Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Mar 8.
Published in final edited form as: Exp Gerontol. 2024 Jan 25;187:112364. doi: 10.1016/j.exger.2024.112364

Exercise and inflammatory cytokine regulation among older adults with myeloid malignancies

Kah Poh Loh a,b,*, Ying Wang c, Chandrika Sanapala d, Nikesha Gilmore e, Colleen Netherby-Winslow e, Jason H Mendler a,b, Jane Liesveld a,b, Eric Huselton a,b, AnnaLynn M Williams a,e, Heidi D Klepin f, Marielle Jensen-Battaglia c, Karen Mustian a,e, Paula Vertino a,g, Martha Susiarjo h, Michelle C Janelsins a,e
PMCID: PMC10923152  NIHMSID: NIHMS1970000  PMID: 38266886

Abstract

Tumor necrosis factor (TNF)α is a major regulator of inflammation. However, the epigenetic regulation of TNFα in the context of an exercise intervention among older adults with cancer is understudied. In this exploratory analysis, we used data from a single-arm mobile health (mHealth) exercise intervention among older adults with myeloid malignancies to 1) assess changes in TNFα promoter methylation, TNFα mRNA expression, serum TNFα and other related-cytokine levels after intervention; and 2) assess correlations between blood markers and exercise levels. Twenty patients were included. From baseline to post-intervention, there was no statistical changes in TNFα promoter methylation status at seven CpG sites, TNFα mRNA expression, and serum TNFα levels. Effect sizes, however, were moderate to large for several CpG sites (−120, −147, −162, and −164; Cohen’s d = 0.44–0.75). Median serum TNFα sR1 levels increased (83.63, IQR 130.58, p = 0.06; Cohen’s d = 0.18) but not the other cytokines. Increases in average daily steps were correlated with increases in TNFα promoter methylation at CpG sites −147 (r = 0.48; p = 0.06) and −164 (r = 0.51; p = 0.04). Resistance training minutes were negatively correlated with TNFα promoter methylation at CpG site −120 (r = −0.62; p = 0.02). All effect sizes were moderate to large. In conclusion, after a mHealth exercise intervention, we demonstrated changes with moderate to large effect sizes in several CpG sites in the TNFα promoter region. Exercise levels were correlated with increases in TNFα promoter methylation. Larger exercise trials are needed to better evaluate TNFα regulation to inform interventions to augment TNFα regulation in order to improve outcomes in older adults with cancer.

Keywords: Myeloid malignancies, Mobile health, Exercise, Inflammation, DNA methylation, Cytokines

1. Introduction

Acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) occur commonly in adults aged ≥60 years (Arber et al., 2016; Ma et al., 2007; Shallis et al., 2019). Up to 88 % of older adults with these myeloid malignancies experience physical impairments prior to and during chemotherapy (Alibhai et al., 2007; Klepin et al., 2013; Klepin et al., 2016; Loh et al., 2018; Loh et al., 2020; Saad et al., 2020), resulting in poor quality of life, treatment interruptions, and reduced survival (Brown et al., 2015; Hurria et al., 2011; Klepin et al., 2013; Saad et al., 2020; Tinsley et al., 2017; Wyatt et al., 2015). Exercise interventions can mitigate physical impairments and improve outcomes in this population (Alibhai et al., 2012; Klepin et al., 2011; Klepin et al., 2022; Loh et al., 2022).

Tumor necrosis factor (TNF)α plays a role in early inflammatory responses and is regarded as the “master regulator” of inflammation (Parameswaran and Patial, 2010; Sedgwick et al., 2000). Preclinical studies demonstrate that cancer and chemotherapy increase serum TNFα and related cytokine levels [e.g., interleukin-1β (IL-1β), interleukin-6 (IL-6), and interleukin-8 (IL-8)] (Asfaha et al., 2013; Elsea et al., 2015; Lamkin et al., 2011; Logan et al., 2008; Smith et al., 2014). Activation of TNFα and related cytokine pathways are associated with physical function decline (Biesmans et al., 2015; Bluthé et al., 1994; Goodman, 1991; Kaster et al., 2012; Palin et al., 2008), which is abrogated by inhibition of these pathways (Koo and Duman, 2009). In older adults with myeloid malignancies, higher baseline serum TNFα levels are associated with shorter survival (Loh et al., 2020). Exercise reduces inflammation, which may represent one pathway by which exercise mitigates physical function decline (Meneses-Echávez et al., 2016; Smart et al., 2011).

DNA methylation, the addition of methyl groups to DNA at CpG sites (where a cytosine nucleotide is followed by a guanine), influences the transcription of DNA to mRNA and subsequent protein expression (Jin et al., 2011). DNA methylation of the TNFα promoter region decreases TNFα mRNA and protein expression (Supplemental Fig. 1) (Sullivan et al., 2007; Uddin et al., 2011). Prior studies have shown that exercise increases TNFα promoter methylation and decreases TNFα mRNA and protein expression (Greiwe et al., 2001; Radom-Aizik et al., 2014; Shaw et al., 2014; Tsukui et al., 2000). However, the epigenetic regulation of TNFα in the context of exercise among older adults with cancer is understudied.

We previously conducted a single-arm study testing the feasibility of a mobile health (mHealth) exercise intervention among older adults with myeloid malignancies receiving outpatient chemotherapy over 2 cycles of chemotherapy (Loh et al., 2022). In this exploratory analysis, we aimed to 1) assess changes in TNFα promoter methylation, TNFα mRNA expression, and downstream TNFα and related inflammatory cytokine levels from baseline to post-intervention and 2) assess correlations between blood markers and exercise levels. We hypothesize that from baseline to post-intervention 1) TNFα gene promoter methylation will increase and TNFα mRNA expression and TNFα cytokine levels will decrease and 2) TNF gene promoter methylation, TNF gene expression, and TNFα levels will correlate with exercise levels. We present individual- and aggregated-level data to better understand the variability of blood markers in order to plan for future studies.

