Comparison of retrospective PET and MRI-DWI (PET/MRI-DWI) image fusion with PET/CT and MRI-DWI in detection of cervical and endometrial cancer lymph node metastases

The aim of this study was to compare the accuracy of retrospective image fusion of PET/MRI-DWI with that of PET/CT and MRI-DWI alone in detecting metastatic lymph nodes in patients with cervical and endometrial carcinoma. Twenty-seven patients with endometrial (n = 14) and cervical (n = 13) cancer who had undergone preoperative MRI-DWI and PET/CT for staging were retrospectively evaluated. The accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of PET/CT, MRI-DWI, and PET/MRI-DWI image fusion were calculated on a per-patient basis and on a per-node basis. Histopathological and follow-up imaging results were used as the gold standard. On a per-patient basis PET/MRI-DWI had the same sensitivity (87.5 %), specificity (84.2 %), diagnostic accuracy (85.1 %), PPV (70 %), and NPV (94.1 %) as PET-CT, but on a per-node basis PET/MRI-DWI showed better sensitivity (89 vs 70.2 %), specificity (91.6 vs 90.5 %), diagnostic accuracy (91.2 vs 87 %), PPV (68.7 vs 60.4 %), and NPV (97.6 vs 93.6 %) than PET-CT. Comparison of the areas under the ROC curves for the detection of metastatic lymph nodes demonstrated a non-significant difference (p = 0.055) between PET/CT and fused PET/MRI-DWI. PET/MRI-DWI may be a valuable technique for N-staging patients with endometrial and cervical cancer, but more studies are needed to investigate its potential clinical utility.