Pre-Treatment Perceived Social Support Is Associated With Chemotherapy-Induced Peripheral Neuropathy in Patients With Breast Cancer: A Longitudinal Study
Article information
Abstract
Objective
Previous studies have reported an association between cancer-related symptoms and perceived social support (PSS). The objective of this study was to analyze whether Chemotherapy-Induced Peripheral Neuropathy (CIPN), a prevalent side effect of chemotherapy, varies according to PSS level using a validated tool for CIPN at prospective follow-up.
Methods
A total of 39 breast cancer patients were evaluated for PSS using the Multidimensional Scale of Perceived Social Support (MSPSS) prior to chemotherapy and were subsequently grouped into one of two categories for each subscale: low-to-moderate PSS and high PSS. CIPN was prospectively evaluated using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Chemotherapy-Induced Peripheral Neuropathy 20 (CIPN20) at five time points. A linear mixed-effects model with square root transformation was employed to investigate whether the CIPN20 scales varied by PSS level and time point.
Results
Statistical analysis of the MSPSS total scale and subscales revealed a significant effect of the friends subscale group and time point on the CIPN20 sensory scale. The sensory scale score of CIPN20 was found to be lower in participants with high PSS from friends in comparison to those with low-to-moderate PSS at 1 month post-chemotherapy (p=0.010).
Conclusion
This is the first study to prospectively follow the long-term effect of pre-treatment PSS from friends on CIPN. Further studies based on larger samples are required to analyze the effects of PSS on the pathophysiology of CIPN.
INTRODUCTION
Chemotherapy-Induced Peripheral Neuropathy (CIPN) is a common side effect experienced by approximately 70% of cancer patients undergoing chemotherapy [1]. CIPN is frequently induced by anticancer drugs that are commonly utilized in the treatment of breast cancer, including docetaxel, paclitaxel, and carboplatin. Consequently, it is reported to occur in approximately 60% of patients with breast cancer [2]. Patients with CIPN frequently report sensory symptoms, such as numbness and tingling, although it can also manifest as motor and autonomic symptoms [3].
CIPN is a side effect with a high symptom burden, manifesting approximately 2 months into chemotherapy and persisting in 60% of patients at 3 months after the end of treatment [3,4]. CIPN has been identified as a dose-limiting toxicity, resulting in the cessation or reduction of chemotherapy treatment and potentially an elevated risk of mortality in cancer patients [5]. CIPN has been demonstrated to result in a decline in activities of daily living and an increased propensity of falls, which in turn has been linked to a decline in self-efficacy in cancer patients [5,6]. Moreover, it has been suggested to markedly diminish quality of life in cancer patients even 5 years after the conclusion of chemotherapy [7].
Nevertheless, the risk factors for CIPN remain poorly defined, impeding the implementation of effective screening and prevention strategies [8]. Prior research has indicated that pre-treatment depression and anxiety are associated with CIPN [9]. However, there is a paucity of studies examining other potential psychosocial risk factors for CIPN [10]. Given that the majority of identified risk factors for CIPN are unmodifiable, such as the cumulative dose of anticancer drugs and age, there is a need for research into potentially modifiable psychosocial risk factors [11].
Perceived social support (PSS) has been identified as a psychosocial risk factor associated with a range of mental and physical symptoms experienced by patients with breast cancer [12]. As posited by the biopsychosocial model, PSS is defined as a holistic concept of the psychosocial profile of the individual from whom social support is provided [12]. This concept encompasses the subjective perception of an individual’s social networks and the instrumental support they receive from these networks [13]. It has been demonstrated that patients diagnosed with breast cancer frequently require substantial social support during the course of their treatment [14]. Furthermore, it has been shown that these individuals often receive the majority of their emotional support from family members and friends, particularly during the early stages of their treatment [14]. The evidence indicates that levels of PSS remain relatively consistent throughout the course of chemotherapy [15]. This observation has prompted the design of prospective studies to investigate the long-term effects of pre-chemotherapy PSS on quality of life, as well as psychological and physical symptoms, in breast cancer patients [15,16].
