Prediction method and system for thrombotic thrombocytopenic purpura, electronic equipment and storage medium

Abstract

The invention specifically discloses a method and a system for predicting thrombocytopenic purpura, electronic equipment and a storage medium, and relates to the technical field of biological medicines. The invention provides a new prediction model (including platelet count, reticulocyte percentage, indirect bilirubin, broken red blood cells and a ratio of lactic dehydrogenase to normal high value) to help clinicians to early identify TTP patients, improve the TTP diagnostic rate, reduce the misdiagnosis rate, and improve the accuracy of TTP diagnosis on the basis of improving the diagnostic rate. The method has important guiding significance in the aspects of early treatment of patients, improvement of prognosis and the like.

Description

Technical Field

[0001] The present invention relates to the field of biomedical technologies, and particularly to a method, a system, an electronic device and a storage medium for predicting thrombotic thrombocytopenic purpura. Background Art

[0002] TTP (thrombotic thrombocytopenic purpura) is a severe disseminated thrombotic microangiopathy, characterized by microangiopathic hemolytic anemia, platelet aggregation and consumptive reduction, and organ damage (such as kidneys, central nervous system, etc.) caused by microthrombus formation. In patients without early recognition and timely plasma exchange, the mortality rate is as high as over 80%. An ADAMTS13 activity less than 10% and a positive ADAMTS13 inhibitor are the basis for diagnosing TTP. Clinically, it is observed that critically ill patients often also present symptoms such as progressive thrombocytopenia, hemolytic anemia, and multi-organ function damage. The clinical manifestations are similar to those of TTP, but the ADAMTS13 activity is normal or low, and there is no production of ADAMTS13 inhibitor, which is called TTP-like syndrome. The pathophysiological mechanism of TTP-like syndrome is unclear and may be related to severe inflammatory reactions leading to vascular endothelial damage or complement activation, etc. The clinical features are difficult to distinguish from those of TTP. It is necessary to perform differential diagnosis on critically ill patients with thrombocytopenia, hemolytic anemia and concomitant organ damage, quickly identify TTP and TTP-like syndrome, and thus give targeted emergency treatment such as plasma exchange to improve the prognosis of patients. For patients with TTP-like syndrome, plasma exchange treatment is not the preferred treatment method for TTP-like syndrome, and multidisciplinary consultation should be considered to formulate a comprehensive treatment plan. The treatment methods for TTP and TTP-like syndrome are shown in Figure 1 。

[0003] Due to the delay in the results of ADAMTS13 activity and antibody detection, the traditional PLASMIC score table (as shown in Table 3) is commonly used in clinical work to predict and diagnose TTP, and it is defined that a score of 0 - 4 is low risk, and the TTP prediction efficiency is <5%; a score of 5 is medium risk, and the prediction efficiency is 5% - 25%; a score of 6 - 7 is high risk, and the prediction efficiency is 60% - 80%. Clinical verification found that the sensitivity of diagnosing TTP for those with a high-risk score is 81.7% and the specificity is 71.4%.

[0004] Table 3 PLASMIC Score Table

[0005]

[0006] Its traditional PLASMIC score table has the following defects:

[0007] 1. The traditional PLASMIC score is a commonly used criterion for identifying patients with high-risk TTP, but it has limitations in differentiating TTP from TTP-like syndromes.

[0008] 2. In clinical practice, there are very few transplanted patients, and most TTP patients have no history of tumors. Therefore, the traditional PLASMIC score cannot be well applied to actual clinical work.

[0009] 3. There is currently no literature study indicating that the traditional PLASMIC score table can be used to distinguish between TTP and TTP-like syndrome patients.

[0010] Although the modified PLASMIC score plus the LDH / ULN ratio in the study by N Zhao et al. may be more suitable for identifying patients with ADAMTS13 deficiency, especially in early diagnosis, it also lacks the identification of TTP-like syndrome patients. Summary of the Invention

[0011] (1) Technical problems to be solved

[0012] In view of this, one of the main purposes of the present invention is to provide a method for predicting thrombotic thrombocytopenic purpura, the method comprising:

[0013] Obtaining data, obtaining the clinical data of a sample to be tested, the clinical data including the percentage of reticulocytes, platelet count, percentage of schizocytes, ratio of lactate dehydrogenase to upper limit of normal, and indirect bilirubin;

[0014] Analyzing data, inputting the clinical data into a constructed prediction model, and the prediction model predicting whether the sample to be tested is a thrombotic thrombocytopenic purpura sample based on the clinical data;

[0015] Outputting a prediction result, predicting and outputting the result of whether the sample to be tested is a thrombotic thrombocytopenic purpura sample based on the output result of the prediction model.

