EP03.01 [18F]FDG PET/MRI FOR ASSESSING TREATMENT RESPONSE OF CNS LYMPHOMA: A PROSPECTIVE CASE SERIES

Abstract BACKGROUND Central nervous system (CNS) lymphoma presents unique diagnostic challenges, as conventional cranial MRI has limitations in assessing treatment response. PET/MRI integration provides simultaneous metabolic and anatomical imaging, potentially overcoming these limitations where post-treatment changes mimic disease progression. This hybrid modality may help differentiate viable tumor from post-treatment changes using 18F-fluorodeoxyglucose [18F]FDG PET metabolic data. It could reduce radiation exposure compared to PET/CT, enable earlier assessment of treatment response, and improve clinical decision-making. The objective is to compare complete response rates according to standard definitions for each modality. MATERIAL AND METHODS In December 2023, we initiated a prospective unicentric study at our institution. Patients with newly diagnosed CNS lymphoma (primary or secondary) who were candidates for methotrexate-cytarabine-thiotepa-rituximab (MATRix) induction were included after informed consent. The study was approved by the institutional ethics committee. Patients are evaluated using cranial +/- spinal [18F]FDG PET/MRI at predefined time points: pre-treatment, after two cycles, at the end of induction, after consolidation with autologous stem cell transplantation (ASCT) or whole brain radiotherapy (WBRT), and every 4 months thereafter. Response assessment followed International Primary CNS Lymphoma Collaborative Group (IPCG) consensus criteria. RESULTS Analysis includes the first five patients. Median age was 53 years (range 45-63); 4 males, with CNS diffuse large B cell lymphoma (4 primary; 1 secondary). Pre-treatment PET/MRI demonstrated a median of 2 lesions (range 1-3), with intense FDG uptake, showing median SUV max of 25 (range 18.9-33) in the largest lesions. Lesions included cerebral hemispheres (n=4), deep brain structures (n=3), and infratentorial region (n=2). All but one patient completed 4 cycles of MATRIX induction. Four patients received ASCT consolidation. One patient developed aspergillus pneumonia and received WBRT as consolidation. After induction (n=5, 100%) and consolidation (n=3, 60%), [18F]FDG PET/MRI demonstrated complete metabolic response in all patients, despite persistent MRI enhancement consistent with partial response in two patients. CONCLUSION This study suggests [18F]FDG-PET/MRI may offer more accurate and timely evaluation of treatment response in CNS lymphoma. Longer follow-up and larger studies are necessary to explore the prognostic value of absent [18F]FDG uptake in residual lesions with mild MRI enhancement. Further research is needed to determine whether PET/MRI enhances confidence in confirming complete response by complementing conventional MRI in post-treatment assessment, potentially leading to more personalized treatment strategies and improved outcomes.