2. Methods

2.1. Study design, setting, and participants

The primary aim of the single-arm pilot study was to assess the feasibility of a mHealth exercise intervention (GO-EXCAP) in older patients with myeloid 2021 (ClinicalTrials.gov identifier: NCT04035499). Findings on feasibility have previously been reported (Loh et al., 2022). Twenty-five participants were recruited from the University of Rochester Medical Center/Wilmot Cancer Institute in upstate New York between February 2020 and July 2021. The study was approved by the University of Rochester Research Subjects Review Board (STUDY00003945), and all participants provided informed consent.

Inclusion criteria were 1) age 60 years or above; 2) diagnosis of a myeloid neoplasm; 3) receiving outpatient chemotherapy; 4) English-speaking; 5) physician-verified Eastern Cooperative Oncology Group (ECOG) performance status of 0–2; (6) able to walk 4 m; 7) no medical contraindications to exercise as determined by the treating oncologist; and 8) able to provide informed consent. Patients with platelet counts ≤10,000 platelets per microliter on the most recent complete blood count who had not received platelet transfusion were excluded.

2.2. Intervention and study procedures

The GO-EXCAP intervention utilizes a mobile app to deliver a home-based exercise intervention [Exercise for Cancer Patients (EXCAP©®)] over 2 cycles of chemotherapy (approximately 8 to 12 weeks). EXCAP©® consists of two individualized components: progressive walking and resistance training. At baseline, an American College of Sports Medicine (ACSM)-certified exercise physiologist provided an individually tailored exercise prescription to each participant. Baseline step counts were collected 4–7 days before the start of the intervention, and progressive step goals were generated (i.e., weekly increases in average daily step counts by 5 % to 20 %). Participants were also instructed to conduct progressive resistance training by starting with one set of 8–15 repetitions of different upper and lower extremity exercises, then gradually increasing intensity (switching to bands with more resistance) and sets (up to 4 sets of 15). Each participant was provided with a kit which included a printed manual, a Garmin Forerunner 35 activity tracker, and three different resistance-level therapeutic bands as well as a tablet preloaded with a mobile app for communication between participants and the study team during the intervention period.

During the intervention period, participants were encouraged to complete the prescribed daily walking and resistance exercises and to enter exercise and symptoms data into the mobile app. The exercise physiologist monitored the data twice per week, sent individualized motivational messages to participants via the mobile app, and adjusted exercise prescriptions when needed.

2.3. Measures

Demographics, including age, gender, race, ethnicity, marital status, and education, were collected at baseline. We also collected blood sample and clinical measures at both baseline and post-intervention (see below).

2.3.1. Biospecimens

Non-fasting blood samples were collected at baseline and post-intervention. All tubes were collected and processed by a trained laboratory technician and then stored at −80 °C. Blood samples were used for TNFα promoter methylation, mRNA expression, and inflammatory cytokine levels.

2.3.1.1. DNA extraction and methylation.

Genomic DNA was isolated from whole blood collected in EDTA tubes with a DNeasy Blood and Tissue Kit (Qiagen, Catalog #: 69516) and bisulfite converted using the EpiTect Bisulfite Kit (Qiagen, Catalog #: 59104). Total methylation of TNFα promoter was performed using pyrosequencing, a high throughout and quantitative sequencing by synthesis system, with PyroMark Advanced Q24 (Qiagen). DNA methylation levels were measured for seven CpG sites in the TNFα gene promoter methylation region from −249 to −100 base pairs (bp; Supplemental Fig. 2). The primer sequences were as follow: forward primer 5′-ATAGGTTTTGAGGGGTATGG-3′, reverse primer 5′-CTACTAACTAAATATACCAACAACTACC-3′. Sequencing primer was 5′-ATATATAAATTAGTTAGTGGTT-3′ for Cps sites −239 and −245 and 5′-TTTGTGTGTTTTTAATTTTTTAAT-3′ for CpG sites −120, −147, −162, −164, −170 (Campión et al., 2009). The selection of these seven CpG sites was based on their responses to a behavioral intervention and their associations with serum TNFα levels in a previous study (Campión et al., 2009). In addition, the majority of known regulatory transcription factor binding sites (e.g., NF-κB, Sp-1, Egr1) within the TNFα gene promoter region are located in this region (Falvo et al., 2010).

2.3.1.2. Gene expression.

RNA was extracted using MagMAX™ for Stabilized Blood Tubes RNA Isolation Kit, compatible with PAXgene™ Blood RNA Tubes (ThermoFisher). RNA concentration and purity were determined with the NanoDrop 1000 spectrophotometer (NanoDrop, Wilmington, DE), and RNA quality was assessed with the Fragment Analyzer (Agilent, Santa Clara, CA). cDNA was prepared using the High-Capacity cDNA Reverse Transcription kit (ThermoFisher Scientific) following manufacturer’s recommendations. Quantitative RT-PCR was performed on all samples in triplicate using 5 ng of cDNA per 10-ul reaction in 384-well plates with TaqMan™ assays and TaqMan™ Universal Master Mix II, no UNG on the QuantStudio 12K Flex™ Real-Time PCR System (Applied Biosystems, Foster City, CA) with the following cycling parameters: −50 °C for 2 m, 95 °C for 10 m, 40 cycles of 95 °C for 15 s and 60 °C for 1 m. Following amplification, CT values were determined using default settings of the QuantStudio 12 K Flex Software v1.2.3. TaqMan assays used for this analysis were 18S (Hs99999901_s1), HPRT (Hs02800695_m1), and TNFa (Hs00174128_m1).

