Molecular typing method and system for chordoma and application of molecular typing method and system

Abstract

The invention belongs to the technical field of biomedicine technologies, and particularly relates to a chordoma molecular typing method and system and application thereof. The invention provides a chordoma molecular typing method which comprises the following steps: performing protein expression analysis on a protein sample of a patient to generate protein expression profile data; performing clustering analysis on the basis of the protein expression profile data, analyzing specific protein expression characteristics obtained by clustering analysis, and identifying molecular subtypes of chordoma; specific protein expression characteristics are provided for each molecular subtype. The invention further provides a chordoma molecular typing system or device. The chordoma molecular typing system or device comprises a proteomics analysis module, a molecular subtype typing module and a typing output module. The methods, systems, or devices identify bone microenvironment dominant, mesenchymal derived, and mesenchymal epithelial transformations, thereby assessing clinical characteristics and prognosis of chordoma of different molecular subtypes.

Description

Technical Field

[0001] The present application belongs to the field of biomedical technology, and specifically relates to a chordoma molecular typing method and system and its application. Background Art

[0002] Chordoma is a rare malignant tumor that originates from residual cells of the notochord that have not degenerated during embryonic development, mainly occurring at the fusion sites of the axial bones of the spine and skull base. Chordoma is usually locally invasive and of low to moderate malignancy. It grows slowly but is prone to recurrence, and shows strong resistance to traditional radiotherapy and chemotherapy, so treatment options are very limited. Although surgical resection is the main treatment modality, the long-term prognosis of patients is often poor due to the complex anatomical location and easy recurrence of the tumor.

[0003] With the development of molecular biology technology, the study of the molecular characteristics of chordoma has gradually deepened, revealing its significant heterogeneity. Chordoma not only shows differences in gene expression levels, but also has different molecular subtypes in protein expression and function. Different chordoma molecular subtypes differ in biological behaviors such as growth and spread, heterogeneity, angiogenesis, and immune escape, resulting in differences in the efficacy of therapeutic drugs for chordomas of different molecular subtypes, which can easily lead to reduced drug therapeutic effects. Traditional histological typing methods cannot fully explain the differences in clinical behavior of chordomas. Therefore, the development of a typing system based on molecular characteristics is of great significance for the diagnosis, treatment, and prognosis evaluation of chordomas, and is conducive to the development of targeted and effective drugs and treatment plans for specific molecular subtypes of chordomas. Summary of the invention

[0004] In order to address the deficiencies of the prior art, this application analyzes the clinical and proteomic data of a large number of chordoma patients, and develops a chordoma molecular typing method, system, and device based on clinical proteomic characteristics, which molecularly types the three molecular subtypes of chordoma, including bone microenvironment-dominated type, mesenchymal type, and mesenchymal epithelial transformation type, and provides molecular features that are obvious in clinical prognosis and biological characteristics. The molecular typing method analyzes the proteomic data of chordoma patients to accurately molecularly type chordoma, thereby revealing the biological differences between different typings, and can provide survival prognosis evaluation, personalized treatment strategies, and new drug development guidance for chordoma patients with different molecular subtypes.

[0005] The first aspect of the present application provides a method for molecular typing of chordoma, comprising the following steps:

[0006] S1) performing protein expression analysis on the patient's protein sample to generate protein expression profile data;

[0007] S2) performing cluster analysis based on the protein expression profile data, analyzing the specific protein expression features obtained by the cluster analysis, and identifying the molecular subtypes of chordoma;

[0008] S3) provides specific protein expression signatures for each molecular subtype.

[0009] Preferably, the protein expression analysis uses tandem mass spectrometry labeling technology.

[0010] Preferably, the cluster analysis adopts a non-negative matrix decomposition algorithm.

[0011] Preferably, the molecular subtypes include any one of a bone microenvironment dominant type, a mesenchymal type, and a mesenchymal-epithelial transition type.

[0012] Preferably, when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of bone microenvironment cells, it is identified as the bone microenvironment-dominated type; when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of mesenchymal cells and mitochondrial oxidative phosphorylation, it is identified as the mesenchymal type; when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of mesenchymal epithelial transformation, epithelial cell proliferation, extracellular matrix decomposition, and tyrosine kinase reaction, it is identified as the mesenchymal epithelial transformation type.

[0013] A second aspect of the present application provides a system or device for molecular typing of chordoma, the system or device comprising:

[0014] A proteomics analysis module, used to perform protein expression analysis on protein samples of chordoma patients to obtain protein expression spectrum data; preferably, the protein expression analysis adopts tandem mass spectrometry labeling technology;

[0015] A molecular subtyping module, used to perform cluster analysis based on the protein expression spectrum data, analyze the specific protein expression features obtained by the cluster analysis, and identify the molecular subtypes of chordoma; preferably, the cluster analysis adopts a non-negative matrix decomposition algorithm;

[0016] The typing output module is used to output the molecular typing results and provide the specific protein expression characteristics of each molecular subtype.

[0017] A third aspect of the present application provides a computer-readable storage medium comprising a computer program, wherein when the computer program is executed by a processor, the chordoma molecular typing method as described in the first aspect is executed.

[0018] A fourth aspect of the present application provides a terminal comprising a memory and a processor, wherein the memory comprises a computer program, wherein when the computer program is executed by the processor, the chordoma molecular typing method as described in the first aspect is executed.

[0019] The fifth aspect of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the chordoma molecular typing method as described in the first aspect.

[0020] The sixth aspect of the present application provides a chordoma molecular typing kit, comprising reagents required for detecting any one of the bone microenvironment dominant type, mesenchymal type and mesenchymal epithelial transition type of a sample.

[0021] Preferably, the reagents required for tandem mass spectrometry labeling technology and / or the reagents required for protein extraction are included.

[0022] The seventh aspect of the present application provides uses of the system or device of the second aspect, or the computer-readable storage medium of the third aspect, or the terminal of the fourth aspect, or the computer program product of the fifth aspect, or the kit of the sixth aspect, wherein the uses include any one or more of the following:

[0023] 1) Prepare chordoma molecular typing products;

[0024] 2) Prepare a product for evaluating the prognosis of chordoma patients;

[0025] 3) Prepare a product for evaluating the clinical characteristics of chordoma patients;

[0026] 4) Prepare products for evaluating chordoma biological behavior;

[0027] 5) Prepare companion diagnostic products for chordoma.

[0028] The beneficial effects of this application are:

[0029] The molecular typing method, system and device developed in this application can not only provide accurate typing diagnosis for chordoma patients, but also perform molecular typing of chordoma based on tumor samples of chordoma patients, and identify three chordoma subtypes with significant molecular characteristics: bone microenvironment-dominated type, mesenchymal type and mesenchymal epithelial transformation type. Based on the specific protein expression characteristics of the provided molecular subtypes, the prognosis and biological behavior of chordomas of different molecular subtypes can be evaluated.

[0030] The molecular typing method and system have good accuracy and stability, and can provide important molecular targets for clinical treatment decisions and new drug development, which is conducive to the development of targeted drugs and treatments for specific chordoma molecular subtypes. These typings can not only provide more accurate prognostic evaluation for chordoma patients, but also guide the formulation of personalized treatment plans and provide potential targets for the development of new drugs, which is of great significance for the accurate diagnosis and personalized treatment of chordoma. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 Schematic diagram of the molecular classification system for chordoma.

