QIN Hu, ZHENG Yi-tong, LIU Wen, MAIMAITI Aierpati, LI Ya-bin, WANG Yong-xin
Objective To investigate the expression of quinolinic acid phosphoribosyltransferase (QPRT) in glioma patients and its associations with clinicopathological features and prognosis. Methods Public databases including The Cancer Genome Atlas (TCGA), TARGET, GTEx, GEPIA2 and UALCAN were used to analyze QPRT expression in glioma and its relationships with disease-specific survival (DSS), overall survival (OS) and progression free survival (PFS). Additionally, 219 pathologically confirmed glioma cases treated at The First Affiliated Hospital of Xinjiang Medical University from July 2010 to December 2018 were enrolled. Immunohistochemical staining was performed to determine QPRT positivity in lower-grade glioma (LGG) and glioblastoma (GBM). Factors affecting survival were identified using Log-rank test and multivariable Cox proportional hazards regression. Univariable and multivariable Logistic regression analyses were conducted to screen risk factors for OS and PFS in glioma patients. Results Bioinformatic analyses showed that QPRT was highly expressed in the combined GBM and LGG (GBMLGG) cohort, LGG and GBM (P=0.000, 0.000, 0.000). QPRT expression was significantly correlated with immune cell infiltration scores in both LGG and GBM. It was positively associated with age in the GBMLGG cohort, LGG and GBM (P=0.000, 0.000, 0.000). It was negatively associated with tumor mutational burden (TMB) in the GBMLGG cohort, LGG and GBM (P=0.000, 0.000, 0.000), and positively associated with microsatellite instability (MSI; P=0.001, 0.019, 0.032). Immunohistochemical staining based on clinical samples showed a QPRT positivity rate was 59.82% (131/219). Compared with the QPRT-negative group, the QPRT-positive group had higher proportions of age > 50 years (P=0.027), receipt of adjuvant chemoradiotherapy (P=0.021), GBM diagnosis (P=0.000), death (P=0.000), disease progression (P=0.000) and tumor recurrence (P=0.000), and had shorter OS (P=0.000) and PFS (P=0.000). Logistic regression showed age > 50 years (OR=2.149, 95%CI: 1.037-4.453; P=0.040), GBM (OR=10.269, 95%CI: 3.635-29.013; P=0.000), and QPRT positivity (OR=5.444, 95%CI: 2.675-11.080; P=0.000) were risk factors for death; GBM (OR=28.821, 95%CI: 3.409-243.648; P=0.002), tumor recurrence (OR=23.538, 95%CI: 2.836-195.336; P=0.003) and QPRT positivity (OR=6.323, 95%CI: 2.850-14.025; P=0.000) were risk factors for progression. Cox proportional hazards regression analysis showed that GBM (RR=8.330, 95%CI: 4.727-14.681; P=0.000) and QPRT positivity (RR=1.692, 95%CI: 1.066-2.685; P=0.026) were risk factors for worse OS, while absence of recurrence was protective (RR=0.507, 95%CI: 0.314-0.819; P=0.006). For PFS, GBM (RR=5.825, 95%CI: 3.436-9.876; P=0.000) and QPRT positivity (RR=1.548, 95%CI: 1.017-2.356; P=0.041) were risk factors. Conclusions QPRT is highly expressed in glioma and serves as an independent risk factor for poorer OS and PFS, suggesting that QPRT may be a potential molecular biomarker for prognostication in glioma.
