2003-241.aug

Combination Leflunomide and Methotrexate (MTX)Therapy for Patients with Active Rheumatoid ArthritisFailing MTX Monotherapy: Open-Label Extension of aRandomized, Double-Blind, Placebo Controlled TrialJOEL KREMER, MARK GENOVESE, GRANT W. CANNON, JACQUES CALDWELL, JOHN CUSH, DANIEL E. FURST, MICHAEL LUGGEN, ED KEYSTONE, JOAN BATHON, ARTHUR KAVANAUGH, ERIC RUDERMAN, PATRICIA COLEMAN, DAVID CURTIS, ELLIOTT KOPP, SETH KANTOR, MICHAEL WEISMAN, JONATHAN WALTUCK, HERBERT B. LINDSLEY, JOSEPH MARKENSON, BRUCE CRAWFORD, INDRA FERNANDO, KAREN SIMPSON, and VIBEKE STRAND ABSTRACT. Objective. To obtain additional safety and efficacy data on leflunomide (LEF) treatment in combi-
nation with methotrexate (MTX) therapy in an open-label extension study in patients with rheuma-
toid arthritis (RA).
Methods. Following a 24 week, randomized, double-blind trial of adding placebo (PLA) or LEF to
stable MTX therapy, patients could enter a 24 week extension. Subjects randomized to LEF and
MTX continued treatment [(LEF/LEF) + MTX]. Subjects randomized to PLA and MTX switched to
LEF (10 mg/day, no loading dose) and MTX [(PLA/LEF) + MTX]. The double-blind regarding
initial randomization was maintained.
Results. For subjects in the extension phase, American College of Rheumatology 20% (ACR20)
responder rates for the (LEF/LEF) + MTX group were maintained from Week 24 (57/96, 59.4%) to
Week 48 (53/96, 55.2%). ACR20 responder rates improved in patients switched to LEF from PLA
at Week 24 [(PLA/LEF) + MTX] from 25.0% (24/96) at Week 24 to 57.3% (55/96) at Week 48.
Patients in the extension who switched from PLA to LEF without a loading dose exhibited a lower
incidence of elevated transaminases compared to patients initially randomized to LEF. Diarrhea and
nausea were less frequent during the open-label extension in patients who did not receive a LEF
loading dose.
Conclusion. Response to therapy was maintained to 48 weeks of treatment in patients who continued
to receive LEF and MTX during the extension. Importantly, ACR20 response rates after 24 weeks
of LEF therapy were similar between patients switched from PLA to LEF without loading dose, and
those who received a loading does of LEF (100 mg/day × 2 days) at randomization. Fewer adverse
events were reported in patients switched to LEF without a loading dose. (J Rheumatol
2004;31:1521–31)
Key Indexing Terms:RHEUMATOID ARTHRITIS LEFLUNOMIDE METHOTREXATE OPEN-LABEL TRIAL COMBINATION DISEASE MODIFYING ANTIRHEUMATIC DRUG THERAPY Rheumatoid arthritis (RA) is a progressive inflammatory antirheumatic drugs (DMARD) has become the standard for disease of unknown etiology that causes severe disability1,2 treatment of RA; however, an incomplete response to and increases mortality2,3. Early use of disease modifying DMARD monotherapy is observed in some patients4-6.
From the Center for Rheumatology, Albany, New York, USA. Lansing, MI; D. Curtis, MD, California Pacific Medical Center, San Supported by Aventis Pharmaceuticals. Francisco, CA; E. Kopp, MD, CARE Center, Raleigh, NC; S. Kantor, MD,Ohio State University, Columbus, OH; M. Weisman, MD, Cedars-Sinai J. Kremer, MD, Center for Rheumatology, Albany, NY; M. Genovese, MD, Medical Center, UCLA School of Medicine, Los Angeles, CA; J. Waltuck, Stanford University Medical Center, Stanford, CA; G.W. Cannon, MD, MD, The Emory Clinic, Atlanta, GA; H.B. Lindsley, MD, Kansas Veterans Affairs Medical Center and University of Utah, Salt Lake City, University Medical Center, Kansas City, KA; J. Markenson, MD, for the UT; J. Caldwell, MD, Florida Arthritis and Allergy Institute, Daytona Study 4001 LEF+MTX Study Group, Hospital for Special Surgery, New Beach, FL; J. Cush, MD, Presbyterian Hospital, Dallas, TX; D.E. Furst, York, NY; B. Crawford, MA, MPH, Mapi Values, Boston, MA; MD, Virginia Mason Research Center, Seattle, WA; M. Luggen, MD, I. Fernando, PhD, Quintiles, Inc., Kansas City, KA; K. Simpson, MD, University of Cincinnati Medical Center, Cincinnati, OH, USA; FACP, Aventis Pharmaceuticals, Bridgewater, NJ; V. Strand, MD, E. Keystone, MD, University of Toronto, Toronto, Canada; J. Bathon, Stanford University School of Medicine, Palo Alto, CA, USA. MD, The Johns Hopkins University School of Medicine, Baltimore, MD;A. Kavanaugh, MD, University of California San Diego School of Address reprint requests to Dr. J.M. Kremer, The Center for Medicine, Division of Rheumatology, Allergy, and Immunology, San Rheumatology, LLP 1367 Washington Avenue, Suite 101, Albany, NY Diego, CA; E. Ruderman, MD, Northwestern University Feinberg School 12206. E-mail: jkremer@joint-docs.com of Medicine, Chicago, IL; P. Coleman, MD, Good Clinical Practice, East Submitted March 20, 2003; revision accepted February 4, 2004. Personal non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved
Personal, non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved.
Kremer, et al: Combination therapy for RA Clinicians now recognize that earlier, more aggressive treat- the protocol. The total duration of therapy, including both double-blind and ment with DMARD is essential for improving the signs and open-label phases, was 48 weeks. All patients who entered the open-labelphase received LEF; patients who were initially randomized to receive LEF symptoms of RA, slowing the progression of the disease, in the double-blind trial continued LEF in addition to their stable dose of and maintaining physical function7-16. This approach is MTX in the open-label phase [termed (LEF/LEF) + MTX], while patients supported by the fact that worsening of the physical compo- initially randomized to receive PLA in the double-blind phase switched to nents of the Health Assessment Questionnaire Disability LEF in the open-label phase in addition to their stable MTX [termed Index (HAQ DI) was observed with standard-care DMARD (PLA/LEF) + MTX]. Although this was an open-label extension, thedouble blind regarding initial randomization to LEF or PLA was main- monotherapy as well as with some combinations of During the open-label extension, patients were started on LEF 10 For the treatment of RA, methotrexate (MTX) is the mg/day in combination with background MTX, regardless of the final dose DMARD most widely used as both monotherapy and in of LEF or PLA at the endpoint of the double-blind portion of the trial. At combination therapy14,18,19. Because in many patients MTX the discretion of the investigator, the dose of LEF could be adjusted to 10mg every other day for tolerability, or to 20 mg/day if 10 mg/day was toler- alone does not adequately control the signs and symptoms ated and active disease persisted. Patients receiving PLA in the double- of RA at tolerated doses, the practice of combination blind phase did not receive a loading dose when beginning LEF at Week 24.
DMARD therapy has increased10,13,20 in an attempt to gain In contrast, patients randomized to the LEF + MTX group during the efficacy while managing toxicity20. A 1997 survey found double-blind phase (Weeks 0–24) had been given a loading dose of LEF that 99% of responding rheumatologists prescribed combi- Patients continuing on LEF in the open-label phase [(LEF/LEF) + nation DMARD therapy in an estimated 24% of all patients MTX] maintained their stable background MTX therapy and were observed for maintenance of effect and any safety issues occurring during MTX has been used in combination with many drugs, the second 24 weeks of combination therapy. Patients switching from PLA including sulfasalazine10,17, sulfasalazine and hydroxy- in the double-blind phase to LEF in the open-label phase [(PLA/LEF) + chloroquine12,22, cyclosporine15,23,24, auranofin25, azathio- MTX] also maintained their stable background MTX therapy and wereobserved for an incremental therapeutic benefit following initiation of LEF, prine26, etanercept27, infliximab28,29, and anakinra30. Limited as well as for additional safety issues during the 24 week open-label phase.
data in abstract form are available on combination therapy Study population. Patients were male or female (age 18 to 75 years; ≥ 19 with leflunomide (LEF) and cyclosporine31, infliximab32, years old in Canada) and diagnosed with RA ≥ 6 months prior to enrollment and sulfasalazine33. In both an open-label trial and a double- in the initial double-blind study. All patients enrolled in the double-blind blind trial (described below), MTX combined with LEF study had active RA, determined at 2 separate examinations 7 to 21 daysapart, despite MTX treatment for at least 6 months (15–20 mg/week or demonstrated a substantial incremental benefit in patients 10–15 mg/week if this was the maximum tolerated dose for the subject).
with RA who had an inadequate response to MTX alone34-37.
Active disease was defined by 3 of 4 criteria: ≥ 6 swollen joints, ≥ 9 tender In a 30-patient open-label pilot study of LEF added to joints, ≥ 45 minutes of morning stiffness, and erythrocyte sedimentation MTX in patients with inadequate response to MTX alone, more than half the patients met the American College of Efficacy endpoints. ACR20 response was defined as at least a 20% Rheumatology 20% (ACR20) response criteria after 36 improvement in tender joint count (TJC) and swollen joint count (SJC), and weeks of therapy, a response rate that was sustained at Week in 3 of 5 of the following measures: physician global assessment, patientglobal assessment, pain intensity assessment, HAQ DI, and an acute phase 4835,36. This LEF + MTX combination was generally well reactant [ESR or C-reactive protein (CRP)]. TJC and SJC were based on 68 tolerated, although an increased risk of elevated hepatic and 66-joint assessments, respectively. Physician and patient global assess- enzymes was observed35,36. No pharmacokinetic interactions ments of RA disease activity were based on a 0 to 100 mm horizontal visual between LEF and MTX were identified35.
analog scale (VAS). ACR50 and ACR70 responses were defined by at least In a randomized, double-blind, placebo controlled trial, 50% and at least 70% improvement, respectively, using the same criteria.
The primary efficacy variable for the intent-to-treat (ITT) population in LEF was added in patients with active RA despite MTX the 24 week, double-blind phase was ACR20 responder-at-endpoint rate at treatment. Adding LEF provided substantial therapeutic Week 24, which required both study completion and ACR20 response at benefit compared with adding placebo (PLA) and it was Week 2434. Patients who discontinued prior to endpoint, or for whom there generally well tolerated34,37. Elevations in liver function were insufficient data to assess ACR20, were considered nonresponders forthis analysis. In patients entering the open-label extension, the ACR20 tests were reversible with discontinuation, dose reduction, responder-at-endpoint rates at Weeks 24 and 48 were assessed to identify or, in mild cases, often with no change in dose34. Our objec- trends with continued LEF + MTX combination therapy or following a tive was to obtain additional descriptive efficacy and safety switch to combination LEF + MTX therapy. data on the combination of LEF and MTX in a 24 week, Secondary efficacy variables assessed in the double-blind phase were open-label extension of the double-blind, placebo controlled also assessed in the open-label phase and included the ACR50 and ACR70responder-at-endpoint rates at Weeks 24 and 48. ACR20, ACR50, and trial34, in patients taking a stable background dose of MTX.
