Cytometry Part B: Clinical CytometryVolume 80B, Issue 5 p. 271-281 Review Article Free Access Flow cytometric characterization of cerebrospinal fluid cells† Marieke T. de Graaf, Department of Neurology, Room H-641, Erasmus University Medical Center, \'s-Gravendijkwal 230, 3015 CE Rotterdam, The Netherlands Department of Medical Oncology, Room E2-80A, Erasmus University Medical Center (Daniel den Hoed), Groene Hilledijk 301, 3075 EA Rotterdam, The NetherlandsSearch for more papers by this authorArjen H. C. de Jongste, Department of Neurology, Room H-641, Erasmus University Medical Center, \'s-Gravendijkwal 230, 3015 CE Rotterdam, The Netherlands Department of Medical Oncology, Room E2-80A, Erasmus University Medical Center (Daniel den Hoed), Groene Hilledijk 301, 3075 EA Rotterdam, The NetherlandsSearch for more papers by this authorJaco Kraan, Department of Medical Oncology, Room E2-80A, Erasmus University Medical Center (Daniel den Hoed), Groene Hilledijk 301, 3075 EA Rotterdam, The NetherlandsSearch for more papers by this authorJoke G. Boonstra, Department of Clinical Chemistry, Erasmus University Medical Center, \'s-Gravendijkwal 230, 3015 CE Rotterdam, The NetherlandsSearch for more papers by this authorPeter A. E. Sillevis Smitt, Department of Neurology, Room H-641, Erasmus University Medical Center, \'s-Gravendijkwal 230, 3015 CE Rotterdam, The NetherlandsSearch for more papers by this authorJan W. Gratama, Corresponding Author j.w.gratama@erasmusmc.nl Department of Medical Oncology, Room E2-80A, Erasmus University Medical Center (Daniel den Hoed), Groene Hilledijk 301, 3075 EA Rotterdam, The NetherlandsDepartment of Medical Oncology, Room E2-80A, Erasmus University Medical Center (Daniel den Hoed), Groene Hilledijk 301, 3075 EA Rotterdam, The NetherlandsSearch for more papers by this author Marieke T. de Graaf, Department of Neurology, Room H-641, Erasmus University Medical Center, \'s-Gravendijkwal 230, 3015 CE Rotterdam, The Netherlands Department of Medical Oncology, Room E2-80A, Erasmus University Medical Center (Daniel den Hoed), Groene Hilledijk 301, 3075 EA Rotterdam, The NetherlandsSearch for more papers by this authorArjen H. C. de Jongste, Department of Neurology, Room H-641, Erasmus University Medical Center, \'s-Gravendijkwal 230, 3015 CE Rotterdam, The Netherlands Department of Medical Oncology, Room E2-80A, Erasmus University Medical Center (Daniel den Hoed), Groene Hilledijk 301, 3075 EA Rotterdam, The NetherlandsSearch for more papers by this authorJaco Kraan, Department of Medical Oncology, Room E2-80A, Erasmus University Medical Center (Daniel den Hoed), Groene Hilledijk 301, 3075 EA Rotterdam, The NetherlandsSearch for more papers by this authorJoke G. Boonstra, Department of Clinical Chemistry, Erasmus University Medical Center, \'s-Gravendijkwal 230, 3015 CE Rotterdam, The NetherlandsSearch for more papers by this authorPeter A. E. Sillevis Smitt, Department of Neurology, Room H-641, Erasmus University Medical Center, \'s-Gravendijkwal 230, 3015 CE Rotterdam, The NetherlandsSearch for more papers by this authorJan W. Gratama, Corresponding Author j.w.gratama@erasmusmc.nl Department of Medical Oncology, Room E2-80A, Erasmus University Medical Center (Daniel den Hoed), Groene Hilledijk 301, 3075 EA Rotterdam, The NetherlandsDepartment of Medical Oncology, Room E2-80A, Erasmus University Medical Center (Daniel den Hoed), Groene Hilledijk 301, 3075 EA Rotterdam, The NetherlandsSearch for more papers by this author How to cite this article: de Graaf MT, de Jongste AHC, Kraan J, Boonstra JG, Sillevis Smitt PAE, Gratama JW. Flow cytometric characterization of cerebrospinal fluid cells. Cytometry Part B 2011; 80B: 271–281. Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URLShare a linkShare onEmailFacebookTwitterLinked InRedditWechat Abstract Flow cytometry facilitates the detection of a large spectrum of cellular characteristics on a per cell basis, determination of absolute cell numbers and detection of rare events with high sensitivity and specificity. White blood cell (WBC) counts in cerebrospinal fluid (CSF) are important for the diagnosis of many neurological disorders. WBC counting and differential can be performed by microscopy, hematology analyzers, or flow cytometry. Flow cytometry of CSF is increasingly being considered as the method of choice in patients suspected of leptomeningeal localization of hematological malignancies. Additionally, in several neuroinflammatory diseases such as multiple sclerosis and paraneoplastic neurological syndromes, flow cytometry is commonly performed to obtain insight into the immunopathogenesis of these diseases. Technically, the low cellularity of CSF samples, combined with the rapidly declining WBC viability, makes CSF flow cytometry challenging. Comparison of flow cytometry with microscopic and molecular techniques shows that each technique has its own advantages and is ideally combined. We expect that increasing the number of flow cytometric parameters that can be simultaneously studied within one sample, will further refine the information on CSF cell subsets in low-cellular CSF samples and enable to define cell populations more accurately. © 2011 International Clinical Cytometry Society White blood cell (WBC) counts and their differential into mononuclear (MNC) and polymorphonuclear cells (PMN) in cerebrospinal fluid (CSF) are critical in the diagnosis of many infectious and inflammatory neurological disorders (1). In acute bacterial meningitis, WBC counts usually range between several hundred to more than 60,000/μl, predominantly PMN (90–95% of WBC count), although early in the disease WBC counts can be lower than 100 WBC/μl (2). In viral meningitis, cell counts are usually between 10–1,000 WBC/μl, but may exceed 1,000/μl. Here, MNC predominate, but in the very acute stages of disease PMN can account for more than 80% of leukocytes (2). In multiple sclerosis (MS), two-thirds of patients have a normal CSF cell count and a low level of mononuclear pleocytosis is found in one-third of the cases (2). The CSF WBC count is usually normal ( 5 leukocytes/μl) in patients with the Guillain-Barré syndrome, whereas in case of an increased WBC count other diagnoses should be considered (3). In paraneoplastic neurological syndromes (PNS), 47% of patients have a lymphocytic pleocytosis before the third month after onset of the neurological symptoms, while after the third month only 28% of patients have elevated cell counts (4). CSF WBC counts are also routinely determined in patients with suspected leptomeningeal metastases of solid or hematological malignancies and half of these patients have a lymphocytic pleocytosis (5, 6). Generally, cell counts and differential can be obtained by evaluating cell number and morphology in microscopic slides, by automatic counting based on cellular scatter properties or by flow cytometry in which antigen expression of cells is assessed in combination with light scatter properties. Specifically, microscopic counting of WBC and red blood cells (RBC) is performed using Neubauer or Fuchs-Rosenthal counting chambers, which contain a microscopically visible counting grid and are used with a fixed sample volume. Staining with Samson or Türk reagent may be added to the procedure to facilitate WBC counting and perform differential (6). However, in CSF samples with low cellularity ( 5 leukocytes/μl), differential by microscopy is not performed. When the WBC count is higher than the upper reference value, most laboratories perform cytospin centrifugation of the sample, followed by Wright or May-Grunwald-Giemsa staining to enable morphological differential of cells in CSF. This technique permits rapid differential between monocytes, lymphocytes, and granulocytes, which is of utmost importance for patients with acute neurological disease (7). Counting of erythrocytes is important to exclude traumatic bleeding as the cause of an elevated WBC count (8). Although microscopic counting and differential has been used for a long time in routine CSF analysis, the clinical laboratory faces several challenges in performing it. First, the analysis is time-consuming and should ideally be performed within 1 hour after lumbar puncture as CSF cell counts decrease rapidly after sampling (9). Second, counting and differential results have relatively high intraobserver and interobserver variation (8), as have other manual microscopic techniques. Furthermore, skilled and continuously trained technicians are required for this assay on a 24 hours/7 day basis (10). Nowadays an increasing number of clinical laboratories replace the microscopic technique by hematology analyzers (HA) for cell counting and differential in CSF (8, 10, 11). HA may provide fast, low-cost, and standardized cell counting of CSF and other body fluids such as ascites or pleural fluid. However, special attention is needed regarding background signal, carry-over, and interference in view of the low cell concentrations in these fluids. Two dedicated, FDA-approved, applications for CSF counting and differential on HA are available, that is, the ADVIA® 120/2120 CSF assay (Siemens AG, München, Germany) and the Body Fluid mode on a XE-5000™ analyzer (Sysmex, Kobe, Japan) (8, 11). ADVIA® CSF assay uses light scatter and absorbance measurement for counting and differential, after mixing the sample with CSF reagent to sphere and fix the cells. Not only RBC, WBC, and PMN/MNC are reported, but also lymphocytes, monocytes, and eosinophils. XE-5000™ Body Fluid mode uses sheath flow impedance for counting RBC, while light scatter and fluorescence intensity after DNA/RNA staining is used to analyze WBC. The application reports WBC, PMN/MNC, and high-fluorescent cells. For CSF WBC counting, comparison of the Fuchs-Rosenthal counting chamber with both the XE-5000™ analyzer and the ADVIA® 120/2120 CSF assay showed linearity between 1 and 10,000 cells per μl (8, 11, 12). Automatic MNC counts also correlated well with manual counts, but PMN counts showed poor agreement being almost two-fold higher using the XE-5000 analyzer (12). Especially in low WBC counts ( 20 cells per μl) high imprecision was observed in both techniques compared to manual counting (8, 11). The detection limit of the XE-5000 is 10 cells per μl and when the WBC count is below that limit, differential into MNC and PMN cannot be made (12). In these cases, manual counting or evaluation of stained cytospin slides should be performed. In automatic cell counting of low-cellular samples, the same problem arises as in flow cytometry (see Rapid decline of leukocyte counts upon lumbar puncture section): due to high imprecision, underestimation or overestimation of CSF WBC counts may lead to erroneous results. In clinical practice, HA are more widely available and have lower material costs than flow cytometers. Morphologic examination of CSF cells is performed on cytospin preparations stained with May-Grunwald-Giemsa (6). Whilst highly specific ( 95%), conventional cytomorphological analysis is associated with only a limited sensitivity with up to 20–60% false-negative results (6, 13). Interpretation of cytological findings may be difficult because of paucity of cells in CSF and possible morphological similarities between benign and malignant cells (14). Cytomorphological examination is used in patients with suspected leptomeningeal dissemination of solid tumors or hematological malignancies. Typically, only 50% of patients have malignant cells identified by cytomorphological examination on the first lumbar puncture (15). This yield is increased to 80% with a second CSF examination (5), but even three lumbar punctures will still miss tumor cells in ∼10% of patients (16). Despite its low sensitivity, CSF cytologic examination has been the gold standard for leptomeningeal metastasis because of its 100% specificity (16). If clinical suspicion is high, gadolinium-enhanced MRI of the brain and spine can provide definitive evidence of leptomeningeal metastasis, even without a positive CSF cytology (17, 18). Immunocytochemistry allows the detection of cell surface antigens on CSF cells by cytospins. For detection of leptomeningeal localization of hematological malignancies, a sensitivity of 89–95% and a specificity of 89–100% were shown by this technique (19). For CSF samples with low cell counts, immunocytochemistry should be used subsequent to cytomorphology and the selection of the antibodies should be determined by the cytological findings in combination with the patient\'s history (20, 21). Alternatively, it is stated that this technique should only be used when CSF cytomorphology fails in patients with a strong suspicion of leptomeningeal metastases (22). Compared to flow cytometry, immunocytochemistry gave similar results in detection of high-density surface markers, whereas for analysis of antigens that are expressed at low density immunocytochemistry may be more reliable in some applications (23). Since flow cytometry is used in the detection of central nervous system (CNS) involvement of hematological malignancies besides cytomorphological analysis, as discussed in the Applications of flow cytometry to study CSF section, immunocytochemistry has no major role anymore. In contrast, immunohistochemistry is still used in combination with cytology in the detection of leptomeningeal metastases of solid tumors. PCR requires the selection of primers specific for tumor cell-derived DNA. In hematological B-cell malignancies, analysis of immunoglobulin heavy chain gene rearrangements in the third complementarity determining region (CDR3) by PCR in blood and bone marrow cells is a state-of-the-art technique for diagnosis, monitoring response to treatment and detection of minimal residual disease (24, 25). Presence of clonally rearranged CDR3 is the molecular signature of malignant B-lymphocytes and is present in 80–95% of B-cell lymphomas and leukemias (26). Until now, this technique has not been generally applied to CSF samples. PCR can also be used for detection of leptomeningeal metastasis in solid malignancies. Although it would be ideal to use primers for DNA sequences common to all metastatic cells, the use of sequences for specific primary cancer histopathologies might provide a more practical option, as many are known already (13, 27-29). Additional molecular tumor markers or oncogenes can be used for other types of cancer and might eliminate the need for biopsy in selected patients (29). In patients with suspected leptomeningeal metastases of hematological tumors, flow cytometry of CSF samples is used in addition to cell counting and cytomorphology. In this review, we will focus on the applications and recent developments of CSF flow cytometry. Although this procedure has only a narrow clinical indication, it has significant prognostic and therapeutic implications in individual patients. The advent of polychromatic flow cytometry, that is, advanced instrumentation and reagent development (30), allows detection of a large spectrum of cellular characteristics, even in samples with small amounts of cells like CSF. Apart from differentiating between major leukocyte subsets by assessing