[0032] Figure 2 Schematic diagram of molecular classification of chordoma based on proteomic analysis; (A) non-negative matrix factorization analysis of three subtypes; (B) NMF clustering based on proteomic data to identify molecular subtypes; (C) PCA diagram showing t-SNE distribution of chordoma samples in the three identified subtypes; (D) Venn diagram for identifying the three subtypes; (E) ternary diagram for identifying the three subtypes.

[0033] Figure 3 Figure 2 shows the overall survival Kaplan-Meier curves of patients with three molecular subtypes; cluster 1 is bone microenvironment-dominant chordoma, cluster 2 is mesenchymal chordoma, and cluster 3 is mesenchymal-epithelial transformation chordoma.

[0034] Figure 4 Figure 3 is a schematic diagram of the three molecular classifications of chordomas based on proteomic analysis and their clinical characteristics; in the three identified clusters, the proportion of chordomas patients with different tumor locations (A), previous surgical treatments (B), T1-weighted imaging signals (C), genders (D), ages (E), KPS scores (F), distant metastases (G), pathological grades (H), tumor sizes (I), Ki-67 (J), and resection methods (K).

[0035] Figure 5 Schematic diagram of NMF clustering based on functional enrichment analysis of differentially expressed proteins (DEPs) for patients with three molecular subtypes.

[0036] Figure 6 Schematic diagram of the biological characteristics of patients with three molecular subtypes; among the three identified clusters, ESTIMATE tumor purity (A), stromal score (B), immune score (C), protein expression of key immune microenvironment markers (D) and bone microenvironment markers (E), estimated proportions of monocytes (F) and CD8 T cells (G), mitochondrial protein expression (H) and mitochondrial protein expression of complex I (I), mitochondrial protein expression of complex II (J), mitochondrial protein expression of complex III (K), mitochondrial protein expression of complex IV (L), mitochondrial protein expression of complex V (M), protein expression of previously reported chordoma mutated genes (N), protein expression of previously reported chordoma chromatin modeling genes (O).

[0037] Figure 7Schematic diagram of the biological characteristics of patients with three molecular subtypes; box plots of chemokine receptor binding to chemokines (A), complement cascade reaction (B), respiratory electron transport (C), complex I biosynthesis (D), FGFR3 binding and activation (E), and PI3K-AKT activation (F) in the three molecular subtypes of chordoma. DETAILED DESCRIPTION

[0038] The following describes the embodiments of the present invention through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention.

[0039] Before further describing the specific embodiments of the present invention, it should be understood that the scope of protection of the present invention is not limited to the specific embodiments described below; it should also be understood that the terms used in the examples of the present invention are for describing the specific embodiments rather than for limiting the scope of protection of the present invention; in the present specification and claims, unless otherwise expressly stated herein, the singular forms "a", "an" and "the" include plural forms.

[0040] When the embodiments give numerical ranges, it should be understood that, unless otherwise specified in the present invention, both endpoints of each numerical range and any numerical value between the two endpoints can be selected. Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those generally understood by those skilled in the art. In addition to the specific methods, equipment, and materials used in the embodiments, according to the grasp of the prior art by those skilled in the art and the record of the present invention, any methods, equipment, and materials of the prior art similar or equivalent to the methods, equipment, and materials described in the embodiments of the present invention can also be used to realize the present invention.

[0041] This application proposes three molecular subtypes of chordoma by analyzing the clinical and proteomic data of a large number of chordoma patients: bone microenvironment-dominated type, mesenchymal type, and mesenchymal epithelial transformation type, and develops a molecular typing method, system, and device for chordoma based on clinical proteomic characteristics, including extracting protein samples from tumor tissues of chordoma patients, using tandem mass spectrometry (TMT) technology to perform in-depth analysis of protein samples, and obtaining protein expression spectrum data. Subsequently, the non-negative matrix decomposition (NMF) algorithm was used to perform cluster analysis on protein expression data, and three chordoma subtypes with significant molecular characteristics were identified: bone microenvironment-dominated type, mesenchymal type, and mesenchymal epithelial transformation type, and the method and system were verified to have good accuracy and stability.

[0042] The present application first provides a chordoma molecular typing method, comprising the following steps:

[0043] S1) performing protein expression analysis on the patient's protein sample to generate protein expression profile data;

[0044] S2) performing cluster analysis based on the protein expression profile data, analyzing the specific protein expression features obtained by the cluster analysis, and identifying the molecular subtypes of chordoma;

[0045] S3) provides specific protein expression signatures for each molecular subtype.

[0046] The chordoma is a rare malignant tumor derived from the residual cells of the notochord that have not degenerated during embryonic development, and mainly occurs at the fusion site of the axial bones of the spine and the base of the skull. Chordoma is locally invasive and usually low to moderately malignant. Although surgical resection is the main method for treating chordoma, since chordoma is often located in a part with complex anatomical structure, it is difficult to completely remove tumor tissue by surgical resection, resulting in a high recurrence rate. In addition, chordoma shows significant tolerance to traditional chemotherapy and radiotherapy, which generally makes the patient's prognosis poor. Chordoma includes but is not limited to classic chordoma, chondroid chordoma, dedifferentiated chordoma, ossifying chordoma, clear cell chordoma, and neurochordoma.

[0047] Molecular typing refers to a classification method based on the molecular characteristics of tumor cells. By analyzing the molecular information of the tumor genome, transcriptome, proteome and epigenetics, the tumor is divided into different molecular subtypes according to its molecular characteristics. These subtypes reflect the heterogeneity of the tumor at the molecular level and are closely related to the biological behavior of the tumor, treatment response and patient prognosis.

[0048] The protein expression analysis refers to a series of experimental techniques used to study and quantify the expression level, type and modification status of proteins in cells or tissues, which may include protein qualitative analysis: determining what types of proteins are present in a sample, and their specific forms (such as splicing variants, isomers, etc.); protein quantitative analysis, measuring the expression level of a specific protein, and understanding its abundance changes under different conditions; protein modification analysis, detecting post-translational modifications of proteins, such as phosphorylation, glycosylation, ubiquitination, etc., which can significantly affect the function of proteins; protein localization analysis, determining the location of proteins, and understanding their spatial distribution within cells; protein interaction analysis, studying the interactions between proteins, and revealing the formation and functional networks of protein complexes.

[0049] The methods used for the protein expression analysis include but are not limited to Western Blot, enzyme-linked immunosorbent assay (ELISA), mass spectrometry, immunohistochemistry (IHC), flow cytometry, protein chip technology, fluorescence resonance energy transfer (FRET). The mass spectrometry includes but is not limited to tandem mass spectrometry labeling technology (TMT), matrix-assisted laser desorption ionization (MALDI), ion mobility mass spectrometry (IMS), high-resolution mass spectrometry (High-Resolution Mass Spectrometry), data-independent acquisition (DIA), single-cell proteomics analysis, mass spectrometry imaging (MSI), and liquid chromatography-mass spectrometry (LC-MS / MS).

[0050] In a specific embodiment of the present application, the protein expression analysis adopts tandem mass spectrometry labeling technology (TMT), which is a mass spectrometry-based proteomics quantitative method that uses stable isotope-labeled reagents to react with peptides in the sample to achieve quantitative analysis of protein expression. Its main feature is the use of labeling reagents in the sample preparation stage, so that the peptides to be quantified are labeled with equal mass labels composed of multiple groups, and in the subsequent detection, the peptides and proteins are quantified based on the signals of the label groups, including the preparation of protein samples and the extraction of proteins from biological samples; proteolysis, the extracted proteins are hydrolyzed into peptides by enzymes such as trypsin; TMT labeling, the peptides are labeled with TMT reagents, which are composed of three parts: a reporter group, a balance group and a peptide reaction group, and can covalently bind to the amino groups of the peptides; sample mixing and liquid chromatography separation, after equal amounts of labeled peptides from different samples are mixed, they are initially separated by liquid chromatography; mass spectrometry analysis, the samples are analyzed using a mass spectrometer, including primary mass spectrometry and secondary mass spectrometry, the primary mass spectrometry is used for the preliminary separation of peptides, and the secondary mass spectrometry is used for the abundance and sequence detection of peptides; data analysis, the qualitative and quantitative information of peptides and proteins is obtained through mass spectrometry data, and then bioinformatics analysis is performed to reveal the differences in protein expression and biological significance.