ACR70 response rates at Weeks 24 and 48 were also analyzed using lastobservation carried forward (LOCF) for patients in the open-label phase MATERIALS AND METHODS
who discontinued prior to Week 48. The open-label phase also assessed Study design. Patients who completed a 24 week, randomized, double-blind mean changes from baseline to Weeks 24 and 48 for individual ACR trial of adding PLA or LEF to stable MTX therapy34 were allowed to enter components and rheumatoid factor (RF). Physical function was assessed by an additional 24 week, open-label, multicenter extension of the study change in HAQ DI from baseline to Week 24 and Week 48. Health related (Figure 1). An institutional review board at each investigative site approved quality of life (HRQoL) was assessed by change in the 36 item Medical Personal non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved
Personal, non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved.
The Journal of Rheumatology 2004; 31:8 Figure 1. Study design of the double-blind and open-label phases of the trial. *Active disease defined as 3 of thefollowing 4: ≥ 9 tender joints; ≥ 6 swollen joints; ≥ 45 minutes morning stiffness; ESR ≥ 28 mm/h.
Outcome Study Short Form Health Survey (SF–36), including the 8 SF–36 MTX; n = 86, or 89% (LEF/LEF) + MTX]. For both groups, domains and mental (MCS) and physical (PCS) component summary the rate of discontinuation appeared to be lower during the scores from baseline to Week 24 and Week 4835.
open-label phase [14.6% for (PLA/LEF) + MTX and 10.4% Safety assessment. Safety was assessed by adverse event (AE) reports, for (LEF/LEF) + MTX] compared to the initial double- physical examinations, and clinical laboratory data. Adverse events weredefined as any sign, symptom, syndrome, or illness that appeared or wors- blind phase (24.8% for PLA + MTX, and 23.1% for LEF + ened during the open-label extension, and that might have impaired the MTX). The most common reason for withdrawal in the open-label phase was the occurrence of AE [2.1% for the Hematology and blood chemistry tests, including liver function tests (PLA/LEF) + MTX group; 5.2% for the (LEF/LEF) + MTX (LFT) examining alanine aminotransferase (ALT) and aspartate amino- transferase (AST), were performed at each study visit. Study visits occurredat Weeks 24 and 48 or at early termination and included a followup visit 6 Clinical and demographic characteristics at baseline for weeks after completion. Clinically significant ALT and AST abnormalities the study population (ITT) have been reported for the double- could have resulted in a reduction or discontinuation of LEF, depending on blind trial34. At the beginning of the double-blind study, the degree and persistence of the elevation31. Occurrence and reversal of the patients had a mean RA duration of 10.5 and 12.7 years in the highest elevation of ALT or AST [> 1.2 to ≤ 2 × upper limits of normal LEF + MTX and PLA + MTX groups, respectively. Although (ULN); > 2 to ≤ 3 × ULN; and > 3 × ULN] were summarized.
radiographic assessment was originally planned as part of the Statistical analysis. Descriptive statistics were used for all efficacy andsafety variables. The treatment groups in the open-label phase are not study, not enough radiographs were completed to make any directly comparable; therefore, no statistical comparisons were made.
analysis or comparison between groups.
ACR20, ACR50, and ACR70 response rates based on both responder-at- Of the 96 patients who received LEF during the initial endpoint and LOCF approaches were provided for each treatment group.
double-blind study and continued in the Week 24–48 open- McNemar’s test was used to compare the ACR20 responder-at-endpointrates at Weeks 24 and 48 within the (LEF/LEF) + MTX group and the label phase, 60 (62.5%) were taking 20 mg/day, 32 (33.3%) (PLA/LEF) + MTX group. Data for changes from baseline to endpoint for were taking 10 mg/day, and 4 (4.2%) were taking 10 mg individual ACR criteria are presented as the mean ± standard deviation every other day at Week 24. Of the 60 patients who were (SD). Safety variables were summarized on the safety-evaluable popula- taking LEF 20 mg/day at the endpoint of the double-blind tion, defined as all patients who received at least one dose of study medica-tion in the open-label extension.
phase, 48 decreased the dosage to 10 mg/day per protocol atthe beginning of the open-label phase, and 12 started the open-label phase on the 20 mg/day dosing schedule. During Patient disposition and demographics. Of the 263 patients the open-label phase, 25 patients increased their LEF dosage enrolled in the initial double-blind study, 200 completed the back to 20 mg/day. All 32 LEF + MTX patients who were initial 24 weeks of therapy; 192 of 200 (96%) patients on a LEF 10 mg/day dosing schedule at the endpoint of the eligible to participate in the open-label phase entered the double-blind phase started the open-label phase with 10 extension. One hundred sixty-eight patients completed the mg/day per protocol. Of the 4 LEF + MTX patients who extension study to Week 48 [n = 82, or 85% (PLA/LEF) + were at 10 mg every other day at the endpoint of the double- Personal non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved
Personal, non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved.
