br Statistical analysis br Di erences between
2.7. Statistical analysis
Diﬀerences between the normally distributed and non-normally distributed variables were analysed via Student’s t-test and the Mann-Whitney U test, respectively. The genotype and allele frequencies of each SNP were compared using the Chi-squared test or Fisher’s exact test. Hardy-Weinberg equilibrium was observed in the patient group. Spearman’s test was used for the correlation analysis. Kaplan-Meier survival curves were compared using the log-rank test. Two-tailed p values < 0.05 were considered statistically significant. The sensitivity and specificity of the IRF5 mRNA levels were evaluated by receiver-operating characteristic (ROC) curve analysis and their diagnostic value was assessed based on the area under the ROC curve (AUC). The di-agnostic value was considered low at AUC values of 0.5-0.7, moderate at AUC values of 0.7-0.9, and high at AUC values above 0.9. Statistical analyses were performed using Prism 5.0 (GraphPad Software, San Diego, CA, USA) and SPSS 18.0 (SPSS Inc., Chicago, IL, USA).
3.1. Association between IRF5 CFTRinh-172 in WBCs and NSCLC stage
Since peripheral WBCs are closely involved in the immune response and IRF5 is an inflammatory response-regulated transcription factor, we addressed the question of whether IRF5 could be an early biomarker of peripheral WBC response to the development of NSCLC by analysing the IRF5 expression in the WBCs of the NSCLC patients and the healthy control individuals. The results indicated that the IRF5 transcription in the WBCs of the NSCLC patients was significantly upregulated (ap-proximately 5-fold) compared to the healthy controls (p < 0.0001; Fig. 1A). Furthermore, the IRF5 mRNA levels were significantly higher in the WBCs of the patients with early stage NSCLC compared to those with progressive NSCLC (p = 0.0337). Thus, the relative IRF5 mRNA expression was > 10 in 55.8% of the early stage NSCLC patients, but only in 37.5% of the progressive stage NSCLC patients (Fig. 1B). ROC analysis revealed that the IRF5 expression in the WBCs had significant diagnostic value for NSCLC (p < 0.0001), which was higher for early stage disease (AUC = 0.904) than for progressive disease (AUC = 0.81) (Fig. 1C). Because some of the NSCLC patients had low levels of IRF5 mRNA, we examined whether the diﬀerence in the IRF5 expression was related to polymorphisms in the IRF5 gene. Genotyping for poly-morphisms at the rs77571059 and rs2004640 loci revealed that the distribution of the rs77571059 SNP in the NSCLC patients was sig-nificantly diﬀerent from that in the healthy controls. Thus, the fre-quencies of both the -/CGGGG genotype and CGGGG allele were higher in the NSCLC patients than in the healthy controls (p = 0.026 and p = 0.036, respectively) (Table 2). Further analysis demonstrated that al-though the distribution of the -/CGGGG genotype was not diﬀerent between the early stage and progressive stage NSCLC patients (Fig. 1D),
Fig. 1. IRF5 mRNA levels in the WBCs of the NSCLC patients. The IRF5 mRNA was ampli-fied from the WBCs of the NSCLC patients or the healthy donors by qRT-PCR and analysed.
(A) IRF5 mRNA levels in the NSCLC patients (NSCLC) and the healthy controls (HC). (B) IRF5 mRNA levels in the NSCLC patients at either early stage (ES) or progressive stage (PS). (C) Diagnostic value of IRF5 mRNA levels in the WBCs for NSCLC. (D) Genotype fre-quency distribution of IRF5 rs77571059 poly-morphism at the NSCLC stages. (E) Correlation of IRF5 rs77571059 polymorphism with IRF5 mRNA levels in the WBCs of the NSCLC pa-tients. NSCLC, non-small cell lung cancer; AUC, area under the receiver-operating char-acteristics curve; CI, confidence interval.
Genotypes and allele frequencies of IRF5 gene polymorphisms at rs77571059 and rs2004640 loci in the NSCLC patients in this study.
Genotypes NSCLC HC OR p χ2,