Reduced dose folinic acid rescue after rapid high-dose methotrexate clearance is not associated with increased toxicity in a pediatric cohort

Methotrexate (MTX) is an antimetabolite that functions as a folic acid antagonist. MTX can be administered over a wide dose range, from 20 mg/m² orally per week in maintenance chemotherapy for acute lymphoblastic leukemia (ALL) to a high intravenous dose of 33,000 mg/m², when combined with folinic acid (FA) rescue [1]. High-dose methotrexate (HD-MTX) therapy given intravenously, defined as doses of 500 mg/m² or higher, can cause significant toxicity [2]. To prevent toxicity, intravenous hydration with urine alkalinization and pharmacokinetically guided FA rescue is used in conjunction with HD-MTX therapy [2].

Despite supportive care, HD-MTX can induce nephrotoxicity, resulting in the accumulation of toxic concentrations of MTX, delayed elimination, and an increased risk of severe toxic adverse events [3]. Prolonged MTX exposure can cause myelosuppression, mucositis, hepatotoxicity, and, in severe cases, multi-organ failure [2]. MTX can also cause acute, subacute, and long-term neurotoxicity, which has been reported in 3.8 to 15.5% of ALL patients receiving HD-MTX [4, 5]. Increased toxicity after HD-MTX has been attributed to the use of inadequate FA doses and to a delay in commencing FA rescue [6]. However, the minimum duration and dose of FA rescue needed to prevent excess toxicity after HD-MTX remains unclear [7]. Furthermore, the use of higher doses of FA after HD-MTX has been associated with an increased risk of leukemic relapse [8, 9].

In the present study, we describe a series of 44 patients with pediatric ALL who received a total of 350 courses of 5 g/m² HD-MTX followed by a pharmacokinetically guided FA rescue, including only one or two 15 mg/m² doses in the case of rapid MTX clearance, with respect to clinical toxicity and length of hospital stay due to the HD-MTX therapy.

Forty-four patients (27 males) were included in the study, with a mean age of 5.7 years (range 1.3 to 15.9 years) at diagnosis. The patients received a total of 352 HD-MTX courses, but a total of 350 courses were analyzed, as information on two courses of HD-MTX administered at a non-participating hospital was missing. The mean (± SD) S-MTX concentrations were 52.2 (± 26.9) μmol/L at 23 h, 1.1 (± 2.6) μmol/L at 36 h, 0.40 (± 0.75) μmol/L at 42 h, and 0.26 (± 0.51) μmol/L at 48 h. A concentration of ≥ 10 μmol/L at 23 h was reached in 347/350 (99.1%) courses. Rapid MTX clearance was observed in 184 (52.6%) of the 350 courses, and S-MTX was ≤ 0.2 μmol/L already at 42 h in 58 (16.6%) of the 350 courses.

This study is the first report of ALL patients receiving only one or two doses of FA after a 5 g/m² HD-MTX infusion. The results suggest that in the context of rapid MTX clearance, a shorter FA rescue may be sufficient. Although guidelines related to HD-MTX courses given as a part of NOPHO ALL-2000 therapy generally recommended at least three doses of FA regardless of MTX clearance speed, the local guidelines used in the hospitals participating in the present study provided no recommendation regarding the minimum number of FA doses. This resulted in administering only one or two FA doses in the case of rapid MTX clearance. Following 181 of 350 5 g/m² HD-MTX courses, only one or two doses of FA were administered due to rapid MTX clearance. In total, 41 out of 44 patients exhibited rapid MTX clearance following at least one course of HD-MTX. A shorter FA rescue following rapid MTX clearance was associated with a shorter hospital stay without increased toxicity. Severe toxicities were uncommon and did not differ between the courses with or without rapid MTX clearance.

All patients received extended hydration with 3000 mL/m²/24 h of fluids including 50 mmol/L NaHCO₃ over 12 h, according to the national practice, with extra NaHCO₃ given if urine pH was < 7. Urine alkalinization has been shown to be the most important factor influencing MTX clearance, and the active alkalinization may partially explain the rapid clearance observed in the present cohort [2]. Extended hydration is deemed important, although a Danish randomized study by Mikkelsen et al. found no difference between 4- and 12-h prehydration with urine alkalinization [14, 15]. However, the study reported markedly higher MTX concentrations after HD-MTX 5 g/m² at hours 23, 36, and 42 compared to the present study as well as a remarkably higher incidence of delayed elimination and nephrotoxicity: 47% and 18.5%, compared to 4.3% and 3.3% in the present study, respectively. The results of these two studies including patients treated with the same chemotherapy regimen are drastically different, implying a role of other unrecognized factors beyond extended hydration and urine alkalinization—such factors could include ethnic differences in genetic polymorphisms in the metabolic and cellular transport pathway of methotrexate, concomitant medication, or differences in other local practices. These differences call for a further study on the effect of extended prehydration in a population different than in Mikkelsen et al. [14].

FA is used to neutralize the effect of MTX on healthy cells after prolonged exposure. It has been reported to be especially effective against mucous membrane and gastrointestinal toxicity, hematological toxicity, and neurotoxicity. Elevated MTX-to-FA ratios at 42 h reportedly increase the risk for leukoencephalopathy [4]. Cohen suggested that neurotoxicity observed after HD-MTX is mainly due to inadequate FA rescue and that FA rescue should commence 24–36 h after MTX exposure [6]. Cohen also stated that higher doses of FA could explain the reduced incidence of neurotoxicity in some protocols [16]. Our study suggests that the pharmacokinetics of MTX following HD-MTX therapy may play a more important role regarding the subsequent toxic effects than the starting time or dosing of FA rescue. The finding of HD-MTX courses with rapid MTX clearance followed by only one or two doses of FA started 42 h from the initiation of HD-MTX infusion being associated with acceptable toxicity might also be explained by spontaneous recovery of folate metabolism, which has been observed by Frickel et al., who reported spontaneously increased levels of 5-methyltetrahydrofolate 42 h after a 24-h infusion of 5 g/m² HD-MTX [17].

There were no clinically significant differences between mucositis, neutropenic fever, nephrotoxicity, or hepatotoxicity between the rapid and non-rapid MTX clearance groups. As blood sampling varied largely between patients, we could not analyze hematological toxicity. No cases of acute or subacute neurotoxicity occurred in the study cohort. Brain MRI revealed no signs of leukoencephalopathy in 22 of the 44 patients undergoing a screening MRI at the end of ALL therapy. However, neuropsychological studies were not routinely performed on all of the patients, and thus, some milder neuropsychological changes may have gone unnoticed. The incidence of mucositis is underestimated, as only the cases of mucositis leading to hospitalization were recorded, and it is likely that a significant proportion of the patients hospitalized due to neutropenic fever also had some degree of mucositis, which had not been documented in the files. However, the rates of both mucositis and infection were lower than in the previously mentioned study by Mikkelsen et al., who reported mucositis in 9% of their cohort [14]. In our cohort, the overall incidence of nephrotoxicity of 4.6% falls in the lower part of the 2–12% range reported in the literature [1]. Three patients developed grade IV hepatotoxicity, with two following rapid MTX clearance. A transient increase in transaminases is common after HD-MTX and is usually not regarded as clinically significant [18]. During the NOPHO ALL-2000 protocol, prior to awareness of the potential interaction, PPIs were commonly used during HD-MTX infusions. Although numerous reports of PPIs increasing the risk of delayed clearance have been published, the use of PPIs did not differ between the clearance groups and did not appear to be associated with delayed elimination in the present study [19].

Glomerular hyperfiltration was common in the subgroup of patients with the result of glomerular filtration rate measurement available, which is in accordance with earlier reports on children with cancer [13]. However, hyperfiltration was equally common in both clearance groups in the present cohort.

There have been concerns that the carry-over effect of FA could decrease the antileukemic effect of HD-MTX therapy. According to Sterba and colleagues, high pretreatment folate concentrations were associated with lower MTX exposure when measured with homocysteine accumulation [20]. In an earlier Nordic study, higher FA doses were associated with a higher risk of relapse [8]. On the other hand, courses with rapid MTX clearance might confer less antileukemic effect due to shortened exposure to MTX. Indeed, in other protocols, shortening HD-MTX infusion times has been shown to increase the risk of relapse [21]. The importance of lowering and postponing the FA dose is still unclear, but it is tempting to assume that a shorter and lower FA rescue had a positive impact on the overall antileukemic effectiveness of the therapy in the present cohort, considering that the EFS and OS were excellent, although the sample size was small and the three early relapses were excluded from the study. Nevertheless, in light of the present study, a reduced FA rescue appears adequate in the context of rapid MTX clearance after HD-MTX infusion. Determining whether extended prehydration plays a role in the rapid MTX clearance and the low incidence of delayed elimination in the present population requires further study.

A pharmacokinetically guided FA rescue of one or two 15 mg/m² doses of FA following HD-MTX courses with rapid MTX clearance results in a shorter hospitalization duration without an increase in toxic effects.