Biomarker combination for pathological osteogenesis diagnosis of ankylosing spondylitis and application

Abstract

The invention discloses a marker combination for pathological osteogenesis diagnosis of ankylosing spondylitis and application, and belongs to the technical field of diagnosis of ankylosing spondylitis. The biomarker combination used for pathological osteogenesis diagnosis of ankylosing spondylitis comprises CDR1, PLPPR2, CMBL, SRXN1 and PPP1R3C, and is characterized in that the biomarker combination is prepared from the CDR1, the PLPPR2, the CMBL, the SRXN1 and the PPP1R3C; the compound can be used for preparing products for evaluating the risk of spinal structure damage progress of patients with ankylosing spondylitis and screening pathological osteogenesis drugs for inhibiting and / or treating ankylosing spondylitis. The biomarker combination for diagnosing the pathological osteogenesis of the ankylosing spondylitis has high sensitivity and specificity when being used for diagnosing the pathological osteogenesis of the ankylosing spondylitis, and the diagnosis accuracy is greatly improved; the method can also be used for spinal structure injury progress risk assessment and screening of pathological osteogenesis drugs for inhibition and / or treatment of ankylosing spondylitis, and has a wide application prospect.

Description

Technical Field

[0001] This invention belongs to the field of ankylosing spondylitis diagnostic technology, specifically relating to a combination of biomarkers for the diagnosis of pathological osteogenic diagnoses in ankylosing spondylitis and their application. Background Technology

[0002] Ankylosing spondylitis (AS) is an immune-mediated disease characterized by chronic inflammation of the axial skeleton and progressive pathological new bone formation (NBF). As the disease progresses, the new bone formation leads to the formation of ligament osteophytes, eventually causing spinal ankylosis and severe functional impairment.

[0003] Current clinical treatments, such as nonsteroidal anti-inflammatory drugs (NSAIDs), tumor necrosis factor (TNF) inhibitors, and interleukin-17A (IL-17A) antagonists, can effectively control systemic inflammation and improve clinical symptoms. However, multiple long-term follow-up studies have shown that these anti-inflammatory treatments have limited and unpredictable effects on halting the radiographic progression of ossification in AS patients. This "inflammation-ossification separation" suggests that pathological ossification in AS may be driven independently by mechanisms other than inflammation.

[0004] Mounting evidence suggests that aberrant osteogenic differentiation of bone marrow mesenchymal stem cells (BMSCs) is a key cellular driver of ectopic bone formation in ankylosing spondylitis (AS). In the pathological environment of AS, BMSCs exhibit enhanced osteogenic potential and reduced plasticity. Recent studies also suggest that cellular senescence and oxidative stress may be involved in this process. However, BMSCs are highly heterogeneous, and it remains unclear which specific subset of BMSCs drives pathological ossification, nor are there molecular markers that can specifically identify this pathogenic subset.

[0005] Current clinical monitoring of ankylosing spondylitis (AS) primarily relies on inflammatory markers (such as CRP and ESR) and imaging scores (such as mSASSS). However, imaging changes typically lag behind changes at the cellular and molecular levels, and inflammatory markers cannot accurately reflect local osteogenic activity. Therefore, there is an urgent need to discover biomarkers that can specifically identify BMSCs subsets that drive ossification, in order to develop new diagnostic tools and treatment strategies. Summary of the Invention

[0006] To overcome the shortcomings of the prior art, the present invention provides a combination of biomarkers for the diagnosis of pathological osteogenic ankylosing spondylitis and their application.

[0007] This invention, using single-cell transcriptome sequencing technology, identifies for the first time a specific senescence-related cluster (SRC) in bone marrow mesenchymal stem cells of AS patients. This subpopulation is closely associated with severe spinal structural damage and exhibits a unique senescence-associated secretory phenotype (SASP) characterized by high glycolysis, oxidative phosphorylation, and secretory activity. A combination of biomarkers associated with this subpopulation was then screened. Detecting the expression levels of this biomarker combination can effectively assess the pathological osteogenic risk and structural progression trends in AS patients, providing new molecular targets and tools for early warning, precise subtyping, and screening of drugs that inhibit osteolysis in AS.

