Cognitive Impairment, Meditative Movement, and Gene Expression in Breast Cancer Survivors

Francisco V. Muñoz ^‡, Linda Larkey ^*

Arizona State University, College of Nursing and Health Innovation, Pomona Valley Hospital Medical Center, Lewis Family Cancer Care Center, USA

‡ Current Affiliation: Casa Colina Centers for Health Care, Research Institute

* Correspondence: Linda Larkey

Academic Editor: Leila Kozak

Special Issue:  Integrative Therapies in Palliative Care

Received: May 20, 2020 | Accepted: July 20, 2020 | Published: July 27, 2020

OBM Integrative and Complementary Medicine 2019, Volume 5, Issue 3, doi:10.21926/obm.icm.2003035

Recommended citation: Muñoz FV, Larkey L. Cognitive Impairment, Meditative Movement, and Gene Expression in Breast Cancer Survivors. OBM Integrative and Complementary Medicine 2019; 5(3): 035; doi:10.21926/obm.icm.2003035.

© 2020 by the authors. This is an open access article distributed under the conditions of the Creative Commons by Attribution License, which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is correctly cited.

1. Introduction

Approximately 78% of breast cancer survivors suffer from adverse effects of cancer treatment [1]. Decrements in cognitive functioning among BCSs is one of these adverse effects that may be due to a combination of factors such as the stress of the cancer diagnosis, treatment effects, and increased sedentary behavior, and is often referred to as chemobrain, or cancer-related cognitive impairment [1].

In addition to cognitive impairment, behavioral comorbidities such as depression, anxiety, and poor sleep quality are also prevalent in cancer survivors, affecting their quality-of-life [2,3,4,5,6]. Inflammation is considered a leading mechanism causing changes in neurobiology that underlie these behavioral conditions [6].

There is growing evidence for mind-body practices such as meditation/mindfulness and meditative movement as well as physical activity to improve quality of life and cognitive function in cancer patients and survivors [7,8,9,10,11,12,13,14,15]. Meditative Movement (MM) has been defined as practices that integrate body movement or postures, a focus on the breath, and a meditative state to cultivate deep state of relaxation [16] and offers the potential of combining both benefits of meditation and exercise for breast cancer survivors. MM combines Tai Chi movements and Qigong into a practice that has been used in research with BCSs because it is very easy to learn and guides participants quickly into a moving meditative state.

Less understood are the molecular signatures that may explain the improvements in cognitive function, anxiety, depression, and sleep quality in breast cancer patients affected by Tai Chi and/or Qigong. Although many studies have examined the effects of mind-body interventions on inflammatory markers, Bowers and Irwin [17] reviewed this body of research and noted that results are mixed and highly dependent upon clinical factors of patients at baseline. In breast cancer survivors, specifically, Janelsins et al., [18] found that the mind-body practice of Tai Chi demonstrated down-regulating effects on pro-inflammatory biomarkers.

For the current study, the targeted gene expression factors in this study were brain derived neurotrophic factor (BDNF), Nuclear Factor kappa B (NF-kB1) and Tumor Protein 53 (TP53). The BDNF gene regulates synaptic plasticity associated with cognitive functioning [19]. Increased levels of BDNF gene expression are associated with improved cognitive functioning [20]. Decreased levels of NF-kB1 gene expression are associated with improvements in managing chronic stress [6,17]. Increased TP53 gene expression is associated with suppressing cancer growth [21]. There is a need for more research exploring the interactions between inflammation, cognitive functioning, and tumor suppression [6].

This pilot study examined the effects of MM on cognitive function and associated symptoms/conditions such as anxiety, depression, and sleep quality. Gene expression factors that may be associated with these changes and/or critical to recovery and future health of BCSs were explored.

Hypothesis 1: In a group of BCSs, 8 weeks of MM will significantly improve cognitive function.

Hypothesis 2: In a group of BCSs, 8 weeks of MM will significantly improve anxiety, depression, and sleep quality.

