Prediction Device, System and Application for the Risk of CAP in ITP Patients

Abstract

The present invention discloses a prediction device, system and application for the risk of CAP in ITP patients. The risk prediction device for CAP in ITP patients includes the following processing modules: a data input module, a data recording module, a data assignment module, a data calculation module, a data grouping module and a conclusion output module. The risk prediction system for CAP in ITP patients includes a risk prediction device for CAP in ITP patients, a platelet count device and a lymphocyte count measurement device. By using the risk prediction device or system for CAP in ITP patients of the present invention, high-risk patients can be identified early and timely treatment can be provided only through medical history inquiry and complete blood cell examination.

Description

Technical Field

[0001] The present invention relates to a prediction device, system and application for the risk of CAP in ITP patients. Background Art

[0002] Idiopathic thrombocytopenia (ITP) is an autoimmune hemorrhagic disease characterized by antibody-mediated platelet destruction and decreased platelet production due to impaired platelet production. Glucocorticoids are the first-line treatment for ITP, and when patients are in an emergency or intolerant to hormones, intravenous immunoglobulin (IVIg) is considered. Rituximab, thrombopoietin (TPO) receptor agonists, immunosuppressants and splenectomy are the second-line treatments for ITP. It is reported that the overall response rate of initial treatment in ITP patients can reach 90%, however, many patients will relapse once they stop using hormones or IVIg, which means that ITP patients need chronic or continuous treatment. Due to the immune disorder of the ITP disease itself and the immunosuppression caused by long-term treatment, some studies have found that the probability of infection in ITP patients is higher than that in normal people, and some studies have explored some risk factors for infection in adult ITP patients, but no consensus has been reached yet.

[0003] Infection is one of the main causes of death in ITP patients, and the mortality related to infection in ITP patients has increased in recent years. According to the existing literature reports, platelets will be activated and consumed during the infection event, so the poor prognosis of ITP patients with infection may be related to the further aggravation of thrombocytopenia in these patients. The lungs are the most common site of infection in ITP patients. Although there is currently no data on the mortality of ITP patients with pneumonia, by comparing with the high mortality of community-acquired pneumonia (CAP) in the general population, the short-term mortality of CAP in the general population can reach 14%-32%, it can be inferred that the short-term prognosis of ITP patients after pneumonia is also not optimistic. Therefore, it is urgent to identify ITP patients at high risk of pneumonia early and treat them in time. However, there is currently no report on the risk factors for pneumonia in ITP patients, and there is no recognized prediction model for the risk of CAP in ITP patients. Summary of the Invention

[0004] The first object of the present invention is to provide a prediction device for the risk of CAP in ITP patients.

[0005] The prediction device for the risk of CAP in ITP patients provided by the present invention includes the following processing modules:

[0006] (1) Data input module: This module is used to input the age value, concomitant disease status, platelet count and lymphocyte count of the tester at the time of ITP diagnosis;

[0007] The concomitant disease status refers to whether the tester has the following diseases: myocardial infarction, congestive heart failure, peripheral vascular wall, cerebrovascular disease, chronic pulmonary disease, ulcer disease, diabetes, hemiplegia, moderate to severe kidney disease, and diabetes with end-organ failure;

[0008] (2) Data recording module: This module is used to receive and store the age value, concomitant disease status, platelet count, and lymphocyte count of the tester when diagnosing ITP output from the data input module;

[0009] (3) Data assignment module: This data assignment module consists of an age data assignment module, a concomitant disease data assignment module, a platelet data assignment module, and a lymphocyte data assignment module:

[0010] The age data assignment module is used to retrieve the age value of the tester stored in the data recording module, assign a value to the age value, and output f(age): when "the age of the tester ≥ 60 years old", f(age) is 1 point; when "the age of the tester < 60 years old", f(age) is 0 point;

