RESEARCH ARTICLE
A JH
Diagnostic utility of cerebrospinal fluid flow cytometry in patients with and without prior hematologic malignancy Alexandra E. Kovach, Michelle E. DeLelys, Abigail S. Kelliher, Laura J. Dillon, Robert P. Hasserjian, Judith A. Ferry, Frederic I. Preffer, and Aliyah R. Sohani* Flow cytometry (FCM) is an adjunct study to routine analysis of cerebrospinal fluid (CSF) to investigate for involvement by a hematologic malignancy. However, in our experience, FCM only infrequently detects abnormalities in CSF. To help optimize resources without forfeiting clinically important data, we sought to determine evidence-based indications and criteria for performing FCM on CSF. FCM results of 316 consecutive CSF specimens were retrospectively reviewed and correlated with clinical history, total nucleated cell (TNC) counts, and results of concurrent cytologic review. Of 255 samples adequate for analysis, 54% were from patients with a prior history of hematologic malignancy, of which 12% (17 cases) were abnormal by FCM. Corresponding TNC counts among samples with abnormal FCM ranged from 0–1050 cells/mL, and only 44% showed abnormal morphology on concurrent cytology. Of the remaining 46% of samples from patients with no known history of hematologic malignancy who had CSF sampling for neurological indications, only one (1%) was abnormal by FCM. This specimen had an elevated TNC count (39 cells/mL) but lacked clearly abnormal findings on concurrent cytology. These results support the use of CSF FCM only in patients with a history of hematologic malignancy or, in the absence of such a history, in samples showing pleocytosis. If these criteria were applied to the current cohort using a TNC count cut-off of >5 cells/mL, 23% of samples would have been deferred from testing, resulting in decreased cost, improved efficiency, and reduction in the need for unnecessary testing without a negative impact on clinical care. C 2014 Wiley Periodicals, Inc. Am. J. Hematol. 89:978–984, 2014. V
䊏 Introduction Flow cytometry (FCM) is a commonly ordered adjunct study to routine cell count and cytologic analysis of cerebrospinal fluid (CSF) at our institution, a large tertiary-care center with significant numbers of patients with hematologic malignancies and/or complex neurologic conditions. The National Comprehensive Cancer Network (NCCN) guidelines specify routine sampling of CSF in the staging and surveillance of patients with a variety of hematologic malignancies that have a high risk for CNS involvement. These include B and T acute lymphoblastic leukemia/lymphoma (ALL), acute myeloid leukemia (AML), primary diffuse large B-cell lymphoma (DLBCL) of the central nervous system (CNS), Burkitt lymphoma/ leukemia (BL), and certain cases of systemic DLBCL not otherwise specified meeting specific clinical or laboratory parameters for high risk of CNS dissemination as defined by the NCCN [1–5]. The NCCN guidelines recommend that FCM be undertaken in addition to morphologic review by cytology, as several studies have established that combining the two modalities improves sensitivity over conventional cytology alone in the detection of hematologic malignancies involving the CSF [5–20]. Guidelines for morphologic review and immunophenotyping of CSF in patients with neurologic symptoms without a prior history of hematologic malignancy are less clear. Pleocytosis (increased CSF white blood cell, WBC, or total nucleated cell, TNC, count) in inflammatory conditions is typically due to an increase in mature reactive T cells, and their immunophenotypic characterization, beyond the routine cell count and differential, is unlikely to change clinical management for primary inflammatory neurological conditions [21]. In addition, normal CSF is paucicellular and not uncommonly yields insufficient cells or events for FCM analysis [22,23]. While methods exist to improve FCM cell yield from paucicellular CSF specimens, such efforts are potentially costly for routine implementation in most clinical laboratories [24–26]. By forgoing FCM analysis of specimens from patients with neurologic symptoms, one might fail to make a primary diagnosis of a hematologic malignancy involving the CNS. However, the CSF as the first site of presentation of a high-grade hematologic malignancy is exceptionally rare, and alternative diagnostic modalities, including imaging studies, are likely better suited for initial work-up of patients with complex neurological symptoms. To optimize the use of supplies and reagents as well as medical technologist and physician labor and time without forfeiting clinically important data, we reviewed our approach to CSF FCM testing over a year-long period in order to determine laboratory-based criteria for clinically appropriate use of FCM on CSF samples at our institution and to identify specimens on which FCM has limited diagnostic utility.
