Managing cancer and living meaningfully (CALM) in adults with malignant glioma: a proof-of-concept phase IIa trial

Abstract

Background

Managing Cancer and Living Meaningfully (CALM) is an evidence-based, brief, semi-structured psychotherapy designed to help patients with advanced cancer cope with the practical and profound challenges of their illness. However, no study to date has investigated its feasibility, acceptability, and preliminary effectiveness in adults with malignant glioma, despite the well-documented incidence of psychological distress in this vulnerable and underserved population.

Methods

Fourteen patients with glioma and elevated symptoms of depression and/or death anxiety enrolled in the trial: 83% glioblastoma, 75% female, M_age = 56 years (SD = 15.1; range = 27–81). Feasibility was assessed based on established metrics. Acceptability was measured by post-session surveys and post-intervention interviews. Preliminary intervention effects were explored using paired t-tests, comparing psychological distress at baseline and post-intervention.

Results

Of the 14 enrolled patients, 12 were evaluable. Nine completed the study (75% retention rate). Three patients withdrew due to substantial disease progression which affected their ability to participate. Participants reported high perceived benefit, and all recommended the program to others. Baseline to post-intervention assessments indicated reductions in death anxiety, generalized anxiety, and depression, and increases in spirituality. Quality of life and fear of cancer recurrence remained stable throughout the study period.

Conclusions

CALM appears feasible for use with adults with malignant glioma. Enrollment and retention rates were high and comparable to psychotherapy trials for patients with advanced cancer. High perceived benefit and reductions in symptoms of death anxiety, generalized anxiety, and depression were reported by participants. These findings are extremely encouraging and support further study of CALM in neuro-oncology.

Introduction

Patients with glioma manifest higher rates of depression and anxiety than both patients with other cancer diagnoses and in the general population [1–6] Death-related distress is common in patients with glioma [7, 8] distress manifests frequently as death anxiety, or the fear of, and preoccupation with, death and dying [9]. The high prevalence this distress in patients with neuro-oncological diseases is understandable, given the certainty of disease progression, decline in functioning, and absence of curative medical treatments [10] However, little is known about the utility of potential treatments to alleviate such distress in this population, given their underrepresentation and exclusion from psycho-oncology research [11].

Malignant glioma presents unique challenges to the development and evaluation of psychological treatments targeting disease-related symptoms of depression and death-related distress. Psychopharmacological approaches are of limited benefit and sometimes contraindicated for individuals with glioma. For example, antidepressant use is associated with lowered seizure thresholds, cognitive decline, and worsened fatigue [12]. Similarly, although cognitive behavioral therapy (CBT) is effective for a variety of psychological concerns, research suggests it is not effective for treating depression and related symptoms in advanced cancer [13–16]. This may be partly due to the rational fears associated with advanced oncological disease, and the difficulty managing cognitive demands of CBT. Therefore, there remains an urgent need to identify novel treatments for patients with glioma who report elevated levels of depression and death-related distress.

An alternative to CBT—Managing Cancer and Living Meaningfully (CALM)—offers promise for ameliorating depression and death-related distress in patients with malignant glioma. CALM is a brief semi-structured psychotherapeutic intervention developed to help patients with advanced cancer adapt to the challenges of their disease and its treatment and remain engaged in life while preparing for end of life [17–19]. Sessions address four broad, interrelated domains: (1) symptom management and communication with healthcare providers, (2) changes in personal relationships, (3) sense of meaning and purpose, and (4) the future, hope and mortality. The CALM process relies on an empathic therapeutic relationship that promotes self-understanding and allows emotions to be experienced and communicated within a tolerable range of intensity. Therapists support patients’ capacity to engage in “double awareness” [20] of life’s possibilities, as well as the eventuality of dying. CALM has proven feasible and acceptable in patients with advanced cancer [21–23], and in a randomized clinical trial, was shown to be effective in treating and preventing depression compared to usual care [19]. Notably, CALM is the only psychotherapeutic intervention that has been shown to significantly reduce death anxiety [24]. However, its feasibility and effectiveness in patients with glioma is unknown, as they have been excluded from previous clinical trials.

