Historically, MBDs were categorized by clinical phenotype or isolated etiologies. However, as the skeleton is increasingly recognized as an endocrine and metabolic organ, classification by core pathological pathways has become essential for understanding disease mechanisms and guiding precision interventions (Karner and Long, 2018; Karsenty and Khosla, 2022). Table 1 summarizes major MBD subtypes according to their dominant pathogenic pathways.

The first category comprises nutritional deficiency–related MBDs. These arise from inadequate intake, malabsorption, or abnormal metabolism of key nutrients, which directly impairs bone matrix mineralization or osteoblast function (Van Gastel and Carmeliet, 2021). Classic examples include osteomalacia in adults and rickets in children due to vitamin D deficiency. Vitamin D regulates intestinal calcium and phosphate absorption and influences the balance between osteoblast differentiation and RANKL/OPG signaling. Severe deficiency leads to defective mineralization of the osteoid, causing bone pain, fractures, and skeletal deformities (Sobh et al., 2022).

The second category involves disorders of mineral ion homeostasis, typically caused by dysregulated renal or intestinal handling of calcium, phosphate, or magnesium (Sprague et al., 2021). For instance, XLH results from PHEX gene mutations that impair renal phosphate reabsorption, leading to hypophosphatemia and defective bone mineralization (Méaux et al., 2022). Similarly, in CKD, reduced phosphate excretion causes hyperphosphatemia. This suppresses 1α-hydroxylase activity, lowers active vitamin D (1,25-(OH)2D) synthesis, and triggers secondary hyperparathyroidism, driving high-turnover bone disease (Tu et al., 2021). These conditions highlight the critical role of the gut–kidney–bone axis in mineral balance.

Furthermore, multiple endocrine hormones regulate bone remodeling equilibrium through direct or indirect pathways (Du Y. et al., 2021). Estrogen deficiency, common in postmenopausal women, leads to increased RANKL expression, decreased OPG, and enhanced osteoclast activation, constituting the primary mechanism of postmenopausal osteoporosis (Hooshiar et al., 2022). Conversely, long-term glucocorticoid exposure inhibits osteoblast proliferation, induces apoptosis, and promotes sclerostin expression. This suppresses the Wnt/β-catenin pathway and causes low-turnover bone loss (Marini et al., 2023). Other conditions, such as hyperthyroidism and abnormalities in the growth hormone/IGF-1 axis, also significantly impact bone turnover rates and quality (Mazziotti et al., 2022).

Simultaneously, bones serve not only as structural support but also as participants in energy metabolism. Insulin resistance, hyperglycemia, and dyslipidemia can impair bone health through multiple mechanisms (Armutcu and McCloskey, 2024). In diabetic bone disease, AGEs accumulate and disrupt collagen cross-linking. Meanwhile, hyperglycemia suppresses osteoblast function and alters the bone marrow microenvironment. Imbalances in adipokines (e.g., adiponectin, leptin) further disrupt remodeling equilibrium, increasing bone fragility even when bone density appears normal (Mangion et al., 2023). Immune-mediated mechanisms also contribute significantly to bone pathology. In chronic inflammatory states, pro-inflammatory cytokines directly target bone cells (Mi et al., 2024). In autoimmune conditions like RA and inflammatory bowel disease, persistently elevated TNF-α, IL-1, and IL-6 stimulate RANKL-driven osteoclastogenesis, causing periarticular bone erosion (Sonmez Kaplan et al., 2023). These cytokines also suppress osteoblast activity, leading to systemic bone loss. Such diseases underscore the importance of the immune–bone axis, or osteoimmunology, in MBD pathogenesis.

Finally, some MBDs result from monogenic mutations that directly disrupt bone matrix formation, mineralization, or cellular function, independent of nutritional or environmental factors. Examples include osteogenesis imperfecta (defective type I collagen synthesis), hypophosphatasia (ALPL mutations causing accumulation of mineralization inhibitors), and XLH (Claeys et al., 2021; Lim et al., 2017). Though rare, these disorders provide crucial insights into bone metabolism and serve as models for targeted therapies such as enzyme replacement or monoclonal antibodies.

