Efficacy and safety of intrathecal human umbilical cord-derived mesenchymal stromal cell therapy for bone cancer pain: a preclinical study

Abstract

Background

Bone cancer pain (BCP) is a severe complication in patients with advanced cancer, for which current therapies remain inadequate and often lead to significant adverse effects. Mesenchymal stromal cell (MSC)-based therapy has recently emerged as a promising strategy for pain management. However, the safety profile of MSC therapy for BCP treatment has not yet been verified in preclinical studies.

Methods

In this study, clinical-grade human umbilical cord-derived MSCs (hUC-MSCs) at doses of 1 × 10⁶ or 4 × 10⁶ were administered intrathecally to BCP rats to evaluate analgesic efficacy, acute toxicity, and in vivo biodistribution. Additional assessments in nude mice investigated the tumorigenicity and tumor-promoting potential of hUC-MSCs.

Results

Intrathecal administration of hUC-MSCs effectively alleviated mechanical allodynia, heat hyperalgesia, spontaneous flinching, and gait abnormalities in BCP rats, while attenuating hyperactivation of neurons and glial cells in the spinal dorsal horn. The higher dose of 4 × 10⁶ hUC-MSCs demonstrated superior analgesic efficacy compared to the 1 × 10⁶ dose. Importantly, comprehensive safety evaluations confirmed a favorable preclinical safety profile for intrathecal hUC-MSC administration, demonstrating absence of acute toxicity, restricted distribution, and no tumorigenicity or tumor-promoting potential.

Conclusions

Intrathecal hUC-MSC therapy demonstrates marked analgesic efficacy and a favorable safety profile in BCP, showing strong potential for clinical translation. This preclinical study provides compelling evidence to support future clinical trials and therapeutic applications of hUC-MSC therapy for BCP.

Background

Cancer pain is one of the most common manifestations of cancer patients, with approximately 65% of patients with advanced cancer experiencing severe pain [1, 2]. Patients with advanced breast, lung, and prostate cancer often experience severe bone destruction and bone cancer pain (BCP) after bone metastases. BCP is one of the most severe and refractory types of chronic pain [3, 4]. The clinical treatment of BCP is mainly drug therapy. The three-step analgesic ladder for cancer pain, which primarily includes non-steroidal anti-inflammatory drugs (NSAIDs) and opioid analgesics, remains the primary clinical treatment for BCP [5]. However, clinical treatment data indicate that about 30% of cancer pain patients who have undergone this therapy do not achieve effective pain relief [1]. Long-term use of NSAIDs can induce several adverse reactions, including coagulation disorders, gastrointestinal ulcers, and perforations, as well as damage to the liver and kidneys. Similarly, long-term use of opioid analgesics carries risks such as respiratory depression, addiction, tolerance, and hyperalgesia [6, 7]. These intolerable adverse reactions prevent effective pain relief for patients with bone cancer, and increasing the dose does not improve the analgesic effect. Currently, there are no effective strategies to mitigate the adverse reactions of traditional analgesics, and no new analgesics have been developed to replace opioid analgesics. This situation represents the biggest obstacle and challenge in the field of cancer pain management. Therefore, there is an urgent need to explore new effective analgesic strategies to improve the quality of life of cancer patients.

As an emerging non-drug therapy in recent years, cell therapy has shown promising efficacy in pain management [8–10]. Mesenchymal stromal cells (MSCs) are adult stem cells derived from mesoderm and have a wide range of tissue sources, including umbilical cord [11], adipose [12], bone marrow [13], gingiva [14], olfactory mucosa [15], and other tissues. Human umbilical cord-derived MSCs (hUC-MSCs) are isolated from umbilical cord tissue obtained after childbirth, which is typically considered medical waste. Compared with MSCs derived from other tissues, hUC-MSCs can be obtained without invasive procedures, involve fewer ethical concerns, and are suitable for standardized large-scale production. Moreover, ​hUC-MSCs have minimal expression of human leukocyte antigen-DR isotype (HLA-DR), endowing them with exceptionally low immunogenicity and a significantly reduced risk of transplant rejection [11, 16, 17]. These advantages make hUC-MSCs the preferred MSC for allogeneic transplantation therapy.

To date, hUC-MSCs have demonstrated remarkable immunomodulatory and tissue-repairing capabilities, leading to their clinical application in various inflammatory and immune-related diseases including rheumatoid arthritis [18], pneumonia [19], acute respiratory distress syndrome [20], systemic lupus erythematosus [21], lupus nephritis [22], and graft-versus-host disease [23, 24]. Additionally, they have shown therapeutic potential for neurological disorders such as hypoxic-ischemic encephalopathy [25], cerebral palsy [26, 27], multiple sclerosis [28], traumatic brain injury [29], and spinal cord injury [30]. In recent years, the potential application of MSC-based therapies in pain management has attracted significant research interest. Clinical trials have reported promising analgesic effects in treating pain such as arthritis and low back pain [8, 31]. Notably, hUC-MSCs exhibit superior analgesic efficacy compared to other MSC sources, such as bone marrow-derived MSCs (BM-MSCs), owing to their unique advantages [32]. However, the application of hUC-MSCs in treating BCP remains at the preclinical research stage.

The safety of hUC-MSC transplantation is a critical prerequisite for clinical application [33]. Although the intrathecal delivery of hUC-MSCs has been demonstrated to be safe and feasible in the clinical treatment of various neurological disorders, the tumor-related characteristics of BCP make it particularly important to validate the safety of hUC-MSC therapy in preclinical studies of the BCP animals [27, 34, 35]. To date, evidence regarding the efficacy and safety of hUC-MSC therapy in BCP animal models remains insufficient. Therefore, this study used clinically grade hUC-MSCs prepared through standardized production processes. hUC-MSCs were administered via intrathecal injection in BCP animals to evaluate the preclinical analgesic efficacy and safety of this therapy. These findings provide critical preclinical evidence to support subsequent clinical trials and facilitate the clinical translation of hUC-MSC therapy for BCP management.

Materials and methods

Animals

Female Sprague-Dawley (SD) rats (8–10 weeks old, 200 ± 20 g) and female BALB/c nude mice (4–6 weeks old, specific pathogen-free grade) were purchased from the Liaoning Changsheng Biotechnology Co., Ltd. (license No. SCXK 2020-0001). Animals were housed in a Specific Pathogen Free environment at the Animal Experimentation Center of Dalian Medical University. Animals were acclimatized for 7 days in polycarbonate cages (3 animals/cage) at 22 ± 2 ℃, 55 ± 5% humidity, 12/12 h light/dark cycle, with free access to standard chow and filtered water. The animals were euthanized by overdose inhalation of isoflurane. No sample size calculation was performed, as sample sizes were chosen based on previous in vivo experiments conducted in the laboratory to ensure sufficient statistical power. Animals were allocated to groups randomly using a random number generator. The order of treatments and measurements for each mouse were randomly assigned. A total of 35 SD rats and 30 BALB/c nude mice were used. To evaluate the preclinical analgesic efficacy and safety of intrathecal hUC-MSC therapy, 35 SD rats were randomly divided into four groups: (1) Sham group (6 mice), (2) BCP group (9 mice), (3) 1 × 10⁶ hUC-MSCs group (10 mice), and (4) 4 × 10⁶ hUC-MSCs group (10 mice). To evaluate the potential tumorigenicity of hUC-MSCs, 18 BALB/c nude mice were randomly divided into three groups: (1) negative control group (PBS group, 6 mice), (2) positive control group (HeLa group, 6 mice), and (3) hUC-MSCs group (6 mice). To evaluate the tumor-promoting effects of hUC-MSCs, 12 BALB/c nude mice were randomly divided into two groups: (1) control group (PBS group, 6 mice) and (2) hUC-MSCs group (6 mice). All animals were used in the analysis. During the outcome evaluation and data analysis stage, the group allocation was made known. All animal experiments were approved by the Animal Ethics Committee of Dalian Medical University (No. AEE24228) and conducted in accordance with the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 80 − 23). All efforts were made to minimize the use of experimental animals and to alleviate their suffering, in strict compliance with the 3R principles. The work has been reported in line with the ARRIVE guidelines 2.0.

Rat BCP model

Walker 256 breast carcinoma cells (CL-0377, Procell, China) were used to establish the rat BCP model. Walker 256 cells were cultured in high glucose (4.5 g/L) Dulbecco’s modified Eagle medium (DMEM, 11965092, Gibco, USA), supplemented with 10% fetal bovine serum (FBS, FSP500, Excell, China), 100 units/mL of penicillin, and 100 µg/mL of streptomycin (15140122, Gibco, USA). Cells were then cultured in the presence of 5% CO₂ at 37 ℃. A total of 5 × 10⁶ ascites tumor cells (0.5 mL) were intraperitoneally injected into SD rats weighing 80 to 100 g. After 6 to 8 days, the ascites was aseptically collected and centrifuged at 300 × g for 5 min. The red blood cells were then removed using a Red Blood Cell Lysis Buffer (R1010, Solarbio, China). The tumor cells were then washed three times with sterile PBS, resuspended in PBS to a final concentration of 1 × 10⁵ cells/µL, and maintained on ice for subsequent model establishment.

Before surgery, all rats underwent baseline assessments of mechanical allodynia and heat hyperalgesia, with only those exhibiting normal thresholds included for BCP model establishment. The establishment of the rat BCP model refers to the previously described methods [36, 37]. Rats were anesthetized with 3% isoflurane, and the left hindlimb was shaved and disinfected with 75% (v/v) ethanol. A superficial incision (0.5–1 cm) was made at the knee joint to expose the patellar ligament. Using a 23-gauge needle inserted into the tibial cavity while avoiding the patellar ligament. Prepared Walker 256 cell suspension (1 × 10⁵ cells/µL, 5 µL) was slowly injected into the tibial cavity using a 10 µL Hamilton microliter syringe. The syringe was maintained in situ for 1 min post-injection to prevent the leakage of tumor cells along the injection track. The injection site was then sealed with sterile bone wax, and the wound was sutured and disinfected with 75% (v/v) ethanol. For Sham group rats, sham surgery was performed by injecting equal volumes of sterile PBS (5 µL) using a similar procedure. Rats that failed to develop pain hypersensitivity post-inoculation of tumor cells, exhibited surgery-related motor dysfunction, or died were excluded from the study.

Preparation and characterization of hUC-MSCs

hUC-MSCs were produced and provided by the Stem Cell Clinical Research Center of the First Affiliated Hospital of Dalian Medical University, as previously described [38]. The quality control of hUC-MSCs was conducted in compliance with Good Manufacturing Practice (GMP) standards and officially certified by China’s National Medical Products Administration (NMPA). hUC-MSCs were characterized prior to use in accordance with the minimal criteria for MSCs established by the International Society for Cell and Gene Therapy (ISCT) [39, 40].

The cryopreserved third-passage hUC-MSCs were cultured in minimum essential medium (MEM, C12571500BT, Gibco, USA) supplemented with 10% fetal bovine serum (FSP500, Excell, China), 100 U/mL penicillin, and 100 µg/mL streptomycin (15140122, Gibco, USA) at 37 ℃ with 5% CO₂. Cell morphology was observed and recorded under a microscope. The surface marker of hUC-MSCs was assessed by flow cytometer (BD FACSAria, USA). hUC-MSCs were collected and incubated at 4 ℃ for 30 min with the following fluorescently labeled antibodies at 4 ℃ for 30 min: APC anti-human CD73 (344005), FITC anti-human CD90 (328107), PE anti-human CD11b (301305), PE anti-human CD19 (302207), PE anti-human CD34 (343505), PE anti-human CD45 (304007), and PE anti-human HLA-DR (307605, Biolegend, USA, 1:100). Multilineage differentiation potential was assessed using osteogenic (XR-hMODM-100, XRBio, China), adipogenic (JY-H1016, SSRCC, China), and chondrogenic (JY-H1017, SSRCC, China) differentiation kits. Differentiation was confirmed by Alizarin Red S (ALIR-10001, OriCell, China), Oil Red O (OILR-10001, OriCell, China), and Alcian Blue 8GX (ALCB-10001, OriCell, China) staining, respectively. For tumorigenicity and tumor-promoting tests, sixth-passage hUC-MSCs were used, while fourth-passage cells were used for all other experiments.

Intrathecal transplantation of hUC-MSCs

BCP rats received a single intrathecal injection of either 1 × 10⁶ or 4 × 10⁶ hUC-MSCs (40 µL). Briefly, rats were anesthetized with 3% isoflurane and positioned prone on a surgical table with a gauze roll placed under the abdomen to arch the lumbar spine for improved intervertebral space identification. After shaving and disinfecting the lumbar skin with 75% (v/v) ethanol, a 50 µL Hamilton syringe was inserted into the L4-L5 intervertebral space. Upon encountering loss of resistance or obtaining cerebrospinal fluid reflux, the hUC-MSC suspension (40 µL) was slowly injected intrathecally. Sham group rats received an equal volume (40 µL) of sterile PBS via the same procedure.

​​Radiographic examination of bone

Radiographic examinations were performed on day 7 after intratibial injection of Walker 256 cells and on day 14 after intrathecal injection of hUC-MSCs. Rats were anesthetized with 3% isoflurane and positioned in ventral recumbency with the left hindlimb extended laterally. Tibial bone destruction was evaluated by radiography using the MultiFocus by Faxitron system (Faxitron Bioptics LLC, USA). The percentage of ​​bone lysis area was calculated as the ratio of the osteolytic area to the total proximal tibial area.

Bone histology

On day 7 after intratibial injection of Walker 256 cells and on day 14 after intrathecal injection of hUC-MSCs, rats were euthanized by overdose inhalation of isoflurane. The left tibiae were collected and fixed in 4% paraformaldehyde (PFA, BL539A, Biosharp, China) for 48 h. Decalcification was performed in EDTA solution (E1171, Solarbio, China) with daily solution changes until no resistance was detected upon needle puncture of the tibial metaphysis. The bone tissues were then dehydrated through a graded ethanol series (70% to 100%), cleared in xylene, and embedded in paraffin (39601006, Leica, Germany). Using a rotary microtome (RM2235, Leica, Germany), 5-µm-thick sagittal sections of trabecular bone were obtained and mounted on poly-L-lysine-coated slides. For hematoxylin-eosin (HE) staining, sections were deparaffinized in xylene and rehydrated through a descending ethanol gradient (100% to 70%). Tissue staining was performed using a HE Staining Kit (C0105S, Beyotime, China), followed by mounting with neutral balsam. Image acquisition and histological analysis were conducted using a bright-field microscope (BX53, Olympus, Japan). The ratio of the tumor area to the total proximal tibial area was calculated.

Behavioral assessments

All behavioral assessments were conducted between 9:00 AM and 3:00 PM by an experimenter blinded to experimental groups. Before baseline testing, rats were acclimated for at least 2 days in a controlled environment with standardized lighting and humidity conditions. Mechanical allodynia, heat hyperalgesia, and spontaneous pain were assessed at specified time points: baseline (BL, 1 day before Walker 256 cell inoculation), on post-inoculation days 1, 3, 5, and 7, and after hUC-MSC injection on days 0, 2, 4, 6, 8, 10, 12, and 14. CatWalk gait analysis was performed on day 7 after hUC-MSCs injection.​​.

Assessment of mechanical allodynia

Mechanical allodynia was assessed using von Frey filaments (Stoelting, USA). Rats were placed in individual plastic compartments on an elevated metal mesh platform and allowed to acclimate for 30 min before testing. A blind experimenter applied a series of von Frey filaments with logarithmically increasing stiffness (1, 1.4, 2, 4, 6, 8, 10, and 15 g) perpendicular to the central plantar surface of the left hind paw for 5–6 s, with 5-minute intervals between applications. A positive response was defined as rapid withdrawal or flinching of the stimulated paw. The 50% paw withdrawal threshold (50% PWT) was determined using Dixon’s up-down method, where the filament strength was decreased following a positive response and increased following a negative response, with five additional stimuli applied after the first response transition [41]. The 50% PWT was calculated using the formula: 50% PWT (g) = 10^(X_f^+κδ), where X_f is the log value of the final von Frey filament used, κ is a tabulated value based on the response pattern, and δ = 0.224, which is the mean logarithmic interval between filaments. Rats unresponsive to the highest filament were recorded as 15 g. Rats with 50% PWT below 4 g were classified as having mechanical allodynia.

Assessment of heat hyperalgesia

Heat hyperalgesia was assessed using the Hargreaves method [42]. Briefly, rats were placed in individual plexiglass chambers and allowed to acclimate for 30 min. A radiant heat source (52 ± 0.2 ℃) was delivered through the glass floor via a plantar test apparatus (IITC Life Science Instruments, USA), with the heat beam focused on the plantar surface of the left hind paw. Paw withdrawal latency (PWL) was measured by an automated timer that initiated upon heat activation and terminated upon detection of withdrawal behaviors, including lifting, licking, or jumping. A 20-second cutoff was enforced to prevent tissue damage. Three PWL measurements were taken for each hind paw at 5-minute intervals, and the mean value was calculated for analysis.

Assessment of spontaneous pain

Spontaneous pain was assessed by quantifying the number of spontaneous flinches (NSF)​​ [43]. Rats were placed in plexiglass chambers, and any lift of the left hind limb unrelated to walking behavior was recorded as a spontaneous flinch. The NSF was counted over a 2-minute observation period. Each rat underwent five repeated assessments at 10-minute intervals, and the mean value was calculated for analysis.

CatWalk gait analysis

CatWalk gait analysis was performed using a CatWalk XT system (Noldus, Netherlands) [44]. Rats were placed at the open end of an enclosed glass platform under a red ceiling light-emitting diode light and were allowed to walk through the walkway voluntarily. A high-speed camera mounted beneath the device captured paw prints during locomotion, with data transmitted to gait analysis software (version 10.0, Noldus, Netherlands). Valid data was defined as a minimum of three consecutive step cycles or complete passes through the walkway. Three gait parameters were identified to evaluate dynamic behaviors associated with BCP: (1) “swing” as the duration of no hind paw contact with glass platform, (2) “max contact area” as the largest print area during hind paw contact with glass platform, and (3) “max contact max intensity” as corresponding to the maximum intensity during maximum hind paw contact [37]. Data were expressed as the percentage of ipsilateral (left)/contralateral (right) hind paw.

