Clinical Pharmacokinetics

, Volume 42, Issue 2, pp 139–151

Pharmacokinetics and Pharmacodynamics of Methotrexate in Non-Neoplastic Diseases


    • Department of PharmacologyCharles University
  • Jaroslav Chládek
    • Department of PharmacologyCharles University
  • Jiřina Martínková
    • Department of PharmacologyCharles University
Review Article Special Populations

DOI: 10.2165/00003088-200342020-00003

Cite this article as:
Grim, J., Chládek, J. & Martínková, J. Clin Pharmacokinet (2003) 42: 139. doi:10.2165/00003088-200342020-00003


Low dose pulse methotrexate (LDMTX) therapy has become effective in the treatment of autoimmune and lymphoproliferative diseases. The pharmacokinetics of LDMTX is individually highly variable, resulting in a different systemic exposure to the drug and a variable therapeutic/toxic effect in patients. The improvements and exacerbations of disease activity in relation to the introductions and discontinuations of LDMTX therapy suggest the possible immunosuppresive and anti-inflammatory properties of the drug. Because of a strong correlation between the drug pharmacokinetics and the therapeutic outcomes (pharmacodynamics), it seems to be possible to individualise the LDMTX therapy according to the results of pharmacokinetic/pharmacodynamic analysis. In the case of psoriasis, pharmacokinetic/pharmacodynamic analysis in our local study revealed a highly significant inverse relationship between PASI (expressed as a percent of the initial value) and a steady-state AUCMTX (area under the curve of methotrexate plasma concentrations; r8 = −0.65, p < 0.001). The considerable inter-individual variability and low intra-individual variability in MTX pharmacokinetics, supports a role for therapeutic monitoring and dose individualisation at the start of pharmacotherapy. The results of this study suggest that a steady-state AUCMTX value of 700 nmol · h/L and higher are associated with a significantly better success rate of antipsoriatic therapy than lower values. The preliminary results in our follow-up study suggest the statistically higher incidence of unwanted effects depending on maximum plasma concentration of the drug. Moreover, statistically significant correlation was found between the toxic effects and exposure to the drug regarding methotrexate plasma concentrations and intracellular storage in erythrocytes. However, the data are still in the process of being completed and are not yet published.

Methotrexate is an antifolate, and has been used as high-dose pulse therapy (HDMTX) for the treatment of malignancies since 1947. The favourable anti-inflammatory effect of low dose pulse methotrexate (LDMTX), given as 7.5–30mg (approximately 0.3 mg/kg) once weekly orally, subcutaneously or intramuscularly, was first reported in the 1950s in patients with psoriasis and psoriatic arthritis. The drug has been commonly used in the therapy of recalcitrant psoriasis since the 1960s. Its use in the treatment of rheumatoid arthritis began during the 1980s.[1]

At present, methotrexate is one of the most frequently used of the disease-modifying antirheumatic drugs (DMARDs), also called slow-acting or symptom-modifying drugs. In the treatment of rheumatoid arthritis, methotrexate has proved to be more effective and less toxic than auranofin and azathioprine, and as effective as but less toxic than sulfasalazine.[1,2] The continuation rate of LDMTX therapy in patients with rheumatoid arthritis has been reported as 70% after 1 year of therapy, 54% after 3 years and 50% after 6 years.[3] These percentages compare favourably with the overall probability of less than 20% for three other DMARDs after 5 years: 19% for sulfasalazine, 17% for penicillamine and 8% for parenteral gold.[4]

LDMTX therapy was also shown in placebo-controlled randomised trials to be efficacious in children with juvenile rheumatoid arthritis (especially the polyarticular form) and systemic onset juvenile rheumatoid arthritis (Still’s disease). Generally, children tolerate higher doses of the drug than adults, up to 0.6 mg/kg. Long-term LDMTX therapy does not induce osteopenia in children, which has been described after HDMTX.[5] Parenteral LDMTX therapy also has a beneficial effect on numerous other inflammatory disorders, including corticosteroid-dependent chronic active Crohn’s disease,[68] antimalarial-resistant lupus arthritis, cutaneous lupus erythematosus,[9] systemic lupus erythematosus,[10] polymyositis, polymyalgia rheumatica, Reiter’s syndrome, sarcoidosis, primary biliary cirrhosis, primary sclerosing cholangitis, scleroderma, graft-versus-host disease and organ allograft rejection.[7,11]

