A Pilot Mind-Body Resiliency Intervention Targeting Fear of Recurrence among Cancer Survivors

Abstract

OBJECTIVE

Interventions for fear of recurrence (FOR) of cancer have nominal effects, perhaps due to limited integration of empirically supported skills. This pilot trial tested the acceptability and feasibility of a multimodal, mind-body resiliency intervention targeting FOR among survivors of various cancers.

METHODS

Early stage cancer survivors 3–30 months post-treatment were recruited to participate in an eight-session in-person mind-body resiliency group intervention that taught relaxation skills, cognitive-behavioral techniques, healthy lifestyle behaviors, mindfulness meditation, and positive psychology skills all targeted for FOR. Primary outcomes were feasibility (enrollment rate, session attendance, survey completion, skills practice) and acceptability (enjoyableness, convenience, helpfulness, relevance). Patient-reported outcomes (FOR, uncertainty intolerance, cancer-related uncertainty, perceived stress, resiliency, positive affect, and coping skills) were collected at baseline, post-intervention, +1 month, and +3 months. Exit interviews assessed survivors’ reported benefits.

RESULTS

Participants (N=4 groups, 23 survivors, enrollment response rate=58%) included survivors of seven common cancer types who were on average 12 months post-treatment. Attendance was high (M=6.1 sessions), and 96% of survivors completed all surveys. Sustained increases in relaxation skills practice 3+ days/week were reported (baseline=16%, post-intervention=76%, +3 months=71%). Most sessions (87%) were rated as highly or very highly acceptable. Moderate-to-large (d=0.87) improvements in FOR severity were observed post-intervention (p<.01) and across assessments (p<.01), with similar changes observed in other patient-reported outcomes. Exit interviews revealed behavioral, cognitive, emotional, and existential benefits.

CONCLUSIONS

The targeted mind-body resiliency intervention shows promising acceptability, feasibility, and favorable changes in FOR and coping skills practice. Further adaptation and testing in a randomized trial are warranted.

Clinical Trials.gov Registration Number

NCT03695406

Background

After active treatment ends, cancer patients enter a stage of surveillance characterized by uncertainty and self-monitoring, often faced with managing lasting somatic concerns and emotional challenges on their own. Chief among these is fear of recurrence (FOR), with as many as two-thirds of cancer survivors living with elevated FOR, a staggering figure considering the growing prevalence of cancer survivors worldwide^1–3. If untreated, FOR levels can persist for years after treatment ends^4,5. Observational studies have demonstrated that FOR severity is independent of cancer type and accounts for up to 44% of cancer survivors’ distress levels^6–9. Not only can FOR result in emotional distress, it may affect how survivors engage in their healthcare, potentially resulting in missed diagnoses, unnecessary or ineffective treatments, and ultimately higher rates of morbidity and mortality, and greater costs^2,10,11.

A recent meta-synthesis of 87 qualitative studies of FOR concluded that “FOR was found to be an intense, difficult, multi-dimensional experience”¹². Distinct from traditional psychiatric anxiety disorders¹³, FOR can be a rational, albeit complex and exaggerated, response after cancer treatments end. Theoretical models indicate that FOR arises and persists due to mind-body interactions between somatic symptoms, psychological factors, and behavioral responses¹⁰. Somatic symptoms that trigger FOR commonly include pain, fatigue, sleep disturbance, numbness, and may be due to illness, treatments, or simply aging¹⁴. Psychological factors include patients’ interpretations of somatic symptoms, tolerance of uncertainty, and optimism about their future health¹⁵. Behavioral responses involve patients’ attempts to assert control over the unpredictability of their health, such as excessive self-examination or unscheduled doctor visits, or behaviors to avoid thinking about uncertainty, such as drinking alcohol or delaying follow-up medical appointments^10,16. If left untreated, these mind-body interactions can form a feedback loop that may account for FOR-related mental and physical health impairments^9,17.

