Abstract
Intracellular thioguanine nucleotides (6-TGN) are the major cytotoxic metabolites of mercaptopurine (6-MP). Red blood cell (RBC) 6-TGN concentrations were measured in a group of 120 consecutive children with lymphoblastic leukemia (ALL) to assess interpatient variability and its clinical importance. Assays were performed after at least 2 months 6-MP maintanance chemotherapy and a minimum 7 days unattenuated protocol dose of 75 mg/m2. Observed 6-TGN concentrations ranged from 126 to 832 pmol/8 x 10(8) RBCs (median, 275). There was a correlation between 6-TGN and neutropenia 14 days postassay (rs = .51; P less than .0005), and an inverse correlation between 6-TGN and the length of time uninterrupted full protocol dose was tolerated without neutropenia (rs = -.3; P less than .01). After a median follow-up of 49 months, 19 children had relapsed, of whom 17 (89%) had 6-TGN concentrations below the group median (log-rank chi 2 = 11.9; P less than .001). Multivariate analysis using Cox's proportional hazards regression showed the 6-TGN effect on disease control to be independent of diagnostic WBC count, sex, age, immunological cell type, French-American-British (FAB) type, variation in other antineoplastic therapy, and duration of remission at the time of 6-TGN assay. Children with ALL taking the same dose of 6-MP show great variability in its measurable cytotoxic effect, and this variability is apparently important in predicting treatment outcome.