In Supplementary table 3, the minor allele frequency (MAF) of the PGF rs8019391 and rs2268615, and TNFAIP2 rs710100 polymorphisms among the case and control groups were listed. The genotype distribution of these SNPs in controls were in accordance with the Hardy-Weinberg equilibrium (p > 0.05). The MAF distribution of PGF rs2268615-A allele and TNFAIP2 rs710100-A allele, were found to be higher in the case group, which have increased risk of cervical cancer (rs2268615, A vs C, OR = 1.27, 95% CI = 1.03–1.58, p = 0.029; and rs710100, A vs G, OR = 1.23, 95% CI = 1.01–1.50, p = 0.043).
Multiple genetic model results adjusted by age were also revealed PGF rs2268615 and TNFAIP2 rs710100 polymorphisms conferred to the increased risk of cervical cancer (Table 2). PGF rs2268615 polymorphism was associated with a significantly increased risk of cervical cancer under heterozygote (OR = 1.39, 95% CI = 1.04–1.86, p = 0.024), dominant (OR = 1.40, 95% CI = 1.06–1.84, p = 0.018) and log-additive (OR = 1.29, 95% CI = 1.03–1.61, p = 0.027) models. For rs710100 in TNFAIP2, GA genotype (OR = 1.44, 95% CI = 1.07–1.95, p = 0.018) and GA + AA genotype (OR = 1.42, 95% CI = 1.07–1.89, p = 0.016) compared with GG genotype increased 1.44-fold and1.42-fold the susceptibility of cervical cancer, respectively. Moreover, the result of additive model also showed an increased risk of cervical cancer (rs710100, OR = 1.23, 95% CI = 1.00-1.50, p = 0.046). However, there were no significant association of cervical cancer susceptibility with PGF rs8019391 variants.
Table 2
Relationships between the candidate SNPs and cervical cancer risk
Gene SNP ID | Model | Genotype | Case | Control | Adjusted by age and gender |
OR (95%CI) | p |
PGF rs8019391 | Genotype | CC | 208 | 327 | 1.00 | |
CT | 119 | 145 | 1.29 (0.96–1.74) | 0.093 |
TT | 15 | 26 | 0.91 (0.47–1.76) | 0.777 |
Dominant | CC | 208 | 327 | 1.00 | 0.150 |
CT-TT | 134 | 171 | 1.23 (0.93–1.64) |
Recessive | CC-CT | 327 | 472 | 1.00 | 0.585 |
TT | 15 | 26 | 0.83 (0.43–1.6) |
Log-additive | --- | --- | --- | 1.13 (0.89–1.42) | 0.324 |
PGF rs2268615 | Genotype | CC | 160 | 273 | 1.00 | |
CA | 156 | 191 | 1.39 (1.04–1.86) | 0.024 |
AA | 26 | 31 | 1.43 (0.82–2.49) | 0.209 |
Dominant | CC | 160 | 273 | 1.00 | 0.018 |
CA-AA | 182 | 222 | 1.40 (1.06–1.84) |
Recessive | CC-CA | 316 | 464 | 1.00 | 0.453 |
AA | 26 | 31 | 1.23 (0.72–2.11) |
Log-additive | --- | --- | --- | 1.29 (1.03–1.61) | 0.027 |
TNFAIP2 rs710100 | Genotype | GG | 118 | 210 | 1.00 | |
GA | 171 | 211 | 1.44 (1.07–1.95) | 0.018 |
AA | 53 | 69 | 1.37 (0.89–2.08) | 0.150 |
Dominant | GG | 118 | 210 | 1.00 | 0.016 |
GA-AA | 224 | 280 | 1.42 (1.07–1.89) |
Recessive | GG-GA | 289 | 421 | 1.00 | 0.576 |
AA | 53 | 69 | 1.12 (0.76–1.65) |
Log-additive | --- | --- | --- | 1.23 (1.00-1.50) | 0.046 |
SNP, single nucleotide polymorphism; OR, odds ratio; 95% CI, 95% confidence interval. |
p values were calculated by logistic regression analysis with adjustments for age and gender. |
p < 0.05 means the data is statistically significant. |
Age stratification displayed PGF rs2268615 and TNFAIP2 rs710100 polymorphisms increased the risk of cervical cancer among women at age ≤ 43 years (Table 3). After calculating the ORs for the allele (OR = 1.38, p = 0.041 and OR = 1.42, p = 0.018, respectively), the genotype (CA vs CC, OR = 1.55, p = 0.039; and AA vs GG, OR = 1.97, p = 0.031, respectively), the dominant (OR = 1.56, p = 0.030; and OR = 1.57, 95%, p = 0.034, respectively), and the log-additive (OR = 1.40, p = 0.042; and OR = 1.42, p = 0.020) genetic models, they all concluded that there were significant association of PGF rs2268615 and TNFAIP2 rs710100 polymorphisms with susceptibility to cervical cancer.
