PAML SCT risk stratification prognosis model and construction method thereof

Abstract

The invention discloses a pAML SCT risk stratification prognosis model and a construction method thereof, and belongs to the field of children acute myelogenous leukemia stem cell transplantation risk prognosis research. Clinical data of a pAML patient are collected and analyzed by adopting a statistical method, a risk stratification model is established according to prognosis factors including combination of diagnosis age, KMT2A-r state and MRD1 state, and through verification, the pAML SCT risk can be effectively predicted by adopting the total lifetime prognosis model and the event-free lifetime prognosis model; the risk group of the patient can be determined according to the diagnosis age, the KMT2A-r state and the MRD1 state, so that a clinician is assisted in judging clinical development and disease outcome of the patient, precise layered treatment is realized, and the doctor can guide and / or identify candidates suitable for stem cell transplantation and formulate a treatment scheme.

Description

Technical Field

[0001] The present invention belongs to the field of risk prognosis research of childhood acute myeloid leukemia stem cell transplantation, and specifically relates to a pAML SCT risk stratification prognosis model and a construction method thereof. Background Art

[0002] Acute myeloid leukemia (AML) is the second most common hematological cancer in children, but it accounts for half of all childhood leukemia deaths. Prominent consortia such as the Children's Oncology Group (COG), the International Berlin-Frankfurt-Münster Study Group (I-BFM-SG), the Medical Research Council (MRC), and the Japan Pediatric Leukemia / Lymphoma Study Group (JPLSG) are currently risk stratifying patients with pediatric acute myeloid leukemia (pAML) based on cytogenetics and molecules (cell molecules). For example, in COG, the tool used to determine the risk stratification of pAML patients is to study defined risk groups. In addition, high risk is defined by the patient's response to induction therapy, which can be measured by minimal residual disease (MRD) tested in the bone marrow.

[0003] For high-risk AML patients with first complete remission (CR1), stem cell transplantation (SCT) is recommended for pAML patients; although SCT for CR1 patients may improve survival in a subset of patients, the benefit for most patients remains unclear. Studies have shown that approximately 40% of pAML patients relapse and die after intensive therapy and SCT. Notably, FMS-like tyrosine kinase3 / internal tandem duplication (FLT3 / ITD) mutations have been identified as important adverse prognostic factors, and patients who exhibit this mutation are also classified as high risk and recommended for SCT. Although SCT may improve the prognosis of patients who exhibit FLT3-ITD mutations, the relapse rate is still higher compared to patients without these mutations. These underscore the importance of establishing a robust prognostic model that can accurately predict outcomes and may help accurately screen SCT candidates.

[0004] Currently, little attention has been paid to developing a specific prognostic prediction model for pAML patients undergoing SCT. The Allogeneic HCT Risk Model remains the only prognostic model for pAML patients undergoing SCT and initially provided an effective means of assessing their prognosis. However, several clinical questions remain to be answered.

[0005] First, the allogeneic HCT risk model incorporates the results of the second complete remission (CR2), which has limited ability to provide prognostic assessment for pAML patients who undergo SCT during CR1; given the importance of the CR1 stage in assessing the suitability, timing, and strategy of SCT in pAML patients, a prognostic model for SCT in CR1 is urgently needed. Second, the allogeneic HCT risk model contains a cellular molecular risk stratification system, and the model needs to be updated accordingly to adapt to its continuous cellular molecular evolution. In addition, patients who are suitable for transplantation are not identified in the allogeneic HCT risk model. Summary of the invention

[0006] In view of the above-mentioned deficiencies in the prior art, the present invention provides a pAML (childhood acute myeloid leukemia) SCT (stem cell transplantation) risk stratification prognosis model and a construction method thereof. The model risk scoring formula constructed by the present invention is used to determine the risk group of the patient. The model has wide applicability in various clinical environments and time ranges, is easy to use in different medical and health care environments, and brings great help in guiding the clinical treatment of childhood acute myeloid leukemia stem cell transplantation.

[0007] The present invention adopts the following technical solutions:

[0008] A pAML SCT risk stratification prognostic model, which includes an overall survival (OS) prognostic model and an event-free survival (EFS) prognostic model; the overall survival prognostic model and the event-free survival prognostic model are both constructed by prognostic factors consisting of diagnosis age, KMT2A-r (KMT2A gene rearrangement) status and MRD1 (minimal residual disease 1) status.