Immunofluorescence

On day 7 after intrathecal injection of hUC-MSCs, rats were euthanized by overdose inhalation of isoflurane. Intracardiac perfusion was performed via left ventricular cannulation using sodium chloride physiological solution until liver pallor was observed, followed by systemic perfusion with 4% PFA (BL539A, Biosharp, China). The L3-L5 spinal cord segments were collected and fixed in 4% PFA for 24 h. Tissues were then sequentially dehydrated in 15% and 30% sucrose-PBS solutions at 4 ℃ until ​​they sank​​ to the bottom of the vial. After embedding in OCT compound (BL557A, Biosharp, China), tissues were stored at − 80 ℃. Tissue was cut into 20 μm thick sections using a cryostat (CM1950, Leica, Germany) and mounted on adhesive slides (188105W, Citotest, China). Sections underwent three 5-minute PBS washes, followed by 1-hour blocking at room temperature with 5% goat serum and 0.3% Triton-X 100 (P0096-100, Beyotime, China). Primary antibody incubation was performed overnight at 4 ℃ with: rabbit anti-cellular proto-oncogene Fos (c-Fos, 226008, Synaptic Systems, Germany, 1:1000), rabbit anti-ionized calcium-binding adapter molecule 1 (IBA1, DF6442, Affinity, China, 1:500), and rabbit anti-glial fibrillary acidic protein (GFAP, DF6040, Affinity, China, 1:500). After three additional PBS washes, sections were incubated for 1 h at room temperature with Alexa Fluor 488-conjugated goat anti-rabbit IgG secondary antibody (AB0141, Abways, 1:500). Following final PBS washes, nuclei were counterstained with DAPI-containing mounting medium (ab104139, Abcam, USA) at room temperature. Fluorescent images were acquired using a high-content analysis system (CellVoyage CQ1, Yokogawa, Japan) with excitation wavelengths of 405 nm (blue) and 488 nm (green).

According to previously described methods, the number of immunoreactive cells was counted by experimenters blinded to group allocation [45]. Image acquisition and analysis parameters were kept consistent. A 500 × 400 μm window was positioned with its central axis perpendicular to the dorsal horn boundary and its medial edge aligned to the junction of white and gray matter. The boundaries of the spinal dorsal horn were delineated using ImageJ software (v1.54, NIH, USA), followed by counting immunoreactive cells. The counts of c-Fos-positive neurons, IBA1-positive microglia, and GFAP-positive astrocytes in the dorsal horn were determined by averaging cell counts from four rats per group.

Western blot

On day 7 post-intrathecal hUC-MSC injection, L3-L5 spinal cord tissues were rapidly harvested from deeply anesthetized rats and frozen in liquid nitrogen. Tissues were homogenized at 4 ℃ in RIPA lysis buffer (P0013B, Beyotime, China, 10 µL/mg tissue) containing protease and phosphatase inhibitor cocktail (P1045, Beyotime, China). Homogenates were centrifuged at 12,000 × g for 15 min at 4 ℃, and supernatants containing total protein extracts were collected. Protein concentrations were determined using a bicinchoninic acid (BCA) protein assay kit (ZJ101, Epizyme, China). Samples containing 30 µg protein were denatured at 95 ℃ for 5 min and separated by 10% or 12.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE, PG112/PG113, Epizyme, China). Proteins were electrotransferred onto polyvinylidene fluoride (PVDF) membranes (IPVH00010, Millipore, USA). Membranes were blocked for 2 h at 22–24 ℃ with Tris-buffered saline containing 0.1% Tween-20 (TBST) and 5% skim milk. Subsequently, membranes were incubated overnight at 4 ℃ with the following primary antibodies diluted in blocking buffer: rabbit anti-c-Fos (226008, Synaptic Systems, Germany, 1:1000), rabbit anti-IBA1 (DF6442, Affinity, China, 1:1000), rabbit anti-GFAP (DF6040, Affinity, China, 1:1000), and rabbit anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH, 10494-1-AP, Proteintech, China, 1:5000). After three 10-minute washes with TBST, membranes were incubated for 2 h at room temperature with horseradish peroxidase (HRP)-conjugated goat anti-rabbit secondary antibody (AS014, Abclonal, China, 1:5000). Protein bands were visualized using chemiluminescent HRP substrate (WBKLS0500, Millipore, USA) on a gel imaging system (ChemiDoc XRS+, BIO-RAD, USA), and quantitative analysis was performed using ImageJ software (v1.54, NIH, USA). The relative expression levels of target proteins were normalized to GAPDH. All normalized data were derived from four independent biological replicates.

Acute toxicity test

In the acute toxicity test, female SD rats aged 8–10 weeks were used. After establishing the BCP model, rats were randomly divided into the BCP group, 1 × 10⁶ hUC-MSCs group, and 4 × 10⁶ hUC-MSCs group (n = 6), receiving intrathecal injection of sterile PBS (40 µL), 1 × 10⁶ cells (40 µL), or 4 × 10⁶ cells (40 µL), respectively. During the 14-day post-injection observation period, daily monitoring included assessment of mental status, physical appearance, behavioral changes, mortality, and local tissue reactions at the injection site. Body weight, body temperature, and food intake were measured at BL (1 day before intrathecal injection) and on post-injection days 1, 3, 7, and 14. Blood samples were collected via jugular vein puncture at each time point. Complete blood counts were performed using an automated hematology analyzer (BC-2800, Mindray, China), with particular focus on white blood cells (WBC), lymphocytes (Lym), monocytes (Mon), neutrophils (Neu), basophils (Bas), and eosinophils (Eos). Blood samples were centrifuged at 1,200 × g for 15 min at 4 ℃ to obtain serum. Serum concentrations of inflammatory cytokines, including tumor necrosis factor alpha (TNF-α), interferon-γ (IFN-γ), interleukin (IL)-1β (IL-1β), IL-4, IL-6, and IL-10, were quantified at BL and day 7 post-injection using commercial ELISA kits (RK00029, RK00199, RK00009, RK00040, RK00020, RK00050, Abclonal, China) according to the manufacturer’s instructions, with standard curves generated for each assay. Serum samples collected at each time point were analyzed for liver and kidney function biomarkers using an automated biochemistry analyzer (BS-240 VET, Mindray, China). The measured parameters included: aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), γ-glutamyl transferase (γ-GT), albumin (ALB), total bilirubin (TBIL), direct bilirubin (DBIL), creatinine (CREA), blood urea nitrogen (BUN), and uric acid (UA). At the experimental endpoint on day 14, all rats were euthanized. Major organs, including the heart, lungs, liver, spleen, kidneys, and spinal cord were collected for gross anatomical and histopathological examinations. Acute toxicity responses to intrathecal hUC-MSC administration were comprehensively evaluated based on all the parameters mentioned above.

Biodistribution study

For in vivo imaging of hUC-MSCs, cells were first fluorescently labeled with 1,1’-dioctadecyl-3,3,3’,3’-tetramethylindotricarbocyanine iodide (DiR) dye (22070, AAT Bioquest, USA). Briefly, harvested hUC-MSCs were washed twice with PBS and incubated with 5 µM DiR dye (1 mL/million cells) at 37 ℃ for 30 min. After centrifugation at 400 × g for 5 min to remove unbound dye, DiR-labeled hUC-MSCs were washed twice with PBS. Two groups of rats (n = 6) were intrathecally injected with DiR-labeled hUC-MSCs at doses of 1 × 10⁶ or 4 × 10⁶ cells (40 µL), respectively. Fluorescent signals were monitored using an in vivo imaging system (IVScope 8500, Clinx, China) on days 1, 7, and 14 post-injection. Following the in vivo imaging on day 14, rats were euthanized, and major organs, including the brain, heart, lungs, liver, spleen, and kidneys, were collected for ex vivo fluorescence signal detection.

Tumorigenicity test

The cervical cancer cell line HeLa (CL-0101, Procell, China) was used as a positive control for the tumorigenicity test. HeLa cells were cultured at 37 ℃ with 5% CO₂ in DMEM (11965092, Gibco, USA), supplemented with 10% FBS (FSP500, Excell, China), 100 units/mL of penicillin, and 100 µg/mL of streptomycin (15140122, Gibco, USA). Cells were harvested at 80% confluence and resuspended in PBS to a concentration of 5 × 10⁷ cells/mL for subsequent experimental procedures. For the tumorigenicity test of hUC-MSCs, female BALB/c nude mice aged 4–6 weeks were randomly divided into the negative control group (PBS group), positive control group (HeLa group), and hUC-MSCs group (n = 6), receiving subcutaneous injection of sterile PBS (100 µL), 5 × 10⁶ HeLa cells (5 × 10⁷ cells/mL, 100 µL), or 4 × 10⁶ hUC-MSCs (100 µL), respectively. Body weight was measured before (BL, 1 day before subcutaneous injection) and after subcutaneous injection. Subcutaneous tumor growth was monitored throughout the study. Positive control mice that reached predefined ethical endpoints (tumor ulceration or volume exceeding 2000 mm³) were euthanized by overdose inhalation of isoflurane. Mice in the negative control and hUC-MSCs groups underwent 16-week observation before euthanasia.

Tumor-promoting test

The breast cancer cell line MDA-MB-231 (CL-0150, Procell, China) was used to establish a tumor-bearing nude mouse model. MDA-MB-231 cells were cultured at 37 ℃ with 5% CO₂ in DMEM (11965092, Gibco, USA), supplemented with 10% FBS (FSP500, Excell, China), 100 units/mL of penicillin, and 100 µg/mL of streptomycin (15140122, Gibco, USA). Cells were harvested at 80% confluence and resuspended in PBS to a concentration of 5 × 10⁷ cells/mL for subsequent experimental procedures. For the tumor-promoting test of hUC-MSCs, a tumor-bearing nude mouse model was established using 4–6-week-old female BALB/c nude mice by subcutaneous inoculation of 5 × 10⁶ MDA-MB-231 cells (5 × 10⁷ cells/mL, 100 µL) into the right flank. Tumor dimensions including the longest diameter (L) and the shortest diameter (W) were measured weekly, and tumor volume (V, mm³) was calculated using the formula: 0.5 × L × W² [46]. When tumors reached approximately 100 mm³, tumor-bearing mice were randomly divided into the control group and hUC-MSCs group (n = 6), receiving intrathecal injection of sterile PBS (10 µL) or 1 × 10⁶ hUC-MSCs (10 µL), respectively. Baseline tumor volume (V₀) was recorded before intrathecal injection, followed by serial measurements (Vₜ) to calculate relative tumor volume (RTV = Vₜ/V₀). Upon reaching predefined ethical endpoints (tumor ulceration or volume exceeding 2000 mm³), all nude mice were euthanized by overdose inhalation of isoflurane to terminate the observation period, followed by tumor tissue collection for Ki67 immunohistochemical analysis.

Immunohistochemistry

The left tibiae of BCP rats and tumor tissues from nude mice were fixed in 4% PFA (BL539A, Biosharp, China), paraffin-embedded, and sectioned at 5 μm thickness. Tissue sections were deparaffinized in xylene and rehydrated through a graded ethanol series (100% to 75%). Antigen retrieval was performed by microwave heating in Tris-EDTA buffer (pH 9.0) at 95 ℃ for 10 min. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide for 15 min at room temperature, followed by blocking of nonspecific binding sites with 10% normal goat serum for 30 min. Sections were incubated overnight at 4 ℃ with rabbit anti-Ki67 antibody (GB111499, Servicebio, China, 1:500), with negative controls processed by primary antibody omission. After PBS washes, sections were incubated for 1 h at room temperature with HRP-conjugated goat anti-rabbit secondary antibody (AS014, Abclonal, China, 1:5000). Color development was performed using diaminobenzidine, followed by hematoxylin counterstaining and mounting. Three representative high-power fields (400× magnification) per section were captured using an optical microscope (Eclipse 80i, Nikon, Japan), avoiding necrotic and marginal areas. Image analysis was conducted with ImageJ software (v1.54, NIH, USA), and the Ki67-positive percentage was calculated as the ratio of positively stained tumor cells to total tumor cells.

Statistical analysis

All statistical analyses were performed using GraphPad Prism 10 (GraphPad Software, Inc., USA). Data are presented as mean ± SD. Normality was assessed by the Shapiro-Wilk test, and homogeneity of variance was confirmed by Brown-Forsythe test. For pain behavioral assessments, body weight measurements, and acute toxicity tests, data were analyzed by two-way ANOVA. Šídák’s post hoc test was used for comparisons between two groups, while Tukey’s post hoc test was applied for multiple comparisons among three or more groups. CatWalk gait analysis, Western blot, and immunofluorescence staining were evaluated by one-way ANOVA with Tukey’s or Dunnett’s post hoc test. For non-normally distributed data, the Mann-Whitney U test was used for two-group comparisons. Statistical significance was set at P < 0.05 (*P < 0.05, **P < 0.01, ***P < 0.001, ns, not significant).

Results

Characterization of hUC-MSCs

The hUC-MSCs used in this study were characterized by flow cytometry, which showed positive expression of CD73 and CD90 (> 98%) and negative expression of CD11b, CD19, CD34, CD45, and HLA-DR (< 3%) (Fig. 1A). Trilineage differentiation assays demonstrated that the hUC-MSCs formed Alizarin Red S-positive calcified nodules after 21 days of osteogenic induction, Oil Red O-positive lipid vacuoles after 14 days of adipogenic induction, and Alcian blue-positive proteoglycan deposition after 28 days of chondrogenic induction (Fig. 1B–D). These results confirmed that the hUC-MSCs met the ISCT criteria for MSC identification.

Fig. 1.

Characterization of hUC-MSCs. A Flow cytometric analysis of hUC-MSC surface markers. Data were normalized to mode. B Osteogenic differentiation of hUC-MSC with Alizarin Red S staining. Scale bar 50 μm. C Adipogenic differentiation of hUC-MSC with Oil Red O staining. Scale bar 50 μm. D Chondrogenic differentiation of hUC-MSC with Alcian blue staining. Scale bar 200 μm

Establishment of a BCP rat model

Implantation of tumor cells into the tibia is a classical method for establishing BCP models. In this study, a rat BCP model was established by injecting Walker 256 cells into the left tibial cavity. After inoculation, bone destruction was verified through radiography and HE staining, while nociceptive behaviors were assessed via mechanical allodynia, heat hyperalgesia, and spontaneous flinching (Fig. 2A). On day 7 post-inoculation, radiographic examination of tibia revealed uniform, continuous, and intact bone structure in the Sham group, whereas the BCP group exhibited extensive radiolucent lesions and significant bone defects in the proximal tibia (Fig. 2B). HE staining demonstrated well-organized trabecular bone with abundant erythrocytes and lymphocytes in the medullary cavity of Sham rats. In contrast, BCP rats showed marked osteolytic resorption pits in trabeculae and massive tumor cell infiltration in the marrow cavity (Fig. 2C). These results indicated that intratibial inoculation of Walker 256 cells induced tumor-associated osteolytic destruction by day 7, a hallmark of this cell line’s bone-invasive phenotype.

Fig. 2.

Establishment of a BCP Rat Model. A Experimental design to validate the BCP rat model. B​​ Radiographic examination of the left tibial bone on day 7 after intratibial injection of Walker 256 cells. ​​C​​ HE staining of the left tibial bone on day 7 after intratibial injection of Walker 256 cells. Blue arrows: bone resorption pits. Black dashed circle: tumor cell infiltration. Scale bars: 200 μm (top row), 50 μm (bottom row). ​​D–F​​ Assessment of mechanical allodynia (von Frey test), thermal hyperalgesia (Hargreaves test), and spontaneous pain (flinching) at BL (1 day before Walker 256 cell inoculation) and on post-inoculation days 1, 3, 5, and 7. Data are presented as mean ± SD. Two-way ANOVA with Šídák’s post hoc test, ***P < 0.001, n = 6 per group

In addition, behavioral assessments demonstrated that Sham rats exhibited transient reductions in PWT and PWL, along with a temporary increase in NSF during post-inoculation days 1–3, followed by gradual recovery to baseline levels by days 3–7. These transient changes were likely attributable to drill-induced tibial trauma during the sham procedure. In contrast, BCP rats exhibited progressive declines in PWT and PWL, accompanied by an escalating increase in NSF during post-inoculation days 1–7, achieving a statistically significant difference from Sham rats by day 7 (Fig. 2D–F). The sustained nociceptive behaviors coupled with tumor-associated osteolytic destruction represent pathognomonic pathological features of BCP. Therefore, the rat BCP model was successfully established.

Intrathecal hUC-MSC therapy alleviated pain in BCP rats

The analgesic effect of intrathecal injection of hUC-MSCs on BCP in rats was evaluated through pain-related behavioral assessments. Stable BCP was established by post-inoculation day 7, which was selected as the intervention time point for intrathecal hUC-MSC administration. Following intrathecal injection of 1 × 10⁶ or 4 × 10⁶ hUC-MSCs, BCP rats underwent serial measurements of PWT, PWL, and NSF (Fig. 3A). Compared to the BCP group, rats receiving intrathecal injection of 1 × 10⁶ hUC-MSCs demonstrated significantly increased PWT from days 2 to 10 post-injection, elevated PWL during days 2–12, and reduced NSF throughout the same period. The 4 × 10⁶ hUC-MSCs group exhibited more pronounced analgesic effects, with significantly higher PWT and PWL alongside lower NSF across the entire observation window of days 2–12. Furthermore, comparative analysis between 1 × 10⁶ and 4 × 10⁶ hUC-MSCs groups revealed that the 4 × 10⁶ group showed superior PWT improvement versus the 1 × 10⁶ group by day 6, while maintaining significant advantages in both PWL and NSF parameters from days 2 through 12 post-injection. These findings collectively indicated that intrathecal injection of hUC-MSCs effectively alleviated BCP in rats, with analgesic effects persisting for 10–12 days post-injection. Notably, the 4 × 10⁶ hUC-MSC dose exhibited superior analgesic efficacy compared to the 1 × 10⁶ dose (Fig. 3B–D).

