CLINICAL THERAPEUTICS® / VOL. 26, NO. 4, 2004 Leflunomide in the Treatment of Rheumatoid Arthritis Edmund K. Li, FRCP, Lai-Shan Tam, MRCP (UK), and Brian Tomlinson, FRCP Department of Medicine & Therapeutics, The Chinese University of Hong Kong, Prince of Wales Hospital, Hong Kong ABSTRACT Background: Current drug therapies for rheumatoid arthritis (RA), including nonsteroidal anti-inflammatory drugs and disease-modifying antirheumatic drugs, help control inflammation but can cause significant toxicity. Drugs are needed that are able to suppress inflammation and modify the underlying immune response with improved tolerability. Leflunomide is an agent that affects the inflammatory process, particularly in RA. Objective: This paper reviews the pharmacology of leflunomide, its approved use in RA, and the results of major clinical trials, including adverse events. Methods: Relevant trials were identified through a search of the English-language literature indexed on EMBASE, MEDLINE, Current Contents, and the Cochrane Controlled Trials Register from January 1980 to November 2003. Search terms were limited to leflunomide. Results: In 3 large Phase III clinical trials (US301, MN301, and MN302), leflunomide had equivalent clinical efficacy and tolerability to methotrexate and sulfasalazine and superior efficacy and tolerability compared with placebo. In US301 (N = 482), the ACR (American College of Rheumatology) 20 response rate (proportion of patients with ≥20% improvement from baseline in tender and swollen joint counts, patient’s assessment of pain, patient’s and physician’s global assessment of disease activity, physical function, and acute-phase reactant value) at 1 year was similar with leflunomide and methotrexate and significantly greater with both active treatments than with placebo (52%, 46%, and 26%, respectively; both, P < 0.001). The efficacy of leflunomide was seen early (after 4 weeks of treatment) and was sustained throughout the study. There was less radiographic damage in both active-treatment groups compared with placebo (leflunomide, P ≤ 0.001; methotrexate, P = 0.02). In MN301 (N = 358), the ACR20 response rate at 6 months was similar with leflunomide and sulfasalazine and significantly greater with both active treatments compared with placebo (55%, 56%, and 29%, respectively; both, P < 0.001). Radiographic progression was also similar with leflunomide and sulfasalazine, both of which were significantly superior to placebo (Larsen score, 0.42, 0.41, and 1.4; both, P < 0.001). An extension of this study revealed maintenance of efficacy at 12 and 24 months. In MN302 (intent-to-treat population, N = 999), 50.5% of patients in the leflunomide group were ACR20 responders at the end of 1 year, compared with 64.8% in the methotrexate group (P < 0.001 vs leflunomide). After 2 years, ACR20 response rates were similar with leflunomide and methotrexate (64.3% and 71.7%). The overall safety profile of leflunomide appears promising, although monitoring for elevations in liver enzymes and bone marrow suppression is recommended. The most common drug-related adverse events associated with leflunomide in these clinical trials were diarrhea, abnormalities in liver enzymes, rash, and hypertension. Accepted for publication January 23, 2004. Printed in the USA. Reproduction in whole or part is not permitted. Copyright © 2004 Excerpta Medica, Inc. 0149-2918/04/$19.00 447 CLINICAL THERAPEUTICS® Conclusions: In the trials reviewed, leflunomide had approximately comparable clinical efficacy and tolerability to methotrexate and sulfasalazine, and was superior to placebo. The onset of treatment efficacy occurred more rapidly with leflunomide than with other active treatments. (Clin Ther. 2004;26: 447–459) Copyright © 2004 Excerpta Medica, Inc. Key words: clinical trial, leflunomide, rheumatoid arthritis. INTRODUCTION Rheumatoid arthritis (RA) is an inflammatory condition that affects 1% of the population worldwide and is associated with substantial morbidity and mortality. Standard pharmacotherapies include nonsteroidal anti-inflammatory drugs (NSAIDs), which reduce arachidonic acid products generated by cyclooxygenase and have significant analgesic and anti-inflammatory effects but are associated with gastrointestinal intolerance, and disease-modifying antirheumatic drugs (DMARDs), which improve symptoms and disease progression with differing degrees of efficacy and toxicity. Available therapies for rheumatic diseases have been inadequate to control inflammation without significant toxicity. Drugs are needed that are able to suppress inflammation and modify the underlying immune response with improved tolerability. For more than 2 decades, no new DMARDs were introduced for the treatment of RA. The standard DMARDs were auranofin, azathioprine, cyclophosphamide, cyclosporine, D-penicillamine, hydroxychloroquine, intramuscular gold, methotrexate, and sulfasalazine. In 1999, 3 new agents—leflunomide,* etanercept,† and infliximab‡—were approved for the treatment of RA by the US Food and Drug Administration. Etanercept and infliximab are biologic interventions that selectively block tumor necrosis factor alpha (TNF-α). Etanercept is a recombinant soluble TNF-Fc fusion protein available for subcutaneous injection, and infliximab is a chimeric (mouse–human) anti-TNF monoclonal antibody for *Trademark: Arava® (Aventis Pharmaceuticals Inc., Bridgewater, New Jersey). †Trademark: Enbrel® (Amgen Inc., Thousand Oaks, California). ‡Trademark: Remicade® (Centocor, Inc., Malvern, Pennsylvania). 448 intravenous infusion. Recently, 2 additional agents became available. Adalimumab§ is a fully humanized monoclonal TNF-α antibody, available for subcutaneous administration, and anakinra?? is a humanized monoclonal antibody and interleukin-1–receptor antagonist, available for subcutaneous injection. This article reviews the preclinical pharmacology, pharmacokinetics, and mechanism of action of leflunomide. The results of major clinical trials of its use in RA, its efficacy compared with current treatments, and adverse events associated with its use are summarized. Relevant trials were identified through searches of the English-language literature indexed on EMBASE, MEDLINE, Current Contents, and the Cochrane Controlled Trials Register from January 1980 to November 2003. The search terms were limited to leflunomide. PRECLINICAL PHARMACOLOGY Leflunomide is a synthetic isoxazole derivative (N-[4trifluoro-methylphenyl]-5-methylisoxazol-4-carboxamide) that was first shown to have disease-modifying properties in the 1980s in animal models of arthritis and autoimmunity. Not only did it inhibit the inflammatory response in adjuvant arthritic rats, but it improved the diminished proliferation of mitogeninduced lymphocytes in diseased animals.1 Although leflunomide has no chemical relationship or mechanism of action in common with any of the conventional immunomodulatory drugs, it has demonstrated prophylactic and therapeutic effects in experimental models of chronic graft-versus-host disease.2 It has also been used successfully to suppress the development of systemic lupus erythematosus in MRL/lpr mice,3 adjuvant arthritis in the Lewis rat,1 and proteoglycan-induced arthritis in genetically susceptible BALB/C mice,4 as well as to suppress the development of autoimmune diabetes in a mouse model.5 Other diseases in which leflunomide has shown benefit in the rat model include tubulointerstitial nephritis,6 glomerulonephritis induced by anti–basement membrane antibody,7 and myasthenia gravis.8 Because of its success in standard animal models of autoimmune disease, leflunomide was tested in patients with RA and was reported to be effective in 1995.9 §Trademark: Humira™ (Abbott Laboratories, Abbott Park, Illinois). ??Trademark: Kineret™ (Amgen Inc.). E.K. Li et al. PHARMACOKINETICS Leflunomide is almost completely absorbed from the gastrointestinal tract and undergoes nonenzymatic conversion in the intestinal mucosa and plasma (and possibly during first pass through the liver) through opening of the isoxazole ring to form the active malononitrilamide metabolite A77 1726 (3 cyano-3hydroxy-N-[4-trifluronmethylphenyl] crotonamide). A77 1726 is highly protein bound in plasma (99.5%), predominantly to albumin, and is not dialyzable. It has a low volume of distribution (0.13 L/kg at steady state).10 Leflunomide’s half-life is ~15 days in humans. It may take 2 to 3 weeks to reach steady-state plasma concentrations after initiation of therapy with a loading dose of 100 mg/d for 3 days, followed by a maintenance dose of 10 to 20 mg/d; without the loading dose, it may take up to 2 months. Plasma concentrations of A77 1726 are dose proportional after administration of both loading and maintenance doses of leflunomide. Mean (SD) steady-state plasma concentrations at 24 hours after receipt of maintenance doses of 5, 10, and 25 mg/d for 24 weeks were 8.8 (2.9), 18 (9.6), and 63 (36) µg/mL, respectively.9 Leflunomide is cleared by hepatic metabolism, mainly to trifluoromethylaniline (TFMA)–oxanillic acid, and by subsequent biliary excretion, which may be affected by hepatic dysfunction. Based on the finding that the drug’s half-life can be decreased to 1 to 2 days by the administration of cholestyramine or activated charcoal, which prevent reabsorption, it is likely that enterohepatic circulation occurs. Leflunomide is excreted in almost equal proportions in the bile as A77 1726 and in the urine as leflunomide glucuronide and TFMA–oxanillic acid.11 MECHANISMS OF ACTION Two mechanisms of action have been described for leflunomide in vitro. They are discussed in the following sections. Inhibition of Dihydroorotate Dehydrogenase Leflunomide’s unique mechanism of action in the treatment of RA is its ability to selectively inhibit the enzyme dihydroorotate dehydrogenase (DHODH), a mitochondrial enzyme critical for de novo biosynthesis of pyrimidine ribonucleotides such as ribonucleotide uridine monophosphate (rUMP) and ribonucleotide uridine triphosphate (Figure 1). This enzyme is used by activated lymphocytes, rap- idly proliferating cells involved in the pathogenesis of RA.12,13 Blockade of pyrimidine synthesis resulting from a reduction in the pyrimidine nucleotide pool interrupts T lymphocyte clonal expansion, arresting the cell cycle between the G1 and S phases (Figure 2). In the resting phase (G0) of lymphocytes and other cell types (eg, hematopoietic cells, gastrointestinal lining cells), the biosynthetic pathway for de novo pyrimidine ribonucleotide synthesis is inactive and can rely on the salvage pathway.11 However, in the presence of antigenic signals, such as occur in inflamed joints, T lymphocytes are stimulated to proliferate. Lymphocytes then progress from G0 to G1 and subsequently to S, with cellular DNA proliferation. Under these circumstances, lymphocytes must accumulate sufficient amounts of nucleic acid precursors (pyrimidines and purines) to provide an energy source for membrane biosynthesis and cell division. The pyrimidine ribonucleotide pool (ie, uridine and cytosine) expands roughly 8-fold,14 compared with only 2-fold expansion of the purine ribonucleotide pool (ie, adenosine and guanosine). After lymphocyte activation, enzyme activity is increased in the salvage pathway of pyrimidine synthesis but is unable to meet the increased demand for pyrimidines; therefore, the de novo pathway occurring in the mid-G1 phase is used to meet this higher demand. The basic step in de novo pyrimidine synthesis is conversion of dihydroorotate to orotate by DHODH. Orotate diffuses into the cytoplasm and is subsequently converted to rUMP by a multi-enzyme complex. rUMP functions in DNA and RNA synthesis, in membrane biosynthesis, and as an energy source for activated cells (Figure 1).14 Leflunomide’s inhibition of DHODH leads to a reduction in the essential pyrimidine precursors (eg, rUMP), causing arrest of the cell cycle at the G1 phase. Indeed, when A77 1726 was added to mitogenstimulated human blood lymphocytes, the proportion of stimulated cells that progressed