Symptom networks of multidimensional symptom experiences in breast cancer survivors: A network analysis

doi:10.21203/rs.3.rs-4939330/v1

Download PDF

Article

Symptom networks of multidimensional symptom experiences in breast cancer survivors: A network analysis

https://doi.org/10.21203/rs.3.rs-4939330/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Objectives

We aimed to construct a symptom network for breast cancer patients, identify its core symptoms, and explore symptom clusters. This network approach may provide valuable insights for precise interventions to improve the overall quality of life in breast cancer patients.

Methods

A total of 462 eligible breast cancer patients were recruited. The severity of patients' symptoms was measured using the EORTC QLQ-C30 Chinese version scale and Zung Self-Rating Depression and Anxiety Scale. A regularized partial correlation network was established, and central symptoms were identified using Strength centrality.

Results

The strongest associations were observed between NV-AP (weight = 0.39), Dep-Anx (weight = 0.38), PA-DY (weight = 0.21), and Anx-SL (weight = 0.20). Fatigue was the most prevalent symptom among breast cancer patients, and fatigue was consistently the central symptom in the network, in addition to anxiety, appitie loss, and pain. DAG indicated that fatigue might influence overall symptoms in breast cancer patients. Three syomtom clusters were indentified: emotional symptoms (depression, anxiety, and insomnia), gastrointestinal symptoms (nausea/vomiting, diarrhea, and loss of appetite), and somatic symptoms (fatigue, pain, and dyspnea).

Conclusions

Fatigue, depression, and anxiety are highly prevalent and central symptoms in breast cancer patients. It is crucial to screen and provide early treatment for these symptoms to effectively manage them and enhance the overall quality of life for breast cancer patients. Future studies should focus on conducting longitudinal research to establish dynamic networks and investigate causal relationships between these symptoms.

Health sciences/Oncology/Cancer/Breast cancer

Biological sciences/Psychology

breast cancer

symptom cluster

fatigue

depression

anxiety

network analysis

Breast cancer stands as the most frequently diagnosed cancer in women globally, comprising 11.7% of all newly reported cases [1]. Despite notable advancements in treatment and care leading to enhanced survival rates, breast cancer survivors frequently endure a spectrum of concurrent adverse physiological and psychological manifestations [2]. Extant studies investigating symptoms among cancer patients reveal a pattern wherein these manifestations seldom manifest in isolation, often sharing common or interconnected etiologies [3]. This observation underscores the potential for a cascading effect, where the presence of one symptom may exacerbate the occurrence and severity of other related symptoms, thereby engendering a deleterious cycle of symptom clusters detrimental to the patient's functional capacity and overall well-being [4].

Prior investigations have elucidated the prevalence of symptoms encountered by breast cancer patients, encompassing fatigue, pain, insomnia, depression, and anxiety [5]. Diverse determinants, including age, disease advancement, and therapeutic modalities, exert influence over the manifestation and aggregation of these symptoms [6]. Despite extensive inquiry into the origins and clustering of symptoms, a paucity of research delves into the nuanced interplay among them [7]. Appreciating the interconnectedness of symptoms bears significance in efficaciously mitigating symptomatology in cancer patients and forestalling the emergence of associated manifestations. A pioneering avenue for probing these intricate relationships lies within network analysis, offering a paradigm to elucidate the intricate connections among these symptoms.

Network analysis is a valuable tool for understanding the internal characteristics of a system by representing it as a network. It allows us to identify important nodes and structural characteristics within the symptom network of cancer patients, shedding light on the complex connections between symptoms and the underlying mechanisms of symptom occurrence. This knowledge can lead to significant improvements in symptom management for cancer patients. Network analysis has gained widespread recognition in recent years for its ability to visually display these complex connections and determine the importance of each symptom. For instance, Rooij et al. demonstrated that fatigue is a common core symptom among survivors of various cancer types. Similarly, Jing et al. found that emotional fluctuations and irritability are strongly linked and serve as core symptoms in breast cancer patients undergoing endocrine therapy By utilizing network analysis to explore these internal connections between symptoms, we can gain a deeper understanding of symptom development and identify precise intervention targets to optimize symptom management for cancer patients.

Network analysis serves as a pivotal instrument in unraveling the intricate inter-relationship of a system by depicting it as a network [8]. Through this approach, we discern pivotal nodes and structural attributes within the symptom network of cancer patients, thereby elucidating the convoluted interrelations among symptoms and the underlying mechanisms governing their onset [9]. Such insights hold promise for refining symptom management strategies for cancer patients. In recent years, network analysis has garnered acclaim for its capacity to visually elucidate these intricate associations and ascertain the significance of each symptom. Rooij et al. underscore fatigue as a prevalent core symptom across various cancer survivor cohorts [10]. Similarly, Jing et al. found that emotional fluctuations and irritability are strongly linked and serve as core symptoms in breast cancer patients undergoing endocrine therapy [11]. By utilizing network analysis to scrutinize these internal symptom interconnections, we stand poised to deepen our comprehension of symptom etiology and pinpoint precise intervention targets, thereby optimizing symptom management strategies for cancer patients.

