As one of the most aggressive malignant tumors, glioblastoma (GBM) has optimal treatment mainly relying on surgical resection combined with radiotherapy or chemotherapy [1,2]. Nevertheless, the recurrence and metastasis of tumors after treatment and the systemic toxic side effects of radiotherapy or chemotherapy are almost inevitable [3]. Cancer immunotherapy can eradicate tumors, therefore considered a promising strategy for improving GBM outcomes [4]. Targeting programmed cell death ligand 1 (PD-L1) and other immunosuppressive proteins, a revolutionary shift in malignancies has been evoked by immune checkpoint blockade (ICB) therapy in recent decades [[5], [6], [7]]. Despite the evidence of overexpressed PD-L1 in GBM suggesting the great potential of ICB in GBM treatment, most GBM patients have poor response rates for ICB therapy in recent clinical trials [8,9]. In the immunosuppressive tumor microenvironment (TME), this phenomenon mainly gave rise to T cell dysfunction [[10], [11], [12]]. Therefore, modulating the composition and metabolic environment of TME holds a key step in improving the application of ICB therapy in GBM.

Poor tumor immunogenicity, T cell exhaustion, high levels of glutathione (GSH), and tumor-supportive macrophages were exhibited in immunosuppressive TME [13,14]. There is an urgent need for an applicable strategy to amplify the immune response in GBM through reprogramming the immunological profile of the tumor from “cold” to “hot”. Reactive oxygen species (ROS) can destroy the tumor mitochondria, inducing cell lysis and release of mitochondrial DNA (mt-DNA). Such a powerful immune stimulation is able to further trigger tumor lysate-specific cytotoxic T cells (CTLs) activation by promoting antigen presentation of dendritic cells (DCs). Meanwhile, oxidative-damaged mt-DNA has been proven to upregulate the related signal pathway of toll-like receptors (TLRs), transforming tumor-supportive macrophages toward tumor-restricted macrophages [[15], [16], [17], [18]]. Photodynamic therapy (PDT) based on photosensitizers is considered an effective approach to triggering the production of ROS. However, PDT is hindered by the multiple defense mechanisms of tumors and has limited immunotherapeutic effects. For instance, to maintain intracellular redox homeostasis, PDT-produced ROS will be rapidly cleared by overexpressed GSH in TME [19]. Recently, ferroptosis activation has gained great attention for its impact on intracellular redox homeostasis. In the ferroptosis pathway, the GSH exhaustion and down-regulated glutathione peroxidase 4 (GPX4) induce ROS generation, which has the potential to enhance PDT in improving tumor immunogenicity and immune “cold” TME [[20], [21], [22]]. Nevertheless, ferroptosis activation is yet to be applied for PDT enhancement to amplify ICB therapy in GBM.

Notably, systemic administration of drugs for brain tumors faces challenges such as depletion of lymphocytes and side effects [23]. Currently, various nano drug delivery systems have shown excellent targeting performance for GBM, commonly used nanocarriers include liposomes, micelles, gold monomers, quantum dots, magnetic iron oxide, silica, etc. However, there are still limitations of inability of low blood-brain barrier (BBB) penetration rate and poor biocompatibility [24,25]. In recent years, exosomes have become a new approach to develop natural biomimetic nanomedicine delivery systems owing to their biocompatibility [26,27]. Additionally, GBM-derived exosome (GBM-Exo) could stimulate the JAK-STAT3 signaling pathway to upregulate the levels of LCN2, thus promoting brain endothelial membrane fluidity for benefiting BBB penetration of nanoparticles [28]. Although the GBM-Exo exhibits favorable homologous tumor targeting and BBB penetration abilities, there remain doubts that exosomes would be distributed in very small numbers in the brain following intravenous injection [29]. Therefore, more means for heightening the brain-targeting abilities of GBM-Exo are required. Magnetic nanoparticles (MNPs), especially iron oxide nanoparticles (IONPs), showed extensive applications in targeted therapy, equipped with unique physical properties of being manipulated by an external magnetic field [30]. IONP-modified exosomes combine the magnetic targeting property of IONP and the homologous homing function of tumor-derived exosomes, and have broad application prospects in targeted therapy [31]. In this context, we explored the combination of GBM-Exo and IONPs in the brain to provide an ideal drug carrier for ferroptosis activation and PDT.

Herein, we designed a novel homologous magnetic targeted immune vesicle to damage ROS homeostasis in TME by ferroptosis activation augmented PDT, thereby enhancing the innate and acquired immunities in GBM (Scheme 1). Above all, the core nanoparticles (PLGA/AI NPs) loaded with ferroptosis inducer ATO and NIR photosensitizer IR780 were developed. To enhance the BBB penetration properties, PLGA/AI NPs were encapsulated into tumor-derived exosomes (Exo/AI). To further potentiate the tumor targeting ability, Exo/AI were surface conjugated with superparamagnetic iron oxide nanoparticles (SPION) to prepare homologous magnetic targeted immune vesicles (Sp-Exo/AI). With favorable membrane fluidity regulation capacity and dual targeting abilities of homologous homing and magnetic navigation, SPION modified GBM-Exo serves as a packaging vehicle to promote both BBB traverse and tumor accumulation of agents. Upon laser irradiation, ROS was generated by IR780-induced PDT and prevented from being eliminated by ATO-mediated GSH depletion and GPX4 inactivation in the ferroptosis pathway. Therefore, neoantigens from tumors undergoing immunogenic cell death (ICD) stimulated specific T cell activation, and mt-DNA damage triggered tumor-associated macrophages (TAMs) polarization. Eventually, Sp-Exo/AI-mediated ferroptosis activation augmented PDT, combining with immunotherapy based on αPD-1, could foster the function of CTLs and activate effector memory T cells, thus enhancing the systemic immune response by overcoming tumor immune escape. To sum up, this homologous magnetic targeted immune vesicle achieved immunosuppressive TME reversal, enhanced tumor immunogenicity and T cell functions to inhibit tumor recurrence and distal metastasis, ultimately amplified the effect of ICB therapy in GBM.