From many clinical studies it is evident that LDMTX treatment is associated with great interindividual variability in the therapeutic response. Regardless of the different immunological characteristics of patients, a significant relationship between pharmacokinetics and pharmacodynamics (i.e. efficacy and toxicity) has been reported.[1215]

1. Pharmacokinetics

1.1 Absorption

1.1.1 Oral Administration

Methotrexate is a weak dicarboxylic organic acid with a molecular weight of 454 daltons. The molecule is negatively charged at neutral pH (pKa1 = 4.84, pKa2 = 5.51), resulting in limited lipid solubility. After oral administration, active absorption of the drug occurs in the proximal jejunum. The process is capacity-limited and decreases nonproportionally with increased oral doses.[1618] Earlier studies indicated that methotrexate absorption was rapid and complete after oral doses of less than 30 mg/m2.[16,19] More recent investigations with larger numbers of patients demonstrated that the rate and extent of absorption are highly variable between patients, and that the absolute bioavailability may be less than 50% for doses as low as 10–15 mg/m2.[15,20,21] The mean absolute bioavailability is about 70–80% and a large interindividual variation from 30–90% has been observed.[15,20,2224] Conversely, only a moderate intra-individual variability in LDMTX pharmacokinetics was noticed during intermediate-term (13 weeks[15]) and long-term (2 years[21]) treatment in patients with psoriasis and rheumatoid arthritis receiving a single weekly dose of methotrexate 15mg.

Under fasting conditions, maximum plasma concentrations of methotrexate (Cmax) range between 0.3 and 1.6 µmol/L, and occur at a tmax of 0.75–2 hours after administration.[15,21,22,24] Food did not significantly influence the bioavailability of methotrexate, but slightly reduced Cmax and prolonged tmax by about 0.4–0.7 hours as a result of delayed gastric emptying.[24]

1.1.2 Parenteral Administration

LDMTX is given parenterally to ensure effective compliance and, presumably, uniform bioavailability.[25] The drug is absorbed more rapidly and reaches higher serum concentrations after intramuscular or subcutaneous administration compared with the oral route.[26,27] Nevertheless, the mean absolute bioavailability is very similar,[28,29] suggesting that the routes of LDMTX administration are interchangeable.[30,31]

LDMTX may also be injected intra-articularly. The mean synovial methotrexate concentration exceeds the serum concentration by a minimum of 10-fold throughout the whole 24-hour post-dose period and ensures the therapeutic effect.[32] The topical application of LDMTX in cream results in drug absorption and accumulation in keratinocytes of psoriatic plaques, but without any histological change.[33]

1.2 Distribution

The volume of distribution of methotrexate is 0.87–1.43 L/kg, which corresponds to the intracellular distribution of the drug.[26,34] In blood, 30–70% of methotrexate is bound to proteins, almost exclusively to albumin.[23,26,27,3436] Edno et al. demonstrated a significantly increased drug plasma concentration for 8 hours following methotrexate administration,[34] reflecting the possible enterohepatic cycling of the drug. The concentrations of methotrexate in the synovial fluid are approximately equal to plasma concentrations at 4 and 24 hours after oral or intramuscular administration.[23]

With regard to the intracellular mechanism of action, it is believed that the most important process is transport of methotrexate into cells and its accumulation within cells in the form of polyglutamates. Transport of methotrexate occurs both by passive transmembrane diffusion and by a carrier-mediated active transport system that methotrexate shares with folates.[18,25] A folate surface receptor responsible for the intracellular transport of both reduced folates and methotrexate has been well described in various in vitro studies.[37] Once inside the cell, up to six glutamate residues may be progressively added to the drug molecule by the folyl-polyglutamate synthetase enzyme. Polyglutamyl derivatives of methotrexate cannot be transported extracellularly unless they are hydrolysed back to the monoglutamate.[16,18] Intracellular accumulation of methotrexate polyglutamates allows drug administration once weekly as a bolus or divided into three equal subdoses.[38] It was found that the pool of folate polyglutamates in liver cells and erythrocytes is gradually replaced by methotrexate polyglutamates during long-term LDMTX therapy.[39,40]