[0016] The prediction method provided by the present invention is mainly applied in clinical work to distinguish critically ill patients presenting with thrombocytopenia, hemolytic anemia, and organ damage, so as to quickly identify TTP and TTP-like syndrome patients at an early stage, and thus implement targeted emergency interventions for the patients, such as plasma exchange treatment, etc., to save the lives of the patients.

[0017] (2) Technical solutions

[0018] To solve the above problems, the present invention provides a method for predicting thrombotic thrombocytopenic purpura, the method comprising:

[0019] Obtain data, obtain the clinical data of the sample to be tested, where the clinical data includes reticulocyte percentage (Reticulocyte count, Ret), platelet count (Platelet count, PLT), schistocyte percentage (Schistocyte count), ratio of lactate dehydrogenase to upper limit of normal (LDH / ULN), and indirect bilirubin (Indirectbilirubin, IBIL);

[0020] Analyze data, input the clinical data into the constructed prediction model, and the prediction model predicts whether the sample to be tested is a thrombotic thrombocytopenic purpura sample based on the clinical data;

[0021] Output the prediction result, predict and output the result of whether the sample to be tested is a thrombotic thrombocytopenic purpura sample based on the output result of the prediction model.

[0022] The present invention provides, in another aspect, a prediction system for thrombotic thrombocytopenic purpura, and the system includes:

[0023] A data acquisition unit, which acquires the clinical data of the sample to be tested, where the clinical data includes reticulocyte percentage, platelet count, schistocyte percentage, ratio of lactate dehydrogenase to upper limit of normal, and indirect bilirubin;

[0024] A data analysis unit, which inputs the clinical data into the constructed prediction model, and the prediction model predicts whether the sample to be tested is a thrombotic thrombocytopenic purpura sample based on the clinical data;

[0025] An output prediction result unit, which predicts and outputs the result of whether the sample to be tested is a thrombotic thrombocytopenic purpura sample based on the output result of the prediction model.

[0026] In one embodiment, it is judged whether the sample to be tested is a thrombotic thrombocytopenic purpura sample based on the following criteria:

[0027] If the reticulocyte percentage, schistocyte percentage, ratio of lactate dehydrogenase to upper limit of normal, and indirect bilirubin are higher than the thresholds, and the platelet count is lower than the threshold, then the sample to be tested is a thrombotic thrombocytopenic purpura sample.

[0028] In one embodiment, the criteria further include:

[0029] If the reticulocyte percentage, schistocyte percentage, ratio of lactate dehydrogenase to upper limit of normal, and indirect bilirubin are lower than the thresholds, and the platelet count is higher than the threshold, then the sample to be tested is a non-thrombotic thrombocytopenic purpura sample.

[0030] In one embodiment, the non-thrombotic thrombocytopenic purpura includes a TTP-like syndrome.

[0031] In one embodiment, the threshold of the reticulocyte percentage is 5.90%;

[0032] and / or the threshold of the schizocyte percentage is 0.55%;

[0033] and / or the threshold of the ratio of lactate dehydrogenase to the upper limit of normal is 2.753;

[0034] and / or the threshold of indirect bilirubin is 24.95 μmol / L;

[0035] and / or the threshold of platelet count is 11.5×10 9 / L.

[0036] In one embodiment, the above threshold may be a level 10%, 20%, 25%, 30%, 40%, 50%, 60%, 70%, 75%, 80%, 90%, 100%, 150%, 200% or even higher than the threshold.

[0037] In one embodiment, the below threshold may be a level 10%, 20%, 25%, 30%, 40%, 50%, 60%, 70%, 75%, 80%, 90%, 100%, 150%, 200% or even lower than the threshold.

[0038] In one embodiment, the sample to be tested is from a subject.