[0009] Furthermore, the above-mentioned overall survival prognostic model and event-free survival prognostic model are specifically as follows: overall survival risk score or event-free survival risk score = 0.612 × whether there is KMT2A-r + 0.643 × whether there is MRD1 - 0.682 × whether the age of diagnosis is less than 5 years old;

[0010] When the age at diagnosis was less than 5 years old or KMT2A-r or MRD1 was present, each prognostic factor was assigned a value of 0; when the age at diagnosis was greater than or equal to 5 years old or KMT2A-r or MRD1 was absent, each prognostic factor was assigned a value of 1;

[0011] When the overall survival risk score or the event-free survival risk score was ≤-0.682, it was judged as the low-risk group; when the overall survival risk score or the event-free survival risk score was >-0.682 and <0.574, it was judged as the medium-risk group; when the overall survival risk score or the event-free survival risk score was ≥0.574, it was judged as the high-risk group.

[0012] The present invention also provides a method for constructing the above-mentioned pAML SCT risk stratification prognostic model, comprising the following steps:

[0013] (1) Collect clinical data of pAML patients, including clinical data, demographic data, and laboratory data;

[0014] (2) Taking the patient's overall survival or event-free survival as the outcome, univariate analysis was performed on the above clinical data to screen the risk factors that affect the patient's overall survival or event-free survival;

[0015] (3) Taking the patient's overall survival or event-free survival as the outcome, a multivariate analysis was performed on the risk factors obtained in step (2) to screen for risk factors that significantly affect the patient's overall survival or event-free survival;

[0016] (4) Taking the patient's overall survival or event-free survival as the outcome, a model was constructed based on the results of the multivariate analysis obtained in step (3); the screened factors were introduced into the logistic regression equation to construct a pAML SCT risk stratification prognostic model.

[0017] Furthermore, the above clinical data included the first event, survival outcome, overall survival, event-free survival, treatment regimen, site of relapse or induction failure, SCT.in.1st.CR (hematopoietic stem cell transplantation after the first complete remission), whether gemtuzumab treatment was performed, MRD1 status, MRD2 (minimal residual disease 2) status, CR1 (complete remission 1) status, CR2 (complete remission 2) status and risk grouping.

[0018] Furthermore, the site of relapse or induction failure is bone marrow, central nervous system, granulocytic sarcoma, cytogenetics or other sites.

[0019] Furthermore, the above demographic information includes gender, race, age at diagnosis and age group.

[0020] Furthermore, the above laboratory data include WBC (white blood cell count), percentage of bone marrow leukemic blasts, blasts in peripheral blood, presence or absence of central nervous system diseases, granulocytic sarcoma, FAB classification, chromosomal abnormality index, cytogenetic complexity, original cytogenetic code, FLT3 / ITD (internal tandem duplication of FLT3 gene), FLT3 / ITD allele ratio, FLT3 point mutation, NPM mutation, CEBPA mutation, WT1 mutation, c-kit gene exon 8 mutation, c-kit gene exon 17 mutation, cell fusion molecular risk, gene fusion and final risk stratification.

[0021] Furthermore, the chromosomal abnormality indicators include abnormalities or rearrangements of t.6.9., t.8.21., t.3.5..q25.q34., t.6.11..q27.q23., t.9.11..p22.q23., t.10.11..p11.2.q23., t.11.19..q23.p13.1., inv.16., del5q, del7q, del9q, monosomy.5, monosomy.7, trisomy.8, trisomy.21, MLL, Minus.Y and Minus.X.

[0022] The beneficial effects of the technical solution of the present invention are:

[0023] The present invention establishes a SCT cell molecular risk stratification non-dependent prognostic model designed specifically for CR1 pAML patients, the model includes relevant prognostic factors, risk groups corresponding to the evaluation scores of different states of prognostic factors, and prognostic states corresponding to each risk group. Through validation group verification, the model can effectively predict the SCT risk of pAML patients. Compared with the existing models the Study Defined Risk Group (research defined risk group) and the Allogeneic HCT Risk Model, the model of the present invention shows good superiority, has better prognostic stratification ability, and also has wide applicability in various clinical environments and time ranges, which can provide a clear framework for determining potential candidates suitable for SCT, thereby assisting clinicians in judging the clinical development and disease prognosis of patients, and realizing accurate stratified treatment. At the same time, the pAML SCT risk stratification prognostic model of the present invention can also be combined with FLT3 / ITD AR, which is beneficial for doctors to guide and / or identify suitable stem cell transplantation candidates and the formulation of treatment plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1The flowchart of the research on the pAML SCT risk stratification prognostic model; A: the process diagram of model establishment; B: the model validation diagram of the prognostic performance of the pAML SCT risk stratification prognostic model by the internal validation group and the external validation group; C: the process diagram of the comparison of the pAML SCT risk stratification prognostic model with the Study Defined Risk Group and the Allogeneic HCT Risk Model prognostic models used to guide SCT; D: the flowchart of the establishment of the pAML SCT risk stratification prognostic model in guiding and / or identifying suitable stem cell transplantation candidates;

[0025] Figure 2 Forest plot for prognostic factors;

[0026] Figure 3 This is the survival probability curve of OS in the experimental group;

[0027] Figure 4 It is the survival probability curve of EFS of the experimental group;

[0028] Figure 5 This is the survival probability curve of OS of the internal validation group;

[0029] Figure 6 This is the survival probability curve of EFS of the internal validation group;

[0030] Figure 7 This is the calibration curve diagram of OS of the internal validation group;

[0031] Figure 8 This is the calibration curve graph of EFS of the internal validation group;

[0032] Fig. 9 Kaplan-Meier plot of OS for the test 1 cohort in the external validation group;

[0033] Fig.10 Kaplan-Meier plot of OS for the test 2 cohort in the external validation group;

[0034] Fig.11 The Kaplan-Meier diagram of the Allogeneic HCT Risk Model in the experimental group;

[0035] Fig.12 This is the Kaplan-Meier plot of the Allogeneic HCT Risk Model in the internal validation group;

[0036] Fig.13The DCA curves of OS at different time points; A: DCA curve of OS at 1 year; B: DCA curve of OS at 3 years; C: DCA curve of OS at 5 years;

[0037] Fig.14 The DCA curves of EFS at different time points; A: DCA curve of EFS at 1 year; B: DCA curve of EFS at 3 years; C: DCA curve of EFS at 5 years;

[0038] Fig.15 Flow chart for the pAML SCT risk stratification prognostic model combined with FLT3 / ITD AR;

[0039] Fig.16 The calibration curves of OS and EFS of the no-effect group; A: calibration curve of OS; B: calibration curve of EFS;

[0040] Fig.17 The calibration curves of OS and EFS of the EFS extension group; A: calibration curve of OS; B: calibration curve of EFS;

[0041] Fig.18 The calibration curves of OS and EFS of the OS and EFS extension groups; A: calibration curve of OS; B: calibration curve of EFS;

[0042] Fig.19 The calibration curves of OS and EFS of the harmful group; A: calibration curve of OS; B: calibration curve of EFS;

[0043] Fig. 20 It is the overall Sankey diagram of the no-effect group, EFS-extended group, OS and EFS-extended group, and harmful group. DETAILED DESCRIPTION

[0044] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those generally understood by those of ordinary skill in the art to which the present invention belongs. It should be noted that the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the exemplary embodiments of the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the term "comprising" or "including" is used in this specification, it indicates the presence of features, steps, operations, devices, components or combinations thereof. In the following specific embodiments, if the experimental method of the specific conditions is not indicated, it is usually in accordance with the conventional methods and conditions in the art, and such techniques and conditions are fully explained in the literature.

[0045] The present invention is further described in conjunction with specific examples. The following examples are only for explaining the present invention and are not intended to limit the content thereof. If the specific experimental conditions are not specified in the examples, they are usually carried out under conventional conditions or under conditions recommended by the sales company; the materials and reagents used in the examples, unless otherwise specified, can be purchased through commercial channels.