The main aim of this study is twofold. Firstly, it aims to establish a symptom network for breast cancer patients, exploring core symptoms and identifying symptom clusters. Secondly, it aims to assess the differences in symptom networks based on various demographic and clinical variables. The objective of this research is to provide a rapid, efficient, scientific, and sustainable basis for managing symptoms in breast cancer patients.

All methods performed in this study were in accordance with the relevant guidelines and regulations.

2.1 Study settings and participants

The study included breast cancer patients who were hospitalized for treatment at the Tumor Hospital of Xinjiang Medical University between June 2016 and September 2017. All paticipants provided written informed consent. The inclusion criteria were: (1) Diagnosis of breast cancer through imaging and pathological examinations; (2) Age between 18 and 80 years. The exclusion criteria were: (1) Patients with severe mental disorders unable to complete the self-assessment scale; (2) Incomplete clinical data or missing responses on the EORTC QLQ-C30 and Zung Depression Anxiety Self-Assessment Scale; (3) Absence of written consent.

2.2 Data collection

This study involved the extraction of basic and clinical data from the medical records of patients. The data collected included age, time since first diagnosis, place of residence, tumor-node-metastasis (TNM) staging, treatment received since diagnosis (surgery, chemotherapy, radiotherapy) since diagnosis, and comorbidity (type-2 diabetes mellitus, chronic heart disease, and hypertension). Additionally, the patients' health-related quality of life (HRQOL) and depression status were evaluated using the Zung Self-Rating Depression Scale (SDS) and the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30).

2.3 Measures

2.3.1 The Zung Self-Rating Depression and Anxiety Scale

The Zung Self-Rating Depression Scale (SDS) and Self-Rating Anxiety Scale (SAS) were utilized to evaluate the depression and anxiety levels of the patients. Each scale comprises 20 items, with scores ranging from 1 to 4 for each item. The sum of the scores of all 20 items represents the raw score, which is then multiplied by 1.25 and rounded off to obtain the standard score. A standard score of ≥ 53 on the SDS indicates the presence of depression, while a standard score of ≥ 50 on the SAS indicates the presence of anxiety [12] [13].

2.3.2 The European Organization for Research and Treatment of Cancer Quality of Life Questionnaire

The European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) was utilized to evaluate patients' symptoms. This scale comprises 15 dimensions, which consist of five functional dimensions, one dimension for overall quality of life and health status, one single-item entry for financial difficulties, and eight dimensions for symptoms. In order to assess patients' symptoms, we specifically included the EORTC QLQ-C30 symptom scale, which covers fatigue, pain, dyspnea, constipation, loss of appetite, diarrhea, and nausea/vomiting [14].

2.4 Statistical analysis

2.4.1 Network estimation

The network analysis was conducted using R 4.2.1 and the R package qgraph [15]. We employed the Gaussian graphical model (GGM) with regularized partial correlations to construct a symptom network consisting of 10 symptoms: fatigue, pain, dyspnea, constipation, loss of appetite, diarrhea, nausea/vomiting, depression, and anxiety [16]. To ensure a concise and easily interpretable model, we utilized the least absolute shrinkage and selection operator (LASSO) [17]. The Fruchterman-Reingold layout was implemented, positioning symptoms with stronger correlations in close proximity to each other [18]. Additionally, factors such as age, disease course, TNM staging, surgery, chemotherapy, and radiotherapy were included as covariates in the overall symptom network. Strength Centrality was employed to identify the central symptoms in the network. This metric quantifies the total connection weight between a specific symptom and other symptoms, thereby revealing its pivotal role and influence within the entire symptom network [19].

2.4.2 Network stability

The accuracy and stability of our estimated network were assessed using the R package bootnet. To evaluate the accuracy and stability of centrality measures, we performed bootstrapping with 1000 iterations. First, we measured the accuracy of the edges by bootstrapping the edge weights and calculating 95% confidence intervals (CI). Second, we determined the centrality stability of the coefficients (CS coefficients) using subset bootstrapping. It is generally recommended that the CS coefficient should not be less than 0.25, and ideally, it should be greater than 0.50 [20].

2.4.3 Network comparison

To formally test for differences between groups in the network, we conducted a network comparison test. Our study aimed to investigate potential differences in network characteristics among different age groups, TNM stages, months of diagnosis, and whether radiotherapy was received. The Network Comparison Test (NCT) was employed, utilizing the Holm-Bonferroni correction of p-values from multiple tests to assess disparities in network structure, global strength, and each edge between two networks [21]. The statistical analysis was performed using the R package NetworkComparisonTest.

