CXCR4-targeted PET imaging in rheumatoid arthritis: a novel approach for monitoring disease activity and therapeutic response

Abstract

Background

Rheumatoid arthritis (RA) is a common chronic, inflammatory autoimmune disease, and early clinical diagnosis is crucial for its treatment. CXCR4 expression was characterized in arthritic mouse models and joints of RA patients, and [¹⁸ F]AIF-NOTA-QHA-04 specificity was assessed in non-malignant cells with elevated CXCR4 expression. This study explored the application of CXCR4-targeted PET probe [¹⁸ F]AIF-NOTA-QHA-04 in monitoring disease activity and therapeutic efficacy in RA. To this aim, the metabolic characteristics of [¹⁸ F]AIF-NOTA-QHA-04 and correlation of [¹⁸ F]AIF-NOTA-QHA-04 uptake with arthritis severity were evaluated by PET imaging in arthritic mice. [¹⁸ F]AIF-NOTA-QHA-04 potential in evaluating therapeutic efficacy was further investigated in arthritic mice following methotrexate (MTX) and etanercept (ETC) treatment.

Results

CXCR4 expression was significantly increased in the inflamed joints of collagen-induced arthritis (CIA) and collagen-antibody induced arthritic (CAIA) mice, and in synovial tissues of RA patients. [¹⁸ F]AIF-NOTA-QHA-04 showed high specificity for CXCR4, with increased probe uptake in arthritic joints that was strongly correlated with arthritis severity scores. PET imaging revealed that increased uptake of [¹⁸ F]AIF-NOTA-QHA-04 in arthritic joints paralleled disease activity, with uptake decreasing upon remission. Furthermore, [¹⁸ F]AIF-NOTA-QHA-04 PET imaging provided earlier and more sensitive assessments of the efficacy of MTX and ETC compared to traditional methods.

Conclusion

The CXCR4-targeted PET probe [¹⁸ F]AIF-NOTA-QHA-04 is a promising tool for RA diagnosis and monitoring, with high specificity and sensitivity. The potential of this probe as a biomarker for disease activity and therapeutic response underscores its value in personalized medication strategies for the management of RA.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13550-025-01203-z.

Introduction

Rheumatoid arthritis (RA) is a chronic, systemic autoimmune disease primarily characterized by erosive, symmetric polyarthritis. The exact pathogenesis is not fully understood, with the fundamental pathological changes involving chronic inflammation of the synovium, pannus formation, and progressive joint cartilage and bone destruction, ultimately leading to joint deformity and functional loss [1]. The foundation of RA treatment lies in suppressing inflammation, alleviating symptoms, maintaining joint function, preventing joint damage, and enhancing the overall quality of life for patients [2, 3]. Accurate assessment of synovial inflammation at the molecular level is crucial for improving the therapeutic outcomes of RA.

Traditional imaging modalities, such as radiography (DR), computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound, provide localized imaging of rheumatic immune diseases and are unable to evaluate the active phase of the disease [4, 5]. Serum laboratory tests are also non-specific and cannot fully reveal the panorama of rheumatic diseases. PET-CT can comprehensively detail the anatomy and functional localization of rheumatic immune diseases, compensating for the shortcomings of other imaging examinations. For example, PET imaging targeting fibroblast activation protein (FAP) can effectively monitor the activation of synovial fibroblasts [6, 7]. However, in addition to fibroblasts, the synovial tissue is also infiltrated by a large number of macrophages, T cells, etc., which promote RA inflammatory reactions and cartilage destruction, with the types of infiltrating cells exhibiting a high degree of heterogeneity [8, 9]. Therefore, the development of more PET probes targeting different markers will be more conducive to the precise monitoring of RA disease activity [10].

CXCR4, known as C-X-C motif chemokine receptor 4, is a G protein-coupled receptor that exerts a critical influence on the migration and positioning of immune cells [11]. In the context of RA pathogenesis, CXCR4 plays a pivotal role in modulating the inflammatory cytokine network, encompassing IL-1, IL-6, IL-10, and TNF-α, thereby directly influencing the initiation and perpetuation of autoimmune responses [12, 13]. The heightened expression of CXCR4 correlates with RA disease activity, underscoring the potential of CXCR4 as a therapeutic target [14]. Targeted interventions against CXCR4 may facilitate a reduction in inflammatory cell infiltration and curtail autoimmune reactions, presenting novel therapeutic avenues for RA [15].

This study assessed the potential value of the Al[¹⁸F]-labeled CXCR4-targeted PET probe in the diagnosis and therapeutic evaluation of RA. Initially, the uptake specificity of the probe for CXCR4 expression was examined both in vitro and in vivo. Subsequently, the correlation between the uptake of CXCR4-targeted probe and the arthritic severity in affected joints was also evaluated. Finally, the potential application of the probe in evaluating the therapeutic effects of RA treatments was assessed.

Materials and methods

Synthesis and characterization of the tracer

The precursor NOTA-QHY-04 was radiolabeled with ¹⁸F following published procedures [16]. Briefly, the ¹⁸F fluoride produced by medical cyclotron (MINITRACE Cyclotron, GE, USA) was trapped on an anion exchange cartridge and eluted with saline. NOTA-QHY-04 in sodium acetate buffer (pH 4, 0.1 M) was mixed with AlCl₃ and acetonitrile, and the saline with ¹⁸F-fluoride was added to the reaction mixture, which was then heated at 110 °C for 10 min. After dilution, the mixture was passed through a preconditioned C18 cartridge, washed with water, and the product was eluted with ethanol. The eluate was reconstituted in saline, filtered through a 0.22 μm filter, and collected in a sterile vial. Quality control analysis via high-performance liquid chromatography showed a radiochemical purity of over 98%.

Cell culture and transfection

293T cells (human embryonic kidney) were obtained from the Fuheng Biotechnology Co., Ltd. (Shanghai, China) and cultured in a humidified incubator at 37 °C with 5% CO₂ using high-glucose Dulbecco’s modified Eagle’s medium (H-DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (PS). To obtain a cell line with high expression of the human C-X-C motif chemokine receptor 4 (CXCR4), the cells were transfected with a human CXCR4 lentiviral expression vector or negative control vector. Following transfection, the cells were subjected to selection pressure using puromycin (10 μg/mL) to isolate and expand the stable cell line expressing human CXCR4 (^CXCR4-293T).

Isolation of peripheral blood mononuclear cells (PBMCs)

PBMCs were isolated from the peripheral blood of healthy human donors using standard procedures [17]. The PBMCs were then cultured in RPMI 1640 medium supplemented with 10% FBS and 1% PS. CD3/CD28 beads and interleukin-2 (IL-2, 100 ng/mL) were used to activate the PBMCs. The activation process was carried out for 9 days, after which the cells were employed in subsequent experimental manipulations. The use of human blood samples was conducted in compliance with ethical guidelines and was approved by the Institutional Review Board (IRB). Informed consent was obtained from all donors prior to sample collection.

Animals

Animal studies were approved by the Institutional Animal Care and Use Committee of Shandong Medicinal Biotechnology Center. Eight-week-old male DBA/1 mice and male 5-week-old BALB/c-nu mice were procured from Huafukang (Beijing, China) and maintained in a specific pathogen-free animal facility. The collagen-induced arthritis (CIA) model was established in DBA/1 mice as follows [18]: on day 0, the mice received an intradermal injection of 0.15 mL Freund’s complete adjuvant (Sigma) emulsified with bovine type II collagen (CII, Chondrex, Redmond, WA, USA) at a dose of 100 μg per mouse, administered in the tail. On day 21, the mice were re-stimulated via an intradermal injection of 0.1 mL CII emulsion. Collagen antibody-induced arthritis (CAIA) was induced in DBA/1 mice using the following protocol: on day 0, the mice received an intraperitoneal (i.p.) injection of 4 mg of a 5-clone anti-collagen monoclonal antibody cocktail (Chondrex). Subsequently, on day 3, the mice were administered 40 μg of lipopolysaccharide i.p., as previously described. On day 16, the mice were euthanized, and the knee joints were harvested for further analysis.

The severity of joint inflammation was assessed every 3–4 days throughout the development of the arthritis model, based on the degree of erythema and swelling. The assessment was performed using a blinded method, with each limb receiving a score ranging from 0 to 4 points, culminating in a maximum possible score of 16 points. Additionally, the thickness of the hind paws was measured using digital calipers to provide an objective measure of joint swelling.

In order to establish a subcutaneous xenograft tissue in mice, the male 5-week-old BALB/c-nu mice were inoculated subcutaneously with 2 × 10^{6 CXCR4−}293T(n = 3) or control 293T cells (n = 3) respectively suspended in 200 μL of extracellular matrix gel. When the tissues reached a volume of 50–100 mm3, the mice were subjected to PET imaging.