granularity and volume of cells, a wide range of cell populations can be specified by immunophenotyping using surface membrane, cytoplasmic, and nuclear antigens (7, 14, 31). However, intracellular staining should be limited to those cases in which it is essential to reach the immunological conclusion, because its use is associated with relatively pronounced cell loss (31). The number of characteristics on one single cell that can be determined in a single tube depends on the number of fluorescent colors available on the flow cytometer used and the number of monoclonal antibodies per tube. The applicability of the assay can even be further enhanced by combining two antigens expressed by nonoverlapping cell subsets on a single fluorochrome (e.g., CD4 present on T lymphocytes and CD19 present on B lymphocytes; 14). In CSF, the simultaneous assessment of 13 parameters (11 colors plus forward and sideward scatter) has been reported in this way (14). However, problems with spectral overlap and color compensation increase when more than six colors are used, but these problems can be reduced if markers and fluorochromes are combined judiciously (32). Frequencies of different WBC populations in CSF are most widely investigated. However, knowledge of absolute numbers of the major cell populations can be of great help to evaluate the sample (33). Due to possible cell loss during concentration and centrifugation steps (31, 34), absolute cell counts may be an underestimation of the real CSF cell number. Addition of counting beads to the monoclonal antibody-stained CSF cell suspension allows accurate enumeration of absolute numbers of cell subsets (31). By using this technique, we showed that PNS patients stood out by highly increased absolute counts of the major lymphocyte subsets in CSF, but above all, by B-lymphocyte counts that had increased more than 20-fold as compared to controls without neurological disease (35). In these patients, the frequency of B-lymphocytes (expressed as fraction of lymphocytes) had increased only three-fold (35). When merely frequencies would have been studied, this enormous B-lymphocyte expansion suggesting an important role for these cells in PNS, would have remained unnoticed. This result indicates that assessment of absolute counts besides frequencies is also important in CSF. To use CSF flow cytometry in research of neuroinflammatory diseases, knowledge of the composition of cells in normal CSF is needed. Because CSF of healthy controls is usually not available due to ethical considerations (36), patients with noninflammatory neurological disorders (NIND controls) are often included instead (37). However, Svenningsson et al. (38, 39) did study normal CSF by assessing the percentages of lymphocyte subsets in CSF of 34 healthy individuals, after informed consent, with 2- or 3-color flow cytometry. In addition, we studied both absolute numbers and percentages of leukocyte, lymphocyte, T lymphocyte, and dendritic cell subsets by six-color flow cytometry in 84 individuals without neurological disease undergoing spinal anesthesia (Table 1; 40). The two published studies (38, 39), as well as our data (40), showed that normal CSF is predominantly composed of CD4+ T lymphocytes, mostly with a central memory phenotype, and in addition contains very low frequencies of B lymphocytes, NK cells, and NKT lymphocytes (Table 1). Dendritic cells, both myeloid as well as plasmacytoid, were also present in normal CSF, although in very low frequencies (Table 1; 40, 41). Absolute numberaa Medians (5th–95th percentiles) of absolute numbers ×106/l are given. Flow cytometry is a sensitive method capable of detecting abnormal monoclonal B lymphocytes, which comprise as little as 0.01% of total lymphocytes (42, 43). The detection rate of CSF involvement in hematological malignancies is up to 86% higher in flow cytometry than in conventional cytomorphological analysis (6, 14, 42-48; Table 2). Although it was previously suggested that this method may not be suitable in the evaluation of samples with low cellularity (23), it has been shown that CSF T lymphocytes, the predominant lymphocyte subset in CSF, can be reliably detected in samples with a cell count lower than 5 leukocytes/μl (46). Kleine et al. (49) showed that precision of lymphocyte flow cytometry is high (coefficient of variance [CV] ≤ 10%) provided that a sufficient number of events has been acquired (50). However, the CV may increase to values up to 30% for the minor subsets in CSF, for example, NK cells and NKT lymphocytes (49). Table 2. Comparison of Flow Cytometry and Conventional Cytomorphology in Detection of CSF Involvement in Hematological Malignancies CSF involvementaa CSF involvement was diagnosed when flow cytometry, conventional cytomorphology or both were positive. FC+/CC+ FC+/CC− FC−/CC+ Detection rate by using CC alonebb Detection rate of CSF involvement by using conventional cytomorphology alone: ([FC+/CC+] + [FC−/CC+])/CSF involvement. Detection rate by using FC alonecc Detection rate of CSF involvement by using flow cytometry alone: ([FC+/CC+] + [FC+/CC−])/CSF involvement. CSF involvement was diagnosed when flow cytometry, conventional cytomorphology or both were positive. Detection rate of CSF involvement by using conventional cytomorphology alone: ([FC+/CC+] + [FC−/CC+])/CSF involvement. Detection rate of CSF involvement by using flow cytometry alone: ([FC+/CC+] + [FC+/CC−])/CSF involvement. Several studies comparing flow cytometry and conventional cytomorphology to detect CSF involvement in hematological malignancies (Table 2) escribe samples in which flow cytometry is positive whilst cytology is negative (6, 14, 43, 51). Presence of neurological symptoms compatible with leptomeningeal disease is highly suggestive for CNS involvement in such patients, whereas absence of symptoms and lack of recurrence of CNS disease during clinical follow-up indicate a false-positive flow cytometric result (6, 14). Results of clinical follow-up were documented in three studies. Sancho et al. (51) found that a flow cytometry-positive, cytology-negative result was associated with a higher probability of CNS relapse in aggressive B-cell lymphomas, as compared to samples with absence of neoplastic cells by both methods, whereas Hegde et al. (43) observed that 5/11 patients with a flow cytometry-positive, cytology-negative result relapsed in the clinical CNS and died despite having received active treatment. In addition, Bromberg et al. (6) described the absence of CNS recurrence in only 1/24 flow cytometry-positive, cytology-negative patients. These follow-up data indicate that flow cytometric analysis of CSF samples has a low risk of being false positive in patients with hematological malignancies. In flow cytometry-negative samples on initial staging, Hegde et al. (43) observed that 3/40 patients at increased risk and 0/41 patients at low risk for CNS involvement relapsed in the CNS. This low risk of a false-negative result in flow cytometry can also be deducted from Table 2, which shows a low number of flow cytometry-negative, cytology-positive samples in all studies. When leptomeningeal localization of a hematologic malignancy is suspected, the presence of a pathological (monoclonal) population and phenotypic characterization of that population can be assessed by using the proper antibody reagent panel adapted to the number of cells and previous histological and immunophenotypical diagnosis or suspected diagnosis together with appropriate gating strategies (31). Pathological cells usually occur at very low