[0039] In one embodiment, the subject includes mammals and non-mammals. Examples of mammals include, but are not limited to, any member of the class Mammalia: humans, non-human primates such as chimpanzees and other apes and monkeys; farm animals such as cows, horses, sheep, goats, pigs; domestic animals such as rabbits, dogs, and cats; laboratory animals, including rodents such as rats, mice, and guinea pigs, etc. Examples of non-mammals include, but are not limited to, birds, fish, or other non-mammals, etc.

[0040] In one embodiment, the subject is a human.

[0041] In one embodiment, the sample to be tested refers to a composition obtained from or derived from a target subject, which contains cell entities and / or other molecular entities to be characterized and / or identified, for example, based on physical, biochemical, chemical, and / or physiological characteristics. The sample can be obtained from the subject's blood and other fluid samples and tissue samples of biological origin, such as biopsy tissue samples or tissue cultures or cells derived therefrom. The source of the tissue sample can be solid tissue, such as from fresh, frozen, and / or preserved organ or tissue samples, biopsy tissues, or aspirates; blood or any blood component; body fluid; cells at any time during individual pregnancy or development; or plasma.

[0042] In one embodiment, the sample to be tested includes any sample of body tissue, cells, or fluid, or any sample derived from the body, such as swabs, washing fluids, aspirates, or rinsing fluids.

[0043] In one embodiment, the sample to be tested includes, but is not limited to, blood, serum, plasma, blood components, synovial fluid, urine, semen, saliva, feces, cerebrospinal fluid, gastric contents, vaginal secretions, mucus, tissue biopsy samples, tissue homogenates, bone marrow aspirates, bone homogenates, sputum, aspirates, wound exudates, swabs, or swab rinsing fluids, such as nasopharyngeal swabs, and other body fluids. Preferably, the sample to be tested is a blood sample or a blood-derived sample, such as serum, plasma, or blood components.

[0044] In one embodiment, the blood sample or blood-derived sample is fresh, frozen, and / or preserved.

[0045] In one embodiment, the subject is healthy.

[0046] In one embodiment, the subject is non-healthy.

[0047] In one embodiment, the subject has thrombotic thrombocytopenic purpura.

[0048] In one embodiment, the subject has a TTP-like syndrome.

[0049] On the other hand, the present invention also provides a predictive electronic device for thrombotic thrombocytopenic purpura, the device including a memory, and more than one program, wherein the more than one program is stored in the memory and is configured to be executed by more than one processors to implement the instructions of the above-mentioned predictive method for thrombotic thrombocytopenic purpura.

[0050] On the other hand, the present invention also provides a non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by a processor of an electronic device, enabling the electronic device to execute the above-mentioned predictive method for thrombotic thrombocytopenic purpura.

[0051] (III) Beneficial effects

[0052] The present invention provides a method for predicting thrombotic thrombocytopenic purpura, and the method includes:

[0053] Obtaining data, obtaining the clinical data of a sample to be tested, where the clinical data includes the percentage of reticulocytes, platelet count, percentage of schizocytes, ratio of lactate dehydrogenase to the upper limit of normal value, and indirect bilirubin;

[0054] Analyzing data, inputting the clinical data into a constructed prediction model, and the prediction model predicts whether the sample to be tested is a thrombotic thrombocytopenic purpura sample based on the clinical data;

[0055] Outputting a prediction result, predicting and outputting the result of whether the sample to be tested is a thrombotic thrombocytopenic purpura sample based on the output result of the prediction model.

[0056] Compared with the prior art, the following beneficial effects are achieved:

[0057] 1. The prediction method provided by the present invention can help clinicians quickly distinguish patients with TTP and TTP-like syndromes for timely treatment. It makes up for the limitations of the traditional PLASMIC score in distinguishing TTP from TTP-like syndromes. The sensitivity and negative predictive value of the traditional PLASMIC score in diagnosing TTP are both 100.00%, while the specificity is only 30.00% and the positive predictive value is 64.10%. The prediction model provided by the present invention has a sensitivity of 100.00%, a specificity of 65.00%, a positive predictive value of 78.13%, a negative predictive value of 100.00%, and a higher prediction ability (96.9%), better meeting the clinical needs for early diagnosis of TTP.

