Detection of suspicious lesions on F-DOPA PET in case of suspected recurrence of medullary thyroid cancer: comparison of intrapatient performance between early and late acquisitions

Introduction: Currently, international recommendations propose to perform 18F-fluorodihydroxyphenylalanine (F DOPA) PET/CT scans in case of suspected recurrence of medullary thyroid cancer (MTC) at 30-60 minutes post-injection. However, several authors have shown that early acquisitions perform better for detection of suspicious lesions of recurrence.

The aim of this retrospective study was to assess the performance in our experience of delayed versus early acquisition of ¹⁸F-FDOPA PET/CT for the detection of MTC suspected lesions of tumoral recurrence.

Methods: We retrospectively collected all the F DOPA PET/CT scans of patients with suspected relapse of MTC in our center between January 2015 and July 2021. Early (carried out between 3 and 20 minutes after injection of the radiotracer) and delayed (between 21 and 40 minutes post-injection) acquisitions were then reviewed independently by three “in training” nuclear medicine physicians. For each suspected lesion, its location and semi quantitative parameters were recorded. All reviewers were blinded to the clinical context and those in training, analysed only one of the acquisition times per patient study in order to avoid information leakage from one acquisition time to another. In a second step, three nuclear medicine experts, independently reviewed all cases to validate or not the lesions found. In case of discrepancy with the first assessment, a consensus categorization was performed. Finally, after consensus round, the discordance between early and late acquisitions was recorded and one acquisition time was considered better than the other if more suspicious of relapse lesions were found.

Results: 185 patients initially were retrieved in our database, finally 81 patients were included, corresponding to 135 dual time-point imaging studies (104 excluded due to lack of available imaging studies or exams not performed in our institution). After review by consensus, early acquisition imaging was better for 29% (39/135 IC95 [22 - 37]), identical for 67% (91/135 – IC95 [59 - 75]) and better at delayed acquisition for 4% (6/135 – IC95 [1 - 9]) of the studies.

Number or nature of lesions found was changed after consensus review in 77% of imaging studies (104/135 – IC95 [69 – 84]. After qualitative examination, the “in training” nuclear medicine physicians frequently reported an excessive number of lesions (e.g. considering reactive lymph nodes in the cervical region as pathological or classifying low uptake foci as a tumor due to the well-known low avidity of medullary thyroid cancer’s metastatic lesions). Moreover, the number of misinterpretation was greater in the delayed acquisition.

Conclusions: For the detection of MTC lesions, in this retrospective analysis, early 18F-FDOPA PET/CT acquisition was equivalent or superior to delayed acquisition in most cases. It is likely that delayed acquisition imaging could be removed from the standard acquisition protocol, thereby optimizing the PET/CT scan workflow by reducing scan time by 50% without compromising detection performance. Prospective studies are needed in this setting. Due to the relatively low avidity of secondary CMT lesions for F DOPA, rigorous training is required to reliably interpret this type of examination.

• Download figure
• Open in new tab
• Download powerpoint