Hypothesis 3: In a group of BCSs, 8 weeks of MM will significantly affect gene expression associated with cognitive impairment, inflammation, and tumor growth.

2. Method

This was a single-group pilot study testing the effects of 8 weeks of MM practice on changes in cognitive function, anxiety, sleep quality, depression, and selected gene expression factors. The enrollment goal was to consent and assign forty BCSs to an eight-week MM program. Measures on cognitive functioning, associated symptoms/conditions, and gene expression data were collected before and after the 8-week MM program. This study utilized the Paired Sample t-test of SPSS [22] to report p-values from the analysis of pre- and post-MM intervention changes in the cognitive functioning, associated symptoms/conditions, and gene expression data. This study was approved by both the IRB committees at Pomona Valley Hospital Medical Center (PVHMC as the primary IRB) and at Arizona State University (ASU as the secondary IRB).

2.1 Study Population

2.1.1 Eligibility Criteria

Inclusion criteria: minimum 45 years of age; female patients diagnosed with breast cancer, stages 0–III; between six months and five years past primary treatment; ability to speak or understand English; and post-menopausal for at least one year. Exclusion criteria: Women who were unable to stand (e.g., wheelchair or walker bound); patients who were too weak or ill; patients on antibiotics; women working on night shift; and patients with anemia or uncontrolled diabetes. Pregnant women, mentally disabled persons, and prisoners were excluded.

2.1.2 Participant Recruitment

On a daily basis, the clinical trials research coordinator (CRC) at the Pomona Valley Hospital Medical Center (PVHMC) Cancer Care Center (CCC) in Pomona, CA, identified potential participants with the support of site oncologists, site oncology nurses, and the site cancer registry. Flyers were displayed at the Breast Cancer Care Center at the CCC and other affiliated breast cancer care centers in the region. The site cancer registry prepared a mailing list based on the eligibility criteria and invitations were mailed. The CRC generated a list of potential participants based on referrals from the CCC staff and from phone calls from potential participants responding to flyers and mailings.

2.1.3 Screening and Consenting

There were two options for screening and consenting potential participants. In the first option, the site PI or the CRC contacted potential participants referred by the oncology team to screen for study interest and eligibility. In the second option, potential participants responded to the mailed invitation sent by the CRC. Study eligibility and enrollment for participants recruited and screened through both options were confirmed upon review of their medical records after obtaining written consent from potential participants in accordance with Good Clinical Practice (GCP) and Health Insurance Portability and Accountability Act of 1996 (HIPAA). As CCC staff members, the CRC and site PI had access to medical records. After reception of signed Informed Consent form and medical chart review, study participants were scheduled to attend the MM classes.

2.2 Study Intervention

2.2.1 The Intervention

The MM (Qigong/Tai Chi Easy) intervention is a standardized, protocol [23] with a formal training program for practice leaders from the Institute of Integral Qigong and Tai Chi (IIQTC) and has been used in previous research with various populations [8,24,25]. This practice is similar to the short, simplified forms used in the majority of Tai Chi research protocols showing health benefits [26]. The protocol is taught as a series of repeated and simple-to-learn movements rather than long chains of choreographed moves that are more difficult to learn (typical of how traditional long-form Tai Chi is taught).

In this study, the MM intervention was implemented over 8 weeks with class sessions once a week at the Pomona Cancer Care Center. Each class session was approximately one hour. The participants learned gentle movements, ranging from mild to moderate levels of exertion. The participants were asked to practice the MM exercises at home, at their own pace, most days of the week, totaling at a minimum, 2 ½ to 3 hours per week and to log their MM practices in a logbook provided to them. A professionally produced DVD and manual demonstrating a core set of 10, and additional exercises for variety were given to participants to help guide their practice at home. The lead investigator, a certified Mind-Body Medicine Practitioner, and certified QG/TCE practice leader assisted the research participants with the MM sessions. The PI and CRC provided support and guidance to the participants during the MM program.

