Dissecting PCD-driven molecular landscapes in AML: a multi-omic framework for prognostication and therapeutic...

Area of Science:

• Hematology
• Oncology
• Molecular Biology

Background:

• Acute myeloid leukemia (AML) is a heterogeneous cancer with poor outcomes, requiring better biomarkers for risk assessment.
• Understanding the molecular drivers of AML heterogeneity is crucial for developing targeted therapies.

Purpose of the Study:

• To investigate the role of programmed cell death (PCD) pathways in AML heterogeneity and prognosis.
• To identify novel biomarkers for risk stratification and potential therapeutic targets in AML.

Main Methods:

• Integration of single-cell RNA-seq and multiomics data from 2,680 AML samples across 10 cohorts.
• Unsupervised clustering to identify PCD-driven AML subtypes.
• Development and validation of a machine learning-based prognostic model (PCDRScore) using PCD-related genes.

Main Results:

• Dysregulated PCD pathways are significantly elevated in AML and correlate with adverse outcomes.
• Two distinct PCD-driven subtypes were identified: Subtype A (high PCD, immunosuppressive microenvironment, poor survival) and Subtype B (lower PCD, immune-active, better prognosis).
• The PCDRScore model demonstrated superior prognostic performance compared to existing models and identified subtype-specific therapeutic vulnerabilities.

Conclusions:

• PCD dysregulation is a key factor in AML heterogeneity, influencing prognosis and treatment response.
• The PCDRScore offers a robust tool for precision risk stratification in AML.
• Findings support the development of immunophenotype-guided therapies and highlight potential targets like immune checkpoint inhibitors.