The aim of this NIH ORBIT model Phase IIa single-arm proof-of-concept trial [25] was to evaluate the feasibility, acceptability, and a signal towards distress reduction of CALM in a neuro-oncology sample. We hypothesized that: (1) patients with malignant glioma would be interested in CALM participation, (2) CALM participants would report high satisfaction with the intervention, and (3) CALM participants would report a reduction in depression and death anxiety symptoms.

Methods

Procedure

Patients were referred for study participation by medical providers or support staff following either direct observation of distress during routine clinic visits or questions to the provider about the terminal nature of the disease and/or death and dying. Eligibility was determined via telephone interview conducted by a trained research assistant. Informed consent was obtained for those who met eligibility criteria. All who enrolled in the study were offered six, biweekly, 50-min telehealth CALM sessions delivered by trained and certified CALM interventionists (KDW & SEB). The process for CALM certification includes multiple training sessions, ongoing supervision, and required fidelity check-points. Caregivers were welcome to participate, if interested. Study interventionists were supervised by the founder of CALM (GR) and a certified CALM interventionist (ARL) in a weekly, group format.

Patients were asked to complete a brief survey assessing acceptability and relevance of CALM domains following each session. Interventionists also completed a brief post-session survey identifying the CALM domains discussed. Session scheduling and attendance were monitored by study staff. If participants missed more than two scheduled sessions at any point during the intervention, a member of the study team attempted to contact the participant to reassess their interest in the study.

Baseline questionnaires were administered via a secure online data collection platform (REDCap) within one week of intervention initiation. These same measures were administered one week following the participant’s completion of the intervention (~ three months from study initiation) per CALM protocol; in addition, a satisfaction questionnaire and the Clinical Evaluation Questionnaire (CEQ) were completed at this timepoint to assess subjective intervention benefit and the patients perceived experience with CALM, respectively. Lastly, within one month following completion of CALM (~ three-to-four months from study initiation), participants completed an exit interview to explore further acceptability and recommendations for intervention optimization. Exit interviews were completed over the phone with a trained research assistant.

Participants

Inclusion criteria:

Participants were patients at an NCI-designated cancer center or surrounding community hospitals meeting the following criteria: 1) malignant glioma diagnosis (grade 3 & 4) confirmed by histopathology report; 2) age ≥ 18 years; 3) ability to participate in sessions via telehealth or telephone; 4) elevated symptoms of depression and/or death anxiety, defined as ≥ 10 on the Patient Health Questionnaire (PHQ-9) and/or ≥ 15 on the Death and Dying Distress Scale (DADDS).

Exclusion criteria:

Individuals were ineligible if they: 1) were deemed to have impaired cognitive functioning as assessed by the Telephone Interview for Cognitive Status (TICS) [26] (score < 21); 2) had indication of major cognitive impairment per chart review; 3) had surgery within the previous two weeks; or 4) were unable to provide informed consent.

Patients undergoing radiation and/or chemotherapy treatment were not excluded as we were interested in the feasibility and acceptability of CALM for malignant glioma patients, many of whom remain on continuous treatment. Further, disease progression was not an exclude criterion, as long as the patient was still able and interested in participating. This was determined collaboratively by the medical provider, CALM interventionist, patient, and caregiver (when applicable).

Measures

Feasibility and Acceptability were operationalized based on published guidelines for Phase IIa proof-of-concept trials and previous peer-reviewed studies adhering to these recommendations, see Table 1 [19, 21, 22, 25, 27].

Table 1.

Feasibility and acceptability evaluation metrics [19, 21, 22, 25, 27]

Domain	Metric	Target
Recruitment feasibility	Enrollment rate	30%
	Screening rate	70%
	Eligibility rate	50%
	Reasons for ineligibility	-
Acceptability of procedures	Baseline questionnaire adherence	75%
	Post-session survey adherence	60%
	Post-CALM questionnaire adherence	60%
	Exit interview adherence	60%
Intervention acceptability	Intervention satisfaction survey benefit rating	Rating ≥ 4/5
	CEQ rating (preliminary)	-
	Retention rates of intent-to-treat participants	70%
	Reason for withdrawal	-
	CALM intervention (6 sessions) adherence	70%

Psychological Distress outcomes were scores on validated, participant-reported questionnaires. Depression and death anxiety were primary outcomes. Generalized anxiety, fear of cancer recurrence, spirituality, and quality of life were secondary outcomes. See Table 2 for outcome measure details.