In summary, classifying MBDs by pathological metabolic pathways clarifies their heterogeneity. This approach provides a scientific foundation for shifting from symptomatic treatment to mechanism-based precision interventions (Figure 1).

Current mainstream drugs for MBDs, including small-molecule inhibitors and biologics, primarily rely on systemic administration. This presents significant pharmacokinetic and therapeutic challenges.

First, most drugs exhibit poor bone targeting (Lagunas-Rangel et al., 2024). After entering systemic circulation, they distribute widely, with only a tiny fraction penetrating the dense bone matrix to reach active remodeling sites like resorption lacunae or bone-forming surfaces. This inefficiency causes substantial off-target effects in non-skeletal tissues, increasing systemic toxicity risk (Verma et al., 2021). Second, the first-pass effect and low bioavailability pose major hurdles. For orally administered drugs, especially peptides or active vitamin D derivatives, gastrointestinal absorption is followed by rapid hepatic or intestinal metabolism and degradation. This reduces systemic concentrations below therapeutic levels, greatly lowering bioavailability (Lewiecki et al., 2008). Third, gastrointestinal intolerance limits several key medications. For instance, bisphosphonates are poorly absorbed and can irritate the gut, leading to esophagitis, gastric ulcers, and other effects that impair patient comfort and long-term adherence (Oliveira et al., 2024). Finally, patient adherence is a critical issue. As chronic diseases, MBDs require lifelong treatment, but prolonged high-frequency dosing, whether daily/weekly oral or subcutaneous injections, often reduces compliance, diminishing efficacy in reducing fracture risk.

These delivery and pharmacokinetic barriers urgently demand novel smart drug delivery systems capable of overcoming biological obstacles and achieving targeted accumulation in the bone microenvironment (Figure 2).

Based on their structural design and loading/release mechanisms, MN systems fall into five principal engineering types: solid microneedles (S-MNs), hollow microneedles (H-MNs), coated microneedles (C-MNs), dissolving microneedles (D-MNs), and hydrogel-forming microneedles (HF-MNs) (Liu L. et al., 2025).

S-MNs function primarily as pre-treatment tools. They create temporary microchannels in the skin, which are then used for drug application via methods like passive diffusion or electroporation. While their high mechanical strength is a distinct advantage, the requirement for a two-step process limits their convenience. Resembling miniature hypodermic needles, H-MNs feature a hollow interior that allows liquid formulations to be precisely injected into the dermal layer via micro-pumps or osmotic pressure. They are particularly suitable for high-dose or continuous infusion delivery. However, manufacturing complexity and the potential for clogging present technical challenges. C-MNs typically consist of an inert core (e.g., metal or rigid polymer) uniformly coated with a drug-matrix mixture. Upon skin implantation, the coating rapidly dissolves or diffuses into the subcutaneous tissue, leaving the needle body intact for subsequent removal. This design is ideal for the rapid release of small molecules, peptides, or vaccines. These are composed entirely of water-soluble or biodegradable polymers (e.g., hyaluronic acid, polyvinyl alcohol) mixed with the active drug. Upon application, the microneedles completely dissolve in the interstitial fluid, releasing their cargo. The primary advantages of D-MNs are their residue-free delivery and exceptional biosafety, making them an ideal platform for precise, dose-controlled release. HF-MNs represent a novel, “capsule-like” design. Composed of hydrophilic polymers, these needles absorb interstitial fluid upon penetration and swell in situ to form a hydrogel lattice. The drug, which is pre-encapsulated within the matrix, is then released in a sustained manner. By combining residue-free application with controlled-release capabilities, HF-MNs are particularly suitable for chronic disease treatments requiring prolonged therapeutic effects (Figure 3).

The performance, penetration efficiency, and safety of MNs depend strongly on material selection. Polymers are the most widely used class of materials. These include natural polymers, such as hyaluronic acid and chitosan, and synthetic polymers, such as polylactic acid (PLA), polyvinyl alcohol (PVA), and polyglycolic acid (PGA) (Lu et al., 2024). Biodegradable polymers like PLA are advantageous because they permit in vivo metabolism, eliminating the need for needle retrieval. Furthermore, their composition can be tuned to precisely control drug release kinetics (Abbasi et al., 2024).

