Abstract

Objectives . To evaluate the long-term impact on physical function of a single course of rituximab in rheumatoid factor, seropositive patients with active rheumatoid arthritis (RA) despite ongoing methotrexate treatment.

Methods . A randomized, controlled trial comparing rituximab alone [1000 mg intravenously (iv) on days 1 and 15, n = 40], or in combination with cyclophosphamide (750 mg iv on days 3 and 7, n = 41) or oral methotrexate (≥10 mg/week, n = 40) with placebo+ methotrexate (≥10 mg/week, n = 40), resulted in significant reductions in disease activity at weeks 24 and 48. Sustained improvements in physical function and standard effect sizes (SES) for changes in components of ACR and EULAR criteria were evaluated over 2 yrs.

Results . More patients receiving rituximab+ methotrexate completed a 2-yr follow-up without further treatment than those receiving placebo+ methotrexate (45% vs 15%, respectively), rituximab alone (10%) or rituximab+ cyclophosphamide (22%). This reflected a higher percentage of patients receiving rituximab+ methotrexate reporting improvements in Health Assessment Questionnaire Disability Index≥ minimum clinically important difference at 1 and 2 yrs (68% and 30%, respectively) compared with placebo+ methotrexate (28% and 15%), rituximab monotherapy (43% and 10%) or rituximab+ cyclophosphamide (39% and 12%). SES were high in all rituximab groups and revealed differing patterns of response over time.

Conclusion . A single course of rituximab with continuing methotrexate in patients with active RA provided clinically meaningful improvements in physical function over 2 yrs, with lower discontinuation rates and larger SES for improvements in ACR and EULAR criteria components.

The clinical symptoms of rheumatoid arthritis (RA)—synovitis and systemic symptoms including fatigue and morning stiffness—are accompanied by radiographic changes that manifest as erosions and joint space narrowing, collectively leading to progressive loss of physical function. As the link between joint damage and disability is well established in RA [ 1 ], treatment should focus not only on reducing the signs and symptoms of active disease and inhibition of radiographic progression, but also on maintenance and improvement in physical function of the patient. In this regard, patient-reported outcomes—pain, global assessment of disease activity and physical function—most effectively distinguish between active and placebo treatment [ 2–4 ].

Despite recent important therapeutic advances in RA, disease management remains challenging both for physicians and patients, and the need for innovative therapeutic approaches remains [ 5 , 6 ]. The inciting events and underlying pathophysiology resulting in synovitis in RA are not completely understood. However, it is clear that B cells contribute to the perpetuation of active disease through mechanisms that include autoantibody production, antigen presentation, cytokine production, T cell–B cell interactions, maintenance of ectopic (synovial) germinal centres and effector cell functions [ 7–11 ].

Rituximab is a B cell-targeted monoclonal antibody that selectively depletes B cells expressing the cell surface antigen CD20. This property of rituximab is utilized in the treatment of relapsed or refractory, low-grade or follicular CD20 positive (CD20+) B cell non-Hodgkin's lymphoma (NHL) in combination with a regimen that includes cyclophosphamide and glucocorticoids [ 12 ]. The binding of rituximab to CD20+ B cells results in depletion of CD20+ B cells through three putative mechanisms of action: antibody-dependent cell-mediated cytotoxicity, complement-dependent cytotoxicity and promotion of B cell apoptosis [ 13–16 ]. As CD20 is not expressed on stem cