Concordance of CNS Leukemia Detection Between Morphology and Flow Cytometry Prior to Stem Cell Transplantation In Patients with MDS, AML, ALL and CMML

Abstract 4842

Accurate diagnosis of CNS involvement is essential for optimizing treatment and maximizing outcomes in patients with leukemia prior to hematopoietic stem cell transplantation (HCT). The most common methods of detecting CNS involvement include cytomorphic analysis and multiparametric flow cytometry (MFC). While cytomorphic analysis still has diagnostic and prognostic importance, MFC may be more than twice as sensitive as standard cytology analyses, raising the question of whether MFC might be able to detect CNS disease in cases where cytology did not.¹ We retrospectively evaluated cerebral spinal fluid (CSF) results among 109 continuous patients with MDS, AML, ALL and CMML who underwent a reduced-intensity HCT at the Fred Hutchinson Cancer Research Center between May 2006 and May 2010. Ninety-seven of these patients had a pre-HCT lumbar puncture on arrival to evaluate for CNS disease. Thirty-five patients had CSF sent for morphology alone, 2 patients had CSF sent for MFC alone and 60 patients had CSF sent for both morphology and MFC. Of the 60 patients with both MFC and morphology results, CSF notable for blasts by concordant morphology and MFC was detected in 6 patients. These patients received treatment with intrathecal chemotherapy with or without craniospinal radiation. One patient had a single atypical cell with blast-like morphology seen by cytology with negative MFC. Fifty-three patients had both negative CSF by morphology and MFC. This data suggests that there is high concordance between morphology and MFC, however, it is difficult to determine the concordance with much confidence due to the small number of patients with CNS disease. Given the infrequency of CNS disease in this patient population, more patients will need to be studied in order to fully assess the value of MFC in addition to cytomorphic analysis. Because it is also difficult to be sure that our patient population was representative, a prospective large study, for example, in consecutive patients presenting for treatment of AML, might be worthwhile.

¹J.E.C. Bromberg, MD, D. A. Breems, MD, PhD, J. Kraan, BSc, G. Bikker, B. van der Holt, MSc, P. Sillevis Smitt, MD, PhD, M. J. van den Bent, MD, M. van't Veer, MD, PhD and J. W. Gratama, MD, PhD. CSF flow cytometry greatly improves diagnostic accuracy in CNS hematologic malignancies. Neurology 2007;68:1674-1679.