[0011] The concomitant disease data assignment module is used to retrieve the concomitant disease status of the tester stored in the data recording module, calculate the Charlson score of the tester, assign a value to the Charlson score, and output f(Charlson score): when "the Charlson score of the tester ≥ 3 points", then f(Charlson score) is 3.5 points; when "the Charlson score of the tester < 3 points", then f(Charlson score) is 0 points; The Charlson score = Charlson-1 score + Charlson-2 score: If the tester has n (n is 0 or 1 or 2 or 3 or 4 or 5 or 6 or 7) of the following diseases: myocardial infarction, congestive heart failure, peripheral vascular wall, cerebrovascular disease, chronic pulmonary disease, ulcer disease, and diabetes, then the Charlson-1 score of the tester is n points; If the tester has m (m is 0 or 1 or 2 or 3) of the following diseases: hemiplegia, moderate to severe kidney disease, and diabetes with end-organ failure, then the Charlson-2 score of the tester is 2m points;

[0012] The platelet data assignment module is used to retrieve the platelet count of the tester stored in the data recording module, assign a value to the platelet count, and output f(platelet): when "the platelet count of the tester < 20000 / uL", f(platelet) is 2 points; when "the platelet count of the tester ≥ 20000 / uL", f(platelet) is 0 points;

[0013] The lymphocyte data assignment module is used to retrieve the number of lymphocytes of the tester stored in the data recording module, assign a value to the number of lymphocytes, and output f(lymphocytes): when "the number of tester lymphocytes < 1000 / uL", f(lymphocytes) is 3.5 points; when "the number of tester lymphocytes ≥ 1000 / uL", f(lymphocytes) is 0 points;

[0014] (4) Data calculation module: This module is used to receive f(age) output from the age data assignment module, f(Charlson score) output from the comorbid disease data assignment module, f(platelets) output from the platelet data assignment module, and f(lymphocytes) output from the lymphocyte data assignment module, and then calculate the ACPA score of the tester according to Equation Ⅰ;

[0015] ACPA score = f(age) + f(Charlson score) + f(platelets) + f(lymphocytes) Equation Ⅰ;

[0016] The ACPA score represents the CAP risk of ITP patients;

[0017] (5) Data grouping module: This module is used to receive the ACPA score of the tester output from the data calculation module, and then group the tester according to the ACPA score and output the risk grouping result;

[0018] The criteria for grouping the tester according to the ACPA score are as follows: testers with an ACPA score of 9 - 10 are in the high-risk group, testers with an ACPA score of 4.5 - 8 are in the medium-risk group, and testers with an ACPA score of 0 - 3.5 are in the low-risk group;

[0019] (6) Conclusion output module: This module is used to receive the risk grouping result output from the data grouping module, and output a conclusion according to the risk grouping result: testers in the high-risk group are ITP patients with a high risk of developing CAP; testers in the medium-risk group are ITP patients with a medium risk of developing CAP; testers in the low-risk group are ITP patients with a low risk of developing CAP.

[0020] In the above prediction device, the platelet count is the platelet count in whole blood, and its unit is / uL.

[0021] In the above prediction device, the number of lymphocytes is the number of lymphocytes in whole blood, and its unit is / uL.

[0022] In the above prediction device, whether the tester has the following diseases: myocardial infarction, congestive heart failure, peripheral vascular wall, cerebrovascular disease, chronic pulmonary disease, ulcer disease, diabetes, hemiplegia, moderate to severe kidney disease, and diabetes accompanied by end-stage organ failure can all be judged according to the relevant diagnostic criteria in the art.

[0023] The second object of the present invention is to provide a method for predicting the risk of CAP in ITP patients.

[0024] The method for predicting the risk of CAP in ITP patients provided by the present invention includes the following steps:

[0025] 1) Obtain the age value, concomitant disease status, platelet count, and lymphocyte count of the tester when diagnosing ITP.

[0026] 2) Assign values according to the following criteria based on the data obtained in step 1) to obtain f(age), f(Charlson score), f(platelets), and f(lymphocytes) of the tester:

[0027] When "the age of the tester ≥ 60 years old", f(age) is 1 point; when "the age of the tester < 60 years old", f(age) is 0 point;

[0028] When "the Charlson score of the tester ≥ 3 points", then f(Charlson score) is 3.5 points; when "the Charlson score of the tester < 3 points", then f(Charlson score) is 0 point; the Charlson score = Charlson-1 score + Charlson-2 score: If the tester has n of the following diseases: myocardial infarction, congestive heart failure, peripheral vascular wall, cerebrovascular disease, chronic pulmonary disease, ulcer disease, and diabetes, then the Charlson-1 score of the tester is n points; if the tester has m of the following diseases: hemiplegia, moderate to severe kidney disease, and diabetes accompanied by end-stage organ failure, then the Charlson-2 score of the tester is 2m points;