2.4.4 Directed acyclic graph (DAG)

The DAG is a method used to encode the relationships between nodes in cross-sectional data and establish causal relationships among them. The R package bnlearn and the Bayesian hill-climbing algorithm were used [22] [23]. The algorithm evaluates the network model's structure by manipulating the edges, such as adding, deleting, or changing their direction, with the goal of optimizing the goodness-of-fit score, which is measured by the Bayesian information criterion. To ensure the stability of the generated DAG, a bootstrap procedure was conducted with 10,000 samples. In this procedure, samples were drawn with replacement and the direction of each edge was determined based on their occurrence in the bootstrapped DAGs. If the direction of a directed edge appeared in more than 51% of the bootstrapped DAGs, it would be included in the final DAG [24] .

2.5 Ethics statement

This study was a retrospective study, approved by Xinjiang Medical University ethnic committee (Number: XJYKDXR20230208001). All sensitive and private information of patients were kept confidential. All paticipants provided written informed consent.

3.1 Characteristics of participants

A total of 462 breast cancer patients, with ages ranging from 24 to 79 years (mean age 49.76 ± 9.42), were included in the analysis. The majority of patients (316, 68.4%) resided in urban areas, and most patients (266, 57.58%) had been diagnosed with the disease for less than a year. The TNM stage of the cancer was classified as stage I-II for the majority of patients (296, 64.07%). Most patients received surgery (451, 97.62%), and chemotherapy (415, 89.83%). However, almost half of the patients (236, 51.08%) didn’t receive radiotherapy. More detailed information on the sociodemographic and clinical characteristics of the patients can be found in Table 1

Table 1

Basic characteristics of cancer patients
	Cases (n)	Ratio (%)
Age
< 55	340	73.59%
≥ 55	122	26.41%
Place of residence
Urban	316	68.40%
Rural	146	31.60%
Marriage Status
Married	410	88.70%
Unmarried	52	11.20%
Education Degree
High school education or below	168	36.40%
University education or above	294	63.60%
Wether to exercise
Yes	284	61.40%
No	178	38.50%
Comorbidities(Type 2 diabetes mellitus, and/or chronic heart disease, and/or hypertension.)
Yes	56	12.10%
No	406	87.80%
Time since fist diagnosis
< 1year	266	57.58%
≥ 1year	196	42.42%
TNM staging
I-II	296	64.07%
III-IV	166	35.93%
Surgery
Yes	451	97.62%
No	11	2.38%
Chemotherapy
Yes	415	89.83%
No	47	10.17%
Radiotherapy
Yes	226	48.92%
No	236	51.08%

3.2 Prevalence and Symptom Scores of Breast Cancer Patients

According to Table 2, the most prevalent symptoms reported by breast cancer patients were fatigue (n = 328, 71%), insomnia (n = 307, 66.45%), depression (n = 197, 42.64%), and anxiety (n = 192, 41.56%). The symptoms with the highest scores were depression (51.29), followed by anxiety (48.31), insomnia (33.69), and fatigue (21.57).

Table 2

Symptom prevalence and mean scores in 462 breast cancer patients.
Symptoms	Prevalence [n (%)]	Mean scores
FA	328 (71.00)	21.57
NV	77 (16.67)	4.08
PA	189 (40.91)	9.63
DY	141 (30.52)	11.54
SL	307 (66.45)	33.69
AP	97 (21.00)	8.73
CO	135 (29.22)	12.34
DI	44 (9.52)	3.39
Dep	197 (42.64)	51.29
Anx	192 (41.56)	48.31

3.3 Network Structure

Figure 1A presents the symptom network of breast cancer patients, composed of eight symptom dimensions from the EORTC QLQ-C30 scale and standard scores from the Zung self rating Depression and Anxiety scales. Among the 45 possible edges in the network, 22 were non-zero, with an average edge weight of 0.07. The strongest associations were observed between NV-AP (weight = 0.39), Dep-Anx (weight = 0.38), PA-DY (weight = 0.21), and Anx-SL (weight = 0.20). The edge with the highest number of connections was FA, showing moderate to strong associations with DY (weight = 0.25), PA (weight = 0.25), AP (weight = 0.21), and Anx (weight = 0.21) among all symptoms. In Fig. 1B, after adjusting for covariates, the connections between symptoms remained largely unchanged, although the edge weights decreased. Notably, the connection between nausea/vomiting and diarrhea was no longer present.

3.4 Node Centrality

As shown in Fig. 2, FA (Str = 1.48, Bet = 2.17, Clo = 1.82) was identified as the most central symptom in the network, while Anx (Str = 1.26, Bet = 0.79, Clo = 1.21), AP (Str = 0.54, Bet = 0.35, Clo = 0.39), PA (Str = 0.51, Bet = 0.70, Clo = 0.11), Dep (Str = 0.00, Bet=-0.24, Clo=-0.72) and DY (Str=-0.02, Bet = 0.47, Clo=-0.30), showed medium node strength. After adjusting for covariates, the node centrality remained largely unchanged

3.4 Network stability

As depicted in Fig. 3A, the network was accurately estimated using the edge weight bootstrap method, as evidenced by a substantial overlap between the 95% confidence intervals (CIs) of the edge weights. The case dropping bootstrap procedure further demonstrated the stability of Strength centrality, even when different proportions of the sample were dropped (Fig. 3B). Additionally, there was a strong correlation coefficient (CS-C) of 0.75 between the edge weight and Strength centrality. The bootstrapped 95% CIs for the estimated edge weights indicated that the majority of edges were both stable and accurate.

