Disease mechanisms and translational barriers guide nanocarrier design for nose to brain delivery in Alzheimer’s disease Authors: Chenqi Liao, Danyang Sun, Xiong Wang Published: 2026 ABSTRACT AbstractAlzheimer’s disease (AD) is a progressive neurodegenerative disorder with limited disease-modifying treatment options, partly because many therapeutic agents show insufficient brain exposure and dose-limiting systemic adverse effects after conventional administration. Nose-to-brain (N2B) delivery has emerged as a non-invasive strategy to transport therapeutics to the central nervous system through the olfactory and trigeminal pathways, thereby partially bypassing the blood-brain barrier. Recent advances in nanomedicine and biomaterial engineering have further improved this approach by enhancing drug stability, nasal residence, mucosal transport, and brain-targeting efficiency. This review examines nanocarrier-enabled N2B delivery strategies for AD from a mechanism-guided perspective, highlighting how AD-related pathological processes shape the selection of therapeutic cargos and formulation designs. We discuss recent progress in the intranasal delivery of repurposed small molecules, natural products, insulin-related agents, peptides and proteins, extracellular vesicles, antibodies, and nucleic acid-based therapeutics. We further summarize major nanocarrier and formulation platforms, including lipid-based systems, polymeric nanoparticles, micelles, extracellular vesicles, in situ gels, and device-assisted delivery technologies. Particular attention is given to the design parameters that influence N2B performance, including particle size distribution/PDI, surface charge, mucus interaction, cargo protection, targeting modification, biodistribution, and deposition reproducibility. Finally, we critically evaluate the translational challenges that continue to limit clinical application, including species differences in nasal anatomy, dose-volume restrictions, device-dependent variability, limited human pharmacokinetic evidence, manufacturing complexity, long-term safety, and regulatory requirements. By integrating disease mechanisms, nanocarrier design, and translational considerations, this review provides a structured perspective for developing more rational and clinically feasible N2B nanodelivery systems for AD.Keywords: Alzheimer’s disease, Nanocarriers, Nose-to-brain delivery, Drug delivery Introduction Alzheimer’s disease (AD) is a progressive neurodegenerative disease characterized by a continuous cognitive decline and memory impairment. It is the most common cause of dementia in older adults [1]. Population ageing is increasing the prevalence of AD and creating a substantial medical and socioeconomic burden [2, 3]. The pathological hallmarks of AD include extracellular β-amyloid protein (Aβ) deposition and neurofibrillary tangles (NFTs) formed by hyperphosphorylated tau. These pathological changes eventually lead to synaptic dysfunction, neuronal loss and widespread brain atrophy [4]. In addition, multiple pathophysiological processes are intertwined, together constituting the complex pathogenesis of AD, such as oxidative stress, neuroinflammation, brain insulin resistance and mitochondrial dysfunction [5, 6]. Although our understanding of the pathological mechanisms of AD has expanded, effective treatment options are still extremely limited. Currently, most approved drugs provide only temporary symptomatic relief and do not prevent or reverse disease progression [4]. Developing disease-modifying therapies for AD remains a major challenge in the field of neuroscience. The blood brain barrier (BBB), formed primarily by tightly connected brain microvascular endothelial cells, remains a major obstacle to AD drug development. It limits the entry of many therapeutic agents, especially macromolecular biologics, from the systemic circulation into the central nervous system (CNS), often resulting in insufficient brain exposure [7, 8]. In order to overcome the limitations of the BBB, researchers have explored a variety of drug administration routes. Among these, nose-to-brain (N2B) delivery has attracted increasing attention in recent years as a non-invasive strategy [7, 9]. N2B delivery utilizes the unique anatomy of the nasal cavity and may facilitate CNS entry through olfactory- and trigeminal-associated pathways, thereby partially bypassing BBB-associated limitations and potentially increasing drug exposure in the CSF and selected brain regions (Fig. 1) [7, 10–12]. This route may enhance regional CNS exposure while reducing peripheral distribution and systemic adverse reactions [9]. Rationale and anatomical routes of N2B delivery for AD. N2B delivery may partially bypass the BBB and transport therapeutics to the CNS through the olfactory and trigeminal pathways Advances in pharmaceutical science and nanotechnology have improved the design of N2B delivery systems. Encapsulating therapeutic agents within nanocarriers, together with mucoadhesive materials, permeation enhancers, or targeting ligands, could reduce rapid mucociliary elimination, prolong formulation retention in the nasal cavity, and facilitate transport across the nasal epithelium toward the brain [13, 14]. This approach may expand the use of therapeutic cargos that are otherwise constrained by conventional administration routes [15]. This review discusses recent advances in N2B nanodelivery for AD from a mechanism-guided and translational perspective. The discussion is organized around the relationship between AD-related pathological processes, therapeutic cargos, delivery barriers, and rational nanocarrier design. Representative N2B strategies are summarized across repurposed small molecules, natural products, insulin-related agents, peptides and proteins, extracellular vesicles, antibodies, and nucleic acid-based therapies. Key nanocarrier and formulation parameters influencing delivery performance are further examined, including particle size distribution/PDI, surface charge, mucus interaction, cargo protection, targeting modification, biodistribution, and device-dependent deposition. Major challenges for clinical translation are also highlighted, including limitations of preclinical models, human nasal anatomy, dosing constraints, delivery-device variability, manufacturability, safety, and regulatory considerations. Mechanism-guided framework for rational N2B nanocarrier design in AD AD is not a single-target disorder, and its major pathological processes impose distinct therapeutic and delivery requirements. Aβ accumulation, tau pathology, neuroinflammation, oxidative stress, insulin resistance, mitochondrial dysfunction, and synaptic impairment are all involved in AD progression, but they differ in target location, cargo type, and delivery requirements [4–6, 16]. For example, Aβ-directed strategies often require extracellular or plaque-associated exposure, whereas tau-directed strategies may require intracellular access or interference with tau propagation [17–19]. Tau-directed delivery is challenging because extracellular tau seeds and intracellular tau aggregates require different delivery strategies. Extracellular tau-targeted approaches need sufficient exposure near affected circuits, whereas intracellular tau clearance requires neuronal uptake and access to intracellular degradation pathways. Neuroinflammation-oriented therapies may benefit from microglia-associated modulation, while insulin-related interventions require reproducible delivery to brain regions involved in metabolic and cognitive regulation [20–23]. Therefore, rational N2B nanocarrier design should be considered not only as a formulation problem, but also as a disease-mechanism-matched delivery problem. Within this framework, therapeutic cargos can be matched to delivery barriers and nanocarrier design principles. Small molecules and natural products often require solubility enhancement, protection from degradation, improved nasal retention, and more favorable brain pharmacokinetics [24–26]. Peptides, proteins, antibodies, and nucleic acids require protection from enzymatic degradation, assistance in epithelial transport, and preservation of biological activity during delivery [27–30]. Extracellular vesicles and other biological nanovesicles provide intrinsic biological functions and cargo-loading capacity, but they also raise challenges related to heterogeneity, scalability, quality control, biodistribution, and long-term safety [31, 32]. Nanocarrier design should therefore be evaluated through parameters directly relevant to N2B delivery, including particle size distribution, polydispersity index (PDI), zeta potential, mucus penetration, mucoadhesion, stability in nasal mucus, epithelial uptake, biodistribution, brain-targeting efficiency, cargo release, and device-dependent nasal deposition [13, 33, 34]. For example, strong mucoadhesion may prolong nasal residence but may also restrict mucus penetration, whereas highly cationic surfaces may enhance epithelial interaction but increase the risk of mucosal irritation or toxicity [33, 35]. In addition, the effectiveness of an optimized nanocarrier depends on whether the formulation can be reproducibly deposited in human-relevant nasal regions, which is strongly influenced by device design, administration posture, dose volume, and interindividual nasal anatomy [36–42]. Based on these considerations, Table 1 summarizes a mechanism-guided prioritization framework linking AD-related pathological mechanisms, therapeutic cargos, N2B delivery barriers, nanocarrier design strategies, representative evidence, and translational bottlenecks (Fig. 2). Table 1Mechanism-guided prioritization of nanocarrier-enabled N2B delivery strategies for ADMechanismCargosN2B barriersDesign logicEvidence gapsCholinergic dysfunctionDonepezil, rivastigmine, galantamineLow brain exposure, gastrointestinal adverse effects, nasal clearanceLipid carriers, chitosan nanoparticles, in situ gelsImproved brain delivery and behavior in animals; clinical advantage unclear [36, 47, 49, 52]Brain insulin resistanceInsulin, metformin, GLP-1 agonistsVariable CNS responsiveness, nasal aging, device variability, hypoglycemia riskDevice-assisted and imaging-guided deliveryHuman insulin data are mixed; dose, device, and patient selection remain key [56, 86, 92, 144]Natural-product anti-Aβ and anti-inflammatory effectsCurcumin, resveratrol, quercetin, berberine, luteolinPoor solubility and bioavailabilityLipid/polymeric carriers and gelsRepresentative animal efficacy; human PK evidence remains weak [68, 69, 71, 78]BACE1-dependent Aβ productionsiBACE1 combinationsNuclease degradation, poor uptake, endosomal trappingEngineered exosomes and targeted LNPsReduced BACE1/Aβ in AD mice; delivery safety and scalability remain unresolved [78, 116, 117]Tau pathology and propagationLansoprazole, anti-tau antibodies, pApoE2, FGF10, NAPIntracellular access, large cargo size, target engagementMicelles, PLGA nanoparticles, ligand-modified chitosan systemsPreclinical tau reduction reported; intracellular delivery and long-term efficacy need validation [17, 18, 64, 121]Neuroinflammation and microglial activationIrbesartan, rapamycin, anti-CD3 antibody, Foralumab, oxytocin, GLENVs, EVsHeterogeneous inflammation, microglial targeting, immune safetyTargeted micelles, EVs, antibody systems, ligand-modified nanoparticlesBroad preclinical anti-inflammatory effects; biomarkers and long-term safety remain needed [14, 23, 60, 73]Synaptic and neurotrophic supportNAP, HB-EGF, FGF10, IGF-2, VGF, NGF, HSP70, YB-1, oxytocinEnzymatic degradation, short half-life, regional exposureProtein delivery, micelles, plasmid delivery, controlled releaseNeuroprotective effects reported; disease-modifying relevance remains uncertain [27, 98, 99, 101]Cell-free regenerative therapyMSC exosomes, hiPSC-NSC EVs, hUSSC exosomes, NSC secretomesHeterogeneity, batch variability, biodistributionNatural or engineered EVs/exosomesStrong animal evidence and early clinical safety signals; GMP production and potency assays remain challenging [22, 32, 110, 115]Antibody-based modulationAnti-Aβ, anti-tau, anti-CD3 antibodies, ForalumabLarge size, poor BBB entry, tissue penetration, immunogenicityIntranasal antibodies, antibody-loaded micelles, CPP-assisted deliveryModel efficacy reported; quantitative brain exposure and target engagement are required [17, 18, 113, 114]Nucleic acid modulationsiRNA, miRNA antagomirs, pApoE2, pVGF, tFNAsNuclease degradation, poor uptake, endosomal escapeLNPs, EVs, PLGA/chitosan carriers, DNA nanostructuresMechanistically precise but limited by delivery, biodistribution, and immunotoxicity [116, 118, 120, 123]Device and deposition constraintsSmall molecules, biologics, antibodies, EVs, nucleic acidsLimited olfactory access, small dose volume, anatomical variabilityDevice-compatible formulations, Aptar-type systems, 3D models, microneedle-assisted deliveryHuman deposition, usability, PK/PD, and device-drug regulation remain central [36, 37, 40, 132]Abbreviations: AD Alzheimer’s disease, BBB Blood-brain barrier, CPP Cell-penetrating peptide, EVs Extracellular vesicles, GLENVs Ganoderma lucidum-derived exosome-like nanovesicles, GLP-1 Glucagon-like peptide-1, GMP Good manufacturing practice, hiPSC-NSC Human induced pluripotent stem cell-derived neural stem cell, LNPs Lipid nanoparticles, MSC Mesenchymal stem cell, N2B Nose-to-brain, PK/PD Pharmacokinetics/pharmacodynamics, PLGA Poly(lactic-co-glycolic acid), tFNAs Tetrahedral framework nucleic acids Mechanism-guided prioritization of nanocarrier-enabled N2B delivery strategies for AD Abbreviations: AD Alzheimer’s disease, BBB Blood-brain barrier, CPP Cell-penetrating peptide, EVs Extracellular vesicles, GLENVs Ganoderma lucidum-derived exosome-like nanovesicles, GLP-1 Glucagon-like peptide-1, GMP Good manufacturing practice, hiPSC-NSC Human induced pluripotent stem cell-derived neural stem cell, LNPs Lipid nanoparticles, MSC Mesenchymal stem cell, N2B Nose-to-brain, PK/PD Pharmacokinetics/pharmacodynamics, PLGA Poly(lactic-co-glycolic acid), tFNAs Tetrahedral framework nucleic acids Mechanism-guided framework for rational N2B nanocarrier design in AD. The framework links AD-related mechanisms, therapeutic cargos, delivery barriers, nanocarrier design strategies, and translational constraints. (Some icons were created with Biorender.com) Therapeutic cargos and mechanism-guided N2B delivery strategies for AD Repositioning of small-molecule drugs Drug repositioning, which explores new indications for approved drugs, represents a key strategy to accelerate the development of novel AD therapies. Many small molecules are being reconsidered for N2B delivery because their pharmacology and safety profiles are relatively well characterized. Intranasal N2B delivery may reduce first pass metabolism and BBB related delivery barriers of these drugs when taken orally, thereby increasing their therapeutic concentration in the CNS. For example, acetylcholinesterase inhibitors (AChEIs) remain widely used symptomatic treatments for AD, but their clinical use may be limited by gastrointestinal adverse effects, systemic exposure, dose-related intolerance, and the need for improved brain-targeted delivery rather than uniformly poor oral bioavailability [9]. Nanoengineered intranasal formulations of acetylcholinesterase inhibitors, such as donepezil [43–48], rivastigmine [36, 49–51] and galantamine [52, 53], have been shown to enhance brain delivery and improve behavioral outcomes in AD models, an effect associated with inhibition of AChE activity [54, 55]. Because impaired insulin signaling in the brain has been increasingly linked to AD pathology, antidiabetic medications are being explored as potential therapeutic options [5]. Metformin, a commonly used glucose-lowering agent, has been reported to improve cognitive performance in streptozotocin (STZ)-induced AD rat models following intranasal administration, even at doses substantially lower than those required for systemic delivery. Its mechanism of action may involve regulating the insulin signaling pathway, such as increasing the levels of phosphorylated Akt (pAkt) and phosphorylated insulin receptor (pIR) [56, 57]. Similarly, angiotensin receptor blockers (ARBs) such as irbesartan have traditionally been used to treat hypertension, but their poor BBB permeability limits their application in neurodegenerative diseases. Delivery of irbesartan through the intranasal pathway has been reported to facilitate its entry into the brain and increase its local availability [58]. In APP/PS1 and LPS-induced AD mouse models, intranasal treatment with irbesartan improved learning and memory behavior