Metals are commonly used as cores for solid or hollow MNs. Stainless steel and titanium alloys provide the high mechanical strength and sharpness necessary for efficient insertion; however, they must be removed after delivery. Ceramics and silicon offer superior hardness and can be fabricated with ultra-sharp tips, but their brittleness increases the risk of fracture, necessitating careful safety assessments. Additionally, carbon nanomaterials are increasingly used as reinforcers or components of drug-carrier composites to enhance mechanical properties or loading capacity (Razzaghi et al., 2024b).

Biocompatibility is essential for clinical translation. Ideal MN materials must support tissue repair with minimal local injury during insertion and degradation. Critically, they should not provoke significant immune reactions, such as inflammatory cell infiltration (Cao et al., 2019). Dissolving MNs are often considered the safest option because they are fully degradable and leave no sharp residues. Conversely, for non-degradable materials like metals or silicon, rigorous surface finishing and retrieval protocols are required to prevent debris retention and minimize the risk of allergic or foreign body responses.

Mechanical performance is essential for reliable and painless skin penetration by MNs. The most important parameters are insertion force, fracture resistance, and tip sharpness. Insertion force should remain within 0.05–0.3 N per needle: values below 0.05 N often result in incomplete penetration, whereas values above 0.3 N heighten pain perception. Fracture resistance requires a Young’s modulus >2 GPa for stainless-steel MNs or >0.5 GPa for polymeric MNs to prevent buckling. Tip sharpness must achieve a radius <20 µm to stay below the nociceptor activation threshold and ensure painless insertion.

Geriatric skin poses distinct challenges owing to age-related changes: elasticity declines markedly (elastic modulus falls 30%–50% in individuals >70 years), fragility increases (tear resistance decreases ∼40%), and stratum corneum thickness becomes more variable (10–40 µm versus 15–20 µm in young adults). To address these characteristics, MNs for elderly patients require specific adaptations, including shorter needle lengths (150–250 µm) that reliably breach the stratum corneum yet avoid dermal vasculature, tapered geometries that gradually reduce diameter to lower insertion force, and flexible backing materials (e.g., polyurethane films) that conform to skin curvature and movement. Ex vivo studies confirm the effectiveness of these modifications: dissolving MNs with 200 µm height achieved 92% successful insertion into elderly human skin, compared with 98% in younger skin, demonstrating that carefully optimized designs can largely overcome age-related barriers.

Because MNs cross the stratum corneum or mucosal barriers without damaging nerves or blood vessels, they are evolving from simple “transdermal conduits” into intelligent platforms that integrate precise delivery, controlled release, and local microenvironment modulation. This shift is reshaping care for bone and joint diseases. A primary example is the painless delivery of biologics. Therapy for rheumatoid arthritis (RA) often relies on monoclonal antibodies or fusion proteins targeting tumor necrosis factor alpha (TNF-α) or interleukin-6 (IL-6). While effective, subcutaneous injections can trigger needle aversion and treatment interruptions. Cao et al. demonstrated that hyaluronic acid crosslinked MNs could deliver etanercept efficiently without loss of bioactivity (Zhang X. et al., 2025). More recently, Zhang et al. integrated a dual-specific TNF-α/IL-6 receptor “fenobody” into gelatin methacrylate (GelMA) MNs. This system suppressed both nuclear factor kappa B (NF-κB) and Janus kinase/signal transducer and activator of transcription 3 (JAK/STAT3) pathways, achieving injection-like efficacy via noninvasive administration (Du G. et al., 2021).

Beyond delivery, the value of MNs extends to active immunomodulation, with demonstrated effects primarily involving targeted suppression of inflammatory pathways and cellular responses. Du et al. and Jin et al. utilized hyaluronic acid or chondroitin sulfate MNs to deliver melittin. Electrostatic binding reduced toxicity and promoted either regulatory T-cell expansion or synoviocyte apoptosis, effectively shifting the paradigm from passive dosing to active immune remodeling (Jin et al., 2025; Xia et al., 2024). Similarly, Xia et al. designed cerium/manganese oxide nanoparticles (Ce/MnO NPs) with enzyme-mimetic activity as carriers. This transdermal system delivered methotrexate (MTX) to suppress inflammation while driving macrophage polarization from M1 to M2, illustrating a compelling “drug plus carrier” dual-therapy concept (Katsumi et al., 2012). These examples highlight specific, experimentally validated immunomodulatory outcomes, such as cytokine inhibition and macrophage repolarization, though broader osteoimmune mechanisms remain under investigation and require further mechanistic delineation.