[0058] (IV) Terms and definitions

[0059] As used herein, the term "threshold" refers to a predetermined number used in an operation. For the determination of the threshold, different measurement techniques will obtain different measurement results, and in other measurement techniques, this value will change, and thus analog conversion may be required, which is within the scope of the skills of those skilled in the art. At the same time, it is easy to understand that the thresholds of different biomarkers are different and need to be determined according to the actual detection situation, and the determination method is familiar to those skilled in the art.

[0060] As used herein, the term "sample to be tested" refers to a composition obtained from or derived from a target subject, which contains cellular entities and / or other molecular entities to be characterized and / or identified, for example, based on physical, biochemical, chemical, and / or physiological characteristics. The sample can be obtained from the subject's blood and other fluid samples and tissue samples of biological origin, such as biopsy tissue samples or tissue cultures or cells derived therefrom. The source of the tissue sample can be solid tissue, such as from fresh, frozen, and / or preserved organ or tissue samples, biopsy tissue, or aspirates; blood or any blood component; body fluid; cells from any time during an individual's pregnancy or development; or plasma.

[0061] As used herein, "sensitivity" refers to the proportion of subjects with a positive outcome who are correctly identified as positive (e.g., correctly identifying those subjects who have the disease or medical condition for which they are being tested). For example, sensitivity can characterize the ability of a method to correctly identify the number of subjects in a population with TTP.

[0062] As used herein, the term "specificity" refers to the proportion of subjects with a negative outcome who are correctly identified as negative (e.g., correctly identifying those subjects who do not have the disease or medical condition being tested). For example, specificity can characterize the ability of a method to correctly identify the number of subjects in a population without TTP.

[0063] As used herein, the term "ROC" or "ROC curve" refers to the receiver operating characteristic curve. The ROC curve can be a graphical representation of the performance of a binary classifier system. For any given method, the ROC curve can be generated by plotting sensitivity against specificity at various threshold settings. Additionally, given at least one of three parameters (e.g., sensitivity, specificity, and threshold setting), the ROC curve can determine the value or expected value of any unknown parameter. A curve fitted to the ROC curve can be used to determine the unknown parameter.

[0064] As used herein, the terms “AUC” or “ROC-AUC” generally refer to the area under the receiver operating characteristic curve. This metric can provide a measure of the diagnostic utility of the method, taking into account the sensitivity and specificity of the method. Generally, ROC-AUC is in the range of 0.5 to 1.0, where a value close to 0.5 indicates that the method has limited diagnostic utility (e.g., lower sensitivity and / or specificity), while a value close to 1.0 indicates that the method has greater diagnostic utility (e.g., higher sensitivity and / or specificity). See, e.g., Pepe et al., “Limitations of the Odds Ratio in Gauging the Performance of a Diagnostic, Prognostic, or Screening Marker,” Am. J. Epidemiol 2004, 159(9):882-890, which is incorporated herein by reference in its entirety. According to Cook, “Use and Misuse of the Receiver Operating Characteristic Curve in Risk Prediction,” Circulation 2007, 11

[0065] 5:928-935, outlines other methods of characterizing diagnostic utility using likelihood functions, likelihood ratios, information theory, predictive values, calibration (including goodness of fit) and reclassification measures, which is incorporated herein by reference in its entirety.

[0066] As used herein, the term “positive predictive value” or “PPV” refers to the probability that a subject has a positive outcome given that the subject has a positive test result.

[0067] As used herein, the term “negative predictive value” or “NPV” refers to the probability that a subject has a negative outcome given that the subject has a negative test result. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0069] Figure 1 is a flow chart of the treatment method for TTP and TTP-like syndromes;

[0070] Figure 2 is an analysis diagram of the underlying diseases of TTP and TTP-like syndromes;

[0071] Figure 3 It is an analysis diagram of systemic inflammatory markers for TTP and TTP-like syndromes;

[0072] Figure 4 It is a receiver operating characteristic (ROC) curve graph for evaluating the diagnostic efficacy of a prediction model in differentiating patients with TTP from those with TTP-like syndromes;

[0073] Figure 5 It is a comparison graph of treatment effects. Detailed implementation manners

[0074] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0075] As used herein, "comprising", "having" or "including" includes "containing", "consisting essentially of...", "consisting substantially of..." and "consisting of"; "consisting essentially of...", "consisting substantially of..." and "consisting of" are subordinate concepts of "comprising", "having" or "including".