[0029] When "the platelet count of the tester < 20000 / uL", f(platelets) is 2 points; when "the platelet count of the tester ≥ 20000 / uL", f(platelets) is 0 point;

[0030] When "the lymphocyte count of the tester < 1000 / uL", f(lymphocytes) is 3.5 points; when "the lymphocyte count of the tester ≥ 1000 / uL", f(lymphocytes) is 0 point;

[0031] 3) Calculate the ACPA score of the tester according to the tester's f(age), f(Charlson score), f(platelets), and f(lymphocytes) according to Formula I; ACPA score = f(age) + f(Charlson score) + f(platelets) + f(lymphocytes) Formula I;

[0032] 4) Perform risk grouping on the tester according to the tester's ACPA score: testers with an ACPA score of 9 - 10 are in the high-risk group, testers with an ACPA score of 4.5 - 8 are in the medium-risk group, and testers with an ACPA score of 0 - 3.5 are in the low-risk group.

[0033] Further, the method further includes the following steps: 5) Predict the risk of developing CAP according to the risk grouping of the tester: that is, testers in the high-risk group are ITP patients with a high risk of developing CAP; testers in the medium-risk group are ITP patients with a medium risk of developing CAP; testers in the low-risk group are ITP patients with a low risk of developing CAP.

[0034] The third object of the present invention is to provide a prediction system for the risk of developing CAP in ITP patients.

[0035] The prediction system for the risk of developing CAP in ITP patients provided by the present invention includes the above-mentioned prediction device for the risk of developing CAP in ITP patients, a platelet count device, a lymphocyte count measurement device, a data input device, and a data output device.

[0036] In the above-mentioned prediction system for the risk of developing CAP in ITP patients, the platelet count device is a platelet count device in whole blood; the platelet count device in whole blood includes reagents and / or instruments for platelet counting. The reagents and / or instruments for platelet counting can be conventional reagents and / or instruments for platelet counting in the prior art. In a specific embodiment of the present invention, the reagents and / or instruments for platelet counting are an automatic hematology analyzer (Sysmex Corporation, model: sysmex XN 9000).

[0037] In the above-mentioned prediction system for the risk of developing CAP in ITP patients, the lymphocyte count measurement device is a lymphocyte count measurement device in whole blood; the lymphocyte count measurement device in whole blood includes reagents and / or instruments for measuring lymphocyte counts. The reagents and / or instruments for measuring lymphocyte counts can be conventional reagents and / or instruments for measuring lymphocyte counts in the prior art. In a specific embodiment of the present invention, the reagents and / or instruments for measuring lymphocyte counts are an automatic hematology analyzer (Sysmex Corporation, model: sysmex XN 9000).

[0038] In the above prediction system for the risk of CAP in ITP patients, the data input device can be any input device known to those skilled in the art in the prior art; specifically, it may include devices for inputting data such as keyboards, mice, joysticks, light pens, touch screens, cameras, scanners, fax machines, etc.

[0039] In the above prediction system for the risk of CAP in ITP patients, the data output device can be any input device known to those skilled in the art in the prior art; specifically, it may include devices for outputting data such as monitors, printers, speakers, plotters, image output systems, voice output systems, magnetic recording devices, etc.

[0040] The fourth object of the present invention is to provide a method for using the above prediction system for the risk of CAP in ITP patients.

[0041] The method for using the above prediction system for the risk of CAP in ITP patients provided by the present invention includes the following steps: obtaining the following data of the test subject when diagnosing ITP: age value, concomitant disease status, platelet count, and lymphocyte count; using the above prediction device for the risk of CAP in ITP patients to judge the risk of CAP in the test subject according to the above prediction method for the risk of CAP in ITP patients based on the data.

[0042] The application of the above prediction device for the risk of CAP in ITP patients or the above prediction system for the risk of CAP in ITP patients in the preparation of products for predicting or assisting in predicting the risk of CAP in ITP patients also belongs to the protection scope of the present invention.