while restoring mitochondrial activity and synaptic integrity, effects that are likely associated with its anti-inflammatory, antioxidative, and neuroprotective actions [59, 60]. Similarly, intranasal delivery of olmesartan using chitosan-coated bilosomes has recently been shown to enhance brain bioavailability, improve cognitive function, and reduce amyloid-β pathology in an AD rat model [61]. Other repurposed small molecule drugs explored via the N2B route include the immunosuppressant rapamycin [14], the anti-asthma drugs montelukast [62] and bambuterol [63], the proton pump inhibitor lansoprazole [64], and the nonsteroidal anti-inflammatory drugs ibuprofen [65] and meloxicam [66, 67]. Rapamycin, delivered via nasal polymeric micelles, can effectively reduce Aβ and proinflammatory cytokine levels and improve cognition in the 3×Tg-AD mouse model [14]. Delivery of montelukast through the intranasal pathway has been reported to produce stronger therapeutic effects than oral dosing, leading to improved memory performance and reduced neuropathological alterations in Aβ-induced AD mouse models [62]. Bambuterol, delivered via nasal liposome gel, has been shown to exhibit neuroprotective effects, reduced tau hyperphosphorylation, and protected synapses in the Aβ-induced mouse model [63]. Lansoprazole inhibited the aggregation of AD brain-derived tau seeds, and its nasal administration improved the motor function and ameliorated tau pathology in AD [64]. These studies collectively indicate that repositioning existing small molecule drugs via the N2B delivery route may be a useful strategy for developing new AD therapies. From a design perspective, repurposed small molecules require formulation strategies that match their physicochemical limitations with disease-relevant target engagement, including improved solubility, nasal retention, controlled release, or regional exposure to Aβ-, tau-, inflammatory-, metabolic-, or synaptic-related targets. However, most evidence remains preclinical, and differences in dose, formulation composition, animal model, and behavioral endpoint make direct comparison across studies difficult. Application of natural products and polyphenols Natural products, particularly polyphenols, have been widely investigated as potential AD-modulating agents because of their anti-inflammatory, antioxidant, anti-amyloid, and neuroprotective properties [24]. Nevertheless, the clinical application of many natural products is constrained by challenges including low bioavailability, poor stability, and limited BBB permeability [25]. The combination of N2B delivery and nanotechnology may help address some of these limitations. For example, intranasal curcumin nanoformulations have been reported to inhibit Aβ aggregation and modulate microglial polarization, suppressing neuroinflammation, Aβ load, and also improving memory impairments in transgenic mice with AD [68]. Similarly, intranasal delivery of resveratrol using chitosan-decorated bovine serum albumin nanoparticles showed favorable safety profiles and therapeutic effects [69]. The delivery of resveratrol has also been found to depolymerize Aβ aggregates and modulate neuroinflammation, which alleviates pathology associated with AD [70]. Quercetin is another flavonoid with reported anti amyloidogenic activity. It substantially enhanced cognitive and spatial learning in STZ-induced AD rat model with its nanocarriers and decreased Aβ aggregation, proinflammatory cytokines, and oxidative stress [71]. Furthermore, incorporating these nanocarriers into in situ gels increased nasal retention time and improved brain-targeting capabilities. Notably, this system showed neuroprotective effects comparable to or greater than those of donepezil in models of TMT-induced neurodegeneration [72]. N2B delivery has also brought new application opportunities for other natural products. Ganoderma lucidum-derived exosome-like nanovesicles (GLENVs) have the capability to traverse the BBB when administered nasally, enhancing memory and cognitive impairments in 5×FAD mice, in addition to reducing Aβ deposition and neuroinflammation [73]. Mangiferin has a limited application for its low bioavailability, but N2B delivery enables it to increase its CNS delivery and can exert its anti-inflammatory, antioxidant, and anti-apoptotic neuroprotective effects [25]. Several other natural active ingredients, such as ferulic acid [74, 75], sinapic acid [76, 77], berberine [78, 79], naringenin [80], luteolin [81, 82], osthole [83] have also shown neuroprotective effects in preclinical models via N2B delivery. Rosemary oil nanoemulsion has also been explored as an intranasal formulation platform, although its AD-specific therapeutic evidence remains less direct [84]. These compounds have been formulated as nanoparticles, nanoemulsions, or in situ gels to improve brain exposure and modulate AD related pathological processes. For example, chitosan-coated sinapic acid solid lipid nanoparticles (Cs-SAc-SLNs) administered nasally showed 3.7 times greater brain targeting efficiency than intravenously injected free sinapic acid and significantly improved biochemical markers in mice with Aβ-induced cognitive impairment [76]. These studies suggest the potential value of N2B delivery strategies in harnessing natural products to treat AD. For natural products and polyphenols, nanocarrier design should mainly address poor solubility, instability, rapid nasal clearance, and weak brain exposure while preserving their anti-amyloid, antioxidant, and anti-inflammatory activities. Nevertheless, the translation of natural-product nanoformulations remains limited by poor pharmacokinetic characterization, variable formulation stability, and limited human evidence. Metabolic regulatory effects of insulin and its analogs Brain insulin resistance and signaling dysfunction are increasingly recognized as important components of AD pathophysiology, leading to the proposed concept of AD as a brain insulin-resistant state, sometimes described as “type 3 diabetes” [16]. Therefore, direct insulin supplementation or modulation of brain insulin signaling has been explored as a therapeutic strategy [5]. Because peripheral insulin administration may cause hypoglycemia and insulin has limited BBB permeability, intranasal administration has been investigated as a non-invasive approach to enhance CNS insulin exposure [20, 21]. Several early preclinical and clinical studies suggested that intranasal insulin (INI) may improve cognitive outcomes or modulate CSF biomarkers in patients with MCI or AD [85, 86]. Mechanistic studies have shown that INI may act through multiple pathways, including improving synaptic plasticity and restoring neural network synchronization [87], regulating regional glucose uptake [21], alleviating AD neuropathology [21, 87], and modulating HPA axis activity and sleep [21]. To validate and optimize the brain delivery of INI, researchers have developed radiolabeled insulin and used positron emission tomography (PET) technology to conduct imaging studies in non-human primates and humans [37–39]. The first human PET study used [68Ga] Ga-NOTA-labeled insulin and was given nasally through a special Aptar administration device. The results provided direct imaging evidence that device-assisted intranasal delivery can achieve measurable insulin uptake in the human brain, including regions relevant to AD pathology [37]. These studies not only verified the reliability of specific drug administration devices, but also found that brain uptake patterns of insulin may vary depending on cognitive status, gender and vascular integrity, which provides a basis for the personalized design of future clinical trials [37]. Although INI is clinically attractive, results from larger clinical trials have been inconsistent and fail to show a significant delaying effect on cognitive decline [85, 88, 89]. These discordant clinical outcomes are unlikely to be explained solely by insulin type, dose, formulation, or device performance [5, 20]. Disease-related factors may also contribute, because impaired brain insulin signaling and central insulin resistance in AD may reduce the biological responsiveness to exogenous insulin even when nasal delivery is technically adequate [90]. To optimize intranasal insulin delivery, a recent study developed a thermo-responsive