Development and validation of a cuproptosis-related prognostic model for acute myeloid leukemia patients using machine learning with stacking.
Ontology highlight
ABSTRACT:
SUBMITTER: Wang X
PROVIDER: S-EPMC10837443 | biostudies-literature | 2024 Feb
REPOSITORIES: biostudies-literature
Publications
Development and validation of a cuproptosis-related prognostic model for acute myeloid leukemia patients using machine learning with stacking.
Wang Xichao X Sun Hao H Dong Yongfei Y Huang Jie J Bai Lu L Tang Zaixiang Z Liu Songbai S Chen Suning S
Scientific reports 20240202 1
Our objective is to develop a prognostic model focused on cuproptosis, aimed at predicting overall survival (OS) outcomes among Acute myeloid leukemia (AML) patients. The model utilized machine learning algorithms incorporating stacking. The GSE37642 dataset was used as the training data, and the GSE12417 and TCGA-LAML cohorts were used as the validation data. Stacking was used to merge the three prediction models, subsequently using a random survival forests algorithm to refit the final model u ...[more]
Similar Datasets
Project description:BackgroundLong non-coding RNAs (lncRNAs) have been reported to have a crucial impact on the pathogenesis of acute myeloid leukemia (AML). Cuproptosis, a copper-triggered modality of mitochondrial cell death, might serve as a promising therapeutic target for cancer treatment and clinical outcome prediction. Nevertheless, the role of cuproptosis-related lncRNAs in AML is not fully understood.MethodsThe RNA sequencing data and demographic characteristics of AML patients were downloaded from The Cancer Genome Atlas database. Pearson correlation analysis, the least absolute shrinkage and selection operator algorithm, and univariable and multivariable Cox regression analyses were applied to identify the cuproptosis-related lncRNA signature and determine its feasibility for AML prognosis prediction. The performance of the proposed signature was evaluated via Kaplan-Meier survival analysis, receiver operating characteristic curves, and principal component analysis. Functional analysis was implemented to uncover the potential prognostic mechanisms. Additionally, quantitative real-time PCR (qRT-PCR) was employed to validate the expression of the prognostic lncRNAs in AML samples.ResultsA signature consisting of seven cuproptosis-related lncRNAs (namely NFE4, LINC00989, LINC02062, AC006460.2, AL353796.1, PSMB8-AS1, and AC000120.1) was proposed. Multivariable cox regression analysis revealed that the proposed signature was an independent prognostic factor for AML. Notably, the nomogram based on this signature showed excellent accuracy in predicting the 1-, 3-, and 5-year survival (area under curve = 0.846, 0.801, and 0.895, respectively). Functional analysis results suggested the existence of a significant association between the prognostic signature and immune-related pathways. The expression pattern of the lncRNAs was validated in AML samples.ConclusionCollectively, we constructed a prediction model based on seven cuproptosis-related lncRNAs for AML prognosis. The obtained risk score may reveal the immunotherapy response in patients with this disease.
| S-EPMC9896718 | biostudies-literature
Project description:BackgroundAcute Myeloid Leukemia (AML) generally has a relatively low survival rate after treatment. There is an urgent need to find new biomarkers that may improve the survival prognosis of patients. Machine-learning tools are more and more widely used in the screening of biomarkers.MethodsLeast Absolute Shrinkage and Selection Operator (LASSO), Support Vector Machine-Recursive Feature Elimination (SVM-RFE), Random Forest (RF), eXtreme Gradient Boosting (XGBoost), lrFuncs, IdaProfile, caretFuncs, and nbFuncs models were used to screen key genes closely associated with AML. Then, based on the Cancer Genome Atlas (TCGA), pan-cancer analysis was performed to determine the correlation between important genes and AML or other cancers. Finally, the diagnostic value of important genes for AML was verified in different data sets.ResultsThe survival analysis results of the training set showed 26 genes with survival differences. After the intersection of the results of each machine learning method, DNM1, MEIS1, and SUSD3 were selected as key genes for subsequent analysis. The results of the pan-cancer analysis showed that MEIS1 and DNM1 were significantly highly expressed in AML; MEIS1 and SUSD3 are potential risk factors for the prognosis of AML, and DNM1 is a potential protective factor. Three key genes were significantly associated with AML immune subtypes and multiple immune checkpoints in AML. The results of the verification analysis show that DNM1, MEIS1, and SUSD3 have potential diagnostic value for AML.ConclusionMultiple machine learning methods identified DNM1, MEIS1, and SUSD3 can be regarded as prognostic biomarkers for AML.
