Screening for differentially expressed genes in retinoblastoma gene chips through the GEO database and validation in clinical settings

AIM: To screen for differentially expressed genes in retinoblastoma (RB) gene chips using GEO2R and validate them clinically. METHODS: The expression profile chip data (GSE110811) was downloaded from the public gene chip database Gene Expression Omnibus (GEO). The GEO2R chip analysis platform was used to identify differentially expressed genes between RB and adjacent normal tissues. According to the International Intraocular Retinoblastoma Classification (IIRC) system, 35 children diagnosed with RB from our hospital and other hospitals were enrolled as the RB group, and 35 healthy children who underwent physical examinations in our hospital were enrolled as the control group. The relative expression levels of Sprouty RTK signaling antagonist 2 (SPRY2) and estrogen-related receptor beta (ESRRB) in the serum of patients were detected by quantitative reverse transcription polymerase chain reaction (qRT-PCR). The diagnostic value of SPRY2 and ESRRB in RB was evaluated by receiver operating characteristic (ROC) curves. Analysis of the relationship between SPRY2/ESRRB expression and clinicopathological features, as well as its correlation with the tumor marker CA199. RESULTS: In the GSE110811 chip, the expression levels of two genes, 16780069 (SPRY2) and 16786783 (ESRRB), showed the most significant differences between RB and normal tissues. The relative expression levels of SPRY2 and ESRRB in the serum of children in RB group (22 males, age 1.64±1.08y) were significantly lower (P0.05) than those in control group (25 males, age 1.54±0.95y). The area under the ROC curve for SPRY2 was 0.735 (95%CI: 0.616-0.854), while that for ESRRB was 0.880 (95%CI: 0.800-0.960). There were statistically significant differences in the expression of SPRY2 and ESRRB with respect to choroidal invasion, optic nerve invasion, differentiation degree, and clinical staging (P0.05). In RB group, the expression levels of SPRY2 and ESRRB decreased gradually with increasing CA199 levels, showing a negative correlation (rSPRY2=-0.593, rESRRB=-0.423; both P0.05). CONCLUSION: The expression of SPRY2 and ESRRB is closely related to the occurrence and development of RB and negatively correlated with the tumor marker CA199. They have the potential to serve as diagnostic biomarkers for RB.