PNPHO nanocarrier that protected insulin from enzymatic degradation, prolonged nasal retention, and achieved a twofold higher brain-to-serum ratio in mice compared to free insulin [91]. Future studies need to explore the mechanism of action of INIs in more depth, optimize the dosing regimen and delivery technology, and combine biomarkers to stratify patients in order to achieve more accurate clinical efficacy [20, 88]. Apart from insulin, compounds that influence insulin signaling, including glucagon-like peptide-1 (GLP-1) receptor agonists, have been explored for intranasal brain delivery. Current evidence supports regional brain uptake of selected incretin receptor agonists after intranasal administration, whereas their therapeutic efficacy in AD models and patients remains to be further established [92, 93]. Therapeutic peptides, proteins and growth factors Therapeutic peptides, proteins, and growth factors are gaining attention as potential interventions for AD due to their potent biological activity and high target specificity. However, their relatively large molecular size restricts BBB penetration. N2B delivery represents a promising strategy for facilitating the entry of these biomacromolecules into the central nervous system. For instance, the neuroprotective peptide NAP (darunatide), an octapeptide derived from activity-dependent neuroprotective protein (ADNP), has been shown to stabilize microtubules and attenuate tau-related pathology [94]. Intranasal delivery of NAP has been reported to enhance cognitive and behavioral outcomes in multiple AD models [27]. In a clinical trial of prodromal AD patients, nasal darunatide showed gender-dependent cognitive improvement effects, especially in males [95]. Similarly, nonapeptides such as JAL-TA9 and ANA-TA9 that hydrolyzed Aβ could effectively enter the CSF and brain parenchyma of rats after intranasal administration, and the brain concentration is significantly higher than that after intraperitoneal or intravenous injection, showing good brain-targeted delivery efficiency [96, 97]. Studies have found that physical exercise can reverse the cognitive impairment in AD mice by upregulating growth factor signals such as heparin-binding epidermal growth factor (HB-EGF), and direct nasal administration of HB-EGF can also enhance memory function in sedentary mice, suggesting that HB-EGF is a potential therapeutic molecule [98]. Fibroblast growth factor 10 (FGF10) levels are decreased in serum of AD patients and in the brain of animal models. Nasal delivery of FGF10 or AAV9-mediated FGF10 gene therapy can improve the cognition of 3×Tg-AD mice and reduce tau protein hyperphosphorylation and neuronal apoptosis [99]. In addition, IGF-2 [100], VGF [101] and neurotrophic factors such as NGF [102–104] have also been shown to have a neuroprotective effect on AD models through nasal delivery. Other protein drugs have also shown therapeutic potential through nasal delivery. For example, Y-box binding protein 1(YB-1) and heat shock protein 70 (HSP70), despite their distinct structures, can regulate the expression of genes related to Aβ and tau metabolism, ribosome biogenesis, neurotransmitter transport, mitochondrial function and antioxidant in the olfactory bulbectomy (OBX) mouse model [105]. Oxytocin, a neuropeptide, can not only improve social and non-social memory impairment in APP/PS1 mice when administered intranasally, but also improve intracranial lymphatic drainage by regulating cerebral hemodynamics and aquaporin-4 (AQP4) polarization, thereby promoting the clearance of Aβ [106, 107]. These studies collectively indicate that N2B delivery provides a feasible clinical application pathway for a variety of therapeutic peptides and proteins. For peptide, protein, and growth-factor therapies, N2B formulations should protect cargos from enzymatic degradation, preserve biological activity, and support sufficient regional exposure to neuronal, synaptic, or trophic signaling targets. Stem cell derivatives: applications of exosomes and secretomes Stem cell-based therapies have attracted attention due to their potential to promote neural regeneration and repair, but direct transplantation of stem cells faces challenges such as safety, immunogenicity, and delivery efficiency [108]. In recent years, research focus has gradually shifted to the use of stem cell derivatives, especially exosomes and secretomes [31]. Exosomes are nanoscale vesicles secreted by cells. They are rich in various bioactive molecules such as proteins, lipids, and nucleic acids, and can serve as an important medium for intercellular communication. Because exosomes can transport bioactive cargos and interact with brain cells, they are being explored as biologically active delivery vehicles for CNS therapy [78]. Intranasal administration provides a direct and non-invasive brain delivery route for these biomacromolecules [32]. A meta-analysis showed that nasal administration of stem cell derivatives, comprising those from bone marrow, adipose tissue, umbilical cord blood, and induced pluripotent stem cells (iPSCs), enhanced cognitive performance via reduction of Aβ deposition and the inhibition of neuroinflammation [32]. This provides supportive preclinical evidence for the nasal administration of stem cell derivatives to treat AD. Exosomes from human umbilical cord mesenchymal stem cells were engineered to be loaded with siRNA targeting β-secretase 1 (BACE1) and berberine, and were reported to deliver to 5×FAD mice brain via nasal administration. These multifunctional exosomes not only target microglia and neurons, but also simultaneously inhibit Aβ production and neuroinflammation, thereby improving cognitive function and promoting neural repair [78]. Similarly, extracellular vesicles (EVs) secreted by neural stem cells (NSCs) derived from human iPSCs (hiPSCs) can effectively enter the brain parenchyma of 5×FAD mice via nasal administration and are rapidly taken up by neurons and microglia, particularly those associated with Aβ plaques [22, 109]. Single-cell RNA sequencing analysis showed that these EVs can induce the transcriptome of microglia and astrocytes to change in the direction of reducing activation state, especially downregulating the expression of genes related to disease-associated microglia (DAM), NLRP3 inflammasome and type I interferon signaling pathway, thereby attenuating the neuroinflammatory cascade, reducing Aβ load and p-tau levels, and ultimately improving cognitive and emotional functions [22]. Exosomes from human unrestricted somatic stem cells (hUSSCs) and iPSC-derived cortical neural stem cell secretomes also showed the function of reducing Aβ accumulation, promoting the expression of synaptic genes, and restoring memory in AD rat or mouse models after intranasal administration [110, 111]. These studies suggest that nasal delivery of stem cell derivatives, especially exosomes/EVs, may represent a promising therapeutic strategy for AD. EV- and secretome-based strategies may combine delivery capacity with intrinsic immunomodulatory or neurotrophic effects. However, their biological complexity makes biodistribution, cellular uptake specificity, batch consistency, potency assays, scalable production, long-term safety, and regulatory classification key translational requirements [32]. Antibody immunotherapy Antibody-based immunotherapy is an important direction in AD treatment, aiming to intervene in the disease process by targeting the clearance of Aβ or pathological tau proteins. Owing to their large molecular size, antibodies cross the BBB inefficiently, limiting the therapeutic effect in the CNS and may require high-dose systemic administration, thereby increasing the risk of side effects [28, 112]. Intranasal administration, as a non-invasive direct brain delivery method, provides new possibilities for antibody treatment of AD [28]. A systematic review analyzed 31 studies on intranasal antibody treatment of rodent models of neurological diseases and found that AD was the most studied disease, and monoclonal antibodies were the main therapeutic agents used [28]. Studies have shown that despite their large molecular weight, intranasal antibodies can still exert neuroprotective effects [28]. For example, a monoclonal antibody (mAb2A7) targeting p-tau217 can effectively enter the brain of PS19 tauopathy mice