Prediction of ISS and R-ISS Stratification in Newly Diagnosed Multiple Myeloma Using Lumbar Spine MRI Radiomics...

Area of Science:

• Radiology
• Oncology
• Medical Imaging Analysis

Background:

• Limited availability of genetic testing hinders accurate staging for newly diagnosed multiple myeloma (ndMM).
• Accurate staging is crucial for treatment decisions and prognosis in ndMM patients.

Purpose of the Study:

• To develop and validate lumbar MRI-radiomics models for predicting International Staging System (ISS) and Revised International Staging System (R-ISS) stages in ndMM.
• To compare the performance of radiomics models, a clinical model, and a fusion model incorporating radiomics and biomarkers.

Main Methods:

• Retrospective analysis of 164 ndMM patients from two centers.
• Development of radiomics models using T1-weighted imaging (T1-WI) and T2-weighted fat-suppressed (T2-FS) features.
• Construction of a clinical model and a fusion model combining radiomics features with peripheral blood biomarkers.
• Evaluation of model performance using AUC, accuracy, sensitivity, and specificity on training, internal, and external test sets.

Main Results:

• The T1_WL radiomics model was effective for ISS stratification (AUCs: 0.743 internal, 0.707 external).
• The cross-region model demonstrated superior R-ISS stratification (AUCs: 0.814 internal, 0.763 external).
• The fusion model significantly outperformed both radiomics and clinical models (AUCs: 0.869 internal, 0.825 external), showing substantial net reclassification improvement.

Conclusions:

• Lumbar MRI-radiomics offers a practical, non-invasive method for risk stratification in ndMM, especially in resource-limited settings.
• The developed fusion model provides enhanced predictive accuracy for MM staging compared to conventional methods.
• This approach addresses the need for accessible and reliable staging tools in multiple myeloma management.