MNs are also moving beyond acting as mere “injection replacements” toward becoming long-acting, regionally targeted, and responsive designs. For osteoporosis, Katsumi et al. used tip-loaded MNs to raise the transdermal bioavailability of alendronate above 90%. The system reduced growth plate degeneration in osteoporotic rat models and avoided cutaneous irritation seen with full-length drug-loaded MNs or intradermal injection, indicating strong safety with efficient delivery (Katsumi et al., 2017; Xu et al., 2025). Xu et al. developed a core–shell MN that enabled sustained release of alfacalcidol for up to 14 days. Twice-monthly dosing matched the bone-protective effect of daily oral therapy while lowering the total systemic dose and improving trabecular structure (Mary et al., 2024). Bisphosphonates such as alendronate are often limited by gastrointestinal irritation and very low oral bioavailability (<1%). MN delivery bypasses the gastrointestinal tract, supports noninvasive absorption, and may reduce local inflammatory risk (Uddin et al., 2024). Addressing environmental and manufacturing concerns, Uddin et al. created a 3D-printed hollow MN array (μNe3dles) using a high bio-based, photo-curable resin composed of isobornyl acrylate and pentaerythritol triacrylate in a 1:1 ratio. The device showed excellent mechanical strength and penetration. In an osteoporotic mouse model, this system delivered the monoclonal antibody denosumab with faster in vitro release and superior in vivo recovery of bone mineral indices compared with subcutaneous injection (Hu et al., 2024).

A further advance is the coupling of MNs with externally triggered or self-responsive systems. Hu et al. constructed an extracorporeal shock wave (ESW)-driven nanomotor combined with dissolving MNs (ESW-NM-MN). Zoledronic acid (ZOL) was loaded into calcium phosphate nanomotors with a high Young’s modulus. After MN insertion, ESW activated nanomotor movement into deeper tissues and triggered disassembly, releasing ZOL and Ca²⁺ to suppress osteoclastogenesis and promote osteogenesis. The platform increased local bone density and reduced fracture risk in vivo, supporting regional precision therapy for osteoporosis (Peng et al., 2025). In another responsive design, Peng et al. engineered an ultrasound-responsive dissolving MN system (MTX-PFP-NPs@DMNs). Poly(lactic-co-glycolic acid) (PLGA) nanoparticles carrying MTX and perfluoropentane (PFP) were embedded in hyaluronic acid MNs. Ultrasound induced a PFP phase transition and cavitation that enhanced drug permeation and diffusion. The MNs dissolved within 20 min ex vivo and penetrated the stratum corneum efficiently. In collagen-induced arthritis (CIA) mice, ultrasound plus MNs reduced joint swelling, bone erosion, cartilage damage, and pro-inflammatory mediators, indicating a high-adherence, synergistic strategy for RA (Li et al., 2025). These examples mark a clear transition from static delivery to dynamic, stimulus-responsive therapy.

Finally, MNs can act as local microenvironment modulators, extending from skin to mucosa and specialized bone–soft tissue interfaces. Li et al. developed hydrogel MNs that penetrate gingival mucosa. In a diabetic periodontitis model, the system combined antibacterial and antioxidant actions with inhibition of JAK-STAT and NF-κB signaling, which promoted alveolar bone regeneration (Shan et al., 2025). Shan et al. achieved precise mucosal delivery of RANKL to periodontal tissue. Localized RANKL drove osteoclast formation and accelerated orthodontic tooth movement (Nguyen et al., 2025). These studies introduce innovative transmucosal strategies for early intervention in metabolism-related bone conditions. In summary, the trajectory of MN technology in MBD care is moving from better adherence to superior efficacy, and ultimately to microenvironment remodeling, based on the demonstrated outcomes above. Continued convergence of materials science, nanotechnology, and disease biology may integrate diagnostics, feedback sensing, and adaptive release within MN platforms. Such systems could enable a true “theranostic” paradigm for precise and minimally invasive orthopedics. Hypothetically, the trajectory toward a “theranostic” MN platform for bone health could involve three progressive stages: (i) Current MNs primarily serve as delivery vehicles with passive release kinetics; (ii) Next-generation systems could integrate wearable biosensors (e.g., flexible electrochemical sensors for serum calcium or bone turnover markers like CTX/P1NP in interstitial fluid) with MN patches to enable biomarker-triggered release; (iii) Future “closed-loop” platforms might combine real-time monitoring of bone microenvironment cues (e.g., local pH changes during active resorption, RANKL concentration gradients) with on-demand drug release.