3.5 Network comparison test

The network comparison tests (Fig. 4) revealed that FA was the most central symptom in almost all subgroups, except for those with time since first diagnosis < 1year. Among patients age ≥ 55, there were stronger connection between depression and fatigue (edge_diif=0.15, P = 0.027), as well as between Dep and AP (edge_diif=0.10, P = 0.039), compared to patients age < 55. Patients with time since first diagnosis < 1year had a stronger connection between Anx and CO (edge_diif diif = 0.16, P = 0.038) than those time since first diagnosis ≥ 1year. Additionally, they had a weaker connection between NV and SL (edge_diif =0.06, P = 0.017) compared to patients with time since first diagnosis ≥ 1year. Patients with TNM stages I-II had a stronger connection between AP and NV (edge_diif =0.31, P = 0.009) than patients with TNM stages III-IV. Conversely, they had weaker connection between AP and FA (edge_diif =0.24, P = 0.002), as well as between NV and DY (edge_diif =0.19, P = 0.002), compared to patients with TNM stages III-IV. Furthermore, patients who received radiotherapy had a stronger connection between PA and Dep (edge_diif =0.14, P = 0.015) compared to those who didn’t receive radiotherapy.

3.6 Directed acyclic graph (DAG)

In Fig. 5A, the importance of each edge to the entire DAG structure is displayed. The most important edges for the network structure were Anx-Dep (BIC: -73.32), FA-Anx (BIC: -52.74), and AP-NV (BIC: -46.97). Meanwhile, the least important edges for the network structure were PA-Dep (BIC: -1.19) and FA-SL (BIC: 0.45). In Fig. 5B, the thickness of an edge represents the proportion of bootstrapped networks in which it points from one node to another. Structurally, FA was positioned at the top of the DAG, directly activating six symptoms: PA (BIC: -33.27; Direction: 0.74), DY (BIC: -34.76; Direction: 0.69), AP (BIC: -38.6; Direction: 0.63), SL (BIC: 0.45; Direction: 0.62), NV (BIC: --3.88; Direction: 0.59), Anx (BIC: -52.74; Direction: 0.55). The directional probability between FA and Anx is close to 0.5, suggesting that the relationship between FA and Anx may be bidirectional. (Supplementary Table S2).

To the best of our knowledge, this is the first study to incorporate both psychological and physiological symptoms into the symptom network of breast cancer patients. Our study results showed that depression, anxiety and eight symptoms of the EORTC-QLQ-C30 questionnaire were highly intercorrelated and could be represented as a symptom network. In this study, we found that FA, Anx, AP, PA, Dep and DY were central symptoms both in total and after adjusting for covariates network and they remained central in patients with all subgroups. We also find three major symptom clusters in breast cancer patients: emotional symptoms (Dep, Anx, and SL), gastrointestinal symptoms (NV, DI, and AP), and somatic symptoms (FA, PA, and DY). Therefore, FA, Anx, AP, PA, Dep and DY may play a crucial role in symptom network in breast cancer patients, which might be important targets for clinical intervention to improve overall symptom burden.

Our findings indicate that fatigue is the most prevalent symptom, affecting 71% of patients, and plays a central role in the overall network, except for those with a disease duration of less than one year. Furthermore, the results of DAG show that fatigue is at the upstream of the DAG and activate other symptoms in the network to varying degrees. Previous studies consistently show that fatigue is the most common and distressing symptom experienced by breast cancer patients [25, 26]. Moreover, fatigue has a more severe negative impact on quality of life compared to other symptoms [27]. Consistent with these findings, Rooij et al. [10] identified fatigue as the most central symptom in the symptom network of all cancer types, including breast cancer, and highlighted its strong associations with other symptoms. Similarly, Berger et al. [28] found that fatigue is the central symptom in breast cancer patients one month after completing chemotherapy. Fatigue exhibits a multidimensional nature within the context of cancer-related symptomatology. Numerous studies have established links between fatigue and mental well-being [29], often proposing psychological interventions as potential remedies. Rha and Lee's findings suggest that fatigue's central position in symptom networks may be attributed to chemotherapy usage and the duration of cancer survivorship [30]. In contrast, Zhu et al. observed a diminished centrality of fatigue in populations with over five years of survivorship [9]. This underscores that cancer therapies, encompassing chemotherapy, radiation therapy, immunotherapy, and surgical interventions, could significantly influence fatigue's prominence within symptom networks. It is important to note that fatigue in breast cancer patients not only leads to physical sleepiness but also contributes to cognitive impairments, such as attention and memory deficits, as well as emotional disturbances, such as depression and anxiety [31, 32]. Therefore, addressing fatigue should be prioritized as a key intervention target to reduce the overall symptom burden and improve the quality of life of breast cancer patients. Long-term symptom management strategies should be implemented to ensure comprehensive care.

