Background

Intravenous high-dose methotrexate (HD-MTX) is commonly used in the treatment of hematological malignancies, particularly in high-grade lymphomas [1]. MTX-related toxicities are common, leading to organ dysfunction that can be very severe, and rarely to death [1]. Acute kidney injury (AKI) is frequently reported, affecting up to 35% of adult patients, mostly in relation with intratubular crystal formation or endothelial injuries [1, 2].

This drug requires a close monitoring and management of MTX-related toxicities relies mostly on preventing measures [1, 3]. AKI is frequent in critically ill patients with newly diagnosed high-grade hematological malignancies [4] and is associated with a high level of frailty. Whether ICU patients may be considered eligible for HD-MTX and risk–benefit ratio in this setting has never been assessed.

This primary objective of this study was to assess outcome in critically ill patients requiring HD-MTX infusion. Secondary objectives were to describe toxicities and risk factors of poor outcome in this setting.

Methods

Patients and data collection

We retrospectively reviewed the medical charts of all consecutive adult patients admitted to the intensive care unit of one university hospital from January, 1st, 2002 to December 31th, 2018, and who received HD-MTX for hematological malignancy or solid tumor. There were no exclusion criteria.

HD-MTX was defined by a single intravenous infusion greater than 500 mg/m2. The different stages of toxicity were defined according to the CTCAE [5]. MTX complete elimination was considered when MTX concentration was lower than 0.1 µmol/L.

This study was approved by a local ethic committee (Société de Réanimation de Langue Française, CE SRLF 19-01). According to French law, need for informed consent was waived.

Statistical analysis

Results are described as medians and interquartile ranges (IQR) for quantitative variables and numbers and percentages for qualitative variables. We used a non-parametric Wilcoxon tests and Fisher exact tests for baseline univariate comparisons between two groups.

Cox regression model were performed to identify factors associated with hospital mortality. Variable selection was performed on a stepwise fashion, backward conditional model according to P value with entry P value of 0.2 and critical removal P value of 0.1. Proportional hazard assumption was checked in the final model.

All tests were two-sided, and P values less than 0.05 were considered significant. Analyses were done using R software version 4.3.4 (R Project for Statistical Computing, Wien, Austria) and with ‘Survival’ packages.

Results

33 patients (24 men, 9 women) were included with a median age of 48 years [34–63]. All the patients had aggressive hematological malignancies and most of them (n = 31/33) had not received any antitumor treatment. B-cell lymphoma had been diagnosed in 31 patients (Burkitt [n = 14]; diffuse large B-cell lymphoma with CNS involvement [n = 9]; primary CNS lymphoma [n = 5]; primary effusion lymphoma [n = 2]; intravascular lymphoma [n = 1]) and T-cell lymphoma in two patients. Twelve (36%) had HIV infection, three had hypertension and one patient was diabetic (Table 1).

Table 1 Characteristics of patients at study inclusion

Patients were mainly admitted to the ICU for coma (n = 14; 42%) or acute kidney injury (n = 8; 24%). Five (15%) patients presented respiratory failure and only one patient had shock. Fifteen patients had a Glasgow Coma Scale of 12 or below and all except one presented with neurologic involvement related to lymphoma. SOFA score was 4 [1–5] at admission and 2 [1–5] the day of MTX infusion. All the patients except one presented a normal renal function at baseline (median serum creatinine = 55 µmol/L [41–74]), 12 experienced AKI episode in the last 3 months, including 8 requiring renal replacement therapy (RRT) (Table 1).

MTX was administered at a median dose of 3.4 g/m2 [2.6–7.4] and the median delay between ICU admission and MTX infusion was 4 days [2–7]. Twenty patients (61%) received concomitant intrathecal MTX. Other most commonly used chemotherapy drugs were cyclophosphamide (n = 23), doxorubicin (n = 23) and vincristine (n = 21). Median body mass index (BMI) was 24.9 [21.4–27.1] and median albumin level was 33 g/L [26–37]. Median creatinine level was 55 µmol/L [41–74] on the day of MTX administration. All patients received parenteral hyperhydration and alkalinization. Only 6 patients failed to achieve urine pH above 7.5 at least once within the first 24 h following MTX infusion. Folinic acid rescue was started 24 h after MTX infusion for all patients except one. The median time required until a complete elimination of MTX was 4 days [4–5]. Fourteen patients had concomitant medication interacting with MTX, mostly piperacillin–tazobactam (n = 8), proton-pump inhibitors (n = 9) or levetiracetam (n = 4). Seven patients presented serous effusions that required fluid removal (pleural effusions, n = 6 and ascites, n = 1).

