CSF fibronectin in Burkitt's lymphoma: an early marker for CNS involvement

The prognostic significance of serum and cerebrospinal fluid MMP-9, CCL2 and sVCAM-1 in leukemia CNS metastasis

Metastasis to the central nervous system (CNS) is the primary obstacle in leukemia treatment. Matrix metalloproteinase-9 (MMP-9), chemokine ligand-2 (CCL2) and soluble vascular adhesion molecule-1 (sVCAM-1) play crucial roles in tumor cell adhesion, motivation and survival, but their roles in leukemia CNS metastasis remain to be elucidated. We investigated the prognostic significance of serum and cerebrospinal fluid (CSF) MMP-9, CCL2 and sVCAM-1 in leukemia patients to explore their potential as predictive biomarkers of the development of CNS leukemia (CNSL). MMP-9, CCL2 and sVCAM-1 were measured in paired CSF and serum samples collecting from 33 leukemia patients with or without CNS metastasis. Other risk factors related to CNSL prognosis were also analyzed. sVCAM-1Serum and CCL2Serum/CSF were significantly higher in the CNSL group than in the non-CNSL group and the controls (p < 0.05). MMP-9Serum was insignificantly lower in the CNSL group than in the non-CNSL group and the controls (p > 0.05). No differences were found for the sVCAM-1Serum, CCL2Serum, and MMP-9Serum levels between non-CNSL patients and controls (p > 0.05). MMP-9CSF was significantly higher in the CNSL group than both the non-CNSL and the control groups (p < 0.05). The indexes of sVCAM-1, CCL2, and MMP-9 in the CNSL group were lower than in the controls (p < 0.05). Positive correlations were determined between the MMP-9CSF and the ALBCSF/BBB value/WBCCSF, between sVCAM-1Serum and the WBCCSF/BBB value. Negative correlations existed between MMP-9Serum and the ALBCSF/BBB value/WBCCSF, and between the CCL2 index and ALBCSF. sVCAM-1Serum was positively associated with event-free survival (EFS), and patients with higher levels of ALBCSF, MMP-9CSF/Serum, CCL2CSF/Serum, and sVCAM-1CSF/Serum had shorter EFS. MMP-9CSF, CCL2CSF and sVCAM-1CSF are the first three principal components analyzed by cluster and principal component analysis. Our data suggest that MMP-9, CCL2 and sVCAM-1 in the CSF may be more potent than serum in predicting the possibility of leukemia metastatic CNS and the outcome of CNSL patients.

Diagnostic strategies to investigate cerebrospinal fluid involvement in haematological malignancies

Central nervous system (CNS) involvement is a fatal complication of certain haematological malignancies with an incidence as high as 25% in specific leukaemia/lymphoma subtypes. It is often accompanied by 'occult' cerebrospinal fluid (CSF) involvement at diagnosis, which is frequently missed by conventional cytology examination. Unfortunately, a diagnostic gold standard is yet unavailable since CSF morphology may be negative for malignant cells in up to 45% of patients with suspected meningeal involvement. New technologies such as flow cytometry, molecular genetics and newer biomarkers may improve sensitivity and specificity facilitating the diagnosis of CNS involvement as well as effective prophylaxis and successful treatment.

Humoral CSF parameters in the differential diagnosis of hematologic CNS neoplasia

Cerebrospinal fluid (CSF) and serum samples of 20 patients with central nervous system manifestations of hematological malignancies including primary cerebral lymphoma (n = 5) and disseminated non-Hodgkin lymphoma (n = 7) were examined for albumin, IgG, IgM, fibronectin, beta 2-microglobulin, interleukin-6, soluble interleukin-2 receptor, tumor necrosis factor alpha, and oligoclonal immunoglobulin bands. Although a broad range of abnormalities were detected, no reliable CSF parameter for the diagnosis of leptomeningeal spread from hematological neoplasias could be identified. An analysis of 61 repeat lumbar punctures added little to the findings of the first CSF examinations. Currently, immunochemical studies of CSF cell surface markers and early biopsy have probably more clinical value than the determination of the humoral CSF parameters included in this study. However, analysis of cytokine synthesis by single CSF cells using molecular biology techniques may improve the differential diagnosis of hematological neoplasia of the brain and spinal cord in the future.

Tumor cell dissemination triggers an intrathecal immune response in neoplastic meningitis

The intrathecal immune response in neoplastic meningitis (NM) was studied by quantitation of immune parameters such as immunoglobulin G (IgG); IgM; interleukins (IL) 1, 2, 4, and 6; soluble IL-2 receptors (sIL-2R); interferon gamma (IFNy); tumor necrosis factor-alpha (TNF alpha); and three tumor markers, carcinoembryonic antigen (CEA), alpha-fetoprotein (AFP), and fibronectin (FN), in 47 paired cerebrospinal fluid (CSF) and serum samples from patients with NM from different carcinomas, malignant melanoma, and lymphoma. Elevated IgG and IgM indices, CSF oligoclonal Ig bands, and CSF IL-6 indicated an intrathecal immune activation in most patients with NM. Results for IL-1, IL-2, and IL-4 were always negative. sIL-2R and IFNy were detected occasionally but not associated with specific malignant neoplasms. CSF TNF alpha was detected only in NM from cases of malignant melanoma. None of the immune parameters proved useful for the differentiation of NM from autoimmune or inflammatory conditions. Immune parameters were not correlated with tumor markers CEA, AFP, or FN. Results for AFP were positive only in a case of glioblastoma. CEA was a useful and specific diagnostic parameter in carcinomatous NM. CSF FN levels frequently were elevated but are not specific for NM.

Increased intrathecal synthesis of fibronectin in bacterial and carcinomatous meningitis

Immunoreactive fibronectin (Fn) was quantified in paired cerebrospinal fluid (CSF) and serum samples from patients with bacterial meningitis (n = 46), tick-borne encephalitis (TBE) (n = 6), HIV infection (n = 6), Guillain-Barré syndrome (n = 5), carcinomatous meningitis (n = 11), multiple sclerosis (n = 15), disk disease (n = 11), and controls (n = 28). A highly significant elevation of CSF Fn was found in bacterial meningitis, TBE, and carcinomatous meningitis. There were no significant differences in serum Fn between any of the groups. An Fn index to estimate the rate of intrathecal Fn synthesis reached the highest value in bacterial meningitis. Our findings suggest that CSF Fn may be an indicator of adequate host reaction and tissue repair. For diagnostic purposes, the determination of CSF Fn probably does not add much to routine CSF laboratory tests.