[0008] The biomarker combination of the present invention can serve as a specific monitoring indicator for pathological osteogenic activity in ankylosing spondylitis and can be used for research on the pathogenesis of pathological osteogenic activity in ankylosing spondylitis and for clinical applications.

[0009] The technical solution adopted by this invention to solve its technical problem is: This invention provides a combination of biomarkers for the diagnosis of pathological osteogenic ankylosing spondylitis, the combination of biomarkers (genes) including CDR1, PLPPR2, CMBL, SRXN1 and PPP1R3C.

[0010] This invention provides a product for the diagnosis of pathological osteogenic ankylosing spondylitis, comprising reagents for detecting the expression levels of the above-mentioned combination of biomarkers (CDR1, PLPPR2, CMBL, SRXN1 and PPP1R3C).

[0011] Preferably, the reagent is a specific detection reagent.

[0012] Preferably, the product is a kit and also includes an instruction manual describing how to use it.

[0013] Preferably, the expression level of the biomarker combination is obtained by detecting the abundance of its transcriptional product (mRNA) or translational product (protein).

[0014] Preferably, reagents for detecting the expression levels of combinations of biomarkers include: (1) Primer pairs, probes, or microarrays of transcripts containing combinations of (specifically identified) biomarkers; Or (2) (specific recognition) the translation product of a combination of biomarkers, antibodies, antigen-binding fragments, or aptamers.

[0015] Preferably, the expression level of the biomarker combination is the abundance of the transcript of the biomarker combination; The product for pathological osteogenic diagnosis of ankylosing spondylitis further includes a data processing module, which processes the abundance data of the transcripts of the biomarker combination, including: Abundance data of transcripts of biomarker combinations were calculated using the formula for risk score P. When the risk score P ≥ 0.57, the patient was identified as a patient with pathological osteoblastic ankylosing spondylitis, and when the risk score P < 0.57, the patient was identified as a patient with pathological osteoblastic non-ankylosing spondylitis.

[0016] Among them, CDR1 △Ct PLPPR2 △Ct CMBL △Ct SRXN1 △Ct and PPP1R3C △Ct The abundance of transcripts for CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C are respectively.

[0017] This invention provides a system for diagnosing pathological osteogenic changes in ankylosing spondylitis, comprising: The data detection module detects the expression levels of a combination of biomarkers in the bone marrow mesenchymal stem cell samples of the subjects; the combination of biomarkers includes CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C. The data processing module processes the expression level data of biomarker combinations and draws conclusions.

[0018] Preferably, the expression level of the biomarker combination is the abundance of the transcript of the biomarker combination; The data processing includes: Abundance data of transcripts of biomarker combinations were calculated using the formula for risk score P. When the risk score P ≥ 0.57, the patient was identified as a patient with pathological osteoblastic ankylosing spondylitis, and when the risk score P < 0.57, the patient was identified as a patient with pathological osteoblastic non-ankylosing spondylitis.

[0019] Among them, CDR1 △Ct PLPPR2 △Ct CMBL △Ct SRXN1 △Ct and PPP1R3C △Ct The abundance of transcripts for CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C are respectively.

[0020] The present invention provides the application of the above-mentioned combination of biomarkers for the diagnosis of pathological osteogenic disease in ankylosing spondylitis, the above-mentioned product for the diagnosis of pathological osteogenic disease in ankylosing spondylitis, or the above-mentioned system for the diagnosis of pathological osteogenic disease in ankylosing spondylitis in the preparation of a product for assessing the risk of progression of spinal structural damage in patients with ankylosing spondylitis.

[0021] Preferably, the product also includes a specification documenting the correlation between test results and the risk of progression of spinal structural damage in ankylosing spondylitis.