Calls for interventions targeting FOR have emphasized the need for evidence-based treatments^1–3, and multimodal interventions (i.e., teaching a variety of targeted skills) may be particularly suited to address the multidimensional nature of FOR. Critically, survivors with higher FOR are more likely to seek training in mind-body practices¹⁶, suggesting the acceptability of these techniques. Mind-body interventions, defined as techniques designed to enhance the mind’s capacity to affect bodily function and symptoms¹⁸, have only recently begun to be studied to address FOR, a concept, however, that has been studied for over three decades¹⁹. A recent systematic review and meta-analysis of RCTs testing mind-body interventions for FOR in cancer patients found that pooled effects were significant, yet small, suggesting the need for further intervention development²⁰. Notably, no trials have evaluated multimodal mind-body interventions that integrate the most commonly studied FOR management skills: cognitive-behavioral techniques, meditation, relaxation, and positive psychology. Mind-body interventions may thus represent an acceptable set of treatments for FOR, with some emerging evidence of their benefits on reducing FOR, although further research is needed to develop targeted multimodal interventions.

The present pilot study was designed to test a multimodal mind-body intervention for reducing FOR in cancer modified from an evidence-based intervention (Relaxation Response Resiliency Program; 3RP²¹) that is multimodal, delivered in groups, and has been targeted to address unique concerns of other populations. Our primary aim was to examine the feasibility and acceptability of the targeted FOR intervention in a proof-of-concept trial. Our exploratory aim was to summarize scores on key FOR and coping measures across baseline (T₀), post-intervention (T₁), +1 month (T₂), and +3 months (T₃). Through exit interviews, we used qualitative methods to evaluate perceived changes from participation.

Methods

Procedure

This pilot study consisted of a single arm intervention registered on clinicaltrials.gov (NCT03695406). Participants were recruited from Beth Israel Deaconess Medical Center in Boston, MA. All procedures were approved and monitored by the institutional review board.

As depicted in Figure 1, participant flow was tracked using a CONSORT diagram. Inclusion criteria were as follows: (a) history of breast, lung, colorectal, hematologic, prostate, melanoma, or gynecological cancer, (b) completion of primary cancer treatment (i.e., radiation, surgery, and/or chemotherapy) between 3 to 30 months prior to enrollment, (c) 18 years of age or older. Exclusion criteria were as follows: (a) self-reported inability to speak and write in English, (b) concurrent participation in weekly, group-based psychosocial or mind-body programs, (c) serious mental illness as defined by history of suicidality, psychosis, and/or psychiatric hospitalization in the past year, (d) inability to travel to necessary study visits, and (e) no e-mail address to access online assessments. Medical history and age were validated using participants’ electronic health record data.

Figure 1:

CONSORT Flow Diagram

Participants were recruited in two, three-month waves (two groups per wave) from July through September 2017 and January through March 2019 using flyers, clinician referral, and word of mouth. After completing phone screening, potentially eligible survivors were invited to meet with study staff in-person to review and provide informed consent. Informed written consent was obtained at the site of the group sessions, which served as a run-in to orient participants to the location. As a token of appreciation, participants received $25 for completing each of the four assessments. Additionally, participants were reimbursed for costs associated with parking or public transportation from/to study visits up to $5 per visit.

Intervention

Prior to the start of a group, participants completed an intake meeting with the group leader as a run-in to discuss the group format, content, and attendance. Each cohort was offered eight, 90-minute sessions delivered weekly. The original core 3RP protocol is multimodal to target the multiple core components of the 3RP theoretical model and flexible so that it may be modified for various medical populations²¹. Sessions all involve a didactic portion as well as opportunities to learn and practice skills. Participants are encouraged to discuss these skills in session, allowing for personalization and troubleshooting barriers to adherence (e.g., scheduling relaxation practice in the morning vs. evening). Across sessions, skills in three core areas are emphasized: relaxation response elicitation, stress awareness, and adaptive strategies. This multimodal, mind body approach emphasizes the interplay between physiological responses to stress (and its counterpoint, the relaxation response), cognitive-behavioral factors (e.g., patterns of distorted and/or negative thinking), social support, health behaviors (e.g., sleep hygiene, gentle chair yoga), and resiliency skills grounded in positive psychology (e.g., logging moments of gratitude, using humor). Each session ends with participants setting individualized goals for applying skills independently.

Critical modifications to the core 3RP content to specifically target FOR included: (a) Targeted relaxation response elicitation (e.g., diaphragmatic breathing for reducing FOR-related distress, mindfulness practice of Leaves on a River to enhance cognitive diffusion of cancer worry); (b) Targeted stress awareness (e.g., incorporating information about cancer-related uncertainty and common triggers of FOR, including somatic concerns/changes, reminders about cancer, and follow-up appointments); and (c) Targeted adaptive strategies (e.g., worry time for reducing interference from FOR, identifying controllable and uncontrollable aspects to concerns about cancer recurrence, using humor skillfully when discussing an uncertain future). We also modified content of the health behavior modules to reflect National Comprehensive Cancer Network survivor health behavior guidelines for diet, physical activity, and sleep. Additional details of the intervention adaptation are elsewhere (Hall et al., in preparation).