The relationship between psychological and physical symptoms experienced by cancer patients and social support can be explained by two main models [17]. The first is the stress buffering model, which states that social support mitigates the negative impact of cancer or cancer treatment on health [18]. The other hypothesis is the main effect model, which states that social support affects disease outcome independently of the stress response [17]. The pathogenesis of CIPN is currently unclear, but neuroinflammation has been hypothesized. Given the recognized long-term anti-inflammatory effects of social support [19], it was hypothesized that it would act as a buffer against the inflammatory response induced by anticancer drugs.
In our preceding study, it was demonstrated that chemotherapy-induced cancer-related symptoms manifest with greater severity in breast cancer patients with lower levels of PSS compared to those with higher levels of PSS [16]. Furthermore, it was demonstrated that this effect is also true for sensory peripheral neuropathy symptoms such as numbness and tingling. However, previous studies have utilized a generalized instrument for the assessment of cancer-related symptoms, thereby precluding a specific assessment of CIPN. Moreover, although CIPN persists long after the conclusion of chemotherapy, no studies have examined the trajectory of CIPN in relation to PSS over an extended period, which constrains the capacity to assess the long-term impact of PSS on CIPN.
The aim of this study was to measure PSS before chemotherapy, to classify breast cancer patients according to their PSS level, and to determine whether the severity of CIPN varies according to the level of PSS through prospective followup before, during, and after chemotherapy. It was hypothesized that the severity of CIPN would be lower in the higher level of PSS group compared to the lower level of PSS group. Furthermore, it was predicted that this effect would remain significant even after adjusting for other psychosocial risk factors, such as depression and anxiety.
METHODS
Settings and participants
The participants for this study were recruited from Seoul National University Hospital in Seoul, Republic of Korea, from December 2019 to January 2023. A total of 136 patients were initially screened for participation at baseline, and 39 provided written informed consent (Figure 1). The present study was predicated on data from the study, “Chemotherapy-Induced Circadian Rhythm Disruption,” which was designed to determine the association of cancer-related symptoms with circadian rhythms and psychological factors. Inclusion criteria were used as follows to control for variables that affect circadian rhythm and cancer-related symptoms: 1) A diagnosis of breast cancer with subsequent anticipation of chemotherapy; 2) premenopausal status; and 3) invasive adenocarcinoma of the breast classified as stages I–III. The exclusion criteria were outlined in a paper that examined the association between circadian rhythms and cancer-related symptoms, which was an another objective of this study [20].
Flowchart of the enrollment and follow-up schedule. *The primary reason for declining to participate was the perceived burden of the hospitalization assessment and blood extraction procedures that were integral components of the study; †Eight participants received a chemotherapy regimen that differed from the typical eight cycles; ‡Due to delays in enrollment resulting from the global pandemic caused by COVID-19, 19 participants were unable to complete the follow-up phase within the approved timeframe for the study. CIPN, Chemotherapy-Induced Peripheral Neuropathy; PSS, perceived social support.
Data collection
One or 2 weeks prior to the commencement of chemotherapy (T0), participants were evaluated for sociodemographic, clinical and psychiatric characteristics, including PSS and known risk factors for CIPN. CIPN was evaluated at T0, on the day of the third (C3) and seventh (C7) cycles of chemotherapy, 1 month post-chemotherapy completion (T1), and at 9 months post-chemotherapy completion (T2). A ±2-week window was allowed for scheduling. This timeline was established in accordance with the conventional eight-cycle chemotherapy regimen typically prescribed to patients with breast cancer (AC → D, doxorubicin 60 mg/m2 and cyclophosphamide 600 mg/m2 once every 3 weeks for four cycles, followed by docetaxel 300 mg/m2 once every 3 weeks for four cycles). In instances where the number of chemotherapy cycles was not eight, assessments were conducted on the third and seventh injection days of an eight-cycle regimen. In cases where the chemotherapy regimen consisted of four cycles, assessments at C7 were not conducted.
This study was undertaken in accordance with the 2013 revised Declaration of Helsinki and with the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use-Good Clinical Practice (ICHGCP) guidelines. Before initiating the recruitment phase of this study, the research protocol was registered with the Clinical Research Registry (www.clinicaltrials.gov) using the identifier NCT04364347. Ethical approval was granted by the Institutional Review Board of Seoul National University Hospital (4.22.2019/No. 1903-052-1017).
