Skip to main content
. 2016 Nov 21;6:237. doi: 10.3389/fonc.2016.00237

Table 4.

Interactions between herbicides and SNPs with prostate cancer risk (p < 0.01).

Exposure
None
Low exposure
High exposure
SNP Pesticide Genotype Ca/Co REF Ca/Co OR (95% CI) Ca/Co OR (95% CI) p-inta q-valueb
Hormonal pathway Gene
rs8192166 SRD5A1 Dicamba CC 133/173 1.0 64/131 0.62 (0.41, 0.93) 49/142 0.44 (0.29, 0.68) 4.0 × 10−5 0.03
CT + TT 189/396 1.0 108/229 0.95 (0.69, 1.30) 127/218 1.15 (0.85, 1.57)
rs3798577 ESR1 Butylate TT 162/256 1.0 10/42 0.40 (0.19, 0.83) 8/42 0.30 (0.14, 0.66) 8.0 × 10−5 0.12
CT + CC 338/647 1.0 42/110 0.74 (0.50, 1.08) 64/97 1.28 (0.90, 1.80)
Hormone GWAS Affected hormone
rs4784336 SHBG Dicamba AA 258/473 1.0 130/273 0.87 (0.65, 1.15) 152/274 1.00 (0.76, 1.32) 3.7 × 10−3 0.90
AC + CC 66/98 1.0 42/87 0.65 (0.39, 1.08) 24/87 0.36 (0.20, 0.65)
rs1017993 DHEAS Alachlor CC 182/406 1.0 135/282 1.07 (0.81, 1.41) 147/273 1.21 (0.92, 1.58) 9.2 × 10−3 1.00
CT + TT 95/140 1.0 65/106 0.92 (0.61, 1.40) 47/113 0.60 (0.39, 0.93)

ORs adjusted for age and state.

aStrata with less than five observations per cell or qualitative interactions excluded. Interactions that were statistically significant (p-int < 0.01) and for which there was a significant association between the pesticide and SNP (p < 0.05) with a monotonic pattern in at least one genotype stratum are included in table.

bThe q-value adjusts the p-value for multiple statistical comparisons using the false discovery rate (FDR) method.