2.3 Measures

Basic demographic data (gender, ethnicity/race, and age) was collected at baseline. Self-report data on cognitive abilities (perceived and objective assessments), anxiety, depression, and sleep quality, and blood draws to examine gene expression, were collected pre- and post- 8-week MM program (within two weeks prior to and subsequent to the intervention).

2.3.1 Cognitive Function (CF) and Cognitive performance (CP) Tests

Cognitive function was assessed using both a self-report and an objective performance test.

Self-reported CF was assessed using the Functional Assessment of Cancer Therapy-Cognitive Function (FACT-COG), 37 items, validated, including 4 subscales including perceived cognitive impairment (PCI), perceptions of effects of cognitive function on quality of life (QOL), and perceived cognitive abilities (PCA) and comments from other (OTH [27,28,29]

CP Tests: Two brief measures of attention/working memory from the Wechsler Adult Intelligence Scale (WAIS-III) Third Edition [30,31] were used to assess CP: Digit Span and Letter-Number Sequencing, with reliability ratings of .90 and .82 respectively.

2.3.2 Profile of Mood States Short Form (POMS-SF)

The POMs-SF consists of 37 items, adjectives scored on a 5-point Likert scale [32]. The POMS is one of the most frequently used and validated scales in studies of psychosocial interventions with BCSs, and has been validated with Hispanics [33] and multicultural populations [34]. POMS consists of the Total Mood Disturbance (TMD) dimensions (tension-anxiety; depression-dejection; anger-hostility; confusion-bewilderment; Cronbach’s alpha = .93). This pilot study reported results for the tension-anxiety and depression-dejection TMD dimensions (12 items).

2.3.3 Sleep Quality

The Pittsburgh Sleep Quality Index (PSQI): 19 items assess sleep, including subscales for subjective sleep quality, sleep latency, sleep duration, sleep disturbance, habitual sleep efficiency, daytime dysfunction and use of sleep medications [35]. A global PSQI score >5 distinguishes good from poor sleepers with 89.6 % sensitivity and 89.5% specificity and demonstrates Cronbach’s α = 0.83 [36].

2.3.4 Process Control, Manipulations and Fidelity

Each participant received a phone call or email midweek between class attendance (whether they attended or not) to remind them to document practice time (session/minutes) and weekly record level of exertion. MM Practice Logs were provided for participants to document home practice. The midweek contact served both as process control evaluation and as a method for encouraging adherence. Whenever a participant missed a scheduled class, she received an additional call to encourage adherence.

2.3.5 Gene Expression Data

The PAXgene Blood RNA Tube (Catalog No. 762165) was at room temperature (18-25C) and properly labeled for patient identification. The CRC collected one 2.5 ml of blood into one tube using the standard technique for BD Vacutainer® Evacuated Blood Collection Tubes.

2.3.6 After Blood Collection

The CRC inverted the PAXgene Blood RNA Tube 8 to 10 times and stored the PAXgene Blood RNA Tube upright at room temperature (18°C–25ºC) for a minimum of 2 hours and a maximum of 72 hours before processing or transferring to refrigerator (28°C) or freezer (−20°C). After 48 hours in freezer (−20°C) the blood samples were transferred and stored in a freezer (-80°C) until they were prepared for RNA sequencing. Transport Collected Blood Samples: Blood samples were transported to SC2 Core lab for processing and library preparation for RNA sequencing.

2.4 Retention Strategies

The PI and CRC maintained contact with the study participants by telephone, in person, or email to encourage retention in the study. In addition, to providing log-book instructions and meeting time reminders during this contact, the CRC or PI encouraged questions and sharing on how they are experiencing the research study with goal of establishing a therapeutic alliance between the research staff and the study participants. The study participants were also encouraged to initiate contact with the PI or CRC with any questions or concerns by phone, email, or in person.