[0046] The establishment and validation of the pAML SCT risk stratification prognostic model, the comparison of the pAML SCT risk stratification prognostic model with existing models, and the clinical decision support were analyzed and validated. The research process is as follows: Figure 1 First, the pAML SCT risk stratification prognostic model of the present invention is a model specifically used for the risk stratification of pAML patients receiving SCT ( Figure 1 A); secondly, the model was validated internally and externally. The internal validation focused on validating the model's risk stratification ability, discrimination ability, calibration ability, and robustness, while the external validation re-examined the model's risk stratification ability, discrimination ability, and universality in different cohorts and subgroups ( Figure 1 B); subsequently, the pAML SCT risk stratification prognostic model was compared with study-defined risk groups and the allogeneic HCT risk model in terms of risk stratification ability, discrimination, and clinical utility ( Figure 1 C). Finally, the pAML SCT risk stratification prognostic model combined with the FLT3 allele ratio was used to accurately screen the most suitable SCT candidates ( Figure 1 D).

[0047] Example 1

[0048] (1) The clinical, demographic, and laboratory data of 1941 patients with primary pAML (excluding M3 subtype) in the Treatment Applicability Research Generating Effective Therapies (TARGET) database before November 11, 2023 were collected and organized. Among them, 308 patients underwent SCT, defined as "TARGET 308", and were divided into an experimental group (n=214) and an internal validation group (n=94) in a ratio of 7:3. The experimental group was used for model construction, and the internal validation group was used for model validation.

[0049] Clinical data included the first event, survival outcome, overall survival, event-free survival, treatment regimen, site of relapse or induction failure, SCT.in.1st.CR, whether gemtuzumab treatment was performed, MRD1 status, MRD2 status, CR1 status, CR2 status, and risk grouping; among them, the site of relapse or induction failure was bone marrow, central nervous system, granulocytic sarcoma, cytogenetic or other sites;

[0050] Demographic data included sex, race, age at diagnosis, and age group;

[0051] Laboratory data included WBC, percentage of bone marrow leukemic blasts, blasts in peripheral blood, presence or absence of central nervous system disease, granulocytic sarcoma, FAB classification, chromosomal abnormality indicators, cytogenetic complexity, original cytogenetic code, FLT3 / ITD, FLT3 / ITD allele ratio, FLT3 point mutation, NPM mutation, CEBPA mutation, WT1 mutation, c-kit gene exon 8 mutation, c-kit gene exon 17 mutation, cell fusion molecular risk, gene fusion and final risk stratification; chromosomal abnormality indicators included t Abnormalities or rearrangements of .6.9., t.8.21., t.3.5..q25.q34., t.6.11..q27.q23., t.9.11..p22.q23., t.10.11..p11.2.q23., t.11.19..q23.p13.1., inv.16., del5q, del7q, del9q, monosomy.5, monosomy.7, trisomy.8, trisomy.21, MLL, Minus.Y, and Minus.X;

[0052] (2) Data from 232 patients from four domestic diagnosis and treatment centers were used as the external validation group, with the first diagnosis time ranging from January 2012 to December 2023; patients from the First Affiliated Hospital of University of Science and Technology of China (n=154) formed the test 1 cohort, while patients from the Second Affiliated Hospital of Dalian Medical University (n=36), Children's Hospital Affiliated to Chongqing Medical University (n=26), and Shengjing Hospital Affiliated to China Medical University (n=16) formed the test 2 cohort; the external validation group was used for model validation;

[0053] (3) Overall survival (OS) was defined as the time interval from the start of the study to death due to any cause; event-free survival (EFS) was defined as the time interval from the start of the study to death or recurrence due to any cause; OS and EFS were calculated in months;

[0054] Taking the patient's OS as the outcome, the above clinical data were subjected to univariate analysis to screen the risk factors that significantly affected the patient's OS. The results showed that the diagnosis age, KMT2A-r (KMT2A gene rearrangement) status and MRD1 (minimal residual disease 1) status showed significant correlation (p<0.05). Taking the patient's OS as the outcome, the risk factors obtained from the above univariate analysis were analyzed and screened by a multivariate proportional hazards regression model (proportional hazards model, Cox model). The results are as follows Figure 2 As shown, it was found that the age of diagnosis, KMT2A-r status and MRD1 status all showed significant correlation (p<0.05); taking the patient's EFS as the outcome, the above clinical data were screened by univariate analysis and multivariate proportional hazard regression model analysis, and it was also found that the age of diagnosis, KMT2A-r status and MRD1 status all showed significant correlation (p<0.05);

[0055] Finally, the independent prognostic factors for OS and EFS were: diagnosis age, KMT2A-r status, and MRD1 status;

