High grade non-Hodgkin's lymphoma and acute myeloid leukemia (AML) may involve the central nervous system (CNS) at initial diagnosis, or relapse (1). Indeed, the CNS is a well known sanctuary site for leukemic cells due to the limited power of penetration displayed by cytostatic drugs across the blood-brain barrier into the cerebrospinal fluid (CSF) and brain parenchyma (2). Patients at high risk of CNS involvement, or CNS relapse, necessitate treatment based on intrathecal injection of cytarabine, administered in a liposomal formulation which has shown similar activity to that of free cytarabine, but with a lower toxicity profile and a more convenient dosage regimen (3). Cytology has been routinely used to monitor the CSF in patients with hematological malignancies to evidence any neoplastic involvement. However, nowadays, flow cytometry (FC) is considered the best tool for the detection of leukemic, or lymphomatous cells in CSF, as it has a sensitivity which is several-fold higher than that of cytology (4). On this basis, the National Comprehensive Cancer Network in the USA has recommended the routine use of FC along with cytomorphologic analysis for the diagnosis of CNS lymphoma (5). Our laboratory routinely receives CSF samples of patients with hematological malignancies for cytological and flow cytometric detection of malignant cells.
We report the unusual flow cytometric detection of liposomes in the cerebrospinal fluid of patients with aggressive non-Hodgkin's lymphoma and acute myeloid leukemia, treated with intrathecal liposomal cytarabine. © 2012 International Clinical Cytometry Society
CASE REPORT AND METHODS
Among a series of 114 CSF samples of patients with hematological malignancies, sent to our laboratory in 2011 and processed as previously reported (6), we detected 10 samples (Table 1), taken from six patients, where atypical event distribution was detectable in the forward (FSC) versus side light scatter (SSC) dot plot (Figs. 1A–1D). This heterogeneous event distribution did not belong to either leukocyte, or erythrocyte populations, as it was CD45 and CD235a negative. No atypical corpuscles were evidenced by the morphologic evaluation carried out with May-Grünwald/Giemsa stained cytospins. The white blood cell (WBC) count was in all, but one case, lower than 5 cell/μl when manually performed by a counting chamber, but significantly higher in some cases, when performed with an automated counter. In most cases, the total protein and glucose values were within normal ranges (Table 1). Wet mount preparation of CSF samples showed the presence of a heterogeneous population of spherules of variable diameter (most of them larger than typical WBCs), each having a prominent inclusion (Fig. 1E). Clinicians referred that all these patients had been given CNS treatment with liposomal cytarabine (DepoCyte®, Mundipharma Pharmaceuticals SL, Cambridge, UK) at a standard dose of 50 mg, that CSF samples for laboratory investigations had been taken before the administration of DepoCyte and that there was a time lapse between last administration and sampling of at least 15 days. No malignant cells were detected by FC or cytomorphology; our conclusion was that the events appearing in the FSC/SSC dot plot belonged to residual DepoCyte liposomes that were still present in CSF samples.
|Patient||Diagnosis||Previous DepoCyte administration||Date of CSF evaluation||WBC/μla||Protein mg/dL (ref. range 10–50)||Glucose mg/dL (ref. range 50–80)|
|1||DLBCL||April 8, 2011||June 16, 2011||3||84||61|
|2||MCL-bl||May 28, 2011||June 13, 2011||2||110||217|
|3 (r)||DLBCL||April 18, 2011||May 19, 2011||1||24||58|
|May 19, 2011||July 1, 2011||2||43||68|
|July 1, 2011||August 18, 2011||2||49||75|
|4 (r)||AML||September 10, 2011||October 10, 2011||20||34||75|
|October 10, 2011||December 6, 2011||4||35||70|
|5 (r)||DLBCL||November 16, 2011||December 16, 2011||3||32||94|
|6 (r)||BL||September 27, 2011||November 7, 2011||1||57||18|
|November 7,2011||December 19, 2011||4||19||76|
As liposomal cytarabine, encapsulated in spherical particles, is slowly released into the CSF, free cytarabine is detectable for up to 14 days after administration (T1/2 = 5.8 days; 7,8). Cytarabine administered in such a way has a prolonged half-life which is able to enhance the therapeutic efficacy of the treatment of lymphomatous meningitis (3, 9). Indeed, particles attributable to liposomal cytarabine, determining a spurious elevation of CSF WBC counts have been previously reported as occasional laboratory findings in patients treated with intrathecal DepoCyte (10–13). However, the spherules were not detectable in cytospin preparations because of their alcohol destruction during the staining procedure (11, 12). DepoCyte liposomes have a size range of 3 to 30 micron (10) similar to that of WBC (12–20 micron) and physical features that a cytofluorimeter displays like a “hyperbole-branch” dot distribution. This kind of event distribution has never been found in CSF samples drawn for laboratory investigations, in patients who did not receive DepoCyte treatment and in treated patient at time zero (prior to starting intrathecal chemotherapy).
To the best of our knowledge, this is the first report in which flow cytometric detection of liposomal cytarabine has been described. The presence of liposomes was detectable in some, but not all the patients who were given intrathecal liposomal cytarabine therapy. The reason for this can most likely be related, not only to the pharmacokinetics of this drug, but also to unknown personal factors depending on interindividual variability (14). As FC is a well established tool for the detection of CNS disease among hematological malignancies, this phenomenon should be brought to the attention of flow cytometrists. We suggest to signaling the presence of liposomes in the FC report so that particular care can be taken by clinicians in the interpretation of CSF results, in patients who have received DepoCyte for prophylaxis or treatment of neoplastic meningitis, mainly in cases with increased leucocyte count.