The pharmacokinetics of methotrexate in erythrocytes have been studied extensively.[39] After a single dose, methotrexate concentrations in plasma and erythrocytes change simultaneously as a consequence of a rapid equilibrium between these compartments. The concentrations of methotrexate both in plasma and erythrocytes usually fall below 10 nmol/L within 24 hours after LDMTX administration. At 3–4 days later, the drug reappears in erythrocytes despite its negligible plasma concentration.[39] Methotrexate polyglutamates in erythrocytes accumulate until a steady-state level is reached after 4–6 weeks of intermittent administration.[15,41] This is a much shorter time than that required for medullar maturation of erythroblasts into erythrocytes (14–18 weeks). The probable explanation is that methotrexate polyglutamates are synthesised mainly in the circulating erythrocytes.[41] The steady-state erythrocyte methotrexate concentration is also highly variable among patients undergoing long-term LDMTX therapy. A range of 10–170 nmol/L erythrocytes was observed following the administration of 7.5–15mg methotrexate once a week.[15,40] No correlation was found between the total cumulative dose of methotrexate and erythrocyte methotrexate concentration.[42]

Although no close relationship was found between folate status and intracellular accumulation of methotrexate, the highest methotrexate concentrations were found in erythrocytes with the lowest folate concentration.[40] Moreover, erythrocyte methotrexate concentration seems to be indicative of hepatic changes that occur during LDMTX therapy. Significantly higher erythrocyte methotrexate concentrations were found in patients with progressive hepatic changes than in patients with no progression. Nevertheless, a critical erythrocyte concentration was not established because of its very large interindividual variability.

Peripheral blood T-lymphocytes also intensively convert methotrexate to polyglutamyl derivatives (tetra- and penta-glutamates). Similarly, methotrexate is highly accumulated in fibroblasts, myeloid precursors in bone marrow and keratinocytes.[1,20,33] Relatively high and equal methotrexate concentrations were found in the synovial membrane, cortical bone and trabecular bone.[43] The intracellular accumulation results in drug-induced apoptosis of T-lymphocytes.[44] On the contrary, the activity of intracellular hydrolases in intestinal epithelium cells is very high, and therefore the accumulation of methotrexate in small intestine mucosa is not significant.[18]

1.3 Elimination

Elimination of methotrexate from plasma was shown to be biphasic or triphasic (dependent on the length of sample collection period) with a mean terminal biological half-life (t½β) of 6–15 hours.[15,18,2224,26,27,29,34] Thus, accumulation of methotrexate in plasma cannot occur after intermittent administration once a week. Extensive sampling of plasma over 1 week after methotrexate administration in nine patients with rheumatoid arthritis allowed estimation of a t½β of 55 hours, reflecting slow release of methotrexate from its intracellular forms.[29] The longer the sampling, the longer the reported t½β of the drug, probably due to intracellular methotrexate storage, polyglutamylation and slow release back to plasma.[16,18] Accumulation of methotrexate in pleural effusion and ascitic fluid is the reason for its slowed elimination in cancer patients receiving HDMTX therapy, but seems to be of no importance after LDMTX.[18]

1.3.1 Metabolism

Three metabolic pathways of methotrexate have been described in humans. First, the drug is metabolised by intestinal bacteria to 4-amino-deoxy-N10-methylpteroic acid. The metabolite usually accounts for less than 5% of the administered dose, and is rarely detectable in human plasma and urine.[18]

Secondly, in the liver, methotrexate is converted to 7-hydroxy-methotrexate. 7-Hydroxy-methotrexate is less water soluble than methotrexate, and it may therefore contribute to acute nephrotoxicity because of its precipitation in acidic urine. The metabolite is a 10-fold less potent inhibitor of dihydrofolate reductase (DHFR), one of the intracellular target enzymes for methotrexate.[16,18,45] The hepatic first-pass effect of methotrexate is low (about 10%), as is its metabolic clearance to 7-hydroxy-methotrexate: 5–7% of the dose was recovered as 7-hydroxy-methotrexate in urine after a broad range of methotrexate doses.[3,15,29] However, due to its slower rate of urinary excretion, plasma concentrations of 7-hydroxy-methotrexate usually exceed those of methotrexate within 8–10 hours after drug administration.[15,29,35] Despite its extensive binding to serum albumin (91–93%), 7-hydroxy-methotrexate does not alter the protein binding of methotrexate.[18,23,26,27,3436] Both compounds compete for the same membrane carriers, intracellular transporters and, subsequently, for folyl-polyglutamate synthetase.[46]

