Dynamic contrast-enhanced MRI model selection for predicting tumor aggressiveness in papillary thyroid cancers
Ramesh Paudyal1, Yonggang Lu2, Vaios Hatzoglou3
1Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, USA.
NMR in Biomedicine
|November 5, 2019
Summary
The extended Tofts model (ETM) best predicts papillary thyroid carcinoma (PTC) aggressiveness using dynamic contrast-enhanced MRI (DCE-MRI) parameters. ETM
Area of Science:
- Radiology
- Oncology
- Medical Imaging Analysis
Background:
- Papillary thyroid carcinoma (PTC) aggressiveness impacts treatment decisions.
- Accurate preoperative assessment of tumor behavior is crucial.
- Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) offers functional insights into tumors.
Purpose of the Study:
- To identify the optimal tracer kinetic model for DCE-MRI data in PTC.
- To evaluate if parameters from the optimal model predict tumor aggressiveness (histopathology) preoperatively.
- To assess the potential of DCE-MRI parameters in differentiating aggressive from non-aggressive PTC.
Main Methods:
- Prospective study of 18 PTC patients undergoing pretreatment 3T DCE-MRI.
- Voxel-wise comparison of four kinetic models (two-compartment, compartmental tissue uptake, extended Tofts model (ETM), standard Tofts model) using corrected Akaike information criterion (AICc).
- Receiver operating characteristic (ROC) curve analysis to evaluate model parameter performance in discriminating aggressive features, specifically extrathyroidal extension (ETE).
Main Results:
- The extended Tofts model (ETM) demonstrated the best performance with the lowest AICc and highest Akaike weight (0.44).
- ETM was the preferred model in 44% of voxels.
- ETM-estimated Ktrans was significantly higher in PTCs with ETE (0.78 min-1) compared to those without (0.34 min-1) (P = 0.005).
- ROC analysis showed high discrimination for Ktrans (AUC = 0.90) with a sensitivity of 86% and specificity of 82% for ETE detection.
Conclusions:
- The extended Tofts model (ETM) is the optimal kinetic model for analyzing DCE-MRI data in PTC.
- ETM-derived Ktrans shows significant potential for preoperatively classifying PTC aggressiveness, specifically identifying extrathyroidal extension (ETE).
- DCE-MRI with ETM analysis may improve preoperative risk stratification and surgical planning for PTC patients.
