Magnetic resonance imaging-based nomograms predict high-risk cytogenetic abnormalities in multiple myeloma: a...

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Multiple myeloma (MM) is a hematologic malignancy.
• Accurate prediction of high-risk cytogenetic abnormalities (HRCAs) is crucial for MM patient management.
• Current methods for identifying HRCAs can be invasive or time-consuming.

Purpose of the Study:

• To investigate the utility of magnetic resonance imaging (MRI) radiomic features for predicting HRCAs in MM patients.
• To develop and validate predictive models for HRCAs using radiomics and clinical data.
• To assess the potential of MRI radiomics in improving clinical decision-making and prognosis evaluation for MM.

Main Methods:

• Retrospective analysis of MRI data from 195 MM patients across two institutions.
• Radiomic features extracted from T1WI, T2WI, and FS-T2WI sequences.
• Logistic regression and 10-fold cross-validation used to build predictive models.
• Model performance evaluated using C-index, accuracy, sensitivity, specificity, and other metrics.
• External validation performed on a separate patient cohort.

Main Results:

• Combined radiomics and age models demonstrated strong performance in differentiating HRCAs.
• The best performing models achieved C-indexes of 0.88 (training) and 0.84 (validation) for multi-sequence MRI.
• External validation showed C-indexes ranging from 0.70 to 0.77 for different radiomics nomograms.
• MRI radiomics effectively predicted HRCAs in MM patients.

Conclusions:

• MRI radiomics is a viable tool for predicting HRCAs in multiple myeloma.
• These radiomic models can assist clinicians in treatment planning and prognostic assessment.
• This non-invasive approach may enhance patient outcomes by enabling timely and informed clinical decisions.