Table 3
Relationships between the candidate SNPs and cervical cancer risk according to the stratification by age
SNP ID | Model | Genotype | > 43 years | ≤ 43 years |
Case | Control | OR (95%CI) | p | Case | Control | OR (95%CI) | p |
PGF rs2268615 | Allele | C | 253 | 393 | 1.00 | 0.303 | 223 | 344 | 1.00 | 0.041 |
A | 99 | 131 | 1.17 (0.87–1.59) | 109 | 122 | 1.38 (1.01–1.88) |
Genotype | CC | 89 | 147 | 1.00 | | 71 | 126 | 1.00 | |
CA | 75 | 99 | 1.25 (0.84–1.86) | 0.282 | 81 | 92 | 1.55 (1.02–2.36) | 0.039 |
AA | 12 | 16 | 1.22 (0.55–2.70) | 0.626 | 14 | 15 | 1.62 (0.74–3.56) | 0.228 |
Dominant | CC | 89 | 147 | 1.00 | 0.269 | 71 | 126 | 1.00 | 0.030 |
CA-AA | 87 | 115 | 1.24 (0.85–1.82) | 95 | 107 | 1.56 (1.05–2.33) |
Recessive | CC-CA | 164 | 246 | 1.00 | 0.795 | 152 | 218 | 1.00 | 0.481 |
AA | 12 | 16 | 1.11 (0.51–2.41) | 14 | 15 | 1.31 (0.61–2.81) |
Log-additive | --- | --- | --- | 1.17 (0.86–1.60) | 0.317 | --- | --- | 1.40 (1.01–1.92) | 0.042 |
TNFAIP2 rs710100 | Allele | G | 217 | 331 | 1.00 | 0.597 | 190 | 300 | 1.00 | 0.018 |
A | 135 | 191 | 1.08 (0.82–1.43) | 142 | 158 | 1.42 (1.06–1.90) |
Genotype | GG | 64 | 111 | 1.00 | | 54 | 99 | 1.00 | |
GA | 89 | 109 | 1.41 (0.93–2.14) | 0.105 | 82 | 102 | 1.46 (0.94–2.27) | 0.092 |
AA | 23 | 41 | 0.96 (0.53–1.74) | 0.890 | 30 | 28 | 1.97 (1.07–3.63) | 0.031 |
Dominant | GG | 64 | 111 | 1.00 | 0.208 | 54 | 99 | 1.00 | 0.034 |
GA-AA | 112 | 150 | 1.29 (0.87–1.91) | 112 | 130 | 1.57 (1.03–2.38) |
Recessive | GG-GA | 153 | 220 | 1.00 | 0.418 | 136 | 201 | 1.00 | 0.103 |
AA | 23 | 41 | 0.80 (0.46–1.38) | 30 | 28 | 1.59 (0.91–2.79) |
Log-additive | --- | --- | --- | 1.07 (0.81–1.41) | 0.635 | --- | --- | 1.42 (1.06–1.90) | 0.020 |
SNP, single nucleotide polymorphism; OR, odds ratio; 95% CI, 95% confidence interval. |
p values were calculated by logistic regression analysis with adjustments for age. |
p < 0.05 indicates statistical significance. |
Subsequently, stratified analyses by tumor stage showed that the risk effect for PGF rs8019391 polymorphism appeared to be more prominent in the subset of patients with stage III + IV (Table 4). Compared with the C allele, rs8019391 T allele was highly represented in patients with III–IV tumor stage as compared to patients with I–II tumor stage under the allele (OR = 2.17, p = 4.58⋅10− 4), heterozygote (OR = 2.34, p = 0.005), homozygote (OR = 5.76, p = 0.015), dominant (OR = 2.59, p = 0.001), recessive (OR = 4.13, p = 0.045), and log-additive models (OR = 2.36, p < 0.001).
