WO2023104163A1 - System for accurate selection of therapeutic dose of tacrolimus for myasthenia gravis patient, and use - Google Patents

Abstract

The present invention relates to a prediction system for accurate selection of the therapeutic dose of tacrolimus for a myasthenia gravis patient, and a use thereof. The prediction system comprises a data acquisition unit and a prediction unit, wherein the data acquisition unit is used for acquiring feature data of a patient, and the prediction unit is used for inputting the feature data into a prediction model to obtain a tacrolimus low concentration predicted probability value. The tacrolimus initial dose and dose adjustment accurate prediction model for the myasthenia gravis patient in the present invention facilitates guiding the myasthenia gravis patient to receive individualized medication in tacrolimus treatment, reduces the situation of a low plasma-drug concentration of the patient at an early stage of treatment caused by under-dosed tacrolimus, significantly improves the efficacy at the early stage, and has important significance for accurate medication in immunosuppression treatment for the myasthenia gravis patient.

Description

System and application of precise selection of tacrolimus treatment dose for patients with myasthenia gravis technical field

The present invention relates to the technical field of drug information, in particular to a system and application for precise selection of tacrolimus treatment dosage for patients with myasthenia gravis.

Background technique

A study in 2017 showed that maintaining the concentration of tacrolimus in the range of 4.8-10ng/ml can make 92% of patients with myasthenia gravis reach the treatment goal after treatment. If the blood concentration is too low, it is easy to cause slow onset or even ineffective treatment. . Therefore, tacrolimus treatment requires regular monitoring of drug concentration, and adjusting the dose of tacrolimus according to the blood concentration results to improve the effectiveness of treatment.

However, drug concentration monitoring cannot guide the selection of the initial dose, and it is difficult to reach the target blood drug concentration in the early stage of treatment. Due to the different medical conditions in different regions, patients in some remote areas cannot monitor drug concentration in time and delay dose adjustment. In order to avoid adverse reactions in the early stage of treatment, the initial dose of tacrolimus for patients with myasthenia gravis is 2 mg/day, and then gradually increase to 3-4 mg/day, and the dose needs to be stabilized for 5-7 days before drug concentration monitoring. Concentration monitoring is often carried out in the 3rd to 4th week after taking the drug, which makes it difficult for some patients to reach the effective concentration range within the first month of tacrolimus treatment. Therefore, analyzing the influencing factors of the blood concentration of tacrolimus and applying it to the selection of the initial dose will help to improve the proportion of early blood drug concentration in patients with myasthenia gravis and improve the early curative effect.

A variety of clinical and genetic factors affect the blood concentration of tacrolimus, which can be used to guide the individualized drug selection of tacrolimus, among which a single nucleotide mutation at the CYP3A5*3 (rs776746) site is the most definite influencing factor. It has been widely used in the optimal dose selection of tacrolimus in the treatment of various diseases such as organ transplantation, inflammatory bowel disease, rheumatic immune disease and myasthenia gravis. However, CYP3A5*3 polymorphisms can only explain 29%-35% of the inter-individual differences in tacrolimus plasma concentrations. In a randomized controlled study of tacrolimus in kidney transplant patients, drug dose selection based on CYP3A5*3 genotype could not help more patients achieve effective blood drug levels or improve prognosis.

For example, Chen Di et al. (Chen Di, Zhang Hua, Yin Jian, Tian Xiaoxin, et al. Monitoring of tacrolimus blood concentration and CYP3A5 gene detection and analysis in patients with myasthenia gravis [J]. Chinese Journal of Neuroimmunology and Neurology, 2020 , 27(03):206-209+218.) The blood concentration of tacrolimus was measured by chemiluminescent microparticle immunoassay, and the CYP3A5*3 gene polymorphism was detected by fluorescent staining in situ hybridization, and multiple linear regression analysis was used Correlation between blood concentration and gender, age, weight, genotype, dosage, combined medication and other factors, the results showed that tacrolimus dosage, CYP3A5 genotype, combined use of omeprazole or esomeprazole and There was a correlation between tacrolimus plasma concentration (both P<0.05). However, this study only analyzed many factors related to tacrolimus plasma concentration through a large number of sample data, such as tacrolimus dose, CYP3A5 genotype, etc., but did not give how to accurately select or even predict other factors. The method of crolimus blood drug concentration, and this study believes that the age of the patient has no significant correlation with the tacrolimus blood drug concentration C/D value, so it is not used as a reference for the study of tacrolimus blood drug concentration More importantly, this study believes that the impact of tacrolimus dose on the monitoring of tacrolimus blood concentration is far greater than the impact of genotype, which is consistent with the international mainstream view: when the CYP3A5 genotype is a fast metabolizer, Contrary to the notion that a significant increase in tacrolimus dose is required to achieve effective concentrations, genotype generally has a much greater effect than tacrolimus dose.

As another example, Chinese patent application CN111662975A discloses the use of products for detecting mutations in CYP3A4rs2242480 and CYP3A4rs4646437 gene sites in the preparation of products for predicting or evaluating the metabolism of patients after taking tacrolimus. However, its application only includes the mutation of a single nucleotide site of the CYP3A4 gene, and does not include the CYP3A5 site, which is the most definite factor affecting tacrolimus, so its clinical practicability is low.

As another example, Chinese patent application CN112786145A discloses a method for accurately predicting the dosage of tacrolimus for organ transplant patients. Model-validated tacrolimus dose prediction model helps to personalize medication for organ transplant patients. However, in this method, the establishment of the model is based on the data of patients with organ transplantation, and there are significant differences in other crolimus treatment regimens, effective concentration ranges, concomitant medications, frequency and time points of drug concentration monitoring, and myasthenia gravis patients. The choice of initial tacrolimus dose cannot be applied to patients with myasthenia gravis.

Therefore, there is still an urgent need for a prediction system/model or prediction method for the precise selection of tacrolimus treatment dose for patients with myasthenia gravis.

In addition, on the one hand, due to differences in the understanding of those skilled in the art; The present invention does not possess the characteristics of these prior art, on the contrary, the present invention already possesses all the characteristics of the prior art, and the applicant reserves the right to add relevant prior art to the background technology.

Contents of the invention

Aiming at the deficiencies of the prior art, the present invention provides a system and application for precise selection of tacrolimus treatment dosage for patients with myasthenia gravis, aiming to solve at least one or more technical problems existing in the prior art.

In order to achieve the above purpose, the inventors constructed an early target prediction system for tacrolimus blood concentration through in-depth research on the factors affecting the blood concentration of tacrolimus in the myasthenia gravis population, and through internal and external The prediction model used was verified and provided a basis for accurately guiding the initial dose selection and early dose adjustment of tacrolimus in patients with myasthenia gravis.

Specifically, the present invention provides a system for accurately selecting the dose of tacrolimus for patients with myasthenia gravis, including:

A data acquisition unit configured to acquire at least one characteristic data of a myasthenia gravis patient;

The prediction unit is configured to determine the predicted probability value of low tacrolimus concentration in patients with myasthenia gravis based on at least one input characteristic data;

Among them, the characteristic data include age, a single nucleotide mutation site of a candidate gene related to tacrolimus metabolism, the selected tacrolimus dose, whether to use five ester capsules, and one or more of hematocrit. In particular, the prior art is usually based on the data of organ transplant patients, in which factors such as tacrolimus treatment regimen, effective concentration range, concomitant medication, frequency and time point of drug concentration monitoring are significantly different from patients with myasthenia gravis, It is not applicable to the initial dose selection of tacrolimus in patients with myasthenia gravis; existing studies on the individual differences in metabolism of tacrolimus in patients with myasthenia gravis only analyze factors related to tacrolimus concentration. On the other hand, a practical concentration prediction model has not been proposed to guide the selection and adjustment of clinical dosage. In the present invention, by establishing an accurate prediction model for the initial dose of tacrolimus and dose adjustment for patients, it is helpful to guide patients with myasthenia gravis to receive individualized medication for tacrolimus treatment, and reduce the risk of insufficient tacrolimus dosage for patients. The resulting insufficient blood drug concentration in the early stage of treatment can improve the early curative effect, which is of great significance for the precise drug use of immunosuppressive therapy in patients with myasthenia gravis. The system for the precise selection of tacrolimus treatment dose for patients with myasthenia gravis provided by the present invention is based on clinical data collection based on clinical cohorts, genetic locus screening based on previous studies, and model verification based on internal cross-validation and external validation. The precision drug research of immunosuppressant therapy for autoimmune diseases provides a model and has important reference value.

Preferably, the data acquisition unit can be an electronic input device, such as a keyboard, mouse, scanner, microphone, etc. The data acquisition unit sends the acquired feature data to the prediction unit. The predicting unit is a data processor, capable of calculating and outputting the predicted probability value of low tacrolimus concentration based on the received characteristic data according to preset computer program instructions. Preferably, the prediction unit is also able to output an instruction to adjust the dose of tacrolimus based on the calculated low tacrolimus concentration prediction probability value, and/or an instruction to use five-ester capsules. Preferably, the predicting unit sends the calculated predicted probability value of low tacrolimus concentration and/or an indication of adjusting the dose of tacrolimus, and/or an indication of whether to use five-ester capsules to the output unit, and the output unit outputs the above information . Preferably, the output unit may be a display, a sound output device, and the like.

Preferably, the prediction model in the present invention is Logit (P low concentration), and Logit (P low concentration) is at least associated with age, tacrolimus dose and at least two candidate genes related to tacrolimus metabolism to be sure. Specifically, as the optimal mode, Logit (P low concentration) = 10.023-0.047 × age - 1.263 × tacrolimus dose - 4.325 × whether to use five ester capsules + 3.039 × rs776746 genotype - 2.111 × rs1045642 genotype - 0.117 × Hematocrit. In particular, in the present invention, the tacrolimus dose and the corresponding coefficients of the rs776746 gene are 1.263 and 3.039, respectively, and the influence of the surface genotype is much greater than that of the tacrolimus dose, which is consistent with the existing mainstream views, and this conclusion can be further improved Better guide physicians to choose the dosage of tacrolimus. In addition, the present invention also determines the correlation between the hematocrit, the synthetic drug five-ester capsule, MDR1 (rs1045642) and the dosage of tacrolimus, and further determines the respective weight coefficients.

Preferably, in the present invention, the predicted probability value of low tacrolimus concentration is calculated and obtained by the principle of logistic regression in a manner related to the prediction model Logit(P low concentration). Specifically, the predicted probability value of low tacrolimus concentration can be calculated by the following formula:

Preferably, when the low concentration probability value is greater than the preset threshold, it is judged that the patient has a high risk of low tacrolimus concentration, and an instruction to increase the dose of tacrolimus or to add five ester capsules is output;

When the low concentration probability value is less than the preset threshold, it is judged that the risk of low tacrolimus concentration in the patient is low, and an instruction to administer according to the selected dose is output. Preferably, in the present invention, the preset threshold may be 50%, or 40%, or 30%, or 20%, or 10%, or any value between 0 and 50%. In particular, it is necessary to increase the dose of tacrolimus or add five ester capsules, and the selected dose is not particularly limited, and a specific dose control plan can be given according to the specific health status of the patient and the experience of the pharmacist to achieve tacrolimus. Division concentration control.

Preferably, the low concentration of tacrolimus means that the patient's blood concentration of tacrolimus is lower than the concentration threshold. In particular, in the present invention, the concentration threshold is 4.8 ng/ml.

Preferably, candidate genes include CYP3A5 and ABCB1 genes.

Preferably, the single nucleotide mutation site of the candidate gene includes a dominant genetic model at the rs776746 site and a recessive genetic model at the rs1045642 site.

Preferably, in the present invention, the prediction model is constructed based on variables through multi-factor analysis, wherein the multi-factor analysis includes but not limited to stepwise regression analysis and binary logistic regression analysis.

Preferably, the prediction model is obtained as follows:

Obtain clinical data information of patients with myasthenia gravis;

Identify factors associated with tacrolimus based on clinical data;

Based on the factors, the variable set correlated with the concentration of tacrolimus was obtained by single factor analysis;

Based on the stepwise regression analysis model, the multi-factor analysis of the variable set is carried out to obtain the characteristic data;

A prediction model for low tacrolimus concentration was constructed based on characteristic data. In particular, clinical data information can be derived from clinical data collection, genetic testing, and laboratory inspection data.

Specifically, univariate analysis may include: normally distributed continuous variables expressed as mean ± standard deviation, non-normally distributed continuous variables expressed as median, and categorical variables expressed as frequency; using Hardy Weinberg balance test Analyze the genetic balance of polymorphism at a single nucleotide site, and test whether the genotype frequency is representative of the population; use Haploview software to analyze the linkage disequilibrium between different single nucleotide sites; use independent sample t test and rank sum test Or chi-square test to compare demographic characteristics, clinical data and single nucleotide sites between groups, and screen to obtain a variable set that can be included in multivariate analysis.

Further, the present invention also relates to the process of internal cohort verification and external cohort verification of the prediction model, wherein the internal cohort verification includes the use of random method, Bootstrap re-sampling method to evaluate the internal validity of the prediction model; the external verification cohort verification includes the use of verification The cohort data is used to verify the established prediction model, and calculate the area under the receiver operating characteristic curve of the prediction results and its 95% confidence interval, specificity, sensitivity, positive predictive rate and negative predictive rate to evaluate the externality of the predictive model validity. In particular, the model of this application has verified the repeatability and universality of the model through internal and external verification methods, especially in the external verification, the area under the receiver operating characteristic curve and its 95% confidence interval are above 0.85, It shows that the prediction system and its model of the present invention can be widely used in this field, without worrying about the low prediction efficiency caused by the difference of population.

Preferably, the factors related to tacrolimus in the present invention include medication regimen, tacrolimus dose, gender, age, height, weight, body mass index, alanine aminotransferase, glutamic acid aminotransferase Enzymes, blood urea nitrogen, creatinine, hematocrit, combined medications, combined diseases, single nucleotide mutation sites of candidate genes related to tacrolimus metabolism.

Preferably, the set of variables correlated with tacrolimus concentration in the present invention includes age, whether to use glucocorticoids, whether to use pentaester capsules, hematocrit, tacrolimus dosage and three single nucleotide mutations Dominant genetic model of locus rs776746 (T/T, T/C vs C/C), recessive genetic model of locus rs2242480 (C/C vs C/T, T/T), recessive genetic model of locus rs1045642 Model (A/A vs A/G, G/G).

Preferably, the prediction system for accurate selection of tacrolimus treatment dose for patients with myasthenia gravis according to the present invention further includes a visualization unit configured to input the characteristic data of patients with myasthenia gravis into the prediction model, so that the prediction model can be visualized. output.

Preferably, the present invention also relates to an electronic device comprising,

one or more processors;

memory for storing one or more computer programs;

When one or more computer programs are executed by one or more processors, one or more processors realize the step of obtaining the predicted probability value of tacrolimus low concentration by using the predictive model: including at least one of the myasthenia gravis patients The feature data is input into the prediction model to obtain the predicted probability value of low concentration of tacrolimus.

Preferably, the present invention also relates to a storage medium containing computer-executable instructions, and the computer-executable instructions are used to realize the step of obtaining the predicted probability value of tacrolimus low concentration by using a prediction model when executed by a computer processor: including At least one characteristic data of the myasthenia gravis patient is input into the prediction model to obtain the predicted probability value of the low concentration of tacrolimus.

Preferably, the present invention also relates to the application of a prediction system for accurate selection of tacrolimus treatment dose for patients with myasthenia gravis in predicting the treatment dose of tacrolimus for patients with myasthenia gravis.

The beneficial technical effects of the present invention include: the present invention provides an accurate prediction system for the initial dose and dose adjustment of tacrolimus for patients with myasthenia gravis, which helps guide patients with myasthenia gravis to receive individualized medication for tacrolimus treatment, reducing the number of patients with myasthenia gravis. Insufficient blood drug concentration in the early stage of treatment caused by insufficient tacrolimus dose can improve the early curative effect, which is of great significance for the precise drug use of immunosuppressive therapy in patients with myasthenia gravis. In addition, the model construction process and method of the present invention, that is, clinical data collection based on clinical cohorts, gene locus screening based on previous studies, model verification based on internal cross-validation and external validation, can be used for the treatment of other autoimmune diseases with immunosuppressants. Precision drug research provides a model and has important reference value.