percentages in the order of 0.01% in CSF. In addition, CSF samples contain a limited number of cells rendering pathologic cells in CSF very rare. To detect low numbers of rare cells, the background fluorescence of the reagents should be minimal, and a sufficient cell number is required to analyze lymphocyte subsets reproducibly. Therefore, the volume of the CSF sample should ideally be minimally 5 ml and cell loss during processing be prevented as discussed in the Technical pitfalls section. Involvement of the CNS is a relatively uncommon complication of leukemia and lymphoma (14, 46), which is suspected in patients who develop neurological symptoms or signs (14) or in patients at high risk of CNS localization (6). It has grave prognostic significance and requires important therapeutic decisions including the administration of intrathecal chemotherapy (6, 46). Leptomeningeal localization is diagnosed by conventional cytomorphological analysis through identification of malignant lymphocytes in CSF (31, 52). However, this technique has a relatively high rate of false-negative results in up to 60% of cases (22, 53). Recent reports suggest that multiparameter flow cytometric assessment of CSF samples could improve the efficiency of detection of CNS involvement, due to its high specificity and greater sensitivity (6, 52, 54). Table 2 gives an overview of studies, which investigated the value of flow cytometry and conventional cytomorphology in detection of CSF involvement in hematological malignancies. These studies showed that the use of flow cytometry alone increased the detection rate of CSF involvement up to 86% compared to the use of cytomorphology alone. Combined use of flow cytometry and cytomorphology increased the detection rate with 17–86% compared to cytomorphology alone. Therefore, the National Comprehensive Cancer Network (USA) has recommended the routine use of flow cytometry in conjunction with cytomorphological analysis for the diagnosis of CNS lymphoma (55). CNS involvement is diagnosed if one of these diagnostic procedures is positive. For detection of hematological malignancies, flow cytometry depends on the analysis of light chain restriction (Fig. 1) and/or aberrant antigen expression, which should be interpreted within the context of the patient\'s histological diagnosis (56). Still, cytomorphological examination of CSF has additional diagnostic and possibly prognostic value and should still be performed in conjunction with flow cytometry (6). Five-color flow cytometric CSF analysis. Example of a five-color flow cytometric analysis for B-lymphocyte clonality in CSF. Each dot represents a single cell. For analysis, debris and nonleukocyte events were excluded by gating on FSC and CD45 (gate 1, panel A). The leukocyte subsets (My = myeloid lineage; Imm = immature lineage; Mo = monocytes; Ly = lymphocytes) were defined with CD45 expression and side scatter (panel B) and show two major subsets: lymphocytes (CD45+, SSClow, FSCintermediate; green dots) and monocytes (CD45+, SSCintermediate, FSChigh, CD4dim; cyan dots). B: lymphocytes were gated using the lineage-specific marker CD19 and side scatter (SSC) (purple dots; gate 2, panel C). Next, the differential light scatter properties of the B-cell subpopulations as a function of their Ig light chain expression are shown in panels D and E. The combined analysis of clonality and light scatter revealed that the larger population of B lymphocytes have relatively high FSC and SSC signals with monoclonal expression of sIgK but not sIgL light chains (panel E, violet dots), compatible with B cell lymphoma, whilst the few B cells with relatively low FSC and SSC signals express either Ig kappa or lambda (panel E, cyan dots). Note that the cyan dots that express both sIgK and sIgL (panel E) are monocytes, as revealed by their CD14 reactivity (panel F). This result is caused by so-called cytophilic Ig binding (i.e., Ig bound through Fc receptors). Flow cytometry of CSF is also used as a research tool in various neuroinflammatory diseases. The distribution of lymphocyte subpopulations in the CSF may be a consistent indicator of the type of immune response active in these diseases (57). Several studies have reported on flow cytometric analysis of lymphocytes and their subsets in CSF of patients with MS. The CSF cell populations in MS patients have been shown to consist of approximately 60% CD4+ T lymphocytes (58) with a higher frequency of the regulatory phenotype (59, 60) and a higher CD4/CD8 ratio (61), while the frequency of NKT lymphocytes is lower (62) than in NIND controls. Compared to blood, CSF of MS patients showed a relative depletion of CD8+ effector memory T lymphocytes (63). In relation to disease activity, patients with active MS had higher percentages of activated CD4+ T lymphocytes (64-67) and lower percentages of activated CD8+ T lymphocytes (64, 65, 67) in their CSF than inactive MS patients. Moreover, the percentage of naïve CD45RA+, CD50+ (ICAM-3) lymphocytes in CSF is significantly increased (68), while cell surface expression of CD54 (ICAM-1) on T lymphocytes in CSF is significantly decreased (69) in patients with relapses compared to patients in remission. Both are suggested to be used as markers of MS disease activity in CSF as well as blood (68, 69). With regard to B lymphocytes in CSF of MS patients, a significant accumulation of mature B lymphocytes and plasma blasts is observed (70, 71). Most B lymphocytes have a memory phenotype (71-74) and more B lymphocytes express CD80 (costimulatory molecule inducing T lymphocyte activation) than in NIND and healthy controls (75, 76). Furthermore, the number of dendritic cells is elevated in CSF of MS patients (41). CSF flow cytometry was also used in evaluation of MS treatment with immunosuppressive drugs. Both rituximab (anti-CD20 monoclonal antibody [mAb]) (77) and natalizumab (anti-α4 integrin mAb) (78) reduced the number of B and T lymphocytes, while high-dose methylprednisolone induced changes in the expression of CD25, CD26, and HLA-DR on CD4+ T lymphocytes (79). We and others have reported on CSF lymphocyte subsets in patients with PNS. In PNS associated with anti-Hu antibodies, CSF is characterized by (i) a very substantial (20-fold) B-lymphocyte expansion and (ii) a three-fold T-cell expansion (including both CD4+ and CD8+ subsets) compared to controls (35). Children with paraneoplastic opsoclonus-myoclonus syndrome had normal CSF cell counts, but higher percentages of B lymphocytes (80, 81), activated T lymphocytes and γδ-T lymphocytes, lower percentages of CD4+ T lymphocytes and a lower CD4/CD8 ratio (81) compared to NIND controls. The low number of cells in CSF (normal range: 5 leukocytes/μl) hampers the use of flow cytometry (82). To analyze lymphocyte subsets reproducibly, measuring a sufficient cell number is required. However, the minimal number of CSF cells required is not universally defined. In literature, the minimal CSF cell number varies between 100 gated lymphocytes in lymphocyte subset characterization (49) and 1,000 cells in suspected CSF localization of lymphoma (54). A subpopulation was reliably identified whenever 13 or more clustered events displaying identical features were present, whereas the presence of fewer than five clustered events could not be related to the presence of a specific cell population (82). In another study, a minimum of 15 events is reported to ascribe them to a specific cell population with a high probability (83). In leptomeningeal metastasis of hematological malignancies, one study describes the presence of at least 