Article Figures & Data
Figures
- Figure 1.
Flow diagram of the selection procedure of studies of the two literature searches. This figure contains a PRISMA flow diagram that describes the inclusion and exclusion of studies included in this review.
- Figure 2.
Estimated SMD comparing study-defined categories of high and low levels of physical activity. This figure presents a forest plot that describes the pooled SMD in DNA methylation between high and low study-defined categories of physical activity.
- Figure 3.
Estimated relative risk of breast cancer comparing study-defined categories of high and low levels of global DNA methylation. This figure presents a forest plot that describes the pooled relative risk of breast cancer between study-defined categories of DNA methylation.
Tables
- Table 1.
Characteristics of studies investigating the association between physical activity and global DNA methylation (n = 12)
Intervention studies (n = 4) Study Country Study design Study population Sample size (n) Mean age (years) Female (%) Duration of intervention Intervention Methylation measure Tissue Barres (2012; ref. 30) Ireland Crossover Healthy individuals 14 25.0 N/A Acute (single bout) Aerobic LUMA Muscle Duggan (2014; ref. 32) USA Randomized controlled trial Postmenopausal overweight women 300 57.9 100.0 Chronic (12 months) Aerobic LINE1 WBC Delgada-Cruzata (2015; ref. 31) USA Crossover Minority breast cancer survivors 24 52.2 100.0 Chronic (6 months) Aerobic + resistance + diet LINE1, Sat2, LUMA WBC King-Himmelreich (2016; ref. 34) Germany Nonrandomized controlled trial Untrained individuals with a leg injury 30 46.1 66.7 Acute (1 month) Aerobic + resistance LINE1 WBC Observational studies (n = 8) Study Country Study design Study population Sample size (n) Mean age (years) Female (%) Duration of exposure Exposure self-reported Methylation measure Tissue McGuinness (2011; ref. 37) Scotland Cross-sectional Healthy individuals 239 49.8 51.0 Chronic Yes Global WBC Zhang (2011; ref. 41) USA Cross-sectional Healthy individuals 161 53.0 62.6 Acute No LINE1 WBC Gomes (2012; ref. 33) Brazil Cross-sectional Healthy older adults 126 70.8 52.4 Acute No Global WBC Morabia (2012; ref. 38) USA Case–control Healthy individuals 180 22.7 36.0 Chronic Yes LINE1 WBC Luttropp (2013; ref. 35) Sweden Cross-sectional Healthy older individuals 1016 70+ N/A Chronic Yes LUMA WBC White (2013; ref. 40) USA and Puerto Rico Cross-sectional Women with a family history of breast cancer 647 54.8 100.0 Chronic Yes LINE1 Whole Blood Perng (2014; ref. 39) USA Cross-sectional Healthy individuals 987 62.4 52.4 Chronic Yes LINE1, Alu WBC Marques-Rocha (2016; ref. 36) Brazil Cross-sectional Healthy individuals 156 23.1 58.3 Chronic Yes LINE1 WBC - Table 2.
Assessment of heterogeneity among reports of the association between physical activity and global DNA methylation
Subgroup Estimates (n) SMD (95% CI); P I2 Meta-regression P Study demographics Mean age <50 4 −0.10 (−0.51–0.31); P = 0.50 84.9% 0.25 50–<60 6 0.37 (−0.04–0.78); P = 0.08 88.4% 60+ 4 0.21 (−0.15–0.57); = 0.25 87.2% % Female <60 5 0.36 (0.02–0.70); P = 0.04 86.1% 0.07 60+ 7 0.00 (−0.12–0.12); P = 0.99 24.6% Country North American 8 0.26 (−0.03–0.55); P = 0.08 85.9% 0.73 Other 6 0.09 (−0.28–0.46); P = 0.64 87.4% Physical activity measure Study design Intervention 6 0.25 (−0.27–0.77); P = 0.34 92.2% 0.66 Observational 8 0.12 (−0.06–0.31); P = 0.19 73.4% Measurement type Self-reported 6 0.16 (−0.07–0.39); P = 0.17 80.7% 0.99 Othera 8 0.19 (−0.21–0.59); P = 0.34 89.4% Duration of physical activityb Long-term 10 0.29 (0.04–0.54); P = 0.02 87.8% 0.11 Acute 4 −0.12 (−0.59–0.36); P = 0.63 83.7% DNA methylation assessment Genomic context Global 2 −0.09 (−0.29–0.12); P = 0.42 0.00% 0.61 Repetitive elements 9 0.25 (0.00–0.49); P = 0.048 85.4% LUMA 3 0.15 (−0.84–1.14); P = 0.77 94.2% Assay type Non-LUMA 11 0.18 (−0.02–0.39); P = 0.08 82.9% 0.92 LUMA 3 0.15 (−0.84–1.14); P = 0.77 94.2% Tissue Blood 13 0.25 (0.05–0.45); P = 0.02 84.2% 0.01 Muscle 1 −0.90 (−1.38 to −0.42); P < 0.01 — ↵aOther includes intervention studies and observational studies with objective assessments via accelerometry.
