A multi-omic single-cell landscape reveals transcription and epigenetic regulatory features of t(8;21) AML.
Comprehensive analysis of single-cell transcriptome and chromatin accessibility will contribute to interpret the heterogeneity of acute myeloid leukemia (AML). We hypothesize that integrating single-cell transcriptomic and chromatin accessibility landscapes underlying t(8;21) AML will provide valuable insights into its heterogeneous cellular properties and gene regulatory programs.
Here, we conducted paired single-cell RNA-sequencing (scRNA-seq) and single-cell ATAC sequencing on bone marrow samples from newly diagnosed t(8;21) AML patients and healthy controls. Genetic signatures extracted from scRNA-seq were built and validated across three independent cohorts (German AMLCG1999, GSE106291 and TCGA LAML).
We identified TCF12, a core component of AML1-ETO-containing transcription factor complex (AETFC), as the most active transcription factor in blast cells, driving a universally repressed chromatin state. Furthermore, we delineated two functionally distinct T cell subsets, revealing that EOMES-mediated transcriptional regulation promotes the expansion of a cytotoxic T cell population (T cells_2; high GNLY, NKG7 and GZMB expression), with an increased clonality and a tendency for drug resistance. In addition, we discovered a novel leukemic CMP-like cluster characterized by high TPSAB1, HPGD and FCER1A expression. Leveraging machine learning-based integration of multi-omic profiles, we identified a robust 9-gene prognostic signature, which demonstrated significant predictive value for AML outcomes across three independent cohorts.
This multi-omics study provides unprecedented insights into the transcriptional and epigenetic heterogeneity of t(8;21) AML, providing a potential actionable tool for clinical risk stratification.
Here, we conducted paired single-cell RNA-sequencing (scRNA-seq) and single-cell ATAC sequencing on bone marrow samples from newly diagnosed t(8;21) AML patients and healthy controls. Genetic signatures extracted from scRNA-seq were built and validated across three independent cohorts (German AMLCG1999, GSE106291 and TCGA LAML).
We identified TCF12, a core component of AML1-ETO-containing transcription factor complex (AETFC), as the most active transcription factor in blast cells, driving a universally repressed chromatin state. Furthermore, we delineated two functionally distinct T cell subsets, revealing that EOMES-mediated transcriptional regulation promotes the expansion of a cytotoxic T cell population (T cells_2; high GNLY, NKG7 and GZMB expression), with an increased clonality and a tendency for drug resistance. In addition, we discovered a novel leukemic CMP-like cluster characterized by high TPSAB1, HPGD and FCER1A expression. Leveraging machine learning-based integration of multi-omic profiles, we identified a robust 9-gene prognostic signature, which demonstrated significant predictive value for AML outcomes across three independent cohorts.
This multi-omics study provides unprecedented insights into the transcriptional and epigenetic heterogeneity of t(8;21) AML, providing a potential actionable tool for clinical risk stratification.