Fig. 3.

Intrathecal hUC-MSC administration alleviated pain in BCP rats. A Experimental design to evaluate the analgesic efficacy of intrathecal hUC-MSC therapy in BCP rats. B–D​​ Assessment of mechanical allodynia (von Frey test), thermal hyperalgesia (Hargreaves test), and spontaneous pain (flinching) before intrathecal hUC-MSC injection (day 0) and on post-injection days 2, 4, 6, 8, 10, 12, and 14. Data are presented as mean ± SD. Two-way ANOVA with Tukey post hoc test, ***P < 0.001 versus Sham group, ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 versus BCP group, ^&P < 0.05, ^&&P < 0.01, ^&&&P < 0.001 versus 1 × 10⁶ hUC-MSCs group, n = 6 per group

CatWalk gait analysis was performed on day 7 after intrathecal injection to evaluate pain-related behavioral changes in rats. Three significantly altered parameters in BCP rats were selected for analysis, including swing, max contact area, and max contact max intensity. In Sham group rats, the percentage of ipsilateral (left)/contralateral (right) hind paw for swing, max contact area, and max contact max intensity were all approximately 100% (Fig. 4A, E–G). BCP rats showed significantly increased swing percentage and significantly decreased max contact area and contact max intensity percentages, representing pain-induced locomotor adaptations (Fig. 4B, E–G). Intrathecal administration of either 1 × 10⁶ or 4 × 10⁶ hUC-MSCs significantly reversed the alterations in these three parameters in BCP rats. The 4 × 10⁶ hUC-MSC dose demonstrated a more pronounced reduction in swing percentage compared to the 1 × 10⁶ dose (Fig. 4C–G).

Fig. 4.

Intrathecal hUC-MSC administration ameliorated gait abnormalities in BCP rats. A–D Representative CatWalk gait analyses, including print view, print area, and print intensity, on day 7 post-intrathecal injection, in Sham, BCP, 1 × 10⁶ hUC-MSCs, and 4 × 10⁶ hUC-MSCs groups. E Swing. F Max contact area. G Max contact max intensity. Data were calculated as the percentage of ipsilateral (left)/contralateral (right) hind paw. Data are presented as mean ± SD. One-way ANOVA with Tukey post hoc test, *P < 0.05, **P < 0.01, n = 6 per group. LF, left front. LH, left hind. RF, right front. RH, right hind

In summary, the results from PWT, PWL, NSF, and CatWalk gait analysis collectively demonstrated that intrathecal administration of either 1 × 10⁶ or 4 × 10⁶ hUC-MSCs effectively alleviated pain in BCP rats. Notably, the higher dose of 4 × 10⁶ hUC-MSCs exhibited superior analgesic efficacy compared to the lower dose.

Intrathecal hUC-MSC therapy attenuated spinal neuronal and glial hyperactivation in BCP rats

Abnormal activation of neurons and glial cells in the spinal dorsal horn contributes to the development and maintenance of BCP. To investigate the effects of intrathecal hUC-MSC administration on the activation of spinal dorsal horn neurons and glial cells in BCP rats, we performed immunofluorescence staining and Western blot analysis on day 7 post-injection to examine the expression of neuronal activation marker c-Fos, microglial marker IBA1, and astrocytic marker GFAP in the spinal dorsal horn. Immunofluorescence results demonstrated significantly increased fluorescence intensity of c-Fos, IBA1, and GFAP in the spinal dorsal horn of BCP rats compared to Sham rats. Intrathecal injection of 1 × 10⁶ hUC-MSCs significantly reduced the fluorescence intensity of these markers in BCP rats. Notably, the 4 × 10⁶ hUC-MSC dose exhibited stronger inhibitory effects on c-Fos, IBA1, and GFAP fluorescence intensity compared to the 1 × 10⁶ dose (Fig. 5A–D). Western blot results corroborated these findings, showing significantly elevated protein expression levels of c-Fos, IBA1, and GFAP in BCP rats compared to Sham rats. Intrathecal injection of ​​1 × 10⁶ hUC-MSCs​​ significantly reduced the expression of ​​c-Fos​​ and ​​IBA1​​, but not ​​GFAP​​, in BCP rats. In contrast, ​​4 × 10⁶ hUC-MSCs​​ significantly decreased the expression of ​​c-Fos, ​​IBA1​​, and ​​GFAP​​, demonstrating a stronger therapeutic effect than the lower dose (​​Fig. 5E–H​​). These results indicated that intrathecal administration of either ​​1 × 10⁶ or 4 × 10⁶ hUC-MSCs​​ attenuated hyperactivation of neurons and glial cells in the spinal dorsal horn of BCP rats, with the higher dose of ​​4 × 10⁶​​ hUC-MSCs exerting more potent inhibitory effects.

Fig. 5.

Intrathecal hUC-MSC administration attenuated spinal neuronal and glial hyperactivation in BCP rats. A Immunofluorescence staining of c-Fos, IBA1, and GFAP in the rat spinal dorsal horn on day 7 after intrathecal injection of hUC-MSCs. Scale bar 100 μm. B Quantitative analysis of the number of c-Fos positive neurons. C Quantitative analysis of the number of IBA1 positive cells. D Quantitative analysis of the number of GFAP positive cells. E Western blot analysis of c-Fos, IBA1, and GFAP protein levels in the rat spinal dorsal horn on day 7 after intrathecal injection of hUC-MSCs. F Relative c-Fos protein level normalized to GAPDH. G Relative IBA1 protein level normalized to GAPDH. H Relative GFAP protein level normalized to GAPDH. Data are presented as mean ± SD. One-way ANOVA with Dunnett post hoc test, *P < 0.05, **P < 0.01, ***P < 0.001 versus BCP group, n = 4 per group

Intrathecal hUC-MSC therapy had no significant acute toxicity in BCP rats

To evaluate the potential toxicity of intrathecal hUC-MSC administration in BCP rats, we conducted assessments using low-dose 1 × 10⁶ and high-dose 4 × 10⁶ cell injections (Fig. 6A). During the 14-day observation period, all rats in each group maintained a normal mental status, showing no observable abnormalities in appearance or behavior, with no recorded mortality. The injection site showed no skin damage or tissue injury. Longitudinal monitoring indicated physiological weight gain compared to pre-injection baselines in all groups, with body temperature and food intake fluctuating within normal ranges. Comparative analyses indicated no statistically significant differences in these parameters between the BCP group and either hUC-MSC dose group at any time points, nor between the two hUC-MSC dose groups themselves (Fig. 6B–D). Hematological profiling demonstrated comparable levels of WBC, Lym, Mon, Neu, Bas, and Eos between the 1× 10⁶ hUC-MSCs group and BCP group across all time points. While the 4 × 10⁶ hUC-MSCs group exhibited transient increases in Mon and Neu counts on day 3 post-injection, along with elevated Mon levels on day 7 compared to BCP controls, all these hematological parameters returned to baseline levels by day 14 (Fig. 6E). Serum cytokine analysis on day 7 showed no statistically significant differences in the concentrations of TNF-α, IFN-γ, IL-1β, IL-4, IL-6, or IL-10 between the BCP, 1 × 10⁶ hUC-MSCs, and 4 × 10⁶ hUC-MSCs groups (Fig. 6F). Liver and kidney function tests revealed no statistically significant differences in the level of ALT, AST, ALP, γ-GT, TBIL, DBIL, ALB, CREA, BUN, and UA between either the 1 × 10⁶ or 4 × 10⁶ hUC-MSCs groups and the BCP group at any post-injection time points (days 1, 3, 7, and 14). Additionally, comparative analysis between the two hUC-MSC dose groups showed no statistically significant differences in these ten biochemical parameters throughout the observation period (Fig. 7A). Gross anatomical and histopathological examinations of major organs, including the heart, lungs, liver, spleen, kidneys, and spinal cord were performed at the end of the 14-day observation period following intrathecal injection. The results showed no significant abnormalities in organ morphology, texture, color, or histological features as evaluated by HE staining across all three groups (Fig. 7B, C). In summary, intrathecal administration of either 1 × 10⁶ or 4 × 10⁶ hUC-MSCs did not induce any observable behavioral abnormalities, systemic inflammatory responses, or pathological organ damage during the 14-day observation period in BCP rats, with no treatment-related mortality recorded. These findings collectively demonstrated that intrathecal hUC-MSC therapy exhibited no significant acute toxicity in BCP rats.

Fig. 6.

Acute toxicity evaluation of intrathecal hUC-MSC administration in BCP rats.​​ A Experimental design to evaluate the safety of intrathecal hUC-MSC therapy in BCP rats. B–D​​ Body weight, temperature, and food intake of rats at BL (1 day before intrathecal injection) and on days 1, 3, 7, 14 post-injection. E​​ Peripheral blood leukocyte counts of WBC, Lym, Mon, Neu, Bas, and Eos at BL and on days 1, 3, 7, 14 post-injection. ​​F​​ Serum cytokine concentrations of TNF-α, IFN-γ, IL-1β, IL-4, IL-6, IL-10 at BL and day 7 post-injection. Data are presented as mean ± SD. Two-way ANOVA with Tukey post hoc test, *P < 0.05, **P < 0.01, ***P < 0.001 versus Sham group, ^#P < 0.05, ^##P < 0.01 versus BCP group, n = 6 per group

Fig. 7.