The intervention was delivered by a licensed clinical health psychologist with post-doctoral training in behavioral medicine, psycho-oncology, and integrative medicine research. The interventionist had prior experience delivering the 3RP and other group-based mind-body interventions in clinical trials with medical patients. To promote fidelity to the treatment manual, the interventionist participated in biweekly peer group supervision with other clinicians delivering 3RP interventions.

Data collection procedures

Patient-reported outcomes were collected using multiple methods. For demographic and psychosocial data, we used the HIPAA-approved electronic data entry and management system called Research Electronic Data Capture (REDCap). Participants were asked to complete REDCap surveys at four timepoints: baseline (T₀), post-intervention (T₁), 1-month post-intervention (T₂), and 3-months post-intervention (T₃). The T₀ and T₁ surveys were offered via electronic tablet in-person. The T₂ and T₃ assessments were e-mailed to participants with a link to the REDCap survey. If participants did not complete the T₂ and T₃ assessments within one week of the initial e-mail, a reminder message was sent. In addition to these surveys, participants were asked to complete a 5-item acceptability questionnaire following each session using pen-and-paper. Lastly, individual exit interviews were conducted with all study participants by a trained qualitative interviewer who was not involved in the design or implementation of the intervention. Interviews were audio recorded and transcribed.

Measures

Sociodemographic and medical characteristics

At T₀, participants self-reported their age, sex, race, ethnicity, education, employment status, income, insurance status, marital status, and number of children. Medical characteristics were self-reported and verified by medical record, including previous cancer history, date of cancer diagnosis, type(s) of treatment, and date of primary treatment conclusion (to determine time since treatment).

Primary outcomes

Feasibility was measured by summarizing the ratio of participants eligible who enrolled, rates of retention at the initial and second follow-up assessments, attendance to sessions, rates of relaxation skills practice, and rates of survey completion within and across assessments. Reasons for ineligibility, refusal, or dropping out were recorded. Reasons for missed sessions were recorded in a post-session attendance log. Acceptability was measured using an in-session rating form. Items asked participants to rate (1=not at all to 5=very high) the enjoyableness, convenience, helpfulness, odds of future use, and overall satisfaction with the session.

Exploratory outcomes

Self-reported psychosocial measures were selected based on the 3RP theoretical model²¹ and our review of extant FOR clinical trials²⁰. The following measures were assessed at each timepoint (T₀-T₃).

FOR and Stress Awareness

FOR-related severity, triggers, functional impairment, and distress

These constructs were measured using subscales of the Fear of Cancer Recurrence Inventory²² (Cronbach’s αs=.86–.90). A FOR severity score ≥16 indicates clinically elevated FOR²².

Intolerance of uncertainty

The Intolerance of Uncertainty Scale-Short Form²³ was used to measure one’s perceived ability to tolerate uncertainty. Higher scores indicate greater intolerance of uncertainty (Cronbach’s α=.90).

Cancer-related uncertainty

Cancer-related uncertainty was assessed using the Survivor version of the Mishel Uncertainty in Illness Scale²⁴, with higher scores indicating greater levels of cancer-related uncertainty during survivorship. Sample items include: “I’m certain they will not find anything else wrong with me” and “The long-term effect of my treatment is clear to me.” (Cronbach’s α=.74).

Perceived stress

The Perceived Stress Scale (PSS-4)²⁵ is a measure of subjective stress over the past month (Cronbach’s α=.78).

Tension awareness

The ability to notice signs of tension in oneself was measured using the Measure of Current Status (MOCS-A)²⁶ tension awareness subscale (Cronbach’s α=.65).

Relaxation Response Elicitation and Adaptive Strategies

FOR-related coping

Use of coping skills in response to FOR was measured using the coping subscale of the Fear of Cancer Recurrence Inventory²² (Cronbach’s α=.89).

Coping using relaxation

The ability to elicit the relaxation response was measured using the MOCS-A²⁶ two-item relaxation subscale.