| S-EPMC10759623 | biostudies-literature
Project description:BackgroundCuproptosis is a type of programmed cell death that is involved in multiple physiological and pathological processes, including cancer. We constructed a prognostic cuproptosis-related long non-coding RNA (lncRNA) signature for acute myeloid leukemia (AML).MethodsRNA-seq and clinical data for AML patients were acquired from The Cancer Genome Atlas (TCGA) database. The cuproptosis-related prognostic lncRNAs were identified by co-expression and univariate Cox regression analysis. The least absolute shrinkage and selection operator (LASSO) was performed to construct a cuproptosis-related lncRNA signature, after which the AML patients were classified into two risk groups based on the risk model. Kaplan-Meier, ROC, univariate and multivariate Cox regression, nomogram, and calibration curves analyses were used to evaluate the prognostic value of the model. Then, expression levels of the lncRNAs in the signature were investigated in AML samples by quantitative polymerase chain reaction (qPCR). KEGG functional analysis, single-sample GSEA (ssGSEA), and the ESTIMATE algorithm were used to analyze the mechanisms and immune status between the different risk groups. The sensitivities for potential therapeutic drugs for AML were also investigated.ResultsFive hundred and three lncRNAs related to 19 CRGs in AML samples from the TCGA database were obtained, and 21 differentially expressed lncRNAs were identified based on the 2-year overall survival (OS) outcomes of AML patients. A 4-cuproptosis-related lncRNA signature for survival was constructed by LASSO Cox regression. High-risk AML patients exhibited worse outcomes. Univariate and multivariate Cox regression analyses demonstrated the independent prognostic value of the model. ROC, nomogram, and calibration curves analyses revealed the predictive power of the signature. KEGG pathway and ssGSEA analyses showed that the high-risk group had higher immune activities. Lastly, AML patients from different risk groups showed differential responses to various agents.ConclusionA cuproptosis-related lncRNA signature was established to predict the prognosis and inform on potential therapeutic strategies for AML patients.
| S-EPMC9585311 | biostudies-literature
Project description:BackgroundAcute myeloid leukemia (AML), a common form of acute leukemia, is due to tumor changes and clonal proliferation caused by genetic variants. Cuproptosis is a novel form of regulated cell death. This study aimed to explore the role of cuproptosis-related genes (CRGs) in AML.MethodsInitially, differentially expressed genes (DEGs) between AML samples and normal samples were obtained by differential analysis, which were further intersected with the cuproptosis score-related genes (CSRGs) acquired by weighted gene co-expression network analysis (WGCNA) to obtain cuproptosis score-related differentially expressed genes (CS-DEGs). Then, a risk model was constructed by Cox analysis and least absolute shrinkage and selection operator (LASSO) analysis. Finally, immune infiltration analysis was performed and the functions and pathways of model genes were explored by single sample gene set enrichment analysis (ssGSEA).ResultsThirty-two CS-DEGs were obtained by overlapping 11,160 DEGs and 132 CSRGs. These 32 CS-DEGs were mainly enriched to cytoplasmic microtubule organization, RNA methylation, mTOR signaling pathway, and notch signaling pathway. Two model genes, PACS2 and NDUFV1, were finally screened for the construction of the risk model. In addition, PACS2 and NDUFV1 were significantly positively correlated with activated B cells, CD56dim natural killer (NK) cells, and negatively correlated with effector memory CD4 T cells and activated CD4 T cells. PACS2 gene was significantly enriched to inositol phosphate metabolism, histone modification, etc. NDUFV1 was mainly enriched to ncRNA metabolic process, 2-oxocarboxylic acid metabolism, and other pathways.ConclusionsA cuproptosis-related risk model consisting of PACS2 and NDUFV1 was built, which provided a new direction for the diagnosis and treatment of AML.
