RT Journal Article SR Electronic T1 Comparison of 18F-FDG PET/MR and PET/CT in the diagnosis of neck recurrence or metastasis of differentiated thyroid cancer after operation JF Journal of Nuclear Medicine JO J Nucl Med FD Society of Nuclear Medicine SP 85 OP 85 VO 62 IS supplement 1 A1 Yangmeihui Song A1 Fang Liu A1 Weiwei Ruan A1 Fan Hu A1 Kevin Fan A1 Jie Ming A1 Tao Huang A1 Weibo Cai A1 Zairong Gao A1 Xiaoli Lan YR 2021 UL http://jnm.snmjournals.org/content/62/supplement_1/85.abstract AB 85Objectives: To explore the clinical value of 18F-FDG PET/MR and PET/CT in neck recurrence or metastasis of patients with differentiated thyroid carcinoma (DTC) after operation. Methods: 25 patients with suspected recurrence or cervical lymph node metastasis after comprehensive treatment of DTC underwent 18F-FDG PET/CT scan and neck PET/MR scan. The gold standard was: histopathological diagnosis of recurrence or metastasis; follow-up results of clinical (Tg and ATg levels) and imaging (neck ultrasound, lung CT, etc.) for 6 months or more. The size, SUVmax and SUVmean of lesions were measured and the detection, diagnostic efficiency and parameter characteristics of 18F-FDG PET/MR and 18F-FDG PET/CT in patients with DTC after operation were analyzed. Results: Among the 25 patients, 19 patients were diagnosed as recurrence or metastasis after operation. No recurrence or metastasis was found in other 6 cases. A total of 94 lesions (7 in the original thyroid region and 87 in the cervical lymph nodes) were included, including 46 malignant lesions (3 in the original thyroid region and 43 in the metastatic cervical lymph nodes), and 48 benign lesions (4 in the original thyroid region and 44 in the benign cervical lymph nodes). The detection rates of malignant lesions by PET/MR and PET/CT were 96% and 89%. Compared with PET/CT, PET/MR presented higher sensitivity, negative predictive value and accuracy (85.71% vs. 69.05%, 86.67% vs. 76.36%, 86.21% vs. 81.61%), but lower specificity and positive predictive value (86.67% vs. 93.33%, 85.71% vs. 90.63%). There was significant difference between the two imaging methods (P < 0.05), but there was no significant difference in sensitivity and specificity (P = 0.210, P = 0.125). SUVmax and SUVmean in PET/MR or PET/CT, and lymph node long diameter values in PET/CT were significantly different in benign and malignant lymph nodes (P < 0.05), while PET/MR lymph node long diameter values were not statistically significant. For the same lesion, the difference in SUVmax measured by PET/MR and PET/CT was statistically significant, and the difference in SUVmax of malignant lesions was the largest (24.3%); SUVmax measured by two imaging methods had significant correlation (r2 = 0.634). The area under the ROC curve (AUC) of PET/MR SUVmax and PET/CT SUVmax for the diagnosis of metastatic aligned lymph nodes were 0.79 and 0.69, and the cut-off values were 2.95 and 1.65. Conclusions: Compared with 18F-FDG PET/CT, 18F-FDG PET/MR can detect more recurrent DTC or cervical lymph node metastasis. Although the diagnostic performance of 18F-FDG PET/MR is equivalent, 18F-FDG PET/MR presented better sensitivity and accuracy, and PET/MR SUVmax has more diagnostic accuracy. Therefore, adding local PET/MR after whole-body PET/CT not only does not increase the absorbed radiation dose, but also can take advantage of the multi-position, multi-parameter, and multi-sequence advantages of MR to diagnose neck recurrence or metastasis of differentiated thyroid cancer after operation. View this table:Table 1 Patient characteristics View this table:Table 2 Number of recurrence or metastasis identified by imaging on per-patient and per-lesion basis View this table:Table 3 Location of dissected nodes and positive neck nodes per level. View this table:Table 4 Performance evaluation of the two modalities as assessed View this table:Table 5 SUVs, long axis of nodes and the coefficients as measured in a lesion-based analysis