Individualized dynamic risk assessment and treatment selection for multiple myeloma

Area of Science:

• Genomics
• Systems Biology
• Computational Biology

Background:

• Accurate risk stratification is crucial for personalized treatment of multiple myeloma (MM).
• Cytogenetic abnormalities significantly impact MM disease progression.
• Existing risk stratification methods need improvement for individualized MM care.

Purpose of the Study:

• To develop a unified framework for cytogenetic subtype-specific risk classification in MM.
• To predict treatment response in MM patients using a machine learning approach.
• To improve individualized risk assessment throughout the MM disease trajectory.

Main Methods:

• Utilized the SYstems Genetic Network AnaLysis (SYGNAL) framework on multi-omics data from 881 MM patients.
• Generated a mmSYGNAL network of transcriptional programs associated with MM progression.
• Applied machine learning to mmSYGNAL program activity profiles for risk and treatment prediction.

Main Results:

• The mmSYGNAL risk models outperformed cytogenetics, ISS, and multi-gene panels in predicting progression-free survival (PFS) across diverse MM patient cohorts.
• Treatment response predictions showed significant concordance with drug efficacy in vitro.
• The model identified potential drug matches for MM patients, including those with relapsed refractory disease.

Conclusions:

• Transcriptional program activities provide superior prognostic and predictive assessments in MM.
• This framework enables more accurate, individualized risk stratification and treatment selection for MM patients.
• The findings support the use of transcriptional network analysis for personalized MM management.