Browse
Submit Data
Databases
API
Help

Dataset Information

13 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

Machine learning applications for prediction of relapse in childhood acute lymphoblastic leukemia.

ABSTRACT: The prediction of relapse in childhood acute lymphoblastic leukemia (ALL) is a critical factor for successful treatment and follow-up planning. Our goal was to construct an ALL relapse prediction model based on machine learning algorithms. Monte Carlo cross-validation nested by 10-fold cross-validation was used to rank clinical variables on the randomly split training sets of 336 newly diagnosed ALL children, and a forward feature selection algorithm was employed to find the shortest list of most discriminatory variables. To enable an unbiased estimation of the prediction model to new patients, besides the split test sets of 150 patients, we introduced another independent data set of 84 patients to evaluate the model. The Random Forest model with 14 features achieved a cross-validation accuracy of 0.827 ± 0.031 on one set and an accuracy of 0.798 on the other, with the area under the curve of 0.902 ± 0.027 and 0.904, respectively. The model performed well across different risk-level groups, with the best accuracy of 0.829 in the standard-risk group. To our knowledge, this is the first study to use machine learning models to predict childhood ALL relapse based on medical data from Electronic Medical Record, which will further facilitate stratification treatments.

SUBMITTER: Pan L

PROVIDER: S-EPMC5547099 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Json Xml

Similar Datasets

Prediction of tumor lysis syndrome in childhood acute lymphoblastic leukemia based on machine learning models: a retrospective study.

Project description:BackgroundTumor lysis syndrome (TLS) often occurs early after induction chemotherapy for acute lymphoblastic leukemia (ALL) and can rapidly progress. This study aimed to construct a machine learning model to predict the risk of TLS using clinical indicators at the time of ALL diagnosis.MethodsThis observational cohort study was conducted at the National Clinical Research Center for Child Health and Disease. Data were collected from pediatric ALL patients diagnosed between December 2008 and December 2021. Four machine learning models were constructed using the Least Absolute Shrinkage and Selection Operator (LASSO) to select key clinical indicators for model construction.ResultsThe study included 2,243 pediatric ALL patients, and the occurrence of TLS was 8.87%. A total of 33 indicators with missing values ≤30% were collected, and 12 risk factors were selected through LASSO regression analysis. The CatBoost model with the best performance after feature screening was selected to predict the TLS of ALL patients. The CatBoost model had an AUC of 0.832 and an accuracy of 0.758. The risk factors most associated with TLS were the absence of potassium, phosphorus, aspartate transaminase (AST), white blood cell count (WBC), and urea levels.ConclusionWe developed the first TLS prediction model for pediatric ALL to assist clinicians in risk stratification at diagnosis and in developing personalized treatment protocols. This study is registered on the China Clinical Trials Registry platform (ChiCTR2200060616).Clinical trial registrationhttps://www.chictr.org.cn/, identifier ChiCTR2200060616.

| S-EPMC10955075 | biostudies-literature

Can Machine Learning Models Predict Asparaginase-associated Pancreatitis in Childhood Acute Lymphoblastic Leukemia.

Project description:Asparaginase-associated pancreatitis (AAP) frequently affects children treated for acute lymphoblastic leukemia (ALL) causing severe acute and persisting complications. Known risk factors such as asparaginase dosing, older age and single nucleotide polymorphisms (SNPs) have insufficient odds ratios to allow personalized asparaginase therapy. In this study, we explored machine learning strategies for prediction of individual AAP risk. We integrated information on age, sex, and SNPs based on Illumina Omni2.5exome-8 arrays of patients with childhood ALL (N=1564, 244 with AAP 1.0 to 17.9 yo) from 10 international ALL consortia into machine learning models including regression, random forest, AdaBoost and artificial neural networks. A model with only age and sex had area under the receiver operating characteristic curve (ROC-AUC) of 0.62. Inclusion of 6 pancreatitis candidate gene SNPs or 4 validated pancreatitis SNPs boosted ROC-AUC somewhat (0.67) while 30 SNPs, identified through our AAP genome-wide association study cohort, boosted performance (0.80). Most predictive features included rs10273639 (PRSS1-PRSS2), rs10436957 (CTRC), rs13228878 (PRSS1/PRSS2), rs1505495 (GALNTL6), rs4655107 (EPHB2) and age (1 to 7 y). Second AAP following asparaginase re-exposure was predicted with ROC-AUC: 0.65. The machine learning models assist individual-level risk assessment of AAP for future prevention trials, and may legitimize asparaginase re-exposure when AAP risk is predicted to be low.

