Mycophenolic acid (MPA), the active drug moiety of mycophenolate, is a potent immunosuppressant agent, which is increasingly being used in the treatment of patients with various autoimmune diseases. An understanding of the pharmacokinetics and pharmacodynamics of mycophenolate in this population should assist the clinician with rational dosage decisions. This review aims to provide an overview of the published literature on the clinical pharmacokinetics of mycophenolate in autoimmune disease and a briefer summary of current pharmacodynamic knowledge, and to identify areas of potential future research in this field. A literature search was conducted using PubMed and EMBASE databases as well as bibliographies of relevant articles and 'on-line early' pages of key journals. Twenty-six pharmacokinetic/pharmacodynamic studies of mycophenolate in people with autoimmune disease were identified and appraised. Twenty-two of these studies used non-compartmental analysis techniques and four used population modelling methods to estimate mycophenolate pharmacokinetic parameters. Seven studies linked mycophenolate exposure to treatment outcomes. Only four studies measured free (unbound) as well as total mycophenolate exposure and only two studies characterised MPA disposition following enteric-coated mycophenolate sodium (EC-MPS) administration. Across all studies MPA displayed erratic and complex pharmacokinetics with substantial between-subject variability. Based on total drug measurement, the dose-normalised MPA area under the plasma concentration-time curve (AUC) from 0 to 12 h post-dose (AUC12) varied at least five- to ten-fold between subjects. Typical values for apparent oral clearance (CL/F) of MPA during nonlinear mixed-effects modelling ranged from 8.3 to 25.3 L/h. Patient renal function, serum albumin levels, sex, ethnicity, food intake, concurrent administration of interacting drugs such as antacids, metal-containing medications and proton pump inhibitors and polymorphisms in genes encoding uridine diphosphate glucuronosyltransferase were identified in some studies as having a significant influence on the pharmacokinetics of mycophenolate. Typical MPA CL/F values in autoimmune disease patients were generally slightly lower than values published previously in population pharmacokinetic studies involving renal allograft recipients, possibly because of usage of ciclosporin, poorer renal function or lower serum albumin levels in the renal transplant cohort. In a single crossover study involving ten subjects only, significantly higher MPA AUC12 and maximum MPA concentration (C max) and lower MPA CL/F were reported following EC-MPS administration compared to mycophenolate mofetil administration. MPA exposure correlated well with treatment efficacy in patients with autoimmune disease (response to treatment, active disease and disease markers); however the relationship between MPA exposure and adverse events (infectious episodes, haematological toxicity and gastrointestinal symptoms) was unclear. Further investigation is required in autoimmune diseases such as chronic plaque psoriasis and rheumatoid arthritis and following EC-MPS administration. The extent of within-subject variability in the pharmacokinetics of mycophenolate is largely unknown and potential covariate influences need to be confirmed in studies with large subject numbers. A relationship between MPA and MPA metabolite exposure and toxicity needs to be established. The contribution of pharmacogenetics to the pharmacokinetics and pharmacodynamics of mycophenolate warrants further investigation, as does the utility of therapeutic drug monitoring. Dosing to achieve a target MPA AUC12 >35 mg·h/L is likely to lead to better efficacy outcomes in patients with autoimmune disease (rather than just giving standard doses, which lead to a wide range of exposures). However, the relationship between mycophenolate exposure and toxicity requires further investigation to determine the upper end of a target AUC range.