Thirdly, the intracellular conversion of methotrexate to polyglutamates represents the most important metabolic pathway regarding efficacy.[16,18] Polyglutamyl derivatives of methotrexate also exhibit more efficient inhibitory properties towards intracellular metabolism of pyrimidines and purines than does the parent drug.[18,25] Alteration of the intracellular folate cycle results in intracellular accumulation of homocysteine and depletion of polyamines such as spermine and spermidine. Polyamines have proinflammatory properties.[47]

1.3.2 Excretion

Renal excretion constitutes the major elimination route for methotrexate. The drug is filtered in renal glomeruli and, additionally, undergoes bidirectional transport across the renal tubules, i.e. active secretion, utilising the general transport mechanism for organic acids, and active reabsorption unaffected by acidic compounds from the distal tubule. At serum concentrations from 0.1–0.4 µmol/L, tubular secretion prevails over reabsorption, which reaches saturation.[16,41] Accordingly, renal clearance (CLR) of LDMTX usually exceeds creatinine clearance (CLCR) by about 2–28%.[26,48] At methotrexate plasma concentrations of 0.6–1 µmol/L, CLR equals CLCR (i.e. 80–120 ml/min), reflecting the saturation of methotrexate active tubular secretion. There is a considerable interindividual variation in the saturation point of both secretion and reabsorption in tubules. Both kinetic processes can occasionally be saturated even at low methotrexate plasma concentrations within the range of 0.1–1 µmol/L. Thus, nonlinear elimination may result following the administration of 7.5–30mg of methotrexate and contribute to the interindividual variability in methotrexate concentrations.[40]

After 6 months of LDMTX therapy, CLR of methotrexate decreased by a mean of 23.8 ml/min and CLCR by 8.6 ml/min.[48] A decrease in glomerular filtration rate has also been reported in rheumatoid arthritis patients taking LDMTX, usually over a period of 2–4 weeks.[29] This important effect of methotrexate has also been observed in HDMTX.[48] It could be explained by an increase in plasma adenosine concentration in extracellular fluid and by subsequent activation of A1 receptors in renal parenchyma, diminishing renal blood flow and salt and water excretion.[47]

In addition, a variable amount of methotrexate is eliminated by active biliary excretion, responsible for 10–30% of methotrexate clearance.[26,36,41,48] However, only about 1–2% of the drug is excreted in faeces, suggesting extensive enterohepatic circulation of methotrexate.[40,45] Biliary excretion can become more important in patients with renal insufficiency.[17,20] Enterohepatic cycling can be interrupted by cholestyramine or charcoal, which can be administered to attenuate potentially life-threatening toxicity of LDMTX in patients with renal insufficiency or after methotrexate poisoning.[49]

1.4 Therapeutic Drug Monitoring

Recently, to provide therapeutic drug monitoring effectively and more practically, limited sampling methods to estimate methotrexate pharmacokinetics using a Bayesian approach and population data modelling programs have been implemented.[5052]

2. Pharmacodynamics

Methotrexate is an analogue of folic acid that was originally designed to inhibit the activity of the enzyme DHFR. This enzyme converts dihydrofolates to tetrahydrofolates, which are involved in single carbon atom transfers in crucial intracellular metabolic pathways such as de novo synthesis of purines, pyrimidines and polyamines and transmethylation of phospholipids and proteins. In oncology, the rationale for the use of HDMTX is that malignant cells become starved of the purine and pyrimidine precursors required for DNA and RNA synthesis, proliferation and cell division. As a result of their inability to synthesise DNA and RNA, the number of malignant cells rapidly falls under such therapeutic conditions.[53]