In the prior art, for the treatment of patients with myasthenia gravis, tacrolimus can only be administered according to an empirical dose. In order to avoid adverse reactions in the early stage of treatment, the initial dose of tacrolimus for patients with myasthenia gravis is 2 mg/day, and then gradually increase to 3-4 mg/day, and the dose needs to be stabilized for 5-7 days before drug concentration monitoring. Concentration monitoring is often carried out in the 3rd to 4th week after taking the drug, which makes it difficult for some patients to reach the effective concentration range within the first month of tacrolimus treatment. The present invention creatively proposes that the patient's low tacrolimus concentration probability value can be calculated in association with the patient's characteristic data, and the initial tacrolimus dose and whether to adjust the tacrolimus low concentration prediction probability value can be given. The indication of Wuzhi Capsules should be used, or the recommended initial dose of tacrolimus and the indication of whether Wuzhi Capsules should be used should be calculated based on the characteristic data combined with the target tacrolimus low concentration probability value, so that more scientific and personalized formulation can be made. The dosage regimen effectively overcomes the problem of poor early curative effect caused by individual differences among patients in the prior art. In addition, this system can also prevent patients from performing multiple blood drug concentration checks. Among them, the characteristic data include age, a single nucleotide mutation site of a candidate gene related to tacrolimus metabolism, the selected tacrolimus dose, whether to use five ester capsules, and one or more of hematocrit. The feature data selected in the present invention is the data most correlated with the possibility of low tacrolimus concentration of the patient, the amount of data is small and the calculation result is accurate. After verification, the accuracy rate of the tacrolimus low concentration probability value of the system disclosed in the present invention is as high as 90%, or even more than 95%, and the accuracy rate of tacrolimus dosage indication is also as high as 90%, or even more than 95%, which is remarkable. Improve the drug treatment effect of patients with myasthenia gravis, statistically speaking, significantly improve the early treatment effect, shorten the treatment cycle of a large number of patients, improve the treatment experience of patients, and save a lot of medical resources at the same time.

Description of drawings

Fig. 1 is a schematic diagram of prediction results of a visual nomogram prediction system according to a preferred embodiment of the present invention;

Fig. 2 is an AUC curve diagram of a prediction system according to a preferred embodiment of the present invention.

Detailed ways

A detailed description will be given below in conjunction with the accompanying drawings.

The prediction method or prediction model and system for the precise selection of tacrolimus treatment dosage for patients with myasthenia gravis of the present invention are used for non-diagnostic and non-therapeutic purposes.

In the present invention, the scope of clinical data information collection is not particularly limited, as long as the collection of patient information can be realized, the scope of collection includes but not limited to myasthenia gravis clinical cohort database, hospital inpatient and outpatient electronic medical record system for the first visit and follow-up The general information of the patient includes at least: patient name and serial number; demographic characteristics: gender, race, permanent residence, age, height, weight, occupation, education level. The patient's medical history data include at least: myasthenia gravis-related information: onset time, disease type, previous treatment regimen, and previous treatment outcomes; underlying diseases: underlying diseases and current treatment medications, including but not limited to diabetes, hypertension, coronary heart disease The name of the therapeutic drug, dosage, and frequency of administration. The detailed medication records of tacrolimus include at least: the time of initiation of medication, time of medication, frequency of administration, dosage, time of dose adjustment, time of drug withdrawal, and adverse drug reactions. The tacrolimus drug concentration monitoring includes at least: collecting venous blood samples, collection date and time, concentration detection method, and tacrolimus drug concentration detection value. In addition, laboratory tests include at least: blood routine, liver function, and kidney function.

In the present invention, the gene detection data is the gene detection result obtained by sequencing genes related to tacrolimus metabolism, including but not limited to CYP3A5, CYP3A4, ABCB1, POR, CYP2C19, NR1L2 genes. Sequencing can use sequencing techniques known in the art, including but not limited to sequencing by synthesis, single-molecule sequencing, and nanopore sequencing.

In the present invention, the term "factors related to tacrolimus" refers to a collection of related factors that are related to the blood concentration of tacrolimus in patients with myasthenia gravis, but without variable screening. The term "variable set correlated with tacrolimus concentration" refers to variables with strong correlation obtained by screening variables through single factor analysis, so as to be used in multivariate analysis. The term "characteristic data" refers to the independent risk factors associated with low concentrations of tacrolimus obtained through multivariate analysis and used to construct a set of factors for predictive models.

Those skilled in the art can understand that the various exemplary embodiments described in the present invention can be implemented by software, or by combining software with necessary hardware. Therefore, the specific embodiment according to the present invention can be embodied in the form of a software product, and the software product can be stored in a non-volatile storage medium or a non-transitory computer-readable storage medium (which can be CD-ROM, U disk, mobile hard disk, etc.) or on the network, including several instructions to make the computing device (which may be a personal computer, server, mobile terminal, or network device, etc.) execute the method according to the present invention.

In an exemplary embodiment, the program product of the present invention may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples of readable storage media include, but are not limited to, electrical connections with one or more conductors, portable disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable Read memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above. Preferably, computer-executable instructions for obtaining the predicted probability value of low concentration of tacrolimus can be deployed in the readable medium: including inputting at least one characteristic data of patients with myasthenia gravis into the prediction system of the present invention, and obtaining tacrolimus Probability values for low concentration predictions.

