Characteristics of Subjects
The study included 403 patients (156 females and 247 males) and 419 cancer free controls (201 females and 218 males), with significant differences in the mean age (59.51±11.37 vs. 58.07±10.86, P=0.02). The distribution of overweight individuals (body mass index [BMI] ≥23.00kg/m2) in the patient group was lower than in the control group (52.51% vs. 64.05%, P<0.01). A lower proportion of patients had completed their education to a level greater than high school than the controls (18.49% vs. 26.98%, P<0.01) and the percentage of white collar jobs among patients was less than controls (8.15% vs. 18.98%) (Table 1).
TABLE 1 Basic Demographic Characteristics of Patients and Controls
Characteristics
|
Number of Cases (%)
|
Number of Controls (%)
|
P value
|
(n=403)
|
(n=419)
|
Gender
|
Male
|
247(61.29)
|
218(52.03)
|
<0.01
|
|
Female
|
156(38.71)
|
201(47.97)
|
|
Age (years)*
|
Mean(SD)
|
59.51(11.37)
|
58.07(10.86)
|
0.02
|
|
≤50
|
69(18.06)
|
88(21.00)
|
0.09
|
|
50-
|
121(31.68)
|
135(32.22)
|
|
|
60-
|
106(27.75)
|
130(31.03)
|
|
|
≥70
|
86(22.51)
|
66(15.75)
|
|
BMI (kg/m2)#*
|
≤18.50
|
33(8.29)
|
19(4.65)
|
<0.01
|
|
18.5-
|
156(39.20)
|
128(31.30)
|
|
|
≥23.00
|
209(52.51)
|
262(64.05)
|
|
Education*
|
No Formal Education
|
107(29.97)
|
79(19.04)
|
<0.01
|
|
Primary
|
104(29.13)
|
115(27.71)
|
|
|
Junior
|
80(22.41)
|
109(26.27)
|
|
|
Senior and above
|
66(18.49)
|
112(26.98)
|
|
Occupational*
|
White Color Jobs
|
30(8.15)
|
78(18.98)
|
<0.01
|
|
Others
|
338(91.85)
|
333(81.02)
|
|
#BMI: reference to Asia standard.
*Age: 21 missing data; BMI: 15 missing data; Education: 50 missing data; Occupation: 43 missing data.
DNA Methylation Status of Genes and CRC Risk
Of the 4 genes, CRC risk was significantly correlated with the methylation of JAK2, STAT1 and SOCS3 (Table 2). After adjusting for gender, age, BMI, education and occupation, statistically significant associations between the methylation of these three genes and the risk of CRC were observed (ORadjusted=1.96, 95% CI, 1.12-3.41, P=0.01 for JAK2; ORadjusted=5.37, 95% CI, 3.74-7.71, P<0.01 for STAT1; ORadjusted=3.30, 95% CI, 1.58-6.87, P<0.01 for SOCS3). However, for STAT3, a significant difference in methylation was not observed between patients and the controls, regardless of the adjustment (Table 2).
TABLE 2 Associations between the Methylation of Individual Genes and CRC Risk
Genes
|
|
Cases No. (%)
|
Controls No.(%)
|
OR (95%CI)
|
P value
|
ORad (95%CI)
|
P value
|
JAK2
|
Negative
|
297(87.61%)
|
323(92.55%)
|
Ref
|
|
Ref
|
|
|
Positive
|
42(12.39%)
|
26(7.45%)
|
1.76(1.05-2.94)
|
0.03
|
1.96(1.12-3.41)
|
0.01
|
STAT1
|
Negative
|
128(31.76%)
|
295(75.84%)
|
Ref
|
|
Ref
|
|
|
Positive
|
211(62.24%)
|
94(24.16%)
|
5.17(3.76-7.12)
|
<0.01
|
5.37(3.74-7.71)
|
<0.01
|
STAT3
|
Negative
|
299(90.88%)
|
350(94.09%)
|
Ref
|
|
Ref
|
|
|
Positive
|
30(9.12%)
|
32(5.91%)
|
1.10(0.65-1.85)
|
0.73
|
0.99(0.57-1.72)
|
0.56
|
SOCS3
|
Negative
|
11(3.38%)
|
42(11.23%)
|
Ref
|
|
Ref
|
|
|
Positive
|
315(96.62%)
|
332(88.77%)
|
3.62(1.83-7.16)
|
<0.01
|
3.30(1.58-6.87)
|
<0.01
|
MCSM
|
Low-level
|
68(29.2%)
|
210(69.1%)
|
Ref
|
|
Ref
|
|
|
High-level
|
165(70.8%)
|
94(30.9%)
|
5.43(3.73-7.87)
|
<0.01
|
4.97(3.34-7.37)
|
<0.01
|
CI: confidence interval; CRC: colorectal cancer; OR: odds ratio.
ORad adjusted for gender, age, BMI, education and occupation.
