In recent years, MSCs have demonstrated promising therapeutic effects in neurological diseases, such as IS, intracerebral hemorrhage, spinal cord injury and Alzheimer's disease (20,21). However, multiple limitations of direct MSC transplantation, including potential tumorigenicity, excessive cell dosage and microvascular embolism risk, have impeded their clinical translation. Research has revealed that the paracrine effects of MSCs exert beneficial outcomes that are partly mediated by soluble factors, particularly exosomes (22). MSC-Exos possess marked biological properties, such as good biocompatibility, high stability, low toxicity and efficient exchange of molecular substances, and confer notable benefits on a variety of diseases and injuries including IS, myocardial infarction, cutaneous wounds and osteoarthritis, via bioactive cargos such as growth-associated proteins and miRNAs (23). In addition to possessing biological functions similar to those of parent MSCs, MSC-Exos have been widely utilized due to advantages such as their small volume, low immunogenicity, and ease of acquisition, storage and management. Compared with MSCs, the use of MSC-Exos in the human body offers several potential advantages. i) Nanoscale exosomes can freely traverse various biological barriers without affecting microvascular circulation (24); ii) while the number of transplanted MSCs rapidly decreases, the delivery of MSC-Exos can sustain their therapeutic effect within the body (25); and iii) because MSC-Exos do not exist as living cells in vivo, this may reduce the risk of tumor formation (although this does not imply that all tumorigenic risks are eliminated; MSC-Exos may still inherit the carcinogenic cargo from parent cells) (26). Beyond direct disease intervention, MSC-Exos can also serve as drug carriers, working synergistically with medications to achieve therapeutic efficacy (27). Therefore, studies across multiple disease models have indicated that MSC-Exos may serve as an alternative to traditional stem cell therapies.

Due to the limitations of traditional separation methods in extracting MSC-Exos (such as cumbersome operation, low efficiency, and low exosome yield and purity), a number of novel separation methods have emerged, including ultrafiltration separation technology, integrated dual-filter microfluidic chips, nano-plasma-enhanced scattering, membrane-mediated exosome separation and chip-based exosome separation (6,28). Table I comprehensively summarizes the advantages and disadvantages of different extraction methods for exosomes. These mainstream separation methods exhibit notable differences in exosome purity, particle yield and impurity content. The differences in technical parameters partially explain the inconsistent therapeutic effects reported in different preclinical studies (28-30). Although advances in new technologies have improved exosome yield and purity to some extent, drawbacks such as high costs and stringent equipment requirements remain. Besides the extraction method, the cultivation method of MSCs, their surrounding environment and the source of MSC-Exos also affect MSC-Exo extraction (30-32). MSC-Exos originating from the bone marrow, adipose tissue and umbilical cord possess distinct cargo compositions, including unique profiles of proteins, cytokines and miRNAs, and therapeutic efficacies against ischemic injury (33); therefore, standardized selection of MSC sources and exosome isolation protocols is crucial for reducing experimental heterogeneity and promoting subsequent clinical translation.

MSC-Exo identification is a critical step to ensure the purity, characteristics and reliability of exosomes in research. The International Society for Extracellular Vesicles (ISEV) established minimum experimental requirements for exosome identification in 2018 (34). Regarding biochemical characteristics, the expression of specific positive marker proteins and the absence of contaminating negative markers must be verified. Specifically, at least two transmembrane proteins (any two from CD9, CD63 and CD81) and one cytoplasmic protein (TSG101 or Alix) should be tested. Additionally, calnexin or GRP94 (endoplasmic reticulum proteins), GM130 (Golgi apparatus protein), at least one mitochondrial protein [such as cytochrome c (Cyt-C)] and one nuclear protein (such as histone H3) must be detected and excluded as contaminants. Regarding individual characterization, transmission electron microscopy and nanoparticle tracking analysis are used to observe the peak size and morphology of exosomes. Exosomes typically appear as saucer-shaped or oval bilayer membrane structures with varying sizes, featuring a low electron density component in the center, clear boundaries and concentrated distribution. The peak particle size distribution ranges between 30 and 150 nm (35). In summary, the identification of MSC-Exos is an integrative process involving multiple parameters and diverse techniques. Only by comprehensively analyzing physical characteristics (such as size and shape), particle concentration and biochemical characteristics (for example, positive and negative marker proteins), and strictly adhering to the core principles of ISEV guidelines can it be reliably confirmed that the isolated nanoparticles are indeed exosomes and relatively pure, suitable for subsequent functional studies and application development.

To date, extensive research results have confirmed the broad application prospects of MSC-Exos in treating neurological diseases. For example, MSC-Exos have been shown to promote neurovascular remodeling in a rat model of MCAO via miR-133b transfer and to regulate neurite outgrowth (36). Furthermore, when injected intravenously into Alzheimer's disease model mice, MSC-Exos pretreated with hypoxia can effectively increase synaptic proteins (such as PSD-95 and Synapsin-I) expression; downregulate inflammatory factors, such as TNF-α and IL-1β; and upregulate anti-inflammatory factors, including IL-4 and IL-10, thereby improving learning and memory function (37). Furthermore, Xu et al (27) comprehensively summarized the mechanisms of MSC-Exos in treating spinal cord injury and optimization strategies to enhance therapeutic outcomes, highlighting that combinatorial interventions can further boost the anti-inflammatory and neuroprotective potency of MSC-Exos. In conclusion, MSC-Exos may effectively promote nerve injury repair and functional reconstruction. Compared with MSCs, MSC-Exos have achieved marked success in treating neurological diseases due to their unique advantages, such as improved safety, enhanced stability and greater targeting properties.

Neuronal apoptosis in ischemic areas typically induces secondary immune/inflammatory responses, leading to brain tissue damage. Therefore, inhibiting inflammatory responses is an important neuroprotective strategy. MSC-Exos can improve post-IS functional deficits by modulating the activity of immune cells in the IS lesion area and alleviating local inflammation. For example, MSC-Exos can alleviate acute brain injury and microglial inflammation by inhibiting the CysLT2/ERK1/2 signaling pathway and inhibiting M1 macrophage activation (40). Additionally, MSC-Exos can reduce neuronal oxidative stress and inflammation levels by inhibiting ferroptosis and regulating the expression of glutathione peroxidase 4 (GPX4) and solute carrier family 7 member 11 (SLC7A11) (41,42). However, it has been reported that although MSC-Exos mitigate secondary injuries caused by local inflammatory responses through their immunomodulatory effects, it remains unclear as to whether MSC-Exos can promote the repair and regeneration of neural tissue in the lesion area (43).

Due to ischemia and hypoxia following IS, neurological function is impaired and neuronal apoptosis is enhanced. Numerous studies have confirmed that MSC-Exos attenuate brain injury by inhibiting apoptosis. Hao et al (44) reported that MSC-Exos markedly reduced the expression levels of Cyt-C and IL-1β, and the apoptotic index in the neurons of IS rats. Zhao et al (45) discovered that when the Toll-like receptor 4 (TLR4)/NF-κB signaling pathway was inactivated, the expression of the key apoptotic protease caspase-3 was also decreased. The underlying mechanism may involve MSC-Exos modulating the TLR4/NF-κB signaling pathway to inhibit vascular endothelial cell apoptosis, thereby exerting anti-cerebral ischemic effects. MSC-Exos can also inhibit the activity of antigen-presenting cells, downregulate the expression of B lymphocytes, T lymphocytes and macrophages at the IS site, participate in the regulation and balance of immune responses and provide a favorable microenvironment for subsequent neural regeneration, thus inhibiting apoptosis (46,47).

Following IS, angiogenesis is a critical step in restoring brain tissue function. Vessel blockage damages the structure of the vessels and alters their permeability, resulting in reduced normal blood supply and oxygen delivery. Notably, it has been shown that MSC-Exos facilitate cerebral tissue remodeling by promoting angiogenesis and neuronal survival (48). Doeppner et al (49) demonstrated that when treating adult male rats with MCAO using bone marrow MSC-derived exosomes (BMMSC-Exos), endothelial cell proliferation was markedly increased and a new capillary network was formed, indicating that BMMSC-Exos could promote post-stroke angiogenesis. In addition, Xin et al (36) administered MSC-Exos via tail vein injection 24 h after constructing an MCAO rat model, and the results showed a notable increase in vascular endothelial cell numbers, indicating that MSC-Exos can improve post-stroke vascular regeneration.

