Discovery of distinct cancer cachexia phenotypes using an unsupervised machine-learning algorithm

• Published in: Nutrition (Burbank, Los Angeles County, Calif.) Vol. 119; p. 112317
• Main Authors: Wu, Hao-Fan, Yan, Jiang-Peng, Wu, Qian, Yu, Zhen, Xu, Hong-Xia, Song, Chun-Hua, Guo, Zeng-Qing, Li, Wei, Xiang, Yan-Jun, Xu, Zhe, Luo, Jie, Cheng, Shu-Qun, Zhang, Feng-Min, Shi, Han-Ping, Zhuang, Cheng-Le
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.03.2024; Elsevier Limited
• Subjects: Algorithms; artificial intelligence; At risk populations; Body mass index; Cachexia; Cachexia - etiology; Cancer; Cancer cachexia; China; Classification; Clinical trials; Cluster analysis; Clustering; Datasets; Hemoglobin; Heterogeneity; Humans; Learning algorithms; Machine Learning; Medical prognosis; Mortality; Neoplasms - complications; Nutrition; Observational studies; Patients; Performance prediction; Phenotype; Phenotypes; Quality of life; Regression analysis; China; Cancer cachexia; Phenotype; Machine learning; Classification
• DOI: 10.1016/j.nut.2023.112317
• ISSN: 0899-9007; 1873-1244

•This is the first establishment of a machine learning-based model for stratifying cancer cachexia.•Our model performed better than the previous method in predicting overall survival.•Our model reflects dynamic changes in body condition during cachexia deterioration. Cancer cachexia is a debilitating condition with widespread negative effects. The heterogeneity of clinical features within patients with cancer cachexia is unclear. The identification and prognostic analysis of diverse phenotypes of cancer cachexia may help develop individualized interventions to improve outcomes for vulnerable populations. The aim of this study was to show that the machine learning–based cancer cachexia classification model generalized well on the external validation cohort. This was a nationwide multicenter observational study conducted from October 2012 to April 2021 in China. Unsupervised consensus clustering analysis was applied based on demographic, anthropometric, nutritional, oncological, and quality-of-life data. Key characteristics of each cluster were identified using the standardized mean difference. We used logistic and Cox regression analysis to evaluate 1-, 3-, 5-y, and overall mortality. A consensus clustering algorithm was performed for 4329 patients with cancer cachexia in the discovery cohort, and four clusters with distinct phenotypes were uncovered. From clusters 1 to 4, the clinical characteristics of patients showed a transition from almost unimpaired to mildly, moderately, and severely impaired. Consistently, an increase in mortality from clusters 1 to 4 was observed. The overall mortality rate was 32%, 40%, 54%, and 68%, and the median overall survival time was 21.9, 18, 16.7, and 13.6 mo for patients in clusters 1 to 4, respectively. Our machine learning-based model performed better in predicting mortality than the traditional model. External validation confirmed the above results. Machine learning is valuable for phenotype classifications of patients with cancer cachexia. Detection of clinically distinct clusters among cachexic patients assists in scheduling personalized treatment strategies and in patient selection for clinical trials.