Additional Supporting Information may be found in the online version of this article. James Homer Wright Pathology Laboratories, Massachusetts General Hospital and Department of Pathology, Harvard Medical School, Boston, Massachusetts.
Conflict of interest: The authors have no financial or other disclosures. *Correspondence to: Aliyah R. Sohani, Department of Pathology, WRN 219, Massachusetts General Hospital, Boston, MA 02114. E-mail: [email protected] Received for publication: 19 May 2014; Revised: 6 July 2014; Accepted: 11 July 2014 Am. J. Hematol. 89:978–984, 2014. Published online: 15 July 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/ajh.23806 C 2014 Wiley Periodicals, Inc. V
978
American Journal of Hematology, Vol. 89, No. 10, October 2014
doi:10.1002/ajh.23806
RESEARCH ARTICLE
Diagnostic Utility of Cerebrospinal Fluid Flow Cytometry
TABLE I. Summary of CSF Test Results by Clinical History of Prior Hematologic Malignancy
Parameter Number of samples (%) Number of patients (%) FCM abnormal (%) FCM negative (%) Mean 6 SD TNC count (cells/mL) TNC count >5 cells/mL (%) Abnormal cytology (%)
Prior hematologic malignancy
Neurological abnormalities only
137 (54) 87 (45) 17 (12) 120 (88) 16 6 96
118 (46) 108 (55) 1 (1)a 117 (99) 46 6 105
NA NA 0.0003
30/135 (22) 8/122 (7)
59/118 (50) 1/111 (1)b
5 cells/mL (%) Abnormal cytology (%)
17 (12) 91 6 263 8/17 (47) 6/16 (38)
120 (88) 5 6 11 22/118 (19) 2/106 (2)
NA 0.0004 0.024
A JH
Diagnostic utility of cerebrospinal fluid flow cytometry in patients with and without prior hematologic malignancy Alexandra E. Kovach, Michelle E. DeLelys, Abigail S. Kelliher, Laura J. Dillon, Robert P. Hasserjian, Judith A. Ferry, Frederic I. Preffer, and Aliyah R. Sohani* Flow cytometry (FCM) is an adjunct study to routine analysis of cerebrospinal fluid (CSF) to investigate for involvement by a hematologic malignancy. However, in our experience, FCM only infrequently detects abnormalities in CSF. To help optimize resources without forfeiting clinically important data, we sought to determine evidence-based indications and criteria for performing FCM on CSF. FCM results of 316 consecutive CSF specimens were retrospectively reviewed and correlated with clinical history, total nucleated cell (TNC) counts, and results of concurrent cytologic review. Of 255 samples adequate for analysis, 54% were from patients with a prior history of hematologic malignancy, of which 12% (17 cases) were abnormal by FCM. Corresponding TNC counts among samples with abnormal FCM ranged from 0–1050 cells/mL, and only 44% showed abnormal morphology on concurrent cytology. Of the remaining 46% of samples from patients with no known history of hematologic malignancy who had CSF sampling for neurological indications, only one (1%) was abnormal by FCM. This specimen had an elevated TNC count (39 cells/mL) but lacked clearly abnormal findings on concurrent cytology. These results support the use of CSF FCM only in patients with a history of hematologic malignancy or, in the absence of such a history, in samples showing pleocytosis. If these criteria were applied to the current cohort using a TNC count cut-off of >5 cells/mL, 23% of samples would have been deferred from testing, resulting in decreased cost, improved efficiency, and reduction in the need for unnecessary testing without a negative impact on clinical care. C 2014 Wiley Periodicals, Inc. Am. J. Hematol. 89:978–984, 2014. V