2.3.1.3. Cytokines/protein expression.

Serum cytokines and their receptors were analyzed on a Luminex Magpix (Luminex Corp., Austin, TX). A median of 50 beaded reactions per well was used to determine concentration per sample (pg/mL). Customized Milliplex xMAP human cytokine and cytokine receptor immunoassay kits (catalog numbers: HSTCMAG-28SK and HSCRMAG-32 K; Luminex Corp.) were used for the study. Cytokines and their receptors analyzed included TNFα, soluble TNF receptor 1 (TNFα sR1), soluble TNF Receptor 2 (TNFα sR2), interferon-gamma IFN-γ, interleukin-1 beta (IL-1β), interleukin-2 (IL-2), soluble interleukin-2 receptor (sIL-2R), interleukin-6 (IL-6), soluble interleukin-6 receptor (sIL-6R), interleukin-8 (IL-8), and interleukin-10 (IL-10). We chose these blood markers based on their significant associations with health outcomes (including physical function, depression, and survival) among older patients with AML as documented in prior literature (Loh et al., 2020). All samples were assayed in duplicates and checked for quality control within reasonable coefficient of variation (≤20 %).

2.3.2. Physical function assessment

Physical function was evaluated using the Short Physical Performance Battery (SPPB). The SPPB consists of three components (i.e., balance, gait speed, and chair stands) with scores ranging from 0 to 12; higher scores indicate better physical function (Gómez et al., 2013; Marsh et al., 2015). A score of 9 or below is considered impaired.

2.3.3. Statistical analyses

Descriptive statistics and box plots were used to describe TNFα promoter methylation, TNFα mRNA expression, and serum inflammatory cytokine levels at baseline and post-intervention. Based on the distribution of change in each blood marker, paired t-tests or Wilcoxon signed-rank tests were used to conduct the comparisons between baseline and post-intervention values. Spearman’s correlation analyses were used to assess relationships between changes in TNFα promoter methylation and TNFα mRNA expression as well as changes in blood marker and exercise levels. We further stratified changes in blood marker from baseline to post-intervention based on changes in average daily steps, average minutes of resistance training, and baseline SPPB (≤9 vs. >9). We applied logarithmic transformations to blood markers with non-normal distributions. To reflect the real changes from baseline to post-intervention at the individual level, all data were visualized on the raw scale without logarithmic transformation. Given the small sample size of this pilot study, hypothesis testing was performed at α = 0.10 (2-tailed) and effect sizes (Cohen’d for group comparisons, r for correlation) were calculated (small effect: 0.20 ≤ d < 0.50 or 0.10 ≤ r <0.30; moderate effect 0.50 ≤ d < 0.80 or 0.30 ≤ r < 0.50; large effect: d ≥ 0.80 or r ≥ 0.50) (Sullivan and Feinn, 2012). We did not correct for multiple testing, given the exploratory nature of the study. All statistical analyses were conducted using SAS statistical software, version 9.4 (SAS Institute Inc., Cary, NC).

3. Results

3.1. Baseline characteristics

A total of 20 patients provided blood samples at baseline and post-intervention and were included in this analysis. Mean age was 71 years (SD: 4.8), 65 % were male, and 60 % had ECOG performance status of 1. Fifty-five percent had AML and 40 % had MDS. Table 1 shows demographic and clinical characteristics.

Table 1.

Demographics and clinical characteristics.

Variables N = 20
Age in years, mean (SD, range) 71.2 (4.8, 62–80)
Gender, n (%) Male 13 (65.0)
Female 7 (35.0)
Race, n (%) White 18 (90.0)
Black or African American 1 (5.0)
Prefer not to say 1 (5.0)
Ethnicity, n (%) Not Hispanic or Latino 19 (95.0)
Prefer not to say 1 (5.0)
Marital status, n (%) Married 13 (65.0)
Divorced or widowed 2 (10.0)
Single 5 (25.0)
Education, n (%) High school or below 2 (10.0)
At least some college 6 (30.0)
College graduate 5 (25.0)
Postgraduate level 6 (30.0)
Prefer not to say 1 (5.0)
Eastern Cooperate Group (ECOG) performance status, n (%) 0 3 (15.0)
1 12 (60.0)
2 5 (25.0)
Diagnosis, n (%) AML 11 (55.0)
MDS 8 (40.0)
MDS/myeloproliferative neoplasm overlap 1 (5.0)
Treatment, n (%) HMA combination treatment (e.g., venetoclax) 11 (55.0)
HMA only 7 (35.0)
Othera 2 (10.0)
Chemotherapy cycle at initiation of intervention, n (%) 1 3 (15.0)
2 9 (45.0)
3 4 (20.0)
≥4 4 (20.0)
Open in a new tab

Abbreviations: AML, acute myeloid leukemia; HMA, hypomethylating agent; MDS, myelodysplastic syndrome.

a

1 received gliteritinib and 1 received low dose cytarabine and venetoclax.

3.2. Relationship between TNFα promoter methylation, TNFα mRNA expression, and cytokine levels at baseline and post-intervention

Fig. 1 and Supplemental Table 1 show the distribution of TNFα promoter methylation status at seven CpG sites, TNFα mRNA expression, and serum TNFα levels, as well as the related cytokines levels. None of the changes were statistically significant except for serum TNFα sR1 levels. From baseline to post-intervention, mean TNFα promoter methylation increased at all seven CpG sites (Cohen d’s ranges from 0.07 to 0.75). Mean mRNA expression (change: −0.14, SD 1.64, p = 0.71; Cohen’d = −0.09) and median serum TNFα levels (change −0.84 pg/mL, IQR 4.36, p = 0.23; Cohen’d = −0.21) decreased from baseline to post-intervention. There was a statistical significant increase in median serum TNFα sR1 levels (median change: 83.63 pg/mL, IQR 130.58, p = 0.06; Cohen’d = −0.18), but not the other cytokines.

Fig. 1.

Fig. 1.

Open in a new tab

Distribution of TNFα promoter methylation status from −120 bp to −245 bp, TNFα mRNA expression, and TNFα levels, as well as the levels of the various cytokines.

Grey diamonds represent the mean values; the top, middle, and lower horizontal lines represent the first quartile, median, and third quartile, respectively. aWilcoxon signed-rank tests were used to compare median level of blood markers at baseline vs. post-intervention. bPaired t-tests were used to compare mean level of blood markers at baseline vs. post-intervention.

