INTRODUCTION: The detection of central nervous system (CNS) disease in hematologic malignancies is important to guide optimal therapeutic approach, refine prognosis and understand patient`s unexplained neurologic symptoms. Newer flow cytometry (FC) techniques are emerging, also there are increasingly reports of higher accuracy than routine cytospin. Moreover, these are becoming incorporated more frequently in clinical work up practices. However, there is still uncertainty on clinical approach of CNS-positive patients, detected only by FC.
OBJECTIVE: To analyze accuracy and clinical outcomes of CNS disease by cytopsin or FC in patients with hematologic malignancies.
MATERIAL AND METHODS: FC cerebrospinal samples and medical charts of 84 consecutive patients evaluated for CNS infiltration by hematologic malignancies from January/2014 to December/2016 were reviewed. Statistical analysis were done with SPSS and STATA softwares.
RESULTS: Baseline patients characteristics were: male (62%), median age 53 years; non-hodgkin lymphoma (52%), Acute Lymphoblastic Leukemia (26%), Acute Myeloblastic Leukemia (15,5%), Multiple Myeloma (6,5%); CNS-positivity rates according to each technique were: Cytopsin-/FC- (71,4%), Cytospin+ (14,3%), Cytospin-/FC+ (14,3%); CNS-disease was detected by FC in 32,3%, while for cytospin was 16,7%. Overall survival was 71,4% and relapse rate 38,1% at 2,5 years of median follow-up. Relapsed (HR: 2,76 p0,023) and CNS-positivity (HR: 2,01 p0,037) patients were significantly associated with an inferior overall survival. Also, progression free survival (PFS) of Cytospin-/FC+ was significantly inferior than CNS-negative subgroup (HR: 2,93 p0,022).
CONCLUSION: FC sensitivity appears to be higher than classicaly cytospin methods to detect CNS disease, also CNS-positivity was associated with a worse prognosis, as well as in the subset of patients Cytospin-/FC+. Further studies with a more homogeneous cohort and larger sample sizes are needed to validate our findings.
No relevant conflicts of interest to declare.