to the S phase was markedly diminished,15 interrupting cell cycle progression. The action of A77 1726 leads to lowered levels of rUMP, which can result in a modification of synovial high endothelial venules that permits homing of activated lymphocytes to the synovium.16 Through the lowering of rUMP, leflunomide may also exert anti-inflammatory effects by inducing increased 449 CLINICAL THERAPEUTICS® Uridine triphosphate ATP + glutamine Carbamoyl phosphate synthetase Uridine diphosphate Uridine diphosphate Carbamoyl phosphate Uridine monophosphate (rUMP) Uridine monophosphate (rUMP) Carbamoyl aspartate Uridine monophosphate (rUMP) Orotidine monophosphate Orotidine monophosphate Dihydroorotate Orotate Orotate DHODH Dihydroorotate Orotate Mitochondria Figure 1. Outline of uridine synthesis. ATP = adenosine triphosphate; DHODH = dihydroorotate dehydrogenase; rUMP = ribonucleotide uridine monophosphate. synthesis of immunosuppressive cytokines such as transforming growth factor beta (TGF-β).17 Tyrosine Kinase Inhibition and Other Actions The other mechanism of action of leflunomide is inhibition of tyrosine kinase.18 Tyrosine kinases are essential for signal transduction, induction of cell growth, and differentiation of activated cells. Although A77 1726 inhibits the activity of tyrosine kinase, the drug concentration required is considerably higher than that required for inhibition of 450 DHODH or than plasma concentrations generally attained in patients with RA15; therefore, in terms of immunomodulation, an effect on tyrosine kinase is unlikely to be the primary antiproliferative action of this drug. Other possible secondary mechanisms of action of leflunomide include inhibition of the activation of nuclear factor κB, a potent mediator of several proinflammatory genes that is induced by TNF.19 Leflunomide also inhibits the expression of cell adhesion molecules, which facilitate cellular interactions E.K. Li et al. Stimulation of cell T cell G0 Preparation for S phase De novo synthesis of pyrimidine ribonucleotides G1 Leflunomide Replication of DNA S G2 Mitosis M M M Figure 2. Cell cycle progression from the G0 (resting) phase to mitosis (M) of activated lymphocytes. In the mid-G1 phase, the pathway for de novo synthesis of ribonucleotides is upregulated. This process must provide sufficient pyrimidines (uridine, thymidine, cytidine) and purines (adenosine, guanosine) for the cell to progress from the G1 phase to the S phase. involved in antigen presentation, secretion of cytokines, and production of matrix metalloproteinases that degrade articular cartilage and bone.20–22 In vitro, leflunomide reduced oxygen freeradical production by human monocytes.23 It also reduced neutrophil chemotaxis in vitro, resulting in a rapid reduction in the number of neutrophils infiltrating the rheumatoid joint cavity.24 Augmentation of the immunosuppressive cytokine TGF-β117 and inhibition of cyclooxygenase-2 activity have also been observed with leflunomide.25 CLINICAL TRIALS OF LEFLUNOMIDE The clinical efficacy of leflunomide in RA was first demonstrated in a German Phase II randomized, double-blind, placebo-controlled trial involving 402 patients with active disease.9 Patients received leflunomide 5, 10, and 25 mg/d or placebo for 6 months. Leflunomide 10 and 25 mg/d were found to be effective based on improvements in both the patient’s and physician’s global assessments (P < 0.05 vs placebo), accepted measures of response in clinical trials of RA. Since that study, there have been 3 multicenter, randomized, double-blind, placebo-controlled Phase III trials of leflunomide in the treatment of RA, referred to as US301,26 MN301,27 and MN302.28 In all 3 studies combined, 1838 patients were randomized to treatment and 816 received leflunomide. Approximately 40% of patients had early RA, with a 451 CLINICAL THERAPEUTICS® disease duration of <2 years; overall, the mean duration of disease was ~5 years. Almost 40% of patients were DMARD naive. Treatment groups in the 3 trials received leflunomide, methotrexate, or placebo,26 leflunomide or methotrexate,28 or leflunomide, sulfasalazine, or placebo.27 Patients randomized to receive leflunomide were given a loading dose of 100 mg for 3 days, followed by a maintenance dose of 20 mg/d. The results of these clinical trials are described in the following sections (Table). Leflunomide Versus Methotrexate and/or Placebo US301, reported by Strand et al,26 was conducted in the United States and Canada and involved 482 patients. Patients were randomized in a 3:3:2 ratio to receive methotrexate 7.5 to 15 mg/wk, leflunomide 20 mg/d, or placebo. Among the clinical end points was the ACR20 response rate, a standard measure defined by the American College of Rheumatology (ACR) as the proportion of patients with ≥20% improvement from baseline in tender and swollen joint counts, patient’s assessment of pain, patient’s and physician’s global assessment of disease activity, physical function, and acute-phase reactant value.29 At 1 year, the ACR20 response rate was 52% in the leflunomide group, 46% in the methotrexate group, and 26% in the placebo group (P < 0.001, both active treatments vs placebo). Leflunomide showed efficacy after 4 weeks of treatment, and efficacy was sustained throughout the study period. Radiographic analyses demonstrated significantly less structural damage in both active-treatment groups compared with placebo (leflunomide, P ≤ 0.001; methotrexate, P = 0.02). Based on these results, the study investigators considered leflunomide and methotrexate to have comparable efficacy. Furthermore, leflunomide was associated with significantly greater patient-reported improvements versus methotrexate in scores on several categories of the Health Assessment Questionnaire (HAQ) (P < 0.01) and the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36) (P < 0.001). The HAQ is an established questionnaire for assessing general function in adults, and the SF-36 was designed for use in population surveys. In an extension of this study to 2 