It has been show that upon occurrence of IS, due to cerebral tissue hypoxia and reperfusion injury, iron ion levels are markedly increased, thus triggering ferroptosis (50). MSC-Exos protect neurons from damage by targeting and regulating ferroptosis-related pathways through carrying miRNAs, proteins and other bioactive substances. For example, miR-214-3p in MSC-Exos can alleviate neuronal injury in mouse models of cerebral ischemia by modulating the expression of ferroptosis-regulating genes, including GPX4 and SLC7A11 (41). Deng et al (51) also revealed that steroid receptor coactivator-3 in MSC-Exos could significantly inhibit lipid peroxidation and ferroptosis in hypoxic glucose-deprived neurons, and reduce infarct volume and brain tissue water content after cerebral ischemia, thereby improving neurological function. Although the specific mechanisms by which MSC-Exos regulate ferroptosis in IS have not been fully elucidated, exosome-mediated regulation of ferroptosis holds promise as a potential therapeutic target for future clinical treatment of IS.

After IS, disruption of the BBB is an important factor leading to further brain tissue injury. In the early stage of IS, BBB disruption manifests as increased permeability and enhanced destruction, subsequently leading to increased penetration of immune cells and inflammatory substances (52). MSC-Exos can protect the integrity of the BBB by enhancing the expression of tight junction proteins (such as ZO-1 and Claudin-5), and reducing the penetration of inflammatory cells and immune factors. Li et al (53) demonstrated that adipose-derived MSC-derived exosomes (ADSC-Exos) could increase the distribution of ZO-1 and Claudin-5 in MCAO rat models, improve the BBB state and alleviate cerebral edema. In addition, MSC-Exos may reduce brain tissue edema and lesion size, lower the levels of blood-derived brain biomarkers (such as S100B and glial fibrillary acidic protein) and enhance BBB integrity by inhibiting the expression and enzymatic activity of MMP-9 (54).

Notably, IS can trigger multiple forms of programmed cell death, not limited to apoptosis and ferroptosis. I/R injury simultaneously activates novel cell death regulatory pathways, such as pyroptosis, necroptosis, PANoptosis, autophagy and cuproptosis. These distinct death pathways interact to amplify oxidative stress and neuroinflammation, collectively resulting in marked neuronal loss in the ischemic penumbra (55,56).

MSC-Exos are capable of successfully crossing the BBB and accumulating at the ischemic lesion site. In addition to the aforementioned effects of suppressing neuroinflammation, antagonizing neuronal apoptosis, regulating iron overload and lipid peroxidation-mediated ferroptosis, and promoting angiogenesis, they can also broadly regulate multiple programmed death pathways by carrying various miRNAs, cytokines and other active substances (55,57,58). Current basic research has also preliminarily suggested that they may improve copper-induced neuronal damage by maintaining mitochondrial homeostasis, although more experimental validation is required (59,60). In conclusion, the synergistic regulation of multiple programmed cell death pathways constitutes an important mechanistic basis for MSC-Exos in improving the prognosis of IS.

Research has shown that a large number of miRNAs in MSC-Exos widely participate in oxidative stress, cellular apoptosis and immune-inflammatory injury after cerebral ischemia, and also serve important roles in vascular remodeling, neuronal regeneration and synaptic plasticity post-ischemia (63,64). Based on core detection indicators and experimental observation focuses from previous studies, the combination of MSC-Exos and miRNAs may promote functional recovery after injury by regulating microglial polarization, alleviating inflammatory responses, promoting angiogenesis and facilitating nerve regeneration. These mechanisms are closely interrelated and mutually reinforcing in the pathological process of ischemic brain injury and inherent biological cross-effects exist (Table II).

Numerous studies have indicated that miRNAs can be loaded and transported to specific target cells in the form of exosomes; thus, MSC-Exos carrying miRNAs possess the ability to regulate microglial activation and anti-inflammatory responses (65-67). MSC-Exos loaded with miR-223-3p have been reported to alleviate pro-inflammatory responses mediated by M1 polarization of microglia, reduce cerebral I/R injury and thereby accelerate functional recovery after IS (65). MSC-Exos loaded with miR-126 can also improve functional recovery in IS model rats by promoting neurogenesis and inhibiting microglial activation (66). Furthermore, MSC-Exos loaded with miR-30d-5p may prevent brain injury by inhibiting autophagy-mediated microglial polarization to the M1 phenotype (67). In addition, MSC-Exos can regulate microglia/macrophage polarization via targeted delivery of miR-21a-5p, protecting the brains of newborn mice from hypoxic-ischemic injury (68).

