Shortening infusion time for high-dose methotrexate alters antileukemic effects: a randomized prospective clinical trial

Abstract

Purpose: To determine whether shortening the infusion duration of high-dose methotrexate (HDMTX; 1 g/m(2)) affects the in vivo accumulation of active methotrexate polyglutamates (MTXPG(1-7)) in leukemia cells and whether this differs among major acute lymphoblastic leukemia (ALL) subtypes.

Methods: From June 2000 through October 2007, 356 children with ALL were randomly assigned to receive initial single-agent treatment with HDMTX (1 g/m(2)) as either a 24-hour infusion or a 4-hour infusion at two pediatric hospitals in the United States. The primary outcome measures were the accumulation of MTXPG(1-7) in leukemia cells and the antileukemic effects (eg, inhibition of de novo purine synthesis in bone marrow ALL cells, and decrease in circulating ALL cells).

Results: The 24-hour infusion resulted in significantly higher amounts of MTXPG(1-7) in bone marrow leukemia cells (median: 1,695 v 1,150 pmol/10(9) cells, P = .0059), and better antileukemic effects. The 24-hour infusion had the greatest effect on MTXPG(1-7) accumulation in hyperdiploid ALL (median: 3,919 v 2,417 pmol/10(9) cells, P = .0038); T-cell ALL exhibited smaller differences in MTXPG(1-7) but greater antileukemic effects with the longer infusion (median decrease in leukemia cells: 88.4% v 51.8%, P = .0075). In contrast, infusion duration had no significant impact on MTXPG(1-7) accumulation or antileukemic effects in ALL with the t(12;21)/(ETV6-RUNX1) chromosomal translocation.

Conclusion: Shortening the infusion time of HDMTX reduces accumulation of active methotrexate in leukemia cells and decreases antileukemic effects, with differing consequences among major ALL subtypes.

Fig 1.

CONSORT flow chart depicting enrollment, random assignment, and analysis. HDMTX, high-dose methotrexate; IV, intravenous; MTXPG_1-7, total intracellular methotrexate polyglutamates; DNPS, de novo purine synthesis.

Fig 2.

Accumulation of total intracellular methotrexate polyglutamates (MTXPG_1-7) in bone marrow acute lymphoblastic leukemia (ALL) cells and antileukemic effects. (A) Accumulation of total MTXPG_1-7 (pmol/10⁹ cells) grouped by the MTX infusion length (4 hour or 24 hour) and the ALL lineage (B-cell ALL [B-ALL] or T-cell ALL [T-ALL]). The significance level comparing 4-hour versus 24-hour infusion length in all patients, B-ALL, and T-ALL are P = .0059, P = .0049, and P = .18, respectively. Medians, quartiles, nonoutlier range (defined as 1.5 times the interquartile range), and outliers (plus signs) are depicted. (B) Percent change in de novo purine synthesis (DNPS) induced by MTX in bone marrow leukemia cells 42 to 44 hours after start of the infusion. The significance level comparing 4-hour versus 24-hour infusion length in all patients, B-ALL, and T-ALL are P = .021, P = .081, and P = .24, respectively. (C) Leukemic response expressed as residual in WBC from day 0 to day 3 grouped by the MTX infusion length (4 hour or 24 hour) and the ALL lineage (B-ALL or T-ALL). The significance level comparing 4-hour versus 24-hour infusion length in all patients, B-ALL, and T-ALL are P = .038, P = .66, and P = .038, respectively. Medians, quartiles, nonoutlier range (defined as 1.5 times the interquartile range), and outliers (plus signs) are depicted.

Fig 3.

Cumulative incidence of leukemic relapse. Illustrates the cumulative incidence of risk of relapse with patients categorized into three groups on the basis of the accumulation of total intracellular methotrexate polyglutamates (MTXPG_1-7) in bone marrow acute lymphoblastic leukemia (ALL) cells. High MTXPG accumulation represented the top quartile in each of the five ALL subtypes, intermediate MTXPG accumulation represented the middle two quartiles, and low MTXPG accumulation represented the bottom quartile.