years, Cohen et al30 reported that efficacy was sustained in both active-treatment groups and that leflunomide produced significantly greater improvements in physical 452 function over 2 years of treatment compared with methotrexate (P = 0.005). In MN302, reported by Emery et al,28 999 patients with active RA were randomized to receive a loading dose of leflunomide 100 mg/d for 3 days, followed by leflunomide 20 mg/d, or methotrexate 7.5 to 15 mg/wk for 1 to 2 years. During the first year, 50.5% (250/495) of patients in the leflunomide group were ACR20 responders, compared with 64.8% (317/489) in the methotrexate group (P < 0.001 vs leflunomide). ACR50 and ACR70 response rates were slightly but not significantly higher in the methotrexate group. The overall improvement with methotrexate in this study was 30% to 70% greater than that reported by Strand et al in US301.26 Two possible reasons were proposed for the greater efficacy of methotrexate: first, folic acid supplementation in all methotrexate recipients in US301 (compared with <10% of patients in MN302) may have decreased the efficacy of methotrexate; second, patients in US301 had a longer history of disease (6.5–7.0 y) compared with those in MN302 (3.7–3.8 y). After 2 years of treatment, however, there was no statistically significant difference in ACR20 response rates between the leflunomide and methotrexate groups in MN302 (64.3% and 71.7%, respectively).28 Radiographic assessment using the Larsen score, an index of bone and joint damage in the hands and feet, indicated equivalent disease progression after 1 year of treatment with leflunomide or methotrexate (mean [SD], 1.25 [0.48] vs 1.29 [0.45]). After 2 years, patients who received leflunomide had no further increase in joint damage, and patients who received methotrexate had a small improvement. The net result was a small but nonsignificant treatment difference for change in radiographic scores between active-treatment groups. However, in the absence of a placebo group, the effect of treatment on disease progression cannot be clearly evaluated. Leflunomide Versus Sulfasalazine and/or Placebo MN301, reported by Smolen et al,27 was a 6-month randomized, double-blind study conducted in Europe, Australia, New Zealand, and South Africa. Three hundred fifty-eight patients were randomized to 1 of 3 treatment arms: a loading dose of leflunomide 100 mg/d for 23 days, followed by leflunomide 20 mg/d; sulfasalazine 0.5 g/d, titrated to 2 g/d at E.K. Li et al. Table. Summary of randomized control trials of leflunomide. In trials that included a placebo group, comparisons are versus placebo; in trials without a placebo group, comparisons are versus the active comparator. Mladenovic et al9* Smolen et al27† LEF/PLA LEF/PLA SSZ/PLA LEF MTX LEF/PLA MTX/PLA LEF SSZ LEF MTX 100/102 133/92 501 498 182/118 80 76 98 101 6 6 12 24 Outcome measures Tender joint count Swollen joint count Patient’s global assessment Physician’s global assessment Duration of morning stiffness Pain score,VAS ESR CRP ACR20 HAQ SF-36 → ↑¶ ↑¶ ↑¶ → ↑¶ ↑¶ ↑¶ ↑‡ NA NA ↑‡ ↑‡ ↑\ ↑\ ↑†† ↑‡ ↑\ ↑‡ ↑‡ ↑‡ NA ↑‡ ↑‡ ↑\ ↑\ ↑†† ↑‡ ↑†† ↑‡ ↑‡ ↑‡ NA ↑§ ↑‡ ↑\ ↑\ ↑¶ ↑¶ ↑¶ ↑¶ ↑‡ ↑¶ NA → ↑# → ↑** → → ↑‡ ↑§§ → → NA ↑\ ↑\ ↑\ ↑\ ↑¶ ↑‡‡ ↑‡‡ ↑\ ↑‡ ↑‡ ↑\ Radiographic measures Larsen score Sharp score NA NA ↑‡ NA ↑‡ NA → NA → NA NA ↑\ No. of patients Duration of follow-up, mo. 133 Emery et al28† Strand et al26† 182 12 Scott et al31† Cohen et al30† 24 (ext. of Smolen) 24 (ext. of Strand) ↑\ ↑\ ↑\ ↑\ → ↑‡‡ ↑‡‡ ↑\ ↑\ ↑‡‡ ↑\ → → ↑\ ↑\ → → → → ↑‡‡ ↑‡‡ NA ↑¶ → → ↑¶ ↑¶ ↑¶ → → → ↑¶ → NA ↑\\ → NA NA → LEF = leflunomide; PLA = placebo; SSZ = sulfasalazine; MTX = methotrexate; → = no change versus PLA or other active treatment; ↑ = improvement; VAS = visual analog scale; ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; ACR20 = American College of Rheumatology 20 response rate (proportion of patients with ≥20% improvement from baseline in tender and swollen joint counts, patient’s assessment of pain, patient’s and physician’s global assessment of disease activity, physical function, and acute-phase reactant value); HAQ = Health Assessment Questionnaire; NA = not available; SF-36 = Medical Outcomes Study 36-Item Short-Form Health Survey. *With leflunomide 10 mg/d. †With leflunomide 20 mg/d. ‡P < 0.001. §P = 0.006. \P = 0.001. ¶P = 0.005. #P = 0.017. **P = 0.015. ††P = 0.004. ‡‡P = 0.01. §§P = 0.008. \\P = 0.02. week 4; or placebo. At 6 months, ACR20 response rates were 55% for leflunomide, 56% for sulfasalazine, and 29% for placebo (P < 0.001, both active treatments vs placebo). Leflunomide and sulfasalazine showed similar efficacy in retarding radiographic progression, as shown by changes in the Larsen score for eroded joint count, and were significantly more effective than placebo (0.42, 0.41, and 1.4, respectively; P < 0.001). Leflunomide produced an earlier response, with 31% of patients having an ACR20 response at week 4, compared with 19% and 9% with sulfasalazine and placebo, respectively. An extension of this study to 2 years reported by Scott et al31 found maintenance of efficacy at 12 and 24 months. At 2 years, receipt of leflunomide was associated with significant improvements versus sul453 CLINICAL THERAPEUTICS® fasalazine in the physician’s global assessment (50% vs 32%, respectively; P = 0.001), patient’s global assessment (46% vs 30%; P = 0.001), ACR20 response rate (82% vs 60%; P < 0.01), and functional ability, as measured by HAQ score (–0.65 vs –0.36; P = 0.01).32 Leflunomide was also significantly superior to sulfasalazine in terms of the more stringent ACR50 response rate (52% vs 25%; P = 0.04). Patients who had benefited from 2 years of leflunomide treatment in MN30127 and MN30228 were invited to participate in an open-label extension study in which they would receive leflunomide for up to 5 years.33 Among the 214 patients who entered the extension study, the improvements in ACR20, ACR50, and ACR70 response rates observed at year 1 (72.9%, 48.3%, and 14.5%, respectively) were maintained (69.2%, 43.0%, and 