Drug resistance is the main cause of glioma recrudescence and death. But its etiology and pathogenesis are still unclear. There is an urgent need to develop a new prognostic and therapeutic method to assess patient survival and reduce mortality. The Cancer Genome Atlas (TCGA) database was used to identify differentially expressed genes(DEGs) between glioblastoma(GBM) and normal. The samples were divided into cluster 1 and cluster 2 subtypes to explore tumor drug resistance according to their IC50 value of drugs for GBM in the GDSC database. The 7 genes were designated as core molecules, which were significantly correlated with immune checkpoints. The dataset CGGA-693 validated that the survival rate of patients in the high-risk group was significantly lower than that in the low-risk group. Moreover, the hub genes were significantly correlated with age, Neoplasm Histologic Grade, IDH status, MGMT status, ATRX status, and BCR status. And the expression level of 7 genes was significantly associated with the grade of glioma, except CAMK2A. RT-PCR revealed that MYBL2, CDC45, CENPA, and E2F7 were upregulated while ATP2B3, KCNJ9, and OPALIN were downregulated in patients. These results demonstrate that the risk model can effectively predict and differentiate the survival and treatment outcomes of glioma patients. MYBL2, CDC45, CENPA, E2F7, ATP2B3, KCNJ9, and OPALIN are expected to be reliable molecular targets and prognostic biomarkers for drug-resistant gliomas.