Cerebrospinal fluid flow cytometry and risk of central nervous system relapse after hyperCVAD in adults with acute lymphoblastic leukemia.

Abstract

Potential involvement of the central nervous system (CNS) by acute lymphoblastic leukemia is typically evaluated by a conventional cytospin (CC) of cerebrospinal fluid (CSF). Multiparameter flow cytometry (MFC) is generally more sensitive and specific than morphology, but data to guide its use versus CC are limited. This study identified 92 patients who had MFC performed on their initial CSF specimen and received at least 4 cycles of hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone alternating with methotrexate and cytarabine (hyperCVAD) as their initial treatment. Eighteen (20%) were CSF+ by MFC at the baseline, and only 6 of these patients were positive by CC. In contrast, 0 of 51 patients who were negative by MFC and had CC available were positive by CC. Despite the receipt of significantly more intra-CSF chemotherapy (P < .001), the cumulative incidence of CNS relapse by MFC was 22% among CSF+ patients versus 5% among those who were CSF- (P = .044). No such association was observed between CNS relapse and CC results (P = .42). None of the 74 CSF- patients became CSF+ during their initial treatment despite being tested a median of 5 times (range, 2-10). CSF positivity by MFC was the factor most strongly associated with CNS relapse in a series of univariate Cox models (hazard ratio, 3.7; P = .067). The initial CSF status by MFC had no significant impact on overall or event-free survival. MFC of CSF is superior to CC of CSF in identifying adults at high risk for CNS relapse after treatment with hyperCVAD. Surveillance of CSF by MFC has limited utility.