Animal studies and proof-of-concept work show that MN systems can deliver many key agents for MBDs with efficient absorption, particularly for gastrointestinally labile biologics and small molecules that benefit from sustained release. Calcitonin, a 32-amino-acid peptide hormone, is rapidly degraded by digestive enzymes after oral dosing and thus has poor bioavailability, while injections suffer from low adherence. Dissolving MN arrays fabricated from biodegradable polymers such as PLGA or HA have been used to encapsulate calcitonin. These MNs preserve peptide integrity, enable efficient transdermal uptake, and produce pharmacokinetic profiles in rodents comparable to subcutaneous injection. They successfully raise systemic exposure and suppress osteoclast activity, supporting feasibility for systemic antiresorptive therapy (Sim et al., 2022).

Significant progress has also been made with parathyroid hormone analogs. MNs have achieved once-weekly administration of teriparatide acetate (TA). Through formulation optimization (incorporating hyaluronic acid and trehalose) and a bilayer structure design (bottom layer composed of pure hyaluronic acid, top layer loaded with TA), a TA-DMN patch demonstrated excellent puncture capability and rapid drug release (87.6% released within 5 min) in ex vivo porcine skin. In rats, this system achieved 66.9% relative bioavailability. Plasma drug concentrations showed a positive correlation with patch quantity, indicating that this system maintains TA activity while demonstrating potential as a patient-friendly delivery platform for long-acting osteoporosis therapy (Zhang et al., 2020). Similarly, two dissolving MN arrays (sCT-DMNA-1 and sCT-DMNA-2) for salmon calcitonin (sCT) further illustrated this concept. Trehalose improved sCT stability under heat and humidity. In vivo, the patches reached ∼70% relative bioavailability and enhanced transdermal delivery while preserving mechanical performance, offering a safe, efficient, and low-pain alternative (Oh et al., 2022). Additionally, teriparatide MNs formulated with sucrose and carboxymethyl cellulose (CMC) increased the maximum plasma concentration (Cmax) and area under the curve (AUC) in rats. In vitro skin-diffusion data predicted in vivo pharmacokinetics, underscoring that MN dissolution rate is a key lever to tune TA transdermal pharmacokinetics (Yang et al., 2024).

Osteoporosis is a prevalent MBD marked by microarchitectural deterioration and fracture risk. Minodronic acid (MA) is a potent antiresorptive agent, but oral dosing has very low bioavailability, notable adverse effects, and poor adherence. MA-loaded dissolving MNs (MA-MNs) improved pharmacokinetics after load optimization. At a minimal dose of 224.9 μg, MA-MNs increased peak concentration and bioavailability by 9-fold and 25.8-fold, respectively, versus oral MA. Furthermore, they prolonged half-life with more stable plasma levels (Salameh et al., 2024). For nutrition-deficiency related bone disease, long-term, steady supplementation is essential. MN systems address solubility-driven barriers to transdermal vitamin delivery (Hamed et al., 2024). Embedding vitamins within the matrices of coated or dissolving MNs made from water-soluble polymers such as PVA and PVP avoids inter-individual variability in gastrointestinal absorption and batch-to-batch issues seen with oral products. Polymer composition can be tuned to modulate release, enabling sustained delivery over days to weeks (Kim et al., 2025; Pramoda et al., 2025).

