3.1 Study subjects
This study included 575 glioma patients and 500 age-matched (p = 0.942) and gender-matched (p = 1.000) healthy controls, and the average ages were 40.53 ± 13.90 years and 40.45 ± 18.08 years respectively. The participants' demographic and clinical information was listed in Table 1, including age, gender, World Health Organization (WHO) grade and classification(18), surgical method, radiotherapy, chemotherapy and survival condition.
3.2 Basic information of the selected SNPs
Five SNPs in AKAP6 (rs1957021, rs2145587, rs2239647, rs4261436 and rs17522122) were genotyped. The basic information of selected SNPs and potential function predicted by HaploReg database about these variants were summarized in Supplementary Table 1. All SNPs conformed to the HWE equilibrium (all p values were more than 0.05). The predicted results from the database showed that these SNPs might function as enhancer histone markers or by changing motifs.
3.3 The SNPs of AKAP6 and the risk of glioma
Multiple inheritance models analysis (allele, genotype, dominant, recessive and additive) for the association between AKAP6 rs2239647 and risk of glioma are showed in Table 2. Our analysis revealed a relationship between AA genotype of rs2239647 and increased glioma risk in genotype model (OR = 1.88, 95% CI: 1.16-3.04, p = 0.010) and recessive model (OR = 1.90, 95% CI: 1.19-3.03, p = 0.007).
In addition, we conducted a stratified analysis to explore the effects of these SNPs on glioma susceptibility in a specific population. The significant results of stratified analysis are showed in Table 3. The results showed that AA genotype at rs2239647 was significantly associated with increased glioma risk in populations over 40 years old (genotype model: OR = 2.60, p = 0.012; recessive model: OR = 2.83, p = 0.006) and in the male population (genotype model: OR = 2.42, p = 0.003; recessive model: OR = 2.49, p = 0.009). And, people with the rs2239647-AA genotype had a higher risk of astroglioma than healthy controls (genotype model: OR = 1.90, p = 0.012; recessive model: OR = 1.92, p = 0.009). Moreover, rs2145587 was associated with an increased risk of glioma in female (genotype model: OR = 1.62, p = 0.016; dominant model: OR = 1.57, p = 0.017).
3.4 The SNPs of AKAP6 and the prognostic of glioma
The log-rank test was applied to analyze the associations between overall survival (OS) or progression free survival (PFS) and clinical factors, and the results indicated that gender, age, WHO grading, and radiotherapy factors were not related to the prognosis of patients (p > 0.05), while surgical methods and chemotherapy were significantly related to the prognosis of patients (p < 0.05) (Supplement table 2 and Supplement Fig. 1). We found that the prognosis of glioma patients undergoing total resection was better than patients who did not undergo complete resection (OS: log-rank p < 0.001, HR = 0.63, p < 0.001; PFS: log-rank p <0.001, HR = 0.59, p < 0.001). The prognosis of patients receiving chemotherapy was better than that of patients not receiving chemotherapy (OS: log-rank p < 0.001, HR = 0.67, p < 0.001; PFS: log-rank p = 0.012, HR = 0.81, p = 0.025).
We evaluated the effect of AKAP6 polymorphisms on the patient survival. Log-rank test and Kaplan-Meier analysis revealed the relationship between rs2239647 and OS and PFS in glioma patients (Table 4 and Fig. 1). We found that AKAP6-rs2239647 significantly affected the PFS of patients with high-level glioma (WHO grade III–IV), and patients with CA genotype had a better prognosis (PFS: log-rank p = 0.045, HR = 1.67, p = 0.034).
Subsequently, we analyzed the effect of AKAP6 polymorphisms on the prognosis of patients with astroglioma (Table 5 and Fig. 2). The results showed that AKAP6-rs4261436 had a significant effect on the OS of patients, and patients with TC genotype had a poor prognosis (OS: log-rank p = 0.038, HR = 0.70, p = 0.045). AKAP6- rs17522122 also had a significant effect on the OS of patients, and patients with TC genotype had a poor prognosis (OS: log-rank p = 0.025, HR = 0.75, p = 0.016).