Method and device for predicting osteosarcoma survival prognosis and electronic equipment

Abstract

The invention belongs to the technical field of bioinformatics, and discloses a method and device for predicting osteosarcoma survival prognosis and electronic equipment, which are used for predicting osteosarcoma survival prognosis by incorporating more cases and detailed clinical baseline data and combining clinical prognosis factors such as primary focus part factor values, tumor staging factor values, tumor diameter factor values and serum alkaline phosphatase factor values. And one or more of the screened super enhancer driving genes LACTB, CEP55, SRSF3, TCF7L2 and FOXP1 with better prediction efficiency are used as gene combinations to jointly construct a prognosis prediction model for survival risk assessment, so that prognosis prediction of a patient corresponding to a to-be-detected sample is realized, the prediction model is more reliable in efficiency, and the prediction efficiency is higher. The prediction accuracy of osteosarcoma survival prognosis can be obviously improved.

Description

Technical Field

[0001] This invention belongs to the field of bioinformatics technology, specifically relating to a method and device for predicting the survival prognosis of osteosarcoma, and electronic equipment. Background Technology

[0002] Osteosarcoma is a common primary malignant bone tumor in adolescents. It frequently occurs during the long bone development stage of adolescence, particularly in the metaphysis of the distal femur and proximal tibia, areas of rapid growth. Rapid tumor growth and lung metastasis contribute significantly to the high mortality rate and severity of osteosarcoma. Current treatment for newly diagnosed osteosarcoma patients primarily involves a comprehensive approach: neoadjuvant chemotherapy, surgical resection, and postoperative treatment. Postoperative treatment further eliminates residual cancer cells, helping to maintain disease-free survival and significantly impacting the prognosis. With advancements in biomedical technology and in-depth research into osteosarcoma, postoperative treatment options have become increasingly diverse, including adjuvant chemotherapy, radiotherapy, targeted therapy, and immunotherapy. However, due to significant differences in patient conditions, unavoidable treatment side effects, and high treatment costs, reliable prognostic assessment tools are urgently needed to help develop rational and effective individualized treatment plans for different patients.

[0003] However, the low incidence of osteosarcoma limits the scale of clinical research and makes specimen collection difficult. Most previous studies on predictive model construction relied on public databases, primarily GEO and TARGET, for their clinical and gene expression information. Limited by the limited number of cases in these databases, the lack of clinical information on patients, and the heterogeneity of data from different databases, previously reported predictive models considered few clinical factors. Furthermore, due to the high heterogeneity of osteosarcoma tumors and the complexity of clinical conditions, the existing clinical prognostic factors are still not ideal for predicting the survival rate of osteosarcoma patients, resulting in insufficient accuracy in predicting osteosarcoma survival prognosis. Summary of the Invention

[0004] The purpose of this invention is to provide a method, device, and electronic device for predicting the survival prognosis of osteosarcoma, which can improve the accuracy of predicting the survival prognosis of osteosarcoma.

[0005] The first aspect of this invention discloses a method for predicting the survival prognosis of osteosarcoma, comprising:

[0006] Obtain the primary lesion site factor value, tumor stage factor value, tumor diameter factor value, serum alkaline phosphatase factor value, and expression index value and weight coefficient of each marker gene of the sample to be tested;

[0007] Based on the expression index values ​​and weighting coefficients of each marker gene, the gene index values ​​of the sample to be tested are calculated using the following formula:

[0008]

[0009] Where SE represents the gene index value, n is the number of marker genes, and n is any positive integer from 1 to 5, w i E represents the weight coefficient of the i-th marker gene. i The expression index value represents the i-th marker gene. The n marker genes include one or more combinations of the following five super enhancer driver genes: LACTB, CEP55, SRSF3, TCF7L2, and FOXP1.

[0010] Based on the primary lesion site factor value, the tumor stage factor value, the tumor diameter factor value, the serum alkaline phosphatase factor value, and the gene index value, the predicted survival risk value for the patient corresponding to the test sample is calculated.

[0011] A second aspect of this invention discloses a device for predicting the survival prognosis of osteosarcoma, comprising:

[0012] The acquisition unit is used to acquire the primary lesion site factor value, tumor stage factor value, tumor diameter factor value, serum alkaline phosphatase factor value, and expression index value and weight coefficient of each marker gene of the sample to be tested.

[0013] The calculation unit is used to calculate the gene index value of the sample to be tested based on the expression index value and weight coefficient of each marker gene using the following formula:

[0014]

[0015] Where SE represents the gene index value, n is the number of marker genes, and n is any positive integer from 1 to 5, w i E represents the weight coefficient of the i-th marker gene. i This represents the expression index value of the i-th marker gene. The n marker genes include one or more combinations of the following five super-enhancer driver genes:

[0016] LACTB, CEP55, SRSF3, TCF7L2, FOXP1;

[0017] The prediction unit is used to calculate the predicted survival risk value of the patient corresponding to the test sample based on the primary lesion site factor value, the tumor stage factor value, the tumor diameter factor value, the serum alkaline phosphatase factor value, and the gene index value.

[0018] A third aspect of the present invention discloses an electronic device, including a memory storing executable program code and a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the osteosarcoma survival prognosis prediction method disclosed in the first aspect.