and bind to the target after intranasal administration, which not only reduces tau pathology and aggregation, but also blocks neuronal apoptosis and brain atrophy, ultimately reversing cognitive deficits [17]. Another study developed a monoclonal antibody TTCM2 targeting the toxic tau conformation and loaded it into micelles (TTCM2-ms) for nasal administration. This preparation can efficiently enter the brain of hTau-tauopathy mice and mediate proteasome degradation by interacting with the intracellular antibody receptor TRIM21 [18]. In addition to targeting core pathological proteins, immunotherapy that regulates neuroinflammation also shows potential. Foralumab, an anti-CD3 monoclonal antibody, was used to treat a moderate AD patient after intranasal administration. PET imaging 3 months later showed a decrease in microglial activation levels in a wide range of brain areas [113]. In the 3×Tg AD mouse model, nasal delivery of anti-CD3 monoclonal antibodies suppressed microglial activation and improved memory, and its effects were independent of Aβ clearance, but were related to brain T cells infiltration and their interaction with microglia [23]. In addition, nasal delivery of antibodies targeting Aβ, combined with the cell-penetrating peptide L-penetratin, was more effective than intravenous injection, improving learning ability and promoting Aβ clearance [114]. Overall, current studies point to intranasal immunotherapy as a potential therapeutic avenue for AD [28]. However, the translational threshold is high. Antibody-based N2B therapy should be matched to target location: Aβ-directed antibodies require extracellular plaque-associated exposure, tau-directed antibodies require sufficient regional exposure and may require neuronal uptake or engagement of intracellular degradation pathways, and immune-modulating antibodies require attention to microglial targeting, biodistribution, and immune safety. For clinical translation, future studies should quantify brain exposure, target engagement, Fc-related immune effects, repeat-dose immunogenicity, mucosal tolerability, and the relationship between delivered dose and regional pharmacodynamic response. Nucleic acid-based therapies Gene therapy intervenes in the disease process by modifying the expression of specific genes, allowing disease mechanisms to be targeted at the level of gene expression. Nucleic acid drugs such as siRNA, miRNA, and plasmid DNA are the main tools for gene therapy, but their efficient delivery into the brain is a major challenge. N2B delivery combined with nanocarrier technology offers a feasible solution for achieving non-invasive gene delivery to the brain [30, 115]. The siRNA targeting BACE1 (siBACE1) is a popular target for AD gene therapy [78, 116]. Researchers co-loaded siBACE1 and the anti-inflammatory drug berberine into engineered stem cell exosomes and achieved BACE1 silencing and neuroinflammation inhibition in the 5×FAD mouse model after intranasal administration, synergistically improving AD pathology [78]. Another study used functionalized lipid nanoparticles (LNPs) to co-deliver siBACE1 and the natural product α-mangostin. After nasal administration, siBACE1 effectively inhibited the expression of BACE1, thereby blocking the synthesis of Aβ [116]. In addition, siBACE1 and rapamycin were co-delivered in dual-target modified nanoparticles, and good therapeutic effects were achieved in AD mice through the nasal route [117]. These studies show that multi-target synergistic therapy can be achieved through nasal co-delivery of siRNA and other drugs. MiRNA plays critical roles in AD pathology, and regulating miRNA expression has also become a therapeutic strategy. For example, miR-206-3p was elevated in AD patients, and inhibiting its function can upregulate brain-derived neurotrophic factor (BDNF). Researchers loaded miR-206-3p antagomirs into engineered EVs, after nasal administration, it was not only superior to the use of antagonists or EVs alone, but also significantly upregulated BDNF levels in the brains of AD mice, and thus improved cognitive impairments, promoted synaptic plasticity [118]. Another study found that nasal delivery of miR-429-3p antagonists can upregulate the expression of its target gene MKP-1, thereby reducing APP amyloid protein processing and Aβ deposition, and improving synaptic function and cognitive deficits via increasing the surface expression of hippocampal AMPARs [119]. Progress has also been made in the delivery of plasmid DNA. For example, a plasmid encoding apolipoprotein E2 (pApoE2), which is believed to effectively bind to and clear Aβ, was co-delivered with the anti-inflammatory drug cannabidiol (CBD) in mannose-modified nanocarriers. After nasal administration, the expression level of ApoE2 in the brain of C57BL6/J mice was significantly increased [120]. Another study used multi-functionalized chitosan micelles to deliver pApoE2. In PS19 tauopathy mice, nasal administration significantly reduced the load of phosphorylated tau [121]. In addition to traditional plasmid DNA, structural DNA nanotechnology has also introduced new therapeutic possibilities. Tetrahedral framework nucleic acids (tFNAs), with their excellent biocompatibility and anti-apoptotic capabilities, have recently been used in the exploration of AD treatment [122]. Studies have shown that intranasally administered tFNAs can successfully enter the brain parenchyma and rescue cognitive and synaptic damage in APP/PS1 mice by inhibiting Aβ-mediated ferroptosis [123]. These studies support the potential of nasal delivery of nucleic acid therapeutics for regulating core AD pathology and neuroprotective pathways. However, effective N2B nanocarriers must not only protect nucleic acids from degradation but also promote cellular uptake, endosomal escape, and sufficient regional biodistribution while minimizing innate immune activation and off-target effects. Compared with small-molecule cargos, clinical translation is further constrained by payload-specific potency assays, repeat-dose immunogenicity, long-term control of gene expression or silencing, scalable GMP manufacturing, and demonstration that molecular target modulation leads to durable pathological or cognitive benefit. A qualitative overview of representative preclinical endpoints across major therapeutic cargo categories is summarized in Fig. 3. Preclinical evidence landscape of nanocarrier-enabled N2B delivery strategies for AD. Major therapeutic cargo categories are summarized according to commonly reported preclinical endpoints, including brain exposure, AD-related pathology, neuroinflammation, neuroprotection, behavioral outcomes, and safety. This figure provides a qualitative overview Pharmaceutical innovations in N2B delivery systems Nanocarrier platform for brain-targeted delivery To improve brain enrichment after intranasal administration, researchers have developed a variety of nanocarrier platforms to utilize their unique physicochemical properties to enhance drug stability and promote drug transport across the nasal mucosal barrier (Fig. 4) [124]. The particle size of these nanocarriers is usually less than 200 nanometers, which can protect drugs from enzyme degradation and improve their bioavailability via improving drug solubility and mucosal permeability [13]. Nanocarrier platforms for intranasal N2B delivery in AD. The major nanocarrier platforms include lipid-based, polymer-based, biological, formulation-supported, and emerging systems In addition to nominal particle size, PDI and zeta potential are critical quality attributes for N2B nanocarrier design. A narrow PDI supports more reproducible mucus interaction, epithelial uptake, and brain exposure, whereas broad size distributions may increase aggregation, nasal clearance, or variable biodistribution. Surface charge also requires careful balance: cationic surfaces may enhance mucoadhesion and epithelial interaction, but excessive positive charge can promote mucus trapping, mucosal irritation, or toxicity. Therefore, particle size, PDI, zeta potential, mucus stability, cargo release, and in vivo biodistribution should be evaluated together rather than as isolated formulation parameters. Lipid-based nanocarriers are one of the most widely studied platforms, including solid lipid nanoparticles (SLNs), liposomes, and nanostructured lipid