Abstract Embryonal tumors with multilayered rosettes (ETMR) are highly aggressive pediatric brain tumors with dismal prognosis. Their early onset and histological features suggest an early developmental origin, though the implications of these developmental patterns for disease progression remain unclear. Recent work utilizing single-cell RNA sequencing (scRNA-seq) has uncovered distinct cell populations within ETMRs, showing that these tumors mirror the developmental trajectories of the brain (manuscript in press). To explore these patterns in a larger cohort, we analyzed DNA methylation profiles from 217 ETMR samples, including paired relapses, to assess their predominant cellular differentiation states. Our aim was to evaluate inter-patient variability in differentiation and its role in driving disease progression. Guided by scRNA-seq insights, we calculated differentiation scores for 26 samples with available bulk gene expression and DNA methylation data. For the remaining 191 profiles without gene expression data, we inferred differentiation scores from methylation data. To achieve this, we trained an elastic net model using machine learning on profiles with matched data types, employing expression-derived differentiation scores as the explanatory variable. Based on unsupervised clustering of the global methylation landscape of ETMRs, distinct subclusters emerged other than those resulting from samples lacking C19MC amplifications. These newly identified subclusters strongly correlated with the inferred predominant differentiation states. We identified an enrichment of differentially methylated CpGs associated with the WNT signaling pathway between stem-like and more differentiated tumors, suggesting a possible role for WNT signaling in driving inter-patient heterogeneity, independent of C19MC status. Furthermore, relapsed ETMR samples had lower differentiation scores, indicating that loss of differentiation is linked to relapse across larger cohorts of patients. In conclusion, our analysis of inferred differentiation scores highlights the role of differentiation loss in ETMR progression, particularly in relapsed cases. These findings underscore the complex interplay between differentiation, genetic alterations and disease progression in ETMRs.
Hof et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: