Allopurinol is used to prevent hyperuricemia in newly diagnosed patients with acute lymphoblastic leukemia (ALL). Although allopurinol has been shown to inhibit de novo purine synthesis (DNPS) in fibroblasts in vitro, this effect has not been assessed in ALL lymphoblasts. We assessed DNPS in ALL lymphoblasts in 46 consecutive patients with ALL. DNPS was determined by 14C-formate incorporation in purine bases both at diagnosis (n = 46) and 44h after MTX therapy +/- allopurinol (n = 31). The 27 patients who had received no allopurinol prior to the diagnostic bone marrow aspirate had significantly higher rates of DNPS (median, 102 fmol new purines/nmol total purines/h) compared to the 12 patients who had received more than one dose of allopurinol (100 mg/m2 orally) (median, 2.3 fmol/nmol/h; P < 0.001); the seven patients who received one dose of allopurinol had intermediate rates of DNPS (median, 58.5 fmol/nmol/h). Among patients who were evaluable for MTX effects at 44h (n = 31), the percent inhibition of DNPS was greater in the eight patients who received concomitant allopurinol (median, 100% inhibition) compared to the 23 patients who received only methotrexate therapy (median, 89% inhibition, P = 0.03). These data indicate that allopurinol suppresses may contribute to the decrease in circulating blasts in patients with newly diagnosed acute leukemias.