| S-EPMC10493802 | biostudies-literature
Project description:BackgroundAcute myeloid leukemia (AML) is a highly heterogeneous cluster of hematologic malignancies. Leukemic stem cells (LSCs) are one of the culprits for the persistence and relapse of AML. The discovery of copper-induced cell death, namely cuproptosis, gives bright insights into the treatment of AML. Analogous to copper ions, long non-coding RNAs (lncRNAs) are not bystanders for AML progression, especially for LSC physiology. Uncovering the involvement of cuproptosis-related lncRNAs in AML will benefit clinical management.MethodsDetection of prognostic relevant cuproptosis-related lncRNAs are carried out by Pearson correlation analysis and univariate Cox analysis with RNA sequencing data of The Cancer Genome Atlas-Acute Myeloid Leukemia (TCGA-LAML) cohort. After the least absolute shrinkage and selection operator (LASSO) regression and the subsequent multivariate Cox analysis, a cuproptosis-related risk score (CuRS) system was derived to weigh the risk of AML patients. Thereafter, AML patients were classified into two groups by their risk property which was validated with principal component analysis (PCA), risk curves, Kaplan-Meier survival analysis, the combined receiver operating characteristic (ROC) curves, and nomogram. Variations in biological pathways and divergences in immune infiltration and immune-related processes between groups were resolved by GSEA and CIBERSORT algorism, respectively. Response to chemotherapies were scrutinized as well. The expression profiles of the candidate lncRNAs were examined by real-time quantitative polymerase chain reaction (RT-qPCR) and the specific mechanisms of lncRNA FAM30A were determined by transcriptomic analysis.ResultsWe fabricated an efficient prognostic signature named CuRS incorporating 4 lncRNAs (TRAF3IP2-AS1, NBR2, TP53TG1, and FAM30A) relevant to immune environment and chemotherapy responsiveness. The relevance of lncRNA FAM30A with proliferation, migration ability, Daunorubicin resistance and its reciprocal action with AUF1 were demonstrated in an LSC cell line. Transcriptomic analysis suggested correlations between FAM30A and T cell differentiation and signaling, intercellular junction genes.ConclusionsThe prognostic signature CuRS can guide prognostic stratification and personalized AML therapy. Analysis of FAM30A offers a foundation for investigating LSC-targeted therapies.
| S-EPMC10248575 | biostudies-literature
Project description:This paper identifies prognosis factors for survival in patients with acute myeloid leukemia (AML) using machine learning techniques. We have integrated machine learning with feature selection methods and have compared their performances to identify the most suitable factors in assessing the survival of AML patients. Here, six data mining algorithms including Decision Tree, Random Forrest, Logistic Regression, Naive Bayes, W-Bayes Net, and Gradient Boosted Tree (GBT) are employed for the detection model and implemented using the common data mining tool RapidMiner and open-source R package. To improve the predictive ability of our model, a set of features were selected by employing multiple feature selection methods. The accuracy of classification was obtained using 10-fold cross-validation for the various combinations of the feature selection methods and machine learning algorithms. The performance of the models was assessed by various measurement indexes including accuracy, kappa, sensitivity, specificity, positive predictive value, negative predictive value, and area under the ROC curve (AUC). Our results showed that GBT with an accuracy of 85.17%, AUC of 0.930, and the feature selection via the Relief algorithm has the best performance in predicting the survival rate of AML patients.