Table 2.

Outcome measures

Measure		Items	Distress	Descriptors
Patient Health Questionnaire [49]	PHQ-9	9	0–5	Minimal
			5–9	Mild
			10–14	Moderate
			15–19	Moderately Severe
			20–27	Severe
Death and Dying Distress Scale [50]	DADDS	15	0–24	Low
			25–46	Moderate
			47–75	Severe
Generalized Anxiety Disorder [51]	GAD-7	7	0–4	Minimal
			5–9	Mild
			10–14	Moderate
			15–21	Severe
Quality of Life at the End-of-Life Cancer Scale [52]	QUAL-EC	17	0–68	-
Functional Assessment of Chronic Illness Therapy -	FACIT-sp	12	0–48	-
Spiritual Wellbeing Scale [53]
Fear of Cancer Recurrence [54]	FCR-7	7	≥17	60th percentile
			≥27	90th percentile
Clinical Evaluation Questionnaire [55]	CEQ	7	-	-

Data Analytic Plan

To explore the feasibility and acceptability of CALM, descriptive statistics and independent samples t-tests or chi-square tests were used. A Shapiro–Wilk test showed normal distribution of data (ps > 0.05). As a result, to investigate distress reduction, paired sample t-tests of within-group change from baseline to post-intervention were conducted using psychological distress measures. Cohen’s d was used to estimate effect sizes such that 0.2 = small, 0.5 = medium, and 0.8 = large effect. Given the preliminary nature of these analyses, no corrections for multiple comparisons were made. Single mean imputation was used to manage individual item missing data points.

Results

Feasibility of Recruitment

Recruitment occurred from November 2020 to September 2021; 18 patients were referred to the study, 15 agreed to screening (83% screening rate), and all met study eligibility (100% eligibility rate). All eligible participants met criteria for elevated death anxiety (100%) and 53% also reported elevated depressive symptoms. None of the participants were eligible for the study based on depression score alone. Of the 15 eligible individuals referred to the study, 14 consented to participate, resulting in a 93% enrollment rate (see Fig. 1). See Table 3 for demographics.

Fig. 1.

CONSORT Diagram

Table 3.

Description of CALM treatment sample (N = 12)

	M (SD)/n (%)
Demographic variables
Age (years)	56.4 (15.1)
Gender
Male	3 (25.0%)
Female	9 (75.0%)
Race
White	11 (91.7%)
Black	1 (8.3%)
Level of education
High School	1 (10.0%)
Some College	2 (20.0%)
Bachelor’s degree	6 (60.0%)
Master’s Degree	1 (10.0%)
Missing	2 (20%)
Marital status
Single	2 (16.7%)
Married/Partnered Cohabitating	7 (58.3%)
Divorced	2 (16.7%)
Widowed	1 (8.3%)
Children
Yes	9 (75%)
No	3 (25%)
Employment (not mutually exclusive)
Full-Time	2 (16.7%)
Part-Time	1 (8.3%)
Not Working	6 (50.0%)
Receiving Disability	5 (41.7%)
Medical variables
Tumor type
Glioblastoma	10 (83.3%)
Astrocytoma	1 (8.3%)
Oligodendroglioma	1 (8.3%)
Tumor grade
3	2 (16.7%)
4	10 (83.3%)
Tumor hemisphere
Left	5 (41.7%)
Right	6 (50.0%)
Bilateral	1 (8.3%)
Tumor Lobe
Frontal	6 (50.0%)
Temporal	5 (41.7%)
Parietal	4 (33.3%)
Time since diagnosis (months)	21.2 (42.7)
Tumor recurrence
Yes	5 (41.7%)
No	7 (58.3%)
Treatment regimens (not mutually exclusive)
Gross total resection	3 (25.0%)
Partial resection	7 (58.3%)
	M (SD)/n (%)
Cranial radiation	9 (75.0%)
Chemotherapy	8 (66.7%)

M mean, SD standard deviation

Acceptability of Study Procedures

Of the 14 participants who enrolled in the trial, 12 completed baseline assessment and initiated treatment; 83% completed post-session process surveys. Retention was 75% at the post-intervention assessment (9/12), and at the exit interview (9/12). Three participants withdrew during the six-session CALM intervention due to substantial disease progression which precluded psychotherapeutic interaction. There were no differences found with respect to sample demographics, tumor-related characteristics, or distress reporting between treatment completers and those who withdrew.