carriers (NLCs) [125, 126]. Liposomes are vesicles composed of phospholipid bilayers that can simultaneously encapsulate hydrophilic and hydrophobic drugs [125]. For example, mitochondria-targeted liposomes co-loaded with α-tocopherol and donepezil penetrated rat neuronal mitochondria and brain slices, and intranasal administration in APP/PS1 mice improved learning performance and reduced Aβ plaque formation [127]. SLNs and NLCs are composed of biocompatible mixtures of solid lipids or solid-liquid lipids, which have good biocompatibility and controllable drug release ability [26]. For example, quercetin-loaded NLCs [71] and ferulic acid-loaded SLNs [75] showed enhanced brain targeting efficiency and therapeutic effects in AD animal models. Co-encapsulating rivastigmine and nilotinib in NLCs resulted in 5-fold drug concentration compared to drug suspensions after nasal administration, showing the potential for synergistic therapy [49]. Polymer-based nanocarriers also exhibit significant advantages, particularly those utilizing biocompatible and biodegradable polymers such as chitosan and its derivatives, as well as poly (lactic-co-glycolic acid) (PLGA) [128]. Chitosan is a natural cationic polysaccharide with excellent mucoadhesion and the ability to transiently open intercellular tight junctions, significantly promoting nasal drug absorption [35, 80]. Chitosan nanoparticles of naringenin [80] and dihydroquercetin [129], administered intranasally, effectively alleviated memory impairment in AD animal models. PLGA nanoparticles possess good biodegradability and controlled drug release properties, and are often used for drug and gene delivery. For example, co-encapsulating CBD and pApoE2 plasmid in mannose-modified PLGA nanoparticles achieved effective co-delivery of drugs and genes to the brain via the nasal route [120]. Polymer micelles are core-shell structured nanocarriers formed by the self-assembly of amphiphilic block copolymers, suitable for delivering hydrophobic drugs [14]. For instance, rapamycin delivered intranasally via targeting peptide-modified PEG-b-PLA micelles successfully improved cognitive function in 3×Tg-AD mice [14]. Beyond conventional lipid- and polymer-based carriers, protein-based nanoparticles [43, 69], and quantum-dot-based systems [130, 131], together with microneedle-assisted nanocarrier delivery [40], have also shown promising applications in AD N2B drug delivery research. Available quantitative comparisons remain limited because studies use different cargos, doses, animal models, analytical methods, and endpoints. Representative data therefore illustrate the range of reported gains rather than a direct platform ranking: intranasal chitosan-coated sinapic acid SLNs showed a 3.7-fold increase in brain bioavailability/drug targeting efficiency compared with intravenous plain sinapic acid [76], rivastigmine/nilotinib NLCs produced a 3.5-5 fold increase in brain drug concentrations compared with drug suspensions [49], a thermoresponsive insulin nanocarrier achieved a twofold higher brain-to-serum insulin ratio than intranasal free insulin [91], and machine-learning-optimized SLN nasal sprays achieved 19.77% olfactory deposition in a nasal cast model [132]. These findings suggest that platform performance is highly cargo-, formulation-, and device-dependent and should be compared using standardized PK/PD and deposition metrics. Overall, no single nanocarrier platform is universally optimal for N2B delivery in AD. Lipid-based systems improve solubility and stability but may face leakage or storage instability. Polymeric and chitosan-based carriers support mucoadhesion and controlled release, but excessive cationic charge or strong mucoadhesion may cause mucus trapping or mucosal irritation. PLGA nanoparticles, micelles, LNPs, extracellular vesicles, DNA nanostructures, and inorganic systems provide additional opportunities for cargo protection, targeting, and functional delivery, but they differ in loading efficiency, biodistribution, safety, scalability, and regulatory complexity. These advantages and limitations are summarized in Table 2. Table 2Advantages and limitations of major nanocarrier platforms for N2B delivery in ADPlatformExamplesAdvantagesLimitationsTranslationLiposomes and nanotransfersomesDonepezil, berberine, curcumin, luteolinBiocompatible; carry hydrophilic and lipophilic drugsLeakage, storage instability, variable loading; size, PDI, zeta potentialNeeds deposition, stability, and tolerability data [45, 68, 79, 82]SLNs and NLCsQuercetin, ferulic acid, sinapic acid, rivastigmine/nilotinibImprove solubility and stability; controlled releaseDrug expulsion, lipid polymorphism; size, PDI, zeta potential, crystallinityPromising for lipophilic cargos; needs standardized PK/PD comparison [49, 71, 75, 76]Nanoemulsions and microemulsionsIbuprofen, rosemary oilImprove solubility and mucosal spreadingSurfactant irritation, phase separation; droplet size, viscosity, osmolalityDevice compatibility and nasal safety require testing [26, 65, 84]Chitosan-based carriersRivastigmine, galantamine, naringenin, luteolin, pVGF, pApoE2Mucoadhesive; opens tight junctions; modifiableCationic irritation, limited mucus penetration; MW, DD, zeta potentialHuman tolerability and material reproducibility are essential [35, 50, 80, 121]PLGA and hybrid polymeric nanoparticlesCBD/pApoE2, antibody systems, nano-in-micro particlesBiodegradable; protects cargo; controlled releaseBurst release, acidic degradation, residual solvent; size, PDI, ligand densityScalable production and brain biodistribution are required [112, 120, 128]Polymeric and chitosan micellesRapamycin, tau antibody, pVGF, pApoE2Solubilize hydrophobic drugs; support ligand targetingDilution instability, premature release; CMC, size, zeta potentialComplex composition may complicate QC and regulation [14, 18, 101, 121]Protein and albumin nanoparticlesDonepezil protein system, resveratrol albumin nanoparticlesBiocompatible; surface decoration feasibleProtein variability, immunogenicity, enzymatic degradationProtein quality control and immunogenicity assessment are needed [43, 69]LNPs and targeted RNA nanoparticlessiBACE1 combinationsProtect nucleic acids; enable co-deliveryMucus penetration, endosomal escape, immune activation; size, PDI, zeta potentialPotency, biodistribution, repeat-dose safety, and GMP manufacturing are needed [29, 116, 117]EVs, exosomes, and exosome-like vesiclesMSC exosomes, engineered exosomes, hiPSC-NSC EVs, GLENVsIntrinsic biological activity; carry proteins, lipids, RNAsHeterogeneity, batch variability, uncertain potencyGMP production, potency assays, and regulatory classification are key [22, 73, 78, 115]Inorganic, nanozyme, quantum-dot, and nanocomposite systemsInsulin quantum dots, graphene quantum dots, MgO nanocomposites, Mn3O4 nanozymeImaging, catalytic, antioxidant, or theranostic functionsMaterial toxicity, unclear biodegradation; coating, clearance, accumulationCNS safety and clearance are major translational barriers [74, 130, 131, 138]DNA nanostructures and tFNAstFNAsProgrammable structure; potential intrinsic activityNuclease degradation, immune activation, uncertain biodistributionToxicology, biodistribution, and manufacturing validation are needed [122, 123]Mucoadhesive and in situ gel systemsDonepezil gel, galantamine gel, quercetin gel, berberine gel, meloxicam systemExtend nasal residence; support sustained releaseExcess viscosity may impair olfactory deposition; pH, osmolality, gelation temperatureHuman deposition, usability, and mucosal safety are needed [44, 52, 66, 72]Device-assisted and microneedle-assisted systemsRivastigmine spray, Aptar insulin delivery, microplasma-assisted galantamine, microneedle-mediated huperzine AImprove deposition reproducibility and olfactory accessAnatomy-dependent performance, usability, mucosal injury riskHuman deposition and device-drug regulatory evaluation are essential [36, 37, 40, 53]Abbreviations: AD Alzheimer’s disease, CBD Cannabidiol, CMC Critical micelle concentration, DD Degree of deacetylation, EVs Extracellular vesicles, GLENVs Ganoderma lucidum-derived exosome-like nanovesicles, GMP Good manufacturing practice, hiPSC-NSC Human induced pluripotent stem cell-derived neural stem cell, LNPs Lipid nanoparticles, MW Molecular weight, NLCs Nanostructured lipid carriers, PDI Polydispersity