Simple and robust methylation test for risk stratification of patients with juvenile myelomonocytic leukemia

Construction of the prediction model for multiple myeloma based on machine learning

INTRODUCTION

The global burden of multiple myeloma (MM) is increasing every year. Here, we have developed machine learning models to provide a reference for the early detection of MM.

METHODS

A total of 465 patients and 150 healthy controls were enrolled in this retrospective study. Based on the variable screening strategy of least absolute shrinkage and selection operator (LASSO), three prediction models, logistic regression (LR), support vector machine (SVM), and random forest (RF), were established combining complete blood count (CBC) and cell population data (CPD) parameters in the training set (210 cases), and were verified in the validation set (90 cases) and test set (165 cases). The performance of each model was analyzed using receiver operating characteristic (ROC) curve, calibration curves, and decision curve analysis (DCA). Accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and area under the ROC curve (AUC) were applied to evaluate the models. Delong test was used to compare the AUC of the models.

RESULTS

Six parameters including RBC (1012/L), RDW-CV (%), IG (%), NE-WZ, LY-WX, and LY-WZ were screened out by LASSO to construct the model. Among the three models, the AUC of RF model in the training set, validation set, and test set were 0.956, 0.892, and 0.875, which were higher than those of LR model (0.901, 0.849, and 0.858) and SVM model (0.929, 0.868, and 0.846). Delong test showed that there were significant differences among the models in the training set, no significant differences in the validation set, and significant differences only between SVM and RF models in the test set. The calibration curve and DCA showed that the three models had good validity and feasibility, and the RF model performed best.

CONCLUSION

The proposed RF model may be a useful auxiliary tool for rapid screening of MM patients.

Layered immunity and layered leukemogenicity: Developmentally restricted mechanisms of pediatric leukemia initiation

Hematopoietic stem cells (HSCs) and multipotent progenitor cells (MPPs) arise in successive waves during ontogeny, and their properties change significantly throughout life. Ontological changes in HSCs/MPPs underlie corresponding changes in mechanisms of pediatric leukemia initiation. As HSCs and MPPs progress from fetal to neonatal, juvenile and adult stages of life, they undergo transcriptional and epigenetic reprogramming that modifies immune output to meet age-specific pathogenic challenges. Some immune cells arise exclusively from fetal HSCs/MPPs. We propose that this layered immunity instructs cell fates that underlie a parallel layered leukemogenicity. Indeed, some pediatric leukemias, such as juvenile myelomonocytic leukemia, myeloid leukemia of Down syndrome, and infant pre-B-cell acute lymphoblastic leukemia, are age-restricted. They only present during infancy or early childhood. These leukemias likely arise from fetal progenitors that lose competence for transformation as they age. Other childhood leukemias, such as non-infant pre-B-cell acute lymphoblastic leukemia and acute myeloid leukemia, have mutation profiles that are common in childhood but rare in morphologically similar adult leukemias. These differences could reflect temporal changes in mechanisms of mutagenesis or changes in how progenitors respond to a given mutation at different ages. Interactions between leukemogenic mutations and normal developmental switches offer potential targets for therapy.

Genomic and Epigenomic Landscape of Juvenile Myelomonocytic Leukemia

Simple Summary

Juvenile myelomonocytic leukemia (JMML) is a rare pediatric myelodysplastic/myeloproliferative neoplasm characterized by the constitutive activation of the RAS pathway. In spite of the recent progresses in the molecular characterization of JMML, this disease is still a clinical challenge due to its heterogeneity, difficult diagnosis, poor prognosis, and the lack of curative treatment options other than hematopoietic stem cell transplantation (HSCT). In this review, we will provide a detailed overview of the genetic and epigenetic alterations occurring in JMML, and discuss their clinical relevance in terms of disease prognosis and risk of relapse after HSCT. We will also present the most recent advances on novel preclinical and clinical therapeutic approaches directed against JMML molecular targets. Finally, we will outline future research perspectives to further explore the oncogenic mechanism driving JMML leukemogenesis and progression, with special attention to the application of single-cell next-generation sequencing technologies.

Abstract

Juvenile myelomonocytic leukemia (JMML) is a rare myelodysplastic/myeloproliferative neoplasm of early childhood. Most of JMML patients experience an aggressive clinical course of the disease and require hematopoietic stem cell transplantation, which is currently the only curative treatment. JMML is characterized by RAS signaling hyperactivation, which is mainly driven by mutations in one of five genes of the RAS pathway, including PTPN11, KRAS, NRAS, NF1, and CBL. These driving mutations define different disease subtypes with specific clinico-biological features. Secondary mutations affecting other genes inside and outside the RAS pathway contribute to JMML pathogenesis and are associated with a poorer prognosis. In addition to these genetic alterations, JMML commonly presents aberrant epigenetic profiles that strongly correlate with the clinical outcome of the patients. This observation led to the recent publication of an international JMML stratification consensus, which defines three JMML clinical groups based on DNA methylation status. Although the characterization of the genomic and epigenomic landscapes in JMML has significantly contributed to better understand the molecular mechanisms driving the disease, our knowledge on JMML origin, cell identity, and intratumor and interpatient heterogeneity is still scarce. The application of new single-cell sequencing technologies will be critical to address these questions in the future.