Skip to main content
. 2022 May 11;148(10):2723–2742. doi: 10.1007/s00432-022-04019-9

Table 4.

Multivariable-adjusted HRsa and 95%-CIs for associations between adiposity measures and CRC for individual mutations and MMR status, by sex and tumor location; NLCS, 1986–2006

Person-years at risk KRASmut PIK3CAmutb BRAFmutb dMMRb
ncases HR (95% CI) ncases HR (95% CI) ncases HR (95% CI) ncases HR (95% CI)
BMI quartiles (kg/m2): range (median)
Men—colon
 < 23.4 (22.2) 7993 61 1.00 (ref.) 27 1.00 (ref.) 25 1.00 (ref.) 19 1.00 (ref.)
23.4–24.9 (24.2) 8343 58 0.89 (0.60–1.32) 38 1.37 (0.81–2.30) 28 1.11 (0.62–1.96) 13 0.68 (0.33–1.42)
25.0–26.6 (25.7) 7683 69 1.12 (0.76–1.65) 42 1.66 (0.98–2.81) 29 1.22 (0.68–2.20) 21 1.21 (0.63–2.34)
 > 26.6 (27.8) 7003 68 1.30 (0.88–1.92) 43 1.97 (1.17–3.32) 23 1.14 (0.61–2.14) 17 1.17 (0.59–2.31)
p-trend 0.112 0.007 0.603 0.378
per 5 kg/m2 31,022 256 1.25 (1.00–1.57) 150 1.38 (1.05–1.82) 105 1.23 (0.87–1.72) 70 1.51 (1.01–2.26)
Men—rectum
 < 23.4 (22.2) 7993 18 1.00 (ref.)
23.4–24.9 (24.2) 8343 13 0.67 (0.31–1.44)
25.0–26.6 (25.7) 7683 19 1.06 (0.53–2.15)
 > 26.6 (27.8) 7003 18 1.21 (0.59–2.47)
p-trend 0.415
per 5 kg/m2 31,022 68 1.17 (0.79–1.73)
Women—colon
 < 22.8 (21.5) 9014 52 1.00 (ref.) 30 1.00 (ref.) 53 1.00 (ref.) 43 1.00 (ref.)
22.8–24.7 (23.8) 8914 46 0.88 (0.58–1.35) 27 0.89 (0.52–1.53) 40 0.76 (0.49–1.19) 34 0.78 (0.49–1.26)
24.8–27.0 (25.7) 8141 65 1.48 (0.99–2.20) 30 1.08 (0.64–1.84) 39 0.82 (0.52–1.29) 24 0.61 (0.36–1.03)
 > 27.0 (29.2) 8158 59 1.37 (0.90–2.08) 29 1.01 (0.60–1.73) 41 0.90 (0.57–1.41) 30 0.78 (0.48–1.30)
p-trend 0.031 0.798 0.678 0.221
per 5 kg/m2 34,228 222 1.31 (1.10–1.57)* 116 1.09 (0.84–1.42) 173 0.99 (0.81–1.22) 131 0.90 (0.70–1.15)
Women—rectum
 < 22.8 (21.5) 9014 18 1.00 (ref.)
22.8–24.7 (23.8) 8914 7 0.40 (0.17–0.98)
24.8–27.0 (25.7) 8141 9 0.57 (0.24–1.33)
 > 27.0 (29.2) 8158 21 1.46 (0.72–2.96)
p-trend 0.312
per 5 kg/m2 34,228 55 1.21 (0.87–1.67)
Clothing size: range (median)
Men—colon
 ≤ 50 (50) 10,903 73 1.00 (ref.) 40 1.00 (ref.) 30 1.00 (ref.) 18 1.00 (ref.)
52 (52) 9750 84 1.26 (0.89–1.78) 52 1.45 (0.94–2.24) 29 1.04 (0.61–1.78) 23 1.37 (0.71–2.64)
54 (54) 5156 48 1.33 (0.88–2.02) 22 1.20 (0.69–2.09) 21 1.40 (0.77–2.55) 12 1.37 (0.64–2.95)
 ≥ 56 (56) 2619 29 1.63 (1.00–2.65) 17 1.80 (0.98–3.31) 9 1.19 (0.54–2.61) 7 1.55 (0.63–3.81)
p-trend 0.040 0.094 0.360 0.280
per 2 sizes 28,428 234 1.24 (1.01–1.54) 131 1.31 (1.01–1.70) 89 1.18 (0.85–1.64) 60 1.33 (0.90–1.96)
Men—rectum
 ≤ 50 (50) 10,903 21 1.00 (ref.)
52 (52) 9750 17 0.90 (0.45–1.77)
54 (54) 5156 17 1.71 (0.86–3.40)
 ≥ 56 (56) 2619 9 1.78 (0.78–4.06)
p-trend 0.073
per 2 sizes 28,428 64 1.17 (0.80–1.73)
Women—colon
 ≤ 40 (40) 6574 35 1.00 (ref.) 24 1.00 (ref.) 29 1.00 (ref.) 32 1.00 (ref.)
42 (42) 8582 54 1.13 (0.72–1.78) 25 0.75 (0.42–1.34) 47 1.23 (0.75–2.03) 35 0.82 (0.49–1.37)
44 (44) 9270 58 1.10 (0.70–1.72) 33 0.84 (0.48–1.45) 45 1.02 (0.62–1.68) 25 0.51 (0.29–0.88)
 ≥ 46 (46) 9454 70 1.33 (0.86–2.05) 30 0.74 (0.42–1.30) 49 1.11 (0.67–1.83) 37 0.76 (0.45–1.28)
p-trend 0.229 0.423 0.957 0.174
per 2 sizes 33,880 217 1.26 (1.03–1.53) 112 1.05 (0.82–1.35) 170 1.04 (0.84–1.27) 129 0.90 (0.71–1.14)
Women—rectum
 ≤ 40 (40) 6574 10 1.00 (ref.)
42 (42) 8582 9 0.65 (0.26–1.62)
44 (44) 9270 12 0.76 (0.32–1.84)
 ≥ 46 (46) 9454 23 1.41 (0.63–3.13)
p-trend 0.243
per 2 sizes 33,880 54 1.07 (0.77–1.51)

HR hazard ratio; CI confidence interval; CRC colorectal cancer; (d/p)MMR mismatch repair (deficient/proficient); NLCS Netherlands Cohort Study; BMI body mass index; p-het p-heterogeneity

*Statistically significant p-heterogeneity, p = 0.008 (reference group: wild-type for KRAS, PIK3CA, and BRAF, and pMMR). Note: other p-heterogeneity tests were not statistically significant

aHazard Ratios were adjusted for age (years; continuous), non-occupational physical activity (minutes/day; continuous), total energy intake (kcal/day; continuous), family history of CRC (yes/no), alcohol consumption (0; 0.1–4; 5–14; > 15 g/day), processed meat intake (g/day; continuous), red meat intake (g/day; continuous). Age was included as a time-varying covariate. BMI models were additionally adjusted for height (cm; continuous)

bAnalyses for subgroups with < 50 cases were not performed