[0019] The beneficial effects of this invention are that by obtaining clinical prognostic factors such as primary lesion site factor value, tumor stage factor value, tumor diameter factor value, and serum alkaline phosphatase factor value of the sample to be tested, and by using one or more of the super enhancer driver genes LACTB, CEP55, SRSF3, TCF7L2, and FOXP1 with good predictive efficacy as gene combinations, a prognostic prediction model for survival risk assessment is jointly constructed to achieve prognostic prediction for the corresponding patient of the sample to be tested. This makes the predictive model of this invention more reliable and significantly improves the accuracy of predicting the survival prognosis of osteosarcoma compared with the prior art. This helps to screen out patients with poor prognosis in the complex osteosarcoma patient population, help to formulate reasonable and effective treatment plans, improve the prognosis of patients, and avoid overtreatment of patients with good prognosis, thereby alleviating the economic and medical resource pressure on patients and society. Attached Figure Description

[0020] The accompanying drawings illustrate specific examples of the technical solutions described in this invention and, together with the detailed embodiments, form part of the specification, serving to explain the technical solutions, principles, and effects of this invention.

[0021] Unless otherwise specified or defined, the same reference numerals in different figures represent the same or similar technical features, and different reference numerals may be used to represent the same or similar technical features.

[0022] Figure 1 This is a graph showing the selective inhibition of gene expression levels by the inhibitor THZ2.

[0023] Figure 2 This is a graph showing the selective inhibition of gene expression levels by the inhibitor THZ531.

[0024] Figure 3 This is a gene function ontology analysis diagram of super-enhancer-related genes suppressed by THZ2;

[0025] Figure 4 This is a gene function ontology analysis diagram of super-enhancer-related genes suppressed by THZ531;

[0026] Figure 5 These are 10 genes that were selected and are associated with typical super enhancers and are significantly downregulated by THZ2 and THZ531;

[0027] Figure 6 This is a graph showing the expression analysis of seven genes that are highly expressed in osteosarcoma;

[0028] Figure 7 This is a graph showing the correlation between gene expression levels and patient survival prognosis.

[0029] Figure 8 This is a graph showing the correlation between immunohistochemical scores and patient survival prognosis.

[0030] Figure 9 These are the participant operating curves for gene combinations targeting 3, 5, and 10-year survival rates;

[0031] Figure 10 It is a predictive model constructed by combining gene combinations and clinical prognostic factors, and the subject working curves for 3, 5 and 10-year survival rates.

[0032] Figure 11 It is a nomogram that visualizes the predicted survival probability of a prediction model;

[0033] Figure 12 It is the calibration curve between the predicted probability and the actual probability of the prediction model in the training set and the validation set;

[0034] Figure 13 This is a screenshot of the user interface of a web-based calculator that predicts the prognosis of osteosarcoma patients.

[0035] Figure 14 This is a flowchart of a method for predicting the survival prognosis of osteosarcoma.

[0036] Figure 15 This is a schematic diagram of a device for predicting the survival prognosis of osteosarcoma.

[0037] Figure 16 This is a schematic diagram of the structure of an electronic device.

[0038] Explanation of reference numerals in the attached figures:

[0039] 100. Acquisition unit; 200. Calculation unit; 300. Prediction unit; 400. Memory; 500. Processor. Detailed Implementation

[0040] To facilitate understanding of the present invention, specific embodiments of the present invention will be described in more detail below with reference to the accompanying drawings.

[0041] Unless otherwise specified or defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. When combined with the technical solutions of the invention in a real-world scenario, all technical and scientific terms used herein may also have meanings corresponding to the purpose of achieving the technical solutions of the invention. The terms "first," "second," etc., used herein are merely for distinguishing names and do not represent a specific number or order. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0042] Unless otherwise specified or defined, the terms “described” or “the” as used herein refer to the technical features or technical content mentioned or described above, which may be the same as or similar to the technical features or technical content mentioned above.

[0043] Undoubtedly, any technical content or feature that is contrary to or clearly contradicts the purpose of this invention should be excluded.

[0044] The embodiments of the present invention include the following stages: marker gene screening stage, construction and efficacy verification analysis of prediction model, and clinical prognosis prediction stage.

[0045] Phase 1: Marker Gene Screening Phase

[0046] During the gene screening phase, osteosarcoma cell samples U2-OS and SJSA-1 were treated for 6 hours with small molecule inhibitors THZ2 (25 nM, 100 nM, and 400 nM) and THZ531 (50 nM, 200 nM, and 800 nM), as well as a control solvent. RNA-seq transcriptome sequencing was then performed to detect changes in intracellular gene mRNA expression, obtaining mRNA data. GSEA analysis of the mRNA data revealed that the two inhibitors, THZ2 and THZ531, selectively inhibited gene expression levels driven by tumor-associated regulatory elements (superenhancers), such as... Figure 1 and 2 As shown.

[0047] Gene ontology (GO) analysis revealed that suppressed super-enhancer-related genes are involved in important processes such as osteosarcoma cell growth, metastasis, and angiogenesis. Figure 3 and 4 As shown.

[0048] Based on the decisive role and characteristics of super-enhancers in tumor malignancy, 10 genes associated with typical super-enhancers and significantly downregulated by THZ2 and THZ531 were screened, such as... Figure 5 The following are listed: LACTB, CEP55, SRSF3, TCF7L2, FOXP1, APOLD1, ERBB2IP, MICAL2, RRP1B, and DNAJB12.

[0049] A search of the public database CCLE revealed that, compared to 24 other tumor types, the super-enhancer driver genes LACTB, ERBB2IP, MICAL2, RRP1B, FOXP1, DNAJB12, and CEP55 were highly expressed in osteosarcoma, exhibiting species-specific characteristics of super-enhancer driver genes. Figure 6 As shown.

[0050] Analysis of data from 127 osteosarcoma patients in the public database GEO revealed a negative correlation between the expression levels of CEP55, SRSF3, TCF7L2, APOLD1, and RRP1B and patient survival prognosis. Figure 7 As shown.