Coping confidence

Confidence in one’s ability to cope was measured using the using the MOCS-A²⁶ coping confidence subscale (Cronbach’s α=.89).

Positive affect

Positive affect was assessed using the subscale of the Positive and Negative Affect Schedule²⁷ (Cronbach’s α=.91).

Resiliency

Resiliency was assessed using an investigator-developed measure adapted from the Post-Traumatic Growth Inventory²⁸ (Cronbach’s α=.93).

Self-efficacy

The perceived ability to effectively manage stressful situations was measured using the General Self-Efficacy Scale²⁹ (Cronbach’s α=.94).

Tolerance of physical discomfort

This 1-item measure assessed one’s subjective ability to tolerate physical discomfort on a 0 (very low) to 10 (very high) rating scale.

Exit interviews

Participants’ perceived changes after participation were assessed through exit interviews. Individual phone interviews lasting approximately 45 minutes employed a semi-structured interview guide with open-ended questions and response probes. Survivors were asked to reflect on their coping with FOR-related triggers and consequences and to identify aspect(s) of the program that contributed to perceived changes. Sample interview questions include: “Did you notice any changes or improvement resulting from the group?” and “One aspect of the sessions involved learning skills to manage uncertainties and fears about potential cancer recurrence. Did this program influence how you manage these uncertainties and fears?”

Data analyses

Primary outcomes

Quantitative analyses were conducted with SPSS software version 25. Feasibility and acceptability measures utilizing quantitative data were summarized using descriptive statistics (means, medians, frequencies, observed ranges, and scatterplots of raw scores). Reasons for missed sessions were summarized for participants with low attendance (i.e., 2/8 sessions attended). A priori feasibility benchmarks included: at least 50% of eligible participants enrolled; participants completed at least 75% (6/8) of the treatment sessions; participants adhered to recommended relaxation practice (3+ days/week); and at least 70% of participants were retained at each follow-up assessment. Acceptability was analyzed by computing frequencies of post-session survey scores. An average response of “high satisfaction” indicated acceptability of intervention content and procedures.

Exploratory outcomes

Cronbach’s alphas were calculated for all exploratory outcomes. Means and standard deviations were computed at each timepoint. Longitudinal changes in exploratory outcomes were analyzed via one-way repeated measures analysis of variance (RANOVA) with four time points: baseline (T₀), post-intervention (T₁), 1-month post-intervention (T₂), and 3-months post-intervention (T₃). Planned contrasts of pre-post intervention changes (T₀ to T₁) were used to calculate Cohen’s d for within-subjects effect sizes of changes. In post-hoc analyses, we sought to identify sociodemographic and medical factors that differentiated responders from non-responders (i.e., FOR severity scores did not decrease from T₀ to T₁).

Exit interviews

Transcriptions of exit interviews were analyzed using thematic content analysis to allow for the emergence of themes within a frame from pre-existing literature³⁰ identifying multidimensional aspects of FOR. Thus, a hybrid approach was used with deductive (i.e., guided by FOR literature) and inductive (i.e., allowing emergent themes) methods. Themes were generated independently by two primary coders (DH and TC). A third coder (GY) provided guidance on refining the extracted themes. Transcripts were stratified by cohort, then randomly selected for analysis in iterative batches of four interviews (one per cohort). After reviewing each batch, primary coders met to compare extracted themes and refine codes. Discrepancies were flagged and resolved by consensus through discussion with the third coder, consideration of common FOR correlates (e.g., coding reassurance-seeking behaviors), and review of transcripts. Using an audit trail of each coding step, this process was repeated until reaching thematic saturation by consensus.

Results

Enrollment and participant characteristics

Of the 65 patients screened, the primary reason for ineligibility was having been treated for cancer longer than 30 months prior to screening (n=19). Seventeen patients who were eligible declined to enroll, chiefly because they were too busy (n=8) or not interested (n=5). As summarized in Table 1, the final sample consisted of 23 cancer survivors who were predominantly middle-aged, college-educated women (Mean age=61 years old) who were on average one year post-treatment. Most participants had been treated with surgery and chemotherapy for a variety of cancer types, most commonly colorectal, breast, and ovarian/uterine cancers. A minority of participants (13%) identified as non-Hispanic Black or African American or multiracial.

Table 1.