Flow cytometry

Following the induction of anesthesia in mice, ocular blood was collected and transferred into tubes containing sodium heparin as an anticoagulant. The anticoagulated blood was subsequently diluted 1:2 with phosphate-buffered saline solution. Subsequently, PBMCs were isolated using a mouse lymphocyte separation medium (Dakewe, Beijing, China). The isolated cells were then resuspended in a flow cytometry staining buffer, and the cell concentration was adjusted to 1 × 10^6 cells per milliliter. To minimize non-specific binding, the cells were incubated with a blocking agent. After washing with the flow cytometry staining buffer to remove unbound reagents, the cells were stained with specific antibodies for flow cytometry analysis. The antibody incubation was conducted for 30 min in the dark to preserve the fluorescence signal integrity. Following incubation, the samples were loaded onto a BD FACS Verse flow cytometer for data acquisition. The acquired data were subsequently analyzed using FlowJo 10.8.1 software to interpret the flow cytometry results.

cell subpopulation	Antibody name	brand	catalog number
macrophage	CD11b-FITC	BioLegend	101,205
	F4/80-Percp	BioLegend	123,125
	CD68-APC	BioLegend	137,007
NK cells	CD3-FITC	BioLegend	100,203
	CD49b-APC	BioLegend	103,515
CD4 + T cells	CD3-FITC	BioLegend	100,203
	CD4-Percp Cy5.5	BioLegend	100,540
CD8 + T cells	CD3-FITC	BioLegend	100,203
	CD8-Percp Cy5.5	BioLegend	100,734
B cells	CD3-FITC	BioLegend	100,203
	CD19-Percp Cy5.5	BioLegend	152,405
	B220-APC	BioLegend	103,211
CXCR4	CXCR4-PE	BioLegend	146,506

Tracer uptake study

293T cells or PBMCs were seeded into 24-well plates at 5 × 10⁵ cells per well and were divided into three groups. For the first group, the cells were incubated with an excess of tracer at 37℃ for 10 min, while the incubation time for the second group was 60 min. For the third group, the cells were pre-incubated with an unlabeled CXCR4 ligand for 30 min before being incubated with the tracer for 60 min. Then, the cells were harvested and counted by a gamma counter(WIZARD 2480, PerkinElmer, USA). The tracer binding per 10⁶ (%AD/10⁶ cells) or 10⁸ cells (%AD/10⁸ cells) was calculated according to the counts.

Small animal positron emission tomography/computed tomography (PET/CT) imaging and biodistribution

Micro-PET imaging was performed on a small-animal PET scanner (IRIS, Inviscan, France) after administration of [¹⁸F]AlF-CXCR4 (3.70–7.40 MBq) via tail vein injection. Dynamic and static PET scanning was performed on arthritic mice after injection. The static PET scans were conducted only at 60 min post-injection. For blocking studies, mice bearing ^CXCR4−293T subcutaneous tissue were injected with the unlabeled precursor (100 nmol) 30 min before tracer injection. All PET images were generated using the 2D/3D ordered subset expectation maximization algorithm (2D/3D OSEM). Tissue radioactivity was measured by regions of interest, and the percentage of injected dose per gram of tissue (% ID/g) was calculated. For ex vivo biodistribution studies, arthritis mice were randomly divided into groups (n = 4). Injections with 3.70 MBq tracer were performed via the tail vein. The animals were euthanized at 30, 60, 90, and 120 min post-injection. Specific organs of interest, as predetermined by the study design, were harvested, weighed, and measured for radioactivity with a gamma counter. The radioactivity uptake was calculated as % ID/g, based on the decay-corrected counts.

Immunofluorescence

Shi fix coverslips were used to facilitate the adherent culture of PBMCs on specially treated materials. After discarding the culture medium, 4% paraformaldehyde was used to fix the cells at room temperature for 15 min. Then, cells were permeabilized for 10 min using permeabilizing solution (0.3% Triton-100) and blocked with 5% BSA. CXCR4 antibody (diluted 1:600) was incubated overnight at 4 ° C. After discarding the antibody diluent, the cells were washed with washing solution and treated with fluorescent secondary antibody (AF488). The cells were incubated at room temperature for 1 h and stained with DAPI staining solution. The images were then obtained under a laser confocal microscope (OLYMPUS FV-3000, Japan).

