Skip to main content

Some NLM-NCBI services and products are experiencing heavy traffic, which may affect performance and availability. We apologize for the inconvenience and appreciate your patience. For assistance, please contact our Help Desk at info@ncbi.nlm.nih.gov.

Annals of Behavioral Medicine: A Publication of the Society of Behavioral Medicine logoLink to Annals of Behavioral Medicine: A Publication of the Society of Behavioral Medicine
. 2021 May 15;56(2):137–145. doi: 10.1093/abm/kaab035

Fatigue Perpetuating Factors as Mediators of Change in a Cognitive Behavioral Intervention for Targeted Therapy-Related Fatigue in Chronic Myeloid Leukemia: A Pilot Study

Kelly A Hyland 1,2, Ashley M Nelson 3, Sarah L Eisel 1, Aasha I Hoogland 1, Javier Ibarz-Pinilla 4, Kendra Sweet 4, Paul B Jacobsen 5, Hans Knoop 6, Heather S L Jim 1,
PMCID: PMC8832107  PMID: 33991085

Abstract

Background

Cognitive behavioral therapy for targeted-therapy related fatigue (CBT-TTF) has demonstrated preliminary efficacy in reducing fatigue in patients treated with tyrosine kinase inhibitors (TKIs) for chronic myeloid leukemia (CML).

Purpose

The aim of the current analyses was to explore whether fatigue perpetuating factors (disturbed sleep/wake cycle, dysregulated activity patterns, maladaptive cognitions about fatigue and cancer, insufficient processing of cancer and treatment, inadequate social support and interactions, heightened fear of cancer progression) changed over time in patients receiving CBT-TTF, and whether the effect of CBT-TTF on fatigue was mediated by these factors.

Methods

Secondary data analyses were conducted from a pilot randomized controlled trial. Patients with CML treated with a TKI who reported moderate to severe fatigue were randomized 2:1 to CBT-TTF delivered via FaceTime for iPad or a waitlist control condition (WLC). Self-report measures of fatigue and fatigue perpetuating factors were obtained before randomization and post-intervention (i.e., approximately 18 weeks later). Mixed model and mediation analyses using bootstrap methods were used.

Results

A total of 36 participants (CBT-TTF n = 22, WLC n = 14) who had baseline and 18-week follow-up data and attended >5 sessions for CBT-TTF were included. Participants randomized to CBT-TTF reported improvements in activity (mental, physical, social, p’s ≤ .023) and cognitions (helplessness, catastrophizing, focusing on symptoms, self-efficacy, p’s ≤ .003) compared to WLC. Mental activity, social activity, self-efficacy, helplessness, and focusing on symptoms, as well as sleep and insufficient processing (avoidance) mediated the relationship between treatment group and fatigue.

Conclusions

CBT-TTF appears to improve TKI-related fatigue in CML patients through changes in behavior (sleep, activity patterns) and cognitions about fatigue and cancer. A larger randomized controlled trial is warranted to confirm these findings.

Keywords: Cancer-related fatigue, ognitive-behavioral therapy, hronic myeloid leukemia, ediation, ognitions, ctivity


Cognitive behavioral therapy for targeted therapy-related fatigue appears to improve fatigue in patients with chronic myeloid leukemia through changes in behavior (sleep, activity) and cognitions about cancer and fatigue.

Introduction

Targeted therapies, including tyrosine kinase inhibitors (TKIs), have transformed the treatment of chronic myeloid leukemia (CML) from a fatal disease to a chronic condition. However, TKIs are not without side effects. Fatigue is one of the most common and debilitating side effects of these medications, with as many as 68% of patients with CML treated with a TKI reporting moderate to severe fatigue [1, 2].

Cognitive behavioral therapy for fatigue (CBTF) has demonstrated efficacy in reducing fatigue in individuals with chronic illness, including people with advanced cancer and post-treatment cancer survivors [3–6]. CBTF is based on the precipitating-perpetuating model of cancer-related fatigue [3, 7]. This model posits that precipitating factors (e.g., cancer and its treatment) initially contribute to increased fatigue. Patients subsequently develop behavioral (e.g., daytime naps, limiting social activities) and cognitive (e.g., focusing on bodily symptoms) responses in an attempt to manage their fatigue. Although these changes may be helpful in the short term, they serve to perpetuate and potentially exacerbate fatigue in the long term. CBTF aims to disrupt this cycle by helping patients address the cognitive, behavioral, and social processes that are believed to be involved in sustaining fatigue, as indicated by the literature and clinical experience.

Given its effectiveness in cancer populations, we adapted CBTF specifically for targeted therapy-related fatigue (CBT-TTF) and evaluated the feasibility, acceptability, and preliminary efficacy of the intervention in a pilot randomized trial [8, 9]. CBT-TTF was found to be acceptable (i.e., study participation rate of 59%) and feasible (intervention completion rate of 79%). Patients who received CBT-TTF demonstrated large improvements in fatigue and quality of life relative to those randomized to a waitlist control (Cohen’s d’s > 1) [9]. Improvements in fatigue and quality of life were both statistically and clinically significant. To our knowledge, this was the first published report of a psychosocial intervention for targeted therapy-related fatigue.

The aim of the current analyses was to investigate mechanisms of improvement in fatigue among patients who received CBT-TTF. Mechanisms underlying decreases in fatigue have been evaluated in other chronic disease groups (e.g., chronic fatigue syndrome, multiple sclerosis, Type I diabetes) [10–12], and evidence suggests that transdiagnostic factors including reduced sleep disturbance, increased activity, and changes in cognitive factors (e.g., increased self-efficacy, decreased focusing on symptoms) may be responsible [10–13]. However, mechanistic analyses in cancer populations are lacking, and the existing evidence is mixed. For example, physical activity-based interventions for fatigue have demonstrated efficacy in reducing fatigue in cancer populations [14], but physical activity measured by actigraphy was not a significant mediator of the relationship between group and fatigue in CBTF with disease-free cancer survivors [15]. Identifying essential components of CBTF in cancer populations is critical for maximizing intervention efficacy and minimizing patient burden.

The current paper describes secondary analyses to evaluate: (a) if perpetuating factors targeted by the intervention changed in the CBT-TTF group compared to WLC, and (b) whether perpetuating factors mediated the observed differences in change in fatigue by the treatment group. We hypothesized that perpetuating factors would change significantly from pre to post-intervention in the CBT-TTF group compared to WLC, and that changes in perpetuating factors would mediate the effect of CBT-TTF on fatigue. Given the pilot nature of the parent study, these findings represent a preliminary evaluation; a larger, randomized controlled trial is warranted to further test these hypotheses.

Methods

Participants

Eligible participants met the following criteria: (a) receiving TKI treatment for chronic phase CML at Moffitt Cancer Center, (b) at least 18 years old, (c) able to speak/read English, (d) able to provide written informed consent, (e) had not received treatment for cancer other than non-melanoma skin cancer in the previous 5 years, (f) on the same oral TKI treatment for at least 3 months, (g) reported new onset or worsening fatigue since starting TKI treatment, (h) reported moderate to severe fatigue in the past week on the Fatigue Symptom Inventory (≥4 on a 0–10 scale), (i) had no other medical condition (e.g., multiple sclerosis, fibromyalgia) that could account for the fatigue, and (j) did not plan to discontinue TKI treatment within the next three months. Participants included in the current analyses were those who completed both Time 1 and Time 2 questionnaires, and CBT-TTF participants who attended at least five intervention sessions (see Participant Flow Chart; Supplementary Figure).

Procedures

The study was approved by the Moffitt Institutional Review Board and registered on clinicaltrials.gov (NCT02592447). Eligible patients interested in study participation provided written informed consent then were randomized 2:1 to receive CBT-TTF or waitlist control (WLC). Both conditions completed self-report primary and secondary outcome measures and measures of fatigue perpetuating factors (Table 1) prior to randomization and approximately 18 weeks later (for CBT-TTF, after intervention completion).

Table 1.

Measures used to determine module assignment in patients receiving CBT-TTF

Module Perpetuating factor Instrument
Sleep Disturbed sleep/wake cycle Sickness Impact Profile – Sleep Rest subscale (SIP-SR) [16, 17]
Activity Dysregulated activity patterns Checklist for Individual Strength (CIS) – Concentration subscale [18]
Sickness Impact Profile – Social Interaction subscale (SIP-SI) [16, 17]
International Physical Activity Questionnaire (IPAQ) – Short Form [19]
Cognition Maladaptive cognitions about fatigue and cancer Illness Cognition Questionnaire (ICQ) – Helplessness subscale [20]
Self-Efficacy Scale 28- Fatigue (SES28-Fatigue) [21]
Fatigue Catastrophizing Scale (FCS) [22]
Illness Management Questionnaire – Focusing on Symptoms (IMQ – Factor III) [23]
Processing Insufficient processing of cancer and treatment Impact of Events Scale – Cancer (IES) – Intrusion and Avoidance subscales [24]
Social support Inadequate social support and interactions Social Support List (SSL) – Interactions, Negative Interaction, and Discrepancies subscales [25]
Fear of progression Heightened fear of cancer progression Cancer Worry Scale (CWS)a [26]

aThe CWS was adapted to a 5-item scale for use with patients with CML.

CBT-TTF

The intervention has been described in detail previously [9]. Briefly, patients randomized to CBT-TTF met with a study therapist for an initial in-person session at Moffitt Cancer Center. The aim of the first session was to establish rapport, introduce the intervention and rationale for CBT-TTF, and orient the patient to the use of the study-provided iPad. Subsequent sessions were conducted via FaceTime for iPad; participants had the option to complete the final session in-person, if desired. The intervention was delivered by the study therapist over 18 weeks, and sessions took place approximately weekly. CBT-TTF consists of six modules targeting factors thought to perpetuate targeted therapy-related fatigue: (a) a disturbed sleep/wake cycle, (b) dysregulated activity patterns, (c) maladaptive cognitions about fatigue and cancer, (d) insufficient processing of cancer diagnosis and treatment, (e) inadequate social support and interactions, and (f) heightened fear of cancer progression. Two core modules were received by all CBT-TTF patients (i.e., sleep/wake cycle and activity patterns). Given the significant variability in presentations of fatigue, the remaining modules were administered only to patients reporting concerns in these areas on self-report measures (listed in Table 1) or during the intake clinical assessment. Perpetuating measures included multiple indicators for each module, and participants received the entire module if at least a portion was indicated. Recommended modules were discussed collaboratively with the participant to confirm appropriateness and facilitate buy-in. Although the number of modules received varied by the participant, the content delivered within each module was standardized. The duration of module delivery was also tailored to the participant, such that the amount of time dedicated to a particular module varied based on the participants’ needs. Module delivery was rolling in nature, such that there was ongoing review and checking in on previous modules as new content was being introduced. This structure was intended to promote flexibility within the manualized treatment to best meet the needs of each participant. Given these factors, the number of sessions also varied by the participant, and termination was dependent upon coverage of indicated module material and participants’ progress toward their goals for therapy.