| S-EPMC8946594 | biostudies-literature

Relapse-specific mutations in NT5C2 in childhood acute lymphoblastic leukemia.

Project description:Relapsed childhood acute lymphoblastic leukemia (ALL) carries a poor prognosis, despite intensive retreatment, owing to intrinsic drug resistance. The biological pathways that mediate resistance are unknown. Here, we report the transcriptome profiles of matched diagnosis and relapse bone marrow specimens from ten individuals with pediatric B-lymphoblastic leukemia using RNA sequencing. Transcriptome sequencing identified 20 newly acquired, novel nonsynonymous mutations not present at initial diagnosis, with 2 individuals harboring relapse-specific mutations in the same gene, NT5C2, encoding a 5'-nucleotidase. Full-exon sequencing of NT5C2 was completed in 61 further relapse specimens, identifying additional mutations in 5 cases. Enzymatic analysis of mutant proteins showed that base substitutions conferred increased enzymatic activity and resistance to treatment with nucleoside analog therapies. Clinically, all individuals who harbored NT5C2 mutations relapsed early, within 36 months of initial diagnosis (P = 0.03). These results suggest that mutations in NT5C2 are associated with the outgrowth of drug-resistant clones in ALL.

| S-EPMC3681285 | biostudies-literature

FPGS relapse-specific mutations in relapsed childhood acute lymphoblastic leukemia.

Project description:Although the cure rate for childhood acute lymphoblastic leukemia (ALL) has exceeded 80% with contemporary therapy, relapsed ALL remains a leading cause of cancer-related death in children. Relapse-specific mutations can be identified by comprehensive genome sequencing and might have clinical significance. Applying whole-exome sequencing to eight triplicate samples, we identified in one patient relapse-specific mutations in the folylpolyglutamate synthetase (FPGS) gene, whose product catalyzes the addition of multiple glutamate residues (polyglutamation) to methotrexate upon their entry into the cells. To determine the prevalence of mutations of the FPGS mutations, and those of two important genes in the thiopurine pathway, NT5C2 and PRPS1, we studied 299 diagnostic and 73 relapsed samples in 372 patients. Three more FPGS mutants were identified in two patients, NT5C2 mutations in six patients, and PRPS1 mutants in two patients. One patient had both NT5C2 and PRPS1 mutants. None of these alterations were detected at diagnosis with a sequencing depth of 1000X, suggesting that treatment pressure led to increased prevalence of mutations during therapy. Functional characterization of the FPGS mutants showed that they directly resulted in decreased enzymatic activity, leading to significant reduction in methotrexate polyglutamation, and therefore likely contributed to drug resistance and relapse in these cases. Thus, besides genomic alterations in thiopurine metabolizing enzymes, the relapse-specific mutations of FPGS represent another critical mechanism of acquired antimetabolite drug resistance in relapsed childhood ALL.

| S-EPMC7374087 | biostudies-literature

Down-Regulated FOXO1 in Refractory/Relapse Childhood B-Cell Acute Lymphoblastic Leukemia.