[0051] Immunohistochemical staining was performed on surgical specimens from 70 osteosarcoma patients using antibodies corresponding to the above 10 genes. Protein expression levels of these 10 genes were assessed based on staining patterns. KM analysis, combined with patient follow-up information, showed that immunohistochemical scores for LACTB, CEP55, SRSF3, TCF7L2, and FOXP1 were negatively correlated with patient survival prognosis. Figure 8 As shown.

[0052] In summary, ten typical super-enhancer driver genes in osteosarcoma cells were identified through multi-omics combined with high-throughput screening. Bioinformatics analysis revealed the specific expression of these ten super-enhancer driver genes (LACTB, CEP55, SRSF3, TCF7L2, FOXP1, APOLD1, ERBB2IP, MICAL2, RRP1B, and DNAJB12) in osteosarcoma cells, and demonstrated their relationship with patient survival. Analysis of samples from 70 osteosarcoma patients validated the prognostic value of these ten genes in osteosarcoma, as well as their potential predictive and therapeutic value in osteosarcoma.

[0053] Phase Two: Construction and Efficacy Validation Analysis of the Predictive Model

[0054] The study included 212 osteosarcoma patients who underwent surgery at the First Affiliated Hospital of Sun Yat-sen University between May 2003 and December 2018. Detailed clinical information was collected, and the patients were randomly assigned by computer to a training set (159 cases) and a validation set (53 cases) at a ratio of 3:1, as shown in Table 1 below.

[0055] Table 1 Clinical Information Table for Training and Validation Sets

[0056]

[0057]

[0058] Univariate analysis of the correlation between the training set and overall survival was performed using the Log-rank test. The results are shown in Table 2. The results indicated that Enneking stage, tumor size, surgical type, serum alkaline phosphatase level, and immunohistochemical scores of five super-enhancer driver genes (LACTB, CEP55, SRSF3, TCF7L2, and FOXP1) were risk factors for survival in osteosarcoma patients. Therefore, these five super-enhancer driver genes (LACTB, CEP55, SRSF3, TCF7L2, and FOXP1) were used as target marker genes to establish a predictive model.

[0059] Table 2 shows the results of the univariate analysis of the correlation between the training set and overall survival.

[0060]

[0061]

[0062] Five significantly correlated super-enhancer driver genes were combined into a gene signature (SE-derived OS-signature, SE) using risk coefficient weighting. Other clinical prognostic factors (such as tumor location) were also included in subsequent analyses. Multivariate Cox analysis was used to determine whether variables associated with patient survival were independent prognostic factors. Five variables (Enneking stage, primary tumor location, tumor size, serum alkaline phosphatase level, and the gene signature SE) were identified as independently associated with patient survival, and the specific results are shown in Table 3.

[0063] Table 3 shows the results of multivariate Cox analysis of the correlation between the training set and overall survival.

[0064]

[0065]

[0066] The receiver operating characteristic (ROC) curve is used to analyze and compare the predictive accuracy of clinical prognostic factors and gene combinations in the model. A larger area under the ROC curve (AUC) indicates more accurate predictions. The analysis results of the gene combinations (LACTB, CEP55, SRSF3, TCF7L2, FOXP1) constructed in this invention are shown below. Figure 9 As shown, the gene combination has good accuracy in predicting the prognosis of osteosarcoma patients, and the predictive accuracy of the gene combination is significantly higher than that of a single gene.

[0067] At the same time, such as Figure 10As shown, the accuracy of the SE-derived OS-signature is higher than that of traditional clinical prognostic factors. The predictive model constructed by combining gene combinations and clinical prognostic factors has good accuracy in predicting the 3-year, 5-year, and 10-year overall survival of patients.

[0068] The established prediction model can be visualized using the R language tool, constructing a nodal plot of predicted survival probabilities, such as... Figure 11 As shown.

[0069] Calibration Validation: Using the bootstrap method, with 1000 repeated samplings, the calibration curve between the predicted probability and the actual probability obtained from the nomogram model for predicting the survival prognosis of osteosarcoma patients established in this invention was validated. The results are as follows: Figure 12 As shown. By Figure 12 It can be seen that the predicted probabilities of the nomograph model constructed in this invention are highly consistent with the actual situation in both the training set and the validation set.

[0070] To facilitate clinical use, we have provided a web-based calculator using R language (version 4.0.3) to predict the prognosis of osteosarcoma patients. The user interface of this web-based calculator is as follows: Figure 13 As shown, users only need to select the corresponding variable parameter on the user operation interface to obtain the corresponding survival rate, which not only makes the operation more convenient for users, but also greatly improves the calculation speed.

[0071] In summary, this invention, through a combination of high-throughput drug screening and gene transcriptome analysis, has revealed the crucial role of super-enhancers in the malignancy of osteosarcoma. Prognostic analysis was conducted on 212 osteosarcoma patients with detailed clinical data, long-term follow-up information, and well-preserved tumor specimens. One or more super-enhancer driver genes from LACTB, CEP55, SRSF3, TCF7L2, and FOXP1, exhibiting good predictive efficacy, were selected as prognostic biomarkers for osteosarcoma patients. Based on these biomarkers, and combined with clinical prognostic factors such as primary tumor site, tumor stage, tumor diameter, and serum alkaline phosphatase levels, a prognostic prediction model for survival risk assessment was constructed. The accuracy and effectiveness of the model were validated using training and validation sets. Therefore, the predictive model of this invention is more reliable and can significantly improve the accuracy of predicting osteosarcoma survival prognosis. Compared with existing technologies, this invention can more accurately predict the overall survival rate of osteosarcoma patients at 3 years, 5 years, and up to 10 years. It helps to identify patients with poor prognosis in the complex osteosarcoma patient population, helps to develop reasonable and effective treatment plans, improve patients' prognosis, and avoid overtreatment of patients with good prognosis, thereby alleviating the economic and medical resource pressure on patients and society.

