Department of Radiology, Affiliated Hospital of Hebei University, Baoding, Hebei, China , hbmufengna@163.com
Abstract: (8 Views)
Background:To investigate the dynamic changes in 18F-FDG PET/CT (18F)-derived metabolic parameters throughout the entire chemotherapy process for diffuse large B-cell lymphoma (DLBCL), and to analyze their clinical value in early diagnosis, curative effect evaluation, and prognosis stratification. Materials and Methods: One hundred and sixty-five patients with suspected DLBCL admitted from April 2022 to May 2023 were included. 18F scans were performed before chemotherapy (baseline), after 3-4 courses of chemotherapy (interim), and 6-8 weeks following chemotherapy completion (post-therapy). SUVmax, SUVmean, and TLG were calculated by delineating the metabolic tumor volume (MTV). Therapeutic efficacy was evaluated using the Deauville 5-point scale. Patients were followed up for 2 years to record their survival outcomes. A combined diagnostic model was constructed using Logistic regression, with its predictive efficiency analyzed using ROC curves. Results: Baseline SUVmax, SUVmean, MTV, and TLG were higher in DLBCL patients than in non-DLBCL cases (P<0.05). The combined model demonstrated an AUC of 0.972, with 93.15% sensitivity and 89.13% specificity. The interim metabolic parameters were highly correlated with Deauville scores. Higher SUVmax, MTV, and TLG levels were observed in the poor-efficacy group than in the good-efficacy group (P<0.05). Besides, the death group exhibited SUVmax, MTV, and TLG elevations compared to the survival group (P<0.05). The area under the ROC curve for the combined detection in predicting the prognostic mortality of DLBCL patients was 0.942 (P<0.001). Conclusion: The temporal variations of metabolic parameters derived from 18F prove valuable for the early detection, therapeutic effect assessment, and prognostic prediction of DLBCL.