Prediction of Central Nervous System Relapse of Diffuse Large B-Cell Lymphoma Using Pretherapeutic [18F]2-Fluoro-2-Deoxyglucose (FDG) Positron Emission Tomography/Computed Tomography

Abstract

Central nervous system (CNS) relapse of diffuse large B-cell lymphoma (DLBCL) is a rare complication, but has a poor prognosis with unknown pathophysiology. Recent trials of CNS prophylaxis have shown to be ineffective, despite patient's selection using several known clinical risk factors. In this study, the authors evaluated the value of pretreatment [18F]2-Fluoro-2-deoxyglucose positron emission tomography in predicting CNS relapse in DLBCL patients. The authors analyzed 180 pathologically confirmed DLBCL patients, retrospectively. Patients underwent [18F]2-Fluoro-2-deoxyglucose positron emission tomography/computed tomography before first line rituximab to cyclophosphamide, doxorubicin, vincristine, and prednisone therapy. Clinical characteristics were evaluated and total lesion glycolysis (TLG) with a threshold margin of 50% was calculated. Among age, sex, Ann Arbor stage, International Prognostic Index, revised International Prognostic Index, high serum lactate dehydrogenase level, presence of B symptoms, bulky disease (≥10cm), extranodal lesion involvement, bone marrow involvement, high metabolic tumor volume (>450mL), and high TLG50 (>2000), the high TLG50 was the only significant prognostic factor for predicting CNS relapse in a multivariate analysis (P=0.04). Kaplan-Meir survival analysis between high TLG50 (>2000) and low TLG50 (≤2000) groups revealed significantly different mean progression free survival (PFS) of 1317.2±134.3 days and 1968.6±18.3 days, respectively (P<0.001). High TLG50 on [18F]2-Fluoro-2-deoxyglucose positron emission tomography/computed tomography is the most significant predictor of CNS relapse in un-treated DLBCL patients.