[0072] Phase 3: Clinical Prognosis Prediction Phase

[0073] like Figure 14 As shown, this invention discloses a method for predicting the survival prognosis of osteosarcoma, comprising the following steps S10 to S30:

[0074] S10. Obtain the primary lesion site factor value, tumor stage factor value, tumor diameter factor value, serum alkaline phosphatase factor value, and expression index value and weight coefficient of each marker gene of the sample to be tested.

[0075] Among them, the values ​​of the primary lesion site factor, tumor stage factor, tumor diameter factor, serum alkaline phosphatase factor, and expression index of each marker gene are all 0 or 1.

[0076] S20. Calculate the gene index values ​​of the sample to be tested based on the expression index values ​​and weight coefficients of each marker gene.

[0077] The gene index values ​​of the sample to be tested are calculated using the following formula (1):

[0078]

[0079] Where SE represents the gene index value, n is the number of marker genes, and n is any positive integer from 1 to 5, w i E represents the weight coefficient of the i-th marker gene. i The expression index value represents the i-th marker gene. The n marker genes include one or more combinations of the following five super enhancer driver genes: LACTB, CEP55, SRSF3, TCF7L2, and FOXP1. Preferably, the weight coefficients of the five super enhancer driver genes are 0.352, 0.195, 0.577, 0.546, and 0.591, respectively.

[0080] S30. Based on the primary lesion site factor value, tumor stage factor value, tumor diameter factor value, serum alkaline phosphatase factor value, and gene index value, calculate the predicted survival risk value for the patient corresponding to the test sample.

[0081] Specifically, the predicted survival risk value for the patient corresponding to the test sample is calculated using the following formula:

[0082] PI=1.02*f1+1.492*f2+0.493*f3+0.645*f4+1.853*SE;

[0083] Wherein, PI represents the predicted survival risk value, and f1, f2, f3, and f4 represent the primary lesion site factor value, tumor stage factor value, tumor diameter factor value, and serum alkaline phosphatase factor value, respectively.

[0084] As an optional implementation, after calculating the predicted survival risk value for the patient corresponding to the test sample, step S40 (not shown) can also be performed:

[0085] S40. Based on the predicted survival risk value, the total risk score of the patient corresponding to the test sample is calculated using the following formula: Point = (PI) * 100 / 1.853.

[0086] In the formula, Point represents the total risk score, PI represents the predicted survival risk value, and 100 / 1.853 represents the score corresponding to each unit of risk coefficient.

[0087] It should be noted that the marker genes used may differ in different embodiments, such as one or more combinations of the five super-enhancer driver genes. In this embodiment, five marker genes are preferably used, i.e., the number n equals 5, including LACTB, CEP55, SRSF3, TCF7L2, and FOXP1.

[0088] In step S10, obtaining the expression index values ​​and weight coefficients of each marker gene may specifically include the following steps S101 to S103:

[0089] S101. Obtain the expression values ​​of the proteins corresponding to each marker gene in the sample to be tested.

[0090] In this process, tumor specimens (i.e., samples to be tested) obtained from surgical resection of osteosarcoma patients can be embedded in paraffin to form tissue blocks, which are then made into multiple tissue sections. Immunohistochemical staining is performed on each tissue section using antibodies corresponding to LACTB, CEP55, SRSF3, TCF7L2, and FOXP1, and the immunohistochemical results of each gene protein in the tissue sections are evaluated and statistically analyzed.

[0091] The scoring criteria for staining intensity in immunohistochemical results are as follows:

[0092] The staining intensity rating is as follows: 0 = unstained; 1 = weakly stained; 2 = moderately stained; 3 = strongly stained.

[0093] The scoring criteria for the proportion of positive cells in the immunohistochemical results are as follows:

[0094] The positive staining percentage was scored as follows: 0 = no positive staining; 1 = 1-25% of cells were positively stained; 2 = 26-50% of cells were positively stained; 3 = 51-75% of cells were positively stained; 4 = 75-100% of cells were positively stained.

[0095] The expression value of the protein corresponding to each marker gene in the patient's tumor specimen = staining intensity score × positive proportion score, with values ​​ranging from 0 to 12.

[0096] Therefore, the expression values ​​of the corresponding proteins of each marker gene—LACTB, CEP55, SRSF3, TCF7L2, and FOXP1—in the test sample can be calculated. Higher expression values ​​indicate a poorer prognosis and survival rate for the patient. Next, the lower and upper limits of expression for each marker gene, which can be preset and stored by the developers, are obtained. Based on the expression values ​​obtained from immunohistochemistry and the preset threshold conditions, the patients corresponding to the test samples are divided into high-risk or low-risk groups, and the expression index values ​​of each marker gene in the test samples are determined. This is step S102.

[0097] S102. Obtain the lower and upper limits of expression for each marker gene. The upper limit of expression for each marker gene is greater than the lower limit.

[0098] Specifically, in this embodiment,

[0099] The lower and upper limits of expression for the marker gene LACTB are {2, 3}, respectively.

[0100] The lower and upper limits of expression for the marker gene CEP55 are {4, 5}, respectively.

[0101] The lower and upper limits of expression for the marker gene SRSF3 are {0, 1}, respectively.

[0102] The lower and upper limits of expression for the marker gene TCF7L2 are {9, 10}, respectively.

[0103] The lower and upper limits of expression for the marker gene FOXP1 are {4, 5}.

