Evaluation methods and systems for predicting the safety of EGFR TKIs monotherapy in NSCLC

Abstract

The application discloses an evaluation method for predicting the safety of EGFR TKIs monotherapy for NSCLC. The method comprises the following steps: obtaining the data of a plurality of patients receiving EGFR TKIs treatment; taking MDRAE (maximum grade of drug-related adverse event) as a safety index, constructing a safety evaluation final model, and the final model is an ordered logistic regression model, the covariates of the final model include normalized steady-state trough concentration, EGFR TKIs treatment history, gender, baseline glutamyltransferase, baseline uric acid, baseline creatine kinase, baseline platelets and baseline lymphocytes; obtaining the data of a target patient, inputting the numerical values of each covariate in the data of the target patient into the final model; determining the first probability of MDRAE of the target patient being grade 3 and above, the second probability of MDRAE being grade 2 and the third probability of MDRAE being grade 1 and below according to the final model, and taking the grade corresponding to the maximum probability in the three probabilities as the predicted grade of MDRAE of the target patient. The application improves the accuracy of predicting the safety of EGFR TKIs monotherapy for NSCLC.

Description

Technical Field

[0001] This invention relates to the field of medication safety technology, and in particular to an assessment method, system, electronic device, and computer-readable storage medium for predicting the safety of EGFR TKIs monotherapy for NSCLC. Background Technology

[0002] Lung cancer is a leading cause of cancer-related deaths, with non-small cell lung cancer (NSCLC) accounting for approximately 85%. Epidermal growth factor receptor (EGFR) is one of the key targets for NSCLC treatment, with a mutation rate of up to 50% in the Chinese population. Currently, several EGFR tyrosine kinase inhibitors (TKIs) are used for first-line treatment of NSCLC.

[0003] However, in clinical practice, the efficacy of EGFR TKIs varies greatly among individuals, primarily due to significant variability in drug exposure levels and adverse events. Clinical trials have shown that over 90% of patients treated with EGFR TKIs experienced drug-related adverse events (DRAEs), with nearly 50% experiencing grade 3 or higher adverse events, severely impacting their quality of life. Previous studies have indicated that higher drug exposure may increase the risk of adverse events. Furthermore, there are significant inter-individual differences in EGFR TKI exposure levels; for example, the area under the concentration-time curves (AUC) of gefitinib can vary by up to 19-fold among different patients, potentially leading to substantial differences in the risk of adverse events. Therefore, a quantitative assessment of the relationship between EGFR TKI drug exposure and adverse events is urgently needed.

[0004] Current technologies are limited to qualitative analysis of the relationship between drug exposure and adverse events, lacking quantitative evidence to clearly explain this relationship. Furthermore, existing technologies treat drug exposure as a single variable to examine its impact on DRAEs, without considering other influencing factors, and typically treat adverse events as binary variables (e.g., whether they occurred or their severity), failing to meet clinical needs for assessing medication safety. Summary of the Invention

[0005] In view of the above problems, the present invention is proposed to provide a method, system, electronic device, and computer-readable storage medium for drug safety assessment based on an ordered logistic regression model that overcomes or at least partially solves the above problems, for the assessment of predicting the safety of EGFR TKIs monotherapy for NSCLC.

[0006] One embodiment of the present invention provides an assessment method for predicting the safety of EGFR TKIs monotherapy in NSCLC, the method comprising:

[0007] Data were obtained from multiple patients who received EGFR TKIs treatment, including drug exposure indicators, demographic information, pathophysiological characteristics, treatment information, hematological data, and MDRAE data.

[0008] Using MDRAE as a safety indicator, a final safety assessment model was constructed. The final model is an ordered logistic regression model. The covariates of the final model include: normalized steady-state trough concentration, EGFR TKI treatment history, gender, baseline gamma-glutamyl transferase, baseline uric acid, baseline creatine kinase, baseline platelet count, and baseline lymphocyte count.

[0009] Acquire target patient data and input the values ​​of each covariate in the target patient data into the final model;

[0010] Based on the final model, the first probability of the target patient being MDRAE level 3 or above, the second probability of MDRAE level 2, and the third probability of MDRAE level 1 or below are determined. The level corresponding to the maximum probability among the first, second, and third probabilities is taken as the predicted MDRAE level of the target patient.