MCSM Methylation Status and CRC Risk
Among the 233 patients and 304 controls included in the present study, 2 patients and 20 controls did not display methylation of the four genes, 66 patients and 190 controls showed methylation of one gene, 136 patients and 71 controls showed methylation of two genes, 28 patients and 21 controls showed methylation of three genes, and 1 patient and 2 controls showed methylation of all four genes. The 527 results were integrated into a multiple CpG site methylation (MCSM) panel. We defined no or one methylation event as a low level of MCSM in the four genes; likewise, two, three or four methylation events were defined as a high level of MCSM.
After adjustments, a high MCSM level was associated with an increased risk of CRC (ORadjusted=4.97, 95%CI, 3.34-7.37, P<0.01 for the high MCSM level). The details are shown in Table 2.
Stratified Analysis
In male, after adjusting above confounders, the methylation of STAT1 and SOCS3 significantly increased the risk of CRC (ORadjusted= 2.46, 95% CI, 1.73-4.39, P<0.01 and ORadjusted=3.60, 95% CI, 1.40-9.25, P<0.01, respectively). In female, compared with the methylation-negative subjects, subjects with JAK2 and STAT1 methylation displayed a higher risk of CRC (ORadjusted=2.76, 95% CI, 1.01-7.54 P=0.05, and ORadjusted=14.05, 95% CI, 7.50-26.34, P<0.01, respectively) (Table 3a).
TABLE 3a Associations between the Methylation of Individual Genes and CRC Risk in Groups Stratified by Gender
|
Genes
|
|
Cases No.
(%)
|
Controls No.
(%)
|
ORad
(95%CI)
|
P value
|
male
|
JAK2
|
Negative
|
178(86.4%)
|
167(90.8%)
|
Ref
|
|
|
|
Positive
|
28(13.6%)
|
17(9.2%)
|
1.75(0.89-3.42)
|
0.10
|
|
STAT1
|
Negative
|
108(53.2%)
|
155(78.3%)
|
Ref
|
|
|
|
Positive
|
95(46.8%)
|
43(21.7%)
|
2.46(1.73-4.39)
|
<0.01
|
|
STAT3
|
Negative
|
175(89.3%)
|
183(92.0%)
|
Ref
|
|
|
|
Positive
|
21(10.7%)
|
16(8.0%)
|
1.36(0.66-2.79)
|
0.40
|
|
SOCS3
|
Negative
|
7(3.4%)
|
23(11.6 %)
|
Ref
|
|
|
|
Positive
|
196(96.6%)
|
175(88.4%)
|
3.60(1.40-9.25)
|
0.01
|
female
|
JAK2
|
Negative
|
119(89.5%)
|
156(94.5%)
|
Ref
|
|
|
|
Positive
|
14(10.5%)
|
9(5.5%)
|
2.76(1.01-7.54)
|
0.05
|
|
STAT1
|
Negative
|
20(14.7%)
|
140(73.3%)
|
Ref
|
|
|
|
Positive
|
116 (85.3%)
|
51(26.7%)
|
14.05(7.50-26.34)
|
<0.01
|
|
STAT3
|
Negative
|
124(93.2%)
|
167(91.3%)
|
Ref
|
|
|
|
Positive
|
9(6.9%)
|
16(8.7%)
|
0.55(0.22-1.40)
|
0.56
|
|
SOCS3
|
Negative
|
4(3.3%)
|
19(10.8%)
|
Ref
|
|
|
|
Positive
|
117(96.7%)
|
157(89.2%)
|
2.97(0.91-9.66)
|
0.07
|
CI: confidence interval; CRC: colorectal cancer; OR: odds ratio.
ORad adjusted for age, BMI, education and occupation.
In the younger (age<60 years) group, STAT1 and SOCS3 methylation were associated with increased risks of CRC (ORadjusted=7.19, 95% CI, 4.18-12.37, P<0.01, and ORadjusted=2.78, 95% CI, 1.05-7.34, P=0.04, respectively). In the senior age (age≥60 years) group, associations between the methylation of the JAK2, STAT1 and SOCS3 genes and CRC risk were observed (ORadjusted=2.25, 95% CI, 1.06-4.77, P=0.04 for JAK2; ORadjusted=4.33, 95% CI, 2.63-7.13, P<0.01 for STAT1; ORadjusted=4.30, 95% CI, 1.38-13.45 , P=0.01 for SOCS3) (Table 3b).
TABLE 3b Associations between the Methylation of Individual Genes and CRC Risk in Groups Stratified by Age
|
Genes
|
|
Cases No.
(%)
|
Controls No.