2.4.1 Sample Size Justification

The enrollment goal of 40 participants followed the guidelines of Browne [37] to formulate samples sizes for pilot studies. This was the target for recruitment with an expected attrition rate of 20%. Although 27 participants were consented, only 14 completed the study. The 14 who completed the study met the standards established by Julious [38] for pilot study sample sizes. Browne and Julious [37,38] provided sample size calculation guidelines for pilot studies designed to collect preliminary data for a clinical trial.

2.4.2 Hypothesis Testing Data Analysis Plan

This study utilized the Paired Sample t-test of the SPSS statistical program to analyze the primary outcome changes in cognitive function anxiety, depression, and sleep quality over an 8-week period. Changes in BDNF, NF-kB1, and TP53 gene expression results were analyzed using the paired sample t-test in SPSS to determine if the means of the pre- and post-MM gene expression data were significantly different.

2.4.3 Missing Data

Participant responses were reviewed for missing data and participants were given an opportunity to complete overlooked questions or indicate a preference not to answer during the data collection session. All analyses were conducted on participants who completed the intervention. If post-intervention data was unavailable, missing data were not imputed. As a small pilot study to detect trends in change from pre- to post-intervention, we do not expect there to be a systematic bias introduced by the data loss of a small number of participants. One study participant did not complete her post-MM program WAIS-III measure and did not submit her MM logbook due to schedule conflicts.

2.4.4 Secondary Outcomes

Peripheral blood samples were collected before and after the 8-week MM program. The gene expression data was processed and analyzed by the bioinformatics team at the Single-Cell, Sequencing, and CyTOF Core (SC2), Children's Hospital Los Angeles (CHLA), Los Angeles. These data were analyzed using the paired sample t-test in the IBM SPSS Statistics program.

RNA sequencing: Sequencing libraries were prepared from previously purified RNA using the Illumina TruSeq Stranded mRNA Library Prep kit following the manufacturer’s instructions. Sequencing was performed on a NextSeq 500 platform using 2×75bp chemistry [39].

Gene Expression data analysis: Quality control and adapter trimming was performed using trim galore (v0.4.2) with default parameters [40]. Reads were aligned to the GRCh38 reference genome and transcriptome using HISAT2, v2.1.0 [41], and transcript quantification was performed using featureCounts, v1.5.1 [42]. Differential expression analysis was performed using the ‘DESeq2’ R package, v1.16.1, [43] and a rank score calculated as -log10(q-val)*sign (log2 FoldChange) was used as input to the GSEA Preranked tool for pathway analysis [44].

The DESeq2 data were used to identify the gene expression changes over the 8-week MM program. DESeq2 (differential expression sequence) files are TMM (weighted trimmed mean of the log expression ratios - trimmed mean of M values) normalized count per million reads.

3. Results

Twenty-seven breast cancer survivors were consented and enrolled into the study. Fourteen completed the MM program in 8-week class cohorts ranging from 5 to 9 participants. All of the 14 study participants were female and ranged in age from 45 to 95 years old. Ten of the study participants were White, three were Latina, and one was Black.

The primary reason 13 of the 27 consented study participants dropped out of the study was schedule conflicts. Other reasons included changes in physical or emotional health related to preexisting conditions, lack of transportation, and perceived prolonged waiting time to begin MM groups after consenting. The research staff maintained contact with those who discontinued participation in case their circumstances changed until they requested no further contact with the study.

Twelve of the fourteen participants attended every weekly session at the Cancer Care Center. Two were absent more than two weekly sessions. Thirteen study participants submitted their logbooks documenting the number of minutes they practiced MM weekly. Their weekly log included their weekly 60-minute MM sessions at the Cancer Care Center. The MM logbook reported mean in minutes per week was M = 91.06, SD = 46.12.