[0011] Optionally, the step of using MDRAE as a security indicator to construct a final security assessment model includes:

[0012] Construct a univariate ordered logistic regression model to examine the impact of each variable on MDRAE;

[0013] A stepwise selection method was used to screen covariates from multiple variables, and the simplest multivariate ordered logistic regression model was selected as the final model for safety assessment, provided that the model evaluation criteria were met.

[0014] Optionally, the model evaluation criteria include the reduction of AIC value during iteration, the reasonableness of parameter estimates, the collinearity of covariates, and the model's parallelism assumptions and fitting performance.

[0015] Feasibly, the final model is as follows:

[0016]

[0017] P MDRAE≤1 =1-P MDRAE＝2 -P MDRAE≥3

[0018] Among them, P MDRAE≥3 For the first probability that MDRAE is level 3 or higher, P MDRAE＝2 For the second probability of MDRAE level 2, P MDRAE≤1 For MDRAE, the third probability is level 1 or lower, α1 and α2 are the intercept terms, and X... k For the covariate value, β k For covariate X k The regression coefficients are given by K, where K is the total number of covariates.

[0019] Optionally, before constructing the univariate ordered logistic regression model, the method further includes:

[0020] Based on the exposure levels of each drug, the IC50 of wild-type EGFR was evaluated. 50 Normalization is performed on the IC. 50 Drug exposure is defined as the drug exposure rate when the drug-target binding inhibition rate reaches 50%.

[0021] Optionally, the IC50 of wild-type EGFR can be calculated using the following formula based on the exposure level of each drug. 50 Normalization is performed:

[0022]

[0023] Where TI is the target inhibition rate, and C DRUG / IC 50 This represents the normalized exposure level.

[0024] Optionally, the method further includes: generating a nomogram with the values ​​of each covariate on the horizontal axis and each covariate on the vertical axis, and determining the MDRAE prediction level of the target patient based on the nomogram.

[0025] Another embodiment of the present invention provides an assessment system for predicting the safety of EGFR TKIs monotherapy for NSCLC, the system comprising:

[0026] The patient data acquisition unit is used to acquire data from multiple patients receiving EGFR TKIs treatment. The patient data includes drug exposure indicators, demographic information, pathophysiological characteristics, treatment information, hematological data, and MDRAE data.

[0027] The final model construction unit is used to construct a safety assessment final model using MDRAE as a safety indicator. The final model is an ordered logistic regression model. The covariates of the final model include: normalized steady-state trough concentration, EGFR TKI treatment history, gender, baseline gamma-glutamyl transferase, baseline uric acid, baseline creatine kinase, baseline platelets, and baseline lymphocytes.

[0028] A numerical input unit is used to acquire target patient data and input the values ​​of each covariate in the target patient data into the final model.

[0029] The prediction unit is used to determine, based on the final model, the first probability that the target patient's MDRAE is level 3 or above, the second probability that the MDRAE is level 2, and the third probability that the MDRAE is level 1 or below, and to use the level corresponding to the maximum probability among the first probability, the second probability, and the third probability as the predicted MDRAE level of the target patient.

[0030] Optionally, the final model building unit is further used for:

[0031] Construct a univariate ordered logistic regression model to examine the impact of each variable on MDRAE;

[0032] A stepwise selection method was used to screen covariates from multiple variables, and the simplest multivariate ordered logistic regression model was selected as the final model for safety assessment, provided that the model evaluation criteria were met.

[0033] Optionally, the model evaluation criteria include the reduction of AIC value during iteration, the reasonableness of parameter estimates, the collinearity of covariates, and the model's parallelism assumptions and fitting performance.

[0034] Feasibly, the final model is as follows:

[0035]

[0036]

[0037] P MDRAE≤1 =1-P MDRAE＝2 -P MDRAE≥3

[0038] Among them, P MDRAE≥3 For the first probability that MDRAE is level 3 or higher, P MDRAE＝2 For the second probability of MDRAE level 2, P MDRAE≤1 For MDRAE, the third probability is level 1 or lower, α1 and α2 are the intercept terms, and X... k For the covariate value, β k For covariate X k The regression coefficients are given by K, where K is the total number of covariates.

