Skip to main content
. 2020 Nov 16;5(1):pkaa109. doi: 10.1093/jncics/pkaa109

Table 2.

Association (RRs, 95% CIs) between 3 methylation-based measures of aging (per 1 SD) and cancer risk in the Melbourne Collaborative Cohort Study

Cancer type Cases, No. PhenoAge
GrimAge
Telomere length
Model RR (95% CI) P RR (95% CI) P RR (95% CI) P
Colorectal cancer 813 Model 1a 1.22 (1.10 to 1.36) <.001 1.20 (1.07 to 1.34) .001 0.98 (0.88 to 1.09) .68
Model 2b 1.22 (1.09 to 1.36) <.001 1.19 (1.03 to 1.36) .02 0.99 (0.89 to 1.10) .90
Model 3c 1.18 (1.05 to 1.32) .01 1.12 (0.96 to 1.30) .15 1.02 (0.90 to 1.14) .78
Gastric cancer 165 Model 1 0.95 (0.77 to 1.18) .65 1.03 (0.83 to 1.27) .80 1.17 (0.92 to 1.48) .19
Model 2 0.96 (0.76 to 1.22) .74 1.05 (0.78 to 1.41) .74 1.25 (0.96 to 1.63) .10
Model 3 0.88 (0.67 to 1.15) .34 0.95 (0.68 to 1.33) .75 1.19 (0.86 to 1.64) .29
Kidney cancer 139 Model 1 1.25 (0.96 to 1.63) .09 1.27 (0.98 to 1.65) .07 1.07 (0.80 to 1.43) .65
Model 2 1.28 (0.94 to 1.76) .12 1.32 (0.91 to 1.91) .15 1.11 (0.78 to 1.57) .57
Model 3 1.25 (0.88 to 1.77) .21 1.28 (0.84 to 1.95) .25 1.19 (0.79 to 1.79) .40
Lung cancer 327 Model 1 1.23 (1.06 to 1.44) .007 1.81 (1.45 to 2.26) <.001 0.90 (0.76 to 1.06) .19
Model 2 1.23 (1.05 to 1.45) .01 1.82 (1.44 to 2.30) <.001 0.88 (0.74 to 1.04) .13
Model 3 1.25 (1.05 to 1.49) .01 2.03 (1.56 to 2.64) <.001 0.88 (0.73 to 1.06) .19
Mature B-cell neoplasms 423 Model 1 1.24 (1.07 to 1.43) .003 0.95 (0.81 to 1.11) .49 0.92 (0.81 to 1.05) .24
Model 2 1.27 (1.09 to 1.47) .002 0.96 (0.78 to 1.17) .66 0.91 (0.79 to 1.05) .20
Model 3 1.23 (1.04 to 1.45) .02 1.03 (0.82 to 1.28) .81 0.95 (0.81 to 1.13) .57
Prostate cancer 846 Model 1 0.98 (0.88 to 1.08) .68 0.88 (0.79 to 0.98) .02 1.06 (0.95 to 1.18) .28
Model 2 0.99 (0.89 to 1.10) .85 0.88 (0.76 to 1.01) .07 1.05 (0.94 to 1.17) .43
Model 3 1.00 (0.89 to 1.11) .96 0.84 (0.72 to 0.98) .02 1.06 (0.94 to 1.20) .35
Urothelial cancers 404 Model 1 1.21 (1.05 to 1.40) .01 1.39 (1.19 to 1.61) <.001 0.90 (0.77 to 1.04) .14
Model 2 1.17 (1.00 to 1.36) .05 1.22 (1.00 to 1.48) .05 0.95 (0.81 to 1.10) .48
Model 3 1.16 (0.99 to 1.37) .07 1.22 (0.98 to 1.52) .08 0.92 (0.78 to 1.09) .33
All types 2994d Model 1 1.14 (1.08 to 1.20) <.001 1.13 (1.06 to 1.19) <.001 0.98 (0.93 to 1.03) .46
Model 2 1.13 (1.07 to 1.19) <.001 1.12 (1.05 to 1.20) .001 0.98 (0.93 to 1.04) .58
Model 3 1.11 (1.05 to 1.18) <.001 1.11 (1.03 to 1.20) .01 1.00 (0.94 to 1.06) 1.00
a

Model 1: No adjustment other than that provided by the matching variables age, sex, country of birth (Australia, northern Europe, or southern Europe), sample type (peripheral blood mononuclear cells, dried blood spots, or buffy coats); lung cancer study: additional matching for smoking status (never; former, quitting less than 10 years; former, quitting 10 years and over; current, smoking less than 15 cigarettes per day; current smoking 15 or more cigarettes per day). CI = confidence interval; RR = rate ratio.

b

Model 2: Additional adjustment for smoking (current, former, or never), smoking pack-years, age at starting smoking (4 categories), time since quitting smoking (4 categories), body mass index (in kg/m2), height (in meters), alcohol consumption (in grams per day), physical activity (categorized score), dietary quality (24), socioeconomic status score (at the local area level, ranging from 1 to 10), education score (ordinal variable ranging from 1, primary school to 8, postgraduate degree).

c

Model 3: Model 2 + additional adjustment for white blood cell proportions estimated using the Houseman algorithm.

d

For the combined analysis, where an individual was diagnosed with several cancers, we included the first diagnosis only (respecting the incidence density sampling procedure), so that participants did not contribute twice to the pooled estimate, resulting in 2994 pairs out of 3117 in total.