10-clustered events with abnormal patterns of antigen expression for diagnosis (14), while another publication prescribes to classify clusters of more than 25 events as positive, 10–25 events as suspicious and below 10 events as negative (31). In our laboratory, we agree with the latter publication and consider 25-clustered events as positive. To obtain a maximal number of cells for analysis, CSF samples have to be concentrated by low-speed centrifugation. No significant cell loss was observed in hypocellular samples ( 10 leukocytes/μl) when CSF cells were enriched by centrifuging at 200g for 15 minutes at 4°C (49). CSF samples containing 10 leukocytes/μl can be stained and analyzed without cell enrichment (36). However, in case of rare event detection, for example, in CNS involvement of lymphoproliferative disorders, CSF cells should also be concentrated in samples containing 10 leukocytes/μl to increase the sensitivity. Another way to deal with the low cell numbers in CSF, is the use of a two step approach (31, 46). First, one third of the sample is analyzed with a screening tube, which in most cases will answer the clinical question. When this first staining is not conclusive, the process should be repeated with the remaining CSF and the same reagent combination. Combining the list mode data of the first and second staining will increase sensitivity by enabling analysis of a larger number of cells. Second, if a pathological population is identified in the first step, immunophenotyping may be extended. Another difficulty of CSF studies is the rapid decay of leukocytes after sampling as described in several studies (Table 3). Within 30 minutes after sampling, the CSF cell number decreases significantly (9, 14, 34, 84-86). Also, differences in survival rate between different leukocyte subsets were observed: monocytes and granulocytes showed a more rapid cell loss than lymphocytes (9, 34, 85; Table 3). In flow cytometric analysis, selective cell losses may cause underestimation of cell counts (23). These errors can affect clinical decisions. For example, in CSF samples with a pleocytosis, underestimation of the cell number may result in a normal cell count and pathology, such as a neuroinflammatory disease, may be considered ruled out. 0aa Spiking: homologous blood was added to cell-free supernatant of CSF samples. 2 and 24 15 and 39  Ref. 85 0bb Spiking: lymphocytes, monocytes, and neutrophils were obtained from peripheral blood and subsequently spiked into CSF samples that had been centrifuged to remove cells or into saline. 2 and 4 Ambient 27cc WBC loss is calculated by adding up the lymphocyte, monocyte, and granulocyte loss.and 44cc WBC loss is calculated by adding up the lymphocyte, monocyte, and granulocyte loss. 12 and 34 20 and 39 50 and 58  Ref. 34 0dd Spiking: after CSF withdrawal, CSF cells were pelleted by centrifugation, and resuspended in CSF or in PBS containing 5% fetal calf serum. The cell number at resuspension was set at 100%. On ice 62cc WBC loss is calculated by adding up the lymphocyte, monocyte, and granulocyte loss.  Ref. 85 0bb Spiking: lymphocytes, monocytes, and neutrophils were obtained from peripheral blood and subsequently spiked into CSF samples that had been centrifuged to remove cells or into saline. Ambient 6cc WBC loss is calculated by adding up the lymphocyte, monocyte, and granulocyte loss.  Ref. 86 0.5ee Sterile physiologic medium (one part Earle\'s balanced salt solution and one part 20% human serum albumin) was added directly after CSF withdrawal. Ambient  Ref. 34 0dd Spiking: after CSF withdrawal, CSF cells were pelleted by centrifugation, and resuspended in CSF or in PBS containing 5% fetal calf serum. The cell number at resuspension was set at 100%. On ice  Ref.9 0.5ff Serum-containing medium (RPMI-1640 with HEPES, L-glutamine, penicillin/streptomycin, heat-inactivated fetal bovine serum, and heparin) was added directly after CSF withdrawal. Ambient Spiking: lymphocytes, monocytes, and neutrophils were obtained from peripheral blood and subsequently spiked into CSF samples that had been centrifuged to remove cells or into saline. Spiking: after CSF withdrawal, CSF cells were pelleted by centrifugation, and resuspended in CSF or in PBS containing 5% fetal calf serum. The cell number at resuspension was set at 100%. Sterile physiologic medium (one part Earle\'s balanced salt solution and one part 20% human serum albumin) was added directly after CSF withdrawal. Serum-containing medium (RPMI-1640 with HEPES, L-glutamine, penicillin/streptomycin, heat-inactivated fetal bovine serum, and heparin) was added directly after CSF withdrawal. This cell loss in native CSF can be reduced by addition of medium to CSF directly after sampling. In an earlier study, we showed that addition of serum-containing medium (RPMI-1640 with HEPES, L-Glutamine, Penicillin/Streptomycin, heat-inactivated fetal bovine serum, and heparin) preserves CSF cells until at least five hours after sampling (9). Another study showed that immediate addition of Earle\'s balanced salt solution with human serum albumin to CSF prevents total WBC loss until at least 24 hour after sampling (86). In addition, addition of TransFix™ (fixative) has been shown to reduce CSF cell loss (14). Other previous reports that investigated CSF cell preservation methods were more laboratory based than clinical. Spiking of CSF cells into 5% fetal calf serum (34) or saline (85) showed no significant cell loss, while spiking into acellular CSF did (34, 85; Table 3). In addition, immediate cooling of the CSF sample (84), a minimum of centrifugation steps (34), and aspiration of the supernatant instead of decanting the sample (31) all reduce cell loss. Furthermore, due to the absence of free immunoglobulins (Ig) in CSF, washing CSF cells before surface-bound Ig staining is not necessary, which minimizes the wash steps in the protocol (31). The function of media, in most studies serum containing, in preserving leukocytes is probably a buffering one. Both increase in pH in CSF after removal from the body (due to diffusion of CO2 out of the sample) and hypotonicity of CSF (causing movement of water and solutes from the extracellular to the intracellular compartments) have been hypothesized to contribute to cell death (85, 87, 88), although the effect of both factors has not been confirmed. It remains to be investigated, which medium is the most effective in preventing CSF cell loss, being both practical for use in clinical settings and inexpensive. Evidently, CSF samples used for cell counting should be handled carefully by (i) collection in a buffering medium to prevent the rapid cell loss; (ii) a minimum of wash and centrifugation steps; (iii) aspirating instead of decanting; and (iv) data acquisition at least within 24 hour after sampling (86), although the maximal storage time has not been determined yet. Nonspecific or \"background” fluorescence may constitute a serious problem, especially in rare event detection in CSF samples. Its cause can be categorized into three groups (89). First, autofluorescence by excitation of naturally occurring cellular components (e.g., granule-associated flavoproteins in granulocytes) other than the antibody bound fluorochrome (89). This problem may be reduced by the use of a 532 nm laser (90) or specific tools like single laser excitation (91) and cell-by-cell autofluorescence correction (92). Second, spectral overlap that becomes significant when more than four colors are used and can be minimized by choosing combination of fluorochromes that have no or little overlap