Correspondingly, based on the same inventive concept, the present invention also provides an electronic device.

In an exemplary embodiment, the electronic device takes the form of a general-purpose computing device. Components of an electronic device may include, but are not limited to, at least one processor, at least one memory, and a bus connecting different system components including the memory and the processor.

Wherein, the memory stores program codes, and the program codes can be executed by the processing unit, so that the processing unit executes the method of the present invention, that is: responding to the input of at least one item of the myasthenia gravis patient Feature data, output the predicted probability value of low tacrolimus concentration. The processor includes at least the data processing unit described in the present invention (sometimes referred to as "module" in the present invention). The memory may include readable media in the form of volatile memory elements, such as random access memory elements (RAM) and/or cache memory elements, and may further include read only memory elements (ROM).

The memory of the present invention may also include a program/utility having a set (at least one) of program modules including, but not limited to, an operating system, one or more application programs, other program modules, and program data, examples of which are Each or some combination of these may include implementations of network environments.

A bus may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus structures.

The electronic device can also communicate with one or more external devices (such as keyboards, pointing devices, Bluetooth devices, etc.), and can also communicate with one or more devices that enable the user to interact with the electronic device, and/or communicate with the electronic A device communicates with any device (eg, router, modem, etc.) that is capable of communicating with one or more other computing devices.

Such communication may occur through input/output (I/O) interfaces. Moreover, the electronic device can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN) and/or a public network such as the Internet) through a network adapter. The network adapter communicates with other modules of the electronic device through the bus. It should be understood that although not shown herein, other hardware and/or software modules may be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage system, etc.

The detection/discrimination value of the system or method of the present invention can be judged by, for example, calculating the evaluation indicators such as the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. AUC is also called the area under the receiver operating characteristic curve, which is defined as the area enclosed by the ROC curve and the coordinate axis, and the value range of the area is between 0.5 and 1. The closer the AUC is to 1.0, the higher the authenticity of the detection method.

Those skilled in the art will understand that as long as the purpose of the present invention can be achieved, other steps or operations can also be included before and after the above steps a, b and (1)-(3), or between the steps, such as further optimization and/or improvement The method described in the present invention.

Example 1

1. Experimental design

Collect the general information, medical history data, tacrolimus detailed medication records, tacrolimus drug concentration monitoring, laboratory examination, and genetic testing data of the patients with myasthenia gravis, and use statistical analysis methods to build a model to identify patients with myasthenia gravis. Influencing factors of inter-individual differences in tacrolimus plasma concentration, and a predictive model for determining the risk of early tacrolimus hypoconcentration.

2. Research object

(1) Meet the following inclusion criteria: Myasthenia gravis patients registered in Xuanwu Hospital's myasthenia gravis clinical cohort (Ministry of Science and Technology National Key Research and Development Program - Precision Medicine Research - Nervous System Disease Special Disease Cohort - Myasthenia Gravis Disease Cohort); Diagnosis criteria for myasthenia (refer to "China Guidelines for the Diagnosis and Treatment of Myasthenia Gravis 2020"); receive oral tacrolimus therapy; complete tacrolimus drug concentration monitoring and genetic testing; patients and their families sign informed consent.

(2) According to the time of joining the study, patients were divided into modeling cohort and external validation cohort: patients who joined this study between July 1, 2017 and December 31, 2019 were the modeling cohort, and a total of 93 patients (the required sample size for cross-sectional study sample size calculation is 83 cases); patients who joined this study between January 1, 2020 and October 1, 2020 are external validation cohorts, with a total of 36 included Patients (the sample size required for the calculation of the sample size of the diagnostic study is 36 cases).

3. Data collection and grouping basis

(1) Tacrolimus regimen: the initial dose is 2 mg/day, and the daily dose is divided into two doses (one in the morning and one in the evening, taken on an empty stomach or 1 hour before meals or 2-3 hours after meals); if there is no obvious For symptoms of discomfort, increase the dose to 3-4mg/day in the 2nd to 3rd week; draw peripheral venous blood in the 3rd to 4th week for blood drug concentration, blood routine, and blood biochemical tests. In the 4th week after taking the medicine, the clinician adjusted the dosage according to the degree of improvement of myasthenia gravis, drug tolerance of tacrolimus, blood drug concentration and related laboratory results. The target range of tacrolimus blood concentration is 4.8-10ng/ml, increase the dose of tacrolimus or give five ester capsules to increase the level of TAC blood concentration, the maximum daily dose should not exceed 5mg.

(2) Tacrolimus drug concentration monitoring: The first blood drug concentration detection is 3-4 weeks after taking the drug, and the blood drug concentration is detected by microparticle enzyme-linked immunoassay; record the single-time drug concentration between 1-3 months after taking the drug. Blood drug concentration measurement value, if there are multiple measurement values, select the first measurement value (the first measurement value more accurately reflects the concentration level in the early stage of treatment, which is conducive to analyzing the influencing factors of early concentration); the target concentration range of blood drug concentration is 4.8 -10ng/ml, less than 4.8ng/ml may cause treatment ineffectiveness, with 4.8ng/ml as the cut-off point, patients are divided into two concentration groups: low concentration group (<4.8ng/ml) and effective concentration group (≥ 4.8ng/ml), as a dichotomous dependent variable for statistical analysis.