[0104] S103. Iterate through each marker gene and judge its expression value to determine the expression index value of each marker gene.

[0105] In step S103, if the expression value of the marker gene is less than or equal to its corresponding lower limit of expression, the survival prognosis of the patient corresponding to the test sample can be divided into a low-risk group, and thus the expression index value of the marker gene can be determined to be 0; if the expression value of the marker gene is greater than or equal to its corresponding upper limit of expression, the survival prognosis of the patient corresponding to the test sample can be divided into a high-risk group, and thus the expression index value of the marker gene can be determined to be 1.

[0106] Specifically, the expression index value E of each marker gene LACTB E CEP55 E SRSF3 E TCF7L2 E FOXP1 The possible values ​​are as follows:

[0107] E LACTBValues: When the expressed value is ≤2, the value is 0; when the expressed value is ≥3, the value is 1.

[0108] E CEP55 Values: When the expressed value is ≤4, the value is 0; when the expressed value is ≥5, the value is 1.

[0109] E SRSF3 Values: When the expressed value is ≤0, the value is 0; when the expressed value is ≥1, the value is 1.

[0110] E TCF7L2 Values: When the expressed value is ≤9, the value is 0; when the expressed value is ≥10, the value is 1.

[0111] E FOXP1 Values: When the expressed value is ≤4, the value is 0; when the expressed value is ≥5, the value is 1.

[0112] After determining the expression index values ​​of the five marker genes, the pre-stored weight coefficients of each marker gene can be retrieved, and then the gene index value (SE-derived OS-signature, SE) of the sample to be tested can be calculated using the following formula: SE = (0.352 * E) LACTB +0.195*E CEP55 +0.577*E SRSF3 +

[0113] 0.546*E TCF7L2 +0.591*E FOXP1 ) / 2.261.

[0114] In addition, the specific implementation method for obtaining the factor values ​​of various clinical prognostic factors of the sample to be tested in step S10, such as the factor value of primary lesion site, tumor stage, tumor diameter, and serum alkaline phosphatase, is as follows:

[0115] Primary tumor location factor value f1: Based on the primary tumor location parameter option selected by the user on the user operation interface, determine whether the primary tumor of the sample to be tested is located in the limb; if the primary tumor is located in the limb, determine the primary tumor location factor value to be 0; if the primary tumor is not located in the limb, determine the primary tumor location factor value to be 1.

[0116] Tumor staging factor value f2: Based on the lung examination data of the patient corresponding to the test sample, determine whether the lung examination data indicates the presence of lung metastases; if it indicates the presence of lung metastases, the tumor staging factor value is set to 0; if it does not indicate the presence of lung metastases, the tumor staging factor value is set to 1; where lung examination data can specifically be: the classification results obtained by identifying and classifying the lung imaging examination results (including ordinary X-ray, CT scan or MRI examination) of the patient corresponding to the test sample, or the tumor staging parameter options selected by the user on the user operation interface, such as Enneking staging data: I / II / III, generally I / II stage is considered to have no lung metastases, and III stage is considered to have lung metastases;

[0117] Tumor diameter factor value f3: The diameter of the primary tumor lesion is determined based on the tumor size data input by the user on the user interface, or based on the tumor size parameter selected by the user on the user interface, or based on the imaging examination results of the patient corresponding to the test sample. If the diameter of the primary tumor lesion is not greater than the diameter threshold, the tumor diameter factor value is determined to be 0; if the diameter of the primary tumor lesion is greater than the diameter threshold, the tumor diameter factor value is determined to be 1. The diameter threshold can be preset and stored, and its value is generally set to 8cm. Of course, in some special cases, it is not excluded that the diameter threshold value can be set to other values, and this invention does not limit this. Based on this, specifically, when the diameter of the primary tumor lesion is ≤8cm, the tumor diameter factor value is determined to be 0; when the diameter of the primary tumor lesion is >8cm, the tumor diameter factor value is determined to be 1.

[0118] Serum alkaline phosphatase factor value f4: Obtain the serum alkaline phosphatase test value and age information of the patient corresponding to the test sample; determine the patient's age information in conjunction with a preset age threshold, and determine the patient's serum alkaline phosphatase test value in conjunction with preset upper and lower detection limits. If the age information is less than the age threshold and the serum alkaline phosphatase test value is less than the upper detection limit, or if the age information reaches the age threshold and the serum alkaline phosphatase test value is less than the lower detection limit, the serum alkaline phosphatase factor value is determined to be 0; where the lower detection limit is less than the upper detection limit. Conversely, if the age information is less than the age threshold and the serum alkaline phosphatase test value reaches the upper detection limit, or if the age information reaches the age threshold and the serum alkaline phosphatase test value is greater than the lower detection limit, the serum alkaline phosphatase factor value is determined to be 1.

[0119] In this embodiment, the age threshold can be set to 18 years old, and the preset upper and lower detection limits can be set to 150 U / L and 110 U / L, respectively. With human development, in certain special circumstances, it is possible that the values ​​of this age threshold, upper and lower detection limits may be set to other values, and this invention does not limit this. Therefore, specifically, if the age information is less than 18 years old and the serum alkaline phosphatase detection value is less than 150 U / L, or if the age information is 18 years old and the serum alkaline phosphatase detection value is less than 110 U / L, the serum alkaline phosphatase factor value is determined to be 0; while if the age information is less than 18 years old and the serum alkaline phosphatase detection value is 150 U / L, or if the age information is 18 years old and the serum alkaline phosphatase detection value is greater than 110 U / L, the serum alkaline phosphatase factor value is determined to be 1.