[0076] The experimental methods used in the following embodiments are all conventional methods unless otherwise specified, and the reagents, methods and equipment used are all conventional reagents, methods and equipment in the technical field of the present invention unless otherwise specified.

[0077] Example 1

[0078] Grouping of samples to be tested:

[0079] The patients were divided into a TTP group and a TTP-like syndrome group according to the test results of ADAMTS13 activity and antibodies sent for inspection by the patients.

[0080] (1) The inclusion criteria for TTP are as follows: (a) The patient shows thrombocytopenia, microangiopathic hemolytic anemia (MAHA), fever or organ dysfunction; (b) ADAMTS13 activity < 10% and ADAMTS13 inhibitor is positive; (c) Regular follow-up is currently being carried out.

[0081] (2) Inclusion criteria for TTP-like syndrome are as follows: (a) patients present with thrombocytopenia, microangiopathic hemolytic anemia (MAHA), fever, or organ dysfunction; (b) ADAMTS13 activity > 10% and ADAMTS13 inhibitor negative; (c) currently undergoing regular follow-up.

[0082] Example 2

[0083] Clinical data analysis:

[0084] According to the grouping method of Example 1, 35 patients among the patients with the test samples were diagnosed with TTP, and 42 patients were diagnosed with TTP-like syndrome (see Table 1). Analyzing their clinical data (clinical characteristics), the results are as follows:

[0085] The median age of the TTP group was 52 years (44 - 64 years), and the median age of the TTP-like syndrome group was 57 years (35 - 68 years).

[0086] The patients in the TTP group had significantly more skin bruising and neurological symptoms (including confusion, agitation, delirium, seizures, and coma) than the TTP-like syndrome group (77.1% vs 24.4%, p < 0.001; 88.6% vs 42.9%, p < 0.001).

[0087] The average ADAMTS13 activity in the TTP group was significantly lower than that in the TTP-like syndrome group (8.30% vs 46.12% respectively).

[0088] Compared with the TTP-like syndrome group, more patients in the TTP group had positive antinuclear antibodies (64.7% vs 33.3%, p = 0.015), more schizocytes (1.4% vs 0.15%, p < 0.001), reticulocytes (10.52% vs 2.87%, p < 0.001), indirect bilirubin (41.20 μmol / L vs 13.00 μmol / L, p < 0.001), lactate dehydrogenase (1637.00 U / ml vs 604.20 U / ml, p = 0.003), LDH / upper limit of normal ratio (5.40 vs 2.49, p = 0.005), and lower red blood cell (p = 0.001) and platelet (p < 0.001) counts, and lower hemoglobin level (p = 0.003). On the other hand, patients in the TTP-like syndrome group had more elevated APTT (35.00 seconds vs 31.90 seconds, p = 0.045), D-dimer (4.18 mg / L vs 2.80 mg / L, p = 0.036), and lower albumin level (31.55 g / L vs 38.40 g / L, p < 0.001).

[0089] Table 1 Clinical characteristics of patients with TTP and TTP-like syndrome

[0090]

[0091]

[0092] Example 3

[0093] Analysis of underlying diseases:

[0094] Analysis of underlying diseases was performed on the TTP group and the TTP-like syndrome group, and the results (see Figure 2 ) are as follows:

[0095] In the TTP group, the main underlying diseases were autoimmune diseases (n = 13, 37.14%), including 5 cases of systemic lupus erythematosus (38.46%), 6 cases of undifferentiated connective tissue disease (46%), and 2 cases of Sjögren's syndrome (15%); other underlying diseases included stroke (n = 11, 31.43%), infection (n = 6, 17.14%), trauma or surgery (n = 2, 5.71%), malignant tumor (n = 1, 2.86%), and unknown cause (n = 6, 17.14%).

