Genetic program activity delineates risk, relapse, and therapy responsiveness in multiple myeloma

Area of Science:

• Genomics
• Bioinformatics
• Cancer Biology

Background:

• Multiple myeloma (MM) treatment faces challenges with patient relapse and therapy resistance.
• Predicting high-risk patients and understanding progression mechanisms are crucial for improving MM outcomes.

Purpose of the Study:

• To construct a transcriptional regulatory network (TRN) for MM using multi-omics data.
• To identify genetic programs and transcriptional states that predict clinical phenotypes and patient outcomes.
• To uncover mechanisms of MM relapse and identify potential therapeutic vulnerabilities.

Main Methods:

• Inferred a TRN from cross-sectional multi-omics data of 881 MM patients.
• Analyzed 124 chromosomal abnormalities and somatic mutations impacting 392 transcription regulators and 8549 genes.
• Identified 141 genetic programs and stratified patients into 25 transcriptional states.

Main Results:

• The TRN accurately predicted clinical phenotypes and patient outcomes.
• Identified genetic programs were more predictive of outcomes than mutations.
• Validated network coherence and risk prediction accuracy in independent datasets.
• Discovered subtype-specific drug vulnerabilities and relapse mechanisms, including immunosuppressive microenvironments.
• Identified an extreme-risk subgroup within the t(4;14) subtype with a median progression-free survival of 5 months.

Conclusions:

• The developed TRN provides a powerful tool for predicting multiple myeloma progression and outcomes.
• Identified genetic programs and transcriptional states offer novel insights into MM pathogenesis and relapse.
• Revealed subtype-specific vulnerabilities and potential therapeutic targets for improved MM management.