Participant characteristics

	N(%)/M(SD)
Age (years)	61.0 (13.6)
Time since treatment (months)	12.3 (7.3)
Gender (female)	20 (87%)
Race/Ethnicity
non-Hispanic White	20 (87%)
non-Hispanic Black or African American	2 (9%)
Multiracial/Other	1 (4%)
Marital Status
Married/partnered	9 (40%)
Divorced/separated/widowed	7 (30%)
Never married	7 (30%)
Education
College graduate	21 (91%)
Some college or technical school	2 (9%)
Employment
Employed (fulltime, part-time, or self)	10 (44%)
Retired	9 (40%)
Out of work/Unable to work	3 (12%)
Homemaker	1 (4%)
Cancer Types *
Breast	9 (39%)
Blood/Hematologic	1 (4%)
Colorectal	9 (39%)
Lung	2 (9%)
Melanoma	1 (4%)
Ovarian/Uterine	3 (13%)
Treatment History *
Surgery	22 (96%)
Chemotherapy	18 (78%)
Radiation	8 (35%)

Note. N=23. M=Mean, SD=Standard deviation. Age and time since treatment values are presented as mean years and months, respectively.

^*Participants could endorse more than one cancer type and treatment history.

Primary outcomes

All a priori feasibility benchmarks were met. As depicted in Figure 1, the majority of screened, eligible cancer survivors enrolled in the pilot trial (58%). The most common reason for ineligibility was being more than 30 months posttreatment. Most participants (74%) attended at least 75% of sessions offered (mean=6.13, median=7, mode=8 sessions). Three participants had low attendance (2/8 sessions), and post-session logs noted that reasons for their missed sessions were high fatigue, having a newly prescribed medication that could interfere with driving to group sessions (i.e., gabapentin), and an unmet expectation of receiving advanced meditation training. Rates of survey completion were excellent, with 96% of survivors completing all four surveys from T₀ to T₃. All participants (100%) completed exit interviews. Regarding adherence to skills taught in the intervention, at baseline, 16% of the sample practiced relaxation three or more days per week. This rate increased to 76% post-intervention, 76% by the 1-month follow-up, and tapered to 71% by the 3-month follow-up.

Sessions were overall rated as highly acceptable. Most (87%) of session ratings were scored as 4 (“High”) or 5 (“Very High”) on items of enjoyment, convenience, helpfulness, odds of future use, and overall satisfaction. The mean and range of scores were similar across sessions: Session 1 (M=4.3, range=2–5), Session 2 (M=4.3, range=2–5), Session 3 (M=4.4, range=1–5), Session 4 (M=4.6, range=2–5), Session 5 (M=4.6, range=2–5), Session 6 (M=4.5, range=3–5), Session 7 (M=4.5, range=2–5), and Session 8 (M=4.8, range=3–5).

Exploratory outcomes

Longitudinal trends in (a) FOR and stress awareness, and (b) relaxation response elicitation and adaptive strategies from baseline (T₀) to post-intervention (T₁), 1-month post-intervention (T₂), and 3-months post-intervention (T₃) are presented in Table 2.

Table 2.

Longitudinal patterns of changes in key patient-reported outcomes

FOR and STRESS AWARENESS	T0 M(SD)	T1 M(SD)	T2 M(SD)	T3 M(SD)	F(3,63)	p	d
FOR severity	19.4 (7.0)	15.0 (5.7)	15.9 (5.9)	15.0 (6.3)	8.3	<0.01	0.87
FOR triggers	17.6 (6.1)	14.7 (5.1)	13.8 (5.5)	14.7 (7.1)	4.2	<0.01	0.51
FOR functional impairment	6.9 (5.1)	3.8 (4.4)	3.3 (4.3)	3.0 (4.0)	10.1	<0.01	0.64
FOR-related distress	6.5 (4.3)	3.6 (3.5)	3.7 (2.8)	3.0 (3.5)	10.8	<0.01	0.69
Intolerance of uncertainty	34.2 (10.9)	29.5 (8.8)	30.8 (10.0)	29.8 (9.3)	10.8	<0.01	0.42
Cancer-related uncertainty	59.2 (9.5)	56.5 (9.7)	59.0 (9.1)	54.8 (8.8)	2.2	0.12	0.23
Perceived stress	6.6 (3.1)	5.0 (3.2)	5.3 (2.7)	4.5 (3.1)	4.4	<0.01	0.63
Tension awareness	6.4 (1.9)	7.3 (2.0)	7.3 (2.1)	7.8 (2.3)	2.7	0.07	0.38
RR ELICITATION and ADAPTIVE STRATEGIES
FOR-related coping	16.0 (8.9)	20.8 (6.0)	20.5 (6.5)	21.0 (8.2)	3.8	0.04	0.50
Coping using relaxation	2.0 (1.7)	4.1 (1.8)	3.9 (1.7)	4.2 (2.0)	20.8	<0.01	1.28
Coping confidence	8.7 (4.2)	11.3 (4.1)	11.8 (3.0)	12.1 (3.4)	10.0	<0.01	0.57
Resiliency	84.0 (19.5)	94.0 (19.9)	94.8 (16.5)	97.8 (16.8)	6.8	<0.01	0.53
Positive affect	33.6 (9.2)	35.8 (8.5)	36.7 (9.1)	37.6 (9.0)	2.5	0.07	0.23
Self-efficacy	29.0 (6.0)	31.5 (4.5)	30.5 (3.8)	32.0 (4.4)	3.9	0.03	0.45
Tolerance of physical discomfort	6.4 (2.8)	8.0 (2.0)	7.2 (2.4)	7.5 (2.6)	2.0	0.12	0.54

Note: N=22. T0=Baseline, T1=Post-intervention, T2=1-month post-intervention, T3=3-months post-intervention. M=Mean, SD=Standard deviation, d=Cohen’s d effect size change from T0 to T1. FOR = fear of recurrence, RR = relaxation response. Statistically significant changes (alpha=.05) are presented in bold.

From T₀ to T₃, there were statistically significant reductions in FOR-related severity (p<.01), triggers (p<.01), functional impairment (p<.01) and distress (p<.01). Notably, FOR severity scores at T₀ were clinically-elevated (i.e., ≥16; M=19.4, SD=7.0), yet no longer fell in the clinically-elevated range by T₁ (M=15.0, SD=5.7), T₂ (M=15.0, SD=5.7), and T₃ (M=15.0, SD=5.7). At T₀, 70% of participants had clinically-elevated FOR severity. Elevated FOR severity at baseline was associated with younger age (r=−.47, p=.02), but not other sociodemographic characteristics (gender, race, ethnicity, employment status, or years of formal education, all p>.05), cancer treatment type (all p>.48), or time since cancer treatment (p=.80). By T₁, 36% of the sample had clinically-elevated FOR, which was generally sustained by T₂ (45%) and T₃ (36%). There were also statistically significant decreases in intolerance of uncertainty (p<.01) and perceived stress (p<.01) across assessments. In contrast, cancer-related uncertainty (p=.12) and tension awareness (p=.07) tended to remain stable from T₀ to T₃.

In terms of relaxation response elicitation and adaptive strategies, from T₀ to T₃ there were statistically significant increases in FOR-related coping (p=.04), coping using relaxation (p<.01), and coping confidence (p<.01). Resiliency (p<.01) and self-efficacy (p=.03) also increased across assessments. However, positive affect (p=.07) and tolerance of physical discomfort (p=.12) did not significantly change over time.

Finally, six participants were identified as non-responders whose FOR severity scores did not decrease from T₀ to T₁. No sociodemographic or medical characteristics differentiated non-responders from responders (all p>.05). However, non-responders were more likely than responders to have subclinical FOR severity scores (≤15) at baseline (p=.02).

Exit interviews

Exit interviews asked participants to identify changes they attributed to participation in the intervention. As depicted in Figure 2, content analyses revealed themes across four domains: behavioral changes, cognitive changes, emotional changes, and existential changes. Behavioral changes included less reassurance-seeking, improved sleep hygiene, and less reactivity. Participants also credited the group structure of the intervention with providing peer reinforcement of behavior change. Cognitive changes included improved meta-cognition (i.e., awareness of thoughts), such as ability to identify negative automatic thoughts and increased awareness of a mind-body connection. Participants also identified gaining a more positive attitude. Following the intervention, participants described their attention as more present-focused. Several participants also described having less fear-based cognitions about their cancer history or experiences of uncertainty. Emotional changes included less distress, in particular less self-blame for developing cancer or the impacts of treatment on loved ones. Participants also described increased feelings of empathy, camaraderie with group members, and greater appreciation and gratitude for their cancer care providers. Finally, a number of existential changes were described, including a heightened perceived ability to separate one’s personal identity from their identity as a cancer survivor, new social role changes (e.g., new work and activities), and increased engagement with others. Participants described feeling more empowered (i.e., increased confidence, a shift to an internal locus of control, containment of stress and fear to an allotted time, and ownership of emotions) and greater self-efficacy for using skills to promote prosocial emotions (i.e., acceptance, appreciation, and gratitude) and to manage anxiety and fear. Finally, when asked which aspect(s) of the intervention accounted for these perceived changes, participants most often identified the multimodal nature of the program. Exemplar quotes regarding perceived benefits from exposure to a variety of mind-body skills for managing FOR are also provided under Figure 2.

