Sustained Effects of Mantra Meditation Compared to Music Listening on Neurocognitive Outcomes of Breast Cancer Survivors: A Brief Report of a Randomized Control Trial

Abstract

Objective:

Cancer-related cognitive impairment is common following the end of adjuvant treatment and there are limited treatment options for it. We compared the sustained cognitive (primary) and psychological (secondary) effects of mantra meditation to classical music listening 8 weeks after interventions ended (Time 3) compared to baseline (Time 1).

Methods:

A two-group parallel random assignment experimental design was used in a community setting. Thirty one breast cancer survivors (ages 21 to 75, received chemotherapy, reported cognitive complaints) were randomly assigned to practice mantra meditation (n=16) or listen to music (n=15) 12 minutes a day for 8 weeks. No blinding was used. Repeated measures analysis of variance models were used to compare Time 1 and Time 3 data for the 26 survivors (13 per group) who completed the interventions and Time 3 data collection.

Results:

Verbal fluency (p < .001, η_p² = 0.58), attention (p = .002, η_p² = 0.33), immediate memory recall (p < .001, η_p² = 0.38), perceived cognitive impairment (p < .001, η_p² = 0.39), and quality of life (p = .001, η_p² = 0.35) improved significantly across time for both groups. The two conditions did not differ significantly in changes across time. There were no adverse effects.

Conclusion:

Daily mantra meditation or classical music listening may be beneficial for cognitive outcomes and quality of life of breast cancer survivors with cancer-related cognitive impairment. The cognitive benefits appear to be sustained beyond the initial intervention period.

Clinical Trials Registration number:

NCT03696056, recruitment status completed.

Introduction

Cancer-related cognitive impairment (CRCI) affects up to 60% of breast cancer survivors (BCS).^1,2 CRCI often occurs in the domains of attention, executive functioning, memory, and processing speed and can negatively impact survivors’ quality of life.^1,3 There are limited treatment options and no standard of care for managing CRCI, but evidence suggests that non-pharmacological treatment options—including cognitive training, exercise, cognitive rehabilitation, and meditation—may be beneficial for managing CRCI.⁴

We conducted a randomized control trial of mantra-based meditation compared to classical music listening in breast cancer survivors who were 3 months to 6 years post chemotherapy completion.⁵ These interventions were chosen based on previous reports^6–11 that indicate they may be beneficial to cognitive functioning and psychological well-being in cognitively vulnerable adults. We previously described positive short-term effects (after 8 weeks of intervention) for both interventions on perceived cognitive impairment, quality of life, immediate memory recall, and verbal fluency.⁵ In this follow-up study, we compared the sustained effects of these two interventions (Time 3: 16 weeks after enrollment/8 weeks after the end of the intervention) on cognitive function (performance and perceived) and psychological functioning (anxiety, depression, stress).

Methods

This study was approved and monitored by the University of Texas Institutional Review Board (protocol #2018–05-0155) and registered with clinicaltrials.gov (NCT03696056). Informed consent was obtained from all individual participants included in the study. A two-group parallel random assignment experimental design was used. No methodological changes were made after the trial started. We followed the CONSORT guidelines, and details of the study design and outcomes at Time 2 have been previously described.⁵ Briefly, BCS who reported cognitive symptoms (reporting 5 or more cognitive symptoms occurring “sometimes” or more, per the Perceived Deficits Questionnaire¹²), were up to 6 years post chemotherapy (although they could be taking adjuvant therapies), could read and write English, and between the ages of 21 and 75 were recruited from community settings. Those who self-reported regular meditation practice (more than 1 time a week for a minimum of one month), with a pre-cancer history of psychiatric or cognitive disorders, and/or inflammatory comorbidities were excluded. The exclusion criteria based on inflammatory comorbidities (e.g., diabetes mellitus type II, rheumatoid arthritis) was related to biomarkers that were also measured in this study, but not reported here.

After enrolling and providing informed consent, baseline data collection (Time 1) occurred which included cognitive testing and self-report questionnaires. Participants were then randomized to one of the two intervention groups—mantra-based meditation (the primary intervention) or classical music listening (the attention control condition) using a randomization software for block randomization on a 1:1 ratio. Each participant was provided an electronic copy of their intervention protocol details by the study Principal Investigator (who did not collect data at Time 1 or Time 2). Research staff assessing outcomes at Time 1 and Time 2 were blinded to group allocation. Each protocol was adapted from earlier studies using the same inventions with older adults with cognitive decline,^6–10 and the details have been previously described.⁵ Briefly, participants were instructed to meditate or listen to music from a digital playlist, sitting quietly, with eyes closed for 12-minutes a day for 8 weeks.