In this study, depression and anxiety were identified as central symptoms in breast cancer patients. Notably, anxiety emerged as the central symptom within the subset of patients diagnosed within one year. These findings underscore the imperative of early detection and intervention to effectively address emotional disorders in breast cancer patients. Moreover, the prevalence of depression and anxiety within this demographic is conspicuously elevated, often ranging from 13–54%. [33]. These affective disorders exhibit a strong correlation with clinical manifestations of pain and fatigue, exerting a profound detrimental impact on the overall well-being of patients [34, 35], and concomitantly elevating the susceptibility to suicidal ideation [36]. Jing et al. [11] and Rooij et al. [10] similarly observed the centrality of emotional symptoms among breast cancer patients. Nonetheless, there is a prevalent tendency among healthcare professionals to inadequately acknowledge the gravity of these emotional disorders, resulting in the frequent underestimation of depression and anxiety concerns in individuals diagnosed with breast cancer [37]. To facilitate comprehensive care and bolster support for breast cancer patients throughout their survivorship journey, heightened awareness regarding the significance of emotional disorders is imperative among healthcare providers, patients, and their families. Regular psychological evaluations are advocated for the timely identification and assessment of symptoms, thus enabling early intervention and management.

This study identified three primary symptom clusters prevalent among breast cancer patients: emotional manifestations encompassing depression, anxiety, and insomnia; gastrointestinal disturbances including nausea/vomiting, diarrhea, and appetite loss; and somatic complaints comprising fatigue, pain, and dyspnea. Prior investigations frequently underscored the association between fatigue and emotional disorders, particularly insomnia [10] [28]. Research posits that cancer and its therapeutic interventions may trigger peripheral pro-inflammatory cytokine networks, eliciting symptoms such as fatigue, pain, insomnia, anxiety, and depression via cytokine-mediated signaling pathways within the central nervous system [38, 39]. A systematic review elucidated a strong correlation between nausea/vomiting and loss of appetite among breast cancer patients, a finding corroborated by our study as well [40]. Nevertheless, constipation did not coalesce into a symptom cluster with gastrointestinal manifestations in our investigation, potentially attributable to the predominance of chemotherapy recipients among our study cohort (89.83%). Catherine et al. [41] noted chemotherapy-induced autonomic dysfunction as a contributor to delayed gastric motility, thereby precipitating constipation in affected patients. Given the likelihood of distinct symptom profiles across varying cancer types and the potential influence of diverse symptom assessment instruments on cluster identification [40], forthcoming research endeavors ought to incorporate cancer-specific assessment tools to enhance the precision of symptom quantification.

This study has several limitations that should be considered. Firstly, the cross-sectional design of the study restricts our ability to establish clear causal relationships between symptoms. Secondly, the use of the EORTC QLQ-C30 scale for symptom measurement may not be as precise as other professional symptom measurement tools. Thirdly, larger sample sizes in network analysis tend to yield more stable networks. While our study had a sufficient sample size and a stable network, it is important to validate the results using different algorithms and larger sample data. Therefore, future research should consider employing more specialized measurement tools and conducting longitudinal studies on larger, specific cancer populations. This will allow for the construction of dynamic networks and a deeper exploration of the causal relationships between symptoms, ultimately providing a stronger foundation for the precise management of cancer symptoms.

In conclusion, our study found that fatigue is the most common and significant symptom among breast cancer patients. This was closely followed by emotional symptoms such as depression and anxiety, which consistently played a central role in the symptom networks. We also identified three major clusters of symptoms: emotional symptoms (depression, anxiety, and insomnia), gastrointestinal symptoms (nausea/vomiting, diarrhea, and loss of appetite), and somatic symptoms (fatigue, pain, and dyspnea). These findings suggest that addressing fatigue, depression, and anxiety could be crucial in reducing the overall symptom burden experienced by breast cancer patients. As these symptoms are interconnected, interventions targeting these central symptoms may have a ripple effect, leading to a reduction in other related symptoms. However, it is important to validate these findings with larger sample sizes and alternative algorithms.

Authors’ disclosures

No.

Conflict of interest

The authors made no disclosures.

Funding

The study was supported by the State Key Laboratory of Pathogenesis, Prevention and Treatment of High Incidence Diseases in Central Asia-Epidemiology (NO. SKL-HIDCA-2020-ER6) and the State Key Laboratory of Pathogenesis, Prevention and Treatment of High Incidence Diseases in Central Asia (NO. SKL-HIDCA-2023-8).

Author Contribution

SM: Data curation, Formal analysis, Investigation, Project administration, Software, Supervision, Validation, Writing – original draft, Writing – review & editing. PM: Conceptualization, Data curation, Funding acquisition, Investigation, Methodology, Software, Supervision, Writing – original draft, Writing – review & editing. KA: Funding acquisition, Project administration, Resources, Visualization, Writing – review & editing. WJ: Conceptualization, Investigation, Software, Supervision, Writing – original draft, Writing – review & editing.