with each other (32). Third, nonspecific antibody binding may occur, which can be eliminated by optimizing antibody concentration using titration assays (93). Red cells present in CSF reflect either CNS bleeding or a traumatic lumbar puncture in which peripheral blood contaminates the CSF. The possibility of blood contamination of CSF samples can be ruled out by absence of cell populations in CSF that are present in blood at normal or high numbers (e.g., neutrophils and erythrocytes). In flow cytometric absolute cell counting of blood contaminated CSF samples, correction of the number of leukocytes has to be performed. We prefer to use the leukocyte/erythrocyte ratio in peripheral blood for correction (94), because this method accounts for a patient\'s individual situation. Alternatively, the CSF leukocyte number may be arbitrarily adjusted by correction according to the CSF erythrocyte count, which results in a correction of 1 leukocyte per 500 erythrocytes present in CSF (95, 96). This latter method is widely used in clinical practice, because information on peripheral blood is not needed. In blood contaminated CSF samples, which are investigated for the presence of CNS involvement of hematological malignancies, detection of a small population of neoplastic cells is only diagnostic of CSF involvement if these cells are not detected in a simultaneously obtained blood sample (31). In acute leukemia, lumbar puncture should not be performed in an acute phase of the disease when the frequency of circulating malignant cells is high. In case of a traumatic lumbar puncture, malignant cells from the blood may become detectable in CSF leading to apparently false-positive CSF results. Moreover, iatrogenic contamination of CSF with malignant cells might be caused (97). Detection of a monoclonal B-lymphocyte population in CSF diagnoses CNS localization of a B-cell lymphoma in patients with hematological malignancies (6, 43). B-lymphocyte clonality can be investigated by flow cytometry by assessing surface immunoglobulin light chain expression on CD19+ B-lymphocytes and comparison of the \"light chain ratio (LCR)” or \"kappa/lambda ratio,” which is determined by dividing the percentage of cells with the dominant light chain by the percentage of cells with the minor light chain (52). Normal ranges for the LCR differ between laboratories. A LCR threshold of two was reported to have a specificity of 92.3% and a sensitivity of 73.1% (98), while other studies proposed thresholds ranging from 2 to 6, with highest specificities and sensitivities around 70% (99-104). This indicates that if, for example, a threshold of two is used, ∼10% of patients with a LCR above 2 are reported to have a monoclonal B-cell population, but do not have a B-cell lymphoma. Therefore, a LCR shift is only suggestive for presence of a monoclonal B-cell population, and further analysis of the CSF has to be performed. Assessment of additional markers may increase specificity as abnormal light scatter patterns and abnormal intensities of CD19 and/or CD20 indicate the presence of a malignant cell population (14), although absence of such an abnormal pattern does not rule out malignancy. Preferably, assessment of monoclonality and abnormal marker expression are combined, for example, looking for monoclonality in a large forward scatter (FSC) or dim CD20 population, as presented in Figure 1. Furthermore, detection of clonally identical rearranged DNA sequences in malignant B-lymphocytes by PCR is suggestive of the presence of B-cell lymphoma (25), but only a minority of clinical laboratories have this technique operational. Another important point of attention is that not every monoclonal B-cell population in CSF indicates symptomatic CNS disease (42). In patients with indolent hematological disorders, including chronic lymphatic leukemia, malignant B cells in the CSF may represent asymptomatic leptomeningeal involvement and may require treatment only when (new) symptoms arise (52, 105, 106). At last, clinical follow-up will aid in diagnosing CNS disease. We expect that the expansion of the number of colors amenable to flow cytometry will enable the simultaneous study of more parameters within the same sample. Using this approach, more refined information on CSF cell subsets will become available and cell populations can be defined more accurately (107). An ongoing challenge is the search for new fluorochromes that can be used in conjunction with current ones and yet do not contribute to significant spectral overlap (108). Even without new fluorochromes or lasers, instruments will improve through advances in software for data processing. However, visualizing these data becomes more and more complex and requires multiple sequential analyses to provide information about each cell subset. Therefore, automated classification systems are being developed (109-112). Additionally, optimization of storage conditions to preserve CSF cells should result in higher cell yields and thereby increase the detection rate of flow cytometry in CSF samples with low cellularity. Importantly, flow cytometry can be combined with molecular techniques including PCR to improve sensitivity in detecting CSF involvement of lymphoma ( 113, 114). Furthermore, broad-spectrum tumor cell-specific antigens could be fluorescently labeled and used in flow cytometric detection of CNS malignancy (13). In future, DNA clonality of the tumor might be identified on biopsy material and then followed by CSF assays along the course of the disease. This allows us to detect whether selection and development of new malignant clones occur in resistant or relapsing disease and will enable us to appraise the therapeutic and prognostic implications of molecular diagnostic testing of CSF (26). However, these future techniques have to be internationally validated and standardized to be used in clinical practice. In summary, these future advances will lead to a higher sensitivity and specificity to detect CNS localization of malignancies, while in neuroinflammatory diseases (e.g., MS), CSF flow cytometry might become an important tool in the diagnosis, prognosis, and follow-up of patients. This work was supported by the Gratama Foundation, Teteringen, The Netherlands.Holroyd-Leduc J. How do I perform a lumbar puncture and analyze the results to diagnose bacterial meningitis? JAMA 2006; 296: 2012– 2022.Posner JB. Diagnosis and treatment of leptomeningeal metastases from solid tumors: Experience with 90 patients. Cancer 1982; 759– 772. Wiley Online LibraryGratama JW. CSF flow cytometry greatly improves diagnostic accuracy in CNS hematologic malignancies. Neurology 2007; 1674– 1679.Van der Planken M. Evaluation of ADVIA 120 CSF assay (Bayer) vs. chamber counting of cerebrospinal fluid specimens. Clin Lab Haematol 2004; 195– 199. Wiley Online LibrarySillevis Smitt P. Addition of serum-containing medium to cerebrospinal fluid prevents cellular loss over time. J Neurol(in press).Dorn-Beineke A. Modifications of haematology analyzers to improve cell counting and leukocyte differentiating in cerebrospinal fluid controls of the Joint German Society for Clinical Chemistry and Laboratory Medicine. Cytometry A 2009; 75A: 688– 691. Wiley Online LibraryReinhoefer M. Evaluation of the XE-5000 for the automated analysis of blood cells in cerebrospinal fluid. Clin Biochem 2009; 684– 691.Lindemans J. Evaluation of the new body fluid mode on the Sysmex XE-5000 for counting leukocytes and erythrocytes in cerebrospinal fluid and other body fluids. Clin Chem Lab Med 2010; 665– 675.Wilson WH. Diagnostic