Organ toxicity assessment of intrathecal hUC-MSC administration in BCP rats. ​​A​​ Serum levels of AST, ALT, ALP, γ-GT, ALB, TBIL, DBIL, CREA, BUN, and UA at BL (1 day before intrathecal injection) and on days 1, 3, 7, 14 post-injection. B​​ Gross morphology of heart, lungs, liver, spleen, and kidneys on day 14 post-injection. ​​C​​ HE staining of heart, lungs, liver, spleen, kidneys, and spinal cord on day 14 post-injection. Scale bar 100 μm. Data are presented as mean ± SD. Two-way ANOVA with Tukey post hoc test, *P < 0.05, **P < 0.01, ***P < 0.001 versus Sham group, n = 6 per group

Biodistribution of intrathecal hUC-MSCs in BCP rats

To assess the migration and distribution of intrathecally injected hUC-MSCs in BCP rats, we pre-labeled hUC-MSCs with DiR dye. Subsequently, we injected either 1 × 10⁶ or 4 × 10⁶ DiR-labeled hUC-MSCs intrathecally into BCP rats. In vivo imaging results from both dose groups demonstrated that on day 1 post-injection, DiR-labeled hUC-MSCs were predominantly localized to the intrathecal space surrounding the injection site. By day 7, while the cells remained primarily restricted to the original injection area, the fluorescence signal intensity and distribution area exhibited significant attenuation compared to day 1. This spatial restriction pattern persisted through day 14, with further ​​reduced​​ signal intensity and spread area, but no significant change in anatomical distribution. The 4 × 10⁶ dose group exhibited slightly greater fluorescence intensity and distribution area compared to the 1 × 10⁶ group at all matched time points (Fig. 8A). Importantly, no fluorescent signals accumulation was detected in major organs, including the brain, heart, lungs, liver, spleen, or kidneys through day 14 post-injection in either dose group (Fig. 8B). These findings collectively ​​indicated​​ that intrathecally delivered hUC-MSCs ​​exhibited​​ limited distribution within the intrathecal space, with no detectable systemic distribution or migration to major organs.

Fig. 8.

Biodistribution of intrathecal hUC-MSCs in BCP rats. A​​ In vivo fluorescence imaging of rats on days 1, 7, 14 post-injection. B​​ Ex vivo fluorescence signal detection of brain, heart, lungs, liver, spleen, and kidneys on day 14 post-injection

hUC-MSC therapy had no tumorigenicity

To evaluate the potential tumorigenicity of hUC-MSCs, we subcutaneously transplanted a high dose of 4 × 10⁶ hUC-MSCs into BALB/c nude mice and monitored tumor formation (Fig. 9A). The results demonstrated that neither the negative control group (PBS group) nor the hUC-MSCs group exhibited tumor formation during the 16-week observation period (tumorigenicity rate: 0% in both groups). Both groups showed physiological weight gain with no statistically significant differences at any time point. In contrast, all mice in the positive control group (HeLa group) that received HeLa cell transplantation developed tumors (tumorigenicity rate: 100%) and reached ethical endpoints by week 12 post-transplantation. The HeLa group exhibited significantly lower body weight compared to both the PBS group and hUC-MSCs group from weeks 4 to 12, correlating with tumor progression (Fig. 9B, C). These findings conclusively demonstrated that subcutaneous transplantation of 4 × 10⁶ hUC-MSCs in BALB/c nude mice had no tumorigenic potential.

Fig. 9.

Tumorigenicity test of hUC-MSCs in nude mice. A Experimental design for tumorigenicity test. B​​ Representative photographs of nude mice at week 12 post-injection in negative control (PBS), positive control (Hela), and hUC-MSCs groups. C Body weight of nude mice at BL (1 day before subcutaneous injection) and weeks 2, 4, 6, 8, 12 post-subcutaneous injection. Data are presented as mean ± SD. Two-way ANOVA with Tukey post hoc test, *P < 0.05, ***P < 0.001 versus Sham group, ^###P < 0.01 versus hUC-MSCs group, n = 6 per group

Intrathecal hUC-MSC therapy has no tumor-promoting effects in BCP rats

To evaluate whether intrathecal injection of hUC-MSCs promotes the proliferation of tumor tissue within the tibia of BCP rats, radiographic examination, HE staining, and Ki67 immunohistochemical staining were performed on the left tibia of BCP rats on day 14 after intrathecal injection of high-dose 4 × 10⁶ hUC-MSCs. The percentage of ​​bone lysis area, the ratio of the tumor area to the total proximal tibial area, and the percentage of Ki67-positive cells were calculated. Radiographic examination of the tibia revealed radiolucent lesions and significant bone defects in the proximal tibia of both the BCP group and the 4 × 10⁶ hUC-MSCs group on day 14 after intrathecal injection. There were no significant differences in the percentage of bone lysis area between the two groups (Fig. 10A, B). HE staining of the tibia showed marked osteolytic resorption pits in the trabecular bone and massive tumor cell infiltration in the marrow cavity in two groups. No significant differences were observed in the ratio of the tumor area to the total proximal tibial area between the two groups (Fig. 10C, D). Ki67 immunohistochemical staining also demonstrated no significant differences in the percentage of Ki67-positive cells in the tibia between the BCP group and the 4 × 10⁶ hUC-MSCs group (Fig. 10E, F). These results indicated that intrathecal injection of high-dose 4 × 10⁶ hUC-MSCs did not promote the proliferation of tumor tissue or exacerbate bone destruction in the tibia of BCP rats. Thus, intrathecal delivery of 4 × 10⁶ hUC-MSCs exhibited no tumor-promoting effects in the BCP rat model.

Fig. 10.

Tumor-promoting test of intrathecal hUC-MSC administration in BCP rats. A​​ Radiographic examination of the left tibial bone on day 14 after intrathecal injection of hUC-MSCs. ​​B Quantitative analysis of the percentages of bone lysis area. C​​ HE staining of the left tibial bone on day 14 after intrathecal injection of hUC-MSCs. Black dashed circle: tumor cell infiltration. Scale bars: 500 μm (top row), 200 μm (bottom row). D Quantitative analysis of the ratio of the tumor area to the total proximal tibial area​​​​. ​​E​​ Ki67 immunohistochemical staining of the left tibial bone. Scale bar 100 μm. F​​ Quantitative analysis of the percentages of Ki67-positive cells. Data are presented as mean ± SD. Mann-Whitney U test, ns, no statistical significance, n = 3 per group

Intrathecal hUC-MSC therapy has no tumor-promoting effects in tumor-bearing nude mice

To further investigate the tumor-promoting potential of intrathecal hUC-MSC administration under conditions that simulate the immunosuppressed microenvironment of clinical cancer patients, a tumor-bearing nude mouse model was established using the human breast cancer cell line MDA-MB-231. The maximum intrathecal capacity dose of 1 × 10⁶ hUC-MSCs was injected to evaluate its potential effects on the proliferation of human-derived tumor tissue in an immunodeficient, tumor-bearing context (Fig. 11A). Results demonstrated that tumor-bearing mice in both the control and hUC-MSCs groups exhibited progressive weight gain at weeks 2, 4, and 6 post-intrathecal injection, with no statistically significant differences between the groups at any time point (Fig. 11B, C). Upon reaching ethical endpoints at week 6 post-intrathecal injection (corresponding to week 12 post-tumor inoculation), mice were euthanized for tumor collection. Comparative analyses revealed no statistically significant differences in tumor volume or RTV progression between hUC-MSC and PBS groups (Fig. 11D–F). Furthermore, immunohistochemical staining for Ki67 showed comparable percentages of Ki67-positive cells in tumor tissues from both groups (Fig. 11G, H). Collectively, these findings indicated that intrathecal delivery of 1 × 10⁶ hUC-MSCs did not exert tumor-promoting effects on subcutaneously implanted MDA-MB-231 breast cancer in nude mice.

Fig. 11.

Tumor-promoting test of intrathecal hUC-MSC administration in tumor-bearing nude mice. A Experimental design for tumor-promoting test. B​​ Representative photographs of nude mice from PBS group and hUC-MSCs group. C Body weight of nude mice at BL (1 day before intrathecal injection) and weeks 2, 4, 6 post-intrathecal injection. ​​​​D​​ Gross morphology and dimensional measurements of tumor tissues. ​​E​​ Tumor volume. ​​F​​ Relative tumor volume (RTV). ​​G​​ Ki67 immunohistochemical staining of tumor tissues. Scale bar 100 μm. ​​H​​ Quantitative analysis of percentages of Ki67-positive cells. Data are presented as mean ± SD. Two-way ANOVA with Šídák’s post hoc test for (C), Mann-Whitney U test for (E, F, H), ns, no statistical significance, n = 6 per group

Discussion

In this study, we aimed to investigate the preclinical efficacy and safety of intrathecally administered clinical-grade hUC-MSCs for BCP management. Prior to this study, no preclinical safety evaluations of hUC-MSCs for cancer pain treatment had been reported. We conducted analgesic efficacy studies in BCP rats, evaluated acute toxicity and in vivo biodistribution of intrathecal hUC-MSC administration. Tumorigenicity and tumor-promoting potential of hUC-MSC were assessed using BCP rats and nude mice. Our results demonstrated that intrathecal hUC-MSC administration effectively alleviated pain in BCP rats and attenuated hyperactivation of spinal neurons and glial cells. Importantly, intrathecal administration of hUC-MSCs showed a favorable preclinical safety profile without tumorigenicity or tumor-promoting effects. Overall, these findings indicated that intrathecal hUC-MSC therapy represents a promising novel analgesic treatment for BCP with good clinical translation prospects.