Acceptability of CALM intervention

All participants who completed CALM reported overall intervention benefit, with an average rating of 4.5/5 (1 = no benefit, 5 = great benefit; mode = 5) during the exit interview. Patients perceived experience of CALM in domains relevant to advanced cancer on the CEQ was also rated high (M = 26.6, SD = 5.8) during the post-intervention surveys. Post-session surveys indicated the participants found the sessions relevant to their distress (4.6/5.0), helpful (4.6/5) and satisfactory (4.6/5). Among those who withdrew (n = 3), post-session surveys were comparable prior to withdrawal: relevance 4.5/5, helpfulness 4.6/5 and satisfaction 4.8/5. Only two of nine participants who completed the intervention recommended a change in session frequency (e.g., preferred more frequent sessions), and three of nine recommended a change in session duration (e.g., 60 vs. 50-min sessions). All four CALM domains were viewed by participants as applicable (1 = not applicable and 5 = very applicable): (1) Symptom management and communication with healthcare providers 4.2/5 (mode = 4); (2) changes in personal relationships 4.1/5 (mode = 5); (3) spirituality, sense of meaning, and purpose 3.9/5 (mode = 4); and (4) the future, hope, and mortality 3.8/5 (mode = 4). From the interventionists’ perspective, Domain 4 ‘the future, hope, and mortality’ was the most discussed (in 83% of sessions), followed by Domain 2 ‘changes in personal relationships’ (81%), Domain 1 ‘symptom management and communication with healthcare providers’ (74%), and Domain 3 ‘spirituality, sense of meaning, and purpose’ (54%). Eight of the nine participants (89%) welcomed a caregiver in at least one CALM session and six of nine (67%) recommended joint sessions with caregivers (n = 5) or family members (n = 6). No one recommended sessions with their medical providers. All reported they would participate in CALM again and would recommend CALM to others (9/9).

Preliminary intervention effects

Preliminary effects were calculated for participants who completed both baseline and post-intervention questionnaires (n = 9). Distributions of scores on distress measures at baseline and post-intervention are presented in Table 4 with a visual representation in Fig. 2.

Table 4.

Preliminary analyses of change from baseline to post-CALM (n = 9)

Outcome variable	Baseline M (SD)	Post-Intervention M (SD)	Δ post-baseline	t	p	d
PHQ-9	13.67 (5.96)	10.44 (5.94)	− 3	1.385	.203	.462
DADDS	44.22 (14.08)	31.00 (17.81)	− 13	3.072	.015	1.024
GAD-7	10.44 (4.30)	6.44 (5.68)	− 4	2.475	.038	.825
QOL-ECRelationship with Healthcare Provider	17.11 (5.01)	18.89 (3.18)	+ 1	− .915	.387	− .305
QOL-ECPreparation for End of Life	10.22 (4.87)	11.67 (3.81)	+ 1	− 1.189	.268	− .275
QOL-ECLife Completion	18.89 (3.85)	18.44 (4.36)	0	.254	.806	− .396
FACIT-SP	22.63 (8.68)	28.00 (6.32)	+ 5	− 4.243	.002	− 1.500
FCR-7	29.71 (10.31)	29.00 (7.44)	0	.175	.867	.066

Paired t-test conducted on baseline to post-CALM

PHQ-9 Patient Health Questionnaire–9, DADDS Death and Dying Distress Scale, GAD-7 Generalized Anxiety Disorder–7, QOL-EC Quality of Life End-of-Life Cancer Scale, FACIT-SP Functional Assessment of Chronic Illness Therapy-Spiritual Well-Being Scale, FCR-7 Fear of Cancer Recurrence −7

Fig. 2.

