Unreliability of beta-2-microglobulin in early detection of central nervous system relapse in acute lymphoblastic leukemia

Diagnosis of acute leukemia in cerebrospinal fluid (CSF-acute leukemia)

Cerebrospinal fluid-acute leukemia (CSF-acute leukemia) is a frequent and serious complication in patients with acute leukemia. One of the major problems of this complication is the diagnosis process itself. CSF cytology is currently considered the gold standard for establishing the diagnosis, a technique which presents various processing limitations, seriously impacting the predictive values. In the last 11 years, studies of CSF flow cytometry analysis done in patients with acute leukemia have demonstrated superiority in comparison with CSF cytology. Although comparative studies between these two techniques have been reported since 2001, no new consensus or formal changes to the gold standard have been established for the CSF acute leukemia diagnosis. The evidence suggests that positive flow cytometry cases, considered as indeterminate cases, will behave like disease in the central nervous system (CNS). Nevertheless, we think there are some variables and considerations that must be first evaluated under research protocols before CNS relapse can be established with only one positive flow cytometry analysis in the setting of indeterminate CSF samples. This paper proposes a diagnostic algorithm and complementary strategies.

Cerebrospinal fluid tumour markers in patients treated for meningeal malignancy

The results of cerebrospinal fluid (CSF) biochemical markers were compared with conventional CSF cytology in patients treated for leptomeningeal metastases from extra cranial malignancies. For lumbar CSF, before treatment, no statistically significant difference of the probabilities of being positive was found between CSF cytology and a classification by linear discriminant analysis, based on patient's age, of beta-glucuronidase and beta 2-microglobulin. During treatment, classification by linear discriminant analysis was found more often positive than cytology. Possible mechanisms for this difference are discussed. For ventricular CSF a correlation was found between CSF cytology and beta-glucuronidase for solid tumours, and between CSF cytology and beta 2-microglobulin for haematological malignancies. Reference values for ventricular protein, CEA beta-glucuronidase and beta 2-microglobulin were obtained for cytological negative samples.

Tumor necrosis factor in the cerebrospinal fluid of children with central nervous system leukemia

To clarify the role of cytokines in cerebrospinal fluid (CSF) in the pathogenesis of central nervous system (CNS) leukemia, three cytokine activities, interleukin 1 (IL-1)-beta, tumor necrosis factor (TNF)-alpha, and interferon (IFN)-gamma, and their correlations with other laboratory studies of the CSF were analysed in 23 children with acute leukemia. These patients were classified into three groups: group A (n = 8)--patients with overt CNS leukemia, group B (n = 5)--patients with CNS leukemia in remission, group C (n = 10)--patients without CNS disease. IFN-gamma in the CSF was undetectable in these 23 patients. There was no difference in IL-1-beta levels among the three groups. However, TNF-alpha levels were significantly higher in group A than in group B, and higher in group B than in group C. By Kendall's rank sum test, high TNF levels in CSF correlated with high CSF leukemic cell counts and low sugar levels. In two patients with overt CNS leukemia, the TNF level in the CSF decreased gradually with intrathecal chemotherapy. These results indicate that TNF released from stimulated cells in the cerebrospinal space may induce CNS leukemia-related symptoms or alter laboratory parameters measured in the CSF. TNF levels in CSF may also prove useful in diagnosing early CNS involvement in children with acute leukemia.

Cerebrospinal fluid B2-microglobulin levels in meningeal involvement by malignancy

Cerebrospinal fluid (CSF) and serum B2-microglobulin (B2m) levels were measured prospectively in 63 patients with hematological malignancies and 14 patients with solid tumours to evaluate the correlation between elevated levels and malignant infiltration of meninges. Serial CSF B2-m levels were also measured in 18 patients who received prophylactic intrathecal cytotoxic treatment. CSF B2-m levels were significantly higher in patients with central nervous system (CNS) involvement than in those without (p less than 0.001). A CSF B2-m level greater than 1.80 mg/L was closely associated with CNS disease (specificity 96%, sensitivity 76%) and CNS infiltration was also likely when the CSF B2-m level exceeded a simultaneously drawn serum level (specificity 98%, sensitivity 46%). Intrathecal methotrexate prophylaxis resulted in a consistent and significant rise in CSF B2-m levels with an average increase of 96% during a course of intrathecal injections. These results suggest that CSF B2-m levels may not be helpful for predicting early CNS relapse in these patients. However the CSF B2-m level and the corresponding serum B2-m level is a useful adjunct to the cytological diagnosis of CNS involvement by malignancy at presentation. Its value in predicting early CNS relapse and documenting response to CNS treatment requires further clarification.

The glomerular sieving of pepsinogen A and C in man

Pepsinogen A (PGA) and pepsinogen C (PGC) are negatively charged, low molecular weight (LMW) proteins with a striking difference in renal handling: PGA (molecular weight 43,500 daltons) shows a high fractional excretion while the fractional excretion of PGC (molecular weight 40,500 daltons) is low, presumably due to tubular reabsorption. As these data suggest a high glomerular sieving of pepsinogens, we assessed the glomerular sieving coefficient (GSC) of PGA, PGC and several other proteins from their renal extractions. For this purpose blood samples were obtained simultaneously from the aorta (A) and right renal vein (V) in nine patients undergoing an elective heart catheterization. After correction of A-V differences for diuresis with the A-V difference of transferrin, GSCs (+/- SEM) for PGA and PGC were 0.90 +/- 0.14 and 0.85 +/- 0.17, respectively, GSC of beta 2-microglobulin being 0.90 +/- 0.12. For albumin and IgG, known to have a low GSC, low values were found. It is concluded that the GSC of a LMW protein in man can be calculated from both its A-V difference over the kidney and the A-V difference of an inert marker with a GSC of 1, provided they are corrected by the A-V difference of an inert marker with a GSC of 0. Our results demonstrate that PGA and PGC are almost freely filtered through the glomerular basement membrane despite their size and negative net molecular charge.