[0039] Optionally, the system further includes a normalization processing unit, used to normalize the IC50 of wild-type EGFR according to each drug exposure level before constructing the univariate ordered logistic regression model. 50 Normalization is performed on the IC. 50 Drug exposure is defined as the drug exposure rate when the drug-target binding inhibition rate reaches 50%.

[0040] Optionally, the normalization unit uses the following formula to adjust the IC50 of wild-type EGFR according to each drug exposure level. 50 Normalization is performed:

[0041]

[0042] Where TI is the target inhibition rate, and C DRUG / IC 50 This represents the normalized exposure level.

[0043] Optionally, the system further includes a column chart generation unit, the column chart generation unit being used for:

[0044] A nomogram is generated with the values ​​of each covariate on the horizontal axis and each covariate on the vertical axis. The MDRAE prediction level of the target patient is determined based on the nomogram.

[0045] Another embodiment of the present invention provides an electronic device, wherein the electronic device includes:

[0046] Processor; and,

[0047] A memory is configured to store computer-executable instructions, which, when executed, cause the processor to perform the methods described above.

[0048] Another embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs that, when executed by a processor, implement the method described above.

[0049] The beneficial effect of this invention is that by establishing an ordered logistic regression model, this invention can quantitatively assess the impact of EGFR TKIs exposure and other covariates on the probability of adverse events of different severity, effectively improving the accuracy of predicting the safety of EGFR TKIs monotherapy for NSCLC, thereby enabling clinical assessment of drug safety.

[0050] In addition, this invention provides a convenient quantitative basis for optimizing clinical medication strategies by generating nomograms that are easy to apply in clinical practice. Attached Figure Description

[0051] Figure 1 This is a flowchart illustrating an embodiment of the present invention for assessing the safety of EGFR TKIs monotherapy in NSCLC.

[0052] Figure 2 This is a nomogram predicting the safety of EGFR TKIs monotherapy for NSCLC according to an embodiment of the present invention;

[0053] Figure 3A This is a calibration curve of the final model for predicting the probability of MDRAE of level 2 or above according to an embodiment of the present invention.

[0054] Figure 3B This is a calibration curve of the final model for predicting the probability of MDRAE of level 3 or above according to an embodiment of the present invention.

[0055] Figure 4 This is a schematic diagram of the structure of an assessment system for predicting the safety of EGFR TKIs monotherapy in NSCLC according to an embodiment of the present invention;

[0056] Figure 5 A schematic diagram of the structure of an electronic device according to an embodiment of the present invention is shown;

[0057] Figure 6 A schematic diagram of the structure of a computer-readable storage medium according to an embodiment of the present invention is shown. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0059] Figure 1 This is a flowchart illustrating an embodiment of the present invention for assessing the safety of EGFR TKIs monotherapy in NSCLC. Figure 1 As shown, the method includes:

[0060] S101: Obtain data from multiple patients receiving EGFR TKIs treatment, including drug exposure indicators, demographic information, pathophysiological characteristics, treatment information, hematological data, and MDRAE data;

[0061] Specifically, drug exposure indicators include steady-state trough concentration (C). MIN,SS ), mean steady-state plasma concentration (C AVG,SS ), steady-state maximum concentration (C)MAX,SS ) and steady-state AUC SS The data includes demographic information such as gender, age, and weight; pathophysiological characteristics such as Eastern Cooperative Oncology Group (ECOG) score, tumor type, and disease stage; treatment information such as treatment drugs, regimens, and previous treatment history; and hematological data such as complete blood count and blood biochemistry.

[0062] MDRAE refers to the maximum grade of drug-related adverse event. In practice, MDRAE is divided into grades 0-4.