[0022] Preferably, the assessment of the risk of progression of spinal structural damage in patients with ankylosing spondylitis includes: (a) Detect the expression levels of CDR1, PLPPR2, CMBL, SRXN1 and PPP1R3C in bone marrow mesenchymal stem cell (BMSC) samples derived from subjects; (b) Compare the expression levels obtained in step (a) with reference values; If the expression level of at least one of the genes CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C is higher than the reference value, the subject is considered to have a risk of progression of spinal structural injury.

[0023] More preferably, the subject is a patient with ankylosing spondylitis.

[0024] Further preferred, the abundance data of the transcripts of CDR1, PLPPR2, CMBL, SRXN1 and PPP1R3C are calculated using the formula for calculating the risk score P. When the risk score P ≥ 0.57, the subject is determined to have the risk of progression of spinal structural injury, and when the risk score P < 0.57, the subject is determined not to have the risk of progression of spinal structural injury.

[0025] Among them, CDR1 △Ct PLPPR2 △Ct CMBL △Ct SRXN1 △Ct and PPP1R3C △Ct The abundance of transcripts for CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C are respectively.

[0026] The present invention provides the use of the above-mentioned combination of biomarkers for the diagnosis of pathological osteogenic ankylosing spondylitis, the above-mentioned product for the diagnosis of pathological osteogenic ankylosing spondylitis, or the above-mentioned system for the diagnosis of pathological osteogenic ankylosing spondylitis in screening drugs that inhibit and / or treat pathological osteogenic ankylosing spondylitis.

[0027] Preferably, the method for screening drugs that inhibit and / or treat pathological osteogenic processes in ankylosing spondylitis includes: (1) Provide a cell model containing SRC subsets, wherein the SRC subsets highly express the combination of biomarkers; (2) Contact the compound to be screened with the cell model; (3) Detect changes in the expression levels of the biomarker combination in the cell model; If the compound can downregulate the expression of at least one of the genes CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C, it indicates that the compound has the potential activity to inhibit pathological ossification, and the compound is determined to be a potential drug for inhibiting and / or treating pathological osteogenic lesions in ankylosing spondylitis.

[0028] The beneficial effects of this invention are: The biomarker combination of the present invention for the diagnosis of pathological osteoblastic ankylosing spondylitis has high sensitivity and specificity, greatly improving diagnostic accuracy; it can also be used for risk assessment of spinal structural injury progression and screening for drugs that inhibit and / or treat pathological osteoblastic ankylosing spondylitis, and has broad application prospects. Attached Figure Description

[0029] Figure 1 This is a single-cell atlas of human bone marrow mesenchymal stem cells (BMSCs) for this invention. (A) Study design diagram: Comparison of patients in the NSD and SSD groups. (B) UMAP dimensionality-reduced clustering diagram of 47,628 BMSCs, identifying 8 subpopulations. The left image shows all cells, while the right image mainly shows the significant differences in cell distribution between the SSD and NSD groups. (C) Proportional distribution diagram: Quantitatively showing the abundance changes of each subpopulation in the NSD and SSD groups.

[0030] Figure 2 This invention provides trajectory inference and differentiation potential analysis of BMSC subsets. (A) RNA velocity analysis: shows the differentiation streamline flowing from progenitor cells (C5) to terminal states (C1 / C2), revealing the dynamic process of cell development. (BC) CytoTRACE2 analysis: assesses cell stemness and differentiation potential; the gradient from red (high potential) to blue (differentiated) verifies the inferred trajectory direction.

[0031] Figure 3This invention presents the transcriptional and functional characterization of eight BMSC subpopulations. (A) Bubble chart: Showing the top three differentially expressed genes (DEGs) defining each subpopulation. (B) Functional annotation diagram: Summarizing the logical relationships from proliferating progenitor cells to lineage-defined or senescent states. (C) Metabolic pathway bubble chart: Revealing the specific metabolic characteristics of different subpopulations (such as glycolysis, oxidative phosphorylation, etc.). (D) Classical biomarker bubble chart: Verifying the identity of the BMSC subpopulations.