19.6%) for a mean of 4.6 years. Improved functional ability, as measured in terms of HAQ score (a change of –0.22 is considered clinically meaningful), was seen at year 1 (mean change, –0.6) and maintained through the end of the study (mean change, –0.5 and –0.5, respectively). Combination Treatment with Leflunomide and Methotrexate The efficacy of leflunomide used in combination with methotrexate has been assessed in 2 trials: an open-label trial in 30 patients34 and a randomized, double-blind, placebo-controlled trial in 263 patients with RA.35 In both studies, leflunomide was added to methotrexate in patients who had failed to respond to methotrexate alone. In the open-label trial,34 the combination of leflunomide and methotrexate produced an improvement in the ACR20 response rate; the response rate reached 53% (16/30 patients) at the end of 1 year, which was considered to indicate therapeutic potential. The combination was generally well tolerated, with the exception of an elevation in liver enzymes in 19 patients; in 68% (13) of these cases, levels of both aspartate and alanine aminotransferase decreased to <1.2 times the upper limit of normal by the end of the study without a reduction in the dose of leflunomide. An analysis of methotrexate pharmacokinetic parameters (ie, maximum plasma concentration, time to maximum concentration, area under the concentrationtime curve) involving 11 of the 30 patients showed 454 no significant changes from baseline (before administration of leflunomide) at any time during leflunomide administration. In the randomized, double-blind, placebocontrolled trial,35 46.2% (60/130) of patients in the leflunomide/methotrexate group and 19.5% (26/133 patients) in the placebo/methotrexate group met ACR20 response criteria at 24 weeks (P < 0.001). ACR50 response rates in the leflunomide/methotrexate and placebo/methotrexate groups were a respective 26.2% and 6.0% (P < 0.001), and ACR70 response rates were 10.0% and 2.3% (P = 0.016). These results suggest that in patients with active RA, combination therapy with leflunomide and methotrexate provides statistically significant and clinically meaningful improvements compared with the combination of placebo and methotrexate. Furthermore, combination therapy appears to be an alternative for patients who have an incomplete response to methotrexate alone. Comparative Efficacy of Leflunomide in Meta-Analyses and Pooled Analyses A study of pooled clinical efficacy data from 6 randomized controlled trials showed leflunomide to be significantly better than placebo, with improvements in ACR20 response rate roughly 2 times that with placebo at both 6 months (relative benefit [RB] = 1.93; 95% CI, 1.51–2.47) and 12 months (RB = 1.99; 95% CI, 1.42–2.77).36 Other clinical measures of disease activity, functional measures, and radiographic scores were also significantly better in leflunomide recipients than in placebo recipients (all, P < 0.05). In another analysis,30 leflunomide was not significantly superior to methotrexate on most outcome measures, with the exception of scores on the modified HAQ (weighted mean difference, –0.14; 95% CI, –0.25 to –0.03), Problem Elicitation Technique scores (weighted mean difference, –3.5; 95% CI, –5.62 to –1.38), and scores on the physical component of the SF-36 (weighted mean difference, –3.0; 95% CI, –5.41 to –0.59). When leflunomide was compared with sulfasalazine,31 there were no statistically significant differences on most clinical outcomes, although leflunomide produced significantly greater ACR20 response rates at 24 months compared with sulfasalazine (RB = 1.37; 95% CI, 1.07–1.75). The effects of leflunomide on retarding radiographic E.K. Li et al. deterioration were not significantly different from those of sulfasalazine and methotrexate,37 although the results of the study by Strand et al26 tended to favor leflunomide over methotrexate, and the results of the study by Emery et al28 favored methotrexate. Thus, based on the available information, leflunomide is not superior to either methotrexate or sulfasalazine in delaying bone erosion or joint damage in patients with RA. SAFETY PROFILE In a recent systematic review and meta-analysis,36 significantly fewer patients receiving leflunomide withdrew from treatment compared with those receiving placebo (RR = 0.70; 95% CI, 0.59–0.83). Rates of withdrawal were similar with leflunomide and sulfasalazine, but more patients discontinued leflunomide than methotrexate (RR = 1.26; 95% CI, 1.08–1.48). Adverse effects were more commonly reported with leflunomide than with methotrexate (RR = 1.43; 95% CI, 1.13–1.83) or placebo (RR, 2.73; 95% CI, 1.67–4.47), but occurred with a similar incidence in the leflunomide and sulfasalazine groups (RR, 0.77; 95% CI, 0.45–1.33). Another study showed a significantly lower rate of discontinuations with methotrexate than with leflunomide (P = 0.001), which had a similar rate to sulfasalazine.38 However, the overall incidence of discontinuations due to adverse effects was lower with leflunomide than with methotrexate or sulfasalazine; discontinuations related to toxicity occurred significantly earlier with leflunomide and sulfasalazine than with methotrexate (P < 0.001). The most common adverse effect associated with leflunomide treatment was diarrhea, reported in 17% of patients.27 Diarrhea was usually transient or responded to a decrease in the leflunomide dose. Nausea occurred in 9.3% of patients, abdominal pain in 4.6%, and vomiting in 2.8%. The frequency and severity of these symptoms were highest during the first 2 weeks of treatment and declined thereafter. Postmarketing surveillance has revealed rare cases of serious hepatic injury, sometimes with a fatal outcome. Most cases have occurred within 6 months of starting therapy in patients with multiple risk factors for hepatotoxicity (eg, liver disease, use of other hepatotoxic drugs).39 There are also reports of elevations in aminotransferases in 15% of leflunomide recipients, compared with 12% for methotrexate and 3% for