Epsylon Laboratory, University Paul Valéry Montpellier 3, Nîmes, France.
Biostatistics and Research Methods Centre, University Hospital and University of Liège, Liège, Belgium.
Cancer Med. 2024 Oct;13(19):e70268. doi: 10.1002/cam4.70268.
Understanding cancer symptom cluster through network analyses is a new approach in oncology, revealing interconnected and influential relationships among reported symptoms. We aimed to assess these relationships using network analysis in posttreatment breast cancer patients, focusing on the five dimensions of cancer-related fatigue (CRF), and on other common difficulties encountered by oncological patients (i.e., pain, anxiety, depression, sleep difficulties, cognitive impairments, and emotion regulation and mental adaptation difficulties).
This study involved a complementary analysis of data from two interventional studies. Participants completed questionnaires before and after the intervention, with baseline scores being used in this article. Partial correlation network analysis modeled the relationships between symptoms in five distinct networks, each of them including one specific dimension of CRF. The core symptom in each network was identified based on the highest centrality indices.
Depression emerged as the core symptom in all networks, strongly associated with all fatigue dimensions (partial correlations ranging from 0.183 to 0.269) except mental fatigue. These findings indicate robust connections between symptoms, as variations in depression scores directly or indirectly influence fatigue and other symptoms within the cluster.
Our results support the multidimensional aspect of CRF, and its links with other common symptoms. To effectively reduce patient CRF, interventions should address not only fatigue but also the closely related symptoms from the cluster, such as depression, given its centrality in the cluster.
ClinicalTrials.gov (NCT03144154 and NCT04873661). Retrospectively registered on May 1, 2017 and April 29, 2021, respectively.
通过网络分析了解癌症症状群是肿瘤学中的一种新方法，揭示了报告的症状之间相互关联和影响的关系。我们旨在使用网络分析评估治疗后乳腺癌患者的这些关系，重点关注癌症相关疲劳（CRF）的五个维度，以及肿瘤患者经常遇到的其他常见困难（即疼痛、焦虑、抑郁、睡眠困难、认知障碍以及情绪调节和心理适应困难）。
本研究对两项干预性研究的数据进行了补充分析。参与者在干预前后完成了问卷，本文使用了基线评分。偏相关网络分析对五个不同网络中的症状之间的关系进行建模，每个网络都包含 CRF 的一个特定维度。根据中心度指数最高的原则，确定每个网络中的核心症状。
抑郁在所有网络中都是核心症状，与所有疲劳维度（偏相关系数范围为 0.183 至 0.269）强烈相关，除了心理疲劳。这些发现表明症状之间存在稳健的联系，因为抑郁评分的变化直接或间接地影响了疲劳和集群中的其他症状。
我们的结果支持 CRF 的多维性及其与其他常见症状的联系。为了有效降低患者的 CRF，干预措施不仅应针对疲劳，还应针对集群中密切相关的症状，如抑郁，因为抑郁在集群中处于中心位置。
ClinicalTrials.gov（NCT03144154 和 NCT04873661）。分别于 2017 年 5 月 1 日和 2021 年 4 月 29 日回顾性注册。