Immunohistochemistry and hematoxylin-eosin (H&E) staining

Patients with RA were diagnosed based on the revised criteria established by the American College of Rheumatology. Both RA and osteoarthritis (OA) patients who underwent knee joint replacement surgery were recruited from the First Affiliated Hospital of Shandong First Medical University. The study received approval from the Institutional Ethics Committee of the Shandong Medicinal Biotechnology Center and was conducted in compliance with the principles outlined in the Declaration of Helsinki. Comprehensive data pertaining to the patients included in this study are provided in Supplementary Table 1. Synovial tissues obtained from RA and OA patients undergoing joint replacement were first fixed with paraformaldehyde. The hind knee joints of CIA mice were fixed in 4% paraformaldehyde and then decalcified in 10% EDTA for three weeks. Following embedding in paraffin, the tissues were sliced into 4-μm sections. The process of immunohistochemistry involves repairing dehydrated slice antigens, blocking endogenous peroxidase, blocking, and overnight application of antibodies (Anti-CXCR4, 1:600, Abcam, Ab181020) at 4 ℃ using a universal two-step immunohistochemistry kit (Zhongshan Jinqiao, China). The experiment was conducted according to the manufacturer’s instructions. DAB staining was performed using the DAB staining kit (Zhongshan Jinqiao, China), and nucleic acid staining was performed using hematoxylin dye (Biosharp, China). The image was obtained under an inverted microscope.

The H&E staining was conducted using a kit from Solarbio (Beijing, China). The tissue sections were first routinely processed and then fluorescently stained with appropriate primary (anti-CXCR4, Cell Signaling Technolog, Danvers, MA, USA) and secondary antibodies. Articular cartilage was stained with safranin-O and toluidine blue staining.

Western blot analysis

For protein extraction, a lysis buffer consisting of RIPA lysate (Beyotime, China), a protein phosphatase inhibitor (MCE, China), and a protease inhibitor (MCE, China) was prepared in a ratio of 100:1:1 to facilitate protein lysis. Following centrifugation, the supernatant containing the protein lysate was collected, and protein concentration was determined using a BCA Protein Assay Kit (Beyotime, China). The protein supernatant was then combined with SDS-PAGE loading buffer (5X) (Beyotime, China) at a 4:1 ratio and denatured by heat treatment. The denatured proteins were loaded onto SDS-PAGE gels at equal protein concentrations. After electrophoresis, proteins were transferred to a 0.45 μm PVDF membrane (Merck, Germany). The membrane was then blocked with 5% non-fat dry milk (Yili, China) and incubated overnight at 4 °C with a primary antibody against CXCR4 (Abcam, 1:2000 dilution, Cat. No. Ab181020). Subsequently, the membrane was incubated with a horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody (Zhongshan Jinqiao, China, 1:10000 dilution) following incubation with a GAPDH antibody (Proteintech, China, 1:5000 dilution, Product No. 10494-1-AP). Finally, enhanced chemiluminescence substrate (Vazyme, China) was added dropwise for signal detection.

Real-time PCR

Total RNA was isolated from joint tissues using Trizol reagent (Vazyme, China) according to the manufacturer’s protocol. The extracted RNA was converted into cDNA with a reverse transcription kit (Toyobo, Japan). qPCR was conducted using the UltraSYBR Mixture Reaction Kit (CWBIO, China) with SYBR Green I dye in a LightCycler 480 II system (Roche, Switzerland). The relative expression levels (using GAPDH mRNA as the internal control) were calculated using the 2^−ΔΔCt method [19]. The primer sequences used for amplification were as follows: CXCR4, forward: GGGCAATGGATTGGTCATCCT, reverse: TGCAGCCTGTACTTGTCCG; GAPDH, forward: GGAGCGAGATCCCTCCAAAAT, reverse: GGCTGTTGTCATACTTCTCATGG. Using TOYOBO reverse transcription kit, the specific steps and procedures are as follows: preparation of reaction solution includes 5 × RT buffer 2 μ L, RT enzyme mixture 0.5 μ L, primer mixture 0.5 μ L and RNA 0.5pg-1 μg, And dilute dd H₂O to 10 μ L. The reaction program for this operation is set to: react for 15 min at 37 ℃, react for 5 min at 98 ℃, and end when the temperature reaches 4 ℃.

ELISA assays

Quantitative measurement of cytokines in serum was performed using a dual-antibody sandwich enzyme-linked immunosorbent assay (ELISA) technique with an ELISA kit (Lianke Biotech, China), following the manufacturer’s instructions.

