Differences in Deauville scores generated using 60- and 120-minute uptake times on Total-Body ¹⁸F-FDG PET/CT scans

Purpose: The 60 min uptake protocol for oncological ¹⁸F-FDG PET imaging has been standard of care since the 1990s. The protocol was primarily determined by factors including the sensitivity of the scanners of the era and the necessity of performing transmission scans prior to the injection of radiotracer. Subsequently, post-injection transmission or CT scanning were adopted, delayed imaging has been demonstrated to improve lesion detectability and characterization of findings in a wide range of clinical situations, and the implementation of Total-Body PET has enabled routine delayed imaging while retaining excellent image quality. However, the Deauville criteria, the recommended scoring system for ¹⁸F-FDG PET/CT lymphoma evaluation, assigns a score between 1 to 5 based on visual assessment of FDG uptake in lesions relative to the liver and blood pool, based on FDG PET scans acquired at 60 mins post injection (PI). In this preliminary analysis, we aim to evaluate the changes in Deauville scores on lymphoma PET scans acquired at 120 mins PI compared to scans acquired at 60 mins PI.

Methods: At our institution, 120 mins is the standard uptake time for oncologic Total-Body PET imaging. For lymphoma patients, we perform an additional scan at 60 mins PI in order to generate Deauville scores. Analysis was performed on data collected prospectively from lymphoma patients who consented to enroll in an IRB-approved registry-type study (#1470016).Thirty-eight ¹⁸F-FDG PET/CT studies were obtained in 22 patients (58.9±16.2 yrs; 12 females) with biopsy-proven lymphoma (8 with diffuse large B-cell lymphoma [DLBCL], 14 with other lymphoproliferative diagnoses). All the studies were acquired as dual-timepoint (TP) total-body scans, 20 min per TP, starting after a median uptake time of 60 min (range: 55-68 min; TP1) and 118 min (range: 115-124 min; TP2). The injected activity was 288.7±44.6 MBq. All scans were reconstructed using 3D OSEM (4 iterations, 20 subsets, 256x256 matrix). Scans were anonymized for subsequent independent review by two physicians (R1 & R2), who were blinded to all clinical data including scan TP. The readers recorded Deauville scores (DS) for each scan. Inter-rater agreement was assessed.

Results: At both TP1 and TP2, R1 & R2 were in agreement on the DS in 28/38 studies (74%), with inter-rater agreement of 0.62 (95%: 0.42-0.81) for TP1 and 0.61 (95%CI: 0.42-0.79) for TP2.Overall, DS from each reader was higher for TP2 compared to TP1: 4.08±1.05 vs. 3.87±0.99 (R1, p=0.048) and 4.24±0.94 vs. 4.08±1.00 (R2, p=0.091).

Conclusions: In this small cohort of lymphoma patients, Deauville scores assessed on delayed timepoint ¹⁸F-FDG Total-Body scans trended slightly higher than those assessed on standard 60-min scans. The difference introduced by the longer uptake time may be clinically relevant in certain instances. Further evaluation of these findings in terms of diagnostic and prognostic impact is under investigation.

• Download figure
• Open in new tab
• Download powerpoint