Acknowledgement

We want to express our gratitude to all the patients, clinicians, and supporters who participated in this study. The views expressed in this article are solely those of the authors and do not necessarily represent the views of their respective institutions or collaborating hospitals.

Data Availability

Data cannot be shared publicly, because data from this study may contain potentially or sensitive patient information. However, data from this study will be made available for researchers who meet criteria for access to confidential data. Requests may be sent to: [email protected]

Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021 May;71(3):209-249. eng. Epub 2021/02/05. doi:10.3322/caac.21660.
Ping Q, Yang CC, Marshall SA, Avis NE, Ip EH. Breast Cancer Symptom Clusters Derived from Social Media and Research Study Data Using Improved K-Medoid Clustering. IEEE Trans Comput Soc Syst. 2016 Jun;3(2):63-74. eng. Epub 2016/06/01. doi:10.1109/tcss.2016.2615850.
Reich RR, Lengacher CA, Alinat CB, Kip KE, Paterson C, Ramesar S, Han HS, Ismail-Khan R, Johnson-Mallard V, Moscoso M, Budhrani-Shani P, Shivers S, Cox CE, Goodman M, Park J. Mindfulness-Based Stress Reduction in Post-treatment Breast Cancer Patients: Immediate and Sustained Effects Across Multiple Symptom Clusters. J Pain Symptom Manage. 2017 Jan;53(1):85-95. eng. Epub 2016/10/11. doi:10.1016/j.jpainsymman.2016.08.005.
Durán-Gómez N, López-Jurado CF, Nadal-Delgado M, Montanero-Fernández J, Palomo-López P, Cáceres MC. Prevalence of Psychoneurological Symptoms and Symptom Clusters in Women with Breast Cancer Undergoing Treatment: Influence on Quality of Life. Semin Oncol Nurs. 2023 Aug;39(4):151451. eng. Epub 2023/05/23. doi:10.1016/j.soncn.2023.151451.
Marshall SA, Yang CC, Ping Q, Zhao M, Avis NE, Ip EH. Symptom clusters in women with breast cancer: an analysis of data from social media and a research study. Qual Life Res. 2016 Mar;25(3):547-57. eng. Epub 2015/10/20. doi:10.1007/s11136-015-1156-7.
So WKW, Law BMH, Ng MSN, He X, Chan DNS, Chan CWH, McCarthy AL. Symptom clusters experienced by breast cancer patients at various treatment stages: A systematic review. Cancer Med. 2021 Apr;10(8):2531-2565. eng. Epub 2021/03/23. doi:10.1002/cam4.3794.
Al Qadire M, Alsaraireh M, Alomari K, Aldiabat KM, Al-Sabei S, Al-Rawajfah O, Aljezawi M. Symptom Clusters Predictive of Quality of Life Among Jordanian Women with Breast Cancer. Semin Oncol Nurs. 2021 Apr;37(2):151144. eng. Epub 2021/03/28. doi:10.1016/j.soncn.2021.151144.
Hevey D. Network analysis: a brief overview and tutorial. Health Psychol Behav Med. 2018 Sep 25;6(1):301-328. eng. Epub 2018/09/25. doi:10.1080/21642850.2018.1521283.
Zhu Z, Sun Y, Kuang Y, Yuan X, Gu H, Zhu J, Xing W. Contemporaneous symptom networks of multidimensional symptom experiences in cancer survivors: A network analysis. Cancer Med. 2023 Jan;12(1):663-673. eng. All authors declare no disclosures. Epub 20220601. doi:10.1002/cam4.4904.
de Rooij BH, Oerlemans S, van Deun K, Mols F, de Ligt KM, Husson O, Ezendam NPM, Hoedjes M, van de Poll-Franse LV, Schoormans D. Symptom clusters in 1330 survivors of 7 cancer types from the PROFILES registry: A network analysis. Cancer. 2021 Dec 15;127(24):4665-4674. eng. Epub 2021/08/14. doi:10.1002/cncr.33852.
Jing F, Zhu Z, Qiu J, Tang L, Xu L, Xing W, Hu Y. Contemporaneous symptom networks and correlates during endocrine therapy among breast cancer patients: A network analysis. Front Oncol. 2023;13:1081786. eng. Epub 2023/04/18. doi:10.3389/fonc.2023.1081786.
Zung WW. A SELF-RATING DEPRESSION SCALE. Arch Gen Psychiatry. 1965 Jan;12:63-70. eng. Epub 1965/01/01. doi:10.1001/archpsyc.1965.01720310065008.
Zung WW. A rating instrument for anxiety disorders. Psychosomatics. 1971 Nov-Dec;12(6):371-9. eng. Epub 1971/11/01. doi:10.1016/s0033-3182(71)71479-0.
