Diagnostic kit based on prognosis and drug sensitivity prediction of acute leukemia

Abstract

The present application relates to the biomedical technology field, and more particularly to a diagnostic kit based on acute leukemia prognosis and drug sensitivity prediction, comprising prediction module I and prediction module II.Prediction module I is used for predicting the clinical survival rate of patients, and one or more of 32 molecules such as CCND3, FERMT3 and PLD4 is used as a prognostic marker; prediction module II is used for predicting the sensitivity of the body to drugs, and one or more molecules of BMP8B, IGF1R, OTULINL, SLC22A15, CERS1 and PDE4A are used as drug sensitivity markers.The present application breaks through the limitations of single biomarker prediction efficiency by constructing a double prediction model, realizes the functional coupling of survival prognosis and multiple drug sensitivity prediction, and provides an integrated tool for clinical transformation for AML precise treatment.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of biomedical technology, and particularly relates to a diagnostic kit based on acute leukemia prognosis and drug sensitivity prediction. BACKGROUND

[0002] Acute leukemia is a malignant clonal disease of hematopoietic stem / progenitor cells, mainly divided into acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML). AML is a highly heterogeneous hematopoietic malignancy, and the molecular characteristics, treatment response and prognosis of different patients are significantly different. The significant difference in patient prognosis is mainly due to the complexity of genomic and molecular lineage. FMS-like tyrosine kinase 3 (FLT3) is a receptor tyrosine kinase located on the surface of hematopoietic stem cells, which regulates cell proliferation, differentiation and survival in normal hematopoiesis. In AML, FLT3 gene mutation is one of the most common driver mutations, which is not only closely related to the occurrence and development of the disease, but also significantly affects the prognosis of patients and the selection of treatment strategies.

[0003] FLT3 inhibitors are key targeted drugs for FLT3 mutations in AML. Although FLT3 inhibitors (such as midostaurin, gilteritinib, and quizartinib) have become an important breakthrough in targeted therapy, there are still two major challenges in clinical practice: first, the initial response rate of FLT3 mutant patients to inhibitors is low, and most of the initial effective patients are prone to acquired drug resistance; second, the current risk stratification system based on the European Leukemia Network (ELN) standard mainly relies on static genetic markers, which is difficult to dynamically capture the clonal evolution trajectory of the disease and the molecular mechanism of drug resistance during treatment.

[0004] Therefore, it is urgent to deeply analyze the biological basis of the heterogeneity of FLT3 inhibitor treatment response in the clinical management of AML in order to achieve precise stratification of patient survival prognosis. Although there are currently biomarker explorations based on a single omics level, such as using FLT3 internal tandem duplication (ITD) mutation status or kinase domain (TKD) mutation in genomics data for prognosis evaluation and drug selection, such strategies have significant limitations:

[0005] Firstly, the multidimensional analysis of drug resistance mechanisms has not been systematized, especially lacking in-depth integration of protein function regulation. Although existing research has established drug resistance / prognosis models or drug sensitivity models using genomic or transcriptomic data, and can partially explain the phenotype of drug resistance, the research is usually limited to these transcriptomic markers / models, and fails to conduct systematic integration analysis at the functional level (for example, exploring the dynamic association, covariation rule and potential causal mechanism between drug resistance related protein expression changes and transcriptomic model predicted scores / molecular subtypes). However, a large amount of evidence shows that the core driving signal of acquired drug resistance often occurs at the protein function regulation level. The current limited analysis mode may lead to insufficient identification of key drug resistance driving targets, and it is difficult to effectively distinguish between driving functional changes and accompanying biological variations, limiting the depth and convertibility of mechanism analysis.

[0006] Secondly, the existing marker system has redundancy and clinical translation barriers. Survival prediction models usually rely on multi-gene combination risk scores, which have high detection costs and are difficult to standardize; and FLT3 inhibitor response prediction markers show drug-specific dispersion phenomenon (such as independent markers required for gilteritinib and midostaurin), making it difficult to build a unified cross-drug universal prediction model. This marker fragmentation phenomenon forces clinical detection to design independent schemes for each drug, significantly increasing the complexity and cost of operation. So far, a universal efficacy prediction model compatible with different FLT3 inhibitors is still blank, which cannot meet the urgent need of clinical dynamic drug selection for individual patients. SUMMARY

[0007] The present application aims to at least solve one of the technical problems in the related art. To this end, the object of the present application is to provide a diagnostic kit based on acute leukemia prognosis and drug sensitivity prediction.