[0051] The protein expression profile data refers to detailed information about the expression levels of all proteins in a biological sample obtained by high-throughput proteomics technology, such as mass spectrometry. These data usually include protein identification information (i.e., which proteins exist) and their relative or absolute quantitative information (i.e., how much protein there is), and can also include post-translational modification (PTM) analysis, changes in protein expression levels, protein network and pathway analysis.

[0052] The cluster analysis refers to an unsupervised learning method that groups objects or samples in a data set into multiple clusters or groups based on similarity. Objects in the same cluster have higher similarity, while objects in different clusters have lower similarity. It includes distance measurement, determining a suitable distance measurement or similarity measurement standard, such as Euclidean distance, Manhattan distance, cosine similarity, etc., to measure the similarity between data points; selecting a clustering algorithm, selecting a suitable clustering algorithm according to the characteristics of the data and the analysis objectives; determining the number of clusters. In some clustering methods, the number of clusters needs to be specified in advance (such as K in K-means), and in other methods, such as hierarchical clustering, the number of clusters can be determined by analyzing the clustering tree; clustering execution: running the clustering algorithm to group the data, clustering result evaluation, and evaluating the quality of the clustering results. Indicators such as the Silhouette coefficient and the Davies-Bouldin index can be used; result interpretation: interpreting the clustering results to understand the potential meaning represented by different clusters. The clustering algorithms include but are not limited to non-negative matrix factorization algorithm (NMF), K-Means clustering algorithm, hierarchical clustering, DBSCAN (Density-Based Spatial Clustering of Applications with Noise), mean shift (Mean Shift), Gaussian mixture model (Gaussian Mixture Model, GMM), spectral clustering, density peak clustering (Density Peak Clustering), fuzzy C-means (Fuzzy C-Means), OPTICS (Ordering Points To Identify the Clustering Structure), BIRCH (Balanced Iterative Reducing and Clustering Using Hierarchies), Affinity Propagation, Mean Shift, and Deep NMF.

[0053] In a specific implementation of the present application, the cluster analysis uses a non-negative matrix factorization algorithm (NMF) to decompose a non-negative matrix V into the product of two non-negative matrices W and H, that is, V≈WH. In which, each column in the V matrix represents an observation, and each row represents a feature; the W matrix is ​​called a base matrix, and the H matrix is ​​called a coefficient matrix or a weight matrix. Through non-negative matrix factorization, the coefficient matrix H can be used to replace the original matrix to achieve dimensionality reduction of the original matrix and obtain a dimensionality reduction matrix of the data features, including initialization, randomly initializing two non-negative matrices W (base matrix) and H (coefficient matrix). The dimensions of W are usually m×k, and the dimensions of H are k×n, where m and n are the number of rows and columns of the original data matrix V, respectively, and k is the rank of the decomposition; select the objective function, determine a suitable objective function to measure the difference between the original matrix V and the reconstructed matrix WH after decomposition. Common objective functions include the squared Frobenius norm (i.e., Euclidean distance) and Kullback-Leibler divergence; iterative optimization, use an iterative algorithm to alternately update W and H to minimize the objective function. This process usually involves the following steps: fix H, update W, adjust the elements of W by solving the optimization problem, fix W, update H, adjust the elements of H by solving the optimization problem. During the iteration process, methods such as multiplication update rules or gradient descent can be used to ensure non-negativity constraints; convergence judgment, set a convergence threshold (such as the change in the objective function is less than a set epsilon value) and a maximum number of iterations. When the improvement of the objective function is less than the threshold or the maximum number of iterations is reached, the algorithm stops iterating; Result evaluation, evaluate the quality of the decomposition, which can be done by calculating the reconstruction error, viewing the characteristics of the basis matrix W or the distribution of the coefficient matrix H, etc.; Application results, use the obtained W and H for subsequent analysis, such as feature extraction, data dimensionality reduction, pattern recognition, etc. The column vector of W can be regarded as a new feature basis of the data, and the row vector of H represents the projection of the original data on these new features; Post-processing, depending on the application requirements, the results may need to be post-processed, such as further analysis of W or H, or using them to build a prediction model; Visualization, visualize the W and H matrices to help understand the underlying structure and pattern of the data. For example, in image processing, the column vectors of the W matrix can be displayed to identify the basis elements of the image.

[0054] In a specific embodiment of the present application, based on the results of cluster analysis, the specific protein expression characteristics obtained by cluster analysis are analyzed, and the molecular subtypes of chordoma are identified.

[0055] The specific protein expression characteristics refer to that in a specific cell type, tissue or physiological state, the expression level of certain proteins is significantly higher or lower than the expression level under other conditions.

[0056] The molecular subtypes of chordoma include any one of bone microenvironment-dominant type, mesenchymal type and mesenchymal-epithelial transition type.

[0057] Through the present application, clinical and proteomic data of a large number of chordoma patients were analyzed, and the present application found that the bone microenvironment-dominant type exhibited a specific protein expression spectrum related to bone destruction, and exhibited the protein expression characteristics of bone microenvironment cells, and the bone microenvironment cells included osteoclasts, osteoblasts, mast cells, neutrophils, NK cells, macrophages, T cells and B cells; specifically, the protein expression characteristics of the bone microenvironment cells may include any one or more immune microenvironment markers such as CD3G, CD8A, MZB1, CD14, CD16a, CD16b, CD47, and CXCL13, and may also include any one or more bone-specific microenvironment markers such as SEMA7A, MMP9, OSCAR, and CA2, and may also include the infiltration level of immune cells such as CD8+T cells or monocytes.

[0058] The mesenchymal type exhibits protein expression characteristics of mesenchymal cells and mitochondrial oxidative phosphorylation, and its characteristic proteins are closely related to energy metabolism; the mitochondrial oxidative phosphorylation includes respiratory electron transfer and complex I biosynthesis; specifically, the protein expression characteristics of mesenchymal cells and mitochondrial oxidative phosphorylation may include any one or more of mitochondrial respiratory chain complex IV, PBRM1 protein, SETD2 protein, and ARID1A protein.

[0059] The mesenchymal epithelial transition type is characterized by the enrichment of epithelial characteristics (mesenchymal epithelial transition, epithelial cell proliferation), the decomposition of the extracellular matrix and the protein expression characteristics of the response to tyrosine kinase (PDGF). The enrichment of epithelial characteristics (mesenchymal epithelial transition, epithelial cell proliferation), the decomposition of the extracellular matrix and the protein expression characteristics of the response to tyrosine kinase (PDGF) can include any one or more of TBXT protein, CDKN2A protein, LYST protein and ACAN protein.

[0060] Therefore, in a specific embodiment of the present application, when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of bone microenvironment cells, it is identified as the bone microenvironment-dominated type; when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of mesenchymal cells and mitochondrial oxidative phosphorylation, it is identified as the mesenchymal type; when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of mesenchymal epithelial transformation, epithelial cell proliferation, extracellular matrix decomposition, and tyrosine kinase reaction, it is identified as the mesenchymal epithelial transformation type.

[0061] In a specific embodiment of the present application, the chordoma molecular subtyping method further comprises providing specific protein expression characteristics of each molecular subtype.

[0062] In a specific embodiment of the present application, the chordoma molecular typing method also includes extracting a protein sample from a tumor tissue of a chordoma patient, wherein the tumor tissue is a chordoma tumor tissue. The method for extracting the protein sample is not limited, and any method that can effectively extract protein from the sample can be used. For example, mechanical homogenization can be used, using a mortar and pestle or a homogenizer to physically grind the tissue into fine fragments, and then adding a lysis buffer to release the protein; chemical lysis, mixing the tissue sample with a lysis buffer containing a surfactant, a protein denaturant, and a reducing agent to dissolve the cell membrane and intracellular structure to release the protein; ultrasonic lysis, using the mechanical action of ultrasound to destroy the cell structure, thereby releasing the protein. This method is fast and effective, but it requires controlled ultrasound conditions to avoid protein degradation; enzymatic hydrolysis, using proteases or other enzymes to treat tissue samples to help break down the extracellular matrix and cell membranes, thereby releasing proteins; organic solvent precipitation, using organic solvents such as methanol, ethanol or acetone to precipitate proteins, and then separating the precipitate by centrifugation; thermal lysis, heating tissue samples to high temperatures to break the cell structure and release proteins; cryolysis, rapidly freezing tissue samples, and then mechanically breaking or using lysis buffer to release proteins; pressure lysis, using a high-pressure homogenizer or pressure cycling technology (Pressure Cycling Technology, PCT) to lyse tissues and release proteins; immunoprecipitation, if the target is a specific protein, specific antibodies can be used for immunoprecipitation to enrich the target protein; affinity chromatography, using specific ligands (such as metal affinity chromatography, lectin affinity chromatography, etc.) to capture the target protein; density gradient centrifugation, separating proteins according to their density differences by centrifugation in a density gradient medium; electrophoretic separation, using techniques such as SDS-PAGE or isoelectric focusing electrophoresis to separate proteins, and then staining and quantitative analysis.

[0063] The present application also provides a system or device for molecular typing of chordoma, the system or device comprising:

[0064] A proteomics analysis module, used to perform protein expression analysis on protein samples of chordoma patients to obtain protein expression spectrum data; preferably, the protein expression analysis adopts tandem mass spectrometry labeling technology;

[0065] A molecular subtyping module, used to perform cluster analysis based on the protein expression spectrum data, analyze the specific protein expression features obtained by the cluster analysis, and identify the molecular subtypes of chordoma; preferably, the cluster analysis adopts a non-negative matrix decomposition algorithm;

[0066] The typing output module is used to output the molecular typing results and provide the specific protein expression characteristics of each molecular subtype.

[0067] The protein sample extraction, protein expression analysis, tandem mass spectrometry labeling technology, cluster analysis, non-negative matrix decomposition algorithm, and specific protein expression characteristics are as described above.

[0068] In a specific embodiment of the present application, the molecular subtype includes any one or more of the bone microenvironment dominant type, the mesenchymal type and the mesenchymal epithelial transition type, and the molecular subtype is as described above.

[0069] In a specific embodiment of the present application, when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of bone microenvironment cells, it is identified as the bone microenvironment-dominated type; when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of mesenchymal cells and mitochondrial oxidative phosphorylation, it is identified as the mesenchymal type; when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of mesenchymal epithelial transformation, epithelial cell proliferation, extracellular matrix decomposition, and tyrosine kinase reaction, it is identified as the mesenchymal epithelial transformation type.

[0070] Specifically, the protein expression characteristics of the bone microenvironment cells may include any one or more immune microenvironment markers such as CD3G, CD8A, MZB1, CD14, CD16a, CD16b, CD47, and CXCL13, or any one or more bone-specific microenvironment markers such as SEMA7A, MMP9, OSCAR, and CA2, or the infiltration level of immune cells such as CD8+T cells or monocytes; the protein expression characteristics of the mesenchymal cells and mitochondrial oxidative phosphorylation may include any one or more of mitochondrial respiratory chain complex IV, PBRM1 protein, SETD2 protein, and ARID1A protein; the protein expression characteristics of mesenchymal epithelial transformation, epithelial cell proliferation, extracellular matrix decomposition, and response to tyrosine kinase (PDGF) may include any one or more of TBXT protein, CDKN2A protein, LYST protein, and ACAN protein.

[0071] The present application also provides a computer-readable storage medium, comprising a computer program, characterized in that when the computer program is executed by a processor, the above-mentioned chordoma molecular typing method is executed.

[0072] The present application also provides a terminal, comprising a memory and a processor, wherein the memory comprises a computer program, and the computer program is characterized in that when the computer program is run by the processor, the above-mentioned chordoma molecular typing method is executed.

[0073] The present application also provides a computer program product, including a computer program, which implements the above-mentioned chordoma molecular typing method when executed by a processor.

[0074] It should be noted that it should be understood that the division of the various modules of the system or device is only a division of logical functions. In actual implementation, they can be fully or partially integrated into a physical entity, or they can be physically separated. And these modules can all be implemented in the form of software called by processing elements, or they can all be implemented in the form of hardware, and some modules can be implemented in the form of software called by processing elements, and some modules can be implemented in the form of hardware. For example: a certain module can be a separate processing element, or it can be integrated in a chip of the following system. In addition, a certain module can also be stored in the memory of the following system in the form of program code, and called by a processing element of the following system and execute the function of the following module. The implementation of other modules is similar. These modules can be integrated together in whole or in part, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the following modules can be completed by an integrated logic circuit of hardware in the processor element or instructions in the form of software.

[0075] For example, these modules in the embodiments of the present application may be one or more integrated circuits configured to implement the above methods, such as: one or more specific integrated circuits (Application Specific Integrated Circuit, referred to as ASIC), or, one or more microprocessors (digital singnal processor, referred to as DSP), or, one or more field programmable gate arrays (Field Programmable Gate Array, referred to as FPGA) or graphics processing units (Graphics Processing Unit, referred to as GPU), etc. For another example, when a certain module above is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (Central Processing Unit, referred to as CPU) or other processors that can call program codes. For another example, these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0076] In a specific embodiment of the present application, the computer-readable storage medium may include a read-only memory, a random access memory, an EEPROM, a CD-ROM or other optical disk storage device, a disk storage device or other magnetic storage device, a flash memory, a USB flash drive, a mobile hard disk, or any other medium that can be used to store a desired program code in the form of an instruction or data structure and can be accessed by a computer. In addition, any connection can be appropriately referred to as a computer-readable storage medium. For example, if the instruction is sent from a website, a server or other remote source using a coaxial cable, an optical fiber cable, a twisted pair, a digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwaves, the coaxial cable, optical fiber cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwaves are included in the definition of the medium. However, it should be understood that the computer-readable storage medium does not include carrier waves, signals, or other temporary media, but is intended to be directed to non-temporary, tangible storage media.

[0077] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to the computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other media that can store program codes.

[0078] In a specific embodiment of the application, the terminal may be a conventional device such as a computer device, a tablet device, a personal digital processing device, a factory background processing device, etc. The terminal may include: at least one processor, a memory, at least one network interface and a user interface. The various components in the device are coupled together through a bus system. It is understood that the bus system is used to achieve connection and communication between these components. In addition to the data bus, the bus system also includes a power bus, a control bus and a status signal bus. In addition, the user interface may include a display, a keyboard, a mouse, a trackball, a click gun, a key, a button, a touch pad or a touch screen, etc.