Study modules are briefly summarized below. Examples of cognitive behavioral intervention strategies to challenge fatigue perpetuating factors have been previously described [3, 6]. The sleep module helped patients to normalize their sleep/wake rhythm using regular bed and wake up times, good sleep hygiene, and cognitive restructuring around patients’ beliefs about rest. The activity module helped patients to more evenly distribute their physical, mental, and social activities across the day, then systematically increase activity as needed using a graded activity program. The cognitions module helped patients to identify and challenge maladaptive thoughts about fatigue and cancer, employ more helpful ways of thinking about fatigue, and increase mindfulness of the present moment. The processing module helped patients to accept that they were being treated for the chronic, “invisible” disease of CML, without preoccupation or avoidance. The social support module helped patients review strengths and deficits in their social support, foster realistic expectations of others, and assert themselves within their social networks. The fear of disease progression module helped patients to determine whether disease-related fears interfered with their daily functioning, and to evaluate whether fears were realistic or helpful.

Measures

Participants self-reported demographic characteristics at baseline. Clinical variables including time since diagnosis and length of current TKI treatment were abstracted from the medical record at baseline and study completion. Fatigue as assessed with the Functional Assessment of Cancer Therapy – Fatigue (FACIT-F) scale was the primary study outcome. The FACIT-F assesses specific quality of life concerns related to fatigue in people with cancer [27]. Additional measures used to determine module assignment are listed in Table 1.

Analyses

Statistical analyses were performed using SAS version 9.4 (Cary, NC). Independent samples t-tests and chi-square tests were conducted to evaluate group differences on demographic and clinical variables. Descriptive statistics were computed to characterize intervention engagement in the CBT-TTF group, including the average number of sessions, average number of modules, and the percentage of participants who received each module. Pearson’s r correlation coefficients were calculated to examine relationships between intervention engagement and fatigue. Due to small cell sizes, analyses focused on comparison at the group level (i.e., CBT-TTF vs. WLC).

Independent samples t-tests were conducted to test for group differences on perpetuating factor measures at baseline. Longitudinal change in continuous perpetuating factor measures as a function of treatment group assignment was evaluated using mixed models. Cohen’s d effect sizes for the magnitude of change in perpetuating factors between groups were computed by dividing mean group change by the pooled standard deviation of both groups. To facilitate interpretation of effect sizes, we provide benchmark values (small (.2 to .5), medium, (.5 to.8) or large (>.8) [28]. Due to the small sample size, benchmark values should be interpreted with caution. We provide additional contextual information to characterize the magnitude of effects within the relevant research context.

Next, PROCESS macro for SAS was used to run simple mediation analyses [29]. Primary mediation analyses included perpetuating factor measures that were significant in mixed model analyses. Given the pilot nature of the study, supplemental mediation analyses were conducted on perpetuating factor measures that were not significant in mixed model analyses. We tested whether the effect of the treatment group (CBT-TTF vs. WLC) on fatigue post-intervention was mediated by the perpetuating factors examined. Perpetuating factor measure scores at Time 2 were entered as the mediator variable. PROCESS estimated the indirect effect and 95% confidence interval for the indirect effect. When the 95% confidence interval for the indirect effect did not include zero, the indirect effect was considered statistically significant.

Results

Sample Characteristics

Sociodemographic and clinical characteristics of the sample are displayed in Table 2. The majority of participants were non-Hispanic White, married, and had a college degree. There was a near equal distribution of gender (53% male), and the mean age was 56 years. On average, participants were diagnosed with CML 5.6 years prior, and had been prescribed their current TKI medication for 2.8 years. Treatment groups did not differ on baseline sociodemographic or clinical characteristics (p’s >.05).

Table 2.

Sociodemographic and clinical characteristics of the sample (N = 36)

CBT-TTF (N = 22) WLC (N = 14) p value
Age: mean (SD) [range] years 53.7 (11.0) 59.6 (12.4) .15
Gender: n (%) female 11 (50%) 6 (43%) .68
Race: n (%) white 20 (91%) 11 (79%) .30
Ethnicity: n (%) non-Hispanic 22 (100%) 14 (100%) 1.00
Education: n (%) college graduate 14 (64%) 7 (50%) .42
Marital status: n (%) married 18 (82%) 10 (71%) .46
Years since diagnosis: mean (SD) 6.7 (6.1) 3.9 (4.1) .14
Time on current TKI: mean (SD) 2.9 (2.7) 2.6 (3.2) .75

CBT-TTF, cognitive-behavioral therapy for targeted therapy-related fatigue; TKI, tyrosine kinase inhibitor; WLC, waitlist control.

Description of CBT-TTF Intervention Engagement

CBT-TTF recipients participated in an average of 13.6 sessions (SD = 1.7, range: 11–17) and received 4.5 intervention modules (SD = 1.1, range: 2–6). By design, all participants (100%) received the sleep and activity modules. Nearly all participants (91%) received the cognition module. Seventy-seven percent received the social module, 59% the fear of progression module, and 18% received the processing module. The number of indicated modules and number of sessions received was positively correlated, such that CBT-TTF participants with more indicated modules had more sessions (r = .43, p = .04). Worse fatigue at baseline was associated with more modules indicated (r = .53, p = .01) and more sessions received (r = .50, p = .02). Number of modules indicated was inversely associated with change in fatigue (r = −.47, p = .03), while number of sessions delivered was not associated with change in fatigue in the CBT-TTF group (r = .05, p = .82).

Change in Perpetuating Factors by Treatment Group

Means, standard deviations, and ranges for perpetuating factor measures at baseline and 18 weeks are presented by the treatment group (Table 3). Groups did not differ significantly on perpetuating factor measure scores at baseline with the exception of fatigue catastrophizing. Significant improvements in multiple indices of activity (i.e., physical, mental, social) and cognitions (i.e., self-efficacy, helplessness, fatigue catastrophizing, and focusing on symptoms) were observed in the CBT-TTF group relative to the WLC group (p’s ≤ .023). Effect sizes for change in the activity and cognition modules were large in magnitude (d’s ≥ .80). Treatment groups did not differ significantly on change in measures of sleep, processing, social, and fear of progression domains (p’s > .05).

Table 3.

Change in fatigue perpetuating factors

CBT-TTF (N = 22) WLC (N = 14) d value p value
Baseline Follow-up Baseline Follow-up
M (SD) [range] M (SD) [range] M (SD) [range] M (SD) [range]
Sleep
Sleep/wake dysregulation (SIP-SR) 183.6 (150.9) [0–499] 54.5 (90.3) [0–395] 195.4 (98.7) [49–395] 144.5 (100.1) [0–335] −.51 .134
Activity
Mental activity (CIS − concentration) 23.9 (6.7) [7–35] 14.5 (6.9) [5–29] 22.5 (7.8) [6–34] 20.3 (9.4) [9–34] −.86 .015
Physical activity (IPAQ – total METS) 1760.3 (2119.0) [0-7866] 4031.3 (4154.7) [198–12,945] 2046.3 (3336.8)
[0−12,240]
1420.1 (2146.5) [0−7,038] −.80 .023
Social Activity (SIP-SI)a 513.9 (266.1) [135–1,091] 108.6 (162.4) [0–652] 562.8 (274.5) [190–997] 524.2 (294.5) [103–973] −1.14 .001
Cognitions
Self-efficacy (SES) 17.7 (3.2) [11–24] 21.4 (2.7) [14–24] 18.4 (3.0) [13–23] 19.1 (3.3) [13–24] 1.19 .001
Fatigue catastrophizing (FCS) 13.7 (6.5) [1–24] 4.9 (6.3) [0–29] 8.6 (5.4) [0–17] 7.1 (5.8) [1–22] −1.22 <.001
Helplessness (ICQ) 14.1 (4.2) [6–22] 8.5 (3.5) [6–18] 13.4 (3.8) [8–19] 12.4 (3.8) [6–19] −1.05 .003
Focusing on symptoms (IMQ) 19.0 (6.9) [6–30] 8.7 (6.0) [0–24] 18.1 (9.0) [9–37] 16.5 (7.8) [6–35] −1.28 <.001
Processing
Intrusive thoughts (IES-I) 8.9 (7.3) [0–23] 4.6 (5.8) [0–22] 10.2 (7.0) [1–23] 8.1 (6.2) [1–25] −.33 .320
Avoidance (IES-A) 13.8 (10.6) [0–32] 6.9 (7.1) [0–22] 16.4 (8.6) [6–32] 15.4 (8.4) [4–32] −.61 .073
Social
Social interaction (SSL-I) 18.5 (5.6) [9–28] 20.0 (4.9) [10–28] 16.9 (3.2) [12–22] 18.0 (3.3) [13–26] .07 .853
Negative interaction (SSL-N) 10.0 (2.5) [7 – 17] 9.0 (2.7) [7–16] 11.0 (4.6) [7–20] 10.1 (3.1) [7–18] −.03 .912
Discrepancy (SSL-D) 11.5 (3.5) [8 – 21] 9.4 (3.6) [8–24] 13.3 (4.8) [8–24] 11.6 (3.5) [8–18] −.07 .826
Fear of progression
Cancer-related worry (CWS – 5 items) 9.9 (2.4) [5–15] 8.2 (1.8) [5–11] 9.6 (3.8) [5–16] 9.6 (1.9) [6–13] −.65 .058

Note: CBT-TTF, cognitive-behavioral therapy for targeted therapy-related fatigue; WLC, waitlist control.

d values refer to effect size of magnitude of change. p values refer to interaction effect for mixed model (group × time). p values <.05 bolded.

aFor SIP-SI, N = 13 for WLC.