[0008] In order to achieve the above-mentioned object, the technical solution adopted by the present application is as follows:

[0009] The diagnostic kit based on acute leukemia prognosis and drug sensitivity prediction comprises prediction module I and prediction module II, the prediction module I is used to predict the clinical survival rate of patients, which can guide the stratified treatment of AML patients, and intensive or combined therapy is needed for high-risk patients; the prediction module II is used to predict the sensitivity of the body to drugs, to support the decision-making of clinical drug use;

[0010] The prediction module I uses one or more of the following molecules as prognostic markers: CCND3, FERMT3, PLD4, TOP1MT, NRGN, RCAN1, ABCD1, ALOX5AP, CCL5, CEBPB, CTSZ, FLOT1, HCK, IL4I1, IL6R, ITGA7, ITGAM, MCOLLN2, NEDD9, PEA15, PECAM1, POU2F2, PSMB9, RNPEP, RRAS, SRGAP2, SYK, THEMIS2, TNFAIP2, UNC13D, VDR, and ZNF385A.

[0011] The prediction module II uses one or more of the following molecules as drug sensitivity markers: BMP8B, IGF1R, OTULINL, SLC22A15, CERS1, and PDE4A.

[0012] Preferably, the prognostic biomarkers include six molecules: CCND3, FERMT3, PLD4, TOP1MT, NRGN, and RCAN1, and the prognostic risk scoring formula is as follows:

[0013] Risk score = Weighting coefficient I × A + Weighting coefficient II × B + Weighting coefficient III × C + Weighting coefficient IV × D + Weighting coefficient V × E + Weighting coefficient VI × F;

[0014] The cutoff value for the risk score is 1 to 1.05;

[0015] A represents the gene expression level of CCND3, B represents the gene expression level of FERMT3, C represents the gene expression level of PLD4, D represents the gene expression level of TOP1MT, E represents the gene expression level of NRGN, and F represents the gene expression level of RCAN1.

[0016] Preferably, the weighting coefficient I is 0.39, the weighting coefficient II is 0.225, the weighting coefficient III is 0.122, the weighting coefficient IV is 0.087, the weighting coefficient V is 0.082, and the weighting coefficient VI is 0.075.

[0017] Preferably, the drug is selected from FLT3 inhibitors.

[0018] Preferably, the FLT3 inhibitor is selected from one or more of quezartinib, crenolatib, giglitinib, midotulin, and sorafenib.

[0019] Preferably, the normalized AUC is used to assess the body's sensitivity to the drug;

[0020] AUC is the area under the drug dose-response curve.

[0021] Preferably, the prognostic biomarkers are obtained by screening based on proteomic data from FLT3 inhibitor-resistant cell models, combined with TCGA-LAML and Target-AML clinical cohort data, using univariate Cox regression analysis.

[0022] TCGA-LAML is for adult patients with acute myeloid leukemia, while Target-AML is for children and adolescents with acute myeloid leukemia.

[0023] Preferably, the drug sensitivity biomarkers are screened using the LASSO-logistic regression machine learning method.

[0024] Preferably, the drug sensitivity includes the body's sensitivity to the drug before and after drug administration.

[0025] Preferably, the acute leukemia is acute myeloid leukemia.

[0026] The above-described one or more technical solutions in the embodiments of the present invention have at least one of the following technical effects:

