Discovery of Distinct Cancer Cachexia Phenotypes Using Unsupervised Machine Learning Algorithm

Hao-Fan Wu,Jiang-Peng Yan,Qian Wu,Zhen Yu,Hong-Xia Xu,Chun-Hua Song,Zeng-Qing Guo,Wei Li,Yan-Jun Xiang,Zhe Xu,Jie Luo,Shu-Qun Cheng,Feng-Min Zhang,Han-Ping Shi,Cheng-Le Zhuang,Investigation on Nutrition Status and Clinical Outcome of Common Cancers Group
DOI: https://doi.org/10.1016/j.nut.2023.112317
IF: 4.893
2023-12-01
Nutrition
Abstract:Purpose Cancer cachexia is a debilitating condition with widespread negative impacts. 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 improving outcomes for vulnerable populations. Methods This was a nationwide multicenter observational study conducted from October 2012 to April 2021 in China. Unsupervised consensus clustering analysis was applied based on a combination of demographic, anthropometric, nutritional, oncological, and quality of life data. Key characteristics of each cluster were identified using the standardized mean difference. One-, three-, five-year and overall mortality were evaluated using logistic and Cox regression analysis. Results Consensus clustering algorithm was performed for 4,329 patients with cancer cachexia in the discovery cohort, and four clusters with distinct phenotypes were uncovered. From Cluster 1 to Cluster 4, the clinical characteristics of patients showed a transition from almost unimpaired to mildly, moderately, and severely impaired. Consistently, an increase in mortality from Cluster 1 to 4 was observed. The overall mortality rate was 32.4%, 39.8%, 54.4%, and 68% and the median overall survival time was 21.9, 18, 16.7, and 13.6 months for patients in Cluster 1 to Cluster 4, respectively. Our machine learning-based model performed better in predicting mortality than traditional model. External validation confirmed above results. Conclusions Machine learning is valuable for phenotype classifications of patients with cancer cachexia. Detection of clinically distinct clusters among cachexic patients assist in scheduling personalized treatment strategies and in patients' selection for clinical trials.
nutrition & dietetics
What problem does this paper attempt to address?