利用人胚胎干细胞对IDH突变型低级别星形细胞瘤进行建模

Introduction: Few existing models accurately recapitulate and sustain the phenotype of IDH-mutant low-grade gliomas (LGGs). Maintaining a stable IDH-mutant glioma line remains a major challenge, as patient-derived cultures rapidly lose mutant IDH expression due to selective pressure favoring more aggressive, IDH wild-type cells. To address this limitation, we developed an inducible human embryonic stem cell (hESC)-derived model for longitudinal studies of IDH-mutant Low-Grade Astrocytoma (LGA) pathogenesis, progression, and therapeutic response. Methods: We engineered hESCs to carry a doxycycline-inducible IDH1-R132H mutation, with or without CRISPR/Cas9-mediated knockouts of TP53 and ATRX. These combinations generated three isogenic lines: IDH-mutant alone, IDH-mutant/TP53-null, and IDH-mutant/TP53-ATRX-null. This allows for the dissection of how common co-mutations influence glioma initiation and evolution. Cells underwent neural induction to the progenitor stage to model early gliomagenesis, with induction of mutant IDH expression being verified via Western blot. Tumorigenic potential and progression are currently being assessed both in vitro and in vivo. Results: Upon differentiation to the neural progenitor stage, IDH1-R132H induction produced hallmark features of LGG biology, including elevated gammaH2AX, a marker of DNA damage response associated with IDH-mutant gliomas. Induced cells exhibited slower proliferation and limited invasiveness compared to non-induced controls, mirroring the slower growth of patient-derived LGGs. In vivo, IDH-mutant xenografts demonstrated delayed tumor expansion relative to IDH-wild-type counterparts. Ongoing studies are comparing therapeutic responses between our hESC-derived model and patient-derived lines to evaluate its translational utility for preclinical drug testing and longitudinal progression studies. Conclusions: We established an inducible hESC-based model with the capability to recapitulate the behavior and characteristics of IDH-mutant LGA. This platform could overcome the instability of patient-derived lines, enabling controlled, long-term studies of glioma evolution and treatment response. Our model provides a powerful system to dissect how IDH and cooperating mutations drive gliomagenesis and to test targeted therapies across disease stages.