Statistical analysis

All data were statistically analyzed using GraphPad Prism V.9.5.0 software. Data are expressed as mean ± standard deviation (SD). The t-test was employed to assess differences between two groups, one-way ANOVA was used for comparisons among three or more groups, and two-way ANOVA was applied to evaluate differences under various treatment conditions. P values of less than 0.05 were considered statistically significant among the experimental groups.

Results

CXCR4 is a potential biomarker for RA

We first characterized the expression profile of CXCR4 in CIA and CAIA mouse models. The expression levels of CXCR4 in the affected joints from the CIA and CAIA mice were significantly higher than in the joints of healthy mice (Fig. 1a). Immunohistochemical analysis further confirmed the high expression of CXCR4 in the arthritic mice (Fig. 1b). Additionally, bioinformatics analysis of the RA patient database (GSE89408) showed that the expression of CXCR4 in the synovial tissue of RA patients was significantly higher than that in patients with osteoarthritis (OA) or healthy controls (Fig. 1c). The above variation in CXCR4 expression between RA and OA was further confirmed in the joint replacement tissues of RA and OA patients (Fig. 1d and e). Furthermore, flow cytometry analysis revealed that CXCR4 was highly expressed in CD4⁺ T cells, B cells, partially expressed in NK cells, and macrophages in arthritic joints, with marked difference compared to healthy joints (Fig. 1f). These results suggest that CXCR4 may serve as a potential biomarker for RA.

Fig. 1.

Expression analysis of CXCR4 in RA. (a) RT-qPCR assay of CXCR4 mRNA levels in articular RNA extracted from healthy control mice (HC) and CIA and CAIA mice (n = 16). (b) Detection of expression of CXCR4 in joints of CIA and CAIA mice by immunohistochemistry. (c) Analysis of CXCR4 expression in RA (n = 152), OA (n = 22), and healthy control (HC) joints (n = 28) using human transcriptome sequencing data from the GEO (GSE89408) database. (d, e) Analysis of CXCR4 expression by (d) RT-qPCR and (e) immunohistochemistry in human synovial tissues isolated from RA and OA after joint replacement surgery (n = 11). (f) CXCR4 expression in immune cells. Synovial tissues of the joints of healthy and CAIA arthritic mice were digested into single-cell suspensions by cathepsin, and CXCR4 expression in each immune cell subpopulation was analyzed by flow cytometry. All RT-qPCR results were normalized with GAPDH. Data are expressed as mean ± SEM. **P<0.01; ***P<0.001

[¹⁸ F]AIF-NOTA-QHA-04 exhibits high specificity for the CXCR4 target

The CXCR4-targeted PET imaging probe [¹⁸ F]AIF-NOTA-QHA-04 was previously designed to offer a precise diagnostic tool for visualizing CXCR4 (Supplementary Fig. 1). To substantiate the probe’s efficacy in non-malignant conditions, a 293T cell line with stable and elevated CXCR4 expression (^CXCR4−293T) was constructed and confirmed by RT-PCR and flow cytometric analysis (Fig. 2a and b). To ascertain the specificity of [¹⁸ F]AIF-NOTA-QHA-04 in vivo, the tracer was incubated with 293T cells with and without ectopic expression of CXCR4. The tracer uptake in ^CXCR4−293T cells was significantly higher than in the control cells at both 10 min (0.530 ± 0.073 vs. 0.124 ± 0.012%AD/10⁶ cells) and 60 min (0.635 ± 0.041 vs. 0.166 ± 0.010%AD/10⁶ cells). Importantly, this elevated uptake at 60 min was effectively blocked by pre-incubation of the cells with unlabeled precursor (0.147 ± 0.006%AD/10⁶ cells) (Fig. 2c). Furthermore, a significant upregulation of CXCR4 expression was observed in activated PBMCs (Fig. 2d and e), and these activated PBMCs exhibited a correspondingly significantly higher uptake of [¹⁸ F]AIF-NOTA-QHA-04 when compared to the control cells (Fig. 2f).

Fig. 2.

