Methotrexate toxicity and glucarpidase: A call for dose optimization

High-dose methotrexate (HDMTX) is a folic acid antagonist that is used to treat acute lymphoblastic leukaemia, lymphoma and osteosarcoma in adult and paediatric patients. HDMTX cytotoxicity is concentration and time-dependent. Due to intracellular accumulation, prolonged exposure to relatively low MTX levels can lead to myelosuppression, mucositis, nephro-, hepato- and neurotoxicity and can ultimately result in multi-organ failure. HDMTX can be safely administered to patients with normal renal function provided that the necessary measures are taken to prevent side effects. These measures include urine alkalinization, hydration and pharmacokinetically guided leucovorin (folinic acid) rescue. Despite these measures, HDMTX-induced renal dysfunction continues to occur in approximately 2%–12% of patients with these diseases.1 As 80%–90% of MTX elimination is via renal clearance, a vicious cycle may occur in which delayed renal clearance increases nephrotoxicity and vice versa, ultimately resulting in prolonged exposure to high MTX blood concentrations. Glucarpidase (brand name Voraxaze®) is used to rapidly reduce high MTX blood concentrations observed in HDMTX treated patients with (potentially) impaired renal function.2 Rapid reduction of methotrexate concentration is considered to prevent further serious methotrexate (nephro)toxicity.3 Timely administration of glucarpidase (within 48–60 h from the start of HDMTX infusion) is important to break the cycle of reduced MTX clearance, prolonged half-life and increased nephrotoxicity.4 Recently, the appropriate dose of glucarpidase has become a topic of debate, as the current recommended dose does not appear to be well established. Doses other than the recommended dose have not been systematically investigated due to a lack of dose-finding studies.4 This raises the question of whether regulatory authorities should require more rigorous clinical pharmacological investigation of the optimal dose when approving antidotes. Glucarpidase is a recombinant carboxypeptidase that catalyses the conversion of methotrexate to the inactive metabolites 4-deoxy-4-amino-N10-methylpteroic acid (DAMPA) and glutamate in the blood. Glucarpidase is mainly found in plasma (volume of distribution in adults = 3.55 L), where the enzyme activity persists for hours (plasma concentration half-life: 9 h, activity half-life: 5.6 h).5 One unit of glucarpidase cleaves 1 μmol of methotrexate in 1 min at 37°C.6 The current-approved dose of glucarpidase is 50 units/kg of body weight (U/kg), irrespective of plasma methotrexate concentration. This clinically recommended dose is based on calculations derived from in vitro studies. According to the manufacturer, 50 U/kg glucarpidase theoretically reduces a plasma methotrexate concentration of 1000 μmol/L to 10 μmol/L in 1 min (calculation based on a blood volume of 4 L, body weight of 60 kg).5 Further translational dose-finding studies are lacking in both the FDA and EMA assessment reports,5, 6 and the 50 U/kg dose has been the most systematically studied dose in the clinical setting. Clinical studies have shown that glucarpidase at the recommended dose effectively reduces plasma methotrexate concentrations, both shortly after glucarpidase administration and over the following days.5, 7 A > 95% reduction within 15 min of glucarpidase administration was observed in 87% of the patients. Plasma methotrexate concentrations were reduced to ≤1 μmol/L 15 min after starting glucarpidase and during the remainder of follow-up, usually up to 8 days, in 59% of the patients (defined as a rapid and sustained clinically important reduction), with a reduction to <10 μmol/L that was observed in almost all patients. Recently, it has been shown that a rapid and sustained reduction is indeed likely to be clinically important, as glucarpidase treatment can improve renal recovery and reduce the incidence of neutropenia and transaminitis in patients with HDMTX toxicity.8 In this study, patients were treated with a standard dose of 50 U/kg (IQR 43–50 U/kg) glucarpidase, and only patients with doses ≥50 U/kg were included for specific sensitivity analyses. The clinical pharmacokinetic properties and the enzyme activity of glucarpidase suggest that the 50 U/kg dose leads to overtreatment. In an 80 kg male with a plasma volume of 2.8 L and a plasma methotrexate concentration of 100 μM, only 280 units of glucarpidase (3.5 U/kg, or 7% of the approved dose) would be needed to reduce plasma methotrexate to concentrations well below 1 μM in 1 min (Figure 1). Even with an extremely high plasma methotrexate concentration of 1000 μM, this patient would require only 35 U/kg (70% of the licensed dose) of glucarpidase. These calculations are very conservative, as the enzymatic nature of glucarpidase allows it to act repeatedly for several hours before it is metabolized and cleared. There is increasing in vivo evidence that glucarpidase is just as effective at lower doses. First, there are several clinical studies describing the use of lower doses of glucarpidase (<50 U/kg, as low as 8 U/kg) in more than 55 patients.5, 9-12 Even the studies used by the FDA and the EMA to approve glucarpidase occasionally used glucarpidase doses lower than their own recommended dose.5, 6 These studies did not report any reduction in glucarpidase efficacy with lower doses of glucarpidase. Effective reductions in methotrexate concentration have been observed with glucarpidase doses of 24.4–32.4 U/kg10), or with a fixed dose of 1000 U (1 vial) of glucarpidase in adults (6.4–14.5 U/kg13), although preglucarpidase methotrexate concentrations differed between studies (up to 182.4 μM vs. 22.6 μM). In addition, a study in a paediatric population found no association between glucarpidase dose and efficacy in paediatric cancer patients receiving doses as low as 13 U/kg of glucarpidase.14 Moreover, a study in obese patients in which dosing was based on ideal body weight showed that a maximum dose of 4000 U (23–40 U/kg) was sufficient to achieve a rapid decrease in methotrexate concentration.15 The efficacy of lower doses of glucarpidase has also been suggested by pharmacokinetic/pharmacodynamic (PK/PD) modelling. In a modified Michaelis–Menten PK/PD model, Kimura et al. showed that an initial preglucarpidase methotrexate plasma concentration of 100 μM was reduced to less than 0.1 μM 70 h after administration of both 20 U/kg and 50 U/kg glucarpidase.16 These studies all show effective reduction of plasma methotrexate concentrations with glucarpidase doses as low as 6.4 U/kg. Weight-based dosing of glucarpidase, as is currently standard practice, is not, in our opinion, the most logical dosing strategy. Given its volume of distribution, glucarpidase does not diffuse out of the blood compartment; it only converts unbound methotrexate in plasma and does not affect intracellular methotrexate, for example, in red blood cells. Body weight varies by a factor of approximately 2 between patients (40 vs. 80 kg), while the plasma premethotrexate concentrations can vary by a factor of 10 to 1000 (1–1000 μM). In this sense, it is more rational to base the glucarpidase dose on the total amount of methotrexate in the blood plasma, similar to digoxin-specific antibodies in the treatment of digoxin toxicity.17 There are several potential benefits of administering lower doses of glucarpidase. First, a lower, more appropriate dose of glucarpidase may improve the efficacy of leucovorin rescue therapy, which directly counteracts cellular toxicity. Since glucarpidase does not hydrolyze intracellular MTX, leucovorin rescue therapy remains important even after glucarpidase administration. However, glucarpidase also converts leucovorin to an inactive form.18 As leucovorin rescue usually starts 2 h after glucarpidase administration, a lower initial dose of glucarpidase will result in less residual glucarpidase activity at the start of leucovorin administration, less leucovorin inactivation and improved leucovorin rescue therapy. Secondly, lower doses of glucarpidase may result in less (neutralizing) antibody formation against glucarpidase, as is suggested by the study by Schaff et al.19 Neutralizing antibodies reduce the efficacy of glucarpidase, resulting in increased methotrexate rebound. Thirdly, although glucarpidase is generally well tolerated, lower doses may result in fewer side effects, such as hypersensitivity reactions, paresthesia, nausea and vomiting. Finally, by reducing the dose, for example, by using a fixed dose of 1000 or 2000 U (1 or 2 vials), logistics, drug preparation and administration may become more efficient. Here, a twofold to fourfold reduction in dose would be achieved, reducing the environmental impact and costs to society. For example, in terms of drug cost savings, for an 80 kg adult, this would mean a cost reduction of approximately €60 000–€90 000 in the Netherlands (cost per vial of 1000 U glucarpidase: €29 950, d.d. 3 February 202520) per glucarpidase treatment. In conclusion, the currently used dose of glucarpidase is effective in rapidly reducing plasma methotrexate concentrations, but leads to unnecessary overtreatment. Similar reductions in plasma methotrexate concentrations are seen with lower doses. Ideally, a dose is given based on the actual preglucarpidase methotrexate concentration rather than the patient's body weight. A pharmacological rationale for the dose used during the clinical development of glucarpidase should have been rigorously evaluated in early phase clinical trials, which should be a future requirement for drug approval by agencies such as the FDA and EMA. Further research is needed to determine the optimal glucarpidase dosing regimen for recovery from methotrexate-induced toxicity without the risk of overdose. All authors contributed to conceptualisation and review/approval of the manuscript. Arjen Koppen prepared the original draft. The authors declare no conflicts of interest. Data sharing is not applicable to this article, no new data were created or analysed.