[0096] The main underlying diseases in the TTP-like syndrome group were infection (n = 23, 54.76%) and malignant tumor (n = 13, 30.95%); pathogens were identified in 18 patients, including fungi (n = 7) (6 cases of Aspergillus infection and 1 case of Candida albicans), Escherichia coli (n = 2), Pseudomonas aeruginosa (n = 1), Acinetobacter baumannii (n = 1), Burkholderia (n = 1), Aeromonas (n = 1), cytomegalovirus (n = 1), and SARS-CoV-2 (n = 1), and 3 patients had multiple pathogen infections; malignant tumors included myelodysplastic syndrome (n = 7), acute leukemia (n = 2), colorectal cancer (n = 1), gastric cancer (n = 1), prostate cancer (n = 1), and craniopharyngioma (n = 1); other underlying diseases included autoimmune diseases (n = 5, 11.90%), pregnancy (n = 4, 9.52%), stroke (n = 3, 7.14%), hematopoietic stem cell transplantation (n = 1, 2.38%), acute myocardial infarction (n = 1, 2.38%), rhabdomyolysis (n = 1, 2.38%), and unknown cause (n = 4, 9.52%).

[0097] Example 4

[0098] Analysis of inflammatory indicators:

[0099] Analysis of inflammatory indicators was performed on the TTP group and the TTP-like syndrome group, and the results (see Figure 3 ) are as follows:

[0100] The systemic inflammatory response in patients with TTP-like syndrome is usually more severe than that in patients with TTP. In contrast, the neutrophil-to-lymphocyte ratio (NLR) in patients with TTP-like syndrome is significantly higher (median 9.18 vs. 5.25, p = 0.005) (see Figure 3 A), the platelet-to-lymphocyte ratio (PLR) (median 22.81 vs. 6.67, p < 0.001) (see Figure 3 B), the monocyte-to-lymphocyte ratio (CLR) (median 75.52 vs. 8.18, p < 0.001) (see Figure 3 C), the systemic inflammation index (SII) (median 203.3 vs. 43.67, p < 0.001) (see Figure 3 D), C-reactive protein (CRP) (median 60.10 vs. 9.59, p < 0.001) (see Figure 3 E), procalcitonin (PCT) (median 0.48 vs. 0.14, p = 0.002) (see Figure 3 F), and D-dimer (median 4.18 mg / L vs. 2.80 mg / L, p = 0.036) (see Figure 3 G) are significantly higher, while the albumin level (ALB) is significantly lower (median 31.55 g / L vs. 38.40 g / L, p < 0.001) (see Figure 3 H). No significant differences were found in the lymphocyte-to-neutrophil ratio (LMR) (p = 0.138) (see Figure 3 I) and white blood cell count (WBC) (p = 0.509) (see Table 1).

[0101] Examples 3 and 4 verified that TTP and TTP-like syndrome are two different diseases. For example, TTP is most often associated with autoimmune diseases, while TTP-like syndrome is often triggered by infections caused by various pathogens. In addition, the levels of inflammatory markers such as NLR, PLR, CRP, PCT, and D-dimer in patients with TTP-like syndrome are significantly elevated, and these findings strongly suggest the presence of severe systemic inflammatory responses in these patients.

[0102] Example 5

[0103] Prediction model establishment and verification:

[0104] 1. Prediction model establishment method:

[0105] Non-parametric Mann-Whitney U test was used to compare continuous variables of clinical data, and chi-square test was used to analyze categorical variables. Variables with significant differences were regarded as potential diagnostic predictors of TTP, and the predictive ability of these variables and their combinations was evaluated by ROC curve analysis and binary logistic regression. The area under the curve (AUC) was calculated to evaluate the predictive ability of variables, where a higher AUC value indicated stronger overall predictive ability. In addition, the differences between individual predictors and combinations of predictors were compared to identify the most effective combination. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the new predictive combination and the traditional PLASMIC score were calculated and compared. At the same time, the optimal cut-off value of the predictor was determined using ROC curve analysis.