[0063] This invention provides data from 277 patients treated with EGFR TKIs, including 69, 70, 44, and 94 patients treated with gefitinib, icotinib, afatinib, and osimertinib, respectively. The number of patients with MDRAE grades 0 and 4 was relatively small, at 18 and 19 respectively, representing 6.5% and 6.9% of the total population. Therefore, patients were divided into three groups based on MDRAE grades: 0 / 1, 2, and 3 / 4.

[0064] S102: Using MDRAE as a safety indicator, a final safety assessment model was constructed. The final model is an ordered logistic regression model. The covariates of the final model include: normalized steady-state trough concentration, EGFR TKI treatment history, gender, baseline gamma-glutamyl transferase, baseline uric acid, baseline creatine kinase, baseline platelet count, and baseline lymphocyte count.

[0065] S103: Obtain target patient data and input the values ​​of each covariate in the target patient data into the final model;

[0066] S104: Based on the final model, determine the first probability that the target patient's MDRAE is level 3 or above, the second probability that the MDRAE is level 2, and the third probability that the MDRAE is level 1 or below, and take the level corresponding to the maximum probability among the first probability, the second probability, and the third probability as the predicted MDRAE level of the target patient.

[0067] The method for predicting the safety of EGFR TKIs monotherapy in NSCLC according to embodiments of the present invention establishes an ordered logistic regression model to quantitatively assess the impact of EGFR TKIs exposure and other covariates on the probability of adverse events of different severity, effectively improving the accuracy of predicting the safety of EGFR TKIs monotherapy in NSCLC, thereby enabling clinical assessment of drug safety.

[0068] Preferably, before constructing the univariate ordered logistic regression model, the method further includes:

[0069] Based on the exposure levels of each drug, the IC50 of wild-type EGFR was evaluated. 50 Normalization was performed to merge and analyze the data from each EGFR TKI. 50 Drug exposure is defined as the drug exposure rate when the drug-target binding inhibition rate reaches 50%.

[0070] Specifically, the IC50 of wild-type EGFR was calculated using the following formula based on the exposure levels of each drug. 50 Normalization is performed:

[0071]

[0072] Where TI is the target inhibition rate, and C DRUG / IC 50 This represents the normalized exposure level.

[0073] It should be noted that, since most current studies suggest that DRAE caused by EGFR TKIs is mainly related to their inhibition of wild-type EGFR, the embodiments of this invention adjust the IC50 of wild-type EGFR based on the exposure level of each EGFR TKI. 50 Normalization was performed to allow for the combined analysis of DRAE data from various EGFR TKIs. Specifically, taking the Emax equation to describe the inhibitory effect produced after drug binding to the target as an example, assuming the maximum inhibition rate of the drug on the target is 100%, the target inhibition rate (TI) is calculated according to the following formula:

[0074]

[0075] Among them, C DRUG IC50 represents the actual drug exposure. 50 Drug exposure is defined as the drug exposure rate when the drug-target binding inhibition rate reaches 50%.

[0076] Due to the difficulty in obtaining in vivo data, the IC50 of various EGFR TKIs against wild-type EGFR is... 50 The values ​​are all from in vitro experiments, as shown in Table 1.

[0077] Table 1. IC50 of EGFR TKIs against wild-type EGFR 50 Summary

[0078]

[0079] Transform the above formula by dividing both the numerator and denominator by IC. 50 Then we get the following formula:

[0080]

[0081] Among them, C DRUG / IC 50 This is the normalized exposure (C). NORM This conversion method ensures that, with the same normalized exposure level, different drugs targeting the same target have the same inhibition rate, thus enabling pooled analysis of different EGFRTKIs.

[0082] In an optional embodiment of the present invention, the step of constructing a final security assessment model using MDRAE as a security indicator includes:

[0083] Construct a univariate ordered logistic regression model to examine the impact of each variable on MDRAE;

[0084] A stepwise selection method was used to screen covariates from multiple variables, and the simplest multivariate ordered logistic regression model was selected as the final model for safety assessment, provided that the model evaluation criteria were met.

[0085] It is understood that the embodiments of the present invention have three dependent variables: 0 / 1, 2, and 3 / 4 levels of MDRAE, and there is a hierarchical relationship between the dependent variables. Therefore, an ordered logistic regression model is used to construct the final security assessment model.