Study measures
PSS
The validated Korean version of the Multidimensional Scale of Perceived Social Support (MSPSS) was employed to evaluate the PSS of the participants [21,22]. The MSPSS is a 12-item scale comprising three subscales: significant others (spouse or partner), family, and friends, as well as a total scale. Participants were required to rate their PSS on a 7-point Likert scale, with one indicating a strong disagreement and seven indicating a strong agreement. Higher scores indicate a higher level of PSS. In a validation study conducted with 349 Korean breast cancer survivors, the mean scores for the total scale, subscales of significant others, family, and friends were reported to be 5.94±1.10, 5.86±1.70, 6.24±1.13, and 5.71±1.36, respectively [22]. The Cronbach’s alpha coefficients for the significant other, family, and friends subscale and total scale were 0.90, 0.94, 0.94, and 0.94, respectively, in the present study.
CIPN
To evaluate CIPN, the validated Korean version of the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Chemotherapy-Induced Peripheral Neuropathy (EORTC QLQ-CIPN20) was employed [23,24]. The EORTC QLQ-CIPN20 is an effective instrument for evaluating alterations in peripheral neuropathy, demonstrating a robust correlation with clinician ratings [25]. The EORTC QLQ-CIPN20 is a 20-item self-report questionnaire comprising three scales: sensory (nine items), motor (eight items), and autonomic (three items). The questionnaire comprises 19 questions for women, with the exception of one question regarding erection from the autonomic scale. Participants were required to rate their symptoms of the past week on a 4-point Likert scale. The scores for the three scales were scaled to 100 points, with higher scores indicating a greater severity of symptoms, in accordance with the scoring manual from the EORTC. Furthermore, a total score was calculated for all items [26]. In this study, the Cronbach’s alpha coefficients for sensory, motor, and autonomic scales, and total score were 0.89, 0.74, 0.60, and 0.91, respectively.
Anxiety and depression
Previous studies have indicated a correlation between pretreatment anxiety and depression and the development of CIPN [11,27]. In this study, the validated Korean version of the hospital anxiety and depression scale (HADS) was employed for the assessment of anxiety and depression at T0 [28,29]. The HADS comprises seven items pertaining to the anxiety subscale and seven items related to the depression subscale. Participants responded to each item using a 4-point Likert scale. The range of scores for each subscale is bounded between 0 and 21. In the context of cancer patients, an anchor point of 7 and 9 for anxiety and depression, respectively, is considered to represent a clinically significant cut-off point [30].
Statistical analyses
In order to examine the effect of PSS on CIPN, two PSS groups were defined on the basis of MSPSS scores. In accordance with the scoring manual from Zimet and colleagues (https://gzimet.wixsite.com/mspss), MSPSS subscales and total scale were classified into three groups (a score of 1 to 2.9 indicating low PSS, 3 to 5 indicating moderate PSS, and 5.1 to 7 indicating high PSS) (Supplementary Table 1). Subsequently, the low and moderate PSS groups were merged to form a new category, designated as the “low-to-moderate” PSS group, given the relatively small proportion of individuals in the low PSS group.
To compare the baseline characteristics of the low-to-moderate and high PSS groups, the student t-test and the Mann-Whitney U-test were employed, in accordance with the normality assumptions derived from the Shapiro–Wilk test. With regard to the categorical variables, Fisher’s exact test and the chi-square test were utilized.
To determine the extent of CIPN severity between the lowto-moderate and high PSS groups, descriptive statistics of all scales and global scores of the EORTC QLQ-CIPN20 were conducted for both groups at each time point (Supplementary Table 2). These were subsequently analyzed using linear mixedeffects models, with age (a continuous variable), chemotherapy regimen (taxane-based regimen vs. others), depression, and anxiety (categorical variables) serving as covariates. Prior to analysis, a square root transformation was applied to each dependent variable, as the residual plots did not fulfill the prerequisite of equality of variance when the EORTC QLQCIPN20 scales and total scores were utilized as raw scores. To ascertain the validity of the normality assumptions, histograms of residuals and normal Q-Q plots were employed for each model. In addition, the Kolmogorov–Smirnov test was employed, and the results demonstrated that the assumptions were satisfied.