3.3. Correlation between TNFα promoter methylation, TNFα mRNA expression, and serum TNFα levels

From baseline to post-intervention, changes in TNFα promoter methylation at all seven CpG sites were negatively correlated with changes in TNFα mRNA expression including several significant relationships (Supplemental Table 2). At the individual patient level, the direction of change in TNFα mRNA expression from baseline to post-intervention was generally the opposite of changes in its gene promoter methylation (i.e., when TNFα promoter methylation at specific CpG sites increased, TNFα mRNA expression decreased or when TNFα promoter methylation at specific CpG sites decreased, TNFα mRNA expression increased).

Changes in TNFα promoter methylation at all seven CpG sites were positively correlated with changes in serum TNFα levels, but these correlations were not statistically significant (Supplemental Table 2). Changes in TNFα mRNA expression were negatively correlated with changes in serum TNFα levels (= −0.51, p = 0.03).

3.4. TNFα promoter methylation, TNFα mRNA expression, cytokines, and exercise levels

Patients walked on average 3289.4 (SD 2056.0, n = 18) daily steps at baseline and 3649.1 (SD 2651.8, n = 18) daily steps at post-intervention. Patients reported performing resistance band exercises for a mean duration of 26.4 (SD 10.21, n = 19) minutes/day, 3.0 (SD 2.3, n = 19) days/week, and they rated their perceived exertion at 3.4 (SD 1.2, n = 18) on a 1–10 Likert scale, indicating low intensity.

Table 2 shows the correlation between changes in TNFα regulations and exercise levels from baseline to post-intervention. Increases in average daily steps were correlated with increases in TNFα promoter methylation at CpG site 147 (r = 0.48, p = 0.06; moderate effect) and −164 (r = 0.51, p = 0.04; large effect). Resistance training minutes were negatively correlated with TNFα promoter methylation at CpG site −120 (r = −0.62, p = 0.02; large effect). Resistance training minutes were positively correlated with change in levels of serum IFN-γ (r = 0.54, p = 0.02; large effect), IL-1β (r = 0.60, p = 0.01; large effect), and IL-2 (r = 0.57, p = 0.01; large effect).

Table 2.

Correlation between blood markers and exercise levels.

Δ Blood markers Δ Average daily steps Resistance training minutes
Δ TNFα promoter methylation at CpG – 120 bpa r = 0.47; p = 0.12
n = 12		 r = - 0.62; p = 0.02*
n = 14
A TNFα promoter methylation at CpG – 147 bpa r = 0.48; p = 0.06*
n = 16		 r = −0.29; p = 0.25
n = 17
Δ TNFα promoter methylation at CpG – 162 bpa r = 0.21; p = 0.41
n = 17		 r = −0.27; p = 0.28
n = 18
Δ TNFα promoter methylation at CpG – 164 bpa r = 0.51; p = 0.04*
n = 17		 r = - 0.18; p = 0.48
n = 18
Δ TNFα promoter methylation at CpG – 170 bpa r = 0.14; p = 0.59
n = 17		 r = − 0.05; p = 0.83
n = 18
Δ TNFα promoter methylation at CpG – 239 bpb r = − 0.15; p = 0.57
n = 17		 r = − 0.32; p = 0.19
n = 18
Δ TNFα promoter methylation at CpG – 245 bpb r = − 0.02; p = 0.94
n = 18		 r = − 0.14; p = 0.57
n = 19
Δ TNFα mRNA expressionb r = − 0.20; p = 0.43
n = 18		 r = - 0.15; p = 0.53
n = 19
Δ TNFαa r = 0.01; p = 0.98
n = 18		 r = 0.08; p = 0.76
n = 18
Δ TNFα sR1a r = − 0.18; p = 0.48
n = 18		 r = − 0.34; p = 0.16
n = 18
Δ TNFα sR2a r = − 0.40; p = 0.10
n = 18		 r = 0.07; p = 0.77
n = 18
Δ IFN-γa r = 0.07; p = 0.78
n = 18		 r = 0.54; p = 0.02*
n = 18
Δ IL-1βb r = − 0.01; p = 0.96
n = 18		 r = 0.60; p = 0.01*
n = 18
Δ IL-2b r = 0.12; p = 0.63
n = 18		 r = 0.57; p = 0.01*
n = 18
Δ sIL-2Ra r = - 0.20; p = 0.42
n = 18		 r = − 0.06; p = 0.80
n = 18
Δ IL-6a r = − 0.22; p = 0.39
n = 18		 r = 0.12; p = 0.62
n = 18
Δ sIL-6Rb r = − 0.31; p = 0.20
n = 18		 r = 0.06; p = 0.82
n = 18
Δ IL-8a r = 0.27; p = 0.29
n = 18		 r = 0.16; p = 0.52
n = 18
Δ IL-10a r = 0.02; p = 0.93
n = 18		 r = 0.04; p = 0.87
n = 18
Open in a new tab
a

log-transformed.

b

raw values in percentages.

*

Statistical significance at p < 0.10.

Figs. 2 and 3 show the changes in TNFα regulation based on median steps and resistance training minutes for selected patients. Patients who had changes in steps at or more than the group median also had greater increases in TNFα promoter methylation status at CpG site −147. They also had lower serum levels of TNFα sR2.

Fig. 2.

Fig. 2.

Open in a new tab

Changes in TNFα promoter methylation at CpG site −147 bp and TNFα sR2 levels based on change in steps from baseline to post-intervention.

aMann-Whitney U tests were used to compare median change from baseline to post-intervention for those who increased their steps ≥ the median value vs. <median.

Fig. 3.

Fig. 3.

Open in a new tab

Changes in TNFα promoter methylation at CpG site −120 bp as well as IFN-γ, IL-1β, and IL-2 based on change in resistance training minutes from baseline to post-intervention.

aMann-Whitney U tests were used to compare median change from baseline to post-intervention for those who increased their resistance training minutes ≥ the median value vs. <median. bTwo-sample t-tests were used to compare mean change from baseline to post-intervention for those who increased their resistance training minutes ≥ the median value vs. <median. # a potential outlier was removed.