[0079] The above-mentioned memory may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM) or a programmable read-only memory (PROM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM) and synchronous static random access memory (SSRAM).

[0080] The above-mentioned processor can be a general-purpose processor, a digital signal processor (DSP, Digital Signal Processor), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor can implement or execute the disclosed methods, steps and logic block diagrams in the embodiments of the present invention. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0081] In various embodiments, the present application can be implemented as a system or method for performing the above-mentioned technology alone or in any combination. In another embodiment, the present application can be implemented as a computer program product including a computer-readable storage medium and a computer program code encoded on the medium for performing the above-mentioned technology. It should be understood that various programming languages ​​can be used to implement the present application and are not limited to a specific programming language.

[0082] The present application also provides a chordoma molecular typing kit, comprising reagents required for detecting any one of the bone microenvironment dominant type, mesenchymal type and mesenchymal epithelial transition type of a sample.

[0083] In a specific embodiment of the present application, the kit may include reagents required for tandem mass spectrometry labeling technology and / or reagents required for protein extraction.

[0084] The reagents required for the tandem mass spectrometry labeling technology may include, for example, TMT labeling reagents (comprising a reporter group, a balance group and a reactive group), peptide lysis buffer (denaturing agents such as urea and CHAPS), reducing agents, alkylating agents, strong acid solvents (such as trifluoroacetic acid), and liquid chromatography solvents (including water, acetonitrile and organic acids such as formic acid or acetic acid).

[0085] The reagents required for protein extraction may include, for example, a lysis buffer (comprising salt, detergent, pH buffer, etc.) for destroying the cell membrane and cell wall to release the protein in the cell; protease inhibitors (such as PMSF (phenylmethylsulfonyl fluoride), AEBSF, Leupeptin, Peptstatin, etc.); A, etc.), in order to prevent protein degradation, it is necessary to add protease inhibitors; phosphatase inhibitors (such as sodium fluoride, sodium orthovanadate, etc.), when extracting phosphorylated proteins, it is necessary to add phosphatase inhibitors to prevent dephosphorylation of phosphorylated proteins; metal ion chelators (such as EDTA or EGTA), used to chelate metal ions and inhibit the activity of metal-dependent proteases; reducing agents (such as DTT (dithiothreitol) or BME (beta-mercaptoethanol)), used to destroy disulfide bonds in proteins and help protein denaturation and unfolding; loading buffer (containing denaturants (such as SDS), reducing agents, glycerol, etc.), used to treat protein samples before electrophoresis to denature them and prepare them for loading; nucleases, used to degrade DNA or RNA in the sample, reduce viscosity, and facilitate protein extraction and subsequent operations; lysozyme, used to decompose bacterial cell walls and improve the extraction efficiency of proteins or nucleic acids; other special reagents (such as specific buffers, salt solutions, organic solvents, etc.) to adapt to specific extraction conditions or the characteristics of the target protein.

[0086] The present application also provides uses of the above-mentioned system or device, computer-readable storage medium, terminal, computer program product, and kit, which include any one or more of the following:

[0087] 1) Prepare chordoma molecular typing products;

[0088] 2) Prepare a product for evaluating the prognosis of chordoma patients;

[0089] 3) Prepare a product for evaluating the clinical characteristics of chordoma patients;

[0090] 4) Prepare products for evaluating chordoma biological behavior;

[0091] 5) Prepare companion diagnostic products for chordoma.

[0092] The product may include, for example, but is not limited to, a drug, a drug composition, a combination of drugs, a kit, a chip, a system, and a device.

[0093] The molecular typing product is used to distinguish whether the chordoma suffered by a chordoma patient is of the bone microenvironment-dominated type, the mesenchymal type, or the mesenchymal-epithelial transformation type.

[0094] The prognosis assessment refers to the process of predicting and evaluating the development of a disease, treatment effects, outcome and possible results based on factors such as the patient's disease characteristics, treatment measures, and individual differences.

[0095] In a specific embodiment of the present application, when the molecular subtype of chordoma is bone microenvironment-dominated, the patient's survival time is predicted to be 158.7-202.2 months; when the molecular subtype of chordoma is mesenchymal, the patient's survival time is predicted to be 46.5-70.4 months; when the molecular subtype of chordoma is mesenchymal-epithelial transformation, the patient's survival time is predicted to be 98.3-132.5 months.

[0096] The clinical characteristics refer to the various symptoms and signs manifested during the natural development of the patient's disease. In a specific embodiment of the present application, molecular typing of chordoma can be used to evaluate the clinical characteristics of chordoma. The clinical characteristics of chordoma may include, for example, tumor site, treatment history, and MRI-T1 signal; wherein the tumor site includes the sacrum and mobile spine, the treatment history includes primary and recurrent, and the MRI-T1 signal includes low signal, equal signal, and high signal.

[0097] For example, when the chordoma molecular subtype is bone microenvironment-dominated, there are more primary chordomas, chordomas with low signal on T1-weighted MRI, and fewer recurrent chordomas, chordomas with equal signal on T1-weighted MRI, and chordomas with high signal on T1-weighted MRI compared with other molecular subtypes; when the chordoma molecular subtype is mesenchymal, there are more mobile spinal chordomas, recurrent chordomas, chordomas with high signal on T1-weighted MRI, and fewer sacral chordomas, primary chordomas, and chordomas with low signal on T1-weighted MRI compared with other molecular subtypes; when the chordoma molecular subtype is mesenchymal-epithelial transformation, there are more sacral chordomas, chordomas with equal signal on T1-weighted MRI, and fewer mobile spinal chordomas.

[0098] The biological behaviors include growth and spread, heterogeneity, angiogenesis, immune escape, etc.; the growth and proliferation refers to the abnormal proliferation ability of tumor cells, which can continuously divide and grow to form tumor masses. Tumor cells can also invade surrounding tissues and spread to other parts; the heterogeneity refers to the heterogeneity of cells in tumor tissues in genes, phenotypes and functions, leading to the emergence of different cell subpopulations in tumors, which may have different growth rates, invasiveness and sensitivity to anticancer treatment; the angiogenesis refers to the fact that tumors need blood supply to provide nutrients and oxygen. In order to meet their own growth needs, tumors will induce angiogenesis and form new vascular networks; the immune escape refers to the fact that tumor cells can escape the attack of the host immune system through various mechanisms. They can express molecules that inhibit the function of immune cells, such as PD-L1, or regulate immune cells in the tumor microenvironment, making them unable to effectively recognize and attack tumor cells.

[0099] The companion diagnostic products are used in combination with specific therapeutic agents (including drugs and biologics) to determine which patients are most likely to benefit from the treatment or which patients may be at risk of serious adverse reactions from the use of a specific treatment.

[0100] In the present application, the methods, systems, devices, computer-readable storage media, terminals, computer program products, kits, and products may be used by mammals diagnosed with chordoma, such as humans; primates, such as monkeys and other primates; livestock, such as sheep, cattle, horses, donkeys, and pigs; pets, such as dogs and cats; laboratory animals, such as mice, rabbits, rats, guinea pigs, and hamsters; captured wild animals, such as foxes and deer; and in a preferred embodiment of the present application, the users are humans.

[0101] In this application, the method for molecular classification of chordoma is for non-disease diagnosis and treatment purposes, such as for dividing confirmed patients into specific molecular subtypes, studying the biological characteristics of tumors, discovering new therapeutic targets, promoting interdisciplinary research, improving the accuracy of tumor classification, studying tumor microenvironment, discovering new biomarkers, education and training, policy making and public health planning.

[0102] The present application is further illustrated by the following examples, but the scope of the present application is not limited thereby. Unless otherwise specified, the experimental methods, detection methods, and preparation methods disclosed in the present invention all adopt conventional molecular biology, biochemistry, chromatin structure and analysis, analytical chemistry, cell culture, recombinant DNA technology, and conventional techniques in related fields in the field of the present technology. These techniques have been fully described in the existing literature. Chordoma tumor tissue was provided by Shanghai First People's Hospital, and the patient was informed and in compliance with ethical regulations.

[0103] Example 1 Sample Collection

[0104] The collection included 102 pathologically diagnosed chordoma samples obtained from surgical specimens between 2016 and 2023 (Shanghai First People's Hospital). All patients or their legal guardians signed written informed consent for tissue and clinical data collection. Soft tissue chordoma samples were processed by formalin-fixed paraffin-embedded (FFPE) method. Events were defined as the first evidence of local recurrence in progression-free survival (PFS) or cancer-related death or cancer-specific survival (CSS) or overall survival (OS) due to any cause. Event time was defined as the interval from the date of surgery to local recurrence, death, or until July 2023 (for living patients). Recurrence / death status and follow-up time were obtained by telephone interview. KM variable survival analysis and visualization were performed using the R packages "survival" and "survminer".

[0105] Imaging studies (including CT and / or MRI) were obtained through an electronic imaging management system or provided by the patient or his / her legal guardian. Chordoma and surrounding structures in imaging studies were identified by oncologists and radiologists. All samples were evaluated by a professional pathologist to determine the pathological subtype of chordoma, including conventional, chondroid, and differentiated. The proportion of tumor cells and the degree of fibrosis / necrosis were identified. Samples used for proteomic analysis were those with a tumor tissue proportion greater than 90% on the slide.

[0106] Example 2 Proteomic Analysis

[0107] This example uses high-throughput proteomics analysis technology based on tandem mass spectrometry tag (TMT) to perform in-depth analysis on the samples of Example 1. During the analysis, protein quantitative detection is used to obtain expression data of a large number of proteins in each sample, and the proteins are separated and identified by a mass spectrometer.

[0108] Experimental methods:

[0109] (1) Dewaxing and hydration preparation steps: Weigh 1 mg of sample and place it in a 1.5 ml Eppendorf (EP) tube. Add 1 ml of n-heptane to the tube, react at 800 rpm and 25°C for 10 minutes, and then discard the solution (this step is repeated once). Subsequently, hydration is performed by adding 1 ml of 100% ethanol, and the reaction conditions are the same (800 rpm, 25°C, 5 minutes), and then the solution is discarded. 200 μl of 90% ethanol and 200 μl of 75% ethanol are added in sequence, and the previous solution is discarded before each addition of new solution (this step is repeated twice, and each reaction is carried out at 800 rpm and 25°C for 5 minutes). Next, 150 μl of 0.1% formic acid is added, and acid hydrolysis is performed at 600 rpm and 30°C for 30 minutes. Then, 150 μl of 0.1 M Tris-HCl (Tris(hydroxymethyl)aminomethane hydrochloride) was added, and alkaline hydrolysis was performed at 600 rpm and 95° C. for 30 minutes.

[0110] (2) Fractionation and mass spectrometry gradient settings: Sample fractionation: A gradient of 5% to 35% acetonitrile containing 10 mM ammonium hydroxide solution (pH = 10.0) was used to generate 60 peptide fractions at a flow rate of 0.5 ml / min. These fractions were then combined into 30 portions and dried separately. The samples were redissolved in 98% H2O, 2% ACN (acetonitrile) and 0.1% formic acid solution for mass spectrometry analysis. Liquid chromatography gradient and mass spectrometry parameters: An UltiMate 3000RSLCnano system was coupled to an Orbitrap Exploris 480 mass spectrometer (Thermo Scientific, San Jose, USA) equipped with a FAIMS Pro TMThe interface was used for data dependent acquisition (DDA) analysis. Mobile phase A consisted of 2% acetonitrile, 98% water, and 0.1% formic acid; mobile phase B consisted of 98% acetonitrile, 2% water, and 0.1% formic acid. All reagents were mass spectrometry grade. The peptides were fed into the precolumn (3 μm, 20mm×75μm i.d.), and then through an analytical column (1.9μm, 150 mm × 75 μm id) at a flow rate of 300 nl / min, using a 60-min LC gradient from 7% to 30% mobile phase B. The mass spectrometer scan settings were as follows: cycle time was 1 second, primary scan range was 375-1800 m / z (resolution at 200 m / z was 60000), normalized AGC target was 300%, and maximum ion injection time was 50 ms. Ions with signal intensity lower than 2e4 were not collected. Secondary resolution was set to 30000, normalized AGC target was 200%, and maximum ion injection time was 86 ms. Turbo-TMT was enabled, the parent ion selection isolation window was 0.7 m / z, and the first mass of the scan was set to 110 m / z 32-34.

[0111] (3) Database and search parameters: Database information: Proteomics analysis used the Homo sapiens species database in the UniProt database, version "2022-10-09-decoys-reviewed-contam-UP000005640.fas" (downloaded from UniProt, containing 20432 reviewed proteins). The search was performed using FragPipe software (version 19.1): trypsin was specified as the enzyme, and 2 sites were allowed to be missed. The precursor mass tolerance and product mass tolerance were both set to 20 ppm, and the fragment mass tolerance was set to 0.02 Da. The FDR (false discovery rate) filter was set to 0.01, and MSFragger was used as the search engine. Peptide-protein uniqueness was defined as Unique+Razor. Median centering was used as the normalization method.

[0112] (4) Missing value interpolation: To solve the problem of missing values ​​of detected proteins, the seqKNN (sequential K nearest neighbor) method was used for interpolation. This method is based on a simple assumption: in a high-dimensional data set, each point (for example, the expression level of a specific protein) can be approximated by its nearest neighbor in the same space. In proteomics, this means that the missing value of a protein can be estimated by observing the expression levels of similar proteins in similar samples. The seqKNN method achieves missing value estimation and interpolation through the following steps:

[0113] (a) Similarity calculation: First, the similarity between each sample and other samples is calculated.

[0114] (b) Select K nearest neighbors: For each protein with missing values, select the K most similar (i.e., nearest neighbors) proteins.

[0115] (c) Missing value estimation: Missing values ​​are estimated by calculating the average of these nearest neighbors.

[0116] In this study, 102 samples were divided into 7 batches, each of which contained a mixed quality control (QC) sample. The mixed QC sample was prepared by mixing equal amounts of peptides from all samples and was aliquoted as quality control samples for TMT analysis in different batches. After obtaining the mass spectrometry quantitative results, the protein signal value of each sample was normalized by dividing the protein quantitative value of each sample by the corresponding protein signal value of the mixed QC sample in the same batch to obtain the corrected protein signal value.

[0117] Example 3 Data processing and typing analysis

[0118] In this example, the protein expression data obtained in Example 2 were clustered using the non-negative matrix factorization (NMF) algorithm to analyze the proteomic data. Through this algorithm, the protein expression profile was divided into different patterns, and finally three molecular subtypes of chordoma were identified: bone microenvironment-dominated type, mesenchymal type, and mesenchymal epithelial transition type.