Perpetuating Factors as Mediators of the Relationship Between Treatment Group and Fatigue

Seven perpetuating measure variables (e.g., mental activity, physical activity, social activity, self-efficacy, helplessness, fatigue catastrophizing, focusing on symptoms) demonstrated differences in change over time by treatment group and were entered into simple mediation analyses (example illustrated in Fig. 1). Bootstrapped estimates and 95% confidence intervals for the mediation models are shown in Table 4. Mental activity, social activity, self-efficacy, helplessness, and focusing on symptoms demonstrated evidence of mediation. Physical activity and fatigue catastrophizing did not mediate the relationship between group assignment and fatigue post-intervention. Across mediation models, standardized effects sizes for the α pathway ranged from β = .34 to .68 and β = .17 to .54 for the b pathway. Exploratory mediation analyses were conducted for the perpetuating measure variables that did not demonstrate differences in change over time by treatment group (Supplementary Table 1). Sleep and processing (avoidance subscale) demonstrated evidence of mediation.

Fig. 1.

Fig. 1.

Perpetuating factors as a mediator of the relationship between treatment group and fatigue post-intervention.

Standardized coefficients shown. c pathway = total effect of the intervention on the dependent variable; c′ pathway = effect of the intervention on the dependent variable when controlling for the mediator (direct effect).

* p< .05, ** p< .01, *** p< .001.

Table 4.

Bootstrapped estimates and confidence intervals for tests of simple mediation

Independent variable Mediator variable Path c (IV to DV)
Total effect B (SE)
Effect of IV on M (a) B (SE) Effect of M on DV (b) B (SE) Direct effects (c’) B (SE) Indirect effect (a x b) B (SE) [95% CI]
Group Mental activity (CIS –Concentration) −15.8 (3.5)*** −5.8 (2.7)* .8 (.18)*** −11.2 (3.0)*** −4.6 (2.6)
[10.6 to0.2]
Physical activity (IPAQ total METS) −15.8 (3.5)*** 2611.3 (1205.0)* −.0006 (.0005) −14.2 (3.8)*** −1.6 (1.2)
[−4.2 to 0.4]
Social activity (SIP-SI) −15.8 (3.5)*** −415.6 (76.0) *** .02 (.01)** −7.5 (4.4) −8.3 (4.2)
[17.7 to1.2]
Self-efficacy (SES) −15.8 (3.5)*** 2.3 (1.0)* −1.4 (.56)* −12.5 (3.5)** −3.3 (1.7)
[6.8 to0.2]
Fatigue catastrophizing (FCS)^ −15.8 (3.5)*** −4.8 (2.0) * .57 (.33) −14.2 (4.1)** −2.7 (3.0)
[−10.3 to 0.1]
Helplessness (ICQ) −15.8 (3.5)*** −3.9 (1.2)** 1.7 (.4)*** −9.1 (3.3)** −6.7 (2.2)
[11.1 to2.2]
Focusing on symptoms (IMQ) −15.8 (3.5)*** −7.8 (2.3)** .51 (.25)* −11.8 (3.9)** −4.0 (1.7)
[7.1 to0.1]

Unstandardized regression coefficients reported. IV, independent variable (treatment group); DV, dependent variable (fatigue at Time 2); M, mediator variable (perpetuating factor measure at Time 2); CI, confidence interval.

^ FCS at Time 1 entered as covariate.

*p ≤ .05, **p ≤ .01, ***p ≤ .001.

Discussion

The objectives of the current study were (a) to investigate the change in fatigue perpetuating factors in participants receiving CBT-TTF compared to WLC, and (b) to test whether perpetuating factors mediate the relationship between treatment group and post-intervention fatigue. Overall, findings are consistent with literature linking changes in fatigue perpetuating cognitions and behaviors to reductions in fatigue [10, 30]. The activity and cognition modules were received by almost all (91%) CBT-TTF participants, and significant group differences were observed on changes in all measures of activity and cognitions. This highlights both patients’ baseline dysfunction related to dysregulated activity patterns and maladaptive cognitions about fatigue and cancer, and the effectiveness of the activity and cognition module content in addressing these multidimensional fatigue perpetuating factors. For example, participants in CBT-TTF experienced significant improvements in subjective reports of activity (e.g., mental, physical, social) compared to WLC, with a particularly large effect size for social activity (d = 1.14). The SIP-SI measures the impact of symptoms (i.e., fatigue) on social interaction. The mean score for participants in the WLC condition remained nearly the same, while the mean score reduced by nearly 80% for participants in CBT-TTF. It is reinforcing that the activity module assisted patients in regulating an activity beyond the physical domain, as the disability caused by fatigue also impacts social and cognitive functioning and wellbeing.

We observed large (d’s > 1) effect sizes for change in elements of cognition in CBT-TTF compared to WLC. This is evident when comparing mean within-group differences at baseline and follow-up, as scores remained relatively stable over 18 weeks in the WLC group, while mean scores improved by at least one standard deviation for each measure of cognition in the CBT-TTF group. Marked treatment group differences in change in fatigue perpetuating cognitions emphasize the integral role that cognitive processes play in maintaining fatigue and suggest the cognition module is a critical element of CBT-TTF. Previous studies have shown that patients’ views of fatigue as negative and tough to change, excessive focus on the experience of fatigue, and dysfunctional beliefs about the relationship between fatigue and activity play a major role in perpetuating fatigue [10, 31], and changes in fatigue- and illness-related cognitions have been shown to produce lasting change in fatigue severity [4]. Therefore, our finding that maladaptive cognitions related to fatigue and cancer significantly improved in CBT-TTF compared to WLC are particularly promising and clinically relevant.