3.5. Exercise-induced changes in TNFα regulation stratified by baseline physical function

Supplemental Fig. 3 demonstrates changes in marker levels stratified by baseline SPPB (≤9: impairment vs. >9: no impairment). Patients with SPPB scores of 9 or above at baseline (e.g. better physical function) were more likely to have an increased TNFα promoter methylation at CpG site −239 (8.1 vs. −4.3, p = 0.06) and −245 (8.3 vs. −5.8, p = 0.03) as well as a lower TNFα mRNA expression (−0.9 vs. 0.5, p = 0.05) from baseline to post-intervention. There were no differences in changes in cytokines as stratified by SPPB score.

4. Discussion

In this single-arm pilot study of a mHealth exercise intervention, from baseline to post-intervention, there was no statistical significant changes in TNFα promoter methylation, TNFα mRNA expression, and serum TNFα levels. Absolute TNFα levels decreased with a small effect size. Effect sizes, however, were moderate to large for several CpG sites (i.e., −120, −147, −162, and −164). There was a statistical significant increase in median serum TNFα sR1 levels but effect size was very small. In terms of correlation between TNFα and exercise levels, increases in average daily steps were correlated with increase in TNFα promoter methylation at CpG sites −147 and −164 (moderate to large effect sizes). Resistance training minutes were negatively correlated with TNFα promoter methylation at CpG site −120 and were positively correlated with change in levels of IFN-γ, IL-1β, and IL-2 (all large effect sizes).

Several studies in the general and cancer populations have demonstrated that exercise decreases serum TNFα levels (Khosravi et al., 2019; Meneses-Echávez et al., 2016; Smart et al., 2011; Tsukui et al., 2000). For example, among cancer survivors during or after treatment, two systematic reviews with meta-analyses showed that exercise training decreased TNFα levels (weighted mean difference, −0.64 and −0.3 pg/mL, respectively, in both meta-analyses) (Khosravi et al., 2019; Meneses-Echávez et al., 2016). The relationship between exercise and TNFα sR1 levels are mixed, with some studies showing increased and some showing decreased TNFα sR1 levels in response to exercise (Arroyo et al., 2020; Conraads et al., 2002; Pischon et al., 2003; Pussieldi et al., 2014). In terms of mRNA expression, in a study of healthy volunteers and older adults (n ≤ 30 individuals), exercise reduced TNFα mRNA expression (Macêdo Santiago et al., 2018; Radom-Aizik et al., 2014). While we did not show significant statistical changes in TNFα levels and TNFα mRNA expression in our study, absolute values decreased (−1.56 pg/mL for TNFα levels with a small effect size; −0.18 for TNFα mRNA expression with a very small effect size). Median TNFα sR1 levels were increased (+83.63 pg/mL) with a very small effect size.

Data on how exercise influences TNFα promoter methylation are more limited. In a study involving eight healthy volunteers, acute aerobic exercise did not change TNFα promoter methylation at CpG sites +197, +202, +214, and +222 bps (these CpG sites differ from those assessed in our studies and direction of changes were not reported by the authors) (Hunter et al., 2019). In the Cardiovascular Health Study that included approximately 390 older adults, physical activity was associated with differential methylation of TNFα, although specific details were not reported by the authors (reported in abstract form only) (Shaw et al., 2014). Our study adds to the literature by investigating seven CpG sites in the TNFα promoter methylation region from −249 to −100 base pairs (bp), specifically at CpG sites −120, −147, −162, −164, −170, −239, and −245 bps (Campión et al., 2009). We found that absolute values for TNFα promoter methylation increased. While these changes were not statistically significant, effect sizes were moderate or large for several CpG sites such as −120, −147, −162, and −164, highlighting the need to further study these CpG sites to better understand how exercise influences inflammatory regulation.

When we correlated blood markers and exercise levels, we demonstrated that increases in average daily steps were correlated with increases in TNFα promoter methylation at CpG sites −147 and −164 with moderate to large effect sizes. Resistance training minutes were negatively correlated with TNFα promoter methylation at CpG site −120 with large effect size (in contrary to our hypothesis). Our findings suggest differential changes in TNFα promoter methylation depending on the CpG sites and exercise modality. This adds to the evidence that these regions may be crucial in the epigenetic regulation of TNFα that intervention studies may target at reducing inflammation. Our study adds to the literature on how aerobic and resistance exercises may be associated with changes in TNFα promoter methylation patterns.

Our study has several strengths. First, we included a group of vulnerable older adults with myeloid malignancies. Second, we were able to measure TNFα promoter methylation, TNFα mRNA expression, and TNFα levels, in addition to other inflammatory markers and clinical measures as part of a clinical trial, which provided an opportunity to investigate important relationships among them. Our study is limited, however, by its small sample size, thereby restricting our ability to statistically control for clinical factors which may influence our blood marker of interest, detect statistical associations, and adjust for multiple testing. Despite this, our study is novel, studying TNFα regulation in the context of an exercise trial. In addition, the majority of patients in the study were on hypomethylating agents, which themselves may affect the epigenetic regulation of TNFα.

In conclusion, while we demonstrated no statistical significant changes in TNFα promoter methylation, TNFα mRNA expression, and serum TNFα levels after a mHealth exercise intervention, we noted large effect sizes in several CpG site (i.e., −120, −147, −162, and −164) in the TNFα promoter region. Increases in average daily steps were correlated with increases in TNFα promoter methylation, specifically at CpG sites −147 and −164, and resistance training minutes were negatively correlated with CpG site −120. Further studies are needed to better evaluate TNFα regulation in the context of an exercise intervention. A pilot RCT (ClinicalTrials.gov Identifier: NCT04981821) that compares the mHealth exercise intervention with an education control is currently ongoing to better understand TNFα regulation in the context of exercise. Results may inform interventions to augment TNFα regulation in order to improve outcomes in older adults with cancer.

