Determinants of cerebrospinal fluid arsenic concentration in patients with acute promyelocytic leukemia on oral arsenic trioxide therapy.

The extent of and factors controlling arsenic penetration into the central nervous system (CNS) remain unclear. Elemental arsenic levels in 67 paired cerebrospinal fluid (CSF) and plasma samples from 9 patients with acute promyelocytic leukemia (APL) on oral arsenic trioxide (As2O3), obtained during intrathecal chemotherapy (treatment of CNS APL, n = 6; prophylaxis, n = 3) were measured. Median arsenic levels of CSF and plasma were 95.8 nmol/L (range, 3.5-318.9 nmol/L) and 498.9 nmol/L (range, 36.3-1892.8 nmol/L). As a group, CSF and plasma arsenic was linearly correlated (P < .001), with CSF at 17.7% the plasma level. The CSF/plasma arsenic ratio, which reflected the arsenic CSF penetration efficiency, varied significantly in individual patients (P < .001). Repeated intrathecal chemotherapy and presence of blasts in CSF did not affect the CSF/plasma arsenic ratio. Plasma arsenic was the only significant determinant of CSF arsenic levels. CSF arsenic was present at therapeutically meaningful levels, implying that As2O3 therapy might be beneficial in CNS APL.

Arsenic in the cerebrospinal fluid of a patient receiving arsenic trioxide for relapsed acute promyelocytic leukemia with CNS involvement.

We report on a 42-year-old patient whose relapse of acute promyelocytic leukaemia (APL) included meningeal infiltration. Since he had previously experienced ATRA syndrome, he received arsenic trioxide (ATO) plus intrathecal therapy with cytarabine, prednisone, and methotrexate. We measured the concentration of arsenic in his cerebrospinal fluid (CSF). Arsenic showed a peak CSF concentration of 0.008 mg/l (0.11 micromol/l) and a nadir of 0.002 mg/l (0.027 micromol/l), both representing about 14% of blood levels. ATO thus crosses the blood-CSF-barrier when administered intravenously, but the concentration in CSF is probably not sufficient for treatment of meningeal leukemia.

Investigations of the metabolites of the trypanocidal drug melarsoprol.

BACKGROUND: Melarsoprol remains the first-choice drug for trypanosomiasis (human African sleeping sickness). To contribute to the sparse pharmacologic data and to better understand the cause of the frequent serious adverse reactions, we investigated the metabolism of this 50-year-old organoarsenic compound. RESULTS: The half-life of melarsoprol determined by HPLC was <1 hour compared with 35 hours determined by bioassay and atomic absorption spectroscopy, indicating the existence of active metabolites. One metabolite, melarsen oxide, was identified by ultraviolet HPLC after incubation of melarsoprol with microsomes. The maximum plasma concentration of melarsenoxide was reached 15 minutes after administration; the clearance was 21.5 mL/min/kg and the half-life of free melarsen oxide was 3.9 hours. Either melarsen oxide or a yet-undiscovered active metabolite is irreversibly bound to proteins, as shown by ultrafiltration, precipitation experiments, and atomic absorption spectroscopy. Because of the poor pharmaceutical properties of melarsoprol, the therapeutic potential of melarsen oxide was investigated. In a rodent model of acute infection, 20 of 20 mice were cured (0.1 to 1 mg/kg intravenously or 2.2 mg/kg intraperitoneally). In a rodent model of central nervous system infection, five of six mice survived for more than 180 days (5 mg/kg intravenously), indicating a sufficient melarsen oxide penetration across the blood-brain barrier. CONCLUSION: The prospects for the future of trypanosomiasis treatment are deplorable. Investigations on the improvement of the use of the old drugs are therefore required. The results of this study may build a basis for further research on the cause of severe adverse reactions.

Concentrations of zinc, iron, molybdenum, arsenic, and lithium in cerebrospinal fluid of patients with brain tumors.

We used flameless atomic absorption spectrophotometry to measure concentrations of Fe, Mo, Li, As, and Zn in cerebrospinal fluid (CSF) of patients with malignant brain tumors benign brain tumors, non-brain malignant tumors and control (non-neoplastic disease) patients. Mean (and SD) concentrations (microgram/L) of these elements in the control group were 62.7 (28.7) for Fe, 6.8 (4.8) for Mo, 0.7 (2.0) for Li, 1.3 (0.7) for As, 7 (5.9) for Zn. We could detect Li in less than 53% of controls. Zn concentrations in CSF of patients with astrocytoma (malignant brain tumor), benign brain tumors, or non-brain tumors were significantly (p less than 0.05) less than in control patients; the ratios for mean concentrations of Zn in tumor patients/control patients for the above groups were 0.3, 0.20, and 0.17, respectively. Concentrations of As in CSF of patients with non-brain malignant tumors were significantly (p less than 0.05) higher than in the controls; the ratio for mean CSF concentration of As in patients with non-brain tumors/control patients was 2.9. Differences in the concentrations of Fe, Li, or Mo among the various groups were nonsignificant.