NYU Rory Meyers College of Nursing, New York, NY, USA.
College of Nursing, 7831University of South Florida, Tampa, FL, USA.
Biol Res Nurs. 2021 Jan;23(1):31-41. doi: 10.1177/1099800420942830. Epub 2020 Jul 23.
To examine a) whether there are significant differences in the severity of symptoms of fatigue, sleep disturbance, or depression between patients with rectal cancer who develop co-occurring symptoms and those with no symptoms before and at the end of chemotherapy and radiation therapy (CRT); b) differences in gut microbial diversity between those with co-occurring symptoms and those with no symptoms; and c) whether before-treatment diversity measurements and taxa abundances can predict co-occurrence of symptoms.
Stool samples and symptom ratings were collected from 31 patients with rectal cancer prior to and at the end of (24-28 treatments) CRT. Descriptive statistics were computed and the Mann-Whitney U test was performed for symptoms. Gut microbiome data were analyzed using R's vegan package software.
Participants with co-occurring symptoms reported greater severity of fatigue at the end of CRT than those with no symptoms. and abundances differed between participants with co-occurring symptoms and those with no symptoms. Our random forest classification (unsupervised learning algorithm) predicted participants who developed co-occurring symptoms with 74% accuracy, using specific phylum, family, and genera abundances as predictors.
Our preliminary results point to an association between the gut microbiota and co-occurring symptoms in rectal cancer patients and serves as a first step in potential identification of a microbiota-based classifier.
检验在接受化疗和放疗(CRT)前后,同时出现疲劳、睡眠障碍或抑郁症状的直肠癌患者与无症状患者之间,其症状严重程度是否存在显著差异;
同时出现症状患者与无症状患者之间的肠道微生物多样性差异;
治疗前的多样性测量和分类丰度是否可以预测症状的同时出现。
收集 31 例直肠癌患者在 CRT 前后的粪便样本和症状评分。
计算描述性统计数据,并使用曼-惠特尼 U 检验进行症状分析。
使用 R 的 vegan 包软件分析肠道微生物组数据。
同时出现症状的患者在 CRT 结束时报告的疲劳严重程度高于无症状患者。
同时出现症状患者与无症状患者之间的分类丰度存在差异。
我们的随机森林分类(无监督学习算法)使用特定门、科和属的丰度作为预测因子,以 74%的准确率预测出现同时出现症状的患者。