[0043] The application of the above prediction device for the risk of CAP in ITP patients, the above platelet counting device, and the above lymphocyte measuring device in the preparation of products for predicting or assisting in predicting the risk of CAP in ITP patients also belongs to the protection scope of the present invention.

[0044] When using the device or system of the present invention to predict the risk of CAP in ITP patients, by obtaining the age and comorbid conditions recorded at the time of ITP diagnosis or reviewed at the time of follow-up for ITP patients, measuring or reviewing the platelet count and lymphocyte count at the time of initial diagnosis of ITP, and using the above-mentioned prediction device for the risk of CAP in ITP patients to judge the risk of CAP in ITP patients, the greater the ACPA score of the ITP patient, the greater the risk of CAP in the ITP patient. Among them, the testers with an ACPA score of 0 - 3.5 are in the low-risk group, the testers with an ACPA score of 4.5 - 8 are in the medium-risk group, and the testers with an ACPA score of 9 - 10 are in the high-risk group: the ITP patients in the high-risk group are ITP patients with a high risk of developing CAP, the ITP patients in the medium-risk group are ITP patients with a medium risk of developing CAP, and the ITP patients in the low-risk group are ITP patients with a low risk of developing CAP.

[0045] In the above text, the age, comorbid conditions, platelet count, and lymphocyte count are all data at the time of initial diagnosis of ITP.

[0046] In the above text, the lymphocyte count is the absolute lymphocyte count.

[0047] In the above text, the testers are all ITP patients, and the ITP patients are adult ITP patients aged 18 years or older.

[0048] In the above text, the CAP is community-acquired pneumonia. The ITP is idiopathic thrombocytopenia.

[0049] The present invention has developed and validated an ACPA score model for predicting the risk of CAP in ITP patients, and it is possible to early identify patients at high risk of developing CAP only through medical history inquiry and complete blood cell examination, so as to provide treatment in a timely manner. Brief Description of the Drawings

[0050] Figure 1 It is a flow chart for patient enrollment, model development, and validation.

[0051] Figure 2 It is the prediction accuracy of the ACPA model. (A) The ROC curve for predicting CAP in ITP patients in the development cohort, with an AUC of 0.853. (B) The ROC curve for predicting CAP in ITP patients in the validation cohort, with an AUC of 0.862.

[0052] Figure 3Calibration curves of the ACPA model for predicting the risk of CAP in ITP patients. (A) The ACPA model in the development cohort. (B) The ACPA model in the validation cohort. The x-axis represents the probability of CAP in ITP patients predicted by the model; the y-axis represents the actual probability of CAP in ITP patients. The ideal calibration plot is represented by a 45° diagonal line. Among them, the solid line represents the actual calibration plot, and the dashed line represents the ideal calibration plot.

[0053] Figure 4 Decision curve analysis of the ACPA model for predicting the risk of CAP in ITP patients in the development cohort. Black line (no benefit): Assume that no ITP patients develop CAP. Gray line (all benefit): Assume that all ITP patients develop CAP. These two lines are used as references. Detailed implementation manner

[0054] The following examples facilitate a better understanding of the present invention, but do not limit the present invention. The experimental methods in the following examples are all conventional methods unless otherwise specified. The test materials used in the following examples are all obtained from regular biochemical reagent stores unless otherwise specified. In the following examples, quantitative tests are all set with three repeated experiments, and the results are averaged.

[0055] Example 1. Development and validation of the ACPA score model for predicting the risk of CAP in ITP patients

[0056] The present invention developed and validated an ACPA score model for predicting the risk of CAP in ITP patients, and the process is as Figure 1 shown. The specific steps are as follows:

[0057] I. Patient samples and research methods for developing the ACPA score model for predicting the risk of CAP in ITP patients