Consistent with previous literature and the well-established efficacy of cognitive behavioral therapy for sleep disorders in cancer patients [30, 32], we expected CBT-TTF to improve disturbed sleep/wake rhythms. While CBT-TTF participants endorsed greater reductions in sleep/wake disturbance than WLC participants, change over time between groups did not achieve statistical significance (p = .134). However, the effect size for the magnitude of change between groups was moderate in size (d = .51). This may reflect the wide confidence intervals for this measure, and suggests that the analysis may have been underpowered to detect some differences. Exploratory mediation analyses demonstrated evidence of sleep as a mediator, suggesting that intervention effects on sleep/wake disturbance may account for group differences in fatigue. This finding warrants further investigation in larger studies. Other measures of sleep quality and utilization of objective measurement tools (e.g., actigraphy) to evaluate sleep and activity patterns over time should be considered for future clinical trials.

The lack of group differences in measures of sleep, processing, social support, and fear of progression may have been partly due to the fact that modules targeting these factors were not indicated or received by most members of the CBT-TTF group. Preliminary evidence from exploratory analyses showed that sleep and processing (avoidance subscale) demonstrated significant indirect effects in simple mediation analyses, supporting further evaluation of these modules as components of CBT-TTF. A larger randomized trial that is powered to evaluate the effects of tailoring (i.e., ability to compare CBT-TTF participants who received a module to those who did not) is warranted.

Mediation analyses revealed that indices of activity (e.g., regulation of mental and social activity) and cognition (e.g., reductions in excessive focus on fatigue, decreased helplessness to influence fatigue, bolstered self-efficacy to manage fatigue) mediate the relationship between treatment group and post-intervention fatigue. Sleep and processing (avoidance) also demonstrated evidence of mediation. Mediation results are consistent with previous findings identifying improvements in perceived activity, sense of control over fatigue, and decreased focusing on symptoms as critical process variables in CBTF [10, 30]. Our finding that physical activity was not a significant mediator is consistent with previous literature suggesting that patients’ perceptions of physical activity, but not level of physical activity, mediate the effect of CBT on fatigue outcomes [10, 15, 30, 33]. This suggests that behavioral increases in physical activity are important to the extent that they facilitate changes in beliefs about activity and fatigue [6, 10]. Consistent with this, our findings suggest that cognitive factors (e.g., self-efficacy, helplessness, focusing on symptoms) play a central role in mediating the effect of the intervention on patients’ reports of fatigue. Changing maladaptive thought patterns may alter patients’ perceptions and subjective experience of fatigue, thereby altering their reports of the experience of fatigue.

Notable strengths of the current analysis include mechanistic insights into an intervention targeting fatigue, a prevalent but difficult-to-treat symptom in people with cancer, in a novel patient population, CML patients taking TKIs. Mediation analyses are largely understudied in the psychosocial oncology literature [34], but are vitally important to help to identify the “active” elements of an intervention and can help us to better understand how treatments work to improve outcomes. In turn, this can inform the refinement of existing interventions and development of future interventions to promote optimal treatment outcomes. Several limitations of this study should also be noted. The current study is a secondary analysis of data from a single-center pilot randomized trial and was not specifically powered for the current analyses. Due to the small sample size and pilot nature of the trial, effect sizes should be interpreted with caution. Mediation analyses should be considered exploratory and did not involve correction for multiple comparisons. Due to the fact that both the mediator and fatigue were measured at the same time, one cannot be certain about the direction of the relationship between them. An additional limitation is a relatively small, homogenous sample of non-Hispanic White participants. Future, larger studies with more racially and ethnically diverse samples of CML patients are warranted to test the generalizability of current exploratory findings. Future randomized trials should be powered to fully examine the effects of intervention engagement and different combinations of modules on fatigue perpetuating factors and fatigue outcomes.

In summary, study findings support change in dysregulated activity patterns and maladaptive cognitions about fatigue and cancer as critical aspects of intervention for targeted therapy-related fatigue in CML patients taking TKIs. Sleep and avoidance of processing the experience of cancer and its treatment may also be important intervention components; further investigation is warranted. These findings can be used to help oncology providers counsel patients about how to effectively manage TKI-related fatigue. Findings can also form the basis for the design of interventions to address fatigue in other cancer populations treated with targeted therapies.

Supplementary Material

kaab035_suppl_Supplementary_Figure
kaab035_suppl_Supplementary_Table

Acknowledgements

This study was funded by the National Cancer Institute (R21 CA191594, R25 CA191314, and P30 CA076292); the views expressed are those of the authors and do not necessarily represent those of the National Cancer Institute. This work was also supported in part by the Population Research, Interventions, and Measurement Core Facility at the H. Lee Moffitt Cancer Center and Research Institute, a National Cancer Institute – designated comprehensive cancer center.

