Report from ISMRM: PET/CT outperforms in whole body compared with MR
Despite playing before a partisan crowd at the International Society for Magnetic Resonance in Medicine meeting ast week in Kyoto, Japan, morphologic MR imaging was unable to surpass the inherent physiological properties of PET/CT in a head-to-head test to determine which is better for whole-body cancer staging.
With one notable exception, whole-body PET/CT was better than moving-table, contrast-enhanced MRI for identifying the extent and distribution of lesions associated with metastatic cancer.
Dr. Florian M. Vogt and colleagues at the University of Essen, Germany, promoted this clash of cross-sectional imaging heavyweights. In the MR corner, Vogt recruited a 1.5T high-performance scanner equipped with a BodySurf rolling-table platform and an integrated torso phased-array coil for whole-body coverage. The protocol featured unenhanced T1- and T2-weighted HASTE imaging of liver and thorax and was complemented by the acquisition of seven overlapping contrast-enhanced axial 3D-VIBE data sets for whole-body coverage.
In the opposite corner, Vogt chose a state-of-the-art PET/CT scanner. Imaging was performed one hour after intravenous injection of 350 MBq of F-18 FDG. CT was enhanced with 140 mL of an iodinated contrast agent.
Whole-body workups with the two modalities were performed on 98 patients with a diverse mix of cancers including bronchial carcinoma, suspected malignant melanoma, head and neck cancer, and carcinoma of unknown origin, Vogt said.
Readers blinded to the results of the competing procedures interpreted the images. Patient outcome, including histology and a clinical follow-up performed an average of 273 days after imaging, served as a reference standard.
Whole-body PET/CT prevailed over MR overall. Radiologists assessing the fusion PET/CT studies correctly determined tumor stage for 75 of the patients, compared with 53 for whole-body MRI. PET/CT influenced a change in clinical management of 12 patients, compared with two alternations of patient management for whole-body MRI, according to Vogt.
A separate assessment of T-stage revealed MRI to be accurate in 24 of 46 of patients, with pathological T-stage verification by PET/CT in 37 of 46 (p = 0.0002). N-stage was correctly determined by PET/CT in 93% of the cases compared with 79% with MR (p = 0.0013).
The two modalities performed equally well for assessing abdominal parenchymal organs.
MRI was superior to PET/CT imaging when determining the presence of bone metastases: 80 bone metastases were detected with MRI in 11 patients, while PET/CT revealed 75 lesions in eight patients.
MRI's superior performance for uncovering bone mets led Vogt to judge PET/CT the winner in a split decision. Rather than recommending the fusion technique in all situations, Vogt suggested that the two modalities play complementary roles. MRI should be used to uncover bone metastases, he said, while PET/CT should be applied to identify metastatic activity elsewhere in the body.
