Background: The role of 18F-fluorodeoxyglucosepositron emission tomography/computed tomography (18F-FDG PET/CT) for staging and response assessment in marginal zone lymphomas (MZL) is controversial, and its routine use is not recommended in the current Lugano classification.

Methods: The present analysis is based on 156 patients with histologically confirmed MZL who were enrolled in the retrospective IELSG44/PIMENTO study conducted across 21 centers in 4 countries. The cohort included 88 extranodal MZL, 43 nodal MZL, 21 splenic MZL, and 4 MZL not otherwise specified. All patients underwent baseline and end-of-treatment (EOT) 18F-FDG PET/CT scans and clinical follow-up. All scans were centrally reviewed by a panel of four expert nuclear medicine physicians. The maximum standardized uptake value (SUVmax) was recorded for all detectable lesions, and post-treatment changes (ΔSUVmax) were calculated. Response was assessed using the Deauville 5-point score (DS), with disagreements resolved by consensus.

Results: After a median follow-up of 2.7 years (IQR: 1.4–4.5), 33 patients experienced disease progression. There were 4 deaths attributed to the disease and 4 from other causes (2 infections, 2 unknown). All patients had at least one FDG-positive lesion detected on baseline PET/CT. The median baseline SUVmax was 8.4 (IQR: 5.7–11.3; range: 1.1–65.5). In 145 of 156 cases (93%), the hottest lesion showed uptake greater than liver activity; in 96 cases (62%), the uptake was more than twice that of the liver. In 150 patients (96%), the diagnostic biopsy was obtained from FDG-positive lesions. In 3 patients, the biopsy site was metabolically negative; 2 of these patients had circulating tumor cells, and 1 had splenic MZL diagnosed after splenectomy (before PET/CT).

A complete metabolic response (CMR), assessed at EOT by the Lugano criteria as DS 1–3 versus 4–5, did not significantly predict outcomes. The 5-year time to progression (TTP) was 66% in patients with DS 1–3 and 63% in those with DS 4–5 (p = 0.629). The 5-year overall survival (OS) was 89% versus 92%, respectively (p = 0.613). This lack of prognostic value remained even when restricting the analysis to patients with baseline lesions showing uptake greater than liver activity. However, defining CMR as DS ≤2 was associated with improved TTP, though not OS. In the overall cohort, the 5-year TTP was 72% in patients with DS ≤2 compared to 60% in those with DS 3–5 (p = 0.092). Among patients with baseline SUVmax greater than liver activity, TTP was 79% versus 59% (p = 0.041), and in those with baseline SUVmax more than twice the liver uptake, it was 80% versus 52% (p = 0.035).

The post-treatment reduction in SUVmax (median: -77.4%; IQR: -43% to -100%) did not predict outcomes when using conventional thresholds from the literature (-66% and -70%). However, it was predictive of progressive disease in the entire cohort, regardless of baseline SUV, when a ROC-defined cutoff of -85.5% was applied, with a 5-year TTP of 82% in patients with reductions beyond this threshold compared to 57% in those below it (p = 0.027).

Conclusion: These findings suggest that 18F-FDG PET/CT may be useful for response assessment, particularly in patients with baseline SUVmax greater than liver activity, supporting its rational use in MZL.

This content is only available as a PDF.
2025
Sign in via your Institution