[0058] 1. Patient samples

[0059] The patient samples in this study included a multicenter retrospective cohort study of 10 large clinical centers in China from 2002 to 2019 ( Figure 1) The target population was adult patients with primary ITP without splenectomy (aged 10 years or older). Patients diagnosed with connective tissue diseases, tumors (solid tumors or leukemia), primary immunodeficiency, liver insufficiency, and other diseases that can cause secondary thrombocytopenia were excluded from the study group. In addition, patients with a history of infection diagnosis before the diagnosis of ITP were also excluded. Finally, a total of 2,094 adult inpatients with ITP without splenectomy from 10 large clinical centers in China were enrolled. The average age of the patients was 52.4 years (age range 18 - 96 years), and the male-to-female ratio was approximately 1.7:1. Among the patient population, 767 patients (36.6%) were newly diagnosed with ITP, and 1,327 patients (63.4%) had persistent or chronic ITP. Among them, 208 inpatients with ITP were diagnosed with CAP. The 10 clinical centers included in this study covered the geographical regions of northern, southeastern, and southwestern China, and the medical records were extracted from the clinical databases of hospital inpatients. If a patient had more than one infection event, the medical record of the first occurrence of the infection was used; if the control group had more than one hospitalization record, the medical record information of the first hospitalization was used for analysis. The patient information collected in this study included the patient's demographic characteristics such as age, gender, concomitant diseases, previous bleeding events, infection-related records during hospitalization, laboratory test results at the time of ITP diagnosis, length of hospitalization, and treatment regimens before hospitalization. This study met the ethical review criteria of the ethics committees of the ten clinical centers.

[0060] To externally validate the model, all patients were divided into a development cohort and a validation cohort. The development cohort consisted of 145 inpatients with CAP and 1,360 inpatients without CAP from five clinical centers (People's Hospital of Peking University, Qilu Hospital of Shandong University, The Second Affiliated Hospital of Shanxi Medical University, The Second Affiliated Hospital of Kunming Medical University, Heping Hospital Affiliated to Changzhi Medical College). The validation cohort consisted of 63 inpatients with CAP and 526 inpatients without CAP from another five clinical centers (Shanxi Academy of Medical Sciences, Beijing Hospital, The Sixth Medical Center of Chinese PLA General Hospital, Peking University Shenzhen Hospital, Peking University First Hospital). The clinical characteristics of the patient population in this study are shown in Table 1. The age, concomitant disease status, and absolute lymphocyte count at the time of ITP diagnosis were comparable between the development cohort and the validation cohort (P > 0.05), while the proportion of patients with persistent or chronic ITP in the development cohort was higher (P < 0.001), and the male-to-female ratio was also higher (P = 0.014). In addition, the initial platelet count at the time of ITP diagnosis in the development cohort was also significantly higher than that in the validation cohort (P < 0.001).

[0061] Table 1. Baseline characteristics of the patients

[0062]

[0063] Note: # There were no initial platelet count records for 15 patients (3 in the development cohort and 12 in the validation cohort); * There were no initial absolute lymphocyte count records for 226 patients (148 in the development cohort and 78 in the validation cohort).

[0064] 2. Definitions and Diagnoses

[0065] All patients meeting the CAP diagnostic criteria included in the analysis were clinically diagnosed within 48 hours of hospitalization. The diagnostic criteria for CAP are as follows: (1) Imaging-confirmed progressive pulmonary infiltrative lesions that cannot be explained by other etiologies (tuberculosis, lung tumors, non-infectious interstitial lung diseases, pulmonary edema, atelectasis, pulmonary embolism, pulmonary eosinophilic infiltration, and pulmonary vasculitis); (2) New respiratory symptoms and signs, including cough, sputum production, dyspnea, abnormal breath sounds or rales, or new non-respiratory symptoms and signs, including altered mental status, exacerbation of underlying chronic diseases, or tachycardia, etc.; (3) Direct evidence of infection, including body temperature ≥ 38 °C, leukocytosis, or decreased oxygenation, etc. If the criteria described in (1) and (2) above are met simultaneously or the criteria described in (1) and (3) above are met simultaneously, the clinical diagnosis of CAP is established. All patients in this study were evaluated by respiratory specialists and hematologists based on the patients' clinical characteristics, laboratory tests, imaging studies, and etiological cultures, and the final clinical diagnosis was obtained by synthesizing the views of all experts. In addition, the Charlson score involved in this study was based on the concomitant diseases of the patients at the time of ITP diagnosis, and the Charlson score = Charlson-1 score + Charlson-2 score. Specifically: If the patient had n (where n is 0 or 1 or 2 or 3 or 4 or 5 or 6 or 7) of the following diseases at the time of ITP diagnosis: myocardial infarction, congestive heart failure, peripheral vascular wall, cerebrovascular disease, chronic lung disease, ulcer disease, and diabetes, then the patient's Charlson-1 score was n points; If the patient had m (where m is 0 or 1 or 2 or 3) of the following diseases at the time of ITP diagnosis: hemiplegia, moderate to severe kidney disease, and diabetes with end-organ failure, then the patient's Charlson-2 score was 2m points.