Project description:Background: Acute lymphoblastic leukemia (ALL) is the most common pediatric cancer, with an overall prevalence of 4/100,000, accounting for 25-30% of all childhood cancers. With advances in childhood ALL treatment, the cure rate for childhood ALL has exceeded 80% in most countries. However, refractory/relapsed ALL remains a leading cause of treatment failure and subsequent death. Forkhead box O1 (FOXO1) belongs to the forkhead family of transcription factors, but its role in B-cell ALL (B-ALL) has not been determined yet. Procedures: RNA sequencing was applied to an ALL case with induction failure (IF) to identify the possible genetic events. A cytokine-dependent growth assay in Ba/F3 cells was used to test the leukemic transformation capacity of MEIS1-FOXO1. The propidium iodide (PI) staining method was used to evaluate the effect of MEIS1-FOXO1 on cycle distribution. FOXO1 transactivity was examined using a luciferase reporter assay. FOXO1 mRNA expression levels were examined using real-time quantitative PCR among 40 children with B-ALL treated with the CCCG-ALL-2015 protocol. Association analysis was performed to test the correlation of FOXO1 transcription with childhood B-ALL prognosis and relapse in a series of GEO datasets. An MTT assay was performed to test the drug sensitivity. Results: In this ALL case with IF, we identified a novel MEIS1-FOXO1 fusion gene. The transactivity of MEIS1-FOXO1 was significantly lower than that of wild-type FOXO1. MEIS1-FOXO1 potentiated leukemia transformation and promoted Ba/F3 cell cycle S-phase entry. Low FOXO1 transcription levels were found to be strongly associated with unfavorable ALL subtype, minimal residual disease (MRD) positivity, and relapse. Lower FOXO1 expression was associated with prednisone and cyclophosphamide resistance. Conclusions: Low FOXO1 transcription was associated with high-risk stratification and relapse in children with B-ALL, probably due to multi-drug resistance.

| S-EPMC7686545 | biostudies-literature

Drug Resistance Biomarkers and Their Clinical Applications in Childhood Acute Lymphoblastic Leukemia.

Project description:Biomarkers are biological molecules found in body fluids or tissues, which can be considered as indications of a normal or abnormal process, or of a condition or disease. There are various types of biomarkers based on their application and molecular alterations. Treatment-sensitivity or drug resistance biomarkers include prognostic and predictive molecules with utmost importance in selecting appropriate treatment protocols and improving survival rates. Acute lymphoblastic leukemia (ALL) is the most prevalent hematological malignancy diagnosed in children with nearly 80% cure rate. Despite the favorable survival rates of childhood ALL (chALL), resistance to chemotherapeutic agents and, as a consequence, a dismal prognosis develops in a significant number of patients. Therefore, there are urgent needs to have robust, sensitive, and disease-specific molecular prognostic and predictive biomarkers, which could allow better risk classification and then better clinical results. In this article, we review the currently known drug resistance biomarkers, including somatic or germ line nucleic acids, epigenetic alterations, protein expressions and metabolic variations. Moreover, biomarkers with potential clinical applications are discussed.

| S-EPMC6978753 | biostudies-literature

Proteogenomics of High hyperdiploid childhood acute lymphoblastic leukemia

Project description:Proteogenomic analysis and genomic profiling, RNA-sequencing, and mass spectrometry-based analysis of High hyperdiploid childhood acute lymphoblastic leukemia.

2019-02-25 | PXD010175 | Pride

Genome-wide association study identifies germline polymorphisms associated with relapse of childhood acute lymphoblastic leukemia.

Project description:With the use of risk-directed therapy for childhood acute lymphoblastic leukemia (ALL), outcome has improved dramatically in the past 40 years. However, a substantial portion of patients, many of whom have no known risk factors, experience relapse. Taking a genome-wide approach, in the present study, we evaluated the relationships between genotypes at 444 044 single nucleotide polymorphisms (SNPs) with the risk of relapse in 2535 children with newly diagnosed ALL after adjusting for genetic ancestry and treatment regimen. We identified 134 SNPs that were reproducibly associated with ALL relapse. Of 134 relapse SNPs, 133 remained prognostic after adjusting for all known relapse risk factors, including minimal residual disease, and 111 were significant even among patients who were negative for minimal residual disease after remission induction therapy. The C allele at rs7142143 in the PYGL gene was associated with 3.6-fold higher risk of relapse than the T allele (P = 6.7 × 10(-9)). Fourteen of the 134 relapse SNPs, including variants in PDE4B and ABCB1, were also associated with antileukemic drug pharmacokinetics and/or pharmacodynamics. In the present study, we systematically identified host genetic variations related to treatment outcome of childhood ALL, most of which were prognostic independent of known risk factors for relapse, and some of which also influenced outcome by affecting host dis-position of antileukemic drugs. All trials are registered at www.clinicaltrials.gov or www.cancer.gov (COG P9904: NCT00005585; COG P9905: NCT00005596; COG P9906: NCT00005603; St Jude Total XIIIB: NCI-T93-0101D; and St Jude Total XV: NCT00137111).

