Ferroptosis‐related gene signature predicts prognosis and immunotherapy in glioma

Aims Glioma is a highly invasive brain tumor, which makes prognosis challenging and renders patients resistant to various treatments. Induction of cell death is promising in cancer therapy. Ferroptosis, a recently discovered regulated cell death, can be induced for killing glioma cells. However, the prognostic prediction of ferroptosis‐related genes (FRGs) in glioma remains elusive. Methods The mRNA expression profiles and gene variation and corresponding clinical data of glioma patients and NON‐TUMOR control were downloaded from public databases. Risk score based on a FRGs signature was constructed in REMBRANDT cohort and validated in other datasets including CGGA‐693, CGGA‐325, and TCGA. Results Our results demonstrated that the majority of FRGs was differentially expressed among GBM, LGG, and NON‐TUMOR groups (96.6%). Furthermore, the glioma patients with low‐risk score exhibited a more satisfactory clinical outcome. The better prognosis was also validated in the glioma patients with low‐risk score no matter to which grade they were affiliated. Functional analysis revealed that the high‐risk score group was positively correlated with the enrichment scores for immune checkpoint blockade‐related positive signatures, indicating the critical role of glioma immunotherapy via risk score. Conclusion A novel FRGs‐related risk score can predict prognosis and immunotherapy in glioma patients.
A random survival forest model was constructed to obtain the risk score based on 59 ferroptosis genes in glioma patients in REMBRANDT cohort and it was validated in other datasets including CGGA‐325, CGGA‐693, and TCGA cohorts. Finally, a novel ferroptosis‐related risk score can predict prognosis and immunotherapy in glioma patients.