[0086] This invention first constructs a univariate ordered logistic regression model to examine the impact of each variable on MDRAE. Then, variables from the univariate ordered logistic regression model that have statistically significant p-values ​​according to the Wald test, or variables that may have clinical relevance to MDRAE of EGFR TKIs, are added to the multivariate ordered logistic regression model. A stepwise selection method is used to screen covariates. If a variable retained in the model has no significant p-value according to the Wald test, it is successively removed, and the model's goodness of fit is evaluated. Under the premise that the model's goodness of fit meets the model evaluation criteria, the simplest multivariate model is selected as the final model.

[0087] In practical applications, the model evaluation criteria include the reduction of the AIC value (Akaike Information Criterion) during the iteration process, the reasonableness of the parameter estimates, the collinearity of covariates, and the model's parallelism assumptions and fit performance.

[0088] In this embodiment of the invention, the variance inflation factor (VIF) is used to assess the collinearity of covariates. The Brant test is used to test the parallelism hypothesis of the model. The Hosmer-Lemeshow goodness-of-fit test and the concordance index (C-index) are used to evaluate the model's fit performance, and the corresponding calibration curve of the Hosmer-Lemeshow test is plotted.

[0089] In this embodiment of the invention, the model is divided into three levels and above, and two levels and above, for comparison based on the value of the dependent variable MDRAE, and described by the following formulas respectively:

[0090]

[0091] Among them, logit(P MDRAE≥3 ) and logit(P MDRAE≥2 ) represent the predicted probabilities of MDRAE levels ≥ 3 and ≥ 2, respectively, after logit transformation, where α1 and α2 are the intercept terms, and X k For the covariate value, β k For covariate X k The regression coefficients are given by K, where K is the total number of covariates.

[0092] The final model can be derived as follows:

[0093]

[0094]

[0095] P MDRAE≤1 =1-P MDRAE＝2 -P MDRAE≥3

[0096] Among them, P MDRAE≥3 For the first probability that MDRAE is level 3 or higher, P MDRAE＝2 For the second probability of MDRAE level 2, P MDRAE≤1 For MDRAE, the third probability is level 1 or lower, α1 and α2 are the intercept terms, and X... k For the covariate value, β k For covariate Xk The regression coefficients are given by K, where K is the total number of covariates.

[0097] Among them, there are 8 covariates, specifically the normalized steady-state valley concentration C. MIN,SS (C NORM EGFR TKI treatment history (Pre-EGFR TKI treatment), sex, baseline gamma-glutamyltransferase (Base GGT), baseline uric acid (Base UA), baseline creatine kinase (Base CK), baseline platelet (Base PLT), and baseline lymphocyte (Base LYMPH).

[0098] Table 2 shows the parameter results of the final model of one embodiment of the present invention.

[0099] Table 2 Parameter results of the final model

[0100]

[0101]

[0102] Here, SE stands for standard error. The p-value is a probability value in statistical hypothesis testing used to determine whether a variable has a significant effect on the model. OR stands for odds ratio. CI stands for confidence interval.

[0103] As shown in Table 2, the first six covariates all had significant effects on the model. The latter two covariates, baseline platelet count and baseline lymphocyte count, had no significant effect on the model, but removing these two covariates worsened the model's fit; therefore, they were ultimately retained. The variance inflation factor (VIF) values ​​of the model were all less than 5, indicating a very low degree of collinearity among the variables.

[0104] The Brant test p-value of 0.98 indicates good model parallelism. The Hosmer-Lemeshow goodness-of-fit test p-value of 0.49 indicates good consistency between the predicted and actual probabilities.

[0105] Preferably, the method further includes: generating a nomogram with the values ​​of each covariate on the horizontal axis and each covariate on the vertical axis, and determining the MDRAE prediction level of the target patient based on the nomogram.

[0106] Understandably, such as Figure 2 As shown, embodiments of the present invention also provide convenient quantitative basis for optimizing clinical medication strategies by generating nomograms that are easy to apply in clinical practice.

