Pharmacokinetic interaction between methotrexate and piperacillin/tazobactam resulting in prolonged toxic concentrations of methotrexate

Sir, Methotrexate is an antimetabolite commonly used to treat a number of malignancies including acute lymphoblastic leukaemia, high-grade lymphomas and germ cell tumours.1 In high doses methotrexate has been shown to result in significant myelosuppression, mucositis, hepatotoxicity and renal failure. Methotrexate is eliminated mainly unchanged in the urine by glomerular filtration and proximal tubular transport. We identified a 50-year-old female with newly diagnosed Burkitt's lymphoma who was treated with alternating monthly cycles of IVAC (ifosfamide, etoposide and high-dose cytarabine) and CODOX-M (cyclophosphamide, doxorubicin, vincristine and high-dose methotrexate). Cerebral spinal fluid disease was treated with alternating weekly intrathecal administration of methotrexate or cytarabine. The patient's course was complicated by a febrile neutropenic event on day +10 of her first cycle of chemotherapy and piperacillin/tazobactam was initiated as empirical therapy. Long-term therapy with piperacillin/tazobactam was necessary due to the formation of cavitary pneumonia secondary to Pseudomonas aeruginosa. A significant drug interaction was suspected during the patient's first cycle of CODOX-M. The serum methotrexate concentration failed to decrease below 0.05 μmol/L and appeared to plateau around 0.2 μmol/L (Figure 1). Cytotoxic methotrexate concentrations were sustained for 8 days and were only abated (i.e. fell below 0.05 μmol/L) by the discontinuation of piperacillin/tazobactam. During the patient's second cycle of CODOX-M, piperacillin/tazobactam was not administered and serum methotrexate levels declined appropriately. The patient's serum creatinine concentrations (average and standard deviation of 0.46 ± 0.07 mg/dL first cycle and 0.47 ± 0.03 mg/dL second cycle) and concurrent medications, aside from piperacillin/tazobactam, did not differ between the two cycles of CODOX-M. Methotrexate total body clearance fell to only 3% of normal in the presence of piperacillin/tazobactam (Table 1). Time course of methotrexate (MTX) elimination in the presence or absence of piperacillin/tazobactam (TZP) in the patient. Expected elimination refers to the literature-anticipated elimination curve for MTX.1 Comparative pharmacokinetic profile of methotrexate (MTX) alone and in the presence of piperacillin/tazobactam (TZP) Vc, volume of distribution in the central compartment; t1/2, half-life of elimination; CL, total body clearance. Comparative pharmacokinetic profile of methotrexate (MTX) alone and in the presence of piperacillin/tazobactam (TZP) Vc, volume of distribution in the central compartment; t1/2, half-life of elimination; CL, total body clearance. Following the patient's next scheduled dose of intrathecal methotrexate, the peak serum methotrexate concentration was measured to be 0.44 μmol/L, well within the cytotoxic range.1 Fortunately, the patient was no longer receiving piperacillin/tazobactam at that time, and serum methotrexate levels fell to below 0.05 μmol/L within 24 h. Many agents are known to prolong methotrexate elimination including probenecid, salicylates, non-steroidal anti-inflammatory drugs (NSAIDs) and weak organic acids.2,3 Penicillin is known to abolish tubular secretion of methotrexate in monkeys and reduce human organic anion transporter (hOAT)-mediated methotrexate elimination in an in vitro mouse model.4 Piperacillin was shown to reduce renal methotrexate clearance in rabbits as early as 1986.5 Only a single clinical case report of a piperacillin–methotrexate interaction has ever been documented.6 The weak organic acid and penicillin derivative, tazobactam, is also eliminated by glomerular filtration and tubular secretion; however, no evidence is available regarding its potential ability to impair methotrexate elimination. The general time course of elimination of methotrexate from the serum following a high-dose intravenous infusion has been reported to be biexponential with an alpha-phase t1/2 of 1.5–3.5 h and a beta-phase t1/2 of ∼8–15 h in patients with normal total body clearance.1 These values are virtually identical to our findings with the only difference being a triexponential decay and a gamma-phase t1/2 of over 180 h in the presence of piperacillin/tazobactam (Table 1). The persistently elevated serum methotrexate levels observed in this correspondence have substantial clinical implications. The relationship between methotrexate concentration and toxicity is well established in the literature and the 8 day plateau observed in this patient (∼0.20 μmol/L) is well above the concentration found to inhibit DNA synthesis in the bone marrow (0.01 μmol/L) and intestinal epithelium (0.005 μmol/L) of mice.1 Furthermore, it has been shown that 0.05 μmol/L of methotrexate for 72 h produces the same cytotoxic effects as 10 μmol/L for 12 h.1 Based on these observations, our patient was exposed to cytotoxic concentrations of methotrexate when it was administered with piperacillin/tazobactam. Fortunately, systemic toxicities were averted by the continued intervention with leucovorin. Serum methotrexate concentrations among patients receiving low-dose methotrexate or intrathecal methotrexate are not routinely measured. In the present case, we observed that intrathecally administered methotrexate resulted in measurable peak serum levels of 0.44 μmol/L, which is well above the cytotoxic threshold. Had our patient remained on piperacillin/tazobactam at that time, then sustained cytotoxic levels could have been present without the clinician's knowledge and without the protection of leucovorin. This is only the second case report characterizing the ability of piperacillin/tazobactam to reduce the renal clearance of methotrexate resulting in prolonged cytotoxic concentrations. Piperacillin/tazobactam and other drugs known to reduce methotrexate elimination should be avoided in patients receiving high-dose intravenous or intrathecal methotrexate. Close therapeutic monitoring of methotrexate levels and intervention with leucovorin is necessary for patients at risk of this drug interaction. This correspondence highlights a significant pharmacokinetic interaction between methotrexate and piperacillin/tazobactam. Given the widespread use of both methotrexate and piperacillin/tazobactam, clinicians must be aware of this interaction so that major toxicity may be averted. We have no conflicts to declare.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.