[0119] The chordoma samples were clustered using the NMF algorithm based on the protein expression matrix. The number of clusters k was set from 2 to 10, the minimum number of members in each subclass was set to 10, and the seed value was set to 123 to ensure the stability of the results. According to the co-occurrence curve and dispersion, the optimal number of clusters was 3, and the samples were re-clustered to obtain 3 clusters (Cluster 1, Cluster 2, and Cluster 3) ( Figure 2 A). The three distinct proteomic subgroups were named cluster 1 (n=26), cluster 2 (n=35), and cluster 3 (n=41). When k=3, the consensus matrix heat map showed distinct and clear boundaries, indicating that the clustering was stable and robust ( Figure 2 B). To verify the cluster assignment, t-SNE analysis was performed and a general consistency was observed between the subtype division and the t-SNE distribution pattern ( Figure 2 C). The Venn diagram and ternary diagram are shown in Figure 2 D and 2E.

[0120] Example 4 CT / MRI evaluation of chordoma

[0121] (1) Bone destruction is assessed by radiologists and clinicians based on CT scan results. Although bone destruction is a common feature of chordomas, some chordomas do not destroy bone structures. Sclerosing chordoma is a rare type of chordoma, and benign chord cell tumors (BNCTs) also show plaque sclerosis between areas of preserved underlying trabecular bone on CT. Unlike BNCTs, sclerosing chordomas have relatively high Ki-67 expression and larger tumor volumes. In addition to the sclerosing chordomas and BNCTs mentioned above, there are also extraosseous chordomas. In addition, some bone-derived chordomas grow mainly into the spinal canal and paraspinal regions. For most recurrent cases, the bone structure has been removed in previous surgeries, and the chordomas have regrown in extraosseous areas. These tumors also belong to chordomas without bone destruction. For chordomas with bone destruction, the tumor tissue comes from bone destruction lesions within the vertebrae. For chordomas without bone destruction, this example selected samples from paraspinal soft tissue tumors.

[0122] (2) Classification of tumor signal intensity is based on the signal contrast between the tumor and surrounding tissue on the MRI image. Signal intensity is affected by many factors, including water molecule concentration, presence of fat molecules, protein content, and cellular structure. These descriptions are relative and are compared to the signal intensity of surrounding tissue or reference substances (such as water or fat).

[0123] (a) Low signal: On T1-weighted images, tumors appear darker than surrounding tissue and are considered low signal. On T2-weighted images, they may also appear low signal or slightly increased signal. This may be related to high cell density, low water content, scarce intercellular fat, or deposition of substances such as iron.

[0124] (b) Isointense: Tumors with similar signal intensities to surrounding tissues on T1- and T2-weighted images are classified as isointense. This may indicate that the tumor has similar water and fat content to the surrounding normal tissue.

[0125] (c) Hyperintensity: Tumors that show increased signal on T1- and / or T2-weighted images are considered hyperintense. This is usually associated with high water content, high vascularity, hemorrhage, or the presence of necrosis within the tumor.

[0126] Example 5 Multivariate Analysis

[0127] In order to explore the relationship between clinical characteristics and prognosis, a risk prediction model was constructed in this example. Lasso regression and Cox proportional hazard regression models were used to screen out clinical characteristics, including identified clusters, previous surgical treatments, pathological grading, metastasis status, T1-weighted MRI, and resection methods as risk factors. After eliminating the interference of confounding factors and excluding patients who underwent subtotal tumor resection and had distant metastases, 74 patients were finally included for analysis to explore the differences in clinical characteristics and prognosis among the three molecular subtypes. In this cohort, all deaths were caused by cancer, and significant prognostic differences were observed in the final analysis (OS / CSS: p=0.001, as Figure 3 The prognosis of cluster 1 was the most favorable (mean OS / CSS: 180.5 months, 95% CI: 158.7-202.2 months), cluster 3 had the shortest OS / CSS (mean OS / CSS: 115.4 months, 95% CI: 98.3-132.5 months), and cluster 2 had the shortest OS / CSS (mean OS / CSS: 58.5 months, 95% CI: 46.5-70.4 months). There were no significant differences among the three clusters in terms of clinical characteristics (patient gender, age, KFS score, distant metastasis), tumor characteristics (pathological grade, tumor size, Ki-67), and resection method. Figure 4 D-4K). Compared with clusters 1 and 2, more chordomas were found in cluster 3 (87.8%), while fewer mobile spine chordomas (12.2%) were found ( Figure 4 A). In cluster 1, primary chordomas (76.9%) and chordomas with low signal intensity on T1-weighted MRI (76.9%) were more common, but recurrent chordomas (23.1%) and isointense chordomas (7.7%) were less common ( Figure 4 B and 4C).

[0128] Example 6 Identification of differentially expressed proteins and functional enrichment analysis

[0129] Clinical features including previous surgical treatment, tumor location, CT imaging, MRI-T1 signal, pathological subtype and Ki67 were grouped. The R package "limma" was used to identify differentially expressed proteins (DEPs) between groups. The identification of DEPs was based on adjusted p-value < 0.05 and |Log2FC| ≥ 1. Subsequently, this example performed gene set enrichment analysis (GSEA) on DEPs using the R package "clusterProfiler". The selected gene sets were from the Molecular Signature Database (MSigDB, http: / / software.broadinstitute.org / gsea / msigdb / index.jsp), including REACTOME (c2.cp.reactome.v7.4.entrez.gmt), WP (c2.cp.wikipathways.v7.4.symbols.gmt), HALLMARK (h.all.v7.4.symbols.gmt) and KEGG (c2.cp.kegg.v7.4.symbols.gmt). In addition, GSVA analysis was performed, and the gene set was the biological process (BP) of GO analysis (c5.go.v7.4.symbols.gmt). Some key pathways with significant differences are shown, and the threshold was selected as p value < 0.05. The differentially expressed proteins of cluster 1 were enriched in characteristic markers of myeloid cells (including osteoclasts, osteoblasts, mast cells, neutrophils, NK cells, macrophages, T cells and B cells), while cluster 2 showed characteristics of mesenchymal cells and mitochondrial oxidative phosphorylation (OXPHOS). Cluster 3 was characterized by enrichment of epithelial features (mesenchymal-epithelial transition, epithelial cell proliferation), extracellular matrix (ECM) disintegration and response to tyrosine kinase (PDGF) ( Figure 5 B). Therefore, chordoma is divided into three molecular subtypes: bone microenvironment-dominant, mesenchymal, and mesenchymal-epithelial transition ( Figure 5 A).

[0130] Example 7 Cell line clustering and immune cell infiltration analysis and mitochondrial protein analysis

[0131] (1) Cell line clustering and tumor-infiltrating immune cells were calculated by CIBERSORTx (https: / / cibersortx.stanford.edu) and displayed as box plots. The clustering markers and immune cell markers used are detailed in the Supplementary Materials. For the ESTIMATE analysis, we used the R package “estimate” to calculate the scores using the estimateScore function with the “affmetrix” platform. Next, we calculated the immune score, stromal score, and tumor purity.

[0132] (2) 1136 human genes with high mitochondrial localization confidence were obtained through MitoCarta3.0 (https: / / www.broadinstitute.org / mitocarta / mitocarta30-inventory-mammalian-mitochondrial-proteins-and-pathways). The expression of the above proteins in the three clusters was calculated separately, and paired t-tests were performed. In addition, we selected proteins involved in five mitochondrial complexes (complexes I, II, III, IV, and V), detected their expression in the three clusters separately, and performed paired t-tests.