In vitro characterization of cellular uptake of CXCR4-targeted tracer. (a, b) CXCR4 expression levels in ^CXCR4−293T cells and control 293T cells were quantified by RT-qPCR (a) and flow cytometry (b). (c) Evaluation of[¹⁸F]AIF-NOTA-QHA-04 uptake by ^CXCR4−293T in 24-well plates at 10 min and 60 min time points. Unlabeled NOTA-QHY-04 was used to assess competitive binding at 60 min (“blocked” sample). Radioactivity was measured by gamma counter. (d, e) CXCR4 expression in control and activated PBMCs. PBMCs were activated by CD3/CD28 beads and IL-2, and CXCR4 expression was evaluated by RT-qPCR (d) and immunofluorescence (e). (f)[¹⁸F]AIF-NOTA-QHA-04 uptake by PBMCs. Uptake by PBMCs and activated PBMCs was compared with and without blocking with unlabeled NOTA-QHY-04. Data are expressed as the mean ± SEM (n = 5). ***P < 0.001

To evaluate the targeting specificity of [¹⁸ F]AIF-NOTA-QHA-04 in vivo, ^CXCR4−293T cells and 293T control cells were inoculated subcutaneously into nude mice. Subcutaneous xenograft tissues overexpressing CXCR4 showed higher uptake of the tracer when compared to the control tumors (2.225 ± 0.302 vs. 0.181 ± 0.058%ID/g), whereas such uptake was effectively blocked by the addition of unlabeled probe precursor (0.199 ± 0.054%ID/g) (Fig. 3a and b). The high expression of CXCR4 in subcutaneous xenograft tissues, corresponding with the high uptake of [¹⁸ F]AIF-NOTA-QHA-04, was further confirmed by immunoblotting (Fig. 3c). The data above indicated the uptake specificity of [¹⁸ F]AIF-NOTA-QHA-04 in RA models.

Fig. 3.

Validation ofin vivo targeting specificity of CXCR4 tracer. (a) Subcutaneous xenograft tissues were constructed by ^CXCR4−293T and control 293T cells. Then, small animal PET imaging was performed following adminstration of [¹⁸F]AIF-NOTA-QHA-04. In the blocked group, 100 μg NOTA-QHY-04 was injected through the tail vein 30 min in advance.[¹⁸F]AIF-NOTA-QHA-04 uptake in these subcutaneous xenograft tissues were quantification (right panel). (b) Analysis of CXCR4 expression levels in subcutaneous growth tissue by western blot. Band intensities were quantified using ImageJ software (right panel). ***P < 0.001

Analysis of dynamic PET scanning in arthritic mice

To characterize the uptake of [¹⁸ F]AIF-NOTA-QHA-04 using the arthritic mouse models, small animal dynamic PET scans were performed. The joints of front and hind limbs in CIA mice showed redness and swelling to varying degrees (Fig. 4a). Next, 60-min dynamic PET scans were conducted after the tail vein injection of [¹⁸ F]AIF-NOTA-QHA-04 in CIA mice at 5 min intervals (Fig. 4b). High tracer uptake was observed in the kidneys, bladder, and arthritic joints. Moreover, the uptake signals were positively correlated with the severity of arthritis (Fig. 4c).

Fig. 4.

Biodistribution of[¹⁸F]AIF-NOTA-QHA-04. (a) Representative images of joint swelling in CIA mice. (b) Dynamic PET imaging of CIA mice following injection of[¹⁸F]AIF-NOTA-QHA-04 (3.74 MBq). (c) Analysis of joint uptake value of[¹⁸F]AIF-NOTA-QHA-04, expressed as relative values. (d) Biodistribution of[¹⁸F]AIF-NOTA-QHA-04 expressed in standardized uptake values per organ. Data were converted to the percentage of injected dose per gram (% ID/g) and are expressed as the mean ± SEM

In order to further clarify the distribution of [¹⁸ F]AIF-NOTA-QHA-04 in the arthritic mice, various tissues, including joints, were harvested to perform a quantitative biodistribution analysis ex vivo at 30 min, 60 min, 90 min, and 120 min post [¹⁸ F]AIF-NOTA-QHA-04 injection. The highest radioactivity was mainly observed in the bladder and kidney, with low radioactivity in the blood, bone, lung, liver, small intestines, and muscle, further suggesting that [¹⁸ F]AIF-NOTA-QHA-04 was predominantly excreted through urinary routes (Fig. 4d). In addition, compared with healthy joints, the joints of arthritis mice showed statistically significant uptake at 30, 60, 90, and 120 min; Regarding the high water-solubility and renal clearance issues of [¹⁸ F]AIF-NOTA-QHA-04, we did observe high levels of radioactivity in the kidneys and bladder, and this state did not significantly decrease after 60 min. We believe that this may be related to the high water-solubility and renal reabsorption mechanism of [¹⁸ F]AIF-NOTA-QHA-04. Although the half-life of 18F is 110 min, the reduction in radioactivity is not significant due to its retention in the kidneys. We discussed in detail the possible mechanisms of this phenomenon in the discussion section and proposed future research directions to develop more effective kidney clearance strategies to reduce the retention time of radiation in the kidneys.