More than 80% of patients (n = 27) experienced at least one MTX-related toxicity (Table 2). The most frequent MTX-related complications were mucositis (n = 21, 64%; median CTCAE grade 3 [2–4]), diarrhea (n = 14, 44%; median CTCAE grade 2 [2–3]) or liver tests disturbance (n = 15, 45%; median CTCAE grade 3 [2–4]). Following MTX infusion, the majority of patients developed neutropenia (n = 26) and acquired bacterial infections (n = 17, 51%). During ICU stay, one-third of patients (n = 11) experienced acute kidney injury (KDIGO stage 2.5 [2–3]) and median onset was reached 3.5 days [2–5] after MTX infusion. Eight patients also received concomitant nephrotoxic agents including contrast media (n = 3) and aminoglycosides (n = 3). Renal toxicity and MTX overdosage lead to carboxypeptidase G2 administration in 2 patients and need for RRT initiation in three.

Table 2 MTX-related toxicities and outcome

Overall, 19 patients (57%) required mechanical ventilation, within a median time of 5 [1–12] days prior to MTX administration, and 10 (30%) vasopressors. Median length of ICU stay was 11 days [6–24]. Overall, ICU and hospital mortality were 18% (n = 6) and 30% (n = 10), respectively. Eighteen patients (55%) were alive 6 months after ICU discharge among whom, 15 (83%) had a complete and sustained hematological remission.

In univariate analysis, mortality was associated with older age (median age 63.5 years [51.25–68.75] vs 38 years [31–51.5, p = 0.013]), lower albumin level (27.5 g/L [24–31.75] vs 36 g/L [31–38], p = 0.045), and higher severity as assessed by SOFA score (6 [4.25–9] vs 2 [1–5], p = 0.022).

After adjustment for patients’ severity, MTX concentration 24 h after administration was independently associated with hospital mortality (HR if concentration above 4.6 μmol/L 6.7; 95% CI 1.6–27.3) (Table 3) (Fig. 1). In non-survivors, creatinine levels were significantly higher the day after MTX administration (p = 0.0017) and during the first week (p = 0.026) (Fig. 2).

Table 3 Variables associated with hospital mortality after adjustment
Fig. 1
figure 1

Adjusted influence of MTX dosage at H24

Fig. 2
figure 2

Relationship between creatinine and hospital mortality within the first week after MTX infusion

Discussion

To our knowledge, this is the first study assessing benefits and risk of HD-MTX in critically ill patients. This study underlines the high rate of risk factors for HD-MTX toxicity and the high rate of MTX-related toxicities. Our results also underline that 6-month survival may be obtained in 55% of the patients and that complete remission may be obtained in 83% of them.

In the literature, classically 2 to 12% of non-ICU patients are reported to develop renal failure following HD-MTX. In fact up to 35% of patients experienced AKI, with a large heterogeneity according to the studied population, HD-MTX protocols and AKI criteria [6]. In critically ill patients, AKI incidence also varies widely from 22 to 67%, discrepancies mainly relying on the definition applied [7]. In our study, we found a high rate of AKI as one in three patients experienced renal failure. This is in line with a previous study in which the authors showed that two-thirds of critically ill patients with newly diagnosed aggressive hematological malignancies developed AKI [4]. Our results highlight that MTX-induced renal toxicity is very frequent in ICU patients.

Survival of patients with hematological malignancies has improved over the past decades and an increasing number of patients may need ICU admission [8]. In previous studies in cancer patients receiving chemotherapy in the intensive care unit, hospital mortality is reported around 40% [9]. Our study, concurrently suggests the feasibility of HD-MTX in this setting, demonstrating that despite the high toxicity rate, a 6-month survival rate of 55% may be achieved, the majority of survivors achieving complete remission.

Dose–toxicity relationship of MTX has been descried previously. The most commonly used threshold is a concentration greater than 10 µmol/L 24 h after MTX infusion or greater than 1 µmol/L at H48 [3]. Evans et al. [10] previously demonstrated that values above 10 µmol/L 24 h after the start of MTX infusion were associated with an increased risk of toxicity. As no published data exist in ICU patients, optimal early MTX concentrations predicting the outcome remain unknown.

This study suffers however several limitations. First, due to its retrospective design, exhaustivity of data was limited. Thus, exact assessment of optimal MTX concentration predicting poor outcome could not be assessed. Second, the small sample size led to limited statistical power, negative findings needing to be interpreted cautiously. Moreover, patients deemed eligible to HD-MTX infusion were likely to be selected according to performance status and clinical severity. Despite these limits, our study demonstrates feasibility of HD-MTX with meaningful chances of long-term survival and complete remission.

Conclusion

This study demonstrates feasibility of HD-MTX in a selected group of critically ill cancer patients. Although the toxicity rate was high, long-term survival was achieved in more than half of the patients and complete remission was achieved in most of these later. Additional studies are needed to allow better identification of patients at high risk of toxicity.