Hypothesis One (Table 1): Cognitive Functioning

Significant improvements, as predicted, were noted in the FACT-COG subscales: Perceived Cognitive Impairments (CogPCI), (t(13) = 3.4, p = .01); Perceived Cognitive Abilities (CogPCA), (t(13) = -2.4, p = .03); Perceived Impairments on Quality of Life (CogQOL), (t(13) = 2.2, p = .04); and Comments from Others (CogOth), (t(13) = 2.3, p = .04).

The significant changes for Perceived Cognitive Impairments (CogPCI) from pre-MM program, mean = 25.4 to post-MM program, mean = 17.3 indicated the BCS perceived their cognitive impairment declined. The BCS also perceived a decline in Perceived Impairments on Quality of Life (CogQOL) indicated by the significant change in pre-MM program, mean = 3.7 to post-MM program, mean = 1.8.

The positive significant changes indicated in the Perceived Cognitive Abilities (CogPCA) subscale suggested the BCSs’ subjective improvement in their cognitive abilities noted in the pre-MM, mean = 16.0 significant change to post-MM, mean = 19.2. Another significant decline reported by the BCS was on the subscale, Comments from Others (CogOth), evidenced by the pre-MM program, mean = 1.5 to post-MM, mean = 0.2. This subscale reflects the BCSs’ experience of what others have said to them regarding perceived cognitive decline.

Significant improvements, as predicted, were noted in the the WAIS-III, Letter/Number sequencing results, (t(12) = -3.2, p = .01). Significant improvements were not noted in the WAIS-III, Digit Span Forward and Backward measure, (t(12) = -0.532, p = .60). One of the two objective WAIS-III assessments of cognitive performance, Letter/Number sequencing, showed significant improvement demonstrated by the pre-MM mean 10.7 increase to the post-MM mean 12.7. The WAIS-III digit span Forward/Backward assessment also demonstrated an increase in the pre-MM mean 19.2 to post-MM mean 19.7 but did not reach significance.

Table 1 Functional Assessment of Cancer Therapy-Cognitive Function (FACT-COG) Subscales.

Hypothesis Two (Table 2): Sleep Quality, Anxiety, and Depression

The Pittsburg Sleep Quality Index (PSQI) scores, evidenced significant improvement, (t(13) = 2.5, p = .03). The POMS/anxiety subscale showed a significant decrease (t(13) = 1.9, p = .05) and the POMS/depression reduction did not reach significance (t(13) = 1.9, p = .08). BCSs’ poor sleep quality significantly decreased from pre-MM program, mean = 1.5 to post-MM program mean = 0.9. Their anxiety levels also significantly decreased from pre-MM program mean = 9.2 to post-MM program mean = 4.2. Depression levels also decreased, pre-MM mean = 4.1 to post-MM mean 2.0, but did not reach significance.

Table 2 The Pittsburgh Sleep Quality Index (PSQI) and Profile of Mood States (POMS).

Hypothesis Three (Table 3): Gene Expression Factors

Significant changes were not found for BDNF (t(13) = -1.1, p = .30), NF-kB1 (t(13) = -0.2, p = .80), and TP53 (t(13) = -0.52, p = .61).

Table 3 Gene Expression Factors.

The mean (sd) differences (Table 3) in BDNF gene expression, 0.19 (-0.2); NF-kB1 gene expression, 2.0 (-3.8); and TP53 gene expression, 0.8 (2.8); all indicated an increase in RNA between pre- and post-MM intervention. Although these differences are not statistically significant.

4. Discussion

The primary outcome expected to improve in response to the intervention as cognitive function, and the assessments showed promising results on this factor, measured both with self-report and one of the objective measures. The direction of change for the FACTCog subscales all indicated a perceived significant improvement in cognitive functioning. The WAIS-III Letter/Number assessment provided significant and objective evidence to support improvement in cognitive functioning. The direction of change for the WAIS-III Digit Span Forward/Backward assessment indicated an improvement in cognitive functioning but did not reach significance.