[0032] Figure 4 This invention relates to the functional reclassification of BMSC subpopulations and the screening of core biomarkers. (A) Functional reclassification UMAP diagram: The eight subpopulations are classified into five major functional subpopulations, and Cluster 4 is identified as the aging-related subpopulation (SRC). (B) Stacked bar chart: Highlights the significant expansion of the SRC (aging-related), NSC (microenvironment support), and IMC (immune regulation) populations in the SSD group. (C) Venn diagram: Shows the screening logic of the biomarker combination of this invention—that is, the intersection of upregulated DEGs and cell type-specific marker genes.

[0033] Figure 5 The ROC curves are for the diagnostic models of marker combination and single marker in Embodiment 4 of the present invention. Detailed Implementation

[0034] The present invention will be further described below with reference to embodiments.

[0035] The following will clearly and completely describe the concept, specific solutions, and technical effects of the present invention with reference to embodiments, so as to fully understand the purpose, features, and effects of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them. Other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are all within the scope of protection of the present invention. The various technical features in the present invention can be combined interactively without contradicting each other.

[0036] This invention utilizes single-cell transcriptome sequencing (scRNA-seq) technology to conduct in-depth analysis of bone marrow mesenchymal stem cells (BMSCs) from AS patients with different degrees of structural damage (Severe SSD group vs. Non-Structural SSD group). The study revealed significant heterogeneity in BMSCs, with a specific subset, the "Senescence-Related Cluster (SRC)," significantly enriched in patients with severe ossification.

[0037] This invention first identifies and validates a combination of biomarkers for identifying this SRC subpopulation, which includes the following genes: CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C.

[0038] CDR1 (Cerebellar Degeneration-Related Protein 1): is associated with cellular stress response.

[0039] PLPPR2 (Phospholipid Phosphatase Related 2): Involved in lipid signaling and inflammatory responses.

[0040] CMBL (Carboxymethylenebutenolidase Homolog): As a p53 target gene, it participates in glycolysis inhibition and metabolic regulation.

[0041] SRXN1 (Sulfiredoxin 1): A key antioxidant stress gene that maintains mitochondrial redox balance.

[0042] PPP1R3C (Protein Phosphatase 1 Regulatory Subunit 3C): Regulates glycogen metabolism and storage.

[0043] Studies have shown that the SRC subset not only highly expresses the above-mentioned genes, but also exhibits a unique "dual phenotype": it has both cellular senescence characteristics (such as high expression of CDKN1A / p21) and hypermetabolic characteristics (high glycolysis and high oxidative phosphorylation), and secretes large amounts of matrix remodeling factors (such as MMP2) and pro-inflammatory factors (such as IL-6), thereby driving pathological ossification of the microenvironment.

[0044] Based on the above findings, the present invention provides: The combination of markers includes CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C.

[0045] Detection kit: Contains primers, probes or antibodies targeting the gene expression products mentioned above.

[0046] Applications: Used to develop diagnostic products to assess the risk of spinal structural progression in AS patients, or to screen for drugs that can reverse the SRC phenotype (such as JAK inhibitors).

[0047] The advantages of this invention are as follows: High specificity: The biomarker combination of the present invention directly targets the core cell subpopulation (SRC) that drives ossification. Its expression level is positively correlated with the patient's mSASSS score and SPARCC-SSS score, and can reflect the risk of ossification more accurately than conventional inflammatory markers.

[0048] Mechanism clearly defined: This combination reveals the "aging-metabolism-inflammation" axis behind pathological ossification in AS, providing a new assessment dimension for clinical practice.

[0049] Drug development value: Drugs screened based on this biomarker combination (such as JAK inhibitors Baricitinib and Filgotinib) have been predicted to effectively reverse the pathogenic phenotype of SRC, demonstrating the practicality of this biomarker in drug development.

[0050] Example 1: Construction of a bone marrow mesenchymal stem cell atlas and clinical sample characteristics in patients with ankylosing spondylitis (1) Sample collection and grouping This invention strictly followed the 1984 revised New York criteria to include AS patients. Based on imaging assessments (mSASSS score and SPARCC-SSS score), the 12 subjects were precisely divided into a severe structural injury group (SSD, n=9) and a non-structural injury group (NSD, n=3).

[0051] (2) Statistical analysis of clinical characteristics Detailed demographic and clinical baseline information for the two groups of patients is shown in Table 1.

[0052] Structural damage differences: The mSASSS score of the SSD group (median 18.00) was significantly higher than that of the NSD group (median 0.00, p=0.024), and the SPARCC-SSS score also showed a significant difference (p=0.044), which verified the effectiveness of the grouping.

[0053] Inflammation and Ossification Separation: Notably, despite severe structural damage in the SSD group, there was no statistically significant difference in the inflammatory marker CRP between the two groups (p=0.644). This clinical data strongly supports the phenomenon of "inflammation-ossification separation" in AS, highlighting that inflammatory markers alone cannot predict ossification risk, and there is an urgent need to find independent ossification markers as described in this invention.

[0054] Table 1. Basic demographic and clinical information of the participants

[0055] (3) such as Figure 1 As shown in A, the design process of this invention covers the entire process from iliac bone marrow aspiration, BMSCs isolation and culture to 10x Genomics single-cell sequencing.

[0056] We conducted an in-depth analysis of bone marrow mesenchymal stem cells from AS patients with different degrees of structural damage (SSD group with severe structural damage vs. NSD group without structural damage) using single-cell transcriptome sequencing (scRNA-seq) technology.

[0057] Single-cell atlas of bone marrow mesenchymal stem cells (BMSCs) as follows: Figure 1 (B) UMAP visualization of 47,628 BMSCs, identifying eight distinct cell subpopulations. The left panel shows all cells, and the right panel shows the distribution after grouping by no structural damage (NSD) and severe structural damage (SSD). (C) Proportional distribution of each BMSC subpopulation within the NSD and SSD groups.

[0058] Trajectory inference and differentiation potential analysis of BMSC subpopulations, such as Figure 2 (A) RNA velocity analysis projecting transcriptional kinetics onto UMAP; streamlines depict differentiation trajectories originating from progenitor cells (C5) and branching towards terminal states (C1 / C2). (B) UMAP visualization of discrete developmental potential states predicted by CytoTRACE2. (C) UMAP visualization of continuous CytoTRACE2 relative differentiation scores, where red indicates high potential (steminess) and blue indicates differentiated state.

[0059] Transcriptional and functional characterization of eight BMSC subsets, as follows Figure 3 (A) A bubble chart defining the top three differentially expressed genes (DEGs) for each subpopulation, where the size of the dots represents the proportion of expressing cells and the color intensity indicates the average expression level. (B) A schematic diagram summarizing the functional annotations and potential developmental relationships of eight BMSC subpopulations, highlighting the transition from proliferating progenitor cells to lineage-defined or senescent states. (C) A bubble chart depicting the metabolic pathway activities of different BMSC subpopulations, revealing the specific metabolic characteristics of each subpopulation. (D) A bubble chart showing the expression patterns of classic BMSC biomarkers grouped by functional categories.

[0060] Functional reclassification of BMSC subgroups, such as Figure 4 (A) UMAP visualization of reclassified subpopulations, divided into five functional types: proliferative matrix subpopulation (PSC, C5), microenvironment support matrix subpopulation (NSC, C1, and C3), bone remodeling subpopulation (ORC, C2, C6, and C8), aging-related subpopulation (SRC, C4), and immune regulatory subpopulation (IMC, C7). (B) Stacked bar chart comparing the proportion of cells of each functional type between the NSD and SSD groups, highlighting the expansion of the SRC, NSC, and IMC populations in the SSD group. (C) Venn diagram showing the overlap between upregulated DEGs and previously identified cell type-specific marker genes for identifying core functional genes.