tools for neoplastic meningitis: Detecting disease, identifying patient risk, and determining benefit of treatment. Semin Oncol 2009;4 Suppl 2): S35– 45.Gonzalez-Barca E, et al. Identification of leptomeningeal disease in aggressive B-cell non-Hodgkin\'s lymphoma: Improved sensitivity of flow cytometry. J Clin Oncol 2009; 1462– 1469.Posner JB. Malignant cells in cerebrospinal fluid (CSF): The meaning of a positive CSF cytology. Neurology 1979; 1369– 1375.Vecht CJ. The diagnostic accuracy of magnetic resonance imaging and cerebrospinal fluid cytology in leptomeningeal metastasis. J Neurol 1999; 246: 810– 814.DeAngelis LM. Neuroimaging and cerebrospinal fluid cytology in the diagnosis of leptomeningeal metastasis. Ann Neurol 1995; 51– 57. Wiley Online LibraryBehringer DM. Clinical follow-up indicates differential accuracy of magnetic resonance imaging and immunocytology of the cerebral spinal fluid for the diagnosis of neoplastic meningitis-a single centre experience. Br J Haematol 2004; 124: 762– 768. Wiley Online LibrarySkoog L. Diagnosis of lymphoma, leukemia, and metastatic tumor involvement of the cerebrospinal fluid by cytology and immunocytochemistry. Diagn Cytopathol 1995; 14– 22. Wiley Online LibraryVecht CJ. Immunohistochemical analysis of the cerebrospinal fluid for carcinomatous and lymphomatous leptomeningitis. Br J Cancer 1990; 653– 654.Heidenreich F. Analysis of cerebrospinal fluid cells by flow cytometry and immunocytochemistry in inflammatory central nervous system diseases: Comparison of low- and high-density cell surface antigen expression. Diagn Cytopathol 1999; 313– 318. Wiley Online LibraryWildemann B. Single-cell PCR analysis of the immunoglobulin heavy-chain CDR3 region for the diagnosis of leptomeningeal involvement of B-cell malignancies using standard cerebrospinal fluid cytospins. J Neurol Sci 2004; 219: 83– 88.Siegal T. CSF analysis of IgH gene rearrangement in CNS lymphoma: Relationship to the disease course. J Neurol Sci 2006; 247: 39– 46.Boerman RH. Early detection of leptomeningeal metastasis by PCR examination of tumor-derived K-ras DNA in cerebrospinal fluid. Clin Chem 2000; 132– 133.Evans TG, et al. Nine-color flow cytometry for accurate measurement of T cell subsets and cytokine responses. Part I: Panel design by an empiric approach. Cytometry A 2008; 73A: 400– 410. Wiley Online LibraryLenkei R, et al. Cytofluorometric methods for assessing absolute numbers of cell subsets in blood. European Working Group on Clinical Cell Analysis. Cytometry 2000; 327– 346. Wiley Online LibraryZimmermann CW. A standardized protocol for flow cytometric analysis of cells isolated from cerebrospinal fluid. J Neurol Sci 1994; 121: 74– 78.Gratama JW. B and T cell imbalances in CSF of patients with Hu-antibody associated PNS. J Neuroimmunol 2008; 195: 164– 170.Khaw AV. Immunophenotypic patterns of T-cell activation in neuroinflammatory diseases. Acta Neurol Scand 2006; 113: 248– 255. Wiley Online LibrarySchmitt M. Flow cytometric analysis of T cell subsets in paired samples of cerebrospinal fluid and peripheral blood from patients with neurological and psychiatric disorders. Brain Behav Immun 2009; 134– 142.Stemme S. Adhesion molecule expression on cerebrospinal fluid T lymphocytes: Evidence for common recruitment mechanisms in multiple sclerosis, aseptic meningitis, and normal controls. Ann Neurol 1993; 155– 161. Wiley Online Library Lymphocyte phenotype and subset distribution in normal cerebrospinal fluid. J Neuroimmunol 1995; 39– 46. Central memory CD4+ T cells dominate the normal cerebrospinal fluid. Cytometry B Clin Cytom 2010; 80B: 43– 50. Wiley Online LibraryCibas ES. Diagnosing lymphoproliferative disorders involving the cerebrospinal fluid: Increased sensitivity using flow cytometric analysis. Diagn Cytopathol 2000; 369– 374. Wiley Online LibraryStetler-Stevenson M, et al. High incidence of occult leptomeningeal disease detected by flow cytometry in newly diagnosed aggressive B-cell lymphomas at risk for central nervous system involvement: The role of flow cytometry versus cytology. Blood 2005; 105: 496– 502.Duhrsen U. Detection of malignant haematopoietic cells in the cerebrospinal fluid by conventional cytology and flow cytometry. Clin Lab Haematol 2006; 22– 29. Wiley Online LibraryElsner B. Lymphoid and myeloid neoplasms involving cerebrospinal fluid: Comparison of morphologic examination and immunophenotyping by flow cytometry. Diagn Cytopathol 2002; 271– 275. Wiley Online LibraryBernacer M. Flow cytometry and the study of central nervous disease in patients with acute leukaemia. Br J Haematol 2001; 112: 381– 384. Wiley Online LibraryGoolsby CL. Enhanced detection of malignant lymphoma in cerebrospinal fluid by multiparameter flow cytometry. Am J Clin Pathol 1998; 110: 341– 346.Shaker S. Immunophenotyping and immunoglobulin heavy chain gene rearrangement analysis in cerebrospinal fluid of pediatric patients with acute lymphoblastic leukemia. Leuk Res 2009; 655– 661.Zofel P. Flow cytometry of cerebrospinal fluid (CSF) lymphocytes: Alterations of blood/CSF ratios of lymphocyte subsets in inflammation disorders of human central nervous system (CNS). Clin Chem Lab Med 1999; 231– 241.Broelsch CE. Flow cytometric \"rare event analysis”: A standardized approach to the analysis of donor cell chimerism. J Immunol Methods 2002; 262: 53– 69.Alonso N, et al. Clinical significance of occult cerebrospinal fluid involvement assessed by flow cytometry in non-Hodgkin\'s lymphoma patients at high risk of central nervous system disease in the rituximab era. Eur J Haematol 2010; 321– 328. Wiley Online Library Clinical significance of monoclonal B cells in cerebrospinal fluid. Cytometry B Clin Cytom 2005; 63B: 23– 27. Wiley Online LibraryRoda CD. Cerebrospinal fluid cytological and biochemical characteristics in the presence of CNS neoplasia. Arq Neuropsiquiatr 2007; 802– 809.Grisold W. FACS analysis-a new and accurate tool in the diagnosis of lymphoma in the cerebrospinal fluid. Clin Chim Acta 2002; 317: 101– 107.Friedman A, Glass J, et al. NCCN clinical practice guidelines in oncology. Central nervous system cancers. V. 1.2008 ed: http://www.nccn.org/professionals/physician_gls/PDF/cns.pdf. Flow cytometric analysis of lymphomas and lymphoproliferative disorders. Semin Hematol 2001; 111– 123.Gold DP. Results of a phase I clinical trial of a T-cell receptor peptide vaccine in patients with multiple sclerosis. I. Analysis of T-cell receptor utilization in CSF cell populations. J Neuroimmunol 1997; 15– 28.Wiendl H. Specific central nervous system recruitment of HLA-G(+) regulatory T cells in multiple sclerosis. Ann Neurol 2009; 171– 183. Wiley Online LibraryStinissen P. Compromised CD4+ CD25(high) regulatory T-cell function in patients with relapsing-remitting multiple sclerosis is correlated with a reduced frequency of FOXP3-positive cells and reduced FOXP3 expression at the single-cell level. Immunology 2008; 123: 79– 89. Wiley Online LibraryMalin JP. Analysis of lymphocyte subpopulations in cerebrospinal fluid and peripheral blood in patients with multiple sclerosis and inflammatory diseases of the nervous system. Acta Neurol Scand 1998; 310– 313. Wiley Online LibraryStemme S. Reduced frequency of memory CD8+ T lymphocytes in cerebrospinal fluid and blood of patients with multiple sclerosis. Autoimmunity 1995; 231– 239.Sindern E. Increase of CD8+ T-effector memory cells in peripheral blood of patients with relapsing-remitting multiple sclerosis compared to healthy controls. J Neuroimmunol 2007; 183: 168– 174.Posnett DN. The activation of memory CD4(+) T cells and CD8(+) T