Purpose: Thioguanine nucleotides (TGNs) mediate the cytotoxicity of mercaptopurine (MP). Methylated MP metabolites (formed by thiopurine methyltransferase [TPMT]) and methotrexate (MTX) polyglutamates can inhibit de novo purine synthesis. We explored whether dose adjustment of MP and MTX by erythrocyte (E) levels of TGN and MTX (including polyglutamates) could improve outcome in childhood acute lymphoblastic leukemia (ALL).
Patients and methods: A total of 538 children with ALL were randomly assigned to have their oral MP/MTX maintenance therapy adjusted by white cell counts (WBC), E-TGN, and E-MTX (pharmacology group), or by WBC only (control group).
Results: After a median follow-up of 7.8 years, 79 patients had relapsed. Cox regression analysis showed an increased risk of relapse for boys (P =.00003), high WBC at diagnosis (P =.03), pharmacology arm (6.6 times increased relapse hazard for girls), high TPMT activity (P =.002), and high average neutrophil counts during maintenance therapy (P =.0009), with a significant interaction between sex and randomization group (P =.0007). For girls, the relapse risk was 5% in the control group and 19% in the pharmacology group (P =.001) because of an increased relapse hazard during the first year after cessation of therapy. TPMT activity was the most significant predictor of relapses among girls in the pharmacology arm (P <.0001). Overall, the TPMT activity was higher for patients who relapsed after cessation of therapy compared with those who stayed in remission (girls 19.5 v 17.4 U/mL, P =.03; boys 19.3 v 18.0 U/mL, P =.04).
Conclusion: Adding pharmacologically guided treatment intensification to dose adjustments by blood counts may not be warranted for girls, whereas new approaches to optimize maintenance therapy are needed for boys.