[0133] (3) The mesenchymal-epithelial transition type showed the highest tumor purity and the lowest stromal score, while the immune score of the bone microenvironment-dominated type was significantly higher than that of other subtypes ( Figure 6 A-6C). In the bone microenvironment-dominant type, overexpression of immune microenvironment markers (CD3G, CD8A, MZB1, CD14, CD16a, CD16b, CD47, CXCL13) and bone-specific microenvironment markers (SEMA7A, MMP9, OSCAR, CA2) and higher levels of immune cell (CD8+ T cells and monocytes) infiltration were identified ( Figure 6 D-6G). For mesenchymal chordomas, the expression of mitochondrial proteins was quantified and found to be high, including mitochondrial respiratory chain complex IV ( Figure 6 H-6M). For previously reported chordoma mutated genes, chordoma markers (TBXT, CDKN2A, LYST, and ACAN) were significantly enriched in the mesenchymal-epithelial transition subtype ( Figure 6 N), while chromatin modeling genes (PBRM1, SETD2, ARID1A) were highly expressed in the mesenchymal subtype, among which SETD2 expression was statistically significant ( Figure 6 O).

[0134] Example 8 Biological characteristics of each molecular subtype

[0135] The three identified molecular subtypes were subjected to detailed biological characterization, e.g. Figure 1 shown.

[0136] (1) Bone microenvironment-dominated type: This subtype exhibits a specific protein expression profile associated with bone destruction and displays characteristics of bone microenvironment cells including osteoclasts, osteoblasts, mast cells, neutrophils, NK cells, macrophages, T cells, and B cells.

[0137] (2) Mesenchymal type: This subtype exhibits the characteristics of mesenchymal cells and mitochondrial oxidative phosphorylation, and its characteristic proteins are closely related to energy metabolism.

[0138] (3) Mesenchymal-epithelial transition: This subtype shows characteristics including enrichment of epithelial features (mesenchymal-epithelial transition, epithelial cell proliferation), degradation of the extracellular matrix and response to tyrosine kinase (PDGF).

[0139] (4) The functional characteristics of chordoma subtypes were annotated by gene set variation analysis (GSVA) ​​of REACTOME enrichment analysis. The results showed that tumor microenvironment (TME) characteristics (chemokines and complement signals) were mainly enriched in the bone microenvironment-dominated type ( Figure 7 A-7B); mitochondrial oxidative phosphorylation (OXPHOS) signature (respiratory electron transport and complex I biogenesis) is highly expressed in mesenchymal chordomas ( Figure 7 C-7D); while RTK-related pathways (FGFR and PI3K-AKT) were enriched in mesenchymal epithelial transition ( Figure 7 E-7F). Therefore, the present application discloses the biological characteristics of three chordoma subtypes, which are respectively related to osteoclastogenesis and immunogenicity, OXPHOS and RTK activation.

[0140] Example 9 System Development and Verification

[0141] This embodiment develops a system for molecular typing of chordoma, and verifies the molecular typing method and system of chordoma. By analyzing independent sample sets, the accuracy and stability of the system are confirmed, ensuring its applicability in different clinical contexts.

[0142] Tumor tissue from another chordoma patient is selected, and in the case of unknown molecular subtypes of the chordoma patient, protein expression analysis is performed on the patient's protein sample using the method or system of the present application to determine which chordoma analysis subtype the patient belongs to. The accuracy, specificity, and stability of this method in determining the molecular subtype of chordoma is determined by comparing the results with the patient's clinical characteristics and the biological characteristics of the tumor.

[0143] When the embodiments give numerical ranges, it should be understood that, unless otherwise specified in the present invention, both endpoints of each numerical range and any numerical value between the two endpoints can be selected. Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those generally understood by those skilled in the art. In addition to the specific methods, equipment, and materials used in the embodiments, according to the grasp of the prior art by those skilled in the art and the record of the present invention, any methods, equipment, and materials of the prior art similar or equivalent to the methods, equipment, and materials described in the embodiments of the present invention can also be used to realize the present invention.

Claims

1. A method for molecular typing of chordoma, comprising the following steps: S1) performing protein expression analysis on the patient's protein sample to generate protein expression profile data; S2) performing cluster analysis based on the protein expression profile data, analyzing the specific protein expression features obtained by the cluster analysis, and identifying the molecular subtypes of chordoma; S3) provides specific protein expression signatures for each molecular subtype.

2. The method for molecular typing of chordoma according to claim 1, characterized in that: The protein expression analysis adopts tandem mass spectrometry labeling technology; And / or, the cluster analysis adopts a non-negative matrix decomposition algorithm.

3. The method for molecular typing of chordoma according to claim 1, characterized in that: The molecular subtypes include any one of bone microenvironment dominant type, mesenchymal type and mesenchymal epithelial transition type; Preferably, when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of bone microenvironment cells, it is identified as the bone microenvironment-dominated type; when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of mesenchymal cells and mitochondrial oxidative phosphorylation, it is identified as the mesenchymal type; when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of mesenchymal epithelial transformation, epithelial cell proliferation, extracellular matrix decomposition, and tyrosine kinase reaction, it is identified as the mesenchymal epithelial transformation type.

4. A system or device for molecular typing of chordoma, comprising: A proteomics analysis module, used to perform protein expression analysis on protein samples of chordoma patients to obtain protein expression spectrum data; preferably, the protein expression analysis adopts tandem mass spectrometry labeling technology; A molecular subtyping module, used to perform cluster analysis based on the protein expression spectrum data, analyze the specific protein expression features obtained by the cluster analysis, and identify the molecular subtypes of chordoma; preferably, the cluster analysis adopts a non-negative matrix decomposition algorithm; The typing output module is used to output the molecular typing results and provide the specific protein expression characteristics of each molecular subtype.

5. The device according to claim 4, characterized in that In the molecular subtyping module, the molecular subtype includes any one of a bone microenvironment dominant type, a mesenchymal type, and a mesenchymal epithelial transition type; Preferably, when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of bone microenvironment cells, it is identified as the bone microenvironment-dominated type; when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of mesenchymal cells and mitochondrial oxidative phosphorylation, it is identified as the mesenchymal type; when the specific protein expression characteristics obtained by cluster analysis are the protein expression characteristics of mesenchymal epithelial transformation, epithelial cell proliferation, extracellular matrix decomposition, and tyrosine kinase reaction, it is identified as the mesenchymal epithelial transformation type.

6. A computer-readable storage medium comprising a computer program, characterized in that: When the computer program is executed by a processor, the chordoma molecular typing method according to any one of claims 1 to 3 is executed.

7. A terminal comprising a memory and a processor, wherein the memory comprises a computer program, characterized in that: When the computer program is executed by a processor, the chordoma molecular typing method according to any one of claims 1 to 3 is executed.

8. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method for molecular typing of chordoma according to any one of claims 1 to 3 is implemented.

9. A kit comprising reagents required for detecting any one or more of the bone microenvironment dominant type, mesenchymal type and mesenchymal epithelial transition type of a sample; preferably, comprising reagents required for tandem mass spectrometry labeling technology and / or reagents required for protein extraction.

10. Use of the system or device according to any one of claims 4 to 5, or the computer-readable storage medium according to claim 6, or the terminal according to claim 7, or the computer program product according to claim 8, or the kit according to claim 9, wherein the use comprises any one or more of the following: 1) Prepare chordoma molecular typing products; 2) Prepare a product for evaluating the prognosis of chordoma patients; 3) Prepare a product for evaluating the clinical characteristics of chordoma patients; 4) Prepare products for evaluating chordoma biological behavior; 5) Prepare companion diagnostic products for chordoma.

Images