Statistical analysis of relation between plasma methotrexate concentration and toxicity in high-dose methotrexate therapy of childhood nonHodgkin lymphoma

Background Plasma monitoring of Methotrexate (MTX) levels is a standard approach to predict MTX‐related toxicities in a high‐dose (HD) MTX monotherapy for childhood acute lymphoblastic leukemia. However, it is uncertain whether plasma MTX levels can predict MTX‐related toxicity in the HDMTX plus additional chemotherapy for childhood B‐cell nonHodgkin lymphoma (B‐NHL). Procedures To statistically analyze the relationship between MTX pharmacokinetic parameters and MTX‐related toxicities, we collected data from patients with delayed MTX elimination (≥1 µM at 48 hr and/or ≥0.5 µM at 72 hr) in the Japanese Pediatric Leukemia/Lymphoma Study Group (JPLSG) BNHL 03 study. Blood MTX levels were measured at 24, 48, and 72 hr after 3 or 5 g/m2 HD‐MTX administration for 24 hr. Results Three hundred and four patients received 2–4 courses of the HDMTX plus additional chemotherapy, and delayed MTX elimination was observed in 165 courses of 127 patients. In those, nephrotoxicity was significantly correlated with plasma MTX levels for each patient (P = 0.03), and also for each course (P = 0.009), but no other toxicities were correlated. Another analysis according to HDMTX courses showed no significant correlation between the first high plasma MTX levels and subsequent MTX levels in later course. It also showed that incidence of liver and gastrointestinal toxicities was most frequent in the first HDMTX course, and then sharply decreased in later courses (P