MN-mediated systemic exposure is highly dependent on molecular properties and formulation design. Small hydrophilic molecules (<500 Da) primarily enter systemic circulation via dermal capillaries, while larger molecules (>5 kDa) and particulate carriers show preferential uptake by dermal lymphatic vessels (Farooq et al., 2025). This distinction critically impacts pharmacokinetics: lymphatic uptake delays systemic appearance (Tmax 2–6 h vs. 0.5–2 h for capillary route) but may enhance bioavailability for molecules susceptible to first-pass metabolism. For bone-targeted biologics delivered via MNs, dermal metabolism presents an underappreciated barrier. Enzymes such as matrix metalloproteinases (MMP-2/9) and cathepsins in the dermal extracellular matrix can degrade peptide therapeutics before systemic entry. Strategies to mitigate this include (i) incorporation of protease inhibitors (e.g., aprotinin): in MN matrices, (ii) PEGylation of therapeutic peptides to sterically hinder enzymatic access, and (iii) rapid-dissolving formulations that minimize dermal residence time. Relative bioavailability versus oral dosing must be distinguished from absolute bioavailability: while MN delivery of alendronate achieved >90% relative bioavailability versus IV injection in rats (Katsumi et al., 2017), absolute bioavailability versus subcutaneous injection remains formulation-dependent (typically 60%–85% for dissolving MNs versus 100% for SC injection).

MN technology can breach skin or mucosal barriers and enable noninvasive delivery of macromolecules. After entry, however, drugs can still undergo nonspecific distribution with insufficient accumulation in bone and joint tissues, which often have relatively limited perfusion. To address this gap, MN platforms are being integrated with active targeting carriers to create an “entry via MNs, navigation via targeting” paradigm. This approach increases local drug concentration at lesions and enables multi-component, multi-pathway synergy.

Decorating drugs or carriers with affinity ligands enhances binding to bone or joint tissues (Liu T. et al., 2024). Several affinity motifs have been incorporated into MN systems. Liu et al. built MTX@PMs MNs using hyaluronic acid (HA) as a CD44 ligand to guide MTX–loaded polymeric micelles toward activated macrophages in inflamed joints of RA. This strategy improved efficacy and reduced systemic toxicity (Cao et al., 2021). Similarly, Cao et al. conjugated a high-affinity DEK-targeting aptamer (DTA6) to cholesterol and embedded it in hydrogel MNs. The platform systemically targeted multiple inflamed joints inhibited neutrophil extracellular trap (NET) formation, and protected joint structure in CIA models, overcoming the limited reach of intra-articular injection for polyarticular disease (Lei et al., 2024). The Arg-Gly-Asp peptide, which binds integrin receptors enriched in bone-remodeling microenvironments, is another promising bone-targeting ligand with potential use in osteoporosis or fracture repair MNs (Lin et al., 2023).

Beyond ligand functionalization, the introduction of biomimetic and smart responsive carriers further enhances the targeting capabilities and therapeutic dimensions of microneedle systems. Lin et al. coated nonsteroidal anti-inflammatory drug nanoparticles with neutrophil membranes and loaded them into MNs. The biomimetic cloak provided inflammation homing and cytokine adsorption, enabling RA lesion “homing” with local cyclooxygenase-2 inhibition and concurrent microenvironment modulation (Li et al., 2024). Li et al. loaded polydopamine-modified exosomes (PDA@Exo) into MNs. By activating the PI3K/Akt/mTOR pathway, the system cleared reactive oxygen species, sustained cartilage homeostasis, promoted M2 macrophage polarization, and supported bone regeneration, yielding disease-modifying effects in osteoarthritis models (Zhou et al., 2025). Zhou et al. used D-RADA16 self-assembling peptides to form an extracellular matrix (ECM)-like nanonetwork in situ within joints, creating a “cellular safe house” that blocked mitochondrial DNA leakage and the cGAS–STING pathway (An et al., 2022). This strategy reduced inflammation and cartilage degeneration in vitro and in vivo, suggesting a precise, durable, and patient-friendly noninvasive option for chronic inflammatory joint disease.