To ascertain the differences in CIPN between the PSS groups at each time point, interaction terms for PSS groups and time points were incorporated into the model. In the event that the interaction was found to be statistically significant, post hoc analyses were conducted for each time point. All analyses were performed using IBM SPSS Statistics, version 29.0 (IBM Corp.). A p value <0.05 (two-tailed) was considered statistically significant for all tests. To account for multiple comparisons, the Bonferroni correction was applied, with a p value <0.05/the number of analyses as the threshold for significance.
RESULTS
Baseline characteristics of the study participants
In total, 39 participants were analyzed, 15 of whom completed follow-up by T2 (Figure 1). Prior to performing linear mixed-effects analysis, a comparison was made between those who were lost to follow-up and those who completed followup. Those who lost follow-up had a higher proportion of a high school diploma or less and neoadjuvant chemotherapy, but there were no significant differences in other baseline characteristics (Supplementary Table 3). The mean age of the participants was 42.10 years and more than 80% of the participants received taxane-based chemotherapy (Table 1). None of the participants had a history of neuropathy. With regard to the friends subscale, 12 participants were classified as lowto-moderate PSS, while 27 were classified as high PSS. No statistical differences were found in sociodemographic, clinical, and psychiatric characteristics between participants with low-to-moderate PSS and those with high PSS from friends. Furthermore, no statistically significant difference was observed between the two groups with regard to PSS from friends for known risk factors for CIPN, including age, history of diabetes mellitus and neuropathy, smoking, chemotherapy regimen and cumulative dose, anxiety, and depression [4].
Baseline characteristics of participants by the level of PSS from friends
PSS of the study participants
The mean total scale score of the MSPSS of the participants was 5.56±1.26 and 82.1% showed a high level of PSS. The mean scores for the MSPSS subscales of significant others, family, and friends were 5.39±1.36, 5.72±1.52, and 5.56±1.60, respectively. The proportion of participants exhibiting low PSS for each subscale was approximately 5%–10%, while high PSS was observed in 69.2% of participants in the friends subscale (Supplementary Table 1).
Severity of CIPN of groups by PSS from friends, by time point
The mean follow-up for CIPN in the low-to-moderate and high PSS groups was 281.10±147.78 and 275.46±163.93 days, respectively (Mann–Whitney U-test p=0.958). Interaction analyses of the group by the level of MSPSS subscales and total scale and time point showed that the groups by PSS from friends showed a difference in the severity of CIPN on the total score and sensory scale of the EORTC QLQ-CIPN20 with square root transformation, following adjustment for age and chemotherapy regimen (Supplementary Table 4). This finding remained statistically significant after additional adjustment for anxiety and depression, which have previously been identified as psychosocial predictors for CIPN (Table 2 and Figure 2). Post-hoc analyses showed that participants with high PSS from friends exhibited significantly lower scores on the sensory scale score of the EORTC QLQ-CIPN 20 with square root transformation compared to those with low-tomoderate PSS from friends at T1 following adjustment for age, chemotherapy regimen, anxiety and depression (p=0.010) (Table 3). The overall score on the sensory scale of the participants with high PSS from friends was found to be lower in comparison with those with the low-to-moderate PSS from friends across all time points (p=0.035).
Effects of PSS on the EORTC QLQ-CIPN20 scores with square root transformation, as a function of groups by PSS level and time points
The EORTC-QLQ CIPN20 scales with square root transformation according to the groups by the level of PSS from friends* (aside Results or Discussion section). A: Total score. B: Sensory scale. C: Motor scale. D: Autonomic scale. *The square root-transformed scores with standard deviation. EORTC-QLQ CIPN20, European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Chemotherapy-Induced Peripheral Neuropathy 20; PSS, perceived social support.
EORTC-QLQ CIPN20 scores with square root transformation of groups by PSS level, by time point
DISCUSSION
This study investigated whether the severity of CIPN differs according to the level of PSS before chemotherapy in breast cancer patients with prospective follow-up before, during, and after chemotherapy.