Compliance With Ethical Standards

Authors’ Statement of Conflict of Interest and Adherence to Ethical Standards H.S.L.J. reports consulting for RedHill BioPharma and Janssen Scientific Affairs. J.I.-P. reports consulting for Novartis, Takeda, Pfizer, and Bristol-Myers Squibb. K.S. reports involvement in speakers bureaus for Jazz Pharma, Celgene, and Novartis; involvement in advisory boards for Novartis, Agios, Astellas, Bristol-Myers Squibb, and AbbVie; consulting for Pfizer; and grants from Incyte outside the submitted work. The views expressed are those of the authors and do not necessarily represent those of the National Cancer Institute. The other authors made no disclosures.

Authors’ Contributions

K.A.H.: methodology, data curation, formal analysis, investigation, writing–original draft, and writing–review and editing. A.M.N.: methodology, data curation, investigation, and writing–review and editing. S.L.E.: formal analysis, writing–original draft, writing–review and editing. A.I.H.: data curation, formal analysis, and writing–review and editing. J.I-.P.: resources and writing–review and editing. K.S.: resources and writing–review and editing. P.B.J.: conceptualization, funding acquisition, investigation, methodology, project administration, supervision, and writing–review and editing. H.K.: conceptualization, investigation, methodology, supervision, and writing–review and editing. H.S.L.J.: investigation, project administration, supervision, writing–original draft, and writing–review and editing.

Ethical Approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed Consent Informed consent was obtained from all individual participants included in the study.