[0107] Understandably, nomograms can present complex statistical prediction models in a concise and intuitive form, predicting the probability of events based on specific combinations of covariates for each patient, thus possessing good clinical applicability. Therefore, this invention, based on the final model results, generates corresponding nomograms, providing a simple and intuitive tool for potential clinical applications.

[0108] Figure 2 Rows 2–9 show the covariates included in the final model and their corresponding value ranges. For each covariate, a specific value is taken, and a perpendicular line is drawn from that scale to the Points axis in row 1; this represents the score corresponding to that variable value. For example, if a patient has no history of EGFR TKI treatment, the Pre-EGFR TKI treatment covariate is considered to be "yes" or "no," and its corresponding score is 9.7. The scores corresponding to each covariate value are summed to obtain the total score, shown in row 10 of the figure. Then, perpendicular lines are drawn from the total score scale to the axes for MDRAE ≥ 2 and MDRAE ≥ 3, respectively, to obtain the probability that the patient's MDRAE is 2 or higher and 3 or higher. This is then converted into the probability that the patient's MDRAE is 3 or higher, 2, or 1 or lower, respectively. The level corresponding to the highest probability is used as the patient's predicted MDRAE level. For example, for a patient, their C... NORM The value was 5.8, the sex was male, and the values ​​of Base GGT, Base UA, Base CK, Base PLT, and Base LYMPH were 36.5 U / L, 219 μmol / L, 81.0 ng / mL, and 318 × 10⁻⁶ U / L, respectively. 9 / L and 3.09×10 9 / L. The scores obtained from the values ​​of each variable for this patient are added together to get a total score of 49.0. The probability of MDRAE ≥ 2 is less than 0.5 and the probability of MDRAE ≥ 3 is less than 0.1. Therefore, the predicted MDRAE grade for this patient is 1 or below.

[0109] Figure 3A This is a calibration curve of the final model for predicting the probability of MDRAE of level 2 or above according to an embodiment of the present invention. Figure 3B This is a calibration curve of the final model for predicting the probability of MDRAE of level 3 or above according to an embodiment of the present invention.

[0110] Figure 3A and Figure 3BThe dashed line on the middle diagonal represents the ideal line, while the solid lines L1 and L2 represent the final model and the predicted probabilities obtained based on 1000 bootstrapping, respectively. Figure 3A and Figure 3B As can be seen, the predicted probability of the final model and the predicted probability of the MDRAE level obtained based on 1000 bootstrapping are similar to the observed probability, and the two curves are quite close to the ideal line. Furthermore, the consistency index of the final model is 0.71. Therefore, the model exhibits good fitting performance.

[0111] Figure 4 This is a schematic diagram of the structure of an assessment system for predicting the safety of EGFR TKIs monotherapy in NSCLC according to an embodiment of the present invention. Figure 4 As shown, this system is a computer system based on a specific computational model, including:

[0112] The patient data acquisition unit 401 is used to acquire data from multiple patients receiving EGFR TKIs treatment. The patient data includes drug exposure indicators, demographic information, pathophysiological characteristics, treatment information, hematological data, and MDRAE data.

[0113] The final model construction unit 402 is used to construct a safety assessment final model using MDRAE as a safety indicator. The final model is an ordered logistic regression model. The covariates of the final model include: normalized steady-state trough concentration, EGFR TKI treatment history, gender, baseline gamma-glutamyl transferase, baseline uric acid, baseline creatine kinase, baseline platelets, and baseline lymphocytes.

[0114] The numerical input unit 403 is used to acquire target patient data and input the values ​​of each covariate in the target patient data into the final model.

[0115] The prediction unit 404 is used to determine, based on the final model, the first probability that the target patient's MDRAE is level 3 or above, the second probability that the MDRAE is level 2, and the third probability that the MDRAE is level 1 or below, and to use the level corresponding to the maximum probability among the first probability, the second probability, and the third probability as the predicted MDRAE level of the target patient.

[0116] The evaluation system for predicting the safety of EGFR TKIs monotherapy in NSCLC, as described in this invention, establishes an ordered logistic regression model to quantitatively assess the impact of EGFR TKIs exposure and other covariates on the probability of adverse events of different severity. This effectively improves the accuracy of predicting the safety of EGFR TKIs monotherapy in NSCLC, thereby enabling clinical assessment of drug safety.