[0056] (4) Taking the patient's OS or EFS as the outcome, a model was constructed based on the results of the multivariate analysis in step (3), and the screened factors were introduced into the logistic regression equation to construct the following pAML SCT risk stratification prognostic model:

[0057] The prognostic models for overall survival and event-free survival were as follows: overall survival risk score or event-free survival risk score = 0.612 × whether there is KMT2A-r + 0.643 × whether there is MRD1 - 0.682 × whether the age of diagnosis is less than 5 years old;

[0058] When the age at diagnosis was less than 5 years old or KMT2A-r or MRD1 was present, each prognostic factor was assigned a value of 0; when the age at diagnosis was greater than or equal to 5 years old or KMT2A-r or MRD1 was absent, each prognostic factor was assigned a value of 1;

[0059] (5) According to the analysis of the values ​​obtained according to the OS and EFS risk scores of the 241 patients in the experimental group, the 5-year OS rates were 76%, 65%, and 35%, respectively (p < 0.0001, Figure 3 ), the 5-year EFS rates in the low-risk, intermediate-risk, and high-risk groups were 65%, 56%, and 32%, respectively (p=0.00045, Figure 4); Variance inflation factor assessment showed no multicollinearity (all variance inflation factors were <1.2, Table 1); As shown in Figures 2 to 4 and Table 1, patients were divided into low-risk, medium-risk and high-risk groups for OS and EFS. When the OS or EFS risk score was ≤-0.682, it was judged as a low-risk group; when the OS or EFS risk score was >-0.682 and <0.574, it was judged as a medium-risk group; when the OS or EFS risk score was ≥0.574, it was judged as a high-risk group; the prognosis of the low-risk group was good, the prognosis of the medium-risk group was average, and the prognosis of the high-risk group was poor;

[0060] Table 1 Variance inflation factor results of prognostic factors in the experimental group

[0061]

[0062] (6) Validation of the constructed model: The above results were internally and externally validated to verify the reliability of the model. The internal validation focused on confirming the risk stratification ability, discrimination ability, calibration and robustness of the model. The external validation re-examined the risk stratification ability, discrimination ability and universality of the model in different cohorts and subgroups. The results were as follows: Figures 5 to 10 As shown:

[0063] In the internal validation group, 27, 42, and 13 cases were in the low-, intermediate-, and high-risk groups, respectively; the pAML SCT model showed satisfactory risk stratification ability; Figures 5-6 It can be seen that the OS of patients with higher risk levels (p = 0.024, Figure 5 ) and EFS (p = 0.00071, Figure 6 ) were significantly reduced; for the low-, intermediate-, and high-risk groups, the 5-year OS rates were 73%, 52%, and 34%, respectively, while the 5-year EFS rates were 74%, 50%, and 23%, respectively;

[0064] The calibration curves at 1, 3, and 5 years were used to evaluate the calibration effect of the model. The calibration curve (indicated by the blue line) showed that there was good consistency between the actual observed OS and the ideal nomogram-prediction OS for 5-year survival ( Figure 7 ); Similarly, the EFS calibration curve for the model (also shown as a blue line) shows good calibration ( Figure 8 ); indicating that the risk stratification ability, differentiation ability, calibration and robustness of the model of the present invention are good;

[0065] The risk stratification ability, discrimination, and generalizability were evaluated in different external validation cohorts; after adopting the same inclusion and exclusion criteria as the Target308 cohort, the entire external validation cohort included 232 pAML patients from four centers, 154, 36, 26, and 16 patients, respectively; in the Test 1 cohort, Kaplan-Meier analysis showed that the pAML SCT model had a significant effect on OS (p = 0.038, Fig. 9 ) effectively distinguished between the three risk groups; similarly, in the test 2 cohort, Kaplan-Meier analysis also showed that the pAML SCT model effectively distinguished OS (p=0.0051, Fig.10 ) of three risk groups;