LDMTX has immunosuppressive and anti-inflammatory properties. Concerning immunosuppressive activity, the assumption for the introduction of LDMTX was that the drug would inhibit proliferation of lymphocytes (notably the CD3 and CD4 subtypes) and other immunocompetent cells (e.g. monocytes-macrophages and polymorphonuclear neutrophils). At concentrations of 0.1–10 µmol/L, methotrexate induces apoptosis of in vitro activated T cells from human peripheral blood.[54] The ability to undergo apoptosis may reflect the capacity of lymphocytes to convert methotrexate to methotrexate polyglutamates, containing four or five glutamyl groups, which were reported to be retained up to 24 hours in breast cancer cells.[6]

According to several lines of evidence, LDMTX does not seem to act only as a cytotoxic agent against immunocompetent cells.[55] In vitro, modulation of the cytokine network by LDMTX increased T helper 2 cytokines, e.g. interleukin-4 (IL-4) and interleukin-10 (IL-10), and decreased T helper 1 cytokines, e.g. interferon-γ (IFNγ) and interleukin-2 (IL-2). This observation could explain the methotrexate-induced anti-inflammatory and immunoregulatory actions in vivo. Once intracellular, methotrexate-polyglutamates bind competitively, and with higher affinity than dihydrofolate and methotrexate, to several enzymes and inhibit their function: DHFR, thymidylate synthetase (TMS) and 5-amino-imidazole-4-carboxamide ribosyl-5-phosphate formyltransferase (AICAR-formyltransferase).[1]

Inhibition of AICAR-formyltransferase leads to intracellular accumulation of 5-amino-imidazole-4-carboxamide ribosyl-5-phosphate (AICAR), even if the enzymatic block is only partial.[21,56] High concentrations of AICAR lead to enhanced release of adenosine into the blood.[57] Additionally, adenosine is synthesised in plasma under conditions of LDMTX therapy.[58] This mediator activates A2a, A2b and A3 extracellular receptors on monocytes-macrophages,[47,59] inhibiting production of tumour necrosis factor α (TNFα), IL-6 and IL-8, promoting transcription of mRNA for an IL-1 receptor antagonist[60,61] and increasing secretion of the potent anti-inflammatory cytokine IL-10.[62] It was also reported that activation of adenosine receptors on human endothelial cells inhibits their production of IL-6 and IL-8 and diminishes expression of E-selectin on the cell surface.[47,60] These observations indicate that adenosine plays a major role in the anti-inflammatory response.

Recent data confirm that enhanced adenosine release may be also responsible for some of the toxicity of LDMTX therapy. Adenosine release in the CNS and its activation of A1 receptors in the brain can be responsible for induction of fatigue and lethargy.[63] A1 receptors are also present in endothelial cells and their activation provokes vasodilatation. This could explain the headache that appears in many patients a few hours after intake of LDMTX and the decrease of CLCR (section 1.3.2).[55]

3. Remission Induction

The maximum effect of LDMTX therapy of rheumatoid arthritis and psoriasis is usually achieved from 4–6 months after the beginning of intermittent drug administration.[7,57,64] To induce and maintain remission of the disease, it is necessary to start intermittent LDMTX therapy at an appropriate dose and mode of administration and to individualise maintenance therapy to preserve the effect under conditions of minimal toxicity.

Increasing information on the severe, progressive and debilitating nature of rheumatoid arthritis and its negative influence on lifespan has prompted the development of improved treatment courses aimed at slowing disease progression. A major change in approach has taken place in that rheumatologists are starting treatment with DMARDs (LDMTX, cyclosporin, hydroxychloroquine, gold salts, penicillamine, sulfasalazine, azathioprine, etanercept and infliximab) as early as possible, before joint damage and loss of function can occur. The American College of Rheumatology recommends that DMARD therapy should be initiated no later than 3 months after diagnosis if a patient has ongoing joint pain, morning stiffness, fatigue, synovitis or persistent elevation of erythrocyte sedimentation rate and C-reactive protein.[65] LDMTX is currently considered to be the most efficacious therapy.

LDMTX therapy is generally started with intermittent oral administration of the drug. The weekly starting dosage is generally 5–7.5mg of methotrexate in adults and 0.3 mg/kg in children.[4] If the response is inadequate and no adverse effects are present, the dose can be increased at 6- to 8-week intervals up to 25mg or 0.6 mg/kg, respectively. Self-administered subcutaneous injections have been recommended in those patients whose absorption of oral methotrexate is insufficient or in those with low tolerance to the tablets.