(3) Selection and detection of gene loci: According to the evidence level of Pharmacogenomics Knowledge Base (PharmGKB) and the guidance of Clinical Pharmacogenetics Implementation Consortium (CPIC), the genes and single nucleotides that affect the blood concentration of tacrolimus were selected Mutation sites, a total of 6 candidate genes related to tacrolimus metabolism were included, including CYP3A5, CYP3A4, ABCB1, POR, CYP2C19 and NR1L2 genes. DNA was extracted from the patient's whole blood by the salting-out method, and NovaSeq6000 sequencer (Illumina Inc.) was used for next-generation sequencing. Bioinformatics analysis obtained information on single nucleotide mutation sites of 120 genes available for analysis. The Hardy-Weinberg balance test was used to analyze the genetic balance of polymorphisms at a single nucleotide locus to test whether the genotype frequency was representative of the population; the Haploview software was used to analyze the linkage disequilibrium between different loci, and finally 14 possible polymorphisms were obtained. Single nucleotide site mutations for statistical analysis.

(4) Collection of clinical data: clinical data such as demographic characteristics, clinical data and auxiliary examination results were collected from the myasthenia gravis clinical cohort database, including medication regimen, tacrolimus dose, gender, age, height, weight, body mass Index, alanine aminotransferase, glutamic acid aminotransferase, blood urea nitrogen, creatinine, hematocrit, combined medications, combined diseases (diabetes, hypertension), a total of 17 clinically relevant factors were included.

4. Statistical analysis to build predictive models

(1) Univariate analysis of screening variables: use independent sample t-test (normally distributed continuous variables), rank sum test (non-normally distributed continuous variables) or chi-square test (categorical variables) for demographic characteristics, clinical data And the inter-group comparison of single nucleotide site mutation (low concentration group vs effective concentration group), the variables with P value ≤ 0.10 or clinically considered strong correlation were included in the multivariate analysis. Five clinical factors (age, use of glucocorticoids, use of Wuzhi capsules, hematocrit, dose of tacrolimus) and 3 single nucleotide mutation sites [rs776746 locus dominant genetic model were obtained in the final screening. (T/T, T/C vs C/C), rs2242480 locus recessive genetic model (C/C vs C/T, T/T), rs1045642 locus recessive genetic model (A/A vs A/G ,G/G)] into multivariate analysis;

(2) Multi-factor analysis to build a model

a. Using stepwise regression analysis, the odds ratio and its 95% confidence interval are used to indicate the strength of correlation between clinical or genetic factors and tacrolimus blood concentration <4.8ng/ml (ie, low concentration group), and 5 results were obtained from the analysis The independent risk factors in the low-concentration group included age, tacrolimus dose, whether to use Wuzhi capsules, rs776746 locus genotype and rs1045642 locus genotype; hematocrit had no statistical difference (P value = 0.068), but was incorporated into the final model of stepwise regression. The final tacrolimus low concentration risk prediction model is Logit (P low concentration) = 10.023-0.047 × (age) - 1.263 × (TAC dose) - 4.325 × (whether to use five ester capsules) + 3.039 × (rs776746 gene type)-2.111×(rs1045642 genotype)-0.117×(hematocrit). In the formula, age is a continuous variable, taking an integer of age, and the unit is years. The dose of TAC is a continuous variable, and the actual dose is taken as the unit of mg. Whether to use Wuzhi capsules is a categorical variable, the value is 0 if Wuzhi capsules are not used, and 1 is used for Wuzhi capsules. The genotype of rs776746 is a categorical variable, the value of C/C genotype is 0, and the value of T/T or T/C genotype is 1. The genotype of rs1045642 is a categorical variable, the value of A/G or G/G genotype is 0, and the value of A/A genotype is 1. Hematocrit is a continuous variable without units, and the value is hematocrit × 100.

Prediction function of the model: According to the conversion according to the following formula, the predicted probability value of the low concentration of tacrolimus can be obtained:

b. Using binary logistic regression, age, tacrolimus dose, whether to use Wuzhi capsule, rs776746 locus genotype and rs1045642 locus genotype, hematocrit as independent variables, odds ratio and its 95% confidence The interval indicates the strength of the correlation between clinical or genetic factors and the blood concentration of tacrolimus <4.8ng/ml (i.e. low concentration group), and the prediction model of tacrolimus concentration is the same as the result of stepwise regression; R( R3.6.0) software rms program package to establish a nomogram prediction model, as shown in Figure 1.

c. Use discrimination and calibration to evaluate the predictive performance of the prediction model. The discrimination evaluation index is the area under the receiver operating characteristic curve and its 95% confidence interval, and the calibration evaluation index is the goodness of fit test.

As shown in Figure 2, the prediction model has good discrimination [area under the receiver operating characteristic curve and its 95% confidence interval = 0.877 (0.809-0.945)] (shown by the dark curve in Figure 2) and calibration [ Test χ2=2.252, P-value=0.972];

The nomogram prediction model has a good degree of discrimination [the area under the receiver operating characteristic curve and its 95% confidence interval = 0.877 (0.810-0.945)] and calibration (the predicted probability in the calibration curve is close to the actual probability);

d. Before patients with myasthenia gravis receive tacrolimus treatment, obtain age, rs776746 genotype, and rs1045642 genotype through clinical data collection, genetic testing, and laboratory tests. Clinicians select the dose of tacrolimus and whether to use it For Wuzhi Capsules, the predicted probability value of low concentration is obtained through the prediction model or nomogram prediction model; when the low concentration probability value is greater than 50%, it indicates that the patient has a high risk of low tacrolimus concentration, and it is necessary to increase the tacrolimus concentration appropriately. The dose of tacrolimus or the addition of Wuzhi capsules; when the low concentration probability value is less than 50%, it indicates that the risk of low tacrolimus concentration in this patient is low, and the selected dose can be administered.

