积（95%置信区间）：0.887（0.866，0.907）；准确率：0.836；灵敏度：0.859；特异性：0.812]和外部验证集[曲线下面积（95%置信区间）：0.863（0.778，0.948）]中表现出良好的PRCC预测性能。WC -1至0对模型输出的影响最大。该模型被证明具有临床实用性和统计学相关性。
本研究提出了一种可解释的ML模型，用于利用简单的体重动态数据早期识别PRCC。研究结果展示了这种方法在改善癌症恶病质管理以优化患者结局方面的潜力。