It has been reported that IL-6 is an important cytokine involved in pro-inflammatory responses that is closely associated with immune responses after IS (69). Ji et al (70) demonstrated that the concentration of serum exosomes, and the median levels of miR-9 and miR-124 in serum exosomes were notably elevated in patients with IS. In addition, the levels of miR-9 and miR-124 were positively associated with National Institutes of Health Stroke Scale (NIHSS) scores, infarct volume and serum IL-6 concentrations. These results suggested that serum exosomal miR-9 and miR-124 may be promising biomarkers for diagnosing IS and assessing the extent of ischemic injury. Xue et al (71) discovered that downregulation of miR-9 inhibited the transcription of inflammatory factors TNF-α and IL-6 by targeting and silencing silent information regulator 1 in the downstream NF-κB signaling pathway, thereby alleviating I/R injury. Similarly, miR-138-5p within BMMSC-Exos has been shown to promote astrocyte proliferation in the I/R area by targeting lipocalin-2 and to suppress inflammatory responses to mitigate neuronal injury (72). Although inflammatory mediators serve as important targets of MSC-Exos at sites of inflammation, the specific mechanisms by which they promote an immunosuppressive microenvironment in damaged areas remain unclear.

Studies have shown that the number of MSC-Exos in the infarcted hemisphere increases in a dose-dependent manner, and MSC-Exos are rich in various proteins and miRNAs related to neurogenesis (73). In acute IS (AIS), miR-126-3p in MSC-Exos can alter brain plasticity, increase neurite growth, and protect nerve cells from damage and apoptosis (74). Zhang et al (75) tracked fluorescence-labeled MSC-Exos and monitored axonal growth of neurons using time-lapse microscopy, revealing that miR-17-92-treated MSC-Exos activated the phosphatase and tensin homolog (PTEN)/mTOR signaling pathway in recipient neurons, further accelerating the growth of cortical neuron axons. Additionally, Hou et al (76) observed apoptosis in the neurons of MCAO model rats; however, overexpression of miR-29b-3p or silencing of PTEN reversed these changes. It has also been reported that the use of custom exosomes enriched with the miR-17-92 cluster can enhance neural plasticity and functional recovery after IS, a mechanism potentially involving targeting PTEN to activate the PI3K/Akt/rapamycin/glycogen synthase kinase 3β signaling pathway (77).

Brain microvascular endothelial cells interact actively with extracellular matrix proteins forming the basal lamina, pericytes, astrocytes and neurons, jointly forming the neurovascular unit (NVU). AIS leads to rapid NVU damage and impairs neurovascular coupling (78). It has been shown that BMMSC-Exos can markedly increase microvascular density after ischemic brain injury by upregulating miR-486 and activating the PTEN/Akt signaling pathway, thereby improving brain ischemic injury (79). Yang et al (80) demonstrated that ADSC-Exos can promote the migration and angiogenesis of brain microvascular endothelial cells under in vitro oxygen-glucose deprivation (OGD) conditions via the miR-181b/TRPM7 axis. Liu et al (81) reported that elevated levels of miR-122 in the blood can reduce neurological deficits and cerebral infarction area in a rat MCAO-induced IS model, by downregulating the expression of cell-adhesion-related genes such as vascular cell adhesion molecule 1 and intercellular cell adhesion molecule 1, which are involved in leukocyte extravasation, while maintaining vascular integrity.

In conclusion, MSC-Exos carrying miRNAs may serve as a means of intercellular communication after cerebral ischemia, providing novel insights into the pathogenesis and treatment of IS using MSC-Exos, and facilitating the development of MSC-Exos-targeted vectors to improve the therapeutic outcomes for IS. Notably, most preclinical studies rely on young-rodent MCAO models with variable protocols for modelling, exosome preparation and outcome assessment, which may produce conflicting results. Furthermore, small sample sizes and incomplete blinding/randomization raise reproducibility risks. Collectively, the current evidence remains preliminary with low certainty.

