Objective: The purpose of this study was to compare the diagnostic accuracy of positron emission tomography (PET)/MRI with PET/CT for determining tumor resectability of non-small cell lung cancer (NSCLC).
Methods: Sequential trimodality PET/CT/MRI was performed in 36 patients referred with the clinical question of resectability assessment in NSCLC. PET/CT and PET/MR images including T1 weighted sequence (T1-Dixon) and respiration gated T2 weighted sequence (T2-Propeller) were evaluated for resectability-defining factors; i.e. longest diameter of the tumor, minimal tumor distance to the carina, mediastinal invasion, invasion of the carina, pleural infiltration, pericardial infiltration, diaphragm infiltration, presence of additional nodules.
Results: There was no significant difference of maximal axial diameter measurements of the primary lung tumors and narrow limits of agreement in Bland-Altman analysis ranging from -11.1 mm to + 11.8 mm for T2-Propeller and from -14.3 mm to + 13.8 mm for T1-Dixon sequence. A high agreement of PET/MR with PET/CT for the different resectability-defining factors was observed (k from 0.769 to 1.000). There was an excellent agreement of T2-Propeller sequence and CT for additional pulmonary nodule detection (k of 0.829 and 0.833), but only a moderate and good agreement using T1-Dixon sequence (k of 0.484 and 0.722).
Conclusion: In NSCLC the use of PET/MRI, including a dedicated pulmonary MR imaging protocol, provides a comparable diagnostic value for determination of tumor resectability compared to PET/CT.
Advances in knowledge: Our findings suggest that whole body PET/MRI can safely be used for the local staging of NSCLC patients. Further studies are warranted to determine whether it is feasible to integrate an imaging sequence in a whole body PET/MRI setting with the potential advantage of detection of liver or brain metastases.