Wei Wei

Persistent fluid despite intravitreal anti–vascular endothelial growth factor (VEGF) treatment remains common in neovascular age‐related macular degeneration (nAMD). Regulatory T cell–derived exosomes (TregExo) have preclinical immunomodulatory activity. To evaluate the safety and maximum tolerated dose (MTD) of intravitreal autologous TregExo plus ranibizumab in refractory nAMD. Design, Setting, and Participants: Single‐centre, prospective, open‐label phase 1a trial using a 3+3 dose‐escalation design at Hebei General Hospital. Four participants with persistent intraretinal or subretinal fluid despite at least 3 consecutive monthly intravitreal anti‐VEGF injections were treated. One intravitreal injection of autologous TregExo (planned doses, 0.05, 0.5, or 1.5 mg) coformulated with ranibizumab, 0.5 mg. The primary outcome was safety and MTD determination. Dose‐limiting toxicity (DLT) was prespecified as severe sterile intraocular inflammation, sustained intraocular pressure greater than 30 mm Hg, loss of more than 15 ETDRS points, or another serious ocular adverse event within 30 days. Exploratory visual and anatomic outcomes were assessed through day 90. Four participants were treated: 3 at 0.05 mg and 1 at 0.5 mg. No DLT or intraocular inflammation occurred in the 0.05‐mg cohort, and intraocular pressure remained stable. Two low‐dose participants improved from ETDRS scores of 3 and 9 at baseline to 35 at day 90; Patient 1 had a macular thickness reduction from 318.2 to 253.9 µm. The 0.5‐mg participant developed sterile intraocular inflammation with vision loss and a large intraretinal cyst, which resolved after local corticosteroid therapy. Dose escalation was stopped, and 0.05 mg was the protocol‐defined MTD. Intravitreal autologous TregExo plus ranibizumab had an identifiable dose ceiling: 0.05 mg was tolerated, whereas 0.5 mg was associated with DLT. Further controlled evaluation of lower‐dose TregExo‐based adjunctive therapy in refractory nAMD is warranted. Trial Registration : Chinese Clinical Trial Registry Identifier: ChiCTR2500095798.

Vitreoretinal fibrosis, a hallmark of proliferative vitreoretinopathy (PVR) triggered by retinal detachment or ocular trauma, necessitates surgery. Through data mining of the vitreoretinal fibrosis microenvironment in PVR patients and mice, showing elevated transforming growth factor β1 (TGFβ1) and M2 macrophage enrichment, we designed and engineered extracellular vesicles that conferred anti-fibrotic efficacy against PVR. These M1 macrophage-derived vesicles (M1evs) were conjugated with anti-TGFβ1 antibodies (aT) via MMP-cleavable linkers (aT-cl-M1ev, termed ACE). Upon intravitreal injection in female PVR model mice, ACE selectively accumulates in lesions, where released antibodies neutralize TGFβ1 and M1evs inhibit M2 macrophage polarization, modulating the microenvironment to diminish vitreoretinal fibrosis. Further incorporating anti-platelet-derived growth factor receptor antibodies yields aTP-cl-M1ev (ACE ^Plus ) to prevent retinal detachment progression in advanced-stage PVR, an efficacy validated in a patient-derived PVR membrane xenograft model. Therapeutic modulation of the vitreoretinal fibrosis microenvironment with the ACE platform provides an efficacious alternative to surgery for PVR.

Bone morphogenetic protein 2 (Bmp2), an FDA-approved osteogenic factor, demonstrates notable potential for enhancing bone regeneration under challenging conditions such as osteoporotic fractures. However, its clinical application is limited by issues including poor in vivo stability and the risk of ectopic ossification. To address these challenges, we developed D-Bmp2@M systems by fusing Bmp2 with a bone-affinity peptide (Asp-Ser-Ser repeated 6 × , DSS6) to generate D-Bmp2, followed by encapsulation of the fusion protein into porous PLGA microspheres. This affinity-based engineering strategy endowed Bmp2 with high bone tissue affinity without compromising bioactivity, thus promoting accelerated fracture healing while reducing the risk of ectopic ossification. Furthermore, the unique multi-chambered porous architecture of the PLGA microspheres enabled a stable, sustained-release profile of D-Bmp2 for approximately 30 days. This release profile not only extends the therapeutic window to align with the fracture healing timeline but also improves treatment efficacy through a cumulative effect, leading to significantly enhanced healing outcomes particularly in osteoporotic fractures. This study presents a PLGA-based sustained-release system that enhances therapeutic efficacy in accelerating fracture healing while reducing Bmp2-associated heterotopic ossification, offering a promising strategy for the treatment of fractures.

Extracellular vesicles (EVs) have emerged as versatile biological carriers capable of transporting diverse therapeutic cargos. Their endogenous biogenesis pathways provide unique opportunities to regulate cargo selection through both environmental modulation and genetic programming. This review outlines how EV‐producing cells can be reprogrammed to load functional proteins and nucleic acids by combining environmental cues with genetic modifications that leverage intrinsic sorting machinery and engineered molecular interactions. For protein cargo, we outline strategies that reshape EV composition through physiological stimuli, as well as genetically encoded systems designed to recruit proteins via scaffold fusion, peptide tags, or fusion‐independent mechanisms. For nucleic acid cargo, we highlight approaches that leverage environment‐driven alterations in RNA abundance, together with targeted loading methods that rely on scaffold‐RNA‐binding proteins (RBPs) fusion constructs, natural sorting motifs, and sorting‐related RBPs to enrich miRNAs, mRNAs, and ribonucleoprotein complexes. We further summarize recent therapeutic applications of these endogenous engineering strategies in cardiovascular, hepatic, neurological diseases, and cancer. Finally, we discuss future directions, including high‐throughput discovery of sorting elements, scalable biomanufacturing, and improved standardization, which together will advance the development of programmable and clinically translatable EVs‐based therapeutics.

Inorganic nanomaterials offer a broader biomedical application landscape for bacterial therapy.