While bone-targeting ligands, such as bisphosphonates, aspartic acid oligomers, and RGD peptides, enhance skeletal accumulation after systemic administration, their application following MN delivery necessitates pharmacokinetic reevaluation. MN delivery generally produces lower peak plasma concentrations but extends exposure time (with increased AUC0-24 h) relative to bolus injection, a profile that may in fact benefit bone targeting since hydroxyapatite-binding ligands rely on sustained exposure for accumulation at remodeling sites rather than high Cmax. As mentioned above, computational modeling showed that for alendronate-conjugated nanoparticles, MN delivery, with 50% lower Cmax but twice the half-life, yielded bone uptake comparable to intravenous injection, owing to prolonged circulation. Nevertheless, significant knowledge gaps persist, including the predominance of intravenous administration in most bone-targeting studies coupled with a scarcity of MN-specific pharmacokinetic data, potential alterations to ligand conformation or binding affinity due to dermal first-pass effects, and reduced targeting efficiency from competition with endogenous calcium-binding proteins in interstitial fluid.

Notably, multi-drug synergism and differentiated release have become key pathways for enhancing therapeutic efficacy. An et al. created a detachable MN with spatially separated materials, enabling sustained release of tocilizumab (TCZ) and rapid release of a TNF aptamer. Dual blockade of IL-6R and TNF outperformed monotherapy in CIA models (Zheng et al., 2024). Zheng et al. integrated dissolving MNs loaded with melittin and a diclofenac sodium transdermal patch to combine anti-inflammatory and immunoregulatory actions with analgesia. The system alleviated paw swelling and reduced synovial, joint, and cartilage injury in RA models, highlighting potential as a combination modality (Liu T. et al., 2025).

Furthermore, the introduction of active delivery mechanisms is breaking through the limitations of traditional microneedles that rely on passive diffusion. Liu et al. embedded a sodium bicarbonate/citric acid gas-generating system inside MNs. Carbon dioxide bubbles generated impulsive thrust that drove deeper drug penetration and improved transdermal efficiency (Meng et al., 2024). Meng et al. designed a near-infrared responsive PDA/GelMA MN that penetrates the annulus fibrosus and delivers combined photothermal and pharmacologic therapy. The platform mounted an “offensive” anti-inflammatory action extracellularly and induced intracellular heat-shock responses for “defense,” restoring ECM synthesis and biomechanics in intervertebral disc degeneration models (Maguire, 2022). In summary, combining MNs with bone and joint-targeted carriers has shifted the field from a single-mode delivery tool to intelligent platforms that integrate spatial targeting, temporal control, pathway synergy, and microenvironment remodeling (Figure 4). As ligand libraries expand, responsive materials improve, and mechanisms of multi-component cooperation are clarified, such “entry-and-navigation” systems are poised to deliver precise, durable, and individualized therapy for bone and joint diseases.

The primary technical challenge stems from the functional limits imposed by the small geometry of microneedles. The effective volume of an MN is typically in the nanoliter to picoliter range. Yet, systemic indications often require large doses or the delivery of high–molecular weight biologics, such as antibodies. Achieving high drug concentrations within such limited space is difficult. The dose ceiling represents a fundamental constraint for MN translation to MBDs. Typical MN arrays (1 cm²) contain 100–400 needles with individual volumes of 10–100 nL, yielding total payload capacity of 1–40 mg depending on drug solubility and matrix composition. This poses challenges for high-dose MBD therapies: denosumab (60 mg/month), zoledronic acid (5 mg/year IV but requires higher doses for transdermal delivery due to incomplete absorption), and teriparatide (20 µg/day but requires chronic administration). Three engineering strategies show promise: (i) Tip-loaded MNs concentrate drug in the distal 20%–30% of needle length that penetrates skin, increasing effective payload by 3–5× while maintaining mechanical integrity; (ii) Multi-layer MN architectures with drug-loaded core and structural shell enable 2–3× higher loading without compromising insertion force; (iii) Larger patch formats (4–9 cm²) with optimized skin adhesion can deliver clinically relevant doses but face patient acceptability challenges, particularly for geriatric populations with fragile skin. Computational modeling suggests that for denosumab delivery, a 4 cm² MN patch with 5% (w/w) drug loading in a rapidly dissolving matrix could achieve therapeutic plasma concentrations, though this requires validation in large animal models with human-relevant skin thickness. Furthermore, high loading can compromise the structural stability of peptides and proteins, leading to aggregation or denaturation and a subsequent loss of bioactivity. Future work must therefore focus on developing nanostructured carriers with ultra-high loading efficiency and built-in molecular stabilization mechanisms.