[0120] To simplify calculations, in this embodiment, for each of the aforementioned clinical prognostic factors, such as primary tumor location, tumor stage, tumor diameter, and serum alkaline phosphatase levels, multiple parameter options are provided on the user interface (including options for whether the primary tumor site is located in the limb, tumor stage, tumor size range, serum alkaline phosphatase level, and age range). Before prediction, the user's selection of any parameter option on the user interface is detected through human-computer interaction to determine the values ​​of the aforementioned factors. This makes the operation more convenient for users and improves calculation speed.

[0121] After determining the values ​​of the aforementioned clinical prognostic factors and the SE value of the gene marker of the sample to be tested, the predicted survival risk value of the sample to be tested is further calculated using the following formula:

[0122] PI = 1.02*f1 + 1.492*f2 + 0.493*f3 + 0.645*f4 + 1.853*SE; where PI represents the predicted survival risk value, f1, f2, f3, and f4 represent the primary tumor site factor value, tumor stage factor value, tumor diameter factor value, and serum alkaline phosphatase factor value, respectively, and SE represents the gene index value.

[0123] After calculating the survival risk prediction value PI, the total risk score is calculated using the following formula (2):

[0124] Point = (PI) * 100 / 1.853 (2)

[0125] In the formula, Point represents the total risk score, PI represents the predicted survival risk, and 100 / 1.853 represents the score corresponding to each unit of risk coefficient, where the maximum risk coefficient of 1.853 is defined as 100 points. By substituting the predicted survival risk value PI into formula (2), the value of the total risk score Point can be calculated. In practical applications, after knowing the value of the total risk score Point, P, P3, P5, and P can be obtained accordingly. 10 The values ​​are more convenient.

[0126] Among them, P3, P5, P 10 , representing the survival probability values ​​of the patients corresponding to the test samples at 3, 5, and 10 years after surgery, respectively, and P representing the linear predictive value. The relationship between P and PI is: P = PI - k0; where k0 is the basic constant. In this embodiment of the invention, the basic constant is calculated based on research experience data, including the average covariance of primary lesion indicators, the average covariance of tumor stage indicators, the average covariance of tumor diameter indicators, the average covariance of serum alkaline phosphatase indicators, and the average covariance of SE.

[0127] For example, the average covariance of the primary lesion site index is 0.07, the average covariance of the tumor stage index is 0.108, the average covariance of the tumor diameter index is 0.611, the average covariance of the serum alkaline phosphatase index is 0.701, and the average covariance of SE is 0.524; then the basic constant k0 = 1.02*0.07 + 1.492*0.108 + 0.493*0.611 + 0.645*0.701 + 1.853*0.524 = 1.956.

[0128] It is understandable that the above-mentioned empirical data may vary at different times, which may lead to changes in the basic constants. This is a normal fluctuation, so the present invention does not impose specific limitations on the value of the basic constants.

[0129] Based on this, after calculating the linear prediction value P based on the total risk score Point, P3, P5, and P' are then calculated based on P. 10 The values ​​can be calculated using the following formulas:

[0130] P3 = S0(3) expP P5 = S0(5) expP ;P 10 =S0(10) expP ;

[0131] Where P represents the linear prediction value, exp is the calculation coefficient; S0(3), S0(5), and S0(10) represent the preset average survival probability of osteosarcoma patients in the 3rd, 5th, and 10th years after surgery, respectively, and are also empirical constants; for example, at present, S0(3) = 0.589, S0(5) = 0.499, and S0(10) = 0.398.

[0132] Optionally, after prediction, a nomogram of the survival probability of the patients corresponding to the tested sample can be output on the user interface, such as... Figure 11 As shown, the survival probability nomogram consists of ten line segments arranged from top to bottom and parallel to each other. Each line segment represents a scale with graduations.

[0133] The scale of the first ruler is 0 to 100, with the scale value 0 at the leftmost end and the scale value 100 at the rightmost end. The scale is divided into equal parts.

[0134] The second scale is the independent variable of the nomogram, the location of the primary lesion; the limbs are at the leftmost endpoint, and the non-limbs are at the rightmost endpoint; the score of the first scale corresponding to the leftmost endpoint is 0 points, and the score of the first scale corresponding to the rightmost endpoint is 55 points.

[0135] The third scale is the independent variable of the nomogram, Enneking period; period I / II is at the leftmost endpoint, and period III is at the rightmost endpoint; the first scale corresponding to the leftmost endpoint has a score of 0, and the first scale corresponding to the rightmost endpoint has a score of 81.

[0136] The fourth scale is the independent variable of the nomogram: tumor size; tumor diameter ≤8cm is at the leftmost endpoint, and tumor diameter >8cm is at the rightmost endpoint; the first scale corresponding to the leftmost endpoint has a score of 0, and the first scale corresponding to the rightmost endpoint has a score of 27.

[0137] The fifth scale is the independent variable of the nomogram: serum alkaline phosphatase (ALP) level; patients aged 18 and older with <110 U / L or under 18 with <150 U / L are at the leftmost endpoint, and patients aged 18 and older with ≥110 U / L or under 18 with ≥150 U / L are at the rightmost endpoint; the first scale corresponding to the leftmost endpoint has a score of 0, and the first scale corresponding to the rightmost endpoint has a score of 35.

[0138] The sixth scale represents the independent variable of the nomogram, the gene indicator SE-derived OS-signature; the scale values ​​range from 0 to 1, corresponding to the calculated values ​​of the indicator. A scale value of 0 is at the leftmost endpoint, and a scale value of 1 is at the rightmost endpoint; the first scale score corresponding to the leftmost endpoint is 0 points, and the first scale score corresponding to the rightmost endpoint is 100 points; the scale is divided into equal parts.