Kremer, et al: Combination therapy for RA blind phase, 2 started the open-label phase on 10 mg/day, ment), respectively. Although the minimum clinically and 2 remained on the same dosing schedule.
important difference (MCID) has not yet been formallydefined for SF–36, several authors39-42 have suggested that changes of 5 to 10 points in domains and 2.5 to 5 points in For the extension study, efficacy analyses included data summary scores are associated with meaningful clinical from those 192 patients who completed the initial 24 weeks improvements. Therefore, changes in PCS well exceeded of therapy and continued in the open-label phase of the the MCID and those for the MCS were within the range study. ACR20 completer-at-endpoint responses obtained in associated with clinical improvements.
the initial double-blind study are included where appropriate (PLA/LEF) + MTX. In the patients who switched from PLA for comparison34. It should be noted that there are differ- to LEF therapy while taking background MTX (n = 96), the ences in reported results at Week 24 for the double-blind ACR20 responder-at-endpoint rate was 25.0% at Week 24, ITT population (n = 263) and the open-label extension which increased to 57.3% at Week 48, a difference of population (n = 192). This is not an unexpected finding in an 32.3%, and was statistically different (p < 0.0001) (Table 1, extension study as the analysis from the double-blind study Figure 2). The 48 week responder rate approximated the likely includes some patients with lower responses to ACR20 responder-at-endpoint rate observed for the therapy who chose not to continue in the open-label exten- (LEF/LEF) + MTX patients at the same timepoint. The sion. Additional detail on efficacy results from the double- ACR50 and ACR70 responder-at-endpoint rates for (PLA/LEF) + MTX patients were 28.1% and 11.5% at Week (LEF/LEF) + MTX. In the patients who continued LEF 48, respectively, which were increased from the rates therapy on background MTX (n = 96), the ACR20 observed at Week 24 on PLA and approached the rates in the responder-at-endpoint rate was 59.4% at Week 24 and (LEF/LEF) + MTX group. Table 2 summarizes changes 55.2% at Week 48, a difference of –4.2% and not statisti- from baseline in individual ACR criteria and RF, and shows cally different (p = 0.4313) (Table 1, Figure 2). Similar improvements in CRP, TJC, and SJC, as well as in assess- trends were noted for ACR50 and ACR70 responder-at- ments for patient global, physician global, pain intensity, endpoint rates at Weeks 24 and 48 (Table 1), indicating and RF at Week 48 of a magnitude similar to that observed maintenance of effect across 48 weeks of combination therapy. Changes from baseline in individual ACR criteria There was a further improvement in the mean change in and RF are summarized in Table 2. Improvements were HAQ DI at Week 48 (–0.33) compared with that seen at observed in TJC and SJC, CRP, RF, and in patient global, Week 24 (–0.15; Table 2) in the (PLA/LEF) + MTX group, physician global and pain intensity assessments.
although the improvement at Week 48 did not reach that Table 2 also summarizes the improvements in physical seen in the (LEF/LEF) + MTX patients at Weeks 24 and 48.
function (HAQ DI) and HRQoL (SF–36 PCS and MCS).
Patients in the (PLA/LEF) + MTX group obtained a clini- The mean change of –0.52 in the HAQ DI at Week 24 was cally important improvement in SF–36 at 48 weeks, with maintained at Week 48 (–0.54). At baseline, 9.6% of improvements in the SF–36 PCS exceeding the MCID of 5 (LEF/LEF) + MTX patients had a HAQ DI score ≤ 0.5, the points and that for SF–36 MCS not exceeding MCID.
best category, which increased to 41.2% at Week 48. Meanchanges in the SF–36 PCS at Week 24 and Week 48 time- points were 8.5 for both (37% improvement), and those for A detailed safety analysis of the double-blind and open-label the SF–36 MCS were 4.5 and 4.2 (12% and 11% improve- studies was performed to assess the longterm safety profile Table 1. ACR responses in the double-blind phase (ITT population) and in the open-label extension.
PLA: placebo, LEF: leflunomide, MTX: methotrexate. * Data for double-blind phase as published34. † p < 0.0001 from McNemar’s test comparing Week 48vs Week 24 for the (PLA/LEF) + MTX group.
Personal non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved
Personal, non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved.
The Journal of Rheumatology 2004; 31:8 Figure 2. Proportion of ACR20 responders over time (by visit): A. Across Weeks 0–24 in the intent-to-treatpopulation (n = 263) (Kremer, et al 2002). B. Across Weeks 24–48 for subjects who continued in the open-label phase (n = 192).
of combination therapy [(LEF/LEF) + MTX to 48 weeks] tunistic), 2 were skin carcinoma (one was reported as being and to compare the AE profile between patients initiating related to study treatment). Three patients discontinued LEF therapy with and without a loading dose. Some data study medication due to AE, one each with intestinal perfo- from the double-blind study have been published34.
ration/sepsis, gastrointestinal hemorrhage, and atrial fibrilla- (LEF/LEF) + MTX. In the first 24 weeks the most tion. No deaths occurred in the open-label phase.