placebo.26,27 Most of these elevations were <2 times the upper limit of normal and were transient or resolved with dose reduction.11,26,28,40 An elevation in transaminases >2 times the upper limit of normal occurred in 2.3% to 11.0% of patients in Phase III trials,26 a similar rate to that seen when folate supplementation was added to methotrexate therapy.28 Three of 108 patients in one study had an elevation in liver enzymes >3 times the upper limit of normal.40 Literature review revealed only a single case of severe liver disease (early cirrhosis) in a patient who had received the combination of leflunomide and methotrexate.41 It was not possible, however, to state whether the liver injury was the effect of either drug alone or of the combination. Liver enzymes must be checked at baseline and monitored at monthly intervals during the first 6 months of leflunomide therapy. If values are stable, monitoring is recommended every 6 to 8 weeks. If values increase <2-fold, the dose should be decreased to 10 mg/d. For persistent elevations 2 to 3 times the upper limit of normal after dose reduction, liver biopsy is recommended if leflunomide therapy is to be continued.42 As discussed earlier, the active metabolite of leflunomide is highly protein bound and is cleared by hepatic metabolism and biliary secretion. Use of leflunomide is not recommended in patients with a history of hepatitis, in those who consume large amounts of alcohol,31,43 or in patients with hypoproteinemia or impaired liver function caused by any stage of cirrhosis. Several other adverse effects can occur with leflunomide. Reversible alopecia or rash has been reported in 8% of patients, and mild weight loss and headache have been noted in some patients.27,28,44–46 Significant leukopenia or thrombocytopenia has not been reported, with the exception of 20 cases of leukopenia in 501 patients over a 2-year period in one study28; none of the leukocyte counts were <2 2 109 white cells/L. Ten cases of pancytopenia have been reported, in 9 of which the patient was taking concomitant methotrexate therapy.47,48 Based on the Naranjo probability scale, the relationship between leflunomide and pancytopenia was probable in 1 case47 and possible in 9 others.48 For this reason, routine hematologic monitoring is suggested in patients taking leflunomide and methotrexate.48 Bone marrow 455 CLINICAL THERAPEUTICS® suppression is a concern, and interruption of therapy may be necessary if a serious infection occurs. Monitoring of the white blood cell count, platelets, and hemoglobin is recommended at the same time intervals as for liver enzyme monitoring.39 Apart from 1 case of nonfatal sepsis,26 no increases in the incidence of opportunistic exogenous infections such as Pneumocystis carinii pneumonia, uncontrolled endogenous infections, or reactivation or dissemination of varicella zoster virus have been reported in the literature.33 This lack of infections reflects the ability of replicating cells in the gastrointestinal and hematopoietic systems to meet their need for pyrimidine ribonucleotides for DNA synthesis through the salvage pathways, independent of de novo nucleotide pathways or DHODH. Unlike methotrexate, which may cause hypersensitivity pneumonitis in 0.7% to 7.0% of patients,49 leflunomide has not been associated with pulmonary toxicity.33 Drug-related hypertension has been noted in patients receiving leflunomide at an incidence ranging from 1.1% to 6.8%.26,28,40,50 In one study,50 30 patients receiving leflunomide had significant increases in both systolic and diastolic blood pressure (P < 0.05); the mechanism is unclear but may be related to an increase in sympathetic drive. Another possible mechanism for a blood pressure increase with leflunomide may involve displacement of the free fraction of concomitant diclofenac or ibuprofen from protein binding, increasing the NSAID’s effect on the distribution of renal blood flow and potentially increasing retention of salt and water.11 A pharmacokinetic study found that steady-state concentrations of A77 1726 in plasma were not affected by renal impairment or hemodialysis.51 Therefore, reduction of the leflunomide dose does not appear to be necessary in patients with chronic renal failure undergoing hemodialysis. Experience with leflunomide in patients with renal impairment is limited. Although leflunomide is not contraindicated, it should be used with caution in patients with any degree of renal insufficiency, as the kidneys play a role in excretion of the metabolite.35 One case of glomerulonephritis leading to renal failure has been reported in a patient with RA; antiglomerular basement membrane antibodies were detected shortly after the initiation of leflunomide.52 Although leflunomide’s exact role in this case of renal failure 456 was unknown, the close temporal relationship between drug administration and the adverse event suggest a causal association. The incidence of rheumatoid vasculitis is not increased with leflunomide compared with methotrexate. Nonetheless, vasculitis has been reported in 2 patients receiving leflunomide, both of whom died.53 Interestingly, leflunomide has been effective in maintaining remission of Wegener’s granulomatosis.54 Leflunomide can potentiate the anticoagulant effect of warfarin, resulting in an increase in the international normalized ratio of warfarin in patients taking warfarin. A77 1726 inhibits the cytochrome P450 (CYP) 2C9 isozyme and can increase the bioavailability of drugs metabolized by this isozyme, such as warfarin,55 phenytoin, tolbutamide, and rifampicin (a nonspecific CYP inducer). Leflunomide has no significant interactions with oral contraceptives, NSAIDs, or cimetidine. Although no animal studies have been conducted to evaluate potential fetal damage, leflunomide is considered teratogenic, and its use is contraindicated in women who are pregnant or who wish to become pregnant.10 Before pregnancy is considered, plasma concentrations of A77 1726 should be <0.2 mg/L on 2 occasions ≥14 days apart. In