Uptake of [¹⁸F]AIF-NOTA-QHA-04 in arthritic joints is associated with arthritis severity

To evaluate the potential application of [¹⁸ F]AIF-NOTA-QHA-04 in arthritis imaging, routine blood tests, blood biochemistry, and body weight were examined for safety evaluation. Following the administration of [¹⁸ F]AIF-NOTA-QHA-04, all data from routine blood tests and biochemical analyses were within the normal range, irrespective of [¹⁸ F]AIF-NOTA-QHA-04 inoculation (Supplementary Tables 2 and 3). Additionally, no significant differences were observed in paw thickness, arthritis score, or body weight between the groups with and without [¹⁸ F]AIF-NOTA-QHA-04 as the arthritic phenotype progressed, indicating the safety of [¹⁸ F]AIF-NOTA-QHA-04 (Supplementary Fig. 2).

Then, the CIA and CAIA arthritis mouse models were used to determine whether [¹⁸ F]AIF-NOTA-QHA-04 could be used to monitor disease progression. Notably, in both models, a significant uptake of [¹⁸ F]AIF-NOTA-QHA-04 was observed in the arthritic joints, with only negligible uptake in the joints of healthy control mice (upper panel, Fig. 5a). Immunohistochemical staining confirmed the higher uptake of [¹⁸ F]AIF-NOTA-QHA-04 in affected joints with relatively higher CXCR4 expression (lower panel, Fig. 5a). Furthermore, a highly significant correlation between [¹⁸ F]AIF-NOTA-QHA-04 uptake and arthritis scores was found in both arthritis models (Fig. 5b). Importantly, the remission of arthritis over time was accompanied by the reduction of tracer uptake in both the CIA (Fig. 5c) and CAIA model (Fig. 5d). These results suggest that PET probe uptake can be used as a sensitive indicator of arthritis activity.

Fig. 5.

Overall evaluation and quantification of [¹⁸F]AIF-NOTA-QHA-04 uptake in arthritic mice. (a) PET imaging and immunohistochemistry analysis of CIA and CAIA arthritic mice compared to healthy controls (HC). Red boxes indicate immunohistochemistry site. (b) Correlation between [¹⁸F]AIF-NOTA-QHA-04 uptake and arthritis score. (c, d) PET images of [¹⁸F]AIF-NOTA-QHA-04 uptake at different time points after CIA (c) and CAIA (d) induction in single representative individual mouse

[¹⁸ F]AIF-NOTA-QHA-04 can aid in evaluating therapeutic efficacy in arthritis

We also explored the value of [¹⁸ F]AIF-NOTA-QHA-04 for evaluating the therapeutic effects of methotrexate (MTX) and etanercept (ETC), two commonly used RA drugs. As shown in Supplementary Fig. 3a, both MTX (5 mg/kg) and ETC (5 mg/kg) could effectively suppress the secretion of inflammatory factors in PBMCs from RA patients, including interleukin-1β (IL-1β), tumor necrosis factor-α (TNF-α), and interleukin-6 (IL-6). Additionally, a significant reduction of [¹⁸F]AIF-NOTA-QHA-04 uptake was observed following treatment with MTX and ETC (Supplementary Fig. 3b). This finding indicates that CXCR4 may serve as an indicator for monitoring drug efficacy in RA.

Subsequent animal experiments showed that MTX and ETC could effectively delay disease progression in CIA models as evidenced by the reduction of arthritis scores and joint swelling and destruction (Fig. 6a and b), as well as the production of inflammatory factors (Fig. 6e, Supplementary Fig. 4). Importantly, when comparing with the control mice (vehicle), the differences in arthritis scores and joint swelling became obvious and significant from the 40th day following treatment with MTX or ETC (Fig. 6d), whereas the uptake of [¹⁸ F]AIF-NOTA-QHA-04 differed markedly from the 28th day onwards (Fig. 6c). These data suggest that PET imaging of CXCR4 is more sensitive and reliable than traditional methods in evaluating the effects of therapy.

Fig. 6.