The choice of administration route for MSC-based therapies is critical and depends on the pathophysiology of the target disease. Intravenous, intrathecal, and local injections, such as intra-articular or intradiscal injections, are commonly used delivery routes for MSCs in the treatment of various diseases and pain conditions [8, 31]. For BCP and other types of pain, the spinal cord serves as a key hub for pain generation and transmission, making intrathecal delivery of hUC-MSCs directly to the perispinal region an effective strategy [10, 47]. Additionally, since BCP possesses distinct tumor-associated characteristics compared to other pain types, selecting the optimal delivery route for hUC-MSCs requires comprehensive consideration of not only analgesic efficacy but also safety and clinical translatability. First, from the perspectives of analgesic efficacy and clinical safety, although intravenous injection is minimally invasive, intravenously delivered hUC-MSCs undergo significant pulmonary retention and require inefficient long-distance migration to reach the central nervous system, which may compromise their analgesic efficacy [48]. More importantly, intravenously injected hUC-MSCs can enter the bone tumor microenvironment via systemic circulation, potentially interacting directly with tumor tissue. Similarly, hUC-MSCs administered via local or intratumoral injection would also come into direct contact with tumor tissue. These delivery routes raise serious concerns about potential tumor-promoting effects, which may compromise clinical safety. In contrast, intrathecal injection delivers hUC-MSCs directly to the perispinal area, significantly improving their targeting of the central nervous system and providing enhanced analgesic effectiveness. Our in vivo imaging results demonstrated that intrathecally delivered hUC-MSCs exhibited a limited range of intrathecal localization. This finding is consistent with a previous study that intrathecally injected bone marrow-derived MSCs in rats with neuropathic pain treatment [47]. This restricted distribution prevents direct contact between hUC-MSCs and tumor tissues, thereby alleviating related safety concerns. Second, from a clinical practice standpoint, intrathecal delivery, including implanted intrathecal drug delivery systems (IDDSs), represents a well-established, effective, and generally well-tolerated clinical analgesic strategy for refractory cancer pain and chronic pain poorly controlled by conventional systemic analgesics [49, 50]. While intrathecal injection is undoubtedly more invasive than intravenous injection and requires specialized skills, extensive clinical experience in pain management suggests that its benefits in analgesic efficacy may outweigh the procedural complexities [50, 51]. Recent Phase II clinical trials investigating intrathecal BM-MSCs for neuropathic pain in complete spinal cord injury also provide evidence supporting the clinical feasibility and effectiveness of this route for MSC-based pain therapy [52]. Future clinical translation of intrathecal hUC-MSC therapy could build upon existing intrathecal drug delivery protocols and clinical experience to facilitate its application in BCP management. Additionally, there have been studies on the intranasal delivery of MSCs to treat chemotherapy-induced peripheral neuropathy and pain [53]. Although intranasal administration of MSCs shows effective analgesic effects in mice, significant differences in the distribution, area, and permeability of the human olfactory epithelium may pose limitations and challenges for the clinical application of intranasal MSC delivery [54]. In summary, when comprehensively considering analgesic efficacy, safety, and clinical translatability, intrathecal injection emerges as a highly promising and clinically viable delivery route for hUC-MSC therapy in BCP.

Previous studies have reported that intrathecally administered MSCs and their exosomes exhibit dose-dependent analgesic efficacy for neuropathic pain [55, 56]. Therefore, we speculated that the analgesic effect of intrathecal injection of hUC-MSCs on BCP would exhibit dose dependency​​. In our study, we compared the analgesic effect of 1 × 10⁶ or 4 × 10⁶ ​​hUC-MSCs per rat​​ in BCP rats. The lower dose of 1 × 10⁶ hUC-MSCs reversed both mechanical allodynia and heat hyperalgesia until post-injection day 10, whereas the higher dose of 4 × 10⁶ hUC-MSCs extended this therapeutic effect through day 12. The CatWalk gait analysis system, which records spontaneous locomotion in animals, has been established as a reliable method for detecting pain-related behaviors [44, 57]. Assessment of spontaneous gait alterations associated with pain can effectively evaluate the analgesic efficacy of hUC-MSCs in BCP rats. Our results showed that intrathecal administration of either 1 × 10⁶ or 4 × 10⁶ hUC-MSCs significantly reversed the abnormal gait parameters induced by BCP, including swing, max contact area, and max contact max intensity. Notably, the 4 × 10⁶ hUC-MSC dose exhibited superior efficacy compared to the lower dose.​ In addition, the development and maintenance of BCP depend on the sensitization of spinal dorsal horn neurons and abnormal activation of glial cells [58–60]. Our results showed that intrathecal injection of hUC-MSCs effectively attenuated the hyperactivation of neurons, microglia, and astrocytes in the spinal dorsal horn of BCP rats, and the inhibitory effect of 4 × 10⁶ hUC-MSCs was stronger than that of the 1 × 10⁶ dose. Consequently, both doses of intrathecally administered hUC-MSCs significantly alleviated the pain of BCP rats, with the 4 × 10⁶ dose exhibiting superior analgesic efficacy compared to the 1 × 10⁶ dose. Although this study only compared two doses of hUC-MSCs, the results suggested that the analgesic effect of hUC-MSCs on BCP would follow a dose-dependent pattern within a certain dosage range, consistent with their effects on other types of pain.

Based on the validated analgesic efficacy of hUC-MSCs, we next evaluated the preclinical safety of intrathecal hUC-MSC administration for BCP treatment. ​​In accordance with the maximum dose principle for preclinical safety studies and considering the anatomical constraints of the rat intrathecal space, we employed the maximum deliverable dose of hUC-MSCs for intrathecal injection. Our preliminary experiments established that, while maintaining hUC-MSC suspension homogeneity, the maximum administrable dose without inducing severe adverse effects or mortality was 4 × 10⁶ hUC-MSCs in a 40 µL total volume. This dosage parameter was further validated by Cao et al. and Yang et al., whose studies demonstrated the feasibility and safety of 40 µL intrathecal hUC-MSC suspensions [61, 62]. Thus, this study designated 4 × 10⁶ hUC-MSCs as the maximum dose for safety evaluation. This dose represents an approximately 20-fold increase over the projected clinical dose (1 × 10⁶ cells/kg, calculated based on a 200 g rat body weight), thereby significantly extending the safety assessment range for clinical dosage translation. In this study, we comprehensively assessed acute toxicity, in vivo biodistribution, tumorigenicity, and tumor-promoting potential of hUC-MSCs.

Acute toxicity test data serve as the foundation for establishing clinical trial dose-escalation protocols. Through high-dose exposure experiments, immediate adverse reactions that may be caused by intrathecal hUC-MSC administration can be identified. This toxicity test evaluated both the maximum dose of 4 × 10⁶ hUC-MSCs and a lower dose of 1 × 10⁶ hUC-MSCs. In addition, this toxicity test focused on acute toxicity up to 14 days after intrathecal injection of hUC-MSCs, without continuing to monitor chronic toxicity over a longer period of time, as extending the observation period for chronic toxicity evaluation was precluded by the natural progression of tumor tissue in BCP rats. Specifically, 3–4 weeks after tumor cell inoculation in the tibia, excessive tumor growth progressively restricted hindlimb mobility, which compromised the validity of pain behavioral assessments. This inherent limitation of the BCP animal model prevented the extension of toxicity observation time. Our results demonstrated that intrathecally administered hUC-MSCs at either dose of 1 × 10⁶ or 4 × 10⁶ hUC-MSCs induced no significant acute toxicity or systemic inflammatory responses throughout the 14-day observation period. Notably, transient elevations in Mon and Neu counts were observed on day 3 after injection of 4 × 10⁶ hUC-MSCs, with Mon levels remaining elevated through day 7. These elevated levels of both Mon and Neu subsequently decreased to baseline on day 14. No such inflammatory cell fluctuations occurred at any time point with the 1 × 10⁶ hUC-MSCs dose. We speculated that this dynamic change in hematologic inflammatory cells would be attributed to a xenogeneic immune response triggered by human MSCs in rodents. This finding does not affect the overall safety conclusion of intrathecal hUC-MSC administration. Importantly, it suggests that future clinical translation should prioritize the lowest effective dose within the analgesic range to ensure safety and avoid potential risks.