A split-dose regimen consisting of three divided doses given at 12-hour intervals may be useful in patients with gastrointestinal complaints, headache and fatigue early after drug ingestion, i.e. smaller doses result in the decrease of acute toxicity of LDMTX therapy.[25,66] The rationale seems to be prevention of excessive release of adenosine in the CNS. Conversely, a fortnightly maintenance schedule may be used once the disease is well controlled, or even in clinical remission. This every-other-week regimen may result in a flare of disease activity.[67] The mode of administration is based on the dose-dependent relationship to clinical outcomes. Indeed, the clinical efficacy of the drug is dose related,[29] although a definitive dose-concentration-effect relationship has not been elucidated and the predictive value of the administered dose for effect is rather poor.[15,68]

With regard to the pharmacokinetics of LDMTX, considerable interindividual variability of drug bioavailability seems to be the most important problem in reaching a significant clinical effect in terms of immunosuppressive and anti-inflammatory therapy. A significant correlation was recently reported between the AUC of methotrexate and both morning stiffness and the Ritchie articular index.[14] However, such correlation was not found with other parameters, including joint pain count, join swelling count, erythrocyte sedimentation rate, liver and renal function tests, and haemoglobin levels. Significant correlation was also found between AUC and decrease in Psoriasis Activity and Severity Index (PASI) score during 3 months of treatment (r2 = −0.91, p < 0.002).[15] This outcome may be significant for individualising LDMTX therapy. Others did not find any correlation between pharmacokinetics and efficacy of LDMTX in patients with rheumatoid arthritis.[69] Capone et al.[70] did not show any difference in LDMTX pharmacokinetics between patients with rheumatoid arthritis who did or did not respond to systemic therapy. However, during the 8-week study, all pharmacokinetic values obtained in responders were higher than those in nonresponders, even if the differences did not reach statistical significance.

The rationale for therapeutic drug monitoring of any drug is based on several prerequisites. LDMTX meets at least three of these: (i) large interindividual variability in kinetics; (ii) concentration-effect relationship; and (iii) a considerable time period for the clinical effect to develop. Our finding of a low intra-individual variability and no effect of the duration of therapy on methotrexate pharmacokinetics in plasma agrees with previous reports from studies over intermediate (13 weeks[15]) and long (6 months to 2 years[21,48]) periods. This suggests that therapeutic drug monitoring and dose individualisation need only be performed at the start of therapy, with a possible additional examination of methotrexate pharmacokinetics in patients showing an unsatisfactory response. According to the individually determined pharmacokinetic parameter values, it might be possible to assure effective exposure of the patient to the drug by changing the route of administration in case of inadequate absorption from the gastrointestinal tract.

The other important issue for the effect of LDMTX seems to be the ability of peripheral blood lymphocytes to convert methotrexate to polyglutamates. It is possible to measure the capacity of intracellular conversion and estimate the clinical outcome. Uptake and intracellular accumulation of methotrexate, along with the irreversibility of its effect on activated lymphocytes, provide a rationale for the intermittent weekly administration of the drug, in contrast to other anti-inflammatory and immunosuppressive agents that must be administered daily because of their short half-life and/or reversible activity.[44]

Decreased formation of methotrexate long-chain polyglutamates is associated with methotrexate resistance, whereas high levels of methotrexate polyglutamate accumulation are found in the blasts of leukaemia patients who respond to the therapy and have improved outcome.[71] The steady-state concentration of long-chain methotrexate polyglutamates depends on the balance of activities of two enzymes: folylpolyglutamyl synthetase (FPGS), which adds glutamate residues to methotrexate with γ-carboxyl linkages, and γ-glutamyl hydrolase (GGH) or conjugase, which sequentially removes the terminal glutamate residue of methotrexate polyglutamates. The ratio of GGH and FPGS activities could be used as a predictor of methotrexate polyglutamylation, response to methotrexate therapy and outcome.

4. Long-Term Maintenance Therapy

Patients usually receive LDMTX therapy for several months or years, which requires strategies to maintain efficacy, increase drug tolerability and prevent chronic toxicity. The most important goal is the prevention of excessive accumulation of methotrexate in tissues and the depletion of endogenous folates.