For ease of understanding, some examples are given to illustrate:

Patient A, aged 40 years, with a TAC dose of 2 mg, did not use Wuzhi capsule (value 0), rs776746 genotype was C/C (value 0), rs1045642 genotype was A/A (value 1), red blood cell The backlog number is 0.45 (value 45), then Logit (P low concentration) = 10.023-0.047×(40)-1.263×(2)-4.325×(0)+3.039×(0)-2.111×(1)- 0.117×(45), the result is -1.705; after formula conversion, the predicted probability value of low tacrolimus concentration is 0.15381, suggesting that the risk of low tacrolimus concentration is low, and the selected dose can be administered.

Patient B, aged 60 years, TAC dose of 2 mg, did not use Wuzhi capsule (value 0), rs776746 genotype was T/T (value 1), rs1045642 genotype was G/G (value 0), red blood cell The backlog number is 0.45 (value 45), then Logit (P low concentration) = 10.023-0.047× (60)-1.263×(2)-4.325×(0)+3.039×(1)-2.111×(0)- 0.117×(45), the result is 2.505; after formula conversion, the predicted probability value of low concentration of tacrolimus is 0.92449, suggesting that the risk of low concentration of tacrolimus is high; according to the predicted result, the dose of tacrolimus is increased If the dose is 3 mg (the final predicted probability is 0.78), there is still a high risk of low concentration; or if the dose of tacrolimus is maintained at 2 mg and Wuzhi capsules are added at the same time (the final predicted probability is 0.139), the risk of low concentration is low.

(3) Internal verification of the model

a. Internal cross-validation of the prediction model: Divide the modeling population into a training set and a verification set at a ratio of 2:1. The training set is used to establish the prediction model, and the internal validity of the prediction model is evaluated in the verification set. Using the simple cross-validation method, repeated 200 random groups, calculated the area under the receiver operating characteristic curve and its 95% confidence interval of the 200-time validation set = 0.799 (0.670-0.908), suggesting that the low concentration prediction model is in the modeling cohort Has good internal validity and repeatability;

b. Internal verification of the nomogram prediction model: using the Bootstrap resampling method, the area under the receiver operating characteristic curve was calculated and = 0.844, suggesting that the low concentration prediction model has good internal validity and repeatability in the modeling cohort sex.

(4) External queue validation of the model:

a. The relevant data of 36 patients in the verification cohort were substituted into the calculation formula of the low concentration risk prediction model, and the results showed that the predicted risk of low concentration in 14 patients was greater than 50%, and the predicted risk of 22 patients was less than or equal to 50%; Result comparison, the prediction model successfully judged the area under the receiver operating characteristic curve of the low concentration group and its 95% confidence interval=0.855 (0.708-1) (as shown in the gray curve of Figure 2), and the sensitivity and specificity were respectively 82.4% and 100%, the positive predictive rate is 100%, and the negative predictive rate is 86.4%, suggesting that the low concentration prediction model has good external validity and is universal to external data;

b. The relevant data of 36 patients in the verification cohort were substituted into the low concentration risk nomogram prediction model, and the results showed that the predicted risk of low concentration in 15 patients was greater than 50%, and the predicted risk of 21 patients was less than or equal to 50%; Compared with the results, the prediction model successfully judged the area under the receiver operating characteristic curve of the low concentration group and its 95% confidence interval = 0.854 (0.712-1), the sensitivity and specificity were 82.4% and 94.7% respectively, and the positive predictive rate The negative predictive rate was 93.3%, and the negative predictive rate was 85.7%, suggesting that the low-concentration prediction model has good external validity and is universal to external data.

In some optional embodiments, the recommended initial dose of TAC can also be calculated according to the preset or desired target tacrolimus low concentration probability value. In particular, the calculation of the initial recommended dose of TAC based on the expected target low tacrolimus concentration probability value can greatly save the lengthy monitoring process of tacrolimus in the early stage, help to improve the proportion of early blood drug concentrations in patients with myasthenia gravis, and improve early curative effect.

In particular, the prediction method provided by the present invention is to correlate the characteristic data of patients with myasthenia gravis with each other, so as to construct a prediction system for predicting or guiding the initial dose selection and early dose adjustment of tacrolimus in patients with myasthenia gravis The technical thinking can also be used to guide the selection and adjustment of the initial dose of corresponding drugs for the treatment of other diseases, or it can also be used to guide the selection and adjustment of the initial dose of tacrolimus in the treatment of other autoimmune diseases.

It should be noted that the above is only the optimal embodiment provided by the present invention. Based on the concept of the present invention, the probability of low concentration of tacrolimus in patients with myasthenia gravis is calculated in association with characteristic data, so as to adjust the drug dosage and whether Wuzhi capsules were used, wherein the characteristic data included age, single nucleotide mutation sites of candidate genes related to tacrolimus metabolism, selected tacrolimus dose, whether Wuzhi capsules were used and/or hematocrit One or more of them, and any other calculation models obtained should fall within the protection scope of the present invention. On the basis of the calculation model disclosed in the present invention, other optional implementations obtained by adjusting parameters, coefficients, etc., do not deviate from the inventive concept of the present invention, and should also fall within the protection scope of the present invention.