The combined application of MSC-Exos with traditional Chinese medicine ingredients also demonstrates favorable therapeutic advantages in the treatment of IS (82). Exosomes have emerged as ideal delivery carriers for herbal active ingredients to target lesion cells and exert therapeutic effects, making them a focus of research (82). Multiple studies have indicated that in the treatment of IS, MSC-Exos loaded with traditional Chinese medicine components such as curcumin, tetramethylpyrazine and gastrodin exert neuroprotective, angiogenic and BBB permeability-regulating effects, with efficacy superior to that of MSC-Exos or drug therapy alone (Table III).

It has been reported that the combination of BMMSC-Exos with tetramethylpyrazine can markedly enhance the neuroprotective effect against IS, a mechanism potentially associated with the inhibition of neuronal apoptosis (83). Furthermore, the combination of BMMSC-Exos with tetramethylpyrazine can notably improve pathological changes in IS, reduce infarct volume, and lower levels of neuronal apoptosis and oxidative stress (84). The combination of BMMSC-Exos with musk ketone has been reported to effectively improve neurological function and reduce infarct volume in MCAO model rats. Further research revealed that this combined strategy can downregulate the expression of pro-apoptotic proteins, such as Bax, and pro-inflammatory factors, including TNF-α and IL-1β, while promoting the expression of the anti-apoptotic protein Bcl-2, thereby effectively inhibiting neuronal apoptosis and microglial activation in rats (85). Additionally, the combination of MSC-Exos with curcumin exhibits anti-lipid deposition, antioxidant and anti-inflammatory effects, protecting against IS by improving endothelial cell permeability and reducing oxidative stress levels (86). It has also been reported that Xionggui decoction enhances axonal regeneration in rats following BMMSC-Exos intervention in IS by regulating the miR-17-92/PTEN signaling pathway (87). This effect is achieved through the improvement of neurological deficits and brain tissue pathological damage, as well as the promotion of oligodendrocyte differentiation and myelin formation.

Ginsenoside Rg1 is a major active component extracted from ginseng roots. Due to its multiple biological functions, such as anti-inflammatory, immune-regulatory and angiogenesis-promoting effects, it is widely used in the treatment of neurological diseases (88). BMMSC-Exos modified with c(RGDyk) peptide, serving as a multifunctional carrier material, load ginsenoside Rg1 through co-incubation and have been reported to activate the PI3K/Akt pathway in a rat model of IS to promote the expression of vascular endothelial growth factor (VEGF) and hypoxia-inducible factor-1α (HIF-1α), thereby promoting angiogenesis and effectively reducing the volume of cerebral infarction (89). Li et al (90) demonstrated that the combination of human umbilical cord mesenchymal stem cell-derived exosomes (hUCMSC-Exos) with Shennao Fuyuan decoction alleviates OGD/reoxygenation-induced damage to neural stem cells by promoting angiogenesis and inhibiting cell apoptosis, which may be associated with activation of the Wnt3a/β-catenin pathway. In addition, compared with treatment with BMMSC-Exos alone, the combination of BMMSC-Exos with Buyang Huanwu decoction shows stronger effects in inducing VEGF secretion by brain tissue endothelial cells and increasing vascular density in cerebral ischemic rats (91). In addition to these drugs, in the treatment of neurological diseases, catalpol can promote post-ischemic neurological recovery, protect the NVU and upregulate VEGF expression (92); however, to the best of our knowledge, whether the combination of catalpol with MSC-Exos is beneficial for angiogenesis after IS has not yet been reported in the literature and requires further exploration.

Studies have shown that single herbal medicines or prescriptions can regulate the BBB through multiple pathways, targets and effects, thereby exerting antioxidant effects, inhibiting neuronal apoptosis, improving cerebral blood flow and other functions. In vivo studies have confirmed that the combination of MSC-Exos and traditional Chinese medicine effectively regulates the permeability of the BBB after cerebral I/R injury, reduces the volume of cerebral infarction and markedly restores neurological function. Gastrodin can alleviate cerebral ischemic injury in mice by enhancing antioxidant and anti-inflammatory activity, and inhibiting the cell apoptosis pathway (93). Nan et al (94) demonstrated that gastrodin can increase the content of miR-20a-5p in circulating exosomes; these MSC-Exos may then penetrate the BBB and upregulate the expression of X-linked inhibitor of apoptosis protein (XIAP) in the ischemic hemisphere, inhibit neuronal apoptosis and promote the recovery of neurological function in MCAO model rats. Yao et al (95) confirmed that the combination of BMMSC-Exos with Qixiong Zuogui compound may markedly reduce the cerebral infarction area in aged MCAO model rats, improve neurological deficit scores and enhance cognitive function; it may also inhibit the expression of MMP-2 and MMP-9, promote the expression of the tight junction proteins ZO-1 and Occludin, and reduce BBB permeability. In addition, MSC-Exos, which are influenced by scutellarin, can protect homocysteine-induced rat microvascular endothelial cells by upregulating the expression of CD63 and BBB tight junction proteins (96).