(%)
|
ORad
(95%CI)
|
P value
|
<60
|
JAK2
|
Negative
|
144(90.0%)
|
168(93.3%)
|
Ref
|
|
|
|
Positive
|
16(10.0%)
|
12(6.7%)
|
1.82(0.79-4.18)
|
0.16
|
|
STAT1
|
Negative
|
63(40.1%)
|
163(78.7%)
|
Ref
|
|
|
|
Positive
|
94(59.9%)
|
44(21.3%)
|
7.19(4.18-12.37)
|
<0.01
|
|
STAT3
|
Negative
|
136(90.1%)
|
187(91.7%)
|
Ref
|
|
|
|
Positive
|
15(9.0%)
|
17(8.3%)
|
1.05(0.48-2.28)
|
0.90
|
|
SOCS3
|
Negative
|
6(3.9%)
|
20(10.3 %)
|
Ref
|
|
|
|
Positive
|
148(96.1 %)
|
174(89.7%)
|
2.78(1.05-7.34)
|
0.04
|
≥60
|
JAK2
|
Negative
|
138(84.1%)
|
155(91.7%)
|
Ref
|
|
|
|
Positive
|
26(15.9%)
|
14(8.3%)
|
2.25(1.06-4.77)
|
0.04
|
|
STAT1
|
Negative
|
58(35.4%)
|
132(72.5%)
|
Ref
|
|
|
|
Positive
|
106(64.6%)
|
50(27.5%)
|
4.33(2.63-7.13)
|
<0.01
|
|
STAT3
|
Negative
|
146(91.3%)
|
163(91.6%)
|
Ref
|
|
|
|
Positive
|
14(8.8%)
|
15(8.4%)
|
0.97(0.43-2.17)
|
0.94
|
|
SOCS3
|
Negative
|
4(2.1%)
|
22(12.2%)
|
Ref
|
|
|
|
Positive
|
153(97.5%)
|
158(87.8%)
|
4.30(1.38-13.45)
|
0.01
|
CI: confidence interval; CRC: colorectal cancer; OR: odds ratio.
ORad adjusted for gender, BMI, education and occupation
Interaction between Methylation of Genes and Environmental Factors
Evaluation of the effects of interactions and combinations between gene methylation and environmental factors revealed a significant antagonistic interactive effect between low levels of coarse grain intake (<50 g/week) and the methylation of JAK2 (ORi=0.31, 95% CI, 0.11-0.87, P=0.04) on the risk of CRC. Antagonistic interactive effects were observed between alcohol consumption (ORi=0.33, 95% CI, 0.13-0.83, P=0.02), smoking (ORi=0.46, 95% CI, 0.22-0.94, P=0.03) and the methylation of STAT1 on the risk of CRC. A high level of MCSM and bean (except soybean) intake (≥4 times/week) also showed an antagonistic interactive effect on CRC risk (ORi =0.26, 95%CI, 0.11-0.59, P<0.01). The details are shown in Table 4. However, the interaction between SOCS3 methylation and environmental factors on the risk of CRC was not assessed, because the frequency of negative methylation in cases was less than 5% (only 3.38%) to lead to unstable results for the effects of interactions and combinations.
TABLE 4 Combined and Interactive Effects between the Methylation of Individual Genes and Environmental Factors on CRC Risk
Gene
|
Coarse Grains Intake (50g/week)
|
≥50
|
<50
|
Interaction
|
OReg (95%CI)
|
ORi (95%CI)
|
P value
|
JAK2 Negative
|
Ref
|
1.81 (1.31-2.50)
|
|
|
Positive
|
3.05 (1.52-6.13)
|
1.70 (0.78-3.68)
|
0.31 (0.11-0.87)
|
0.04
|
|
Smoking
|
|
No
|
Yes
|
Interaction
|
|
OReg (95%CI)
|
ORi (95%CI)
|
P value
|
STAT1 Negative
|
Ref
|
2.45 (1.51-3.96)
|
|
|
Positive
|
8.16 (4.86-13.70)
|
9.13 (5.25-15.86)
|
0.46 (0.22-0.94)
|
0.03
|
|
Alcohol Intake
|
|
No
|
Yes
|
Interaction
|
|
OReg (95%CI)
|
ORi (95%CI)
|
P value
|
Negative
|
Ref
|
1.60 (0.86-2.99)
|
|
|
Positive
|
16.51 (8.60-31.96)
|
8.67 (4.24-17.74)
|
0.33 (0.13-0.83)
|
0.02
|
|
Bean (except soybean) Intake (average times/week)
|
|
<4
|
≥4
|
Interaction
|
|
OReg (95%CI)
|
ORi (95%CI)
|
P value
|
MCSM Low level
|
Ref
|
1.20 (0.66-2.20)
|
|
|
High level
|
9.41 (5.35-16.55)
|
2.94 (1.67-5.17)
|
0.26 (0.11-0.59)
|
<0.01
|
Additionally, significant combination effects between JAK2, STAT1, STAT3, and MCSM methylation and nine environmental factors on CRC risk existed, and the data are shown in Supplemental Tables S1-S9.