[0061] The top 20 genes with high expression specific to the SRC subset are shown in Table 2.

[0062] Table 2. List of the top 20 genes highly expressed specifically in SRC subsets

[0063] Differentially expressed genes for different degrees of structural damage (SSD group vs. NSD group) are shown in Table 3.

[0064] Table 3. DEGs between SSD and NSD

[0065] Example 2: Screening for SRC subgroup-specific highly expressed genes (combination of core biomarkers) Identification and distribution of SRC subgroups (combined with...) Figure 1 B in Figure 1 C and Figure 2 ): Dimensionality reduction clustering analysis of single-cell data (e.g.) Figure 1 As shown in Figure B), this invention identified eight BMSC subsets. Among them, Cluster 4 showed a significantly increased cell density in the SSD group (e.g., ...). Figure 1 (As shown in C). Further combining Figure 2 Inference of RNA Velocity trajectory of A in the data and Figure 2 CytoTRACE2 differentiation potential analysis of C cells revealed that Cluster 4 is at a critical juncture in the transition from progenitor cells to terminal differentiation.

[0066] Among all identified BMSC subsets, the C4 cluster exhibited the most distinctive transcriptional and metabolic characteristics. Figure 3 C4 is characterized by the expression of CDR1, AL078639.1, and FGF14, exhibiting a transitional state between osteogenic differentiation and senescence signals, which differs from proliferative or immune-related subsets. Figure 3 (B in the original text). Single-cell metabolic analysis revealed the metabolic state of C4. Compared with the quiescent stem cell-like cluster (C8), C4 maintained a "hypermetabolic" phenotype characterized by high glycolysis and high oxidative phosphorylation. Figure 3 (C and D in the text). This high metabolic activity, maintained even in the absence of proliferation markers, is consistent with the high energy demand characteristic of the senescence-associated secretory phenotype (SASP). Based on these characteristics, we define C4 as the senescence-associated cluster (SRC). Figure 4 In group A), the SSD group showed a significant amplification of SRC (in this case). Figure 4 (B in the text), this specific accumulation suggests that SRC may be one of the key factors driving the pathological progression of ossification in AS.

[0067] Screening of core markers (combined with) Figure 4(See Tables C, 2, and 3 in the original text). To pinpoint the specific molecular tag of SRC, this invention employs... Figure 4 The Venn diagram strategy shown in C selects genes that simultaneously meet the criteria of "high expression specific to SRC subgroup (Table 2)" and "significant upregulation in SSD group (Table 3)," namely CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C.

[0068] The gene combinations obtained in this invention—CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C—all showed extremely high specificity in Table 2 (LogFC values ​​of 1.61, 1.21, 1.15, 1.01, and 1.12, respectively).

[0069] Functional annotations of these genes (combined with...) Figure 3 Functional analysis showed that the SRC subgroup possesses both high metabolic activity (CMBL, PPP1R3C) and a strong oxidative stress / aging response (SRXN1, CDR1).

[0070] Example 3: Drug prediction and screening based on SRC features Combination Figure 3 Based on the high glycolytic and oxidative phosphorylation metabolic characteristics of the SRC subpopulation revealed by C in Table 2, this invention utilizes the Drug2cell algorithm and the CMAP database to perform reverse drug matching prediction based on the SRC characteristic gene profile in Table 2. As shown in Table 4, JAK inhibitors (Filgotinib and Baricitinib) dominate the prediction list (Score>7.8) and account for over 90% of the non-zero group (Pct NZ Group). This suggests that JAK inhibitors not only inhibit inflammation but may also block the ossification process by reversing the metabolic and aging phenotypes of the SRC subpopulation.