commonly reported AE (Table 3) in patients taking LEF + Mild to moderate decreases in leukocyte count and MTX (n = 130) were associated with the digestive system: neutrophil count were observed in Weeks 24–48 with diarrhea (25.4%) and nausea (16.2%). Rash (7.7%), (LEF/LEF) + MTX. As in the first 24 weeks, no patient exhib- alopecia (6.2%), and hypertension (4.6%) were also ited leukopenia < 2.0 103/mm3, neutropenia < 0.5 103/mm3, or reported. Fewer LEF + MTX patients (40.8%) had infec- low platelet count (below normal range of 140.0–440.0 tions than PLA + MTX patients (51.9%). No infection was 103/mm3. The mean change from baseline in hemoglobin was opportunistic, and no patient withdrew from treatment not significant (–0.02); 3.2% of patients had hemoglobin In patients continuing a second 24 weeks of treatment The mean increase in liver enzyme concentrations from with LEF + MTX (n = 96), diarrhea occurred at a lower rate baseline to Week 48 among patients continuing LEF into the (3.1%) compared with that in the first 24 weeks of treat- open-label phase (ALT 3.6 U/l; AST 3.7 U/l) was less than ment, as did alopecia (1.0%) and hypertension (2.1%); the that observed for the first 24 weeks of LEF treatment (ALT incidence of rash was similar (7.3%; Table 3). The incidence 9.3 U/l; AST 6.7 U/l). The overall incidence of ALT and of all infections combined in the open-label phase (35.4%) AST elevation over the 48 week period in the (LEF/LEF) + was similar to the incidence in the double-blind phase MTX group was 33.1% in the first 24 weeks and 19.2% in among patients continuing LEF. No increase in toxicity or new type of AE was apparent during the second 24 weeks of The occurrence of various degrees of LFT, based on a subject’s highest elevation for the first and second 24 weeks In the first 24 weeks, 11 LEF + MTX patients had 11 of treatment, is summarized in Table 5. As reported31, serious AE (Table 4); 2 of the 11 serious AE reported for adding LEF in patients tolerating background MTX LEF + MTX treated patients (one case of cellulitis and one increased the risk of liver enzyme elevation compared to diagnosis of breast carcinoma) were considered by the adding PLA, as seen in the Week 0–24 double-blind phase.
investigator to be at least possibly related to study treatment.
The incidence of ALT and AST elevations was lower in the Events leading to treatment discontinuation in more than second 24 weeks of combination treatment for patients one LEF + MTX treated subject were diarrhea (4 patients, continuing on combination therapy in the open-label phase, 3.1%), abnormal LFT (3 patients, 2.3%), and rash (2 compared with the first 24 weeks of treatment in the LEF + MTX group (ALT 13.7% and 31.5%; AST 6.3% and 16.9%, During the open-label phase, 15 (LEF/LEF) + MTX respectively). During the first 24 weeks, all ALT and AST patients had 21 serious AE: 3 were infections (none oppor- elevations in LEF + MTX patients normalized with no inter- Personal non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved
Personal, non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved.
Kremer, et al: Combination therapy for RA Table 2. Changes from baseline in individual efficacy measures at Weeks 24 and 48 (mean ± SD).
Mean change indicates the mean change from baseline. PLA: placebo, LEF: leflunomide, MTX: methotrexate,HAQ DI: Health Assessment Questionnaire Disability Index, PCS: Short-Form 36 physical component summaryscore, MCS: mental component summary score. *Open-label patients with non-missing values at baseline,double-blind phase endpoint, and open-label phase endpoint.
Personal non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved
Personal, non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved.
The Journal of Rheumatology 2004; 31:8 Table 3. Adverse events across treatment groups in Weeks 0–48.
vention, or a dose reduction, or discontinuation of study During the double-blind phase, 8 patients had 9 serious medication at or before the end of the study. Similarly, in the AE while taking PLA + MTX; 3 of the 9 were considered by open-label phase, ALT and AST elevations > 2 × ULN the investigator to be treatment related (one case each of normalized after LEF was reduced or discontinued. No pyogenic arthritis, gastritis, and cellulitis). Adverse events patient discontinued due to elevated LFT in the Week 24–48 leading to discontinuation in more than one PLA + MTX open-label phase. Mild elevations in ALT or AST (< 2 × treated patient during the double-blind phase were abnormal ULN) normalized after dose reduction or discontinuation in LFT (2 patients, 1.5%) and nausea (2 patients, 1.5%). After switching from placebo to LEF in the open-label phase, 13 (PLA/LEF) + MTX. In the first 24 weeks of PLA + MTX patients had 18 serious AE. No patient died during the open- treatment (n = 133), commonly reported AE were upper label phase. In the first 24 weeks, one subject taking PLA + respiratory infection (24.1%), diarrhea (13.5%), nausea MTX had an abnormally low platelet count (≥ 100.0 to < (11.3%), headache (8.3%), rash (8.3%), alopecia (3.8%), and 120.0 103/mm3), and 6.0% of patients had hemoglobin hypertension (3.0%). In the open-label, Week 24–48 exten- values < 10 g/dl. No clinically relevant decreases in leuko- sion phase (n = 96), when LEF treatment was initiated at cytes or neutrophils were observed. During the Week 24–48 Week 24 without a loading dose, diarrhea was the most phase, 5.2% of patients had hemoglobin levels < 10 g/dl.
common gastrointestinal event (16.7%). The incidence of Table 5 summarizes the number of patients with ALT or nausea during the first 24 weeks of combination therapy was AST elevations categorized by the patient’s highest value higher in the LEF + MTX patients in the double-blind study during the first and second 24 weeks of LEF treatment.
(received a loading dose) compared with patients in the Patients who switched from PLA to LEF for the second 24 (PLA/LEF) + MTX group who switched to LEF without a weeks of treatment without a loading dose exhibited an loading dose. Other common AE in the (PLA/LEF) + MTX incidence of elevated transaminase enzymes (ALT 14.6%; group during the extension included rash (6.3%), alopecia AST 13.7%) that was lower than in those initially random- (8.3%), and hypertension (3.1%); incidences of these events ized to LEF with a loading dose (ALT 31.5%; AST 16.9%), were similar to those in the first LEF + MTX group in the but higher than in patients initially randomized to PLA (ALT double-blind trial. Three (PLA/LEF) + MTX patients discon- 6.8%; AST 4.6%). All elevations of transaminase enzymes tinued study medication due to AE (maculopapular rash, in this group reversed with no intervention, or a dose reduc- infection, and joint disorder) during the open-label phase.
tion, or discontinuation of study medication at or before the Personal non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved
Personal, non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved.
Kremer, et al: Combination therapy for RA Table 4. Serious adverse events across treatment groups in Weeks 0–48.
* Considered related to study drug. ** One case considered related to study drug. Numbers in each columncannot be summed because a patient may have had more than one serious adverse event in the same body system.
end of the study. No patient initiating LEF at Week 24 discon- during the second 24 weeks of combination therapy, with tinued due to elevated LFT during the open-label phase.
the exception of ESR. Baseline ESR was only mildlyelevated in these subjects who were taking background DISCUSSION
MTX therapy, which may in part explain the lack of Our study continues to support the rationale for combined improvement despite clinical improvement in other ACR LEF + MTX treatment. The efficacy of LEF + MTX was first reported in an open-label trial in which 57% of patients The ACR20 response rate was significantly lower in the were ACR20 responders after 36 weeks of therapy — a PLA + MTX group (27.1%) compared with the LEF + percentage of improved patients that remained relatively MTX group (59.4%) in the initial 24 weeks. When patients constant for the remainder of the 48 week study35. In the 24 receiving placebo had LEF added at Week 24, they achieved week randomized controlled trial preceding this extension an ACR20 response rate at Week 48 of the same magnitude study, adding LEF in patients with active disease despite (58.3%) as that attained by patients originally randomized to MTX treatment provided significant benefit compared with LEF + MTX. This is especially interesting, as the patients adding placebo. The benefits were documented by signifi- who switched from PLA to LEF did so without a loading cant improvement in ACR20, ACR50, and ACR70 response dose. The finding cannot be attributed to the fact that all rates, as well as in quality of life measures34 (Table 2).
patients knew they were receiving open-label combination Individual components of the ACR response criteria also therapy, because the double blind regarding their initial followed a pattern of maintained or further improvement Personal non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved
Personal, non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved.
The Journal of Rheumatology 2004; 31:8 Table 5. Highest liver enzyme elevations and normalization during the first and second 24 weeks of treatment.
ULN: upper limit of normal, ALT: alanine aminotransferase, LFT: liver function tests, AST: aspartate aminotransferase. * In the PLA + MTX group, 3 of 9patients with ALT elevations and 1 of 6 patients with AST elevations did not normalize to ≤ 1.2 × ULN prior to or at the final study visit. ** Resolved aftertermination of study. § Discontinuations due to AE of abnormal LFT: 3 LEF + MTX patients discontinued in Weeks 0–24 due to elevations of ALT and AST,which normalized by the followup visit. † Two PLA + MTX patients discontinued due to elevations of ALT and AST, which remained elevated at the followupvisit but normalized several months later. ¶ No (LEF/LEF) + MTX patient in Weeks 24–48 discontinued due to an AE of abnormal LFT. # No patient switchingfrom PLA to LEF [(PLA/LEF) + MTX] discontinued in Weeks 24–48 due to an AE of abnormal LFT.
The improvement seen in HAQ DI at Week 48 for patients the ability of combined LEF and MTX therapy to greatly switching from PLA to LEF at Week 24 did not reach the reduce functional impairment and disability over time.
magnitude seen in the group originally randomized to combi- A safety concern of combining LEF and MTX is poten- nation therapy for the first 24 weeks (Table 2); nonetheless, tial hepatotoxicity. Liver enzyme elevations that occurred in HAQ DI improved by –0.33, a clinically important improve- patients receiving combination LEF + MTX during Weeks ment. Failure to achieve the same magnitude of improvement 24–48 normalized after a reduction or discontinuation of in HAQ DI level may possibly have been related to the delay LEF, as seen in the earlier double-blind trial. Three patients of 24 weeks prior to the addition of LEF.
whose elevations normalized had a reelevation to > 1.2 to ≤ For patients in the (LEF/LEF) + MTX group, the mean 2 × ULN. After initiating LEF at Week 24 without a loading change of –0.52 in the HAQ DI at Week 24, which was dose, fewer patients had elevated LFT in a 24 week period maintained at Week 48 (–0.54) (Table 2), exceeded the than did those who added LEF with a loading dose at the MCID of –0.22 points for HAQ DI39. The distribution of beginning of the double-blind phase. Similarly, the inci- HAQ DI scores at baseline and at Weeks 24 and 48 showed dence of both diarrhea and nausea was less during the open- Personal non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved
Personal, non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved.
Kremer, et al: Combination therapy for RA label phase when LEF was added without a loading dose, monotherapy is inadequate45, and provides further insight on compared with that seen during the first 24 weeks of LEF + how to lessen toxicity when LEF is added to MTX.
MTX when a loading dose of LEF was given.
Although the reversibility of mild liver enzyme eleva- ACKNOWLEDGMENT
tions in a clinical trial setting is reassuring, the potential for The authors thank Marilyn Stearns, MD, for editorial assistance in the increased hepatic toxicity with the use of LEF and MTX combination should be recognized, confirming the need forregular liver enzyme monitoring. We recommend moni- REFERENCES
1. Yelin E, Meenan R, Nevitt M, Epstein W. Work disability in toring monthly for the first 6 months, and then every 4 to 8 rheumatoid arthritis: effects of disease, social, and work factors.
weeks, as described in the guidelines for monitoring MTX43.