[0066] (7) Compare the OS risk stratification prognostic model with the existing models, the Study Defined Risk Group and the Allogeneic HCT Risk Model. Figures 11 to 14 As shown in the experimental group, there was no significant difference among the four risk groups of the AllogeneicHCT Risk Model (p = 0.051, Fig.11 ); in the internal validation group, although there was a significant difference (p = 0.019, Fig.12 ), but there was significant overlap between the high-risk group and the intermediate-risk group; DCA curves were used to further evaluate the decision-making utility of the model in the Target 308 cohort; OS ( Fig.13 ) and EFS( Fig.14 ) showed that the pAML SCT model has a good prospect for clinical application, and its clinical application in predicting patient death at different time points (1 year, 3 years and 5 years) after SCT is better than that of the Study Defined Risk Group and the Allogeneic HCT Risk Model; this indicates that the pAML SCT model has a higher net benefit in actual clinical decision-making and is more suitable for clinical practice; comparison with different models shows that the pAML SCT model of the present invention shows good superiority, indicating that the constructed model has better prognostic stratification ability;

[0067] In the validation group, three patients were given: ① A male patient, 1 year old, CR1 was on March 25, 2016, MRD1 was negative, and there was no KMT2A-r; the patient was determined to be in the low-risk group according to the model of the present invention; the results showed that as of the last follow-up date of May 11, 2024, the patient was still alive, with an OS of 100 months and a PFS of 0 months, and the prognosis was good; ② A male patient, 8 years old, CR1 was on April 14, 2016, MRD1 was positive 6%, and there was no KMT2A-r; the patient was determined to be in the low-risk group according to the model of the present invention It is the high-risk group; the results showed that as of the last follow-up date of July 23, 2016, the patient died, with an OS of 6.97 months, a PFS of 6.97 months, and a very poor prognosis; ③ A male patient, 0 years old, CR1 was on October 17, 2022, with KMT2A-r (MLL / AF9+); according to the model of the present invention, the patient was determined to be in the medium-risk group; the results showed that as of the last follow-up date of May 11, 2024, the patient was still alive, with an OS of 18.36 months, a PFS of 18.36 months, and a general prognosis.

[0068] Example 2

[0069] To further aid pre-transplant decision-making in patients with low- to moderate-risk disease, we incorporated clinically important and well-recognized prognostic risk factors into our model; since pAML patients have a high FLT3 / ITD mutant allele ratio (>0.4; FLT3 / ITD AR is generally considered high risk and recommended for SCT 23,30, the present invention uses FLT3 / ITD AR to improve the selection process of SCT candidates. A total of 237 pAML patients had FLT3 / ITD AR data available for further analysis. The combination of the pAML SCT model with FLT3 / ITD AR resulted in the final pAML SCT model, which divided patients into four subgroups, such as Fig.15 Shown: no effect group: FLT3 / ITD AR low-risk group; EFS extension group: low-risk group with high FLT3 / ITD AR and intermediate-risk group with low FLT3 / ITD AR; OS and EFS extension group: intermediate-risk group with high FLT3 / ITD AR; harmful group: high-risk group.

[0070] The Kaplan-Meier survival curves of the four groups are shown in Figures 16 to 19 As shown. Recommended EFS extension group ( Fig.17 ) and OS and EFS extension groups ( Fig.18) underwent SCT, and the SCT patients in the EFS-extended group had a longer EFS than those without SCT (p<0.001); there was no significant difference in OS between the two groups in the EFS-extended group (p=0.058); in the OS and EFS-extended groups, the OS (p=0.00069) and EFS (p<0.0001) of SCT patients were longer than those of non-SCT patients, and these two groups can prolong the OS and / or EFS of patients. However, SCT is not recommended for patients in the ineffective group and the harmful group. For the ineffective group, there was no significant difference in OS (p=0.13) and EFS (p=0.4) between patients who received SCT and those who did not receive SCT, and SCT did not improve OS or EFS ( Fig.16 ), which suggests that these patients can potentially avoid unnecessary SCT surgery and related risks. For the harmful group, the OS of SCT patients in the harmful group was shorter than that of non-SCT patients (p = 0.0053), and there was no statistically significant difference in EFS between the two groups (p = 0.89). SCT may lead to a decrease in OS ( Fig.19 ), regardless of whether the FLT3 / ITD AR was high or low. Clearly, our final pAML SCT model effectively identified individuals who were not suitable for SCT and those who would benefit from this procedure. Among patients who underwent SCT, the OS and EFS-extended groups had the best prognosis, while the EFS-extended group ranked second. Patients in the deleterious group had the worst prognosis.

