Variable correlation between 6-mercaptopurine metabolites in erythrocytes and hematologic toxicity: implications for drug monitoring in children with acute lymphoblastic leukemia

Nineteen pediatric patients affected by acute lymphoblastic leukemia (ALL) were examined weekly with respect to 6-mercaptopurine nucleotide (6-MPN) and 6-thioguanine nucleotide (6-TGN) levels in erythrocytes during the course of maintenance treatment with 6-MP 50 mg/m2 per d and results were related to various parameters of bone marrow function to assess, in the same individual, the level of reliability of 6-MP metabolites in predicting a later change in peripheral blood cell counts. Median values for 6-MPN and 6-TGN were 57 and 200 pmol/8 x 10(8) erythrocytes, respectively, as measured by reversed-phase high-performance liquid chromatography (HPLC). 6-TGN levels in erythrocytes were inversely related with white blood cell count (r = -0.463, p < 0.0001, n = 361), absolute neutrophil count (r = -0.386, p < 0.0001, n = 347), erythrocyte (r = -0.354, p < 0.0001, n = 287), and platelet counts (r = -0.24, p < 0.0001, n = 319) in the majority of patients (n = 10-12), while no correlation was found for 6-MPN. In the remaining children, no evidence of correlation was demonstrated between 6-TGN levels and myelotoxicity. The results confirm the role of 6-TGN as the reference cytotoxic metabolite for evaluating the exposure to 6-MP and identifying treatment compliance in ALL children but indicate the limits of a follow-up based solely on metabolite levels and suggest that a more correct approach remains the double monitoring of 6-TGN and blood cell count with differential.