Aaronson NK, Ahmedzai S, Bergman B, Bullinger M, Cull A, Duez NJ, Filiberti A, Flechtner H, Fleishman SB, de Haes JC, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst. 1993 Mar 3;85(5):365-76. eng. Epub 1993/03/03. doi:10.1093/jnci/85.5.365.
Epskamp S, Cramer AOJ, Waldorp LJ, Schmittmann VD, Borsboom D. qgraph: Network Visualizations of Relationships in Psychometric Data. Journal of Statistical Software. 2012 05/24;48(4):1 - 18. doi:10.18637/jss.v048.i04.
Borsboom D, Cramer AO. Network analysis: an integrative approach to the structure of psychopathology. Annu Rev Clin Psychol. 2013;9:91-121. eng. Epub 2013/03/30. doi:10.1146/annurev-clinpsy-050212-185608.
van Borkulo CD, Borsboom D, Epskamp S, Blanken TF, Boschloo L, Schoevers RA, Waldorp LJ. A new method for constructing networks from binary data. Scientific Reports. 2014 2014/08/01;4(1):5918. doi:10.1038/srep05918.
Fruchterman TMJ, Reingold EM. Graph drawing by force-directed placement. Software: Practice and Experience. 1991;21(11):1129-1164. doi:https://doi.org/10.1002/spe.4380211102.
Armour C, Fried EI, Deserno MK, Tsai J, Pietrzak RH. A network analysis of DSM-5 posttraumatic stress disorder symptoms and correlates in U.S. military veterans. J Anxiety Disord. 2017 Jan;45:49-59. eng. Epub 2016/12/10. doi:10.1016/j.janxdis.2016.11.008.
Epskamp S, Borsboom D, Fried EI. Estimating psychological networks and their accuracy: A tutorial paper. Behav Res Methods. 2018 Feb;50(1):195-212. eng. Epub 2017/03/28. doi:10.3758/s13428-017-0862-1.
van Borkulo CD, van Bork R, Boschloo L, Kossakowski JJ, Tio P, Schoevers RA, Borsboom D, Waldorp LJ. Comparing network structures on three aspects: A permutation test. Psychol Methods. 2022 Apr 11. eng. Epub 2022/04/12. doi:10.1037/met0000476.
Choi T. Bayesian networks with examples in R [https://doi.org/10.1111/biom.12369]. Biometrics. 2015 2015/09/01;71(3):864-865. doi:https://doi.org/10.1111/biom.12369.
Scutari M. Learning Bayesian Networks with the bnlearn R Package. Journal of Statistical Software. 2010 07/16;35(3):1 - 22. doi:10.18637/jss.v035.i03.
Briganti G, Scutari M, McNally RJ. A tutorial on bayesian networks for psychopathology researchers. Psychol Methods. 2022 Feb 3. eng. Epub 2022/02/04. doi:10.1037/met0000479.
Ruiz-Casado A, Álvarez-Bustos A, de Pedro CG, Méndez-Otero M, Romero-Elías M. Cancer-related Fatigue in Breast Cancer Survivors: A Review. Clin Breast Cancer. 2021 Feb;21(1):10-25. eng. Epub 2020/08/21. doi:10.1016/j.clbc.2020.07.011.
Álvarez-Bustos A, de Pedro CG, Romero-Elías M, Ramos J, Osorio P, Cantos B, Maximiano C, Méndez M, Fiuza-Luces C, Méndez-Otero M, Martín S, Cebolla H, Ruiz-Casado A. Prevalence and correlates of cancer-related fatigue in breast cancer survivors. Support Care Cancer. 2021 Nov;29(11):6523-6534. eng. Epub 2021/04/29. doi:10.1007/s00520-021-06218-5.
Cheng KK, Lee DT. Effects of pain, fatigue, insomnia, and mood disturbance on functional status and quality of life of elderly patients with cancer. Crit Rev Oncol Hematol. 2011 May;78(2):127-37. eng. Epub 2010/04/21. doi:10.1016/j.critrevonc.2010.03.002.
Fox RS, Ancoli-Israel S, Roesch SC, Merz EL, Mills SD, Wells KJ, Sadler GR, Malcarne VL. Sleep disturbance and cancer-related fatigue symptom cluster in breast cancer patients undergoing chemotherapy. Support Care Cancer. 2020 Feb;28(2):845-855. eng. Epub 2019/06/05. doi:10.1007/s00520-019-04834-w.
Abrahams HJG, Gielissen MFM, Verhagen C, Knoop H. The relationship of fatigue in breast cancer survivors with quality of life and factors to address in psychological interventions: A systematic review. Clin Psychol Rev. 2018 Jul;63:1-11. eng. Epub 20180517. doi:10.1016/j.cpr.2018.05.004.
Rha SY, Lee J. Stable Symptom Clusters and Evolving Symptom Networks in Relation to Chemotherapy Cycles. J Pain Symptom Manage. 2021 Mar;61(3):544-554. eng. Epub 20200820. doi:10.1016/j.jpainsymman.2020.08.008.
Janz NK, Mujahid M, Chung LK, Lantz PM, Hawley ST, Morrow M, Schwartz K, Katz SJ. Symptom experience and quality of life of women following breast cancer treatment. J Womens Health (Larchmt). 2007 Nov;16(9):1348-61. eng. Epub 2007/11/16. doi:10.1089/jwh.2006.0255.
Yang S, Chu S, Gao Y, Ai Q, Liu Y, Li X, Chen N. A Narrative Review of Cancer-Related Fatigue (CRF) and Its Possible Pathogenesis. Cells. 2019 Jul 18;8(7). eng. Epub 2019/07/22. doi:10.3390/cells8070738.
An Y, Fu G, Yuan G. Quality of Life in Patients With Breast Cancer: The Influence of Family Caregiver's Burden and the Mediation of Patient's Anxiety and Depression. J Nerv Ment Dis. 2019 Nov;207(11):921-926. eng. Epub 2019/09/14. doi:10.1097/nmd.0000000000001040.
O'Connor M, Weir J, Butcher I, Kleiboer A, Murray G, Sharma N, Thekkumpurath P, Walker J, Fallon M, Storey DJ, Sharpe M. Pain in patients attending a specialist cancer service: prevalence and association with emotional distress. J Pain Symptom Manage. 2012 Jan;43(1):29-38. eng. Epub 2011/06/18. doi:10.1016/j.jpainsymman.2011.03.010.
Bower JE, Asher A, Garet D, Petersen L, Ganz PA, Irwin MR, Cole SW, Hurvitz SA, Crespi CM. Testing a biobehavioral model of fatigue before adjuvant therapy in women with breast cancer. Cancer. 2019 Feb 15;125(4):633-641. eng. Epub 2018/12/19. doi:10.1002/cncr.31827.
Walker J, Hansen CH, Martin P, Symeonides S, Ramessur R, Murray G, Sharpe M. Prevalence, associations, and adequacy of treatment of major depression in patients with cancer: a cross-sectional analysis of routinely collected clinical data. Lancet Psychiatry. 2014 Oct;1(5):343-50. eng. Epub 2015/09/12. doi:10.1016/s2215-0366(14)70313-x.
Nolan TS, Frank J, Gisiger-Camata S, Meneses K. An Integrative Review of Psychosocial Concerns Among Young African American Breast Cancer Survivors. Cancer Nurs. 2018 Mar/Apr;41(2):139-155. eng. Epub 2017/02/22. doi:10.1097/ncc.0000000000000477.
Saligan LN, Olson K, Filler K, Larkin D, Cramp F, Yennurajalingam S, Escalante CP, del Giglio A, Kober KM, Kamath J, Palesh O, Mustian K. The biology of cancer-related fatigue: a review of the literature. Support Care Cancer. 2015 Aug;23(8):2461-78. eng. Epub 2015/05/16. doi:10.1007/s00520-015-2763-0.
Kim S, Miller BJ, Stefanek ME, Miller AH. Inflammation-induced activation of the indoleamine 2,3-dioxygenase pathway: Relevance to cancer-related fatigue. Cancer. 2015 Jul 1;121(13):2129-36. eng. Epub 2015/03/03. doi:10.1002/cncr.29302.
Nguyen J, Cramarossa G, Bruner D, Chen E, Khan L, Leung A, Lutz S, Chow E. A literature review of symptom clusters in patients with breast cancer. Expert Review of Pharmacoeconomics & Outcomes Research. 2011 2011/10/01;11(5):533-539. doi:10.1586/erp.11.55.
Cherwin CH. Gastrointestinal symptom representation in cancer symptom clusters: a synthesis of the literature. Oncol Nurs Forum. 2012 Mar;39(2):157-65. eng. Epub 2012/03/01. doi:10.1188/12.Onf.157-165.

No competing interests reported.

SupplementaryMaterials.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Symptom networks of multidimensional symptom experiences in breast cancer survivors: A network analysis

Status:

Version 1

Abstract

Objectives

Methods

Results

Conclusions

Figures

1 Introduction

2 Materials and Methods

2.1 Study settings and participants

2.2 Data collection

2.3 Measures

2.3.1 The Zung Self-Rating Depression and Anxiety Scale

2.3.2 The European Organization for Research and Treatment of Cancer Quality of Life Questionnaire

2.4 Statistical analysis

2.4.1 Network estimation

2.4.2 Network stability

2.4.3 Network comparison

2.4.4 Directed acyclic graph (DAG)

2.5 Ethics statement

3 Results

3.1 Characteristics of participants

3.2 Prevalence and Symptom Scores of Breast Cancer Patients

3.3 Network Structure

3.4 Node Centrality

3.4 Network stability

3.5 Network comparison test

3.6 Directed acyclic graph (DAG)

4 Discussion

5 Limitation

6 Conclusion

Declarations

Authors’ disclosures

Conflict of interest

Funding

Author Contribution

Acknowledgement

Data Availability

References

Additional Declarations

Supplementary Files

Status:

Version 1