Aminopterin (AMT) was replaced by methotrexate (MTX) in the therapy for patients with acute lymphoblastic leukemia (ALL) because of fears that the former antifolate produced excessive and unpredictable toxicity. Regardless of the presumed risk, AMT has greater accumulation and metabolism by leukemic blasts than MTX in vitro and had 100% bioavailability in early clinical trials. Therefore we are using AMT as a replacement for MTX to treat newly diagnosed patients with ALL at high risk (HR) of relapse by NCI criteria. The Cancer Institute of NJ ALL trial (CINJALL) is based an earlier regimen using intensive antimetabolite therapy (100 mg/m2 MTX in 4 divided doses weekly or every other week for 76 weeks) delivered orally on an outpatient basis. With 7 years median follow up, the 5-year EFS for the 106 patients treated on this earlier protocol is 84% ±3% (personal communication, N. Winick, UT Southwestern Medical Center). HR patients on CINJALL are non-randomly assigned to receive a single delayed intensification and oral AMT at its MTD, 4 mg/m2 in two divided doses 12 hours apart, in place of oral MTX (given to standard risk (SR) patients). There is no intravenous MTX or AMT on either arm. To date, 44 children have enrolled: 15 on the SR (MTX) arm (median age 3 years; range 1–9) and 29 on the HR (AMT) arm (median age 14 years; range 2–21). As of this analysis, median follow up is 85 weeks (range 0–177 weeks). Six HR and one SR patient have completed all scheduled therapy. Twenty HR and 14 SR patients have completed the 12-week consolidation, which includes weekly divided-dose antifolate therapy. The mean AMT AUC, 0.64 ± 0.24 mmol·hr/L, was not statistically different from the AUC after IV AMT at the same dose on the Phase I trial, confirming complete bioavailability. RBC antifolate was measured after a minimum of 8 weeks of therapy as an indication of marrow exposure. The lack of significant difference between mean RBC MTX and RBC AMT (240 ± 63 vs. 211 ± 22 pmol/ml RBC), despite a 25-fold difference in weekly dose, supports the pharmacodynamic equivalence of the two drugs at this dose and schedule. Mean CSF MTX 2-6h after oral MTX was significantly greater than CSF AMT (15.8 ± 2.3 vs 3.8 ± 1.3 nM; P < 0.001). Focusing on toxicity in consolidation, where the the two arms differ only in the antifolate given, there was no significant difference between the MTX and AMT arms in mean WBC count, hemoglobin or platelets. HR patients were no more likely than SR patients to experience >10% body weight loss, admission to the hospital or treatment interruption for treatment-related toxicity, neurotoxicity, SGOT >5 x normal, bilirubin > 2x normal, ANC <750/mL, or hemoglobin <7 g/dL. HR patients had a higher frequency of severe thrombocytopenia (platelets < 75,000/mL) than those on the MTX arm (9.9 vs. 4.8% of CBCs), but this difference was not statistically significant. In summary, this early data show that AMT, in its current formulation and at this dose and schedule, can be safely given to children with ALL in place of systemic MTX. In contrast with reports from the early history of antifolate use, we are finding no significant differences between AMT and MTX in the magnitude or coefficient of variation of clinical toxicity. In addition, lower CSF accumulation of AMT may result in decreased frequency and/or severity of subacute or delayed neurotoxicity relative to MTX. The proper role for AMT in therapy for patients with ALL deserves further study.