| S-EPMC8294525 | biostudies-literature
Project description:Acute myeloid leukemia (AML) is a clonal malignant proliferative blood disorder with a poor prognosis. Ferroptosis, a novel form of programmed cell death, holds great promise for oncology treatment, and has been demonstrated to interfere with the development of various diseases. A range of genes are involved in regulating ferroptosis and can serve as markers of it. Nevertheless, the prognostic significance of these genes in AML remains poorly understood. Transcriptomic and clinical data for AML patients were acquired from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO). Univariate Cox analysis was performed to identify ferroptosis-related genes with prognostic value, and the least absolute shrinkage and selection operator (LASSO) algorithm and stepwise multivariate Cox regression analysis were utilized to optimize gene selection from the TCGA cohort (132 samples) for model construction. Tumor samples from the GEO database (136 samples and 104 samples) were used as validation groups to estimate the predictive performance of the risk model. Finally, an eight-gene prognostic signature (including CHAC1, CISD1, DPP4, GPX4, AIFM2, SQLE, PGD, and ACSF2) was identified for the prediction of survival probability and was used to stratify AML patients into high- and low-risk groups. Survival analysis illustrated significantly prolonged overall survival and lower mortality in the low-risk group. The area under the receiver operating characteristic curve demonstrated good results for the training set (1-year: 0.846, 2-years: 0.826, and 3-years: 0.837), which verified the accuracy of the model for predicting patient survival. Independent prognostic analysis indicated that the model could be used as a prognostic factor (p ≤ 0.001). Functional enrichment analyses revealed underlying mechanisms and notable differences in the immune status of the two risk groups. In brief, we conducted and validated a novel ferroptosis-related prognostic model for outcome prediction and risk stratification in AML, with great potential to guide individualized treatment strategies in the future.
| S-EPMC8803125 | biostudies-literature
Project description:BackgroundNeutrophil extracellular traps (NETs) play pivotal roles in various pathological processes. The formation of NETs is impaired in acute myeloid leukemia (AML), which can result in immunodeficiency and increased susceptibility to infection.MethodsThe gene set variation analysis (GSVA) algorithm was employed for the calculation of NET score, while the consensus clustering algorithm was utilized to identify molecular subtypes. Weighted gene coexpression network analysis (WGCNA) revealed potential genes and biological pathways associated with NETs, and a total of 10 machine learning algorithms were applied to construct the optimal prognostic model.ResultsThrough the analysis of multiomics data, we identified two molecular subtypes with high and low NET scores. The low-NET score subgroup exhibited increased infiltration of immune effector cells. Conversely, the high-NET score subtype presented an abundance of monocytes and M2 macrophages, accompanied by elevated expression levels of immune checkpoint genes. These findings suggest that a pronounced immunosuppressive effect is associated with a significantly worse prognosis for this subtype. The optimal risk score model was selected by employing the C-index as the criterion on the basis of training 10 machine learning algorithms on 9 multicenter AML cohorts. Survival analysis confirmed that patients with high-risk scores had considerably poorer prognoses than those with lower scores. Receiver operating characteristic (ROC) curve and Cox regression analyses further validated the strong independent prognostic value of the risk score model. The nomogram, which was constructed by integrating the risk score model and clinicopathological factors, demonstrated high accuracy in predicting the overall survival of AML patients. Moreover, patients with refractory or chemotherapy-unresponsive AML had significantly higher risk scores. By analyzing drug therapy data from in vitro AML cells, we identified a subset of drugs that demonstrated increased sensitivity in the high-risk score group. Additionally, patients with a high risk score were also predicted to exhibit a favorable response to anti-PD-1 therapy, suggesting that these individuals may derive greater benefits from immunotherapy.ConclusionThe NET-related signature, derived from a combination of diverse machine learning algorithms, has promising potential as a valuable tool for prognostic prediction, preventive measures, and personalized medicine in patients with AML.