Participants were asked to record their progress in a hard copy of a practice log. Facilitators for each group provided weekly reminders and were available to answer questions about the protocols upon request. After 8 weeks of the daily sessions, Time 2 data collection occurred. Then between Time 2 and Time 3, participants were told they could continue the daily practice if they wanted to, and to log any additional practice sessions in their logs. Practice log data were entered in REDCap (Vanderbilt, TN) by research staff. Time 3 data collection occurred 8 weeks after Time 2 and included cognitive testing and surveys. All data were stored in REDCap.

Primary outcomes.

Cognitive performance was measured with three standardized neuropsychological tests of immediate and delayed memory recall with the Hopkins Verbal Learning Test,¹³ verbal fluency with the Controlled Oral Word Association Test,¹⁴ and attention and executive functioning with the Trail Making Test A & B (where low scores represent better performance).¹⁵ Perceived cognitive impairment and quality of life were measured with the Functional Assessment of Cancer Therapy-Cognitive Function Instrument version 3 (FACT-Cog), Perceived Cognitive Impairments (PCI) and the FACT-Cog Quality of Life (QOL) subscales.¹⁶

Secondary Outcomes.

The Patient-Reported Outcomes Measurement Information System¹⁷ Short Form 8a scales were used to measure anxiety, feelings of depression, and fatigue. Stress was measured using the Perceived Stress Scale.¹⁸ The PROMIS Pain Interference-Short Form 6 assessed pain.¹⁹ No changes were made to trial outcomes after the trial commenced.

Data analyses.

As previously described in detail,⁵ the sample size was determined by the primary aim of the study, to determine feasibility. Descriptive statistics were used to check assumptions. We used repeated measures analysis of variance (RM-ANOVA) to evaluate within and between group changes in cognitive and psychological outcomes from Time 1 to Time 3 (2 levels), not including Time 2 as these data were previously reported.⁵ For this 2 level RM-ANOVA, the assumption of homogeneity was verified for each model using the Levene’s test. Effect sizes were calculated using partial eta squared (η_p²). Alpha was set at .05 for all statistical tests.

Results

Recruitment occurred between October 2018 and July 2019. Thirty one BCS were enrolled, thirteen BCS completed the mantra mediation intervention and Time 3 data collection, and thirteen BCS completed the music listening intervention and Time 3 data collection (See Consort Flow Diagram, Supplemental Figure 1). On average, participants were 52.22 +/− 10.18 years of age, 29.93 (+/− 19.22) months since the end of chemotherapy treatment, reported 15.89 (+/− 1.95) years of education, had a history of stage II or III breast cancer (83.07%), were employed at the time of study (81.48%), and were post-menopausal either naturally or treatment induced (77.78%). Baseline demographic and clinical factors did not differ between the two groups and have been previously described in detail.⁵ Four participants in the music group and three in the meditation group reported additional practice time in the post intervention period (mean total music minutes: 13.93, standard deviation: 22.16; mean total meditation minutes: 6.39, standard deviation: 14.52; t = −1.03, p = .31).

No significant group by time effects were found in the RM-ANOVA models of the cognitive or psychological outcome variables. For the cognitive outcomes, there were significant time effects found between T1 and T3 for verbal fluency (F(1,24) = 33.54, p < .001, η_p² = 0.58), attention (F(1,24) = 11.77, p = .002, η_p² = 0.33); immediate recall memory (F(1,24) = 14.87, p < .001, η_p² = 0.38); and perceived cognitive impairment (F(1,24) = 15.59, p < .001, η_p² = 0.39, Table 1, Figure 1). For the psychological outcomes, there was a significant time effect for quality of life (F(1,24) = 13.01, p = .001, η_p² = 0.35, Table 1, Figure 2). No harms were identified in either intervention group.

Table 1.

Repeated measures (from Time 1 to Time 3) analysis of variance with cognitive and psychological outcomes as dependent variables