Supplementary Material

1

Acknowledgement

We thank the Cancer Control Psychoneuroimmunology Lab (CCPL) and Human Biophysiology Shared Resource (HBSR) at Wilmot Cancer Institute for assistance with Luminex profiling and data interpretation and for guidance in study design and protocol development. We thank the University of Rochester Genomics Shared Resource for assistance with gene expression analyses. We wish to acknowledge Susan Rosenthal, MD, for her editorial assistance.

Funding sources

This study was funded by the National Cancer Institute at the National Institutes of Health (K01CA276257 to NK; UG1CA189961 to KM; K99CA237744 and R00CA237744 to KPL), National Institute of Aging (R33AG059206 to HDK), Conquer Cancer Foundation American Society of Clinical Oncology-Walther Cancer Foundation Career Development Award (to KPL), and Wilmot Research Fellowship Award (to KPL). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Footnotes

Human rights

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The study was approved by the University of Rochester Research Subjects Review Board.

Informed consent

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

Welfare of animals

This article does not contain any studies with animals performed by any of the authors.

Transparency statements

This study is registered at ClinicalTrials.gov (identifier: NCT04035499). The analysis plan was not formally pre-registered. Data, analytic code used to conduct the analyses presented in this study, and study materials (intervention protocol, survey instruments and items) are available by emailing the corresponding author.

CRediT authorship contribution statement

Kah Poh Loh: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Writing – original draft, Writing – review & editing. Ying Wang: Data curation, Formal analysis, Visualization, Writing – review & editing. Chandrika Sanapala: Data curation, Project administration, Writing – review & editing. Nikesha Gilmore: Investigation, Writing – review & editing. Colleen Netherby-Winslow: Investigation, Methodology, Resources, Writing – review & editing. Jason H. Mendler: Investigation, Writing – review & editing. Jane Liesveld: Investigation, Writing – review & editing. Eric Huselton: Investigation, Writing – review & editing. AnnaLynn M. Williams: Investigation, Writing – review & editing. Heidi D. Klepin: Conceptualization, Investigation, Writing – review & editing. Marielle Jensen-Battaglia: Data curation, Writing – review & editing, Formal analysis. Karen Mustian: Conceptualization, Investigation, Writing – review & editing. Paula Vertino: Conceptualization, Investigation, Writing – review & editing. Martha Susiarjo: Conceptualization, Investigation, Methodology, Writing – review & editing. Michelle C. Janelsins: Conceptualization, Investigation, Methodology, Resources, Supervision, Writing – review & editing.

Declaration of competing interest

Dr. Loh has served as a consultant to Pfizer and Seagen and has received honoraria from Pfizer. All other authors have no relevant conflicts of interest to report.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.exger.2024.112364.

Data availability

Data will be made available on request.