↵bLong-term physical activity exposure was defined as a physical activity intervention that lasted 6 or more months in duration or an observational study employing self-reported physical activity levels reflective of typical levels in the past year. Acute physical activity exposure was defined as a physical activity intervention lasting less than 6 months or an observational study that assessed levels of physical activity over periods shorter than 2 months using objective measures.
- Table 3.
Characteristics of studies investigating the association between global DNA methylation and breast cancer risk (n = 12)
Prospective studies (n = 5) Study Country Study design Study population Sample size (cases/controls) Age (years)a Blood timing Methylation assay Methylation measure Tissue Brennan (2012; ref. 42) United Kingdom Nested case–control BGS 242/241 54/54 N/R Pyrosequencing LINE1 WBC Italy EPIC 263/232 52/52 Australia and New Zealand KConFab 218/153 50/60 Deroo (2014; ref. 43) USA and Puerto Rico Case–cohort Sister study 294/646 <60 1.3 years avg Pyrosequencing LINE1 Whole blood Severi (2014; ref. 44) Australia Nested case–control MCCS 420/420 64/64 50% > 8.9 years Illumina 450k Betas Whole blood van Veldhoven (2015; ref. 45) Italy Nested case–control EPIC 166/166 54.4/54.2 50% > 3.8 years Illumina 450k Betas WBC Norway NOWAC 192/192 55.4/55.4 50% > 2.1 years Illumina 450k United Kingdom BGS 548/548 52.0/52.0 N/R WGBS Sturgeon (2017; ref. 46) USA Nested case–control PLCO (control arm of intervention study) 428/419 60+ 68.6% > = 4 years %5-mdC Global WBC Nonprospective studies (n = 7) Study Country Study design Study population Sample size (cases/controls) Age (years) a Blood timing Methylation assay Methylation measure Tissue Choi (2009; ref. 47) USA Family-based case–control American women 176/173 < 60 At diagnosis 5-mdC Global WBC Cho (2010; ref. 48) Turkey Case–control Turkish women 40/40 50.8/48.3 At diagnosis MethylLight (%) LINE1 WBC ALU SAT2 Wu (2012; ref. 49) USA Family-based case–control Sisters discordant for breast cancer 266/334 49.5/48.0 At diagnosis MethylLight (%) LINE1 WBC ALU SAT2 Xu (2012; ref. 50) USA Population based case–control American women (LIBCSP) 1064/1101 60+ At diagnosis LUMA (% methylation) LUMA WBC Pyrosequencing LINE1 Kitkumthorn (2012; ref. 51) Thailand Population-based case–control Women from Thailand 36/144 50.28/47.72 At diagnosis COBRA (%) LINE1 WBC Delgado-Cruzata (2012; ref. 52) USA Family-based case–control Sisters discordant for breast cancer 263/321 49.6/48.2 At diagnosis LUMA (% methylation) LUMA WBC 3H-methyl Global Kuchiba (2014; ref. 53) Japan Hospital-based case–control Japanese women 384/384 53.9/54.1 At diagnosis LUMA (% methylation) LUMA WBC ↵aAverage age for cases/controls if available and age range if not available.
- Table 4.
Assessment of heterogeneity among studies examining the association between global DNA methylation and breast cancer risk
Subgroup Estimates (n) RR (95% CI); P I2 Meta-regression P Study demographics Mean age <50 5 0.83 (0.58–1.19); P = 0.31 0.0% 0.08 50–<60 5 0.47 (0.32–0.69); P = 0.02 64.7% 60+ 4 1.01 (0.54–1.90); P = 0.97 92.0% Country North American 10 0.83 (0.55–1.25); P = 0.37 86.6% 0.14 Other 4 0.47 (0.28–0.79); P = 0.004 67.9% Study design Type Prospectivea 5 0.62 (0.43–0.87); P = 0.007 58.1% 0.49 Case–control 9 0.77 (0.45–1.32); P = 0.34 90.1% Blood draw type >3 years prior to diagnosis 3 0.52 (0.28–0.97); P = 0.04 71.5% 0.37 <3 years prior to diagnosis 11 0.77 (0.49–1.20); P = 0.24 88.8% DNA methylation assessment Genomic context Illumina 450k 3 0.53 (0.27–1.06); P = 0.07 69.8% 0.26 Global 3 0.59 (0.33–1.04); P = 0.07 69.9% Repetitive elements 6 0.84 (0.63–1.13); P = 0.25 54.0% LUMA 4 1.02 (0.28–3.76); P = 0.98 96.4% Assay type Non-LUMA 12 0.64 (0.50–0.83); P < 0.001 58.3% 0.25 LUMA 3 1.02 (0.28–3.76); P = 0.98 96.4% ↵aIncludes nested case–control studies.
Supplementary Data
- Table S1 - Sensitivity analysis assessing the impact of mis-specification of intra-individual correlation of pre- and post-test global DNA methylation measures
- Figure S1 - Supplementary Figure 1 presents a funnel plot of studies investigating the association between physical activity and DNA methylation. This plot is used to assess the presence of publication bias.
- Figure S2 - Supplementary Figure 2 presents a funnel plot of studies investigating the association between DNA methylation and breast cancer. This plot is used to assess the presence of publication bias.
Volume 27, Issue 11