[0139] The 7th scale represents the total risk score Point, with a scale value of 0 to 280. The scale value of 0 is at the leftmost end, and the scale value of 280 is at the rightmost end. The scale is divided into equal parts.

[0140] The 8th scale represents the probability of 3-year survival for osteosarcoma patients, P3, with a scale value of 0.9 to 0.1. The scale value of 0.9 is at the leftmost end, and the scale value of 0.1 is at the rightmost end. The scale distribution on the scale is based on P3 = S0(3). expP The linear prediction value P is obtained by performing an exponential transformation.

[0141] The 9th scale represents the probability of 5-year survival for osteosarcoma patients, P5, with a scale value of 0.9 to 0.1. The scale value of 0.9 is at the leftmost end, and the scale value of 0.1 is at the rightmost end. The scale distribution on the scale is based on P5 = S0 (5). expP The linear prediction value P is obtained by performing an exponential transformation.

[0142] The 10th scale represents the probability of 10-year survival (P) for osteosarcoma patients after surgery. 10 The scale values ​​range from 0.85 to 0.1, with the 0.85 value at the leftmost endpoint and the 0.1 value at the rightmost endpoint. The scale distribution on the ruler is based on P. 10 =S0(10) expP The linear prediction value P is obtained by performing an exponential transformation.

[0143] In the survival probability nomogram, the primary tumor site, Enneking stage, tumor diameter, serum ALP level, and SE-derived OS-signature correspond to different risk score ranges, as detailed in Table 4 below.

[0144] Table 4. Risk scores of independent prognostic factors affecting postoperative survival in osteosarcoma patients.

[0145]

[0146] Draw a vertical line at the total risk score. The intersection of this line and the 3-year survival rate line represents the patient's 3-year survival probability. The same method can be used to obtain the patient's 5-year and 10-year survival rates. Different total risk scores correspond to different 3-year, 5-year, and 10-year survival rates.

[0147] For example, if a patient's primary tumor is located in the limbs (0 points), the Enneking stage is I / II (0 points), the tumor diameter is >8cm (27 points), the patient is under 18 years old and has serum ALP ≥150 (35 points), and the SE-derived OS-signature score is 0.45 (45 points), then the patient's total score is 107 points, and the corresponding 3-year, 5-year, and 10-year survival rates are 58%, 49%, and 39%, respectively.

[0148] like Figure 15 As shown in the figure, an embodiment of the present invention discloses a device for predicting the survival prognosis of osteosarcoma, including an acquisition unit 100, a calculation unit 200, and a prediction unit 300, wherein,

[0149] The acquisition unit 100 is used to acquire the primary lesion site factor value, tumor stage factor value, tumor diameter factor value, serum alkaline phosphatase factor value, and expression index value and weight coefficient of each marker gene of the sample to be tested.

[0150] The calculation unit 200 is used to calculate the gene index value of the sample to be tested based on the expression index value and weight coefficient of each marker gene using the following formula:

[0151]

[0152] Where SE represents the gene index value, n is the number of marker genes, and n is any positive integer from 1 to 5, w i E represents the weight coefficient of the i-th marker gene. i The expression index value represents the i-th marker gene. The n marker genes include one or more combinations of the following five super enhancer driver genes: LACTB, CEP55, SRSF3, TCF7L2, and FOXP1.

[0153] The prediction unit 300 is used to calculate the predicted survival risk value of the patient corresponding to the test sample based on the primary lesion site factor value, tumor stage factor value, tumor diameter factor value, serum alkaline phosphatase factor value, and gene index value.

[0154] Furthermore, the prediction unit 300 is specifically used to calculate the predicted survival risk value of the sample to be tested using the following formula:

[0155] PI=1.02*f1+1.492*f2+0.493*f3+0.645*f4+1.853*SE;

[0156] Wherein, PI represents the predicted survival risk value, and f1, f2, f3, and f4 represent the primary lesion site factor value, tumor stage factor value, tumor diameter factor value, and serum alkaline phosphatase factor value, respectively.

[0157] In this embodiment, the prediction unit 300 is also used to calculate the total risk score by the following formula after calculating the predicted survival risk value of the sample to be tested: Point=(PI)*100 / 1.853;

[0158] In the formula, Point represents the total risk score, PI represents the predicted survival risk value, and 100 / 1.853 represents the score corresponding to each unit of risk coefficient.

[0159] like Figure 16 As shown, an embodiment of the present invention discloses an electronic device, including a memory 400 storing executable program code and a processor 500 coupled to the memory 400;

[0160] The processor 500 calls the executable program code stored in the memory 400 to execute the osteosarcoma survival prognosis prediction method described in the above embodiments.

[0161] The purpose of the above embodiments is to reproduce and derive the technical solution of the present invention by way of example, and to fully describe the technical solution, purpose and effect of the present invention. The purpose is to enable the public to have a more thorough and comprehensive understanding of the disclosure of the present invention, and not to limit the scope of protection of the present invention.

[0162] The above embodiments are not an exhaustive list based on the present invention, and there may be many other embodiments not listed. Any substitutions and improvements made without departing from the concept of the present invention are within the protection scope of the present invention.

Claims

1. A method for predicting the prognosis of osteosarcoma survival, characterized by, The method comprises the following steps: Obtain the primary site factor value, tumor stage factor value, tumor diameter factor value, serum alkaline phosphatase factor value, and expression index value and weight coefficient of each marker gene of the to-be-tested sample; According to the expression index value and weight coefficient of each marker gene, the gene index value of the to-be-tested sample is calculated by the following formula: wherein SE represents the gene index value, n is the number of marker genes, n is any positive integer from 1 to 5, w i represents the weight coefficient of the i-th marker gene, E i represents the expression index value of the i-th marker gene, and the n marker genes include one or more combinations of the following 5 super-enhancer driver genes: LACTB, CEP55, SRSF3, TCF7L2, FOXP1. According to the primary site factor value, tumor stage factor value, tumor diameter factor value, serum alkaline phosphatase factor value, and gene index value, the survival risk prediction value of the patient corresponding to the to-be-tested sample is calculated.