index, PK/PD Pharmacokinetics/pharmacodynamics, PLGA Poly(lactic-co-glycolic acid), QC Quality control, SLNs Solid lipid nanoparticles, tFNAs tetrahedral framework nucleic acids Advantages and limitations of major nanocarrier platforms for N2B delivery in AD Abbreviations: AD Alzheimer’s disease, CBD Cannabidiol, CMC Critical micelle concentration, DD Degree of deacetylation, EVs Extracellular vesicles, GLENVs Ganoderma lucidum-derived exosome-like nanovesicles, GMP Good manufacturing practice, hiPSC-NSC Human induced pluripotent stem cell-derived neural stem cell, LNPs Lipid nanoparticles, MW Molecular weight, NLCs Nanostructured lipid carriers, PDI Polydispersity index, PK/PD Pharmacokinetics/pharmacodynamics, PLGA Poly(lactic-co-glycolic acid), QC Quality control, SLNs Solid lipid nanoparticles, tFNAs tetrahedral framework nucleic acids Formulation strategies for overcoming the physiological barriers of the nasal cavity Although the N2B delivery route offers the unique advantage of bypassing the BBB, its clinical translation is still hindered by major physiological barriers. First, the mucociliary clearance mechanism rapidly removes substances from the nasal cavity. Second, the nasal mucosal epithelial cell layer restricts the paracellular transport of many therapeutics due to the dense tight junctions [33]. Therefore, developing formulation strategies that can overcome these barriers is crucial for improving the efficiency of intranasal drug delivery. To prolong the residence time of drugs in the nasal cavity, researchers have widely applied mucoadhesive strategies. By incorporating mucoadhesive polymers such as chitosan [80], polyacrylic acid [133], sodium alginate [44], and hydroxypropyl methylcellulose (HPMC) [66, 134] into formulations, the preparations can adhere to the nasal mucosa surface, resisting mucociliary clearance. In situ gelling is another effective strategy for extending residence time [135]. These formulations are low-viscosity liquids that are easy to spray or drip when administered. Upon entering the nasal cavity, they rapidly transform into a semi-solid gel under physiological temperature or ionic triggers, thus achieving in situ retention and sustained drug release [66, 136]. Poloxamer-based thermosensitive in situ gels are commonly used in intranasal delivery systems to incorporate nanoparticle formulations. For example, galantamine-loaded thiolated chitosan nanoparticles [52] and donepezil-loaded lipid microspheres [44] have been formulated within these gels. This combination helps prolong residence on the nasal mucosa and facilitates subsequent transport of drugs toward the brain, resulting in higher drug levels in brain tissue [44, 137]. However, stronger mucoadhesion does not necessarily translate into better N2B transport, because prolonged retention may also restrict mucus penetration and reduce access to the olfactory region. Thus, realistic transport efficiency should be assessed using mucus stability, epithelial transport, nasal deposition, and in vivo brain biodistribution data. In addition, modifying the surface of nanocarriers has been investigated to improve their interaction with the nasal epithelium. Such surface engineering can promote active targeting and enhance transport across the epithelial barrier [34]. Introducing targeting ligands onto nanocarrier surfaces enables selective interactions with receptors on nasal epithelial cells, which can initiate receptor-mediated transcytosis and promote the transport of therapeutic agents into the brain [43, 116]. Commonly used targeting ligands include transferrin (Tf) or lactoferrin (Lf), which can bind transferrin receptors (TfR) or lactoferrin receptors (LfR) expressed on nasal epithelial or brain-associated barriers [43, 74, 116]; mannose, which has been used to enhance brain targeting through GLUT1- or mannose-associated uptake mechanisms depending on the carrier and biological context; and specific cell-penetrating peptides (CPPs), such as penetratin, which can facilitate cellular internalization of nanocarriers [101, 121]. For example, lactoferrin-functionalized LNPs used for the co-delivery of α-mangostin and siBACE1 showed enhanced brain targeting efficiency after nasal administration [116]. Similarly, chitosan micelles double-modified with mannose and cell-penetrating peptides delivered pVGF plasmids efficiently into the brains of mice [101]. These surface modification strategies improved N2B delivery performance in preclinical settings. Novel drug delivery devices and technologies Beyond formulation design, progress in delivery devices and administration technologies also plays a critical role in improving intranasal drug targeting and delivery consistency [9]. Conventional nasal drops or sprays frequently lead to deposition of a large proportion of the formulation in the anterior nasal cavity. From this region, drugs are readily removed by mucociliary clearance or drained into the gastrointestinal tract through the nasopharynx, limiting their ability to reach the olfactory region, which contains abundant olfactory nerve endings and serves as an important gateway for N2B transport [36, 41]. Therefore, developing devices capable of precisely delivering drugs to the olfactory region is a current research focus. Studies have shown that patient position during drug administration has a significant impact on drug distribution. For example, using a syringe with a straight tip for collimated liquid delivery in the lateral decubitus position can deliver most of the therapeutic liquid to the olfactory region more effectively than conventional sprays, avoiding the head tilt or head-down posture that may be difficult for elderly patients [41]. Human nasal cavity models made using 3D printing technology provide a powerful tool for in vitro evaluation of the effects of different devices and drug delivery parameters on drug deposition and distribution, which helps to optimize dosing regimens [36, 41, 42, 132]. Recently, machine learning has been integrated with such 3D-printed nasal casts to further optimize olfactory deposition of nanocarrier-based sprays, achieving a nearly 20% deposition rate in the cast and demonstrating therapeutic efficacy in an AD model [132]. Furthermore, Aptar’s Cartridge Pump System (CPS) has been shown to reliably deliver insulin to multiple brain regions in PET imaging studies of non-human primates and humans, validating the device’s clinical application potential [37, 38]. Relatively more invasive, device-assisted physical enhancement techniques have also been explored to improve intranasal drug transport. Among these, dielectric barrier discharge (DBD) microplasma technology represents an emerging approach. In one study, a spiral DBD microplasma device was applied approximately 2 mm inside the rat nasal cavity for 4 min before drug administration. This pretreatment significantly increased galantamine distribution in both cerebral hemispheres compared with untreated animals [53]. Soluble microneedle patches represent another emerging strategy. These toothbrush-shaped patches contain hyaluronic-acid-based microneedle tips loaded with drug nanocarriers. Upon contact with the nasal mucosa, the tips dissolve within seconds and release the encapsulated drug, while the patch base can be removed. This approach enables delivery directly into the submucosal layer and reduces potential disturbance to nasal cilia and microbial communities. In preclinical experiments, the system achieved substantial therapeutic effects through efficient delivery of huperzine A [40]. These novel devices and technologies provide new solutions for achieving more precise and efficient N2B delivery, but their safety and clinical feasibility still need further verification. Translational medicine: from basic research to clinical applications Evidence and limitations of preclinical animal models Preclinical animal models are fundamental to evaluating the safety and efficacy of intranasal drug delivery for AD therapies, providing crucial proof-of-principle evidence for advancing to human trials. Currently, a variety of AD animal models are widely used in research, including transgenic models such as APP/PS1 [59, 123], 5×FAD [73, 138], 3×Tg-AD [14, 139], and tau P301S [19, 140] mice, which respectively mimic Aβ and/or tau pathology. In addition, there are