It should be noted that the above-mentioned specific embodiments are exemplary, and those skilled in the art can come up with various solutions inspired by the disclosure of the present invention, and these solutions also belong to the scope of the disclosure of the present invention and fall within the scope of this disclosure. within the scope of protection of the invention. Those skilled in the art should understand that the description and drawings of the present invention are illustrative rather than limiting to the claims. The protection scope of the present invention is defined by the claims and their equivalents. The description of the present invention contains a number of inventive concepts, such as "preferably", "according to a preferred embodiment" or "optionally" all indicate that the corresponding paragraph discloses an independent concept, and the applicant reserves the right to propose a division based on each inventive concept right to apply.

Claims (15)

1. A system for accurately selecting the dose of tacrolimus for patients with myasthenia gravis, characterized in that it includes:

   A data acquisition unit configured to acquire at least one characteristic data of the myasthenia gravis patient;

   The predicting unit is configured to determine the predictive probability value of the low tacrolimus concentration of the myasthenia gravis patient based on at least one feature data input;

   Wherein, the characteristic data include age, single nucleotide mutation site of candidate genes related to tacrolimus metabolism, selected tacrolimus dosage, whether to use five ester capsules and one or more of hematocrit .
2. The system for accurately selecting the dose of tacrolimus for patients with myasthenia gravis according to claim 1, wherein the predicted probability value of the low concentration of tacrolimus is based on the principle of logistic regression according to the prediction model Logit(P low concentration ) are calculated in a related way.
3. The system for accurately selecting tacrolimus treatment doses for patients with myasthenia gravis according to claim 2, wherein the predictive model Logit (P low concentration) is at least related to age, tacrolimus dose and at least two Candidate genes involved in tacrolimus metabolism were identified in association.
4. The system for accurately selecting the dose of tacrolimus for patients with myasthenia gravis according to claim 2, characterized in that when the low concentration probability value is greater than a preset threshold, it is judged that tacrolimus for patients with myasthenia gravis The risk of low concentration is high, output instructions to increase the dose of tacrolimus or add five ester capsules;

   When the low concentration probability value is less than the preset threshold value, it is judged that the risk of low tacrolimus concentration in the myasthenia gravis patient is low, and an instruction is output according to the selected dosage;

   Preferably, the preset threshold is 50%, or 40%, or 30%, or 20%, or 10%, or any value between 0 and 50%.
5. The system for accurately selecting the therapeutic dose of tacrolimus for patients with myasthenia gravis according to claim 2, wherein the low concentration of tacrolimus refers to the blood concentration of tacrolimus for patients with myasthenia gravis Less than the concentration threshold, preferably, the concentration threshold is 4.8 ng/ml.
6. The system for accurately selecting the dose of tacrolimus for patients with myasthenia gravis according to claim 1, wherein the candidate genes include CYP3A5 and ABCB1 genes.
7. The system for accurately selecting the dose of tacrolimus for patients with myasthenia gravis according to claim 1, wherein the single nucleotide mutation site of the candidate gene includes the rs776746 site dominant genetic model and the rs1045642 site recessive genetic model.
8. The system for accurately selecting the dose of tacrolimus for patients with myasthenia gravis according to claim 1, wherein the prediction model is constructed based on the variable through multivariate analysis, wherein the multivariate analysis includes stepwise Regression analysis or binary logistic regression analysis.
9. A system for precise selection of tacrolimus treatment dose, characterized in that it includes:

   A data acquisition unit configured to acquire at least one characteristic data of the patient;

   A predicting unit configured to, in response to at least one item of characteristic data input, determine a low tacrolimus concentration prediction probability value related to the characteristic data;

   Wherein, the feature data includes age, single nucleotide mutation site of candidate genes related to tacrolimus metabolism, selected tacrolimus dose, whether to use five ester capsules and/or hematocrit.
10. The system for precise selection of tacrolimus therapeutic dose according to claim 10, wherein the characteristic data include age, single nucleotide mutation sites of candidate genes related to tacrolimus metabolism, selected other Crolimus dose, whether to use five ester capsules and at least two or more of hematocrit.
11. The system for accurately selecting tacrolimus therapeutic dose according to claim 10, wherein said determining the predicted probability value of low tacrolimus concentration related to said characteristic data comprises determining the low concentration value of tacrolimus Logit(P low concentration), wherein the tacrolimus low concentration value Logit(P low concentration) is at least associated with age, tacrolimus dose and at least two candidate genes related to tacrolimus metabolism ground to determine.
12. The system for accurately selecting the tacrolimus treatment dose according to claim 11, wherein the predicted probability value of the low tacrolimus concentration is calculated according to the logit( P low concentration) was calculated in a related way.
13. An electronic device, characterized in that, comprising,

    one or more processors;

    memory for storing one or more computer programs;

    When the one or more computer programs are executed by the one or more processors, the one or more processors implement the step of using a predictive model to obtain a predicted probability value of tacrolimus low concentration: including the At least one characteristic data of the myasthenia gravis patient is input into the system according to any one of the preceding claims to obtain the predicted probability value of low concentration of tacrolimus.
14. A storage medium containing computer-executable instructions, characterized in that, when the computer-executable instructions are executed by a computer processor, the step of using a prediction model to obtain a predicted probability value of tacrolimus low concentration: comprising converting the At least one characteristic data of the myasthenia gravis patient is input into the system according to any one of the preceding claims to obtain the predicted probability value of low tacrolimus concentration.
15. The system for accurately selecting the therapeutic dose of tacrolimus for patients with myasthenia gravis according to any one of claims 1 to 9, or the system for accurately selecting the therapeutic dose of tacrolimus according to any one of claims 10 to 12 Application in predicting the therapeutic dose of tacrolimus in patients with myasthenia gravis.

Images