It should be noted that LEF was initiated at lower dose in the 2. Pincus T, Callahan LF, Sale WG, Brooks AL, Payne LE, Vaughn study than that recommended for monotherapy. It is impor- WK. Severe functional declines, work disability, and increased tant to use proper selection to avoid the combination in mortality in seventy-five rheumatoid arthritis patients studied overnine years. Arthritis Rheum 1984;27:864–72.
patients with known hepatic disease and/or other hepatic 3. Pincus T, Callahan LF. Taking mortality in rheumatoid arthritis risk factors. In addition, there should be a higher level of seriously — predictive markers, socioeconomic status and vigilance for adverse effects, with regular hepatic enzyme comorbidity. J Rheumatol 1986;13:841–5.
4. Kremer JM, Lee JK. The safety and efficacy of the use of Given the progressive nature of RA, most double-blind methotrexate in long-term therapy for rheumatoid arthritis. ArthritisRheum 1986;29:822–31.
placebo controlled studies in RA are of limited duration (6 5. Scott DL, Symmons DP, Coulton BL, Popert AJ. Long-term months or less), as prolonged treatment with placebo is outcome of treating rheumatoid arthritis: results after 20 years.
considered unethical in patients with active RA. While extension studies provide clinicians with valuable longterm 6. Weinblatt ME, Trentham DE, Fraser PA, et al. Long-term safety and efficacy information on RA therapies, it should prospective trial of low-dose methotrexate in rheumatoid arthritis.
Arthritis Rheum 1988;31:167–75.
be recognized that there are inherent weaknesses in such 7. American College of Rheumatology Subcommittee on Rheumatoid trials. Since extension studies usually exclude patients who Arthritis Guidelines. Guidelines for the management of rheumatoid do not complete the initial trial, efficacy responses in the arthritis. 2002 Update. Arthritis Rheum 2002;46:328-46.
extension may be biased, as patients with a good response to 8. Haagsma CJ, van Riel PL, de Jong AJ, van de Putte LB.
therapy are more likely to continue in the extension study Combination of sulphasalazine and methotrexate versus the singlecomponents in early rheumatoid arthritis: a randomized, controlled, (and subjects with a poorer response more likely to drop double-blind, 52 week clinical trial. Br J Rheumatol out). In addition, it is not clear if outcome measures reported in the initial trial are the most appropriate measures to be 9. Luong BT, Chong BS, Lowder DM. Treatment options for used in an extension trial. A review by Landewe and van der rheumatoid arthritis: celecoxib, leflunomide, etanercept, and Heijde discusses in depth the role and limitations of exten- infliximab. Ann Pharmacother 2000;34:743–60.
10. Mottonen T, Hannonen P, Leirisalo-Repo M, et al. Comparison of sion studies in RA44. Thus, it is important that this extension combination therapy with single-drug therapy in early rheumatoid study provides new insight and information regarding safety arthritis: a randomised trial. FIN-RACo trial group. Lancet of LEF combination therapy, as well as information regarding the safety and efficacy of LEF given in the 11. O’Dell JR. Combination DMARD therapy for rheumatoid arthritis: a step closer to the goal. Ann Rheum Dis 1996;55:781–3.
12. O’Dell JR, Haire C, Erikson N, et al. Efficacy of triple DMARD In summary, the therapeutic benefit of combination LEF therapy in patients with RA with suboptimal response to + MTX for the treatment of RA in patients with active methotrexate. J Rheumatol 1996;23 Suppl 44:72–4.
disease taking MTX alone, including improvements in the 13. Pincus T, Stein CM, Wolfe F. “No evidence of disease” in signs and symptoms (ACR response), physical function rheumatoid arthritis using methotrexate in combination with other (HAQ DI), and HRQoL (SF–36), was maintained to 48 drugs: a contemporary goal for rheumatology care? Clin ExpRheumatol 1997;15:591–6.
weeks. Adverse events after adding LEF were similar to 14. Pincus T, O’Dell JR, Kremer JM. Combination therapy with those reported in the LEF monotherapy studies. Elevated multiple disease-modifying antirheumatic drugs in rheumatoid liver enzymes, diarrhea, and nausea were less frequent in the arthritis: a preventive strategy. Ann Intern Med 1999;131:768–74.
24 weeks after adding LEF without a loading dose than they 15. Stein CM, Pincus T, Yocum D, et al. Combination treatment of severe rheumatoid arthritis with cyclosporine and methotrexate for were after adding LEF with a loading dose. While the time forty-eight weeks: an open-label extension study. The required to achieve an ACR20 response without a loading Methotrexate-Cyclosporine Combination Study Group. Arthritis dose is uncertain, it appears that a useful strategy to lessen toxicity would be to decrease or omit a loading dose of LEF 16. Verhoeven AC, Boers M, Tugwell P. Combination therapy in 100 mg on Days 1 and 2 when LEF is added to MTX. Our rheumatoid arthritis: updated systematic review. Br J Rheumatol1998;37:612–9.
findings support the more recent paradigm of combined 17. Boers M, Verhoeven AC, Markusse HM, et al. Randomised DMARD therapy for RA treatment when control on comparison of combined step-down prednisolone, methotrexate and Personal non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved
Personal, non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved.
The Journal of Rheumatology 2004; 31:8 sulphasalazine with sulphasalazine alone in early rheumatoid treatment from an open randomized study [abstract]. Ann Rheum arthritis. Lancet 1997;350:309–18.
18. Kremer JM. Combination therapy with biologic agents in 32. Hansen KE, Cush J, Singhal A, et al. The safety and efficacy of rheumatoid arthritis: perils and promise. Arthritis Rheum leflunomide in combination with infliximab in rheumatoid arthritis [abstract]. Arthritis Rheum 2001;44 Suppl:S84.