A second critical issue is controlling release kinetics. Many MBD therapies, particularly osteoanabolic agents, require sustained or pulsatile release over weeks to months. Conversely, dissolving MNs often exhibit rapid burst release, which exhausts the therapeutic payload in a short time. Achieving near zero-order kinetics will likely require advanced polymer engineering. Potential solutions include the development of multilayer or porous MN architectures, the integration of microfluidic channels, or encapsulation within microcapsules or nanocapsules with staggered degradation rates to precisely tune diffusion and dissolution.

In addition to drug loading and release, mechanical reliability is paramount. MNs must be strong enough to pierce the stratum corneum reliably, yet not so rigid that insertion becomes painful. Moreover, inter-individual differences in skin thickness, elasticity, and hydration produce significant variability in penetration depth and absorption rates. At an industrial scale, product reliability depends on the tight control of tip sharpness, needle height uniformity, and dose content uniformity across batches. Ensuring these parameters are maintained during mass manufacturing is essential for clinical success.

Clinical translation requires a systematic evaluation of patient safety, long-term efficacy, and manufacturing practices. Although MNs are minimally invasive, repeated or prolonged use can increase the risk of local irritation, erythema, or inflammatory reactions. This is especially relevant when non-degradable materials are used. For dissolving or hydrogel MNs, the rate of skin-barrier recovery after complete needle degradation serves as a key safety endpoint. Therefore, robust sterilization and biocompatibility testing are essential to minimize infection risks.

Particular attention must be paid to the biological risks associated with improper application. For instance, frequent microneedling combined with topical bone-marrow–derived cytokines for cosmetic skin care carries potential hazards. Such procedures are often promoted by non-dermatology clinicians or non-physician providers who may lack formal training in cutaneous immunology. MNs can trigger sterile inflammation. If asepsis is inadequate, the dense skin microbiota may enter the dermis through microchannels and amplify local or systemic inflammation. Furthermore, pro-inflammatory bone-marrow–derived cytokines can sustain chronic inflammation. This contrasts with adipose- or skin-derived mesenchymal stromal cell products, which may aid resolution. Consequently, improper use can disrupt cutaneous homeostasis (Meng et al., 2024).

Beyond biological safety, practical usability determines real-world adoption. For MN patches designed to remain in place for hours to days, skin adhesion must withstand routine activities such as movement, perspiration, and bathing to ensure continuous dosing. Patient acceptance also depends on perceived minimal invasiveness, comfort, and ease of self-administration. Ultimately, MN platforms must demonstrate a clear quality-of-life advantage over oral or injectable therapy to achieve widespread adoption.

Finally, manufacturing presents distinct complexities. Current MN fabrication methods, including micromolding and photolithography, lack harmonized guidance under Good Manufacturing Practice (GMP). Regulatory expectations for this novel delivery class are still evolving. This includes undefined standards for quality attributes, in vitro and in vivo performance assays, and requirements for long-term stability. These gaps create significant uncertainty for industrial scale-up.

To date, no MN-based therapies have received regulatory approval specifically for metabolic bone diseases. However, the clinical pipeline provides valuable context: (i) Vaxxas’ high-density MN patch for influenza vaccine completed Phase I/II trials (NCT03251874) demonstrating safety and immunogenicity; (ii) Zosano Pharma’s Qtrypta® (zolmitriptan MN patch) received FDA approval in 2020 for migraine, validating MN technology for systemic delivery; (iii) Micron Biomedical’s dissolving MN vaccine platform entered Phase I trials in 2023 (NCT05687391). For bone-related applications, preclinical studies dominate the landscape. As mentioned above, MN loaded with targeted drugs has shown superior efficacy in rodent osteoporosis models compared to oral/injection routes, but has not yet progressed to IND-enabling studies.

Key translational barriers specific to MBDs include: (i) dose requirements (e.g., denosumab requires 60 mg monthly; current MNs typically deliver <10 mg/patch), (ii) need for chronic administration (>5 years for osteoporosis), and (iii) absence of established regulatory guidelines for MN-biologic combination products targeting skeletal endpoints. The FDA’s 2023 draft guidance on “Transdermal and Topical Delivery Systems” provides partial framework, but bone-specific endpoints (BMD changes, fracture risk reduction) require novel clinical trial designs.