	Mantra meditation (n=13)		Music listening (n=13)
Dependent Variables	Time 1 Mean (SD)	Time 3 Mean (SD)	Time 1 Mean (SD)	Time 3 Mean (SD)	F statistic from RM ANOVA time effect	P value for time effect	ηp2 for time effect
Perceived Cognitive Impairment	49.92 (17.28)	58.54 (16.53)	37.15 (21.11)	50.69 (19.35)	15.59	< .001	.39
Quality of Life	11.39 (3.99)	13.69 (2.43)	9.85 (5.06)	12.08 (4.21)	13.01	.001	.35
Immediate Memory	29.15 (2.85)	32.23 (2.13)	29.77 (2.62)	31.08 (3.15)	14.87	< .001	.38
Delayed Memory	11.33 (1.16)	11.75 (.62)	11.5 (.80)	11.5 (.80)	1.54	.23	.07
Verbal Fluency	37.77 (8.67)	43.85 (9.51)	39.62 (10.10)	47.15 (15.51)	33.54	< .001	.58
Attention	25.85 (6.95)	20.92 (2.60)	27.15 (12.07)	22.15 (7.09)	11.77	.002	.33
Executive Function	58.39 (13.48)	51.39 (11.59)	52.08 (23.61)	49.46 (20.79)	3.80	.06	.14
Anxiety	15.69 (8.57)	14.92 (4.80)	18.62 (6.69)	16.31 (5.66)	1.62	.22	.06
Depressive	13.31 (7.32)	12.77 (6.67)	12.29 (3.93)	12.46 (5.29)	.85	.85	.00
Fatigue	20.77 (5.95)	19.54 (4.05)	24.85 (8.24)	22.00 (8.98)	1.86	.19	.07
Stress	12.92 (7.82)	12.15 (7.65)	13.92 (7.76)	11.92 (8.51)	1.13	.30	.05
Pain	11.92 (6.75)	10.39 (4.46)	10.92 (4.73)	8.69 (5.68)	3.86	.06	.14

Footnote: For each variable, time was the random factor and group assignment the fixed factor. Immediate and delayed memory were measured with the Hopkins Verbal Learning Test,¹³ verbal fluency with the Controlled Oral Word Association Test,¹⁴ attention and executive functioning with Trail Making Tests Part A and B.¹⁵ Perceived cognitive impairment and quality of life were measured with the Functional Assessment of Cancer Therapy-Cognitive Function Instrument version 3 (FACT-Cog), Perceived Cognitive Impairments (PCI) and the FACT-Cog Quality of Life (QOL) subscales.¹⁶ The Patient-Reported Outcomes Measurement Information System¹⁷ Short Form 8a scales were used to measure anxiety, feelings of depression, and fatigue. Stress was measured using the Perceived Stress Scale.¹⁸ The PROMIS Pain Interference-Short Form 6 assessed pain intensity.¹⁹ The Levene’s Test of homogeneity was not violated for any of the models.

Figure 1.

Descriptive plots for cognitive outcomes across Time 1 (baseline), Time 2 (8 weeks post interventions) and Time 3 (8 weeks after Time 2). Bars represent standard error of the mean. A) Verbal fluency was measured with the Controlled Oral Word Association Test, higher scores are better. B) Attention was measured with the Trails A, lower scores are better. C) Executive function was measured with Trails B, lower scores are better. D) Immediate recall was measured with the Hopkins Verbal Learning Test, higher scores are better. E) Delayed Memory was measured with the Hopkins Verbal Learning Test, higher scores are better. F) Perceived Cognitive Impairment was measured with the Functional Assessment of Cancer Therapy Cognitive Function version 3, Perceived Cognitive Impairment Subscale. Higher scores are better.

Figure 2.

Descriptive plots for psychological outcomes across Time 1 (baseline), Time 2 (8 weeks post interventions) and Time 3 (8 weeks after Time 2). Bars represent standard error of the mean. A) Quality of life (QOL) was measured with the Functional Assessment of Cancer Therapy Cognitive Function version 3, Quality of Life Subscale. Higher scores are better. B) Stress was measured with the Perceived Stress Scale, lower scores are better. C) Depressive Symptoms were measured with the Patient-Reported Outcomes Measurement Information System (PROMIS) scales short form Emotional Distress Depressive symptoms 8a, lower scores are better. D) Fatigue was measured with the PROMIS short form Fatigue 8a, lower scores are better. E) Anxiety was measured with the PROMIS short form Emotional Distress Anxiety 8a, lower scores are better.

Discussion

To our knowledge, this is the first study to compare daily remote delivery of mantra meditation to classical music listening (for 12 minutes a day) to improve cognitive functioning and psychological symptoms in BCS. We previously described short-term positive effects of both interventions on immediate memory recall, verbal fluency, perceived cognitive impairment, and quality of life after just 8 weeks of the daily practice.⁵ Here we found evidence of sustained positive effects on cognitive outcomes in both intervention groups 8 weeks beyond the end of the intervention that did not differ by group across time. In addition to the maintenance of effects originally reported,⁵ we also found additional positive effects of the mantra meditation and music listening protocols on attention at Time 3, suggesting a delayed positive impact of meditation and/or music listening on these outcomes.

Our findings are consistent with Innes et al.²⁰ who evaluated the same intervention protocols (mantra meditation and music listening 12 minutes a day) in older adults with subjective cognitive decline (n=60), but for a longer duration (12 weeks) with sustained effects that did not differ by group at 24 weeks. Notably, the effects we found on perceived cognitive impairment in our sample (approximately 9 points decrease in the meditation group and 13 points in the music group on the FACT-Cog PCI from Time 1 to Time 3) were two and three times, respectively, the clinically meaningful change that have been previously reported in this population using the same measure.²¹

Research utilizing brain imaging to study both meditation²² and music listening^23–25 have reported structural and functional changes suggestive of neuroplasticity. CRCI in breast cancer survivors has been consistently associated with altered neural structure and function especially in the prefrontal cortex, hippocampus, and regions involved in the default mode and attention networks.^26,27 Activation in these regions has been reported for both focused meditation and music listening interventions,^22,28–32 suggesting that the positive cognitive outcomes in this study could be a function of beneficial neural reorganization. There is also overlap in cognitive function and emotional regulation in terms of neurobiology^33,34 and symptom co-occurrence.^35,36 It is possible that the interventions affected both, either simultaneously or via indirect pathways resulting in neural reorganization, and could explain why these effects were maintained at Time 3. It is also possible that our findings were a function of practice effects on the cognitive tests. These mechanistic hypotheses need to be further evaluated using neuroimaging methods and with a waitlist control group to evaluate practice effects.

We previously reported positive effects of music listening on perceived stress, and improvements in anxiety and feelings of depression (that did not reach significance) immediately following the 8-week interventions.⁵ However, there were no effects on psychological outcomes at T3 for either group, Both meditation and music listening have been previously associated with improved well-being and mood, and reduced stress, anxiety, and depressive symptoms.^37,38 It is possible that these psychological effects are transient, or dependent on continued exposure to either daily meditation or music listening, and may explain why we did not find sustained effects for these outcomes in either intervention group at Time 3.

The participants in our sample had lower continued daily practice during the “optional” time period compared to Innes et al.²⁰ who reported 71% of their sample continued to practice. In our sample we found more uptake of music listening compared to meditation, but this difference was not significant. It is unclear why most of our participants did not choose to continue the intervention during the optional period, there were no significant demographic or clinical variable differences between those who chose to continue compared to those who chose not to. It is also possible that some participants did continue some level of activity, but did not record this on their logs.

Limitations to the study must be considered. First, the sample size was small, as the original aim of the study was to establish feasibility and acceptability of the two intervention protocols. We implemented a limited cognitive battery, so the interventions may have affected other cognitive domains not measured here. We controlled for prior exposure to meditation, but not music listening behaviors, so participants’ normal music listening behaviors may have biased the results. We did not include a waitlist control group and therefore cannot account for the potential effects of time and/or attention on study outcomes. In this small exploratory study, we did not correct for multiple comparisons, and therefore findings could be influenced by type I error. Finally, the sample may be biased toward those most interested in these interventions, so results may not generalize to the population of breast cancer survivors who experience cognitive limitations.

Despite these limitations, our findings add to a growing evidence base for the efficacy of meditation interventions on improving quality of life, psychological and cognitive functioning in breast cancer survivors.^39,40 Here we add new findings on sustained effects of two daily interventions, mantra meditation and classical music listening, beyond the initial 8 weeks, providing additional evidence supporting these cost-effective, easy to learn, and accessible interventions that require 12 minutes a day. These intervention characteristics are important, as the majority of studies evaluating meditation for improving neurocognitive outcomes in cancer survivors have been resource intensive (e.g., large time requirements; travel to in person sessions),⁴¹ limiting accessibility for survivors unable to take time away from their social responsibilities and travel to the research site.

In the present study we examined the effects of classical music, modeled after previously reported studies,^6–11 however research supports positive cognitive⁴² and psychological³⁸ effects of various types of instrumental music listening. It is possible that listening to other genres of music have similar effects, but more research is needed. Future studies of these protocols should include a waitlist control group arm, investigate underlying neural mechanisms of meditation and music listening protocols, and explore effects of other types of music and/or meditation. Our findings provide new evidence on positive neurocognitive effects of an at-home, remotely delivered intervention that BCS can complete on their own time which is especially relevant for survivors who are working or have family responsibilities. Furthermore, these interventions could be easily adapted for telehealth delivery, making them even more accessible.

Key Message:

There are limited treatment options for managing cancer-related cognitive impairments. Daily mantra meditation or classical music listening for 12 minutes a day may improve cognitive outcomes and quality of life for cancer survivors, with no negative side effects.