Participants with high PSS from friends had lower scores on the sensory scale of the EORTC QLQ-CIPN20 compared to those with low-to-moderate PSS at 1 month after the end of chemotherapy (T1), after adjusting for age, chemotherapy regimen, anxiety and depression. As the participants’ chemotherapy treatment typically consisted of eight cycles, with a 3-week interval between each cycle, the trajectory of CIPN in this study was consistent with that reported in previous studies, which indicate that CIPN predominantly onset around the second month of chemotherapy [3]. Furthermore, a proportion of subjects indicated the presence of relatively elevated levels of CIPN symptoms 9 months following the conclusion of chemotherapy. This observation indicates that CIPN persists beyond the termination of chemotherapy, thereby underscoring the necessity of prolonged follow-up to investigate the correlates of CIPN.
The discrepancy in CIPN severity between the low-to-moderate and high PSS from friends group was notable on the sensory scale. This finding may be attributed to the fact that the majority of patients received a taxane-based regimen, an anticancer drug that is primarily associated with sensory neuropathy [4]. An approximation of the raw score for the sensory subscale at T1 was derived by squaring the estimated mean from the model. This calculation yielded a difference between the two groups estimated to be approximately 12.1, which exceeds the minimal clinically important difference for the sensory scale (6.9–16.4 as scaled score) [31]. The fact that this difference was statistically significant even after adjustment for age, chemotherapy regimen, anxiety and depression sug-gests that PSS may be an independent potential risk factor for CIPN, rather than a mediator.
The MSPSS total scale and subscale scores documented in this study exhibited congruence with or were marginally lower than the outcomes reported by Kim et al. [22] in a study of 349 Korean breast cancer survivors. This discrepancy may be attributable to the fact that the study by Kim et al. [22] was conducted among patients who had completed cancer treatment and had a spouse, whereas the present study was conducted among patients who were about to undergo chemotherapy, the proportion of unmarried participants was over 20%, and the number of coresidents was relatively small.
The friends subscale of the MSPSS exhibited a significant group effect on CIPN in this study. This finding is consistent with the findings from the previous study, which demonstrated a significant correlation between the level of PSS from friends and the severity of numbness and tingling, whereas no such relationship was observed for other subscales including significant others and family [16]. The present study builds on the findings of a previous investigation, which was conducted on an outpatient basis and comprised a sample of over 180 participants. Despite the implementation of more stringent inclusion criteria and a reduced number of participants in the present study, the findings were similar to those of the earlier study.
In a study by Roy et al. [19] it was discovered that the immune system response to chemotherapy in breast cancer patients was associated with the level of social support received from friends prior to chemotherapy, with the effect differing according to the social support provided by a romantic partner. This indicates that the biological effects of social support may be contingent on the source of the support, that is to say, the type of the relationship with the patient.
In contrast to men, who have been shown to receive high levels of social support from their spouses, women have been found to receive high levels of social support from their friends [32]. These supportive friendships have been associated with increased satisfaction and intimacy in women’s relationships [33]. The impact of a supportive friendship on well-being has been demonstrated to be positive, both physically and psychologically [34]. In the context of non-white breast cancer patients, the presence of friendships has been identified as a significant predictor of survival [35]. The hormonal response that is involved in women’s coping responses to stress is proposed to be the mechanism by which support from friends exerts this physical effect [35]. In addition to the fight-or-flight response, women exhibit behavioral changes that strengthen their social network of friends and relatives by activating oxytocin-induced befriending in stressful situations, which is known to activate the estrogen and endogenous opioid systems [36]. The inhibitory effect of endogenous opioids on the mechanical hypersensitivity of CIPN suggests a plausible mechanism through which social support from friends contributes to the pathogenesis of CIPN [37].
The absence of significant discrepancies in CIPN severity based on the level of PSS provided by significant others or family members may be attributed to several factors. First, women prioritize social support from friends over spouses, which may reduce the effect of PSS from significant others on CIPN [32]. Secondly, the number of participants experiencing low-to-moderate PSS from family was the smallest of the subscales (eight participants), which may have resulted in insufficient statistical power. Further study with a larger sample is required to elucidate the effect of PSS from family and significant others on CIPN.
The mechanisms by which high PSS exerts a protective effect against CIPN warrant further investigation. The results of the present study suggest that PSS may offer a protective effect against the side effects of chemotherapy, as indicated by the stress-buffering model [38]. As social pain from social exclusion interacts with physical pain by sharing somatosensory representations, PSS may act as a protective factor against physical pain, which is one of the sensory symptoms of CIPN [39]. Furthermore, neuroinflammation has been postulated as a pivotal mechanism underlying CIPN [4]. The evidence suggests that high levels of social support are associated with lower levels of inflammatory markers [40]. The protective effect of PSS against inflammation has been reported to persist over a 10-year follow-up period, suggesting a potential link between the pathogenesis of CIPN and the anti-inflammatory effects of PSS [41].
Strength and limitations
This is the inaugural study to investigate the trajectory of CIPN in relation to the PSS, with follow-up extending up to 9 months post-chemotherapy. The study employed the EORTC QLQ-CIPN20, a validated patient-reported outcome measure for the assessment of CIPN. The EORTC QLQCIPN20 is a sensitive tool for detecting changes in symptoms and is particularly effective in follow-up studies [3]. However, this study is subject to the following limitations. Firstly, the small number of participants and the heterogeneity in the chemotherapy regimens limit the statistical power for evaluating PSS as a risk factor. Secondly, patient enrollment was delayed due to the impact of the global pandemic of COVID-19, which resulted in a lack of sufficient observations of CIPN at T2. Thirdly, the study was conducted in premenopausal women from the Korean population, and inclusion and exclusion criteria were designed to control factors that affect circadian rhythm and cancer-related symptoms, which limits its generalizability. In light of the younger age of diagnosis among Asian breast cancer patients and the varying social support needs of different ethnic groups, there is a need for research in diverse populations [42,43].
Clinical implications
From the perspective of psychosomatic medicine, the findings of this study support the hypothesis that CIPN, a side effect of biological therapy, is influenced by psychosocial factors. Recent studies have indicated that PSS exerts a protective effect on the immune system and inflammatory response, and the association between CIPN and the inflammatory response has been acknowledged [4,19,41]. The aforementioned studies indicate that PSS may prove useful in identifying individuals at an elevated risk of developing CIPN. Moreover, for a treatment paradigm that is currently underdeveloped in terms of both prevention and treatment for CIPN, the implementation of a supportive care approach based on a biopsychosocial model may contribute to a reduction in the symptom burden associated with CIPN [4].
Conclusion
The findings of this study indicate that participants with high PSS from friends prior to chemotherapy exhibited lower level of CIPN compared to those with low-to-moderate PSS, as observed in a prospective follow-up. Further investigation into the mechanistic effects of PSS on CIPN in larger populations is recommended.
Supplementary Materials
The Supplement is available with this article at https://doi.org/10.30773/pi.2024.0336.
Distribution of participants by PSS levels
Raw scores of the EORTC QLQ-CIPN20 scales of groups by PSS from friends, by time point
Baseline characteristics of participants who were lost to follow-up and those who completed follow-up
Effects of PSS on the EORTC QLQ-CIPN20 scores, with square root transformation, as a function of groups by PSS level and time points without adjustment for anxiety and depression
Notes
Availability of Data and Material
The data supporting the findings of this study are available on request from the corresponding author. Due to the presence of privacy or ethical restrictions, the data are not publicly available.
Conflicts of Interest
The authors have no potential conflicts of interest to disclose.
Author Contributions
Conceptualization: Joon Sung Shin, Sanghyup Jung, Geun Hui Won, Sun Hyung Lee, Jaehyun Kim, Saim Jung, Chan-Woo Yeom, Kwang-Min Lee, Kyung-Lak Son, Bong-Jin Hahm. Data curation: Joon Sung Shin, Sanghyup Jung, Geun Hui Won. Formal analysis: Joon Sung Shin, Sanghyup Jung. Funding acquisition: Bong-Jin Hahm. Investigation: Joon Sung Shin, Chan-Woo Yeom, Kyung-Lak Son. Methodology: Sanghyup Jung, Geun Hui Won, Jang-il Kim, Sook Young Jeon, Han-Byoel Lee, Bong-Jin Hahm. Resources: Jang-il Kim, Sook Young Jeon, Han-Byoel Lee. Supervision: Chan-Woo Yeom, Kwang-Min Lee, Kyung-Lak Son, Bong-Jin Hahm. Writing—original draft: Joon Sung Shin. Writing—review & editing: all authors.
Funding Statement
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (Ministry of Science and ICT, No. 2019R1A2C1003675).
Acknowledgments
None