the event of pregnancy or toxicity, the elimination of leflunomide can be hastened by administering cholestyramine 8 g TID for 11 days. Leflunomide should not be used during lactation, because it is not known whether leflunomide is excreted in breast milk. Similarly, use of leflunomide is not recommended in men who wish to father a child.26,45 There is no evidence of an increased incidence of lymphoproliferative disorders or malignancies in patients receiving leflunomide for RA.10 The incidence has been similar to that in patients receiving methotrexate or sulfasalazine. CONCLUSIONS Leflunomide is a synthetic isoxazole derivative that acts predominantly on activated lymphocytes and prevents them from initiating the inflammatory processes that lead to cartilage and bone destruction in RA. In the trials reviewed, leflunomide appeared to improve all clinical outcomes and to delay radiologic progression of RA compared with placebo. After 5 years of treatment, its efficacy and adverse-event profile were comparable to those of sulfasalazine and methotrexate. E.K. Li et al. REFERENCES 1. Bartlett RR, Schleyerbach R. Immunopharmacological profile of a novel isoxazol derivative, HWA 486, with potential antirheumatic activity—I. Disease modifying action on adjuvant arthritis of the rat. Int J Immunopharmacol. 1985;7:7–18. 2. Popovic S, Bartlett RR. The use of the murine chronic graft vs host (CGVH) disease, a model for systemic lupus erythematosus (SLE), for drug discovery. Agents Actions. 1987;21:284–286. 3. Popovic S, Bartlett RR. Disease modifying activity of HWA 486 on the development of SLE in MRL/1-mice. Agents Actions. 1986;19:313–314. 4. Glant TT, Mikecz K, Bartlett RR, et al. Immunomodulation of proteoglycan-induced progressive polyarthritis by leflunomide. Immunopharmacology. 1992; 23:105–116. 5. Stosic-Grujicic S, Dimitrijevic M, Bartlett RR. A novel immunomodulating agent—leflunomide inhibits experimental autoimmune diabetes in mice. Transplant Proc. 1996;28:3072–3073. 6. Thoenes GH, Sitter T, Langer KH, et al. Leflunomide (HWA 486) inhibits experimental autoimmune tubulointerstitial nephritis in rats. Int J Immunopharmacol. 1989;11:921–929. 7. Ogawa T, Inazu M, Gotoh K, et al. Therapeutic effects of leflunomide, a new antirheumatic drug, on glomerulonephritis induced by the antibasement membrane antibody in rats. Clin Immunol Immunopathol. 1991;61:103–118. 8. Vidic-Dankovic B, Kosec D, Damjanovic M, et al. Leflunomide prevents the development of experimentally induced myasthenia gravis. Int J Immunopharmacol. 1995;17:273–281. 9. Mladenovic V, Domljan Z, Rozman B, et al. Safety and effectiveness of leflunomide in the treatment of patients with active rheumatoid arthritis. Results of a randomized, placebo-controlled, phase II study. Arthritis Rheum. 1995;38:1595–1603. 10. Leflunomide [package insert]. Bridgewater, NJ: Aventis Pharmaceuticals Inc; 2000. 11. Fox RI, Herrmann ML, Frangou CG, et al. Mechanism of action for leflunomide in rheumatoid arthritis. Clin Immunol. 1999;93:198–208. 12. Cherwinski HM, McCarley D, Schatzman R, et al. The immunosuppressant leflunomide inhibits lymphocyte progression through cell cycle by a novel mechanism. J Pharmacol Exp Ther. 1995;272:460–468. 13. Greene S, Watanabe K, Braatz-Trulson J, Lou L. Inhibition of dihydroorotate dehydrogenase by the immunosuppressive agent leflunomide. Biochem Pharmacol. 1995;50:861–867. 14. Fairbanks LD, Bofill M, Ruckemann K, Simmonds HA. Importance of ribonucleotide availability to proliferating T-lymphocytes from healthy humans. Disproportionate expansion of pyrimidine pools and contrasting effects of de novo synthesis inhibitors. J Biol Chem. 1995;270:29682–29689. 15. Cherwinski HM, Cohn RG, Cheung P, et al. The immunosuppressant leflunomide inhibits lymphocyte proliferation by inhibiting pyrimidine biosynthesis. J Pharmacol Exp Ther. 1995;275:1043–1049. 16. Fox RI. Mechanism of action of leflunomide in rheumatoid arthritis. J Rheumatol Suppl. 1998;53: 20–26. 17. Cao WW, Kao PN, Aoki Y, et al. A novel mechanism of action of the immunomodulatory drug, leflunomide: Augmentation of the immunosuppressive cytokine, TGF-beta 1, and suppression of the immunostimulatory cytokine, IL-2. Transplant Proc. 1996;28:3079– 3080. 18. Xu X, Blinder L, Shen J, et al. In vivo mechanism by which leflunomide controls lymphoproliferative and autoimmune disease in MRL/MpJ-lpr/lpr mice. J Immunol. 1997;159:167–174. 19. Manna SK, Aggarwal BB. Immunosuppressive leflunomide metabolite (A77 1726) blocks TNF-dependent nuclear factor–kappa B activation and gene expression. J Immunol. 1999;162:2095–2102. 20. Deage V, Burger D, Dayer JM. Exposure of T lymphocytes to leflunomide but not to dexamethasone favors the production by monocytic cells of interleukin-1 receptor antagonist and the tissue-inhibitor of metalloproteinases-1 over that of interleukin-1beta and metalloproteinases. Eur Cytokine Netw. 1998;9:663–668. 21. Ruckemann K, Fairbanks LD, Carrey EA, et al. Leflunomide inhibits pyrimidine de novo synthesis in mitogen-stimulated T-lymphocytes from healthy humans. J Biol Chem. 1998;273:21682–21691. 22. Kraan MC, Reece RJ, Barg EC, et al. Modulation of inflammation and metalloproteinase expression in synovial tissue by leflunomide and methotrexate in patients with active rheumatoid arthritis. Findings in a prospective, randomized, double-blind, paralleldesign clinical trial in thirty-nine patients at two centers. Arthritis Rheum. 2000;43:1820–1830. 457 CLINICAL THERAPEUTICS® 23. Schorlemmer HU, Kurrle R, Schleyerbach R. Leflunomide’s active metabolite A77-1726 and its derivatives, the malononitrilamides, inhibit the generation of oxygen radicals in mononuclear phagocytes. Int J Immunother. 1998;14:213–222. 24. Kraan MC, de Koster BM, Elferink JG, et al. Inhibition of neutrophil migration soon after initiation of treatment with leflunomide or methotrexate in patients with rheumatoid arthritis: Findings in a prospective, randomized, double-blind clinical trial in fifteen patients. Arthritis Rheum. 2000;43:1488–1495. 