Table 4
Relationship of clinical stage with PGF rs8019391 polymorphism in cervical cancer patients adjusted by age
SNP ID | Model | Genotype | I-II | III-IV | OR (95%CI) | p |
rs8019391 | Allele | C | 110 | 220 | 1.00 | 4.58⋅10− 4 |
T | 50 | 44 | 2.27 (1.43–3.62) |
Codominant | CC | 37 | 91 | 1.00 | |
CT | 36 | 38 | 2.34 (1.29–4.25) | 0.005 |
TT | 7 | 3 | 5.76 (1.41–23.52) | 0.015 |
Dominant | CC | 37 | 91 | 1.00 | 0.001 |
CT-TT | 43 | 41 | 2.59 (1.46–4.60) |
Recessive | CC-CT | 73 | 129 | 1.00 | 0.045 |
TT | 7 | 3 | 4.13 (1.04–16.45) |
Log-additive | --- | --- | --- | 2.36 (1.45–3.86) | < 0.001 |
SNP, single nucleotide polymorphism; OR, odds ratio; 95% CI, 95% confidence interval. |
p values were calculated by logistic regression analysis with adjustments for age. |
p < 0.05 indicates statistical significance. |
Subsequently, MDR analysis was implemented to assess the impact of the SNP-SNP interactions. Association of higher order interactions with cervical cancer risk was analyzed by MDR analysis as summarized in Fig. 1. The interaction information analysis revealed additive effect between TNFAIP2 rs710100-GA, PGF rs2268615-CA, and PGF rs8019391-CT on conferring risk towards the progression of the cervical cancer. The dendrogram and the Fruchterman-Reingold interaction analysis of our data showed that PGF rs2268615, TNFAIP2 rs710100, PGF rs8019391 exhibited a strong synergy effect on the risk of cervical cancer development as shown in Fig. 2. Table 5 showed that TNFAIP2 rs710100 was the best single-locus model to predict the risk of cervical cancer (testing accuracy = 0.508, CVC = 6/10, p = 0.014). The best two-locus model was the combination of PGF rs2268615 and TNFAIP2 rs710100 (testing accuracy = 0.536, CVC = 9/10, p < 0.0001. The three-locus model included TNFAIP2 rs710100, PGF rs2268615, and PGF rs8019391 (testing accuracy = 0.550, CVC = 10/10, p < 0.0001).
Table 5
SNP–SNP interaction models of the PGF and TNFAIP2 genes analyzed by the MDR method
Model | Training Bal. Acc. | Testing Bal. Acc. | CVC | OR (95% CI) | p |
TNFAIP2 rs710100 | 0.544 | 0.508 | 6/10 | 1.46 (1.08–1.97) | 0.014 |
PGF rs2268615, TNFAIP2 rs710100 | 0.564 | 0.536 | 9/10 | 1.91 (1.41–2.59) | < 0.0001 |
PGF rs2268615, TNFAIP2 rs710100, PGF rs8019391 | 0.587 | 0.550 | 10/10 | 2.11 (1.56–2.84) | < 0.0001 |
MDR, multifactor dimensionality reduction; Bal. Acc., balanced accuracy; CVC, cross–validation consistency; OR, odds ratio; CI, confidence interval. |
p values were calculated using χ2 tests. |
p < 0.05 indicates statistical significance. |