| S-EPMC11891243 | biostudies-literature
Project description:BackgroundAcute myeloid leukemia (AML) is a cancer arising in the bone marrow and is the most common type of adult leukemia. AML has a poor prognosis, and currently, its prognosis evaluation does not include immune status assessment. This study established an immune-related long non-coding RNA (lncRNA) prognostic risk model for AML based on immune lncRNAs screening.MethodsTo construct training and validation cohorts, The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) public databases were accessed to obtain gene expression profiles and clinical data. The correlation between lncRNAs and immunity genes was analyzed using the "limma" package, and the immune-related lncRNAs were obtained. Through least absolute shrinkage and selection operator regression, a prognostic model was established with immune-related lncRNAs. Using the median risk score, patients were divided into high- and low-risk groups. The Kaplan-Meier method was used for survival analysis, whereas the accuracy of the risk model was evaluated using time-dependent receiver operating characteristic curves, risk score distribution, survival status, and risk heat maps. We utilized univariate and multivariate Cox regression to examine the association between risk score and clinical variables and AML survival and prognosis.ResultsIn the immune-related lncRNA prognostic risk model, the prognosis was better for low-risk than for high-risk patients, indicating risk score of this model as an independent indicator of prognosis. The area under the curve value for 1-, 3-, and 5-year survival of TCGA patients was 0.817, 0.859, and 0.909, respectively, whereas that of GEO patients (of dataset GPL96-GSE37642) was 0.603, 0.652, and 0.624, respectively. Gene set enrichment analysis revealed the enrichment of multiple pathways, such as antigen processing, B-cell receptor signaling pathway, natural killer cell-mediated cytotoxicity, and chemokines, in high-risk patients.ConclusionsIn this study, immune-related lncRNA prognostic risk models effectively predicted AML survival and provided potential treatment targets.
| S-EPMC10774049 | biostudies-literature
Project description:BackgroundAlthough it is well-known that adult and pediatric acute myeloid leukemias (AMLs) are genetically distinct diseases, they still share certain gene expression profiles. The age-related genetic heterogeneities of AMLs have been well-studied, but the common prognostic signatures and molecular mechanisms of adult and pediatric AMLs are less investigated.AimTo identify genes and pathways that are associated with both pediatric and adult AMLs and discover a gene signature for overall survival (OS) prediction.MethodsThrough mining the transcriptome profiles of The Cancer Genome Atlas (TCGA) data sets of adult cancers and The Therapeutically Applicable Research to Generate Effective Treatments (TARGET) data of pediatric cancers, we identified genes that are commonly dysregulated in both pediatric and adult AMLs, further discovered a common gene signature, and built two risk score models for TCGA and TARGET cohorts, respectively with L 0 regularized global AUC (area under the receiver operating characteristic curve) summary maximization.ResultsWe identified 57 genes that are differentially expressed and prognostically significant in both adult and childhood AMLs. The top 4 Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways enriched with those 57 genes include transcriptional misregulation, focal adhesion, PI3K-Akt signaling pathway, and signaling pathways regulating pluripotency of stem cells. We further identified a 6-gene signature including genes of ADAMTS3, DNMT3B, NYNRIN, SORT1, ZFHX3, and ZG16B for risk prediction. We constructed a risk score model with one dataset (either TCGA or TARGET) and evaluated its performance with the other. The test AUCs for the risk prediction of TCGA data with a 2-year and 5-year OS cutoffs are 0.762 (P = 2.33e-13, 95% CI: 0.69-0.83) and 0.759 (P = 7.26e-08, 95% CI: 0.66-0.85), respectively, while the test AUCs of TARGET data with the same cutoffs are 0.71 (P = 3.3e-07, 95% CI: 0.62-0.79) and 0.72 (P= 5.25e-09, 95% CI: 0.65-0.80), respectively. We further stratified patients into 3 equal sized prognostic subtypes with the 6-gene risk scores. The P-values of the tertile partitions are 1.74e-07 and 3.28e-08 for the TARGET and TCGA cohorts, respectively, which are significantly better than the standard cytogenetic risk stratification of both cohorts (TARGET: P = 1.64e-06; TCGA: P = 1.79e-05). When validated with two other independent cohorts, the 6-gene risk score models remain a significant predictor for OS. Investigating the common gene expression program is significant in that we may extrapolate the findings from adults to children and avoid unnecessary pediatric clinical trials.
| S-EPMC9556437 | biostudies-literature