Nausea and fatigue are symptoms of the acute toxicity of LDMTX; however, they become more prominent with the duration of therapy. The incidence (about 30% of patients on LDMTX therapy) and significance of these clinical symptoms are probably related to intracellular depletion of folates, resulting in increased adenosine production and hyperhomocysteinaemia during long-term LDMTX administration.[1,55] Folic/folinic acid substitution therapy and division of the weekly dose of methotrexate into three equal doses given at 12-hour intervals is considered to alleviate the clinical significance of such symptoms.[1,38,72]

Hepatotoxicity is the major problem of LDMTX therapy, mainly in patients with psoriasis.[38] Elevation of hepatic enzymes, especially transaminases, occurs in about 20% of patients shortly after initiation of LDMTX therapy, but generally resolves within 1–3 weeks of treatment. However, the risk of developing severe hepatotoxicity (grade III or IV) occurs later and has been reported in approximately 3–25% of psoriatic patients. Hepatotoxicity is related to the dose given each week and to the cumulative dose of methotrexate (table I).[50,7381] Patients with rheumatoid arthritis on LDMTX pulse therapy rarely develop severe hepatotoxicity.[38,45]
Table I

Liver biopsy findings during methotrexate therapy of psoriasis with a single oral dose per week and three weekly oral doses at 12-hour intervals

Alcoholism is the major predisposing factor for LDMTX-induced hepatotoxicity. Other important risk factors include underlying liver disease and obesity, mainly in the presence of diabetes. Relative risk factors are renal insufficiency and drug-drug interactions during concomitant administration of aspirin, probenecid and sulfonamides. Abnormal liver histological findings correlate with serial elevations of hepatic transaminases (ALT, AST) and bilirubin. However, about 35% of biopsy-proven cases of fibrosis or cirrhosis have normal liver function tests. Neither biochemistry nor ultrasound is selective enough to distinguish between hepatic steatosis or fibrosis. The preferred method of assessing severity of methotrexate hepatotoxicity is liver biopsy (table II).[38]
Table II

Guidelines for liver biopsy during therapy of psoriasis with low-dose methotrexate[38]

Continuous intracellular depletion of tetrahydrofolate derivatives by LDMTX can result in hyperhomocysteinaemia, a well-known risk factor for teratogenicity, vascular ischaemic disease and thrombosis.[1] Patients with significantly higher plasma homocysteine levels were found to have a significantly more frequent incidence of gastrointestinal toxicity.[82,83] Interruption of methylation reactions and methionine biosynthesis could also participate in drug-induced hepatotoxicity (i.e. insufficient choline synthesis, resulting in lipid intracellular accumulation in hepatocytes). Recently a C677T mutation in the gene for methylenetetrahydrofolate reductase (MTHFR) was discovered, which has a prevalence of about 8% in the population.[1,84] This gene plays an important role in folate metabolism. The mutation decreases the enzymatic activity of MTHFR, resulting in hyperhomocysteinaemia.[84] However, no relationship was found between the change in homocystein concentration and the presence or absence of the C677T mutation in the MTHFR gene.[85]

Adenosine, via activation of A1 receptors, may also cause progressive subcutaneous nodulosis and bronchoconstriction.[1,4,86] Moreover, the immunosuppressive properties of adenosine can explain the existence of the parallel between severe combined immunodeficiency syndrome (SCID), caused by adenosine deaminase deficiency resulting in the accumulation of adenosine in the body, and LDMTX therapy.

Many reports describe opportunistic infections caused by Pneumocystis carinii in patients on LDMTX therapy. Several reports associate these infections with pancytopenia,[56,87,88] whereas others describe pulmonary infection only.[8991] Very often, these patients are elderly with decreased CLCR and infections of the urinary tract treated with cotrimoxazole (trimethoprim-sulfamethoxazole).