Visual representation of distress change from baseline to post-CALM

Note. PHQ-9 = Patient Health Questionnaire - 9, DADDS = Death and Dying Distress Scale; GAD-7 = Generalized Anxiety Disorder - 7, QOL-EC = Quality of Life End-of-Life Cancer Scale, FACIT-SP = Functional Assessment of Chronic Illness Therapy - Spiritual Well-Being Scale; FCR-7 = Fear of Cancer Recurrence - 7.

Discussion

CALM is a brief, semi-structured psychotherapy designed to help advanced cancer patients cope with the practical and profound challenges of their illness. Previous trials have found CALM to be feasible, acceptable, and effective for reducing depression and death anxiety in advanced cancer patients. However, patients with gliomas were excluded from these trials. This single-arm, Phase IIa proof-of-concept study is the first to assess feasibility, acceptability, and preliminary intervention effects of CALM in adults with malignant glioma. Results suggest that CALM meets feasibility and acceptability standards in this patient population, shows promising preliminary reductions in death anxiety, generalized anxiety, and depressive symptoms, and improvements in spiritual wellbeing. Larger and randomized controlled trials are warranted to assess further the effectiveness of CALM as an evidence-based treatment for the psychological challenges experienced by patients with malignant glioma.

Screening and enrollment in this trial proved to be feasible in patients with malignant glioma. Most patients referred to CALM agreed to screening, and all who were screened reported elevated levels of distress that met inclusion criteria for the trial. This high eligibility rate is not surprising given the documented high prevalence of death anxiety and other psychological distress in brain tumor populations [1, 7, 8, 28–30]. All of those who were screened endorsed at least moderate symptoms of death anxiety (100%), but only about half (53%) reported equally severe depressive symptoms. This aligns with the neuro-oncology literature documenting higher levels of existential distress than general psychological distress and highlights that death anxiety is a unique, independent dimension of psychological distress that warrants tailored intervention [7]. Limiting eligibility criteria to depression alone would have yielded a much smaller sample of participants who might benefit from this intervention, and would have overlooked the greater end-of-life concerns of these individuals. Almost all of those who met inclusion criteria for the trial consented to participate (93%), which is higher than that found in the original feasibility trials of CALM [21] and other behavioral interventions in neuro-oncology [31, 32].

Overall, study procedures were found to be acceptable based on survey completion rates and participants’ feedback. Post-session survey completion was adequate (83%), although participants required frequent reminders to complete questionnaires at each stage of the trial. It will be important to consider less burdensome assessment protocols for future trials, although in the present study this was not linked to reason for any withdrawal. Minor changes to the protocol merit consideration, including more frequent meetings and extending sessions by 10 min. In future trials, exploring the benefit of more individualized and flexible session frequency, length, and duration, matching the original trial of CALM in non-CNS cancers in this already burdened population will be worthwhile.

The perceived benefit of CALM was consistently rated high and the retention rate of 75% is comparable to, or higher than, that of other psycho-oncology intervention trials in advanced cancer [33, 34]. All participants were willing to complete CALM again (100%) and all would recommend it to others (100%), further underscoring the acceptability of CALM in neuro-oncology. All withdrawals in this trial were due to significant disease progression, highlighting some of the difficulty retaining a population with progressive impairment and a limited life expectancy. However, individuals who withdrew due to disease progression may have still benefited from CALM even without completing the standard protocol of six sessions. Indeed, prior to withdrawal, all participants reported relevance, helpfulness, and satisfaction in post-session surveys.

Exit interviews revealed that all four CALM domains were perceived to be applicable to our participants; however, symptom management and communication with healthcare providers (Domain 1) and changes in personal relationship (Domain 2) were rated as most applicable in post-session surveys. In contrast, study interventionists reported that the future, hope, and mortality (Domain 4) were the most-discussed topics in session. Future qualitative research investigating the experience of patients and psychotherapists and of the content of their conversations might help to elucidate this issue. Another important line of inquiry could investigate the profile of patients who have particular interest in talking about existential issues, the optimal timing of such conversations and therapeutic interventions that facilitate or limit such discussions.