Although the combination of MSC-Exos and traditional Chinese medicine exhibits certain advantages in treating IS, most existing preclinical evidence comes from a limited number of research teams and independent verification across different teams is still lacking. Almost all data originate from young-adult rodent MCAO models, and evidence from aged or comorbid animals is scarce. In addition, variability in herbal and exosome protocols hampers cross-study comparison and the pre-clinical evidence certainty remains low. Therefore, these encouraging experimental results should be interpreted with caution and more research is needed to translate them into clinical applications to confirm their reproducibility.

It has been reported that the in vivo therapeutic efficacy of natural MSC-Exos is limited by their short half-life, poor lesion accumulation and insufficient tissue targeting ability. Notably, hydrogels loaded with MSC-Exos have attracted widespread attention as an emerging therapeutic strategy in the fields of tissue engineering and regenerative medicine, and are regarded as an effective means to promote nerve regeneration and functional recovery in patients following a stroke (97,98). The good biocompatibility and adjustability of MSC-Exos combined with hydrogels can effectively deliver exosomes to individual cells, thereby extending their retention time at the site of tissue injury (99).

As aforementioned, the main pathogenesis of IS involves ischemia, oxidative stress, inflammatory response and neuronal apoptosis pathways (100). The novel exosome-hydrogel combination therapy directly delivers exosomes to the injury site, providing a scaffold for tissue and nerve repair, thereby exerting anti-inflammatory, antioxidant, axon growth-stimulating and myelin-regenerating effects (101). Lu et al (102) confirmed that the combination of BMMSC-Exos with hydrogel materials can markedly improve neurological functional recovery in rats with traumatic brain injury and can partially relieve the occurrence of neural inflammation. Hydrogels can also mimic different repair microenvironments after peripheral nerve injury, producing a synergistic effect with MSC-Exos to promote neurite outgrowth and neurological defect repair (103). Li et al (104) reported that when using a hydrogel loaded with 3D-MSC-Exos to treat AIS, a 30% Pluronic-F127 hydrogel had good thermosensitivity, was fluid at low temperatures and rapidly transformed into a soft gel near body temperature, and stably released MSC-Exos for >6 days. Notably, this combination can be safely injected into the ischemic core of MCAO model rats, effectively alleviating the inflammatory response in the ischemic hemisphere and promoting the recovery of neurological function in MCAO model rats.

Functionally, MSC-Exos not only participate in immune regulation and homeostasis maintenance but also in blood coagulation, nerve regeneration and intercellular signal transduction. Zhang et al (22) developed a supramolecular hydrogel capable of effectively sustained release of exosomes. After application in an MCAO mouse model, it promoted nerve repair after stroke. Immunofluorescence staining indicated that the neurological function in MCAO model mice was markedly improved, and angiogenic and neuroregenerative capabilities were enhanced. In behavioral tests, motor coordination and balance abilities of these mice were notably improved and the number of apoptotic cells was also reduced. This demonstrates the efficacy of exosomes combined with hydrogels in promoting post-stroke neurological functional recovery. Furthermore, the latest research reports that hydrogels can precisely deliver bioactive molecules to specific regions of the brain while supporting the survival and proliferation of transplanted stem cell-derived exosomes, thereby improving neuroinflammation, vascular remodeling and sensorimotor functional recovery after IS (105). It has also been reported that injectable adhesive hydrogels composed of catechol-grafted hyaluronic acid, prepared via photocuring, can be combined with exosomes to achieve sustained exosome release, which may be used to improve neurological function and angiogenesis after IS (106).