LDMTX therapy can also induce interstitial lung injury, especially in patients with rheumatoid arthritis. The X-ray film generally shows linear and reticulonodular basal and/or midlung field infiltrates. A biopsy specimen shows monocellular cell desquamation into the alveoli, along with a tendency to form noncaseating granulomas that contain multicellular giant cells. About 40% of patients have concomitant eosinophilia and 17% have a rash.[45] The strongest predictor for lung injury is old age (≥60 years), diabetes mellitus, rheumatoid pulmonary involvement, previous use of DMARDs and hypoalbuminaemia.[92]

The risk for all types of adverse effects is permanent, requiring long-term monitoring of LDMTX therapy.[4,9395] Plasma methotrexate concentrations at 24 hours post-dose usually range from 0.01–0.1 µmol/L in patients on LDMTX therapy, regardless of their age.[39,54,96] Although 24-hour plasma methotrexate concentrations ≤0.1 µmol/L would give an individual patient a wide margin of safety, monitoring this concentration has been recommended in case of a potential decrease in methotrexate clearance.[97] This advice is supported by the increased toxicity of methotrexate in patients with renal impairment. It also seems that individual susceptibility and folate deficiency play a major role in methotrexate tolerability.[21,98]

A special problem concerns the possible oncogenicity of LDMTX therapy. To evaluate the incidence of cancer caused by LDMTX therapy is rather difficult, since the autoimmune diseases themselves, notably rheumatoid arthritis, carry a higher risk of developing non-Hodgkin’s and Hodgkin’s lymphomas (approximately 2-fold higher risk compared with the healthy population), multiple myeloma, leukaemia, lung cancer and nonmelanoma skin cancer.[99] The immunosuppressive and pro-oncogenic role of other immunosuppressants, notably azathioprine and cyclophosphamide, is well documented, essentially for bladder carcinomas. However, no higher risk for developing malignant disease was noted in patients treated with LDMTX.[99,100] LDMTX therapy of psoriasis does not correlate with a higher risk of inducing lymphomas and leukaemias, although a positive correlation exists between combined therapy with PUVA (psoralens plus UV light) and LDMTX and the incidence of skin squamous spinocellular carcinoma.[101]

5. Supplementation with Folic and Folinic Acids

To prevent the intracellular depletion of folates and the consequent hyperhomocysteinaemia, nausea and megaloblastic anaemia, the most rational method seems to be the supplementation of LDMTX therapy with at least 1mg of folic acid daily.[72,101,102] Administration usually begins 24 hours after the LDMTX dose, with the goal of not diminishing the effect of immunosuppressive therapy.[103] However, recent studies suggest that 1mg of folic acid given the same day as LDMTX does not interact with the immunosuppressive and anti-inflammatory effect.[98,104] On the other hand, late administration of folic acid, i.e. 3 to 4 days after LDMTX, was no more effective than placebo in controlling nausea.[1] Recently, Bressolle et al.[52] reported increases of methotrexate total clearance and volume of distribution in patients taking folic acid supplementation therapy. This observation was interpreted as increased cellular uptake of methotrexate, with the folic acid supplements promoting the intracellular sequestering of methotrexate. It is also possible that the reduced AUC is associated with increased intracellular folate levels.

Folinic acid (leucovorin) may also be used for oral supplementation therapy. If doses of LDMTX are 5–12.5mg weekly, 2.5mg of folinic acid is administered the day after LDMTX. If the dose is 15–30mg weekly, the amount of folinic acid is increased to 5mg, given the day after methotrexate.[38,105] Clinical studies reveal a decrease of about 50% in reported adverse effects in patients receiving supplementation therapy, especially with respect to nausea and fatigue.[101,106]

6. Conclusion

A better understanding of LDMTX pharmacokinetics and mechanism of action enables more rational and safer clinical use of the drug as monotherapy and/or in combination with other immunosuppressive and anti-inflammatory drugs. Because a relationship exists between the pharmacokinetics and pharmacodynamics of methotrexate, it is possible to individualise both the mode of LDMTX administration and the dosage regimen according to the pharmacokinetics in the individual patient. Tailored LDMTX treatment and folic/folinic acid supplementation may enable a reduction of acute and chronic toxicity, maintain efficacy, and increase the tolerability of the drug.


We wish to acknowledge the contribution of the personnel in the Department of Pharmacology, Charles University in Hradec Králové, Czech Republic, for their assistance in completing this overview. The data used in the manuscript were obtained from MEDLINE.

Copyright information

© Adis International Limited 2003