[0071] Table 4. List of drug predictions for SRC subgroups

[0072] Example 4: Comparative Analysis of Diagnostic Performance of Biomarker Combinations and Single Biomarkers To verify the clinical application value of the biomarker combination described in this invention, this invention conducted an in-depth comparison of the diagnostic efficacy of a single biomarker and a 5-gene combination model based on qPCR detection data from 60 subjects (30 patients with severe structural lesions (SSD) vs. 30 controls with no structural lesions (NSD)).

[0073] (1) Detection method Total RNA was extracted from BMSCs, and the mRNA transcription abundance of five biomarkers and the internal reference GAPDH was detected by qPCR to obtain raw Delta Ct data (see Table 5).

[0074] Table 5. Raw qPCR data from 60 subjects

[0075] (2) Model construction and formula A predictive model was constructed using binary logistic regression. The calculated formula for the risk score P of the biomarker combination is as follows:

[0076] Note: The variable in the formula is the Delta Ct value of the corresponding gene.

[0077] (3) Threshold setting According to the ROC curve ( Figure 5 Based on the principle of maximizing the Yoden index, the optimal diagnostic cutoff value for the biomarker combination prediction model was determined to be 0.57.

[0078] P ≥ 0.57: Classified as high risk (SSD).

[0079] P < 0.57: classified as low risk (NSD).

[0080] (3) Performance comparison (Table 6) To clearly evaluate model performance, this embodiment defines the misclassification situation as follows: FN (False Negative): refers to a patient who actually has severe structural damage (SSD) but is incorrectly classified as low risk (NSD) by the model.

[0081] FP (False Positive): refers to a subject who actually has no structural lesions (NSD) but is incorrectly classified as high-risk (SSD) by the model.

[0082] Limitations of single biomarkers: The AUC of single biomarkers is generally between 0.80 and 0.83. Taking CDR1 as an example, although the AUC reaches 0.81, its specificity is only 70.0% at the optimal cutoff value, which means that 9 NSD controls were misclassified as positive (FP=9), and the sensitivity is only 80.0% (FN=6).

[0083] Combination Advantages: By integrating multi-dimensional expression signals, the gene combination model significantly improves diagnostic efficacy. The combination model achieves an AUC of 0.98, with sensitivity increasing to 90.0% and specificity reaching 96.7% at Cut-off=0.57. Compared to single biomarkers, the combination model drastically reduces the total number of misclassified samples from 15 (CDR1) to 4 (3 FN / 1 FP), demonstrating extremely high clinical translational value.

[0084] Table 6. Comparison of diagnostic performance between biomarker combinations and single biomarkers

[0085] Example 5: Reagent kit preparation and clinical application Based on the five genes screened in Example 2, specific primers and probes targeting the expression products of CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C were designed, and a detection kit was prepared.

[0086] Application process: Extract total RNA from patient BMSCs → Perform qPCR detection using the kit in this example → Calculate the comprehensive expression score.

[0087] Judgment criteria: If the score is significantly higher than the reference threshold for people with no structural damage (NSD), combined with Figure 1 The positive correlation between the SRC subgroup C shown in the figure and structural damage indicates the presence of an active pathological osteogenic subgroup in the patient, suggesting a high risk of spinal fusion. It is recommended to use JAK inhibitors (see Table 4) in addition to routine anti-inflammatory treatment for intervention.

[0088] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. A combination of biomarkers for the diagnosis of pathological osteogenic changes in ankylosing spondylitis, characterized in that, The biomarker combination includes CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C.

2. A product for the diagnosis of pathological osteogenic changes in ankylosing spondylitis, characterized in that, Includes reagents for detecting the expression levels of the biomarker combination of claim 1.

3. The product for diagnosing pathological osteogenic changes in ankylosing spondylitis according to claim 2, characterized in that, Reagents for detecting the expression levels of combinations of biomarkers include: (1) Primer pairs, probes or microarrays of transcripts of biomarker combinations; Or (2) antibodies, antigen-binding fragments or aptamers of the translation products of a combination of biomarkers.