[0027] The present invention provides a diagnostic kit for predicting the prognosis and drug sensitivity of acute leukemia, comprising prediction module I and prediction module II. Prediction module I is used to predict the clinical survival rate of patients, and it uses one or more of the following molecules as prognostic markers: CCND3, FERMT3, PLD4, TOP1MT, NRGN, RCAN1, ABCD1, ALOX5AP, CCL5, CEBPB, CTSZ, FLOT1, HCK, IL4I1, IL6R, ITGA7, ITGAM, MCOLLN2, NEDD9, PEA15, PECAM1, POU2F2, PSMB9, RNPEP, RRAS, SRGAP2, SYK, THEMIS2, TNFAIP2, UNC13D, VDR, and ZNF385A. Prediction module II is used to predict the body's sensitivity to drugs, and it uses one or more of the following molecules as drug sensitivity markers: BMP8B, IGF1R, OTULINL, SLC22A15, CERS1, and PDE4A. This invention overcomes the limitations of single biomarker predictive efficacy by constructing a dual prediction model, achieving functional coupling between survival prognosis and prediction of multiple drug susceptibility, and providing an integrated tool that can be clinically translated for precision treatment of AML. Simultaneously, it employs a proteomics-driven biomarker dimensionality reduction strategy to overcome the redundancy problem of traditional biomarker detection systems, simplifying the operation process, reducing clinical testing costs, and improving efficiency.

[0028] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0029] Figure 1 This is a heatmap of differential protein clustering in drug-resistant cell lines obtained from proteomics detection, provided in Example 1 of this invention.

[0030] Figure 2 This is a schematic diagram of the screening process for 32 risk proteins that are related to both drug resistance and patient survival, provided in Embodiment 1 of the present invention.

[0031] Figure 3 This is a graph showing the relationship between the combination of biomarkers used to construct the survival model and the area under the model curve in the TCGA data provided in Embodiment 1 of the present invention.

[0032] Figure 4 This is the AUC curve corresponding to the optimal combination of markers provided in Embodiment 1 of the present invention.

[0033] Figure 5 This is a survival curve diagram of the optimal biomarker combination model provided in Embodiment 1 of the present invention.

[0034] Figure 6 This is the performance of the optimal marker combination model provided in Embodiment 1 of the present invention on the Target AML validation set.

[0035] Figure 7 This is the optimal combination of biomarkers for establishing predictive drug sensitivity models for the five FLT3 inhibitors provided in Example 1 of this invention.

[0036] Figure 8 This is the AUC curve corresponding to the optimal biomarker combination of the five FLT3 inhibitors provided in Example 1 of this invention.

[0037] Figure 9 This is the AUC curve of six optimal biomarker combinations provided in Embodiment 1 of the present invention, using data corresponding to gilteritinib as the test set.

[0038] Figure 10 The AUC curves provided in Embodiment 1 of the present invention correspond to the six optimal biomarker combinations, using data from quezartinib, keranolanib, midotolamine, and sorafenib as validation sets. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments. Obviously, the described embodiments are only some embodiments of this invention, not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention. The following embodiments are used to illustrate this invention, but cannot be used to limit the scope of this invention.

[0040] Example 1

[0041] I. Construction of Prognostic Risk Model

[0042] To elucidate the clinically relevant proteomic dynamics in drug resistance evolution, this invention performs K-means clustering on differentially expressed proteins, identifying five clusters with clearly defined functional characteristics, such as... Figure 1 As shown in the figure, the expression of modules C2 and C5 is high in the wild type and decreases with the development of drug resistance (early to late stage); while the expression of modules C1, C3, and C4 is low in the wild type and increases with the development of drug resistance.

[0043] Among them, the C1 cluster contains 111 proteins, whose functions include: innate immune response, carboxylic acid metabolism, lymphocyte activation, type II interferon response, and positive regulation of cell-matrix adhesion.

[0044] The increasing expression trend of cluster C1 suggests enhanced immune response and cell adhesion, which may help drug-resistant cells escape immune surveillance or remodel the microenvironment.

[0045] Cluster C2 contains 71 proteins, whose functions include: regulation of reactive oxygen species metabolism, positive regulation of cysteine-type endopeptidase activity, mitochondrial assembly, negative regulation of cell-cell adhesion, and alcohol biosynthesis.

[0046] The decreasing expression trend of cluster C2 suggests that mitochondrial assembly and reactive oxygen species metabolism-related functions are suppressed as drug resistance develops, possibly reflecting adaptive changes in cellular energy metabolism.

[0047] Cluster C3 contains 370 proteins, whose functions include: innate immune response, inflammatory response, endoplasmic reticulum stress response, glycoprotein metabolism, and oxidative stress response;

[0048] The increasing expression trend of cluster C3 suggests that inflammation and endoplasmic reticulum stress activation reflect the compensatory mechanisms of drug-resistant cells in response to stress (such as drug stimulation).

[0049] Cluster C4 contains 118 proteins, whose functions include: acute phase response, positive regulation of cell adhesion, regulation of inflammatory response, regulation of MAPK cascade reaction, and regulation of interleukin-6 (IL-6) production.

[0050] The increasing expression trend of cluster C4 suggests that upregulation of inflammatory signals (such as MAPK and IL-6) and cell adhesion may promote drug resistance through a pro-inflammatory microenvironment or signaling pathway reprogramming.

[0051] Cluster C5 contains 350 proteins, whose functions include: oxidative phosphorylation, mitochondrial electron transport (ubiquinone → cytochrome C; cytochrome C → oxygen), mitochondrial transport, and oxidative stress response.

[0052] The expression trend of cluster C5 decreased, which indicates that the core function of the mitochondrial respiratory chain (oxidative phosphorylation) was significantly downregulated, suggesting that drug-resistant cells may rely on alternative metabolisms such as glycolysis.

[0053] In the figure, wild-type represents initial drug sensitivity and no drug resistance, but with FLT3-ITD mutation; early drug resistance represents drug resistance after 3 to 6 months of drug use, with FLT3-TKD mutation added on the basis of FLT3-ITD mutation; late drug resistance represents drug resistance after 6 months of drug use, with NRAS mutation added on the basis of early drug resistance mutation.

[0054] like Figure 2 As shown, the analytical process from protein clustering to prognostic risk protein screening is described in detail below:

[0055] 1. Using the K-means clustering algorithm, the proteins are divided into two groups:

[0056] C3 and C4 clusters: contain a total of 488 proteins. The upward arrow indicates that this group of proteins is positively correlated with the "drug resistance" phenotype (drug resistance proteins).

[0057] Clusters C2 and C5: Contain a total of 421 proteins. The downward arrows indicate that this group of proteins is negatively correlated with the "drug resistance" phenotype (drug sensitivity proteins).

[0058] II. Univariate Cox Regression Analysis: Univariate Cox proportional hazards model analysis was performed on RNA expression in two AML datasets (TCGA-LAML and Target-AML) to screen proteins associated with survival prognosis.

[0059] Thirty-two “risk proteins” (hazard ratio > 1, indicating that their high expression is associated with poor prognosis) were screened out; only one “beneficial protein” (hazard ratio < 1, indicating that its high expression is associated with good prognosis) was identified.

[0060] III. Identifying core risk proteins using Venn diagrams:

[0061] The intersection of the three sets of protein data is shown by Venn diagram. The three sets of data are drug resistance proteins of C3 and C4 clusters, risk transcripts of Target AML data, and risk transcripts of TCGA-LAML data.

[0062] The central intersection (32 molecules): Molecules that belong to all three groups mentioned above represent the most stable and core prognostic risk molecules (those that belong to the drug resistance cluster in the cluster and are also verified as risk factors in two independent datasets).

[0063] The above screening process begins by identifying phenotypic differences (high / low risk clusters) through "clustering," then uses "survival analysis" to screen for prognostic-related proteins, and finally uses "multi-dataset intersection" to identify core risk proteins, ensuring the reliability and generalizability of the results (cross-dataset validation).

[0064] By integrating transcriptomic and survival data from independent clinical cohorts of TCGA-LAML (n=151) and Target-AML (n=1074), the clinical relevance of differentially expressed proteins screened by the drug sensitivity model was validated: using a univariate Cox proportional hazards regression model, 32 risk proteins that were significantly negatively correlated with overall patient survival (hazard ratio >1, p<0.05) were screened out, as shown in Tables 1 and 2.

[0065]

[0066]

[0067] To overcome overfitting in high-dimensional data, the LASSO regression algorithm was applied for feature compression: the regularization parameter λ was optimized through 10-fold cross-validation, and six core prognostic biomarkers were selected.

