Table 1.

SNP with smallest P for interaction with total, dietary, or supplemental calcium for colorectal cancer risk

Interaction resultsb,c
Calcium analysisSNP with smallest P for interactionLocusPosition (bp)aFunction classGenetic regionMinor alleleAlt alleleMAFMean RsqVariableOR-int (95% CI)PPhet
Totalrs19337556q23.1130925767IntergenicTMEM200A/EPB41L2CT0.100.93Total0.84 (0.78–0.90)1.5E−064.1E−01
Dietary0.86 (0.80–0.92)5.2E−053.2E−01
Suppl0.81 (0.65–1.02)7.4E−025.5E−01
Dietaryrs68558854q22.192039007IntronicFAM190AAG0.500.94Total1.09 (1.04–1.13)9.0E−052.5E−01
Dietary1.11 (1.06–1.16)1.9E−065.5E−01
Suppl0.93 (0.81–1.07)3.0E−012.4E−01
Supplementalrs10281664q34.3182813298IntergenicAGA/TENM3GA0.310.85Total1.07 (1.02–1.12)6.7E−033.8E−01
Dietary1.02 (0.97–1.07)5.2E−017.4E−01
Suppl1.49 (1.27–1.74)7.3E−072.9E−01

Abbreviations: Alt, alternate; MAF, minor allele frequency; OR-int, odds ratio for interaction; Phet, P for heterogeneity across studies; Rsq, imputation Rsq.

  • aOn the basis of NCBI build 37 data.

  • bCorresponds to each additional copy of the minor allele (i.e., assuming additive genetic effects) and each quartile increase in calcium intake (total, dietary) or ≥500 versus <500 mg/d (supplemental). Genotyped SNPs were modeled as 0, 1, or 2 copies of the minor allele; imputed SNPs were modeled as the expected number of copies of the minor allele (the genotype “dosage”; refs. 3, 7).

  • cOn the basis of multivariable logistic regression adjusted for age (continuous), sex (F/M), study center (indicators), energy consumption (continuous), first 3 principal components of genetic ancestry (continuous), SNP main effect, and calcium main effect.