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 crossvalidation 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.Acute lymphoblastic leukemia (ALL) is the most common malignant cancer among children 1 . Current risk-adapted treatments and supportive care have increased the survival rate to over 90% in the developed countries 2, 3 . However, approximately 20% of children who relapse have a poor prognosis, making ALL the leading cause of cancer mortality in pediatric disorders 4 . A major challenge in childhood ALL management is to classify patients into appropriate risk groups for better management. Stratifying chemotherapeutic treatment through the early recognition of relevant outcomes is critically important in order to mitigate poor disease courses in these patients 5 .Previous group-level studies have identified many potential prognostic factors for childhood ALL, such as white blood cell (WBC) counts, age at diagnosis, response to prednisone and some gene fusions like BCR-ABL, TEL-AML1 and E2A-PBX1. Moreover, immunophenotype (T cell or B cell), percentage of lymphoblast in bone marrow (BM) on day 15 and day 33, level of minimal residual disease (MRD) may also help to identify the probability of relapse risk for patients at early therapy 3,6,7 . However, despite insight into various prognostic features, there is no clear consensus regarding how and which of these features should be combined for prediction. Clinicians still lack accurate tools to estimate a patient's risk of ALL relapse in the early course of treatment.Machine learning is a data-driven analytic approach that specializes in the integration of multiple risk factors into a predictive tool 8 . The application of different techniques for feature selection and classification in multidimensional heterogeneous data can provide promising tools for inference in medicine. Over the past several decades, such ...
Quality assessments were performed with the Newcastle-Ottawa Scale. Heterogeneity was evaluated by Cochran's Q test and source of heterogeneity was detected by subgroup analysis and sensitivity analysis. Results: A total of seven studies involving 5183 participants were included in the meta-analysis. VDD was associated with an increased incidence of anemia (OR ¼ 2.25, 95% CI ¼ 1.47-3.44), with significant evidence of heterogeneity among these studies (p for heterogeneity 50.001, I 2 ¼ 84.0%). The subgroup and sensitivity analysis confirmed the stability of the results and no publication bias was detected. Conclusion: Our outcomes showed that VDD increased the risk of developing anemia. More researches are warranted to clarify an understanding of the association between VDD and risk of anemia.
Precursor B cell acute lymphoblastic leukemia (B-ALL) is a B cell-derived, malignant disorder with the highest incidence among children. In addition to the genetic abnormality, a dysregulated immune system also has an important role in the pathogenesis of B-ALL. Myeloid-derived suppressor cells (MDSCs) represent one of the key drivers in immune tolerance against tumor cells, including various solid tumors and hematologic malignancies. The role of MDSCs in B-ALL remains poorly understood. Here, we showed that the granulocytic (G)-MDSC population was significantly elevated in both the peripheral blood and BM of patients with B-ALL, when compared with age-matched healthy controls. G-MDSCs levels correlated positively with clinical therapeutic responses and B-ALL disease prognostic markers, including minimal residual disease, and the frequencies of CD20 and blast cells. The immunosuppressive function of B-ALL-derived G-MDSCs was mediated through the production of reactive oxygen species and required direct cell-cell contact, with the potential participation of STAT3 signaling. Overall, the results of our study support accumulation and activation of G-MDSCs as a novel mechanism of immune evasion of tumor cells in patients with B-ALL and may be a new therapeutic target.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.