Akut Lösemi ve Agresif B Lenfoma (Dbbhl, Burkitt Lösemi Lenfoma) SSS Tutulumu: Sıklık, Risk Faktörleri ve Prognozu

Abstract

The management of central nervous system (CNS) involvement is relatively clear in acute lymphoblastic leukemia (ALL) and Burkitt's leukemia/lymphoma (BL). However, the incidence, risk factors, prevention, and treatment of CNS involvement in diffuse large B-cell lymphoma (DLBCL) and acute myeloid leukemia (AML) are controversial. Therefore, in this study, we aimed to evaluate the incidence, clinical features, risk factors and prognosis of CNS involvement in patients with ALL, AML, BL and DLBCL. Patients and Methods: The medical records of 477 DLBCL, 134 ALL, 282 AML, 54 BL and 16 biphenotypic leukemia cases were retrospectively analyzed. Categorical variables were evaluated by chi-square test; continuous variables were evaluated by Mann-Whitney U test and t-test for univariate analyses. Survival analyses were done by the Kaplan-Meier method. Using logistic regression analysis, a scoring system that enables the prediction of CNS involvement in DLBCL patients has been developed. Results: CNS involvement risk was significantly different between groups at the time of diagnosis (p = 0.035) and follow-up (p <0.001). At the time of diagnosis, β2 microglobulin level and risky extralymphatic involvement (RELI); during follow-up, higher LDH level, higher ECOG performance score, diffuse bcl-2 staining and RELI were identified as independent risk factors for CNS involvement in DLBCL cases. The risk scoring system included these independent variables. Using this scoring system, it was possible to identify DLBCL patients who developed CNS involvement during follow-up with 100% sensitivity and 38.7% positive predictivity. Independent risk factors for CNS involvement could not be identified in ALL, AML and BL cases. Median survival of the patients with CNS involvement at the time of diagnosis or relapse was shorter than the patients without this complication. Conclusion: CNS involvement is associated with a particularly poor prognosis in DLBCL. Definition of the risk factors for CNS involvement during follow-up is important in this respect. We developed a highly predictive risk scoring system. If confirmed in an independent cohort of patients with an appropriate study, our risk model can be applied in routine practice.