Low-dose pulse methotrexate has emerged as one of the most frequently used slow-acting, symptom-modifying antirheumatic drugs in patients with rheumatoid arthritis (RA) because of its favourable risk-benefit profile. Methotrexate is a weak bicarboxylic acid structurally related to folic acid. The most widely used methods for the analysis of methotrexate are immunoassays, particularly fluorescence polarisation immunoassay. After oral administration, the drug is rapidly but incompletely absorbed. Since food does not significantly affect the bioavailability of oral methotrexate in adult patients, the drug may be taken regardless of meals. There is a marked interindividual variability in the extent of absorption of oral methotrexate. Conversely, the intraindividual variability is moderate even over a long time period. Intramuscular and subcutaneous injections of methotrexate result in comparable pharmacokinetics, suggesting that these routes of administration are interchangeable. A mean protein binding to serum albumin of 42 to 57% is usually reported. Again, the unbound fraction exhibits a large interindividual variability. The steady-state volume of distribution is approximately 1 L/kg. Methotrexate distributes to extravascular compartments, including synovial fluid, and to different tissues, especially kidney, liver and joint tissues. Finally, the drug is transported into cells, mainly by a carrier-mediated active transport process. Methotrexate is partly oxidised by hepatic aldehyde oxidase to 7-hydroxymethotrexate. This main, circulating metabolite is over 90% bound to serum albumin. Both methotrexate and 7-hydroxy-methotrexate may be converted to polyglutamyl derivatives which are selectively retained in cells. Methotrexate is mainly excreted by the kidney as intact drug regardless of the route of administration. The drug is filtered by the glomeruli, and then undergoes both secretion and reabsorption processes within the tubule. These processes are differentially saturable, resulting in possible nonlinear elimination pharmacokinetics. The usually reported mean values for the elimination half-life and the total body clearance of methotrexate are 5 to 8 hours and 4.8 to 7.8 L/h, respectively. A positive correlation between methotrexate clearance and creatinine clearance has been found by some authors. Finally, the pharmacokinetics of low-dose methotrexate appears to be highly variable and largely unpredictable even in patients with normal renal and hepatic function. Furthermore, studies in patients with juvenile rheumatoid arthritis provide evidence of age-dependent pharmacokinetics of the drug. These features must be considered when judging the individual clinical response to methotrexate therapy. Various drugs currently used in RA may interact with methotrexate. Aspirin might affect methotrexate disposition to a greater extent than other nonsteroidal anti-inflammatory drugs without causing greater toxicity. Corticosteroids do not interfere with the pharmacokinetics of methotrexate, whereas chloroquine may reduce the gastrointestinal absorption of the drug. Folates, especially folic acid, have been shown to reduce the adverse effects of methotrexate without compromising its efficacy in RA. Finally, both trimethoprim-sulfamethoxazole (cotrimoxazole) and probenecid lead to increased toxicity of methotrexate, and hence should be avoided in patients receiving these drugs. A relationship between oral dosage and efficacy has been found in the range 5 to 20mg methotrexate weekly. The plateau of efficacy is attained at approximately 10 mg/m2/week in most patients. No clear relationship between pharmacokinetic parameters and clinical response has been demonstrated. Overall, the dosage must be individualised because of interindividual variability in the dose-response curve. This variability is probably related, at least in part, to the wide interindividual variability in the disposition of the drug.