Application of [¹⁸F]AIF-NOTA-QHA-04 PET imaging in monitoring the therapeutic response in CIA mice. Mice treated with vehicle, MTX (5 mg/kg), or ETC (5 mg/kg) underwent PET imaging and histological evaluation. (a, b) On days 28–56 after post-immunization, arthritis scores and foot swelling of CIA mice were assessed and recorded every four days. (c, d) Representative [¹⁸F]AIF-NOTA-QHA-04 images of CIA mice receiving MTX or ETC treatment (c) and analysis of probe uptake (d). (e) H&E staining, toluidine blue staining, and safarin-O/green staining of mouse joints were performed at 56 days post-CIA induction. Data are expressed as the mean ± SEM (n = 5). In a, b and d, MTX vs. Vehicle, ^*P < 0.05, ^***P < 0.001. ETC vs. Vehicle, ^###P < 0.001

Discussion

The synovial tissues in RA exhibit a pronounced heterogeneity, characterized by the infiltration of a multitude of immune cells, including macrophages, B cells, T cells, and synovial fibroblasts. This complexity necessitates novel diagnostic tools, such as PET molecular imaging, to predict disease progression and remission. The probe [¹⁸ F]AIF-NOTA-QHA-04, which targets CXCR4, has been previously evaluated in oncological contexts. For example, ⁶⁸Ga-PentixaFor PET/CT has been used to detect solid tumors like ovarian cancer, small cell lung cancer, and adrenocortical carcinoma, showing high image contrast and specific tumor uptake [20, 21]. Additionally, CXCR4 PET imaging is proving useful in diagnosing benign conditions, such as subtyping primary aldosteronism, thereby enhancing diagnosis and treatment [22, 23]. However, the role of CXCR4 in autoimmune diseases has attracted widespread attention, especially in conditions such as RA and systemic lupus. In this study, we evaluated the application of this probe for monitoring RA.

The upregulation of CXCR4 within the inflamed joints of CIA and CAIA mouse models mirrored the differential expression of this receptor observed in human synovial tissues. This concordance, coupled with the observed upregulation of CXCR4 in immune cells such as CD4⁺ T cells and macrophages, underscores the pivotal role of CXCR4 in the inflammatory cascade of RA [24]. After treatment with the commonly used drugs MTX and ETC in clinical practice for RA [25, 26], PET confirmed the role of this imaging agent in evaluating the progression of RA. These observations provide a compelling rationale for the development of CXCR4-targeted imaging modalities. The innovative application of a CXCR4-targeted PET probe in RA, underscores the potential of this probe as an alternative monitoring tool for assessment of disease activity and therapeutic response in RA.

However, this study is not devoid of limitations. Although we did not detect any detrimental effects of the probe on the mice in our study, the long-term stability and safety profile of the PET probe necessitate further investigation. Moreover, the efficacy of [¹⁸ F]AIF-NOTA-QHA-04 in human RA patients requires clinical validation. Additionally, the potential of CXCR4 antagonists to delay the progression of arthritis in mice [27, 28], as suggested by some studies, raises intriguing question about whether this imaging technology could identify RA patients who might benefit from targeted CXCR4 therapy.

Conclusion

This study pioneered the exploration of the CXCR4-targeted PET probe [¹⁸ F]AIF-NOTA-QHA-04 in the context of a non-neoplastic disease, specifically RA. The findings demonstrate the utility of CXCR4-targeted PET probes in monitoring disease activity and evaluating treatment efficacy in RA. Future research endeavors should focus on the clinical translation of this probe for autoimmune diseases and the development of personalized treatment strategies, which are of paramount importance for the advancement of personalized medicine and early intervention strategies.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

LG wrote the manuscript. YH, SW, JL, BD and SC acquired the data. JP and YG analyzed and interpreted the data. KC, LW, and LG conceptualized and designed the study. All authors read and approved the final manuscript.

Funding

This study was funded by the National Natural Science Foundation of China (Grant Nos. 82072850, 81772760, 82101903, 81901666, 82171801, 82271842), the Natural Science Foundation of Shandong Province (Grant No. ZR2023MH013), the Key Research and Development Project of Shandong Province (No. 2021ZDSYS27), the Innovation Project of Shandong Academy of Medical Sciences (2021), the Youth Innovation Technology Plan of Shandong University (Grant No. 2019KJK003), and Academic Promotion Programme of Shandong First Medical University (Grant No. 2019LJ001).

Data availability

All data relevant to the study have been included in the article or uploaded as supplementary information and are available from the corresponding author upon reasonable request.

Declarations

Ethical approval

All animal studies were approved and performed according to the guidelines of the animal care and use committee of Shandong Medicinal Biotechnology Center.

Consent to participate

Written informed consent for participation was provided by the patient involved in this study.

Consent for publication

All patients obtained informed consent in the study and the data provided were approved for publication.

Conflict of interest

The authors declare no competing interests.