[0066] 3. Data Analysis

[0067] In this study, the high-risk factors for infection in ITP patients described in previous literature were investigated. These factors could be obtained through detailed medical history taking and routine laboratory tests. The finally included factors were the patient's gender, age at ITP diagnosis (whether ≥60 years old), comorbidities, platelet count (whether <20,000 / μL), and absolute lymphocyte count (whether <1,000 / μL). All this information was retrospectively collected from the electronic medical records of each center. In the logistic regression analysis, variables with more than 30% missing values were excluded, and only complete cases were used to develop and validate the prediction model. Through univariate and multivariate logistic regression analysis of the development cohort, variables with P < 0.10 in the univariate analysis were included in the multivariate regression model. The final selection of the prediction model adopted the stepwise logistic regression method based on the Akaike information criterion. Based on the regression coefficients of each factor in the results of the multivariate analysis, scores were assigned to each factor for the model. The model was internally validated by bootstrap (n = 1,000) and geographically externally validated in both the development cohort and the validation cohort. The validation indicators included discrimination, calibration, and clinical net benefit. All data analyses were completed using IBM SPSS 24.0 and R software.

[0068] II. Obtaining and validating the ACPA score model for predicting the risk of CAP in ITP patients

[0069] 1. Risk factors for CAP

[0070] Univariate analysis of baseline characteristics was performed on CAP and non-CAP patients in the development cohort. Finally, 5 factors with P<0.10 entered the multivariate analysis: male gender (P<0.001), age at diagnosis of ITP ≥60 years (P<0.001), Charlson score at diagnosis of ITP ≥3 (P<0.001), platelet count at diagnosis of ITP <20,000 / μL (P<0.001), absolute lymphocyte count at diagnosis of ITP <1,000 / μL (P<0.001). Multivariate analysis finally confirmed four independent risk factors (Table 2): age at diagnosis of ITP ≥60 years (OR 1.901, 95%CI 1.242-2.910; P = 0.003), Charlson score at diagnosis of ITP ≥3 (OR 9.328, 95%CI 5.240-16.604; P<0.001), platelet count at diagnosis of ITP <20,000 / μL (OR 3.776, 95%CI 2.425-5.879; P<0.001), absolute lymphocyte count at diagnosis of ITP <1,000 / μL (OR 8.932, 95%CI 5.893-13.539; P<0.001).

[0071] Table 2. Model regression coefficients and scores based on independent risk factors in the development cohort

[0072]

[0073] 2. Model establishment

[0074] Scores were assigned based on the regression coefficients of the independent risk factors for CAP in ITP patients identified, thus establishing a model (Table 2): Age ≥ 60 years at the time of ITP diagnosis was assigned 1 point, and age < 60 years was assigned 0 points; Charlson score ≥ 3 at the time of ITP diagnosis was assigned 3.5 points, and Charlson score < 3 was assigned 0 points; Platelet count < 20,000 / μL at the time of ITP diagnosis was assigned 2 points, and platelet count ≥ 20,000 / μL at the time of ITP diagnosis was assigned 0 points; Absolute lymphocyte count < 1,000 / μL at the time of ITP diagnosis was assigned 3.5 points, and absolute lymphocyte count ≥ 1,000 / μL at the time of ITP diagnosis was assigned 0 points. Based on the identified risk factors, this model was named the ACPA model. Scoring each patient in the development cohort according to the ACPA model, it was observed that as the ACPA score increased, the risk of CAP also increased (Table 3): Among 1,118 patients with a total ACPA score of 0 - 3.5 points, 46 had CAP (4.1%); among 227 patients with a total score of 4.5 - 8 points, 77 had CAP (33.9%); among 12 patients with a total score of 9 - 10 points, 11 (91.7%) had CAP (Table 3). The CAP prediction risks at each score in the ACPA model are shown in Table 4. The risk of CAP in patients with a score of 0 was approximately 1.4%, while the risk of CAP in patients with a score of 10 was approximately 99.9%.

