A prospective study of a simple algorithm to individually dose high-dose methotrexate for children with leukemia at risk for methotrexate toxicities

Abstract

Purpose: High-dose methotrexate (HDMTX) is critical to the successful treatment of pediatric acute lymphoblastic leukemia (ALL) but can cause significant toxicities. This study prospectively evaluated the effectiveness of a fixed algorithm which requires no real-time pharmacokinetic modeling and no previous patient exposure to HDMTX, to individualize HDMTX dosing for at-risk patients with the aim of avoiding methotrexate-related toxicities. Methods: We developed a simple algorithm to individualize HDMTX infusions with 0–2 rate adjustments based on methotrexate levels during the infusion. This was a prospective, open-label, study; eligible patients were identified and referred by their oncologist. Results: Fifty-four evaluable cycles of HDMTX (5 g/m 2 over 24 h) were administered to 22 patients. Blood samples were obtained in 21 patients to examine single nucleotide polymorphisms (SNPs) related to methotrexate disposition. Twelve (54.5%) subjects had a history of previous HDMTX toxicities including seven (31.8%) who previously required glucarpidase rescue and seven (31.8%) with an entry glomerular filtration rate < 80 ml/min/1.73 m 2 . 107/110 (97.2%) of methotrexate levels were drawn properly and 100% of algorithm dosing instructions were performed correctly at the bedside. Thirty-five (64.8%) of all cycles and 24 of 33 (72.7%) cycles that required a dose-adjustment had an end 24-h methotrexate level (Cp ss ) within our goal range of 65 ± 15 µM with only 3 (5.6%) resulting in Cp ss higher than goal. Grade 3/4 toxicities were rare; no patients developed > Grade 1 acute kidney injury. Conclusion: This algorithm is a simple, safe and effective method for individualizing HDMTX in pediatric patients with ALL. Clinicaltrials.gov Registry: NCT02076997.