References

  1. Alibhai SM, Leach M, Kermalli H, Gupta V, Kowgier ME, Tomlinson GA, Brandwein J, Buckstein R, Minden MD, 2007. The impact of acute myeloid leukemia and its treatment on quality of life and functional status in older adults. Crit. Rev. Oncol. Hematol 64, 19–30. [DOI] [PubMed] [Google Scholar]
  2. Alibhai SM, O’Neill S, Fisher-Schlombs K, Breunis H, Brandwein JM, Timilshina N, Tomlinson GA, Klepin HD, Culos-Reed SN, 2012. A clinical trial of supervised exercise for adult inpatients with acute myeloid leukemia (AML) undergoing induction chemotherapy. Leuk. Res 36, 1255–1261. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Arber DA, Orazi A, Hasserjian R, Thiele J, Borowitz MJ, Le Beau MM, Bloomfield CD, Cazzola M, Vardiman JW, 2016. The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood 127, 2391–2405. [DOI] [PubMed] [Google Scholar]
  4. Arroyo E, Laudato JA, Gibson BM, Dulaney CS, Vaughan JA, Followay BN, Glickman EL, Jajtner AR, 2020. Tumor necrosis factor-α, TNF receptor, and soluble TNF receptor responses to aerobic exercise in the heat. Cytokine X 2, 100033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Asfaha S, Dubeykovskiy AN, Tomita H, Yang X, Stokes S, Shibata W, Friedman RA, Ariyama H, Dubeykovskaya ZA, Muthupalani S, Ericksen R, Frucht H, Fox JG, Wang TC, 2013. Mice that express human interleukin-8 have increased mobilization of immature myeloid cells, which exacerbates inflammation and accelerates colon carcinogenesis. Gastroenterology 144, 155–166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Biesmans S, Bouwknecht JA, Ver Donck L, Langlois X, Acton PD, De Haes P, Davoodi N, Meert TF, Hellings N, Nuydens R, 2015. Peripheral administration of tumor necrosis factor-alpha induces neuroinflammation and sickness but not depressive-like behavior in mice. Biomed. Res. Int 2015, 716920. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bluthé RM, Pawlowski M, Suarez S, Parnet P, Pittman Q, Kelley KW, Dantzer R, 1994. Synergy between tumor necrosis factor alpha and interleukin-1 in the induction of sickness behavior in mice. Psychoneuroendocrinology 19, 197–207. [DOI] [PubMed] [Google Scholar]
  8. Brown JC, Harhay MO, Harhay MN, 2015. Physical function as a prognostic biomarker among cancer survivors. Br. J. Cancer 112, 194–198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Campión J, Milagro FI, Goyenechea E, Martínez JA, 2009. TNF-alpha promoter methylation as a predictive biomarker for weight-loss response. Obesity (Silver Spring) 17, 1293–1297. [DOI] [PubMed] [Google Scholar]
  10. Conraads VM, Beckers P, Bosmans J, De Clerck LS, Stevens WJ, Vrints CJ, Brutsaert DL, 2002. Combined endurance/resistance training reduces plasma TNF-alpha receptor levels in patients with chronic heart failure and coronary artery disease. Eur. Heart J 23, 1854–1860. [DOI] [PubMed] [Google Scholar]
  11. Elsea CR, Kneiss JA, Wood LJ, 2015. Induction of IL-6 by cytotoxic chemotherapy is associated with loss of lean body and fat mass in tumor-free female mice. Biol. Res. Nurs 17, 549–557. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Falvo JV, Tsytsykova AV, Goldfeld AE, 2010. Transcriptional control of the TNF gene. Curr. Dir. Autoimmun 11, 27–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gómez JF, Curcio CL, Alvarado B, Zunzunegui MV, Guralnik J, 2013. Validity and reliability of the short physical performance battery (SPPB): a pilot study on mobility in the Colombian Andes. Colomb. Med. (Cali) 44, 165–171. [PMC free article] [PubMed] [Google Scholar]
  14. Goodman MN, 1991. Tumor necrosis factor induces skeletal muscle protein breakdown in rats. Am. J. Phys 260, E727–E730. [DOI] [PubMed] [Google Scholar]
  15. Greiwe JS, Cheng B, Rubin DC, Yarasheski KE, Semenkovich CF, 2001. Resistance exercise decreases skeletal muscle tumor necrosis factor alpha in frail elderly humans. FASEB J 15, 475–482. [DOI] [PubMed] [Google Scholar]
  16. Hunter DJ, James L, Hussey B, Wadley AJ, Lindley MR, Mastana SS, 2019. Impact of aerobic exercise and fatty acid supplementation on global and gene-specific DNA methylation. Epigenetics 14, 294–309. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hurria A, Togawa K, Mohile SG, Owusu C, Klepin HD, Gross CP, Lichtman SM, Gajra A, Bhatia S, Katheria V, Klapper S, Hansen K, Ramani R, Lachs M, Wong FL, Tew WP, 2011. Predicting chemotherapy toxicity in older adults with cancer: a prospective multicenter study. J. Clin. Oncol 29, 3457–3465. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Jin B, Li Y, Robertson KD, 2011. DNA methylation: superior or subordinate in the epigenetic hierarchy? Genes Cancer 2, 607–617. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kaster MP, Gadotti VM, Calixto JB, Santos AR, Rodrigues AL, 2012. Depressive-like behavior induced by tumor necrosis factor-α in mice. Neuropharmacology 62, 419–426. [DOI] [PubMed] [Google Scholar]
  20. Khosravi N, Stoner L, Farajivafa V, Hanson ED, 2019. Exercise training, circulating cytokine levels and immune function in cancer survivors: a meta-analysis. Brain Behav. Immun 81, 92–104. [DOI] [PubMed] [Google Scholar]
  21. Klepin HD, Danhauer SC, Tooze JA, Stott K, Daley K, Vishnevsky T, Powell BL, Mihalko SL, 2011. Exercise for older adult inpatients with acute myelogenous leukemia: a pilot study. J. Geriatr. Oncol 2, 11–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Klepin HD, Geiger AM, Tooze JA, Kritchevsky SB, Williamson JD, Pardee TS, Ellis LR, Powell BL, 2013. Geriatric assessment predicts survival for older adults receiving induction chemotherapy for acute myelogenous leukemia. Blood 121, 4287–4294. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Klepin HD, Tooze JA, Pardee TS, Ellis LR, Berenzon D, Mihalko SL, Danhauer SC, Rao AV, Wildes TM, Williamson JD, Powell BL, Kritchevsky SB, 2016. Effect of intensive chemotherapy on physical, cognitive, and emotional health of older adults with acute myeloid leukemia. J. Am. Geriatr. Soc 64, 1988–1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Klepin HD, Tooze JA, Rejeski J, Mihalko S, Pardee TS, Demark-Wahnefried W, Powell BL, Geiger AM, Kritchevsky S, 2022. Tailoring a physical activity intervention to older adults receiving intensive chemotherapy for acute myeloid leukemia (AML): one size does not fit all. J. Geriatr. Oncol 13, 511–515. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Koo JW, Duman RS, 2009. Evidence for IL-1 receptor blockade as a therapeutic strategy for the treatment of depression. Curr. Opin. Investig. Drugs 10, 664–671. [PMC free article] [PubMed] [Google Scholar]