䊏 Introduction Flow cytometry (FCM) is a commonly ordered adjunct study to routine cell count and cytologic analysis of cerebrospinal fluid (CSF) at our institution, a large tertiary-care center with significant numbers of patients with hematologic malignancies and/or complex neurologic conditions. The National Comprehensive Cancer Network (NCCN) guidelines specify routine sampling of CSF in the staging and surveillance of patients with a variety of hematologic malignancies that have a high risk for CNS involvement. These include B and T acute lymphoblastic leukemia/lymphoma (ALL), acute myeloid leukemia (AML), primary diffuse large B-cell lymphoma (DLBCL) of the central nervous system (CNS), Burkitt lymphoma/ leukemia (BL), and certain cases of systemic DLBCL not otherwise specified meeting specific clinical or laboratory parameters for high risk of CNS dissemination as defined by the NCCN [1–5]. The NCCN guidelines recommend that FCM be undertaken in addition to morphologic review by cytology, as several studies have established that combining the two modalities improves sensitivity over conventional cytology alone in the detection of hematologic malignancies involving the CSF [5–20]. Guidelines for morphologic review and immunophenotyping of CSF in patients with neurologic symptoms without a prior history of hematologic malignancy are less clear. Pleocytosis (increased CSF white blood cell, WBC, or total nucleated cell, TNC, count) in inflammatory conditions is typically due to an increase in mature reactive T cells, and their immunophenotypic characterization, beyond the routine cell count and differential, is unlikely to change clinical management for primary inflammatory neurological conditions [21]. In addition, normal CSF is paucicellular and not uncommonly yields insufficient cells or events for FCM analysis [22,23]. While methods exist to improve FCM cell yield from paucicellular CSF specimens, such efforts are potentially costly for routine implementation in most clinical laboratories [24–26]. By forgoing FCM analysis of specimens from patients with neurologic symptoms, one might fail to make a primary diagnosis of a hematologic malignancy involving the CNS. However, the CSF as the first site of presentation of a high-grade hematologic malignancy is exceptionally rare, and alternative diagnostic modalities, including imaging studies, are likely better suited for initial work-up of patients with complex neurological symptoms. To optimize the use of supplies and reagents as well as medical technologist and physician labor and time without forfeiting clinically important data, we reviewed our approach to CSF FCM testing over a year-long period in order to determine laboratory-based criteria for clinically appropriate use of FCM on CSF samples at our institution and to identify specimens on which FCM has limited diagnostic utility.
Additional Supporting Information may be found in the online version of this article. James Homer Wright Pathology Laboratories, Massachusetts General Hospital and Department of Pathology, Harvard Medical School, Boston, Massachusetts.
Conflict of interest: The authors have no financial or other disclosures. *Correspondence to: Aliyah R. Sohani, Department of Pathology, WRN 219, Massachusetts General Hospital, Boston, MA 02114. E-mail: [email protected] Received for publication: 19 May 2014; Revised: 6 July 2014; Accepted: 11 July 2014 Am. J. Hematol. 89:978–984, 2014. Published online: 15 July 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/ajh.23806 C 2014 Wiley Periodicals, Inc. V
978
American Journal of Hematology, Vol. 89, No. 10, October 2014
doi:10.1002/ajh.23806
RESEARCH ARTICLE
Diagnostic Utility of Cerebrospinal Fluid Flow Cytometry
TABLE I. Summary of CSF Test Results by Clinical History of Prior Hematologic Malignancy
Parameter Number of samples (%) Number of patients (%) FCM abnormal (%) FCM negative (%) Mean 6 SD TNC count (cells/mL) TNC count >5 cells/mL (%) Abnormal cytology (%)
Prior hematologic malignancy
Neurological abnormalities only
137 (54) 87 (45) 17 (12) 120 (88) 16 6 96
118 (46) 108 (55) 1 (1)a 117 (99) 46 6 105
NA NA 0.0003
30/135 (22) 8/122 (7)
59/118 (50) 1/111 (1)b
5 cells/mL (%) Abnormal cytology (%)
17 (12) 91 6 263 8/17 (47) 6/16 (38)
120 (88) 5 6 11 22/118 (19) 2/106 (2)
NA 0.0004 0.024