In summary, the combined application of MSC-Exos and hydrogels represents a revolutionary advancement in the field of IS repair therapy. By addressing the critical obstacle of rapid clearance of exosomes in vivo, this approach achieves synergistic effects on neurovascular repair. Current research has transitioned from basic mechanistic exploration to clinical translation, with various innovative hydrogel systems (such as thermosensitive, adhesive and active interface materials) demonstrating unique advantages. It is worth noting that currently, almost all supporting data for hydrogel-assisted MSC-Exos treatment of IS come from rodent models, and such preclinical results cannot be directly applied to human stroke. A number of unresolved issues remain regarding intracranial hydrogel delivery for human IS, including the determination of safe injection volume, optimization of injection sites and assessment of in vivo biocompatibility. Risks such as local inflammation and mass effects that may occur after hydrogel implantation also need to be fully validated in large animal models before human trials can be initiated. Limited sample sizes cause reproducibility concerns and these findings cannot be directly extrapolated to humans. In the future, with the development of engineered exosomes, smart responsive materials and personalized treatment strategies, this field is expected to break through the limitations of neural repair and provide novel solutions for the treatment of disabling IS.

Currently, exosome-based combination therapies have entered early-stage clinical trials. To the best of our knowledge, the ChiCTR2500099772 trial (registered at Chinese Clinical Trial Registry, https://www.chictr.org.cn/showproj.html?proj=266284), registered at The First Affiliated Hospital of Xiamen University (Xiamen, China), is the first to explore the safety and efficacy of hUCMSC-Exos nasal spray in post-stroke cerebral infarction following smoking-related diseases. This trial has employed a single-arm design with a planned enrollment of eight patients; the drug is administered via the nasal mucosa to bypass the BBB, and the primary endpoints are improvements in NIHSS and Modified Rankin Scale scores. An investigational stem cell product for chronic stroke developed by Jiuzhitang Maker (Beijing) Cell Technology Co., Ltd. has completed Phase II patient enrollment. The preliminary Phase I data, presented orally at the 2025 Wuzhen Health Congress, showed a 35% improvement in Fugl-Meyer motor score and 28% improvement in Barthel index (Jiuzhitang Maker Cell Technology Co., Ltd., 2025 Wuzhen Health Congress, Wuzhen, 2025). The conference proceedings remain unpublished, and the detailed quantitative data were only presented orally and have not undergone peer review. The company intends to submit an application for conditional marketing approval in 2026. Although some early clinical trials have reported encouraging preliminary results, these should not be over-interpreted. Current clinical evidence is limited to small-sample studies, and there is still a lack of safety and efficacy data from large-scale clinical verification, which restricts the routine clinical application of MSC-Exos in IS.

Although MSC-Exos exert notable neuroprotective effects against IS in preclinical experiments, the overall methodological quality and experimental rigor of current studies remain suboptimal, leading to multiple potential risks of bias and limiting clinical translation (123,124). First, a number of the included preclinical studies share common methodological flaws, including limited sample sizes, incomplete blinding and randomization procedures, and inconsistent exosome identification criteria, which may introduce potential bias risks and undermine the reliability of the experimental results. No unified optimal dosage has been established in either animal or human studies; notably, excessive dosing may trigger off-target inflammatory responses, whereas insufficient dosing fails to achieve satisfactory neuroreparative effects (125). Regarding administration routes, intravenous injection is the most convenient option, but intranasal and intracerebral administration exhibit higher brain targeting efficiency, accompanied by greater operational risks (126-128). The therapeutic time window is another critical factor; early intervention within hours of cerebral ischemia yields a better prognosis, whereas delayed administration during the rehabilitation phase primarily focuses on neuroplasticity (59).

In vivo biodistribution and pharmacokinetic profiles determine the retention time and efficacy of exosomes in the ischemic penumbra (129). Furthermore, standardized good manufacturing practice (GMP)-level production is a prerequisite for clinical trials (130). Differences in cell sources, culture conditions and isolation methods inevitably lead to batch-to-batch variability in exosomes (131). Almost all preclinical studies tend to report positive therapeutic effects, whereas negative or ineffective experimental data are rarely published, resulting in publication bias.

Notably, the potential oncogenic risks cannot be ignored. MSC-Exos may encapsulate oncogenic proteins or oncogenic miRNAs derived from parent MSCs; these oncogenic cargoes can be transported to recipient cells, posing potential safety hazards (132). Therefore, strict quality assessment of the parent cell source is obligatory prior to clinical translation. There is a need to establish standardized GMP-level exosome preparation methods, a unified quality control system and strictly standardized in vivo experimental protocols to reduce bias and enhance the experimental rigor of future research.