  26. Lamkin DM, Lutgendorf SK, Lubaroff D, Sood AK, Beltz TG, Johnson AK, 2011. Cancer induces inflammation and depressive-like behavior in the mouse: modulation by social housing. Brain Behav. Immun 25, 555–564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Logan RM, Stringer AM, Bowen JM, Gibson RJ, Sonis ST, Keefe DM, 2008. Serum levels of NFkappaB and pro-inflammatory cytokines following administration of mucotoxic drugs. Cancer Biol. Ther 7, 1139–1145. [DOI] [PubMed] [Google Scholar]
  28. Loh KP, Ramsdale E, Culakova E, Mendler JH, Liesveld JL, O’Dwyer KM, McHugh C, Gilles M, Lloyd T, Goodman M, Klepin HD, Mustian KM, Schnall R, Mohile SG, 2018. Novel mHealth app to deliver geriatric assessment-driven interventions for older adults with cancer: pilot feasibility and usability study. JMIR Cancer 4, e10296. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Loh KP, Tooze JA, Nicklas BJ, Kritchevsky SB, Williamson JD, Ellis LR, Powell BL, Pardee TS, Goyal NG, Klepin HD, 2020. Inflammatory biomarkers, geriatric assessment, and treatment outcomes in acute myeloid leukemia. J. Geriatr. Oncol 11, 410–416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Loh KP, Sanapala C, Watson EE, Jensen-Battaglia M, Janelsins MC, Klepin HD, Schnall R, Culakova E, Vertino P, Susiarjo M, Lin PJ, Mendler JH, Liesveld JL, Huselton EJ, Taberner K, Mohile SG, Mustian K, 2022. A single-arm pilot study of a mobile health exercise intervention (GO-EXCAP) in older patients with myeloid neoplasms. Blood Adv 6, 3850–3860. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Ma X, Does M, Raza A, Mayne ST, 2007. Myelodysplastic syndromes: incidence and survival in the United States. Cancer 109, 1536–1542. [DOI] [PubMed] [Google Scholar]
  32. Macêdo Santiago L, Neto LGL, Borges Pereira G, Leite RD, Mostarda CT, de Oliveira Brito Monzani J, Sousa WR, Rodrigues Pinheiro, A.J. M, Navarro F, 2018. Effects of resistance training on immunoinflammatory response, TNF-alpha gene expression, and body composition in elderly women. J. Aging Res 2018, 1467025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Marsh AP, Wrights AP, Haakonssen EH, Dobrosielski MA, Chmelo EA, Barnard RT, Pecorella A, Ip EH, Rejeski WJ, 2015. The virtual short physical performance battery. J. Gerontol. A Biol. Sci. Med. Sci 70, 1233–1241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Meneses-Echávez JF, Correa-Bautista JE, González-Jiménez E, Schmidt Río-Valle J, Elkins MR, Lobelo F, Ramírez-Vélez R, 2016. The effect of exercise training on mediators of inflammation in breast cancer survivors: a systematic review with meta-analysis. Cancer Epidemiol. Biomark. Prev 25, 1009–1017. [DOI] [PubMed] [Google Scholar]
  35. Palin K, McCusker RH, Strle K, Moos F, Dantzer R, Kelley KW, 2008. Tumor necrosis factor-alpha-induced sickness behavior is impaired by central administration of an inhibitor of c-jun N-terminal kinase. Psychopharmacology 197, 629–635. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Parameswaran N, Patial S, 2010. Tumor necrosis factor-α signaling in macrophages. Crit. Rev. Eukaryot. Gene Expr 20, 87–103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Pischon T, Hankinson SE, Hotamisligil GS, Rifai N, Rimm EB, 2003. Leisure-time physical activity and reduced plasma levels of obesity-related inflammatory markers. Obes. Res 11, 1055–1064. [DOI] [PubMed] [Google Scholar]
  38. Pussieldi GA, Gomes EC, Veneroso CE, De Paz JA, Fonseca TR, Mendes TT, Rossi AG, Teixeira MM, Teixeira AL, Alessandri AL, 2014. Soluble tumour necrosis factor receptor-1 (sTNFR1) levels are positively associated with exercise intensity in athletes after strenuous off-road cycling. J. Sports Med. Phys. Fitness 54, 225–231. [PubMed] [Google Scholar]
  39. Radom-Aizik S, Zaldivar FP Jr., Haddad F, Cooper DM, 2014. Impact of brief exercise on circulating monocyte gene and microRNA expression: implications for atherosclerotic vascular disease. Brain Behav. Immun 39, 121–129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Saad M, Loh KP, Tooze JA, Pardee TS, Ellis LR, Powell B, Bhave RR, Geiger A, Williamson JD, Kritchevsky S, Klepin HD, 2020. Geriatric assessment and survival among older adults receiving post-remission therapy for acute myeloid leukemia. Blood 136 (23), 2715–2719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Sedgwick JD, Riminton DS, Cyster JG, Körner H, 2000. Tumor necrosis factor: a master-regulator of leukocyte movement. Immunol. Today 21, 110–113. [DOI] [PubMed] [Google Scholar]
  42. Shallis RM, Wang R, Davidoff A, Ma X, Zeidan AM, 2019. Epidemiology of acute myeloid leukemia: recent progress and enduring challenges. Blood Rev 36, 70–87. [DOI] [PubMed] [Google Scholar]
  43. Shaw B, Wc L, Hs T, Al F, Saxton J, Belshaw N, 2014. A Change in Physical Activity Level Affects Leukocyte DNA Methylation of Genes Implicated in Cardiovascular Disease in the Elderly
  44. Smart NA, Larsen AI, Le Maitre JP, Ferraz AS, 2011. Effect of exercise training on interleukin-6, tumour necrosis factor alpha and functional capacity in heart failure. Cardiol. Res. Pract 2011, 532620. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Smith LB, Leo MC, Anderson C, Wright TJ, Weymann KB, Wood LJ, 2014. The role of IL-1β and TNF-α signaling in the genesis of cancer treatment related symptoms (CTRS): a study using cytokine receptor-deficient mice. Brain Behav. Immun 38, 66–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sullivan GM, Feinn R, 2012. Using effect size-or why the P value is not enough. J. Grad. Med. Educ 4, 279–282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Sullivan KE, Reddy AB, Dietzmann K, Suriano AR, Kocieda VP, Stewart M, Bhatia M, 2007. Epigenetic regulation of tumor necrosis factor alpha. Mol. Cell. Biol 27, 5147–5160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Tinsley SM, Sutton SK, Thapa R, Lancet J, McMillan SC, 2017. Treatment choices: a quality of life comparison in acute myeloid leukemia and high-risk myelodysplastic syndrome. Clin. Lymphoma Myeloma Leuk. 17s, S75–s79. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Tsukui S, Kanda T, Nara M, Nishino M, Kondo T, Kobayashi I, 2000. Moderate-intensity regular exercise decreases serum tumor necrosis factor-alpha and HbA1c levels in healthy women. Int. J. Obes. Relat. Metab. Disord 24, 1207–1211. [DOI] [PubMed] [Google Scholar]
  50. Uddin M, Koenen KC, Aiello AE, Wildman DE, de los Santos R, Galea S, 2011. Epigenetic and inflammatory marker profiles associated with depression in a community-based epidemiologic sample. Psychol. Med 41, 997–1007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Wyatt G, Sikorskii A, Tesnjak I, Victorson D, Srkalovic G, 2015. Chemotherapy interruptions in relation to symptom severity in advanced breast cancer. Support Care Cancer 23, 3183–3191. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

1
NIHMS1970000-supplement-1.docx (10MB, docx)

Data Availability Statement

Data will be made available on request.

RESOURCES