[0068] The weighting coefficients for the six prognostic biomarkers CCND3, TOP1MT, RCAN1, NRGN, PLD4, and FERMT3 are 0.39, 0.255, 0.122, 0.087, 0.082, and 0.075, respectively. A prognostic risk scoring formula is established as follows:

[0069] Risk score = 0.39 × CCND3 gene expression level + 0.255 × TOP1MT gene expression level + 0.122 × RCAN1 gene expression level + 0.087 × NRGN gene expression level + 0.082 × PLD4 gene expression level + 0.075 × FERMT3 gene expression level;

[0070] In the TCGA-AML training set, the high-risk group (n=66) and the low-risk group (n=66) were divided using the median risk score (risk score = 1.027056) as the cutoff value. The 5-year survival rate of the high-risk group was only 9.85%, while the 5-year survival rate of the low-risk group reached 35.83% (hazard ratio = 2.380, 95% CI: 1.512–3.746, p<0.001).

[0071] The results were replicated in the Target-AML validation set (hazard ratio = 1.360, 95% CI: 1.204–1.536, p < 0.001).

[0072] Six prognostic biomarkers and prognostic models (CCND3, FERMT3, PLD4, TOP1MT, NRGN, RCAN1), such as Figure 3 As shown;

[0073] The prediction results in the TCGA LAML test set are as follows: Figure 4 and Figure 5 As shown, the results indicate that these six prognostic biomarkers have excellent predictive efficacy.

[0074] Prediction results in the Target AML validation set, such as Figure 6 As shown, the results indicate that these six prognostic biomarkers still maintain high predictive efficiency (AUC=0.836).

[0075] In summary, the prognostic risk scoring formula based on the aforementioned six prognostic biomarkers demonstrated superior stratification capabilities on both the training set (TCGA-LMAL) and the validation set (Target-LMAL), effectively classifying patients into high-risk and low-risk groups. Its predictive power was significantly better than the traditional ELN classification. Therefore, using these six prognostic biomarkers as a predictive model can serve as risk stratification biomarkers to identify patients who may benefit from drug resistance proteomics-guided combination therapy.

[0076] II. Construction of drug sensitivity model.

[0077] This study employed a dose-escalation protocol to construct early resistance (ER) and late resistance (LR) pathological cell models for mechanistic research and the development of reversal strategies. Compared to parental (WT) cells, ER cells exhibited an IC50 value that was more than 10-fold higher and were accompanied by FLT3-TKD mutations, while LR cells showed more than 100-fold resistance due to secondary NRAS mutations.

[0078] To predict FLT3 inhibitor sensitivity, pharmacogenetic data from primary AML samples (n=671) in the BeatAML database were integrated. The area under the dose-response curve (AUC) of five inhibitors—quezatinib, crenolatib, giglitinib, midotulin, and sorafenib—was collected as a sensitivity indicator (lower AUC represents higher sensitivity). For each inhibitor, the AUC value was normalized to between 0 and 1 using the formula 1 - AUC / 300 (higher normalized AUC represents higher sensitivity). Patients were then grouped according to their normalized AUC values: the upper third was the sensitive group, and the lower third was the resistant group. Differences between the sensitive and resistant groups for each inhibitor were compared to identify differentially expressed molecules. Predictive models were independently constructed for each inhibitor using differentially expressed molecules: iterative LASSO-logistic regression (L1 regularization coefficient α=1, 10-fold cross-validation to optimize λ) was employed. For each drug, 22–31 biomarker combinations were selected, and the predictive power AUC of each model was >0.85 (e.g., the AUC of the giglitinib model was 0.87). Figure 7 As shown.

[0079] The predictive power of highly predictive biomarkers was identified for the molecular profiles of each FLT3 inhibitor-sensitive and resistance cohort, such as... Figure 8 As shown (AUC=0.82~0.90).

[0080] The intersection of biomarkers for each drug was used to identify six common core biomarkers as drug susceptibility markers. These six biomarkers are BMP8B, IGF1R, OTULINL, SLC22A15, CERS1, and PDE4A.

[0081] The data corresponding to gilteritinib was used as the test set to test the model, and the results are as follows: Figure 9 As shown, this result indicates that the prediction efficiency can reach 0.733;

[0082] Using quezartinib, crenolatib, midotolin, and sorafenib as the validation set, the results are as follows: Figure 10 As shown (AUC 0.73~0.81), it can still maintain stable predictive performance.

[0083] This invention provides a diagnostic kit for predicting prognosis and drug sensitivity in acute leukemia (AML). It utilizes a dual-prediction module based on cross-validation of the drug resistance proteome and clinical transcriptome: an acquired drug-resistant cell model of AML is established, and deep proteomics analysis is used to screen for drug resistance-related protein targets. Simultaneously, clinical transcriptome data of AML patients from public databases such as TCGA are integrated to validate the survival relevance of these targets (Cox regression analysis), ensuring that the biomarkers possess both explanatory power for drug resistance and predictive value for overall survival (OS). This approach overcomes the limitations of traditional single-omics methods, achieving a direct mapping between functional protein mechanisms and clinical phenotypes. A survival risk scoring model is established based on a concise biomarker set (≤10 core targets): a multivariate Cox regression system is used to generate a quantitative risk assessment system. A cross-inhibitor response prediction model is developed: drug sensitivity data (AUC values) of five FLT3 inhibitors (quezartinib, keranoprazole, giglitinib, midotulin, and sorafenib) from the beatAML database are used, and a unified prediction framework is trained using machine learning algorithms to achieve simultaneous evaluation of multi-drug efficacy. This system utilizes a proteome-driven biomarker dimensionality reduction strategy to overcome the redundancy problem of traditional biomarkers. At the same time, through the universal association of core targets of drug resistance mechanisms, it achieves for the first time the functional coupling of survival prognosis and multidrug response prediction, providing an integrated tool that can be clinically translated for precision treatment of AML.

[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A diagnostic kit for predicting the prognosis and drug sensitivity of acute leukemia, characterized in that, It includes a prediction module I and a prediction module II, wherein the prediction module I is used to predict the clinical survival rate of patients, and the prediction module II is used to predict the body's sensitivity to drugs; The prediction module I uses six molecules—CCND3, FERMT3, PLD4, TOP1MT, NRGN, and RCAN1—as prognostic biomarkers, and the prognostic risk scoring formula is as follows: Risk score = Weighting coefficient I × A + Weighting coefficient II × B + Weighting coefficient III × C + Weighting coefficient IV × D + Weighting coefficient V × E + Weighting coefficient VI × F; Weight coefficient I is 0.39, weight coefficient II is 0.225, weight coefficient III is 0.122, weight coefficient IV is 0.087, weight coefficient V is 0.082, and weight coefficient VI is 0.075; The cutoff value for the risk score is 1 to 1.05; A represents the gene expression level of CCND3, B represents the gene expression level of FERMT3, C represents the gene expression level of PLD4, D represents the gene expression level of TOP1MT, E represents the gene expression level of NRGN, and F represents the gene expression level of RCAN1. The prediction module II uses six molecules—BMP8B, IGF1R, OTULINL, SLC22A15, CERS1, and PDE4A—as drug susceptibility markers. The drug is selected from FLT3 inhibitors.

2. The diagnostic kit for predicting the prognosis and drug sensitivity of acute leukemia as described in claim 1, characterized in that, The FLT3 inhibitor is selected from one or more of quezartinib, crenolanib, giglitinib, midotoxorin, and sorafenib.

3. The diagnostic kit for predicting the prognosis and drug sensitivity of acute leukemia as described in any one of claims 1 or 2, characterized in that, Normalized AUC was used to assess the body's sensitivity to the drug; AUC is the area under the drug dose-response curve.

4. The diagnostic kit for predicting the prognosis and drug sensitivity of acute leukemia as described in claim 2, characterized in that, The prognostic biomarkers were obtained by screening based on proteomic data from FLT3 inhibitor-resistant cell models, combined with TCGA-LAML and Target-AML clinical cohort data, using univariate Cox regression analysis. TCGA-LAML is for adult patients with acute myeloid leukemia, while Target-AML is for children and adolescents with acute myeloid leukemia.

5. The diagnostic kit for predicting the prognosis and drug sensitivity of acute leukemia as described in claim 1, characterized in that, The drug susceptibility markers were screened using the LASSO-logistic regression machine learning method.

6. The diagnostic kit for predicting the prognosis and drug sensitivity of acute leukemia as described in claim 1, characterized in that, The drug sensitivity includes the body's sensitivity to the drug before and after drug administration.

7. The diagnostic kit for predicting the prognosis and drug sensitivity of acute leukemia as described in claim 1, characterized in that, The acute leukemia mentioned is acute myeloid leukemia.

Images