[0117] In an optional embodiment of the present invention, the final model construction unit 401 is further configured to:

[0118] Construct a univariate ordered logistic regression model to examine the impact of each variable on MDRAE;

[0119] A stepwise selection method was used to screen covariates from multiple variables, and the simplest multivariate ordered logistic regression model was selected as the final model for safety assessment, provided that the model evaluation criteria were met.

[0120] Furthermore, the model evaluation criteria include the reduction of AIC value during iteration, the reasonableness of parameter estimates, the collinearity of covariates, and the model's parallelism assumptions and fitting performance.

[0121] Specifically, the final model is as follows:

[0122]

[0123] P MDRAE≤1 =1-P MDRAE＝2 -P MDRAE≥3

[0124] Among them, P MDRAE≥3 For the first probability that MDRAE is level 3 or higher, P MDRAE＝2 For the second probability of MDRAE level 2, P MDRAE≤1 For MDRAE, the third probability is level 1 or lower, α1 and α2 are the intercept terms, and X... k For the covariate value, β k For covariate X k The regression coefficients are given by K, where K is the total number of covariates.

[0125] Furthermore, the system also includes a normalization processing unit, used to normalize the IC50 of wild-type EGFR according to each drug exposure level before constructing the univariate ordered logistic regression model. 50 Normalization is performed on the IC. 50 Drug exposure is defined as the drug exposure rate when the drug-target binding inhibition rate reaches 50%.

[0126] Specifically, the normalization processing unit uses the following formula to adjust the IC50 of wild-type EGFR according to each drug exposure level. 50 Normalization is performed:

[0127]

[0128] Where TI is the target inhibition rate, and C DRUG / IC 50 This represents the normalized exposure level.

[0129] Preferably, the system further includes a column chart generation unit, the column chart generation unit being used for:

[0130] A nomogram is generated with the values ​​of each covariate on the horizontal axis and each covariate on the vertical axis. The MDRAE prediction level of the target patient is determined based on the nomogram.

[0131] It should be noted that the evaluation system for predicting the safety of EGFR TKIs monotherapy for NSCLC in the above embodiments can be used to perform the methods in the above embodiments, and therefore will not be described in detail one by one.

[0132] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0133] It should be noted that:

[0134] The algorithms and displays provided herein are not inherently related to any particular computer, virtual device, or other equipment. Various general-purpose devices can also be used in conjunction with the teachings herein. The required structure for constructing such devices is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the invention.

[0135] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0136] Similarly, it should be understood that, in order to simplify the invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof. However, this disclosure should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the following claims, inventive aspects lie in fewer than all features of a single foregoing disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into this detailed description, wherein each claim itself is a separate embodiment of the invention.

[0137] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0138] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are intended to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments can be used in any combination.

[0139] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components in the apparatus according to embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0140] For example, Figure 5 A schematic diagram of an electronic device according to an embodiment of the present invention is shown. The electronic device conventionally includes a processor 51 and a memory 52 arranged to store computer-executable instructions (program code). The memory 52 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Figure 1Storage space 53 for program code 54 of any method steps shown and in any of the embodiments. For example, storage space 53 for storing program code may include various program codes 54 for implementing the various steps in the methods above. This program code can be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, CDs, memory cards, or floppy disks. Such computer program products are typically, for example, Figure 6 The aforementioned computer-readable storage medium. This computer-readable storage medium may have the same characteristics as... Figure 5 The memory 52 in the electronic device is similarly arranged as a storage segment, storage space, etc. The program code can be compressed, for example, in a suitable form. Typically, the storage space stores program code 61 for performing the steps of the method according to the invention, that is, program code that can be read by a processor 51, which, when run by the electronic device, causes the electronic device to perform the various steps of the method described above.

