Overexpression of G6PD Represents a Potential Prognostic Factor in Clear Cell Renal Cell Carcinoma

Abstract

Glucose-6-phosphate dehydrogenase (G6PD) participates in glucose metabolism and it acts as the rate-limiting enzyme of the pentose phosphate pathway (PPP). Recently, G6PD dysregulation has been found in a variety of human cancers. Through analyzing published data in The Cancer Genome Atlas (TCGA), our pilot study indicated that G6PD mRNA expression was significantly higher in advanced Fuhrman grade in clear cell renal cell carcinoma (ccRCC). These clues promoted us to further evaluate the expression profile of G6PD and its prognostic impact in patients with ccRCC. In this study, G6PD expression levels were analyzed in 149 human ccRCC and normal tissues using immunohistochemistry. The results showed that compared with that in the normal renal samples, G6PD was found highly expressed in 51.0% of ccRCC (p<0.05). High expression of G6PD was significantly correlated to tumor extent, lymph node metastasis, Fuhrman grade, and TNM stage of ccRCC (all p<0.05). Moreover, positive G6PD expression was associated with poorer overall survival in ccRCC (p<0.001). In Cox regression analyses, high expression of G6PD also could be an independent prognostic factor for overall survival in ccRCC (p=0.007). This study suggests that overexpression of G6PD is associated with advanced disease status and therefore may become an important prognosticator for poor outcomes in ccRCC, as well as a potential therapeutic target for developing effective treatment modalities.

Keywords: TCGA data mining; glucose-6-phosphate dehydrogenase; immunohistochemistry; prognosis biomarker; renal cell carcinoma.

Figure 1

mRNA expression of G6PD and its association with survival in ccRCC based on TCGA data mining. (A) Box plot of G6PD mRNA levels in normal renal tissues, Fuhrman tumor grade 1 and 2 (G1/2) and G3/4 of ccRCC patients. Significant differences were observed by one-way ANOVA. (ns, not significant; ***p<0.001, vs. the control) (B) Stepwise expression of G6PD mRNA associated with significant increase in advanced Fuhrman grade in ccRCC. One-way ANOVA was applied to calculate the p value. (n, a, vs G1 group; b, b1 vs G2 group; c, vs G3 group. n, not significant; a, p=0.001; b, p=0.008; b1, c, p<0.001) (C) Kaplan-Meier analysis of G6PD mRNA expression and overall survival in ccRCC. Patients are stratified as low and high expression group by using mean G6PD mRNA expression level as cutoff. p values were calculated using a log-rank test (p<0.0001, vs. G6PD low group)

Figure 2

Representative images of immunohistochemical staining for G6PD in ccRCC and adjacent non-tumor renal tissues. (A) Immunohistochemistry in human ccRCC specimens and matched adjacent normal tissues by using IgG antibody as isotope negative control. (B-D) Immunohistochemistry for G6PD expression in matched adjacent normal tissues (upper row) and ccRCC tissue samples (lower row), showing weak (B), moderate (C), and strong (D) levels of G6PD staining. Images were captured using a 20 × objective lens. Scale bars=100 μm.

Figure 3

Statistical analysis of G6PD immunoreactivity in human ccRCC and paired normal tissue. (A) Percentage of ccRCC specimens and matched adjacent normal tissues with different expression levels of G6PD was analyzed. (B) Statistical analysis of G6PD expression in ccRCC and matched adjacent normal tissues by χ² test according to IHC staining scores. (* p<0.05, vs. matched adjacent normal tissues)

Figure 4

Correlation between G6PD expression and clinical parameters in ccRCC. Percentage of ccRCC patients in different status (A-F left panel). G6PD expression level in different ccRCC status compared with each control according to IHC staining scores (A-F right panel). Statistical analysis was carried out with χ² test (A, B, D-F) or Fisher exact test (C). (ns, not significant; *p<0.05, **p<0.01, ***p<0.001, vs. each control)

Figure 5

Association of G6PD expression level with survival in ccRCC. Stratification of ccRCC into low and high G6PD expression groups was based on an optimal threshold determined by immunohistologic analysis. Kaplan-Meier overall survival curves for all 117 patients (A; p<0.001), 83 patients with stage I/II ccRCC (B; p<0.05), and 34 patients with stage III/IV ccRCC (C; p<0.05) were shown. Significance measures were based on log-rank test of the p value (vs. G6PD low group).