2. The method for predicting the survival prognosis of osteosarcoma according to claim 1, wherein According to the primary site factor value, tumor stage factor value, tumor diameter factor value, serum alkaline phosphatase factor value, and gene index value, the survival risk prediction value of the patient corresponding to the to-be-tested sample is calculated, comprising: The survival risk prediction value of the patient corresponding to the to-be-tested sample is calculated by the following formula: PI = 1.02 * f1 + 1.492 * f2 + 0.493 * f3 + 0.645 * f4 + 1.853 * SE; Wherein, PI represents the survival risk prediction value, f1, f2, f3, f4 respectively represent the primary site factor value, tumor stage factor value, tumor diameter factor value and serum alkaline phosphatase factor value.

3. The method of predicting the survival prognosis of osteosarcoma according to claim 2, wherein After the survival risk prediction value of the patient corresponding to the to-be-tested sample is calculated, the method further comprises: The total risk score of the patient corresponding to the to-be-tested sample is calculated by the following formula: Point = (PI) * 100 / 1.853; In the formula, Point represents the total risk score, PI represents the survival risk prediction value, and 100 / 1.853 represents the score corresponding to each unit risk coefficient.

4. The method of predicting the survival prognosis of osteosarcoma according to claim 3, wherein After the total risk score of the patient corresponding to the to-be-tested sample is calculated, the method further comprises: According to the total risk score and the basic constant, a linear prediction value is calculated; According to the linear prediction value, the survival probability values of the patient corresponding to the to-be-tested sample after 3, 5 and 10 years of surgery are calculated respectively.

5. The osteosarcoma survival prognosis predicting method according to any one of claims 1 to 4, characterized in that, The weight coefficients of the 5 super-enhancer driver genes LACTB, CEP55, SRSF3, TCF7L2 and FOXP1 are 0.352, 0.195, 0.577, 0.546 and 0.591 respectively.

6. The method of predicting the survival prognosis of osteosarcoma according to claim 5, wherein Obtaining the expression index value of each marker gene of the to-be-tested sample comprises: Obtaining the expression value of each marker gene corresponding protein in the to-be-tested sample; Obtaining the expression lower limit value and the expression upper limit value of each marker gene; wherein the expression upper limit value of each marker gene is greater than the expression lower limit value; Iterating each marker gene to judge its expression value to determine the expression index value of each marker gene, wherein if the expression value of the marker gene is less than or equal to the corresponding expression lower limit value, the expression index value of the marker gene is determined as 0; if the expression value of the marker gene is greater than or equal to the corresponding expression upper limit value, the expression index value of the marker gene is determined as 1.

7. The osteosarcoma survival prognosis predicting method according to any one of claims 1 to 4, characterized by, Obtaining the primary site factor value of the to-be-tested sample comprises: According to the tumor primary site parameter option selected by the user on the user operation interface, it is judged whether the tumor primary site of the to-be-tested sample is located in the limbs; If the tumor primary site is located in a limb, the primary site position factor value is determined as 0; If the tumor primary site is not located in a limb, the primary site position factor value is determined as 1.

8. The osteosarcoma survival prognosis predicting method according to any one of claims 1 to 4, characterized by, The serum alkaline phosphatase factor value of the to-be-tested sample is obtained, including: The serum alkaline phosphatase detection value and age information of the patient corresponding to the to-be-tested sample are obtained; If the age information is less than the age threshold value and the serum alkaline phosphatase detection value is less than the detection upper limit value, or the age information reaches the age threshold value and the serum alkaline phosphatase detection value is less than the detection lower limit value, the serum alkaline phosphatase factor value is determined as 0; wherein the detection lower limit value is less than the detection upper limit value; If the age information is less than the age threshold value and the serum alkaline phosphatase detection value reaches the detection upper limit value, or the age information reaches the age threshold value and the serum alkaline phosphatase detection value is greater than the detection lower limit value, the serum alkaline phosphatase factor value is determined as 1.

9. An osteosarcoma survival prognosis predicting device characterized by comprising: The method comprises: An acquisition unit is configured to acquire a primary site position factor value, a tumor staging factor value, a tumor diameter factor value, a serum alkaline phosphatase factor value, and an expression index value and a weight coefficient of each marker gene of a to-be-tested sample; A calculation unit is configured to calculate a gene index value of the to-be-tested sample according to the expression index value and the weight coefficient of each marker gene by the following formula: wherein SE represents the gene index value, n is the number of marker genes, n is any positive integer from 1 to 5, w i represents the weight coefficient of the i-th marker gene, E i represents the expression index value of the i-th marker gene, and the n marker genes include one or more combinations of the following 5 super-enhancer driver genes: LACTB, CEP55, SRSF3, TCF7L2, FOXP1. A prediction unit is configured to calculate a survival risk prediction value of a patient corresponding to the to-be-tested sample according to the primary site position factor value, the tumor staging factor value, the tumor diameter factor value, the serum alkaline phosphatase factor value, and the gene index value.

10. An electronic device, comprising: The method comprises a memory storing executable program codes and a processor coupled with the memory; the processor invokes the executable program codes stored in the memory to execute the osteosarcoma survival prognosis prediction method of any one of claims 1 to 8.

Images