It has been recommended that caregivers be given the opportunity to participate in sessions [35]. Previous research in neuro-oncology demonstrates that a brain tumor diagnosis significantly affects the lives of both patients and their loved ones [36]. In our trial, most participants completed at least one session with a loved one (89%) and most participants recommended joint sessions (67%). The extent to which caregivers benefitted from participating in CALM was not evaluated in this trial. In view of the significant emotional distress reported by caregivers [36] and the obstacles they face in accessing mental health services [37, 38]. CALM might offer them important benefits. Future trials should collect data on caregiver participation in and experience of CALM.

Psychological benefit was assessed in terms of reduction of general and cancer-related distress, and improvement of quality of life and spiritual wellbeing. There were reductions in the primary outcomes, with a medium effect in depression and a large effect in death anxiety following CALM. The high prevalence of death anxiety at baseline and its responsiveness to the intervention are consistent with observations about the importance of existential issues and death-related distress in individuals with advanced cancer [24]. However, despite its relevance for this population, death anxiety is rarely studied as an empirical outcome in neuro-oncology trials [11], and has only recently emerged as an area of psychological concern for glioma patients [7]. There was also a meaningful reduction and large effect in the present trial in symptoms of generalized anxiety, suggesting there may be shared features between these two constructs, such as fear, worry, or rumination.

Interestingly, the outcome that exhibited the largest effect size pre-to-post intervention was spiritual well-being, although CALM is not an inherent religious intervention, this finding is consistent with previous examinations of CALM in patients with advanced cancer [22]. In examining the 12-item FACIT-SP, many prompts are aligned with Domains 3 (sense of meaning and purpose; spirituality) and 4 (the future, hope, and mortality) of CALM. For example, items include “I have a reason for living,” “My life has been productive,” and “Whatever happens with my illness, things will be okay.” Only three items explicitly refer to “faith or spiritual beliefs.” The large change seen in this measure suggests that CALM does indeed produce improvements in the domains it intends to target.

Similar to other CALM trials, quality of life remained stable from baseline to post-assessment. The current trial might have been underpowered to detect such differences as the psychological and hastened end-of-life changes that accompany a brain tumor diagnosis might lead to a lower quality of life relative to other types of cancer [1, 39, 40]. Indeed, only two psychosocial interventions to date have led to an improvement in quality of life for patients with a brain tumor[41–43]. Factors worth investigating include the involvement of caregivers and alternative health professionals (e.g., acupuncturists). Fear of cancer recurrence also remained stable throughout the study period. Fear of cancer recurrence may be inevitable in this population, who face a high likelihood of tumor progression. It is also plausible there would be benefit from a greater focus on fear of recurrence in CALM sessions–particularly in Domain 4. One option could be to add an additional domain of CALM for neuro-oncology focused on how to manage symptoms of fear; another option could be to implement a separate intervention specifically targeting fear of cancer recurrence. There are multiple evidence-based interventions focused on treating this specific fear (e.g., ConquerFear, IN FOCUS, FORT) [44–46], which have yet to be studied in glioma patients.

This study has several limitations. In accordance with the ORBIT model, this single-arm, Phase IIa proof-of-concept trial did not include an active or wait-list control group, limiting our ability to determine to the extent to which pre- to post-intervention changes are attributable to CALM. Further, the satisfaction interviews and effectiveness measures were not completed by those who withdrew, limiting conclusions that can be made about the reasons for these withdrawals. However, no withdrawals were attributed to intervention disinterest. Finally, the majority of the sample were women although glioma is more common in men [47], reducing the generalizability of the findings. This may be because women are more likely to participate in psychotherapy [48] or that clinicians were more likely to approach women about the study. Other concerns with external validity include race/ethnicity and tumor type, which should be considered in recruitment strategies in future trials.

In conclusion, this proof-of-concept trial suggests that CALM and associated study procedures are feasible and acceptable for malignant glioma patients with elevated distress, and the perceived benefit, relevance, and satisfaction of CALM are high. Preliminary evidence supports reduced death anxiety, generalized anxiety, and depressive symptoms following intervention participation. These results complement previous research regarding the feasibility, acceptability, and impact of CALM [19, 21–23] and provide a strong rationale for further study of CALM in neuro-oncology.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.