4. The product for diagnosing pathological osteogenic changes in ankylosing spondylitis according to claim 2, characterized in that, The expression level of the biomarker combination is the abundance of the transcripts of the biomarker combination; The product for pathological osteogenic diagnosis of ankylosing spondylitis further includes a data processing module, which processes the abundance data of the transcripts of the biomarker combination, including: Abundance data of transcripts of biomarker combinations were calculated using the formula for risk score P. When the risk score P ≥ 0.57, the patient was identified as a patient with pathological osteoblastic ankylosing spondylitis, and when the risk score P < 0.57, the patient was identified as a patient with pathological osteoblastic non-ankylosing spondylitis. Among them, CDR1 △Ct PLPPR2 △Ct CMBL △Ct SRXN1 △Ct and PPP1R3C △Ct The abundance of transcripts for CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C are respectively.

5. A system for diagnosing pathological osteogenic changes in ankylosing spondylitis, characterized in that, include: The data detection module detects the expression levels of a combination of biomarkers in the bone marrow mesenchymal stem cell samples of the subjects; the combination of biomarkers includes CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C. The data processing module processes the expression level data of biomarker combinations and draws conclusions.

6. The system for diagnosing pathological osteogenic changes in ankylosing spondylitis according to claim 5, characterized in that, The expression level of the biomarker combination is the abundance of the transcripts of the biomarker combination; The data processing includes: Abundance data of transcripts of biomarker combinations were calculated using the formula for risk score P. When the risk score P ≥ 0.57, the patient was identified as a patient with pathological osteoblastic ankylosing spondylitis, and when the risk score P < 0.57, the patient was identified as a patient with pathological osteoblastic non-ankylosing spondylitis. Among them, CDR1 △Ct PLPPR2 △Ct CMBL △Ct SRXN1 △Ct and PPP1R3C △Ct The abundance of transcripts for CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C are respectively.

7. The use of the combination of biomarkers for the diagnosis of pathological osteogenic ankylosing spondylitis as described in claim 1, the product for the diagnosis of pathological osteogenic ankylosing spondylitis as described in any one of claims 2-4, or the system for the diagnosis of pathological osteogenic ankylosing spondylitis as described in any one of claims 5-6 in the preparation of a product for assessing the risk of progression of spinal structural damage in patients with ankylosing spondylitis.

8. The application according to claim 7, characterized in that, The assessment of the risk of progression of spinal structural damage in patients with ankylosing spondylitis includes: (a) Detect the expression levels of CDR1, PLPPR2, CMBL, SRXN1 and PPP1R3C in bone marrow mesenchymal stem cell samples derived from subjects; (b) Compare the expression levels obtained in step (a) with reference values; If the expression level of at least one of the genes CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C is higher than the reference value, the subject is considered to have a risk of progression of spinal structural injury.

9. The use of the combination of biomarkers for the diagnosis of pathological osteogenic ankylosing spondylitis as described in claim 1, the product for the diagnosis of pathological osteogenic ankylosing spondylitis as described in any one of claims 2-4, or the system for the diagnosis of pathological osteogenic ankylosing spondylitis as described in any one of claims 5-6 in screening for drugs that inhibit and / or treat pathological osteogenic ankylosing spondylitis.

10. The application according to claim 9, characterized in that, The method for screening drugs that inhibit and / or treat pathological osteogenic processes in ankylosing spondylitis includes: (1) Provide a cell model containing SRC subsets, wherein the SRC subsets highly express the combination of biomarkers; (2) Contact the compound to be screened with the cell model; (3) Detect changes in the expression levels of the biomarker combination in the cell model; If the compound can downregulate the expression of at least one of the genes CDR1, PLPPR2, CMBL, SRXN1, and PPP1R3C, then the compound is determined to be a potential drug for inhibiting and / or treating pathological osteoblastic ankylosing spondylitis.

Images