25. Hamilton LC, Vojnovic I, Warner TD. A771726, the active metabolite of leflunomide, directly inhibits the activity of cyclo-oxygenase-2 in vitro and in vivo in a substrate-sensitive manner. Br J Pharmacol. 1999;127: 1589–1596. 26. Strand V, Cohen S, Schiff M, et al, for the Leflunomide Rheumatoid Arthritis Investigators Group. Treatment of active rheumatoid arthritis with leflunomide compared with placebo and methotrexate. Arch Intern Med. 1999;159:2542–2550. 27. Smolen JS, Kalden JR, Scott DL, et al, for the European Leflunomide Study Group. Efficacy and safety of leflunomide compared with placebo and sulphasalazine in active rheumatoid arthritis: A double-blind, randomised, multicentre trial. Lancet. 1999;353: 259–266. 28. Emery P, Breedveld FC, Lemmel EM, et al. A comparison of the efficacy and safety of leflunomide and methotrexate for the treatment of rheumatoid arthritis. Rheumatology (Oxford). 2000;39:655–665. 29. Felson DT, Anderson JJ, Boers M, et al. American College of Rheumatology. Preliminary definition of improvement in rheumatoid arthritis. Arthritis Rheum. 1995;38:727–735. 30. Cohen S, Cannon GW, Schiff M, et al, for the Utilization of Leflunomide in the Treatment of Rheumatoid Arthritis Trial Investigator Group. Twoyear, blinded, randomized, controlled trial of treatment of active rheumatoid arthritis with leflunomide compared with methotrexate. Arthritis Rheum. 2001;44: 1984–1992. 31. Scott DL, Smolen JS, Kalden JR, et al. Treatment of active rheumatoid arthritis with leflunomide: Two year follow up of a double blind, placebo controlled trial versus sulfasalazine. Ann Rheum Dis. 2001;60:913–923. 32. Kalden JR, Scott DL, Smolen JS, et al, for the European Leflunomide Study Group. Improved functional ability 458 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. in patients with rheumatoid arthritis—longterm treatment with leflunomide versus sulfasalazine. J Rheumatol. 2001;28:1983–1991. Kalden JR, Schattenkirchner M, Sorensen H, et al. The efficacy and safety of leflunomide in patients with active rheumatoid arthritis: A five-year followup study. Arthritis Rheum. 2003;48:1513–1520. Weinblatt ME, Kremer JM, Coblyn JS, et al. Pharmacokinetics, safety, and efficacy of combination treatment with methotrexate and leflunomide in patients with active rheumatoid arthritis. Arthritis Rheum. 1999;42:1322–1328. Kremer JM, Genovese MC, Cannon GW, et al. Concomitant leflunomide therapy in patients with active rheumatoid arthritis despite stable doses of methotrexate. A randomized, double-blind, placebocontrolled trial. Ann Intern Med. 2002;137:726–733. Osiri M, Shea B, Robinson V, et al. Leflunomide for the treatment of rheumatoid arthritis: A systematic review and metaanalysis. J Rheumatol. 2003;30: 1182–1190. Sharp JT, Strand V, Leung H, et al, for the Leflunomide Rheumatoid Arthritis Investigators Group. Treatment with leflunomide slows radiographic progression of rheumatoid arthritis: Results from three randomized controlled trials of leflunomide in patients with active rheumatoid arthritis [published correction appears in Arthritis Rheum. 2000;43:1345]. Arthritis Rheum. 2000;43:495–505. Aletaha D, Stamm T, Kapral T, et al. Survival and effectiveness of leflunomide compared with methotrexate and sulfasalazine in rheumatoid arthritis: A matched observational study. Ann Rheum Dis. 2003;62:944– 951. 2003 Safety alerts for drugs, biologics, medical devices, and dietary supplements. Available at: http://www.fda. gov/medwatch/SAFETY/2003/safety03.htm#arava. Accessed November 23, 2003. Smolen JS. Efficacy and safety of the new DMARD leflunomide: Comparison to placebo and sulfasalazine in active rheumatoid arthritis. Scand J Rheum Suppl. 1999;112:15–21. Weinblatt ME, Dixon JA, Falchuk KR. Serious liver disease in a patient receiving methotrexate and leflunomide. Arthritis Rheum. 2000;43:2609–2611. Sanders S, Harisdangkul V. Leflunomide for the treatment of rheumatoid arthritis and autoimmunity. Am J Med Sci. 2002;323:190–193. E.K. Li et al. 43. Kremer JM. Rational use of new and existing diseasemodifying agents in rheumatoid arthritis. Ann Intern Med. 2001;134:695–706. 44. Herrmann ML, Schleyerbach R, Kirschbaum BJ. Leflunomide: An immunomodulatory drug for the treatment of rheumatoid arthritis and other autoimmune diseases. Immunopharmacology. 2000;47:273–289. 45. Wallace CA. On beyond methotrexate treatment of severe juvenile rheumatoid arthritis. Clin Exp Rheumatol. 1999;17:499–504. 46. Strand V. Approaches to the management of systemic lupus erythematosus. Curr Opin Rheumatol. 1997;9:410–420. 47. Auer J, Hinterreiter M, Allinger S, et al. Severe pancytopenia after leflunomide in rheumatoid arthritis. Acta Medica Austriaca. 2000;27:131–132. 48. Hill RL, Topliss DJ, Purcell PM. Pancytopenia associated with leflunomide and methotrexate. Ann Pharmacother. 2003;37:149. 49. Alarcon GS. Methotrexate: Its use for the treatment of rheumatoid arthritis and other rheumatic disorders. 50. 51. 52. 53. 54. 55. In: Koopman WJ, ed. Arthritis and Allied Conditions. 14th ed. Baltimore, Md: Williams & Wilkins; 2001: 743–768. Rozman B, Praprotnik S, Logar D, et al. Leflunomide and hypertension. Ann Rheum Dis. 2002;61:567–569. Beaman JM, Hackett LP, Luxton G, Illett KF. Effect of hemodialysis on leflunomide plasma concentrations. Ann Pharmacother. 2002;36:75–77. Bruyn GA, Veenstra RP, Halma C, Grond J. Antiglomerular basement membrane antibody–associated renal failure in a patient with leflunomide-treated rheumatoid arthritis. Arthritis Rheum. 2003;48:1164– 1165. Bruyn GA, Griep EN, Korff KJ. Leflunomide for active rheumatoid arthritis. Lancet. 1999;353:1883. Gross WL. New concepts in treatment protocols for severe systemic vasculitis. Curr Opin Rheumatol. 1999; 11:41–46. Lim V, Pande I. Leflunomide can potentiate the anticoagulant effect of warfarin. BMJ. 2002;325:1333. Address correspondence to: Edmund K. Li, FRCP, Department of Medicine & Therapeutics, The Chinese University of Hong Kong, Prince of Wales Hospital, Hong Kong. E-mail: edmundli@cuhk.edu.hk 459