The PSQI and POMS/anxiety subscale also provided significant and positive subjective changes in sleep quality and anxiety levels. The POMS/depression subscale evidenced positive changes but these changes did not reach significance. This lack of significance may be due to the majority of the BCSs in this study not reporting any major issues with depression.

The research design for the GME study was based on the intent to perform a pilot study and examine a trend for effects of the intervention on gene expression and symptom changes. Regarding the enrolled participants failing to initiate, we have further explained that conducting this study in a small cancer center, recruitment was slow, thus often extending the time from consent to a new class group start date. This is now discussed in the limitation section.

Limitations. The MM suggested time goal for the study participants was between 2.5 and 3 hours per week. The MM logbooks showed a mean of 1.5 MM hours per week for the study participants. Lower than projected time engagement in the intervention may have attenuated results across all of the assessments. Increasing the amount of engagement time between the research team and the study participants using phone calls, texts or other types of encouragement may increase the MM practice hours per week. The consented participants who did not begin the study may have perceived too long a wait time for the MM groups to begin. The wait time often resulted in loss of interest or change in other circumstances. For these consented participants, consistent communication from the research team may have maintained their engagement with the study until the MM groups were started.

Despite the small number of study participants, the results of this study parallel the results of larger studies conducted on the effects of meditation, exercise, and mind-body practices such as Yoga or Tai Chi/Qigong on the cognitive functioning, anxiety and quality of life factors for BCS [7,8,9,10,11,12,13,14,15,45]. The absence of significant change in the means for pre- and post- intervention BDNF, TP53 and NF-kB1 gene expression may indicate that there is no effect of the intervention, that the intervention dose was insufficient, or that the sample size was too small. Another study that examined NF-kB gene expression in response to a 12-week MM intervention, Tai Chi Chih (TCC), but with higher dose (2 hours/week), reported less increase in NF-kB gene expression compared to control, but also not reaching significance [46]. Differences between the two studies may due to type of study participants (older adults vs. BCSs), and length and dose of intervention.

Given such a small sample size, we did not examine potential associations of the biomarkers assessed with the improvements in cognitive functioning and quality of life factors. In larger studies in the future, it would be important to examine these relationships as possible mediators of symptomatic change. Rather than a focus on the target genes themselves, an analysis of the biological pathways of cognitive functioning, tumor suppression, and pro-inflammation may yield more insight on potential effects of the MM program on the molecular signatures and gene expression within the study participants.

Further, while many studies of MM in cancer patients or survivors have examined (and found) reductions in inflammatory cytokines, very few have assessed the gene expression pathways to that inflammation, and none, to our knowledge, have explored BDNF specifically. Although significance changes were not found, these novel targets were worth exploring, and in fact, show promise for studies that are better powered for significance testing.

5. Conclusion

This pilot study tested the effects of the MM program on changes in cognitive functioning and associated symptoms/conditions such as anxiety, depression, and sleep quality. This study also utilized gene expression factors to improve understanding of the biological mechanisms potentially associated with these changes. The patient self-reported data indicated the MM program was perceived as improving cognitive functioning, in support of other study results. This perception was further supported by an objective cognitive performance measure showing significant improvements on one of the two assessments. The patient self-reported data also indicated the MM program was perceived as improving anxiety and sleep quality. The gene expression data did not evidence significant results to support association with positive changes in cognitive performance, anxiety, and sleep quality. The major limitation of this study is the underpowered sample size that does not allow for the definitive testing of the hypotheses. This study provided preliminary data for a future powered randomized control trial to study psycho-behavioral outcomes and associated genomic expression factors.

Acknowledgements

This project was supported by a grant from the National Cancer Institute (1R01CA182901- 01A1).

Author Contributions

Dr. Munoz conducted the study, analyzed, and interpreted results while taking the lead in writing the manuscript. Dr. Larkey provided study oversight, expertise for MM intervention, and editorship for manuscript. Both authors provided critical analysis as they shaped the research and manuscript.

Competing Interests

The authors have declared that no competing interests exist.