[0075] Table 3. Number of ITP patients with CAP in different risk stratification intervals

[0076]

[0077] Note: * 148 patients in the development cohort and 78 patients in the validation cohort had no ACPA scores due to lack of records of absolute lymphocyte counts at the time of ITP diagnosis.

[0078] Table 4. Predicted risks of CAP in ITP patients at each ACPA score

[0079]

[0080] 4. Internal and external validation of the model

[0081] The established ACPA score model was validated by the bootstrap method with 1000 repetitions based on the development cohort and the validation cohort. By analyzing the discrimination and calibration, the performance of the ACPA score model of the present invention was evaluated. Among them, the discrimination was calculated from the ROC curve (area under the curve (AUC)). The calibration was evaluated using a calibration plot, and a perfect calibration plot was represented by a 45° diagonal line. The net benefit was evaluated by decision curve analysis (DCA).

[0082] The results showed that: in the development cohort, the ACPA score model showed good discrimination in predicting the risk of CAP in ITP patients, with an AUC of 0.853 (95% CI 818 - 0.889)( Figure 2 A). In addition, the calibration curve showed good consistency between the actual probability and the prediction of the model of the present invention( Figure 3 A). In the validation cohort, the AUC of the AHC score model was 0.862 (95% CI 0.807 - 0.916), indicating good discrimination( Figure 2 B). Figure 3 Figure B shows the calibration curve of the validation cohort, which reflects relatively good consistency between the actual risk and the predicted risk( Figure 3 B). In addition, similar to the development cohort, the ACPA score range in the validation cohort was from 0 to 10. Among the 387 patients with scores of 0 - 3.5, 14 (3.6%) had CAP; among the 106 patients with scores of 4.5 - 8, 32 (30.2%) had CAP; among the 18 patients with scores of 9 - 10, 16 (88.9%) had CAP (Table 3). According to the results obtained from the development cohort and the validation cohort, based on the ACPA score model of the present invention, the risk of CAP in ITP patients was defined into three categories: an ACPA score of 0 - 3.5 indicated a low risk of CAP in ITP patients; an ACPA score of 4.5 - 8 indicated a medium risk of CAP in ITP patients; an ACPA score of 9 - 10 indicated a high risk of CAP in ITP patients, that is, the scoring system of the ACPA score model of the present invention (referred to as the ACPA scoring system for short). The risk of CAP in ITP patients with an ACPA score of 4.5 - 8 was higher than that of patients with an ACPA score of 0 - 3.5; the risk of CAP in ITP patients with an ACPA score of 9 - 10 was higher than that of patients with an ACPA score of 4.5 - 8. The results of the DCA analysis showed that: using the ACPA score model, patients could have certain clinical benefits( Figure 4 ).