cells in patients with multiple sclerosis. J Neurol Sci 2005; 235: 11– 17.Saida T. Differences in systemic and central nervous system cellular immunity relevant to relapsing-remitting multiple sclerosis. J Neurol 2005; 252: 908– 915.Martinez-Caceres EM. CD4(+)CD45RO(+)CD49d(high) cells are involved in the pathogenesis of relapsing-remitting multiple sclerosis. J Neuroimmunol 2000; 111: 215– 223.Kaps M. CD45RA+ ICAM-3+ lymphocytes in cerebrospinal fluid and blood as markers of disease activity in patients with multiple sclerosis. Acta Neurol Scand 2000; 102: 326– 332. Wiley Online LibraryDorndorf W. Soluble and cell surface ICAM-1 as markers for disease activity in multiple sclerosis. Acta Neurol Scand 1998; 102– 109. Wiley Online LibraryReindl M. Cerebrospinal fluid B cells correlate with early brain inflammation in multiple sclerosis. PLoS One 2008; e2559.Hemmer B. Short-lived plasma blasts are the main B cell effector subset during the course of multiple sclerosis. Brain 2005; 128(Pt 7): 1667– 1676.Monson N. Cerebrospinal fluid B cells from multiple sclerosis patients are subject to normal germinal center selection. J Neuroimmunol 2007; 183: 189– 199.Racke MK. Effect of rituximab on the peripheral blood and cerebrospinal fluid B cells in patients with primary progressive multiple sclerosis. Arch Neurol 2005; 258– 264.Pistoia V. Recapitulation of B cell differentiation in the central nervous system of patients with multiple sclerosis. Proc Natl Acad Sci U S A 2004; 101: 11064– 11069.Ryder LP. Costimulatory CD80 (B7–1) and CD86 (B7–2) on cerebrospinal fluid cells in multiple sclerosis. J Neuroimmunol 1998; 179– 187.Andersen O. Increased expression of B7–1 costimulatory molecule on cerebrospinal fluid cells of patients with multiple sclerosis and infectious central nervous system disease. J Neuroimmunol 1997; 59– 68.Lyons JA. Rituximab reduces B cells and T cells in cerebrospinal fluid of multiple sclerosis patients. J Neuroimmunol 2006; 180: 63– 70.Hemmer B, et al. Immune surveillance in multiple sclerosis patients treated with natalizumab. Ann Neurol 2006; 743– 747. Wiley Online LibraryFrederiksen JL. Immunological effects of oral high-dose methylprednisolone in acute optic neuritis and multiple sclerosis. Eur J Neurol 2000; 281– 289. Wiley Online LibraryVerhulst SJ. CSF B-cell expansion in opsoclonus-myoclonus syndrome: A biomarker of disease activity. Mov Disord 2004; 770– 777. Wiley Online LibraryVerhulst SJ. B- and T-cell markers in opsoclonus-myoclonus syndrome: Immunophenotyping of CSF lymphocytes. Neurology 2004; 1526– 1532.Orfao A. Flow cytometric analysis of cerebrospinal fluid samples and its usefulness in routine clinical practice. Am J Clin Pathol 2002; 117: 952– 958.Recktenwald D. Model study detecting breast cancer cells in peripheral blood mononuclear cells at frequencies as low as 10(-7). Proc Natl Acad Sci U S A 1995; 537– 541.Wilcox MH. Effect of pH changes in cerebrospinal fluid specimens on bacterial survival and antigen test results. J Clin Pathol 1996; 249– 253.Vernau KM. Effects of time, initial composition, and stabilizing agents on the results of canine cerebrospinal fluid analysis. Vet Clin Pathol 2006; 72– 77. Wiley Online LibrarySutherland DR. Considerations for the control of background fluorescence in clinical flow cytometry. Cytometry B Clin Cytom 2009; 76B: 355– 364. Wiley Online LibraryHerzenberg LA. A single laser method for subtraction of cell autofluorescence in flow cytometry. Cytometry 1987; 114– 119. Wiley Online LibraryMurphy RF. Cell-by-cell autofluorescence correction for low signal-to-noise systems: Application to epidermal growth factor endocytosis by 3T3 fibroblasts. Cytometry 1986; 558– 565. Wiley Online LibraryJacobberger JW. Streptavidin-based quantitative staining of intracellular antigens for flow cytometric analysis. Cytometry 1992; 711– 721. Wiley Online Library Novak RW. Lack of validity of standard corrections for white blood cell counts of blood-contaminated cerebrospinal fluid in infants. Am J Clin Pathol 1984; 95– 97.Hopf HC. Influence of artificial blood contamination of the analysis of cerebrospinal fluid. J Neurol 1981; 226: 187– 193.de Graaf SS. Prognostic significance of blasts in the cerebrospinal fluid without pleiocytosis or a traumatic lumbar puncture in children with acute lymphoblastic leukemia: Experience of the Dutch Childhood Oncology Group. J Clin Oncol 2006; 2332– 2336.Takezako N, et al. The diagnostic value of kappa/lambda ratios determined by flow cytometric analysis of biopsy specimens in B-cell lymphoma. Clin Lab Haematol 2002; 33– 36. Wiley Online LibraryTorlakovic E. Evaluation of lymphoid cell populations in cytology specimens using flow cytometry and polymerase chain reaction. Diagn Mol Pathol 1999; 183– 188.Bennett G. Flow-cytometric algorithm on fine-needle aspirates for the clinical workup of patients with lymphadenopathy. Diagn Cytopathol 1998; 274– 278. Wiley Online LibraryStetler-Stevenson M. Flow cytometric analysis of kappa and lambda light chain expression in evaluation of specimens for B-cell neoplasia. Cytometry 1996; 243– 252. Wiley Online LibraryFallon MA. The role of flow cytometry in the diagnosis of lymphoma: A critical analysis. Ann Clin Lab Sci 1994; 6– 11.Normansell DE. Quantitative criteria for clonality in the diagnosis of B-cell non-Hodgkin\'s lymphoma by flow cytometry. Mod Pathol 1993; 155– 161.Gonchoroff NJ. Rapid immunotyping of B-cell non-Hodgkin\'s lymphomas by flow cytometry. A comparison with the standard frozen-section method. Am J Clin Pathol 1990; 280– 286.van den Bent MJ. Incidence of central nervous system involvement in chronic lymphocytic leukemia and outcome to treatment. J Neurol 2008; 255: 828– 830.Louis DN. Chronic lymphocytic leukemia and the central nervous system: A clinical and pathological study. Neurology 1996; 19– 25.Orfao A. Generation of flow cytometry data files with a potentially infinite number of dimensions. Cytometry A 2008; 73A: 834– 846. Wiley Online LibraryMaecker HT. Standardization and optimization of multiparameter intracellular cytokine staining. Cytometry A 2008; 73A: 984– 991. Wiley Online Library Unsupervised immunophenotypic profiling of chronic lymphocytic leukemia. Cytometry B Clin Cytom 2006; 70B: 124– 135. Wiley Online LibraryOrfao A. A new automated flow cytometry data analysis approach for the diagnostic screening of neoplastic B-cell disorders in peripheral blood samples with absolute lymphocytosis. Leukemia 2006; 1221– 1230.Van Dongen JJ, et al. Automated pattern-guided principal component analysis vs expert-based immunophenotypic classification of B-cell chronic lymphoproliferative disorders: A step forward in the standardization of clinical immunophenotyping. Leukemia 2010; 1927– 1933.Levy NB. A comparison of polymerase chain reaction examination of cerebrospinal fluid and conventional cytology in the diagnosis of lymphomatous meningitis. Cancer 1996; 543– 548. Wiley Online LibraryThiel E. CSF evaluation in primary CNS lymphoma patients by PCR of the CDR III IgH genes. Neurology 2002; 390– 396. The full text of this article hosted at iucr.org is unavailable due to technical difficulties. Please check your email for instructions on resetting your password. If you do not receive an email within 10 minutes, your email address may not be registered, and you may need to create a new Wiley Online Library account. Can\'t sign in? Forgot your username? Enter your email address below and we will send you your username If the address matches an existing account you will receive an email with instructions to retrieve your username