Figure 2:

Identified themes and exemplary quotes from qualitative exit interviews

Discussion

This pilot study provides initial evidence for the feasibility and acceptability of a multimodal, mind-body intervention that targeted content to address fear of recurrence (FOR) among survivors of various cancers. Within-subjects changes suggest significant, clinically meaningful reductions in FOR severity (70% above clinical cut-off at baseline, 36% post-intervention and 3-month follow-up) and multiple distress, coping, and resiliency variables that were sustained at a three-month follow-up assessment. Exit interviews substantiated these findings by revealing multiple behavioral, cognitive, emotional, and existential benefits that were attributed to being exposed to multiple FOR management skills. These findings provide support for the use of a multimodal, mind-body approach for treating FOR and have several important implications for cancer survivorship and resiliency interventions.

To the best of our knowledge, the present study is the first to provide initial evidence that a group-delivered FOR intervention with multiple mind-body practices integrated into one cohesive intervention is feasible and acceptable for survivors of various cancers. Although most eligible survivors enrolled, completed our survey batteries, and were able to attend in-person, a subset of cancer survivors were ineligible due to limited mobility/ability to travel to sessions (n=4). Among those enrolled, three participants were only able to attend two out of eight (25%) sessions, chiefly due to medical and psychiatric comorbidities that precluded their ability to attend in-person. While most participants were able to attend all eight sessions, it is possible that other delivery modalities, such as telehealth, could make it more feasible for more survivors facing the greatest burdens to participate. We also demonstrate feasibility for survivors to learn relaxation techniques and to practice them independently; at baseline, 16% of the sample reported practicing relaxation three or more days per week, which increased to 76% after completing the program and was sustained three months later. Ratings of acceptability collected after each session demonstrated that participants enjoyed sessions, found them helpful and convenient, and believed they would apply the skills in the future. Ratings were also similar across each of the eight sessions. These findings reflect a trend among cancer survivors more broadly, many of whom are increasingly seeking integrative modalities that use holistic approaches with the primary purpose for reducing risk of recurrence, managing stress, and enhancing psychological support^31–33.

Surveys from baseline through 3-month follow-up revealed significant within-subjects improvements across multiple measures of FOR and resiliency (i.e., stress awareness, relaxation response elicitation, and adaptive strategies²¹). Recently, the efficacy of interventions for reducing FOR was examined in two meta-analyses, and existing programs were found to yield small-to-medium sized reductions in FOR^20,34. The most common mind-body practices taught in these programs were skills rooted in cognitive-behavioral therapy (e.g., cognitive restructuring, worry time) or in meditation instruction (e.g., mindfulness, yoga), and to a lesser extent relaxation training (e.g., diaphragmatic breathing) and positive psychology skills (e.g., gratitude journaling). Across trials, greater improvements tended to come from (a) multimodal interventions and (b) group delivery. In contrast to existing theory-driven FOR interventions, we based this targeted program on a resiliency framework that has proven efficacious in reducing distress and promoting psychological growth across a variety of medical populations²¹. Using this model as a basis for adaptations, we incorporated cognitive-behavioral techniques, mindfulness meditation, relaxation skills, positive psychology practices, and NCCN health behavior information³⁵ into an eight-session multimodal protocol that was designed to address the FOR as a “complex experience involving emotional, perceptual, conceptual, bodily and behavioral dimensions”¹².

