Prevalence of the rs8067378 SNP Among all Patients with SCC and Controls
The values for the χ
2 test of the Hardy–Weinberg equilibrium were 0.329 and 0.181 for the patients and controls, respectively. The statistical evaluation of the rs8067378 genotype prevalence in cases and controls are shown in Table 2. For all patients with SCC, the p
trend value calculated for the rs8067378 transition was statistically significant (p
trend = 0.0019). The logistic regression analysis, which adjusted for the effects of age, parity, oral contraceptive use, tobacco smoking, and menopausal status, also demonstrated a contribution of the rs8067378 SNP to cervical SCC development (Table 2). We observed that the G/G vs. A/A genotype is a risk factor of cervical SCC with an adjusted OR of 1.304 (95% CI 1.080–1.574, p = 0.0057). We observed the same risk effect of the G/A + G/G vs. A/A genotype, with an adjusted OR of 1.444 (95% CI 1.064–1.959, p = 0.0181). However, we did not find an association with cervical carcinogenesis for the G/A vs. A/A genotype, adjusted OR of 1.302 (95% CI 0.944–1.794, p = 0.1069).
Table 2 Prevalence of the rs8067378 polymorphism among patients with squamous cell carcinoma and controls
Distribution of the rs8067378 SNP among SCC Patients with Different Tumor Stage and Grade of Differentiation
Stratified analyses revealed an association of the rs8067378 genotypes with tumor stages III, IV, and grade of differentiation G3 (Table 2). The p
trend value calculated for the rs8067378 SNP in cervical SCC patients with stages III and IV was statistically significant (p
trend < 0.0001 and p
trend = 0.0452, respectively). Adjusting for the effect of age, parity, oral contraceptive use, tobacco smoking, and menopausal status in patients with stages III, the logistic regression analysis demonstrated that the G/G vs. A/A genotype is a risk factor of cervical carcinogenesis with an adjusted OR of 1.674 (95% CI 1.260–2.225, p = 0.0004). There was also a risk effect of SCC for the G/A vs. A/A genotype, with an adjusted OR of 2.180 (95% CI 1.316–3.61, p = 0.0024) and for the G/G + G/A vs. A/A genotype, with an adjusted OR of 2.696 (95% CI 1.668–4.367, p = 0.0001) in stage III patients.
In patients with grade of differentiation G3, the p
trend value calculated for the rs8067378 SNP was statistically significant (p
trend = 0.0004). We found a risk effect of the G/G vs. A/A genotype, with an adjusted OR of 2.004 (95% CI 1.305–3.078, p = 0.0014) for the G/A vs. A/A genotype, with an adjusted OR of 2.346 (95% CI 1.180–4.666, p = 0.0148), and for the G/G + G/A vs. A/A genotype, with an adjusted OR of 2.782 (95% CI 1.437–5.389, p = 0.0024). However, the logistic regression analysis did not show any association of the rs8067378 SNP with tumor stage I, II, and IV and grade of differentiation G1, G2, and GX. Moreover, there was no contribution of the rs8067378 SNP with HPV strains to either SCC or tumor stage I, II, and IV and grade of differentiation G1, G2, and GX (data not shown).
Distribution of the rs8067378 SNP among SCC Patients and Controls with a History of Parity, Oral Contraceptive Use, Tobacco Smoking, or Menopausal Status
The stratified analysis for the rs8067378 polymorphism revealed a risk role of this SNP among patients with a positive history of parity, oral contraceptive use, smoking, and among women of postmenopausal age (Table 3). The age-adjusted OR for women with a history of parity for G/G vs. A/A was 1.323 (95% CI 1.083–1.615, p = 0.0059) and for G/G + G/A vs. A/A was 1.484 (95% CI 1.071–2.055, p = 0.0174). The age-adjusted OR for women with a history of oral contraceptive use for G/G vs. AA was 1.376 (95% CI 1.043–1.816, p = 0.0232). The age-adjusted OR for women with a history of tobacco smoking for G/G vs. AA was 1.666 (95% CI 1.195–2.322, p = 0.0024) and for G/G + G/A vs. A/A was 1.958 (95% CI 1.133–3.384, p = 0.0157). The age-adjusted OR among women of postmenopausal age for G/G vs. A/A was 1.334 (95% CI 1.059–1.679, p = 0.0138).
Table 3 The distribution of rs8067378 genotypes among squamous cell carcinoma risks: parity, oral contraceptive use, tobacco smoking, and menopausal status
Effect of the rs8067378 Polymorphism on GSDMB1–4 Transcript Isoform Levels in SCC and Non-Cancerous Tissues
We observed statistically significant increased GSDMB1 isoform transcript levels in the SCC cervical tissues from carriers of the GG vs. A/A (p = 0.00001) and G/A vs. A/A (p = 0.017) (Fig. 1a). We also found a statistically significant increase in the GSDMB1 transcript levels in the non-cancerous cervical tissues from carriers of the GG vs. AA (p = 0.000001) and G/A vs. A/A (p = 0.033) (Fig. 1b). There were also statistically significant increased GSDMB2 and 3 isoform transcripts in the SCC cervical tissues from carriers of the G/G vs A/A genotype (p = 0.010, p = 0.035, respectively) (Fig. 1a).
However, we did not find significant differences in GSDMB2 and 3 isoform transcript levels in the SCC cervical tissues for the GA vs. AA genotype (Fig. 1a). There were also no significant differences in GSDMB2 and 3 isoform transcript levels in non-cancerous cervical tissues from carriers of the G allele as compared with carriers of the A/A genotype (Fig. 1b). We also did not find significant differences in GSDMB4 isoform transcript levels in SCC cervical tissues and non-cancerous cervical tissues from carriers of the G allele as compared with carriers of the A/A genotype (Fig. 1b).