19. Kremer JM. Methotrexate and leflunomide: biochemical basis for 33. Dougados M, Combe B, van Riel P, et al. Efficacy and safety of combination therapy in the treatment of rheumatoid arthritis. Semin leflunomide in combination with sulfasalazine versus sulfasalazine in patients with rheumatoid arthritis: Results from the RELIEF 20. Paulus HE. The use of combinations of disease-modifying study. Ann Rheum Dis 2002;61 Suppl 1:209.
antirheumatic agents in rheumatoid arthritis. Arthritis Rheum 34. Kremer JM, Genovese MC, Cannon GW, et al. Concomitant leflunomide therapy in patients with active rheumatoid arthritis 21. O’Dell J. Combination DMARD therapy for rheumatoid arthritis: despite stable doses of methotrexate. A randomized, double-blind, apparent universal acceptance [abstract]. Arthritis Rheum 1997;40 placebo-controlled trial. Ann Intern Med 2002;137:726–33.
35. Weinblatt ME, Kremer JM, Coblyn JS, et al. Pharmacokinetics, 22. O’Dell JR, Haire CE, Erikson N, et al. Treatment of rheumatoid safety, and efficacy of combination treatment with methotrexate arthritis with methotrexate alone, sulfasalazine and and leflunomide in patients with active rheumatoid arthritis.
hydroxychloroquine, or a combination of all three medications. N 36. Mroczkowski PJ, Weinblatt ME, Kremer JM. Methotrexate and 23. Tugwell P, Pincus T, Yocum D, et al. Combination therapy with leflunomide combination therapy for patients with active cyclosporine and methotrexate in severe rheumatoid arthritis. The rheumatoid arthritis. Clin Exp Rheumatol 1999;17:S66–8.
Methotrexate-Cyclosporine Combination Study Group. N Engl J 37. Furst D, Luggen M, Thompson A, Coleman J. Adding leflunomide to patients with active rheumatoid arthritis while receiving 24. Bensen W, Tugwell P, Roberts RM, et al. Combination therapy of methotrexate improves physical function and health-related quality cyclosporine with methotrexate and gold in rheumatoid arthritis (2 of life [abstract]. Arthritis Rheum 2000;43 Suppl:S344.
pilot studies). J Rheumatol 1994;21:2034–8.
38. Ware JE, Kosinski M, Keller SD. SF-36 physical and mental health 25. Williams HJ, Ward JR, Reading JC, et al. Comparison of auranofin, summary scales: a user’s manual. Boston: The Health Institute, methotrexate, and the combination of both in the treatment of rheumatoid arthritis. A controlled clinical trial. Arthritis Rheum 39. Kosinski M, Zhao SZ, Dedhiya S, Osterhaus JT, Ware JE Jr.
Determining minimally important changes in generic and 26. Willkens RF, Urowitz MB, Stablein DM, et al. Comparison of disease-specific health-related quality of life questionnaires in azathioprine, methotrexate, and the combination of both in the clinical trials of rheumatoid arthritis. Arthritis Rheum treatment of rheumatoid arthritis. A controlled clinical trial.
40. Kosinski M, Martin R, Henkenius S, Wanke LA, Buatti M.
27. Weinblatt ME, Kremer JM, Bankhurst AD, et al. A trial of Determining clinically meaningful improvement in SF-36 scale etanercept, a recombinant tumor necrosis factor receptor:Fc fusion scores for treatment studies in rheumatoid arthritis [abstract].
protein, in patients with rheumatoid arthritis receiving methotrexate. N Engl J Med 1999;340:253–9.
41. Samsa G, Edelman D, Rothman ML, Williams GR, Lipscomb J, 28. Lipsky PE, van der Heijde DM, St. Clair EW, et al. Infliximab and Matchar D. Determining clinically important differences in health methotrexate in the treatment of rheumatoid arthritis. Anti-Tumor status measures: a general approach with illustration to the Health Necrosis Factor Trial in Rheumatoid Arthritis with Concomitant Utilities Index Mark II. Pharmacoeconomics 1999;15:141–55.
Therapy Study Group. N Engl J Med 2000;343:1594–602.
42. Strand V, Bombardier C, Maetzel A, Scott D, Crawford B. Use of 29. Maini R, St. Clair EW, Breedveld F, et al. Infliximab (chimeric minimum clinically important differences in evaluating patient anti-tumour necrosis factor alpha monoclonal antibody) versus responses to treatment of RA [abstract]. Arthritis Rheum 2001;44 placebo in rheumatoid arthritis patients receiving concomitant methotrexate: a randomised phase III trial. ATTRACT Study 43. Kremer JM, Alarcon GS, Lightfoot RW Jr, et al. Methotrexate for rheumatoid arthritis: suggested guidelines for monitoring liver 30. Bresnihan B. The safety and efficacy of interleukin-1 receptor toxicity. Arthritis Rheum 1994;37:316–28.
antagonist in the treatment of rheumatoid arthritis. Semin Arthritis 44. Landewe R, van der Heijde D. Follow up studies in rheumatoid arthritis. Ann Rheum Dis 2002;61:479-81.
31. Karanikola G, Charalambopoulos D, Andrianakos A, et al.
45. Kremer JM. Rational use of new and existing disease-modifying Cyclosporin vs leflunomide vs combination of two drugs in agents in rheumatoid arthritis. Ann Intern Med 2001;134:695–706.
rheumatoid arthritis: long-term optimal results of combination Personal non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved
Personal, non-commercial use only. The Journal of Rheumatology. Copyright 2004. All rights reserved.
Kremer, et al: Combination therapy for RA

Source: http://jrheum.com/subscribers/04/08/tables/PDF/1521.pdf