[0083] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A prediction device for the risk of CAP in ITP patients, comprising the following processing modules: (1) Data input module: This module is used to input the age value, concomitant disease status, platelet count, and lymphocyte count of the tester at the time of ITP diagnosis; The concomitant disease status is whether the tester has the following diseases: myocardial infarction, congestive heart failure, peripheral vascular wall, cerebrovascular disease, chronic lung disease, ulcer disease, diabetes, hemiplegia, moderate to severe kidney disease, and diabetes with end-organ failure; (2) Data recording module: This module is used to receive and store the age value, concomitant disease status, platelet count, and lymphocyte count of the tester at the time of ITP diagnosis output from the data input module; (3) Data assignment module: This data assignment module consists of an age data assignment module, a concomitant disease data assignment module, a platelet data assignment module, and a lymphocyte data assignment module: The age data assignment module is used to retrieve the age value of the tester stored in the data recording module, assign a value to the age value, and output f 年龄 : when "the age of the tester ≥ 60 years old", f 年龄 is 1 point; when "the age of the tester < 60 years old", f 年龄 is 0 point; The concomitant disease data assignment module is used to retrieve the concomitant disease status of the tester stored in the data recording module, calculate the Charlson score of the tester and assign a value to the Charlson score, and output f Charlson评分 : when "the Charlson score of the tester ≥ 3 points", then f Charlson评分 is 3.5 points, when "the Charlson score of the tester < 3 points", then f Charlson评分 is 0 points; the Charlson score = Charlson-1 score + Charlson-2 score: if the tester has n of the following diseases: myocardial infarction, congestive heart failure, peripheral vascular wall, cerebrovascular disease, chronic lung disease, ulcer disease, and diabetes, then the Charlson-1 score of the tester is n points; the n is 0 or 1 or 2 or 3 or 4 or 5 or 6 or 7; if the tester has m of the following diseases: hemiplegia, moderate to severe kidney disease, and diabetes with end-organ failure, then the Charlson-2 score of the tester is 2m points; the m is 0 or 1 or 2 or 3; The platelet data assignment module is used to retrieve the platelet count of the tester stored in the data recording module, assign a value to the platelet count, and output f 血小板 : when "the tester's platelet count < 20,000 / μL", f 血小板 is 2 points; when "the tester's platelet count ≥ 20,000 / μL", f 血小板 is 0 points; The lymphocyte data assignment module is used to retrieve the lymphocyte count of the tester stored in the data recording module, assign a value to the lymphocyte count, and output f 淋巴细胞 : When "the tester's lymphocyte count < 1000 / uL", f 淋巴细胞 is 3.5 points; when "the tester's lymphocyte count ≥ 1000 / uL", f 淋巴细胞 is 0 points; (4) Data calculation module: This module is used to receive f output from the age data assignment module 年龄 , f output from the concomitant disease data assignment module Charlson评分 , f output from the platelet data assignment module 血小板 and f output from the lymphocyte data assignment module 淋巴细胞 , and then calculate the ACPA score of the tester according to Formula Ⅰ; ACPA integral = f 年龄 + f Charlson评分 + f 血小板 + f 淋巴细胞 Formula Ⅰ; (5) Data grouping module: This module is used to receive the ACPA scores of the test subjects output from the data calculation module, then group the test subjects according to the ACPA scores, and output the risk grouping results; The criteria for grouping the test subjects according to the ACPA scores are as follows: Test subjects with ACPA scores of 9-10 are in the high-risk group, test subjects with ACPA scores of 4.5-8 are in the medium-risk group, and test subjects with ACPA scores of 0-3.5 are in the low-risk group; (6) Conclusion output module: This module is used to receive the risk grouping results output from the data grouping module, and output a conclusion based on the risk grouping results: namely, test subjects in the high-risk group are ITP patients at high risk of developing CAP; test subjects in the medium-risk group are ITP patients at medium risk of developing CAP; test subjects in the low-risk group are ITP patients at low risk of developing CAP.

2. The device according to claim 1, wherein: The platelet count is the platelet count in whole blood.

3. The device according to claim 1 or 2, wherein: The lymphocyte count is the lymphocyte count in whole blood.

4. A prediction system for the risk of CAP in ITP patients, comprising the device according to any one of claims 1 - 3, a platelet count device, a lymphocyte count measurement device, a data input device, and a data output device.

5. The system according to claim 4, wherein: The platelet counting device is a platelet counting device for whole blood.

6. The system according to claim 4 or 5, wherein: The lymphocyte count measuring device is a lymphocyte count measuring device for whole blood.

7. A method of using the system according to any one of claims 4 - 6, comprising the following steps: obtaining the following data of the tester at the time of ITP diagnosis: age value, concomitant disease status, platelet count, and lymphocyte count; predicting the risk of CAP in the tester using the device according to any one of claims 1 - 3 based on the data.

8. Use of the device according to any one of claims 1 - 3 or the system according to any one of claims 4 - 6 in the preparation of a product for predicting or assisting in predicting the risk of CAP in ITP patients.

9. Use of the device according to any one of claims 1 - 3, the platelet counting device according to any one of claims 4 - 6, and the lymphocyte count measuring device according to any one of claims 4 - 6 in the preparation of a product for predicting or assisting in predicting the risk of CAP in ITP patients.

Images