Conceptually, it is noteworthy that levels of cancer-related uncertainty remained stable from our initial assessment through three-month follow-up. In this intervention, uncertainty was normalized through psychoeducation about triggers of cancer-related uncertainty (e.g., screening, new aches/pains)^14,36 and acceptance of an unpredictable future. In contrast, tolerance of uncertainty was framed as controllable using cognitive-behavioral and mindfulness skills³⁷. Although (unmanaged) cancer-related uncertainty may trigger anxiety, fear, and distress, prior studies have shown that coping skills for managing uncertainty can mitigate these concerns^38,39. While uncertainty did not change, intolerance of uncertainty decreased significantly, suggesting that participants may have learned to manage – rather than attempt to reduce or avoid – their experiences of uncertainty.

Although findings overall supported the feasibility and acceptability of the targeted intervention, our findings also suggest the need for adaptation in several areas. One component requiring adaptation is the delivery modality, as several barriers made it challenging for certain participants to attend. Potentially, a virtual program could offer greater access particularly for cancer survivors struggling with late effects of treatments (e.g., fatigue, insomnia, pain), especially since these survivors may also have the highest FOR^6,9. It would be prudent to learn about preferences for a virtually-delivered, multimodal mind-body program from cancer survivors experiencing elevated FOR. In addition, future adaptations might also assess participants’ expectations from the mind-body program. One participant only attended the first two sessions because she had an unmet expectation that she would be exposed to advanced meditation training. To mitigate this risk in the future, recruitment for mind-body intervention trials such as ours might assess expectations about exposure to mind-body practices during screening and offer referrals for meditation resources as appropriate. Cancer survivors with low FOR could also play a critical role in informing future adaptations, for instance by providing feedback on strategies they find most and least helpful for managing recurrence-related fears. Indeed, recent adaptations of FOR interventions using formative qualitative interviews highlight the benefits of soliciting and integrating feedback from cancer survivors when developing and refining FOR intervention materials and procedures^40,41.

Study Limitations

Limitations of this study included recruitment from a single institution, resulting in a sample largely comprised of white, college-educated women, which may limit generalizability to other groups. Future adaptations of this program would be strengthened by including feedback from more diverse survivors, including a broader range of racial and gender identities. Due to the pilot nature of this study, the small sample (total N=23 among four intervention cohorts) meant we were underpowered to explore potential mediators (e.g., stress management skills, tolerance of uncertainty) and moderators (e.g., age). The lack of a control group precludes our ability to make causal inferences when interpreting the significant within-subjects changes observed on patient-reported outcomes. In future trials, comparative effectiveness could also be tested by inclusion of a usual care study arm (e.g., referral to cancer support group). In this study, participants were not screened on FOR severity prior to enrollment. Several participants had low FOR upon enrollment, potentially restricting our ability to detect changes in FOR in this subsample. Future trials could exclude survivors with subclinical FOR or consider stratification by FOR severity. Although baseline FOR severity was not associated with time since treatment, we also acknowledge that survivors were on average one year since completing treatment, a time when many survivors can experience “scanxiety” leading up to follow-up oncology visits and tests. Future studies could assess the specific timing of these scans/visits in the context of timing the assessment for this program, particularly as these visits may also offer valuable opportunities for navigating uncertainty using mind-body skills. Finally, this study did not assess participants’ healthcare utilization. Given the relationships between FOR and healthcare engagement behaviors (i.e., seeking reassurance, avoidance), it would be prudent for future trials to ask cancer survivors about how, when, and with whom they seek medical care when they are feeling worried about cancer recurrence.

Clinical Implications

As cancer survivors grow in prevalence, more clinicians will have opportunities to treat patients affected by cancer. Screening for FOR in oncology and primary care clinics is feasible and acceptable for clinicians, support staff, and patients^3,42. Furthermore, clinicians treating cancer survivors are well positioned to provide appropriate referrals to address FOR, including empirically-supported psychological and mind-body practices^20,34. The present study is an initial step to demonstrating that an intervention incorporating many of these practices into a group-based program is helpful for cancer survivors.

Highlights.

• This pilot trial demonstrated the initial acceptability and feasibility of a theory-guided, multimodal, mind-body resiliency intervention targeting fear of recurrence (FOR) among survivors of various cancers.

• Longitudinal assessments of patient-reported outcomes revealed favorable improvements in FOR and key coping skills that persisted by a 3-month follow-up, especially among survivors with elevated FOR at study entry.

• Exit interviews substantiated changes in patient-reported outcomes and highlighted the benefits of being taught multiple resiliency skills in a group-based intervention.

• Considerations for further adaptation include screening on baseline FOR severity levels and the use of synchronous remote (i.e., virtual) delivery technologies.