[0071] Based on the pAML SCT model combined with FLT3 / ITD AR, we identified 8 patients who were not suitable for transplantation and underwent SCT, including 10 patients in the no-effect group (16.39%) and 48 patients in the harmful group (16.61%). In addition, 81 patients who were suitable for transplantation but did not receive SCT. Among them, 37 patients in the OS+EFS extension group (45.12%) and 44 patients in the EFS extension group (40%) ( Fig. 20 ). These results suggest that the final pAML SCT model could provide more valuable guidance for clinical decision-making compared with current SCT strategies.

[0072] Through the above verification in the validation group, the overall survival prognostic model of the present invention can effectively predict the risk of pAML SCT. Compared with the existing scoring criteria, the present invention determines the patient's risk group according to the diagnosis age, KMT2A-r status and MRD1 status, so as to assist clinicians in judging the patient's clinical development and disease prognosis, realize accurate stratified treatment, and is more conducive to the doctor's guidance and / or identification of suitable stem cell transplantation candidates and the formulation of treatment plans.

[0073] The present invention has been described according to the above embodiments. It should be understood that the above embodiments do not limit the present invention in any form. Any technical solutions obtained by equivalent replacement or equivalent transformation shall fall within the scope of the present invention.

Claims

1. A pAML SCT risk stratification prognostic model, characterized in that: The pAML SCT risk stratification prognostic model includes an overall survival prognostic model and an event-free survival prognostic model; the overall survival prognostic model and the event-free survival prognostic model are both constructed by prognostic factors consisting of diagnosis age, KMT2A-r status and MRD1 status.

2. The pAML SCT risk stratification prognostic model according to claim 1, characterized in that: The overall survival prognostic model and the event-free survival prognostic model are specifically as follows: overall survival risk score or event-free survival risk score = 0.612 × whether there is KMT2A-r + 0.643 × whether there is MRD1 - 0.682 × whether the age of diagnosis is less than 5 years old; When the age of diagnosis is less than 5 years old or there is KMT2A-r or MRD1, each prognostic factor is assigned a value of 0; when the age of diagnosis of the prognostic factor is greater than or equal to 5 years old or there is no KMT2A-r or MRD1, each prognostic factor is assigned a value of 1; When the overall survival risk score or event-free survival risk score was ≤-0.682, it was judged as a low-risk group; When the overall survival risk score or event-free survival risk score was > -0.682 and < 0.574, it was judged as the medium-risk group; When the overall survival risk score or event-free survival risk score was ≥ 0.574, it was judged as a high-risk group.

3. The method for constructing a pAML SCT risk stratification prognostic model according to any one of claims 1 to 2, characterized in that: The following steps are involved: (1) Collect clinical data of pAML patients, including clinical data, demographic data, and laboratory data; (2) Taking the patient's overall survival or event-free survival as the outcome, univariate analysis was performed on the above clinical data to screen the risk factors that affect the patient's overall survival or event-free survival; (3) Taking the patient's overall survival or event-free survival as the outcome, a multivariate analysis was performed on the risk factors obtained in step (2) to screen for risk factors that significantly affect the patient's overall survival or event-free survival; (4) Taking the patient's overall survival or event-free survival as the outcome, a model was constructed based on the results of the multivariate analysis obtained in step (3); the screened factors were introduced into the logistic regression equation to construct a pAML SCT risk stratification prognostic model.

4. The method for constructing a pAML SCT risk stratification prognostic model according to claim 3, characterized in that: The clinical data include the first event, survival outcome, overall survival, event-free survival, treatment regimen, site of relapse or induction failure, SCT.in.1st.CR, whether gemtuzumab treatment was performed, MRD1 status, MRD2 status, CR1 status, CR2 status and risk grouping.

5. The method for constructing a pAML SCT risk stratification prognostic model according to claim 3, characterized in that: The demographic data included sex, race, age at diagnosis, and age group.

6. The method for constructing a pAML SCT risk stratification prognostic model according to claim 3, characterized in that: The laboratory data included WBC, percentage of bone marrow leukemic blasts, blasts in peripheral blood, presence or absence of central nervous system disease, granulocytic sarcoma, FAB classification, chromosomal abnormality index, cytogenetic complexity, original cytogenetic code, FLT3 / ITD, FLT3 / ITD allele ratio, FLT3 point mutation, NPM mutation, CEBPA mutation, WT1 mutation, c-kit gene exon 8 mutation, c-kit gene exon 17 mutation, cell fusion molecular risk, gene fusion and final risk stratification.

Images