The pharmacokinetics of low dose methotrexate (MTX) were evaluated in 12 rheumatoid arthritis patients in the presence and absence of steady-state levels of salicylic acid (ASA) and sulindac (SU). Using a Latin square design, patients were given MTX plus ASA (mean 3.4 g/day), MTX plus SU (mean 400 mg/day), or MTX alone. On a background of at least one year of regular MTX therapy, patients received 10 mg/m2 MTX iv (mean 17.8 mg) given after at least 2 weeks of treatment with each of the above regimens. Plasma concentrations of MTX and 7-hydroxymethotrexate (7-OH-MTX) were measured using HPLC. No differences in MTX clearance (Cl) were found comparing MTX alone, MTX + ASA, and MTX + SU. However, if one particular subject that had a very low clearance when receiving MTX alone was excluded, there was a statistically significant decrease in MTX clearance when either ASA or SUL were present. It is also noteworthy that ASA significantly increased the exposure of the subject to 7-OH-MTX and, to a lesser extent, so did sulindac. Since 7-OH-MTX has been shown to be an active metabolite when given for cytotoxic effects at higher doses and because it has been show to be nephrotoxic at doses a thousand-fold greater than used in rheumatoid arthritis, nonsteroidal anti-inflammatory drugs should be used cautiously with MTX until further large scale safety studies are conducted. The data indicate that if a clinically significant interaction were to occur, ASA is more likely than SU to interact with MTX.