Fluorescent labeling combined with in vivo imaging can track the migration and distribution of intrathecally injected hUC-MSCs in rats, facilitating the identification of off-target organ accumulation and assessment of potential risks. Our in vivo imaging results revealed that intrathecally delivered hUC-MSCs remained localized to the perispinal region near the injection site until gradual clearance, and the intrathecal diffusion range expanded with increasing hUC-MSC dose. There was no distribution or accumulation in other organs at both doses of 1 × 10⁶ and 4 × 10⁶ cells. The integrity of the blood-spinal cord barrier (BSCB) is a critical factor in determining the systemic biodistribution of intrathecally injected hUC-MSCs. Unlike pathological conditions such as spinal cord and brain injury that have substantial structural damage, the integrity of BSCB in pain seems to be largely preserved. Although some studies have suggested that peripheral nerve injury and C-fiber stimulation may transiently increase BSCB permeability, Lu et al. found that neither inflammatory pain in rats established by plantar injection of carrageenan and Freund’s complete adjuvant nor neuropathic pain in rats established by spinal nerve ligation caused significant BSCB disruption [63, 64]. Our findings in BCP rats are consistent with these results, indicating that BCP does not induce sufficient BSCB disruption to permit leakage of intrathecally delivered hUC-MSCs into systemic circulation. Importantly, it prevents direct contact between intrathecal hUC-MSCs and tibial tumors. Therefore, this predictable and limited biodistribution range provided compelling evidence supporting the safety of intrathecal hUC-MSC delivery in BCP.

The inherent self-renewal capacity and multilineage differentiation potential of MSCs theoretically confer a risk of abnormal proliferation and tumor formation. The potential for tumorigenicity is the most concerning risk in the clinical translation of hUC-MSC therapy. Our 16-week experimental observations revealed that intrathecal injection of 4 × 10⁶ hUC-MSCs did not exhibit tumorigenic potential. These findings are consistent with​ previous studies by Yang et al. and Wang et al., who confirmed the absence of tumorigenicity with 5 × 10⁶ hUC-MSCs over 13-week and 180-day observation periods, respectively [62, 65]. Notably, BCP possesses distinct tumor-associated pathological characteristics that differentiate it from other pain types. When applying hUC-MSC therapy to BCP, although intrathecally administered hUC-MSCs avoid direct contact with bone tumor tissue, whether they might indirectly influence bone tumors through central-to-peripheral neural signaling remains unclear. Therefore, evaluating the potential tumor-promoting effects of intrathecal hUC-MSCs is particularly important. This study investigated the tumor-promoting potential of intrathecally administered hUC-MSCs using two in vivo models, BCP rats and tumor-bearing nude mice. In BCP rats, tibial bone destruction and tumor proliferation were assessed on day 14 after intrathecal injection of high-dose 4 × 10⁶ hUC-MSCs. The results demonstrated that 4 × 10⁶ hUC-MSCs did not promote tumor proliferation within the tibia of BCP rats. Furthermore, considering the potential immunodeficiency of clinical cancer patients, a tumor-bearing nude mouse model was established using the human breast cancer cell line MDA-MB-231 to evaluate the tumor-promoting effects of intrathecal hUC-MSCs under immunosuppressed conditions. Owing to the anatomical limitations of the intrathecal space in nude mice, we utilized a dose of 1 × 10⁶ hUC-MSCs. Our results showed that intrathecal injection of 1 × 10⁶ hUC-MSCs did not promote the proliferation of primary tumors over a 12-week observation period. These findings aligned with a previous study by Wang et al., who demonstrated that intravenous injections of 1 × 10⁶ hUC-MSCs weekly for five consecutive doses showed no tumor-promoting effects in tumor-bearing nude mice [65]. Evidence from these two animal models collectively demonstrates that intrathecal hUC-MSC administration has no tumor-promoting effects in BCP rats and tumor-bearing nude mice. Therefore, these comprehensive assessments of both tumorigenicity and tumor-promoting risks provide critical safety evidence for clinical translation of intrathecal hUC-MSC therapy in BCP management.

In summary, these safety evaluations demonstrated that intrathecal administration of hUC-MSCs induced no significant acute toxicity, exhibited a predictable biodistribution range, and had no tumorigenicity or tumor-promoting potential. Therefore, intrathecal injection of hUC-MSCs is a safe therapy with reliable preclinical safety for BCP management. It is worth noting that high doses of hUC-MSCs may carry increased safety risks while providing enhanced analgesic efficacy. Therefore, dose selection in clinical treatment should comprehensively weigh the analgesic efficacy and safety of different doses of hUC-MSCs, preferentially using the lowest effective dose within the analgesic range. Furthermore, single-dose administration remains the optimal clinical strategy, considering the benefit-risk ratio for the patient.

While the findings of this study are encouraging, this study has some limitations. First, this study primarily focused on preclinical validation of the analgesic efficacy and safety of intrathecal hUC-MSC administration, rather than investigating the underlying mechanisms of hUC-MSC-mediated pain relief in BCP, which will be explored in future studies. Secondly, given breast cancer’s high incidence and mortality in females, we specifically established a breast cancer bone metastasis model using female animals. While breast carcinoma cells are widely recognized and frequently employed for modeling BCP, additional preclinical studies are needed to evaluate therapeutic efficacy and safety in cancers with high prevalence in males, such as lung and prostate cancers [66]. Both the mechanism of pain occurrence and optimal analgesic strategies have sex differences [67–69]. Therefore, including both female and male animals in preclinical studies holds significant research implications. Additionally, the rat model of breast cancer BCP used in this study employed the classic method of intra-tibial inoculation of tumor cells. Regarding the establishment of BCP animal models, the most primitive method of intracardiac injection of tumor cells has significant limitations. These include uncontrollable tumor metastasis, difficulty in standardizing bone metastasis conditions, and severe systemic spread of cancer cells, which compromises animal health and may affect experimental outcomes [70, 71]. In contrast, the intra-tibial transplantation of Walker 256 breast cancer cells effectively mimics key pathological features of breast cancer bone metastasis in patients, including osteolytic bone destruction, sustained glial cell activation, and neuroinflammation [72, 73]. Compared to intracardiac injection, this method demonstrates better stability, reproducibility, and is easier to standardize. Consequently, the intra-tibial tumor cell transplantation method has largely replaced intracardiac injection in BCP research [74]. However, it should be acknowledged that this bone metastasis model cannot fully replicate the natural process of primary tumor metastasis to bone, and the tumor microenvironment and host immune response in this model may differ from those in spontaneous bone metastasis. Finally, this preclinical study primarily employed rodent models. Before clinical application of this hUC-MSC therapy in BCP patients, additional efficacy and safety evaluations in non-human primates, which share greater physiological and anatomical similarities with humans, may be necessary.

Conclusions

This study demonstrates that intrathecal injection of clinical-grade hUC-MSCs effectively alleviates BCP in rats. Comprehensive safety studies confirm that intrathecal injection of hUC-MSCs has no tumorigenicity and tumor-promoting potential, no significant acute toxicity, and a predictable distribution range, indicating good preclinical safety. These preclinical studies on the analgesic efficacy and high safety of hUC-MSCs provide good preclinical support and guidance for the future clinical translation of intrathecal hUC-MSC therapy in BCP management.

Abbreviations

BCP

Bone cancer pain

MSCs

Mesenchymal stromal cells

hUC-MSCs

Human umbilical cord-derived MSCs

NSAIDs

Non-steroidal anti-inflammatory drugs

HLA-DR

Human leukocyte antigen-DR isotype

GMP

Good Manufacturing Practice

ISCT

International Society for Cell and Gene Therapy

PWT

Paw withdrawal threshold

PWL

Paw withdrawal latency

NSF

Number of spontaneous flinches

c-Fos

Cellular proto-oncogene Fos

IBA1

Ionized calcium-binding adapter molecule 1

GFAP

Glial fibrillary acidic protein

BSCB

Blood-spinal cord barrier

Authors contributions

JZ performed experiments and data analysis, wrote the main manuscript text, and prepared figures. YQL performed experiments and data analysis. PHR and MNJ performed data analysis and prepared figures. PW designed the work and revised the manuscript. QPW designed the work and contributed to funding acquisition. All authors reviewed and approved the final manuscript.

Funding

This work was supported by the United Foundation for Dalian Institute of Chemical Physics, Chinese Academy of Sciences and the First Hospital of Dalian Medical University (DMU-1&DICP UN202402) and the Liaoning Provincial Key Research and Development Program (2024JH2/102500038).

Data availability

All data and materials are available in the manuscript.

Declarations

Ethics approval and consent to participate

Animal Ethics declaration: All animal experiments were approved by the Animal Ethics Committee of Dalian Medical University (Title of the approved project: Study on the Efficacy and Safety of Human Umbilical Cord-Derived Mesenchymal Stromal Cells in Treating Bone Cancer Pain in Rats, and the Therapeutic Effects of Combined Use with Fentanyl; Approval number: AEE24228; Data of approval: 19 November 2024). The work has been reported in line with the ARRIVE guidelines 2.0. The hUC-MSCs used in this study were all produced and provided by the Stem Cell Clinical Research Center of the First Affiliated Hospital of Dalian Medical University. The umbilical cord tissue was obtained from healthy pregnant donors. All donors signed written informed consent. The use of human umbilical cord tissue was approved by the Ethics Committee of the First Affiliated Hospital of Dalian Medical University (Approval number: LCKY2016-59; Data of approval: 16 November 2016). Clinical trial number: not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.