References

  • 1. Efficace F, Baccarani M, Breccia M, et al. Chronic fatigue is the most important factor limiting health-related quality of life of chronic myeloid leukemia patients treated with imatinib. Leukemia. 2013;27:1511–1519. [DOI] [PubMed] [Google Scholar]
  • 2. Phillips KM, Pinilla-Ibarz J, Sotomayor E, et al. Quality of life outcomes in patients with chronic myeloid leukemia treated with tyrosine kinase inhibitors: a controlled comparison. Support Care Cancer. 2013;21:1097–1103. [DOI] [PubMed] [Google Scholar]
  • 3. Gielissen MF, Verhagen S, Witjes F, Bleijenberg G. Effects of cognitive behavior therapy in severely fatigued disease-free cancer patients compared with patients waiting for cognitive behavior therapy: a randomized controlled trial. J Clin Oncol. 2006;24:4882–4887. [DOI] [PubMed] [Google Scholar]
  • 4. Gielissen MF, Verhagen CA, Bleijenberg G. Cognitive behaviour therapy for fatigued cancer survivors: long-term follow-up. Br J Cancer. 2007;97:612–618. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Abrahams HJG, Gielissen MFM, Donders RRT, et al. The efficacy of Internet-based cognitive behavioral therapy for severely fatigued survivors of breast cancer compared with care as usual: a randomized controlled trial. Cancer. 2017;123:3825–3834. [DOI] [PubMed] [Google Scholar]
  • 6. Poort H, Peters MEWJ, van der Graaf WTA, et al. Cognitive behavioral therapy or graded exercise therapy compared with usual care for severe fatigue in patients with advanced cancer during treatment: a randomized controlled trial. Ann Oncol. 2020;31:115–122. [DOI] [PubMed] [Google Scholar]
  • 7. Donovan KA, Small BJ, Andrykowski MA, Munster P, Jacobsen PB. Utility of a cognitive-behavioral model to predict fatigue following breast cancer treatment. Health Psychol. 2007;26:464–472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Poort H, Meade CD, Knoop H, Gielissen MFM, Pinilla-Ibarz J, Jacobsen PB. Adapting an evidence-based intervention to address targeted therapy-related fatigue in chronic myeloid leukemia patients. Cancer Nurs. 2018;41:E28–E37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Jim HSL, Hyland KA, Nelson AM, et al. Internet-assisted cognitive behavioral intervention for targeted therapy-related fatigue in chronic myeloid leukemia: results from a pilot randomized trial. Cancer. 2020;126:174–180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Knoop H, Prins JB, Moss-Morris R, Bleijenberg G. The central role of cognitive processes in the perpetuation of chronic fatigue syndrome. J Psychosom Res. 2010;68:489–494. [DOI] [PubMed] [Google Scholar]
  • 11. Knoop H, Kessel KV, Moss-Morris R. Which cognitions and behaviours mediate the positive effect of cognitive behavioural therapy on fatigue in patients with multiple sclerosis? Psychol. Med. 2012; 42(1): 205– 213. doi: 10.1017/S0033291711000924 [DOI] [PubMed] [Google Scholar]
  • 12. Menting J, Tack CJ, Bleijenberg G, et al. Is fatigue a disease-specific or generic symptom in chronic medical conditions? Health Psychol. 2018;37:530–543. [DOI] [PubMed] [Google Scholar]
  • 13. Menting J, Tack CJ, Donders R, Knoop H. Potential mechanisms involved in the effect of cognitive behavioral therapy on fatigue severity in Type 1 diabetes. J Consult Clin Psychol. 2018;86:330–340. [DOI] [PubMed] [Google Scholar]
  • 14. Kessels E, Husson O, van der Feltz-Cornelis CM. The effect of exercise on cancer-related fatigue in cancer survivors: a systematic review and meta-analysis. Neuropsychiatr Dis Treat. 2018;14:479–494. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Gielissen MF, Wiborg JF, Verhagen CA, Knoop H, Bleijenberg G. Examining the role of physical activity in reducing postcancer fatigue. Support Care Cancer. 2012;20:1441–1447. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Bergner M, Bobbitt RA, Pollard WE, Martin DP, Gilson BS. The sickness impact profile: validation of a health status measure. Med Care. 1976;14:57–67. [DOI] [PubMed] [Google Scholar]
  • 17. Gilson BS, Gilson JS, Bergner M, et al. The sickness impact profile. Development of an outcome measure of health care. Am J Public Health. 1975;65:1304–1310. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Worm-Smeitink M, Gielissen M, Bloot L, et al. The assessment of fatigue: Psychometric qualities and norms for the Checklist individual strength. J Psychosom Res. 2017;98:40–46. [DOI] [PubMed] [Google Scholar]
  • 19. Booth M. Assessment of physical activity: an international perspective. Res Q Exerc Sport. 2000;71Suppl 2:114–120, [DOI] [PubMed] [Google Scholar]
  • 20. Evers AW, Kraaimaat FW, van Lankveld W, Jongen PJ, Jacobs JW, Bijlsma JW. Beyond unfavorable thinking: the illness cognition questionnaire for chronic diseases. J Consult Clin Psychol. 2001;69:1026–1036. [PubMed] [Google Scholar]
  • 21. Prins JB, Bleijenberg G, Bazelmans E, et al. Cognitive behaviour therapy for chronic fatigue syndrome: a multicentre randomised controlled trial. Lancet. 2001;357:841–847. [DOI] [PubMed] [Google Scholar]
  • 22. Jacobsen PB, Andrykowski MA, Thors CL. Relationship of catastrophizing to fatigue among women receiving treatment for breast cancer. J Consult Clin Psychol. 2004;72:355–361. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Ray C, Weir W, Stewart D, Miller P, Hyde G. Ways of coping with chronic fatigue syndrome: development of an illness management questionnaire. Soc Sci Med. 1993;37:385–391. [DOI] [PubMed] [Google Scholar]
  • 24. Horowitz M, Wilner N, Alvarez W. Impact of Event Scale: a measure of subjective stress. Psychosom Med. 1979;41: 209–218. [DOI] [PubMed] [Google Scholar]
  • 25. Bridges KR, Sanderman R, van Sonderen E. An English language version of the social support list: preliminary reliability. Psychol Rep. 2002;90:1055–1058. [DOI] [PubMed] [Google Scholar]
  • 26. Custers JA, van den Berg SW, van Laarhoven HW, Bleiker EM, Gielissen MF, Prins JB. The Cancer Worry Scale: detecting fear of recurrence in breast cancer survivors. Cancer Nurs. 2014;37:E44–E50. [DOI] [PubMed] [Google Scholar]
  • 27. Yellen SB, Cella DF, Webster K, Blendowski C, Kaplan E. Measuring fatigue and other anemia-related symptoms with the Functional Assessment of Cancer Therapy (FACT) measurement system. J Pain Symptom Manage. 1997;13:63–74. [DOI] [PubMed] [Google Scholar]
  • 28. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Abingdon, UK: Routledge; 2013. doi: 10.4324/9780203771587 [DOI] [Google Scholar]
  • 29. Hayes A. Introduction to Mediation, Moderation, and Conditional Process Analysis: Second Edition: A Regression-Based Approach. New York, NY: Guilford Press; 2013. [Google Scholar]
  • 30. Heins MJ, Knoop H, Burk WJ, Bleijenberg G. The process of cognitive behaviour therapy for chronic fatigue syndrome: which changes in perpetuating cognitions and behaviour are related to a reduction in fatigue? J Psychosom Res. 2013;75:235–241. [DOI] [PubMed] [Google Scholar]
  • 31. Schellekens MPJ, Wolvers MDJ, Schroevers MJ, Bootsma TI, Cramer AOJ, van der Lee ML. Exploring the interconnectedness of fatigue, depression, anxiety and potential risk and protective factors in cancer patients: a network approach. J Behav Med. 2020; 43(4): 553– 563. doi: 10.1007/s10865-019-00084-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Garland SN, Johnson JA, Savard J, et al. Sleeping well with cancer: a systematic review of cognitive behavioral therapy for insomnia in cancer patients. Neuropsychiatr Dis Treat. 2014;10:1113–1124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Wiborg JF, Knoop H, Stulemeijer M, Prins JB, Bleijenberg G. How does cognitive behaviour therapy reduce fatigue in patients with chronic fatigue syndrome? The role of physical activity. Psychol Med. 2010;40:1281–1287. [DOI] [PubMed] [Google Scholar]
  • 34. Stanton AL, Luecken LJ, MacKinnon DP, Thompson EH. Mechanisms in psychosocial interventions for adults living with cancer: opportunity for integration of theory, research, and practice. J Consult Clin Psychol. 2013;81:318–335. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

kaab035_suppl_Supplementary_Figure
kaab035_suppl_Supplementary_Table

Articles from Annals of Behavioral Medicine: A Publication of the Society of Behavioral Medicine are provided here courtesy of Oxford University Press

RESOURCES