| S-EPMC3501717 | biostudies-literature

CREBBP HAT domain mutations prevail in relapse cases of high hyperdiploid childhood acute lymphoblastic leukemia.

Project description:Despite their apparently good prognosis ?15% of high hyperdiploid (HD) childhood acute lymphoblastic leukemia (ALL) cases relapse. To search for responsible risk factors we determined copy number aberrations as well as copy neutral loss of heterozygosity (LOH) in 13 matched diagnosis and relapse samples and added the data of the only three available cases from the literature. Deletions and copy neutral LOH in 3 and 2 of the 16 cases directed us to the histone-modifying CREB-binding protein (CREBBP) gene, whose functional impairment is implicated in drug resistance. We therefore screened all samples for mutations in this gene and discovered 9 acquired sequence mutations in 7/16 cases, leading to an overall frequency of somatic CREBBP aberrations in HD ALL relapse cases of 63% that is considerably higher than that of the reported, mainly non-HD ALL (18.3%). Moreover, mutations in HD cases occur almost exclusively in the HAT domain (8/9; 89%). Hot spot mutations are present at diagnosis in 18.8% of relapsing HD ALL cases but in none of 40 respective cases remaining in long-term remission. Thus, the particular high incidence of CREBBP mutations in relapse-prone HD ALL cases could eventually be exploited for refined risk stratification and customized treatment in this genetic subgroup.

| S-EPMC4194312 | biostudies-literature

Biologic pathways associated with relapse in childhood acute lymphoblastic leukemia: a Children's Oncology Group study.

Project description:Outcome for children with childhood acute lymphoblastic leukemia (ALL) who relapse is poor. To gain insight into the mechanisms of relapse, we analyzed gene-expression profiles in 35 matched diagnosis/relapse pairs as well as 60 uniformly treated children at relapse using the Affymetrix platform. Matched-pair analyses revealed significant differences in the expression of genes involved in cell-cycle regulation, DNA repair, and apoptosis between diagnostic and early-relapse samples. Many of these pathways have been implicated in tumorigenesis previously and are attractive targets for intervention strategies. In contrast, no common pattern of changes was observed among late-relapse pairs. Early-relapse samples were more likely to be similar to their respective diagnostic sample while we noted greater divergence in gene-expression patterns among late-relapse pairs. Comparison of expression profiles of early- versus late-relapse samples indicated that early-relapse clones were characterized by overexpression of biologic pathways associated with cell-cycle regulation. These results suggest that early-relapse results from the emergence of a related clone, characterized by the up-regulation of genes mediating cell proliferation. In contrast, late relapse appears to be mediated by diverse pathways.

| S-EPMC1895482 | biostudies-literature

OmicsDI is part of the ELIXIR infrastructure

OmicsDI is an Elixir interoperability service. Learn more ›

Tweets

OmicsDI Databases

PRIDE
PeptideAtlas
MassIVE
JPOST Repository
Physiome Model Repository

EGA
EVA
ENA
LINCS
PAXDB
Cell Collective

MetaboLights
Metabolomics Workbench
MetabolomeExpress
GNPS
BioModels
FAIRDOMHub

ArrayExpress
dbGaP
ExpressionAtlas
GEO
NODE

Information

Databases
Help
API
Contact us
Code on GitHub
Terms of use
Submit Data