Preliminary clinical assessment of dynamic 18F-fluorodeoxyglucose positron-emission tomography/computed tomography for evaluating the clinicopathological grade in patients with non-Hodgkin's lymphoma: a prospective study.

Objective: We prospectively assessed the diagnostic capacity of dynamic and dual-time-point F-fluorodeoxyglucose PET/computed tomography and explored the most appropriate scan timing for clinicopathological discrimination of non-Hodgkin's lymphoma.

Methods: Thirteen patients underwent dynamic scans dynamic scans (5-15, 15-25, and 25-35 minute postinjection) and consecutive dual-time-point scans (1- and 2-hour postinjection). For five indolent and 16 aggressive lymphomas, we statistically compared the maximum standardized uptake value (SUVmax) and the retention index of the SUVmax (RI-SUVmax) for each scan and explored the diagnostic capacities using receiver operating characteristic (ROC) curve analyses.

Results: SUVmax increased progressively over time in all lymphomas and was significantly higher for aggressive lymphoma than for indolent lymphoma in each timephase. RI-SUVmax of dynamic scans (RI-SUV-dynamic) was significantly higher in aggressive than in indolent lymphoma. The cutoff values obtained a sensitivity of 94%, positive-predictive value of 94%, and accuracy of 91% for SUVmax analyses of the dynamic second and third phases, 1-hour early phase, and 2-hour delayed phase, and a sensitivity of 88%, positive-predictive value of 93%, and accuracy of 86% for RI-SUV-dynamic. In contrast, for the dynamic first phase, the cutoff value of SUVmax yielded moderate sensitivity of 69%, a positive-predictive value of 92%, and an accuracy of 71%. The area under the ROC curve (AUC) of the RI-SUV-dynamic had the highest value (0.938), whereas the AUCs of the other ROC analyses were not significantly different.

Conclusion: The dynamic second and third phase could potentially provide good predictors of clinicopathological discrimination, as can the early and delayed phases.