[0141] The above description is merely a specific embodiment of the present invention. Under the teachings of the present invention, those skilled in the art can make other improvements or modifications based on the above embodiments. Those skilled in the art should understand that the above specific description is only to better explain the purpose of the present invention, and the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for predicting the safety of EGFR TKIs monotherapy in NSCLC, characterized in that, include: Data were obtained from multiple patients who received EGFR TKIs treatment, including drug exposure indicators, demographic information, pathophysiological characteristics, treatment information, hematological data, and MDRAE data. Using MDRAE as a safety indicator, a final safety assessment model was constructed. The final model is an ordered logistic regression model. The covariates of the final model include: normalized steady-state trough concentration, EGFR TKI treatment history, gender, baseline gamma-glutamyl transferase, baseline uric acid, baseline creatine kinase, baseline platelet count, and baseline lymphocyte count. Acquire target patient data and input the values ​​of each covariate in the target patient data into the final model; Based on the final model, the first probability of the target patient being MDRAE level 3 or above, the second probability of MDRAE level 2, and the third probability of MDRAE level 1 or below are determined. The level corresponding to the maximum probability among the first, second, and third probabilities is taken as the predicted MDRAE level of the target patient.

2. The method according to claim 1, characterized in that, The construction of a final security assessment model using MDRAE as a security indicator includes: Construct a univariate ordered logistic regression model to examine the impact of each variable on MDRAE; A stepwise selection method was used to screen covariates from multiple variables, and the simplest multivariate ordered logistic regression model was selected as the final model for safety assessment, provided that the model evaluation criteria were met.

3. The method according to claim 2, characterized in that, The model evaluation criteria include the reduction of AIC value during iteration, the reasonableness of parameter estimates, the collinearity of covariates, and the model's parallelism assumptions and fitting performance.

4. The method according to claim 2, characterized in that, The final model is as follows: P MDRAE≤1 =1-P MDRAE＝2 -P MDRAE≥3 Among them, P MDRAE≥3 For the first probability that MDRAE is level 3 or higher, P MDRAE＝2 For the second probability of MDRAE level 2, P MDRAE≤1 For MDRAE, the third probability is level 1 or lower, α1 and α2 are the intercept terms, and X... k For the covariate value, β k For covariate X k The regression coefficients are given by K, where K is the total number of covariates.

5. The method according to claim 2, characterized in that, Before constructing the univariate ordered logistic regression model, the method further includes: Based on the exposure levels of each drug, the IC50 of wild-type EGFR was evaluated. 50 Normalization is performed on the IC. 50 Drug exposure is defined as the drug exposure rate when the drug-target binding inhibition rate reaches 50%.

6. The method according to claim 5, characterized in that, The IC50 of wild-type EGFR was calculated using the following formula based on the exposure levels of each drug. 50 Normalization is performed: Where TI is the target inhibition rate, and C DRUG / IC 50 This represents the normalized exposure level.

7. The method according to claim 1, characterized in that, The method further includes: generating a nomogram with the values ​​of each covariate on the horizontal axis and each covariate on the vertical axis, and determining the MDRAE prediction level of the target patient based on the nomogram.

8. An assessment system for predicting the safety of EGFR TKIs monotherapy in NSCLC, characterized in that, include: The patient data acquisition unit is used to acquire data from multiple patients receiving EGFR TKIs treatment. The patient data includes drug exposure indicators, demographic information, pathophysiological characteristics, treatment information, hematological data, and MDRAE data. The final model construction unit is used to construct a safety assessment final model using MDRAE as a safety indicator. The final model is an ordered logistic regression model. The covariates of the final model include: normalized steady-state trough concentration, EGFR TKI treatment history, gender, baseline gamma-glutamyl transferase, baseline uric acid, baseline creatine kinase, baseline platelets, and baseline lymphocytes. A numerical input unit is used to acquire target patient data and input the values ​​of each covariate in the target patient data into the final model. The prediction unit is used to determine, based on the final model, the first probability that the target patient's MDRAE is level 3 or above, the second probability that the MDRAE is level 2, and the third probability that the MDRAE is level 1 or below, and to use the level corresponding to the maximum probability among the first probability, the second probability, and the third probability as the predicted MDRAE level of the target patient.

9. An electronic device, characterized in that, The electronic device includes: Processor; and, A memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs that, when executed by a processor, implement the method of any one of claims 1-7.

Images