Transforming Multiple Myeloma Care with AI and Digital Twin Technology
Overview
Multiple myeloma (MM) remains incurable due to its clinical and biological heterogeneity, complicating precise risk stratification and treatment selection. Artificial intelligence (AI) and digital twin (DT) technologies offer promising tools to integrate complex data, enabling dynamic risk assessment and personalized therapeutic strategies to improve patient outcomes.
Background
MM is a hematological malignancy characterized by malignant plasma cell proliferation with an annual global incidence of approximately 188,000 cases. Despite advances in novel and immune-based therapies, about 20% of newly diagnosed patients experience early relapse within 24 months, highlighting the need for better risk stratification. Traditional prognostic models have limitations in specificity and adaptability to evolving therapies. AI and DT technologies can integrate multi-omics, imaging, and longitudinal clinical data to create patient-specific virtual models that simulate disease progression and treatment response in real time.
Data Highlights
Risk Stratification System
Key Features
Year
Durie–Salmon Staging (DSS)
Tumor burden focused
1975
International Staging System (ISS)
Incorporates clinical parameters
2005
Revised ISS (R-ISS)
Includes cytogenetics and LDH
2015
Second Revision ISS (R2-ISS)
Further refinements
2022
Mayo Additive Staging System (MASS)
5-factor genomic abnormalities including del(17p), TP53 mutation
2022
IMWG/International Myeloma Society Definition
High-risk defined by del(17p) >20%, TP53 mutation, biallelic del(1p32), or combined intermediate-risk abnormalities
2025
Key Findings
MM exhibits significant clinical and biological heterogeneity, complicating risk stratification and treatment.
Approximately 20% of newly diagnosed MM patients relapse within 24 months, indicating functional high-risk disease.
Traditional prognostic models (DSS, ISS, R-ISS, R2-ISS, MASS) have evolved but still lack precision and adaptability.
AI can integrate large-scale multi-omics, imaging, and clinical data to develop advanced predictive models for MM.
Digital twins serve as dynamic virtual replicas of patients, enabling real-time simulation of disease progression and treatment response.
Continuous data acquisition synchronized with AI-driven digital twins allows iterative risk recalibration reflecting evolving disease biology.
Clinical Implications
Incorporating AI and digital twin technologies into clinical practice can enhance personalized risk stratification and optimize therapy selection in MM. These tools facilitate dynamic monitoring and prediction of disease progression, potentially improving outcomes by identifying high-risk patients earlier and tailoring treatment strategies accordingly.
Conclusion
AI and digital twin technologies represent transformative advances in managing multiple myeloma by addressing the challenges of heterogeneity and evolving disease biology. Their integration into clinical workflows promises more precise risk assessment and personalized therapeutic decision-making.
References
IMWG/International Myeloma Society 2025 -- Updated High-Risk MM Definition
Mayo Additive Staging System 2022 -- Genomic Risk Stratification in MM
International Staging System (ISS) 2005 and Revisions -- MM Prognostic Models
