ALBERT

Beschreibung: Abstract 4792 Background / Aim Although prognostic factors (PF) for conventional lymphoma therapy are well known and used in clinical practice and medical research, the PF for radioimmunotherapy (RIT) have not been fully defined until now. The aim of this prospective multicenter trial is to identify patient- and lesion-specific PF for a standard RIT using 90Yttrium-labeled anti-CD20-antibodies in relapsed or refractory follicular lymphoma (FL) by means of clinical and image data including FDG-PET and CT. To prove the feasibility of the multicenter web-based data collection and to assess the imaging data we performed an analysis on retrospective data. Material and methods This retrospective analysis included clinical and image data of patients with FL from 3 German centers (UKS Homburg, TU Munich, SK Karlsruhe). Clinical data were documented using the ”International Registry on Radioimmunotherapy“ (RIT-Registry). PET and CT data were uploaded in pseudonymous form into an online archive designed for the purpose of this study (Hermes Medical Solutions®). Treatment response was evaluated both on patient and lesion basis. To do so, clinical parameters as well as the documented patient and image data were analysed for every measurable lesion in terms of location, standardized uptake value and volume. PF were evaluated using a uni- and multivariate generalized mixed linear model (MGMLM). Results Web-based data capture comprised a total of 32 patients (aged 44 up to 86 years) documented retrospectively in the RIT-Registry. From this group, 16 patients with at least 1 PET examination before and after RIT were further analysed. Altogether, 159 lesions were measured corresponding to 1 up to 25 lesions per patient. In regard to the patients response, 5/16 patients achieved a CR, 8/16 patients a PR and 3/16 patients remained with NC. 6 patients showed divergent findings (21/159 lesions). In the MGMLM evaluation, the number of lesions per patient (p=0.009), the maximum lesion volume (p=0.004), total volume of lesions per patient (p=0.001) and the FLIPI (p=0.01) were identified as prognostic factors for patient's response (CR, PR). Concerning the lesional response (CR, PR), initial small lesion volume (in PET and CT) and high metabolic activity (PET) were identified as prognostically relevant variables (both p=0.04). Only the maximum SUV (p