chemically induced models, such as the STZ intracerebroventricular injection model, used to simulate the brain insulin resistance characteristics of sporadic AD [56, 87]; and models that simulate Aβ toxicity and neuroinflammation through direct injection of Aβ oligomers or LPS [60, 62]. These models provide important in vivo platforms for verifying the brain targeting, pharmacokinetic characteristics, and therapeutic effects of intranasally delivered drugs on core pathology, neuroinflammation, and cognitive function. Although these animal models provide abundant data support for N2B drug delivery research, their inherent limitations are also significant factors hindering the translation of research findings into clinical applications. First, most transgenic models are based on mutations associated with familial AD, while over 95% of AD patients have sporadic AD, with a more complex etiology. Therefore, these models may not fully reproduce the pathological complexity and disease progression of human sporadic AD [141]. Non-transgenic animal models, including those induced by STZ or Aβ injection, can reproduce certain pathological features of AD; however, they do not fully capture the long-term progressive nature of the disease [141]. Furthermore, the differences in nasal anatomy, mucosal physiology and immune system between species may cause the delivery efficiency and therapeutic effect observed in rodents to not be directly extrapolated to humans [33]. For example, compared with humans, the olfactory epithelium of mice accounts for a significantly larger proportion of the total nasal cavity. Therefore, the olfactory pathway delivery efficiency data obtained from the mouse model may be overestimated when extrapolating to humans. Finally, due to the differences between the endpoint indicators commonly used in preclinical research and the complex cognitive functions of human beings, it is often difficult to effectively predict the clinical benefits of humans by the behavioral improvement of animal models alone. In short, we must be cautious when interpreting the results of preclinical research. The development of animal models closer to human pathological characteristics, such as further research on non-human primates, can provide research data closer to the human pathological state, thus helping to narrow the transformation gap between basic research and clinical applications [38, 39]. For clinical translation, dose volume is also a practical constraint that is often underestimated in preclinical studies. Human intranasal administration generally allows only a limited volume per nostril, and excessive volume may increase anterior leakage, swallowing, mucociliary clearance, and systemic exposure. Therefore, clinically feasible N2B formulations should achieve sufficient drug loading, stability, and brain exposure within a small, device-compatible dosing volume. Emerging clinical research evidence and human trials Although most N2B delivery strategies for AD remain at the preclinical stage, several promising approaches have entered human trials and provided preliminary clinical evidence. Among them, INI stands out as the most extensively studied candidate. A 12-month randomized, double-blind clinical trial showed that INI treatment significantly delayed the progression of white matter hyperintensity (WMH) volume in the deep brain and frontal lobe regions, providing direct imaging evidence for the neuroprotective effect of intranasal administration [142]. Furthermore, in a study of elderly surgical patients, perioperative INI administration was shown to significantly improve cognitive scores and effectively suppress the levels of inflammatory markers [143]. However, a large-scale, multisite Phase 2/3 trial failed to observe a significant delay in cognitive or functional decline with INI delivery [144]. Randomized controlled trials of glulisine (rapid-acting insulin) have also shown that it did not demonstrate the expected advantage in improving cognition, function, or mood, and was accompanied by a higher incidence of nasal irritation [89]. Furthermore, a meta-analysis of 16 randomized controlled trials found no significant overall cognitive benefit of intranasal insulin over placebo [85]. These findings suggest that improving delivery device reliability, as well as refining dosage regimens and patient selection criteria, may be critical in achieving therapeutic benefit. Apart from INI, stem cell-derived therapies have also entered early-stage clinical trials. A phase I/II clinical trial tested intranasal administration of allogeneic human adipose mesenchymal stem cell-secreted exosomes in patients with mild to moderate AD. The results showed that after 12 weeks of treatment, the therapy demonstrated good safety and tolerability, with no serious adverse events reported. The moderate dose group showed a transient improvement in cognitive scores after treatment discontinuation. Although these preliminary data suggest acceptable short-term safety in a small cohort, this therapeutic approach remains at a very early clinical stage. Its efficacy should therefore be interpreted cautiously, and larger, controlled, multi-center trials are needed to validate reproducible clinical benefit [115]. Other candidates, including intranasal NAP [27, 95] and Foralumab [113], are also being investigated for their translational potential. While clinical investigations remain limited, existing studies provide preliminary human data supporting the feasibility of N2B delivery as a treatment strategy for AD. Future research will need to address several challenges, including improving delivery efficiency, identifying reliable biomarkers, and optimizing patient stratification. Conclusion N2B delivery provides a non-invasive strategy for improving CNS exposure of therapeutic agents for AD, particularly for cargos limited by poor BBB permeability or systemic adverse effects. When combined with nanocarrier engineering, this route may improve drug stability, nasal residence, mucosal transport, cargo protection, and regional brain exposure. Nevertheless, N2B delivery should not be viewed as a universal solution for AD drug delivery. Its performance depends on the interaction among disease mechanisms, cargo properties, formulation attributes, nasal deposition, and biological barriers. Preclinical studies have provided encouraging evidence for a range of therapeutic cargos, including repurposed small molecules, natural products, insulin-related agents, peptides and proteins, extracellular vesicles, antibodies, and nucleic acid-based therapeutics. However, most supporting evidence is still derived from rodent models, whereas human evidence remains limited mainly to intranasal insulin and a few biologic or peptide-based approaches. Future studies should therefore move beyond demonstrating brain distribution or behavioral improvement alone and provide quantitative evidence of target engagement, dose response, long-term nasal and CNS safety, device reproducibility, and human-relevant nasal deposition. Future N2B nanodelivery systems for AD should move from empirical formulation screening to mechanism-guided and translation-oriented design. Standardized characterization of particle size distribution/PDI, zeta potential, mucus interaction, cargo loading and release, biodistribution, and repeat-dose tolerability will be essential. These systems should also be evaluated as device-formulation combinations under human-relevant dosing and deposition conditions. Together with scalable manufacturing and biomarker-guided clinical evaluation, such evidence will help identify candidates with genuine clinical potential. Author contributions Chenqi Liao and Danyang Sun drafted the manuscript. The study was supervised by Xiong Wang. Funding This work was supported by the National Natural Science Foundation of China (82572661 to Xiong Wang). Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Declaration of generative AI and AI-assisted technologies in the manuscript preparation process During the preparation of this work, the authors used Gemini and ChatGPT for grammar and English language editing. After using these tools, the authors reviewed and edited the content as needed and take full responsibility for the content of the manuscript. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.