[0106] 2. Performance analysis of single clinical data prediction model

[0107] As Figure 4 shown, the AUC values of individual variables ranged from 0.6933 to 0.8515 in the diagnosis of TTP. When the single predictor was a platelet count < 11.5 x 10^9 / L, the specificity was 73.81%, the sensitivity was 85.71%, and the predictive ability for diagnosing TTP was 82.01% (see Figure 4 A). When the single predictor was a reticulocyte count > 5.90%, the specificity was 85.30%, the sensitivity was 80.70%, and the predictive ability was 85.15% (see Figure 4 B). When the single predictor was an indirect bilirubin > 24.95 μmol / L, the specificity was 81.00%, the sensitivity was 74.30%, and the predictive ability was 72.83% (see Figure 4 C). When the single predictor was an LDH / ULN > 2.753, the specificity was 56.80%, the sensitivity was 84.80%, and the predictive ability was 69.33% (see Figure 4 D). When the single predictor was a percentage of schizocytes > 0.55%, the specificity was 79.20%, the sensitivity was 74.20%, and the predictive ability was 78.36% (see Figure 4 E).

[0108] 3. Performance analysis of combined prediction model of multiple clinical data

[0109] The predictive ability of the combination of creatinine, mean corpuscular volume, and reticulocyte percentage for diagnosing TTP was only 67.7%. The predictive ability of the combination of reticulocyte percentage, platelet count, and LDH / ULN was 91.8%, while the predictive ability of the combination of reticulocyte percentage, platelet count, and indirect bilirubin was 93.8%; the predictive ability of the combination of reticulocyte percentage, platelet count, LDH / ULN, and indirect bilirubin was 92.9%; the predictive ability of the combination of reticulocyte percentage, platelet count, and schizocyte percentage was 96.1%. Additionally, the five-variable model (reticulocyte percentage, platelet count, schizocyte percentage, LDH / ULN, and indirect bilirubin) showed the highest predictive ability of 96.9% (see Figure 4 F).

[0110] 4. Comparison of the performance of the new predictive model (five-variable model) with the PLASMIC score

[0111] Subsequently, a new TTP diagnostic scoring system was developed using these five indicators (reticulocyte percentage, platelet count, schizocyte percentage, LDH / ULN, and indirect bilirubin). Based on the risk stratification of the traditional PLASMIC score, low-risk patients were defined as those with scores of 0–4, and medium- and high-risk patients were defined as those with scores of 5–7. Similarly, for the new scoring system, low-risk patients were defined as those with scores of 0–2, and medium- and high-risk patients were defined as those with scores of 3–5. The indicators and scores of the new scoring system are shown in Table 4:

[0112] Table 4 New scoring system

[0113]

[0114] The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the new and old scoring systems were compared. The survival probability was estimated according to the Kaplan–Meier curve. Statistical analysis was performed using SPSS 26.0 and R statistical software, and the criterion for considering differences to be statistically significant was p < 0.05.

[0115] In the high-risk and low-risk classifications, the sensitivity of the traditional PLASMIC score was 100.00%, the specificity was 30.00%, the positive predictive value (PPV) was 64.10%, and the negative predictive value (NPV) was 100.00%; the sensitivity of the previously modified PLASMIC score (reticulocyte percentage, platelet count, LDH / ULN, and indirect bilirubin) was 88.46%, the specificity was 70.00%, the positive predictive value was 79.31%, and the negative predictive value was 82.35%. However, the sensitivity of the new scoring system was 100.00%, the specificity was 65.00%, the positive predictive value was 78.13%, and the negative predictive value was 100.00%. The component comparison of the traditional PLASMIC or the new scoring system in patients with TTP and TTP-like syndromes is shown in Table 2.

[0116] Table 2

[0117]

[0118]

[0119] Example 6

[0120] Analysis of the efficacy of previous cases:

[0121] Effective treatment was defined as a platelet count ≥ 150 × 10^9 / L for more than two days without obvious clinical symptoms.

[0122] In the TTP group, 23 patients (65.71%) responded to treatment, while in the TTP-like syndrome group, only 22 patients (52.38%) had a positive response to treatment. Overall, 16 patients (20.77%) died, including 3 (8.57%) in the TTP group and 13 (30.95%) in the TTP-like syndrome group. Kaplan-Meier survival analysis showed a significant difference in survival rates between TTP patients and TTP-like syndrome patients, with the overall survival rate (OS) at 180 days being 90.6% (95% CI: 73.4% - 96.6%) and 60.9% (95% CI: 41.7% - 75.5%) respectively (p = 0.009) (see Figure 5 A).

[0123] For the surviving patients, the cumulative incidence of platelets increasing to ≥ 150 × 10^9 / L at 60 days was significantly higher in the TTP group than in the TTP-like syndrome group [80% (95% CI: 62.6% - 89.9%) vs 40.5% (95% CI: 25.8% - 54.7%)] (p < 0.001) (see Figure 5 B).

[0124] Plasma exchange therapy significantly improved the 180-day overall survival rate in the TTP group compared with the TTP-like syndrome group [90.6% (95% CI: 73.4% - 96.6%) vs 65.6% (95% CI: 26.0% - 87.6%)] (p = 0.054) (see Figure 5 C). However, in the TTP-like syndrome group, there was no significant difference in survival rate between patients who received plasma exchange therapy and those who did not [65.6% (95% CI: 26.0% - 87.6%) vs 54.0% (95% CI: 34.8% - 69.9%)] (p = 0.590) (see Figure 5 D).

[0125] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the said element.

[0126] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting thrombotic thrombocytopenic purpura, characterized in that: The method comprises: Acquiring data, acquiring clinical data of the sample to be tested, wherein the clinical data includes reticulocyte percentage, platelet count, schistocyte percentage, lactate dehydrogenase to high normal ratio and indirect bilirubin; Analyze the data, input the clinical data into a constructed prediction model, and the prediction model predicts whether the sample to be tested is a thrombotic thrombocytopenic purpura sample based on the clinical data; Output a prediction result, and predict and output whether the sample to be tested is a thrombotic thrombocytopenic purpura sample result based on the output result of the prediction model.

2. A prediction system for thrombotic thrombocytopenic purpura, characterized in that: The system comprises: An acquisition data unit is used to acquire clinical data of the sample to be tested, wherein the clinical data includes reticulocyte percentage, platelet count, schistocyte percentage, lactate dehydrogenase to high normal value ratio and indirect bilirubin; A data analysis unit inputs the clinical data into a constructed prediction model, wherein the prediction model predicts whether the sample to be tested is a thrombotic thrombocytopenic purpura sample based on the clinical data; The prediction result output unit predicts and outputs whether the sample to be tested is a thrombotic thrombocytopenic purpura sample result based on the output result of the prediction model.

3. The prediction method according to claim 1 or the prediction system according to claim 2, characterized in that: Whether the sample to be tested is a thrombotic thrombocytopenic purpura sample is determined based on the following criteria: If the reticulocyte percentage, schistocyte percentage, lactate dehydrogenase to high normal value ratio and indirect bilirubin are higher than the threshold value, and the platelet count is lower than the threshold value, the sample to be tested is a thrombotic thrombocytopenic purpura sample.

4. The prediction method or prediction system according to claim 3, characterized in that: The standards also include: If the reticulocyte percentage, schistocyte percentage, lactate dehydrogenase to high normal value ratio and indirect bilirubin are lower than the threshold value, and the platelet count is higher than the threshold value, the sample to be tested is a non-thrombotic thrombocytopenic purpura sample.

5. The prediction method or prediction system according to claim 3 or 4, characterized in that: The threshold value of the reticulocyte percentage is 5.90%; and / or the threshold for schistocyte percentage was 0.55%; and / or the threshold value of the lactate dehydrogenase to high-normal ratio is 2.753; and / or the threshold for indirect bilirubin is 24.95 μmol / L; and / or platelet count threshold of 11.5 × 10 9 / L.

6. The prediction method according to claim 1 or the prediction system according to claim 2, characterized in that: The sample to be tested comes from a subject.

7. The prediction method or prediction system according to claim 6, characterized in that: The subject suffers from thrombotic thrombocytopenic purpura.

8. The prediction method or prediction system according to claim 6, characterized in that: The subject suffers from TTP-like syndrome.

9. An electronic device for predicting thrombotic thrombocytopenic purpura, characterized in that: The device includes a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors to implement the instructions of the prediction method for thrombotic thrombocytopenic purpura as described in any one of claims 1 and 3-8.

10. A non-transitory computer-readable storage medium, characterized in that: When the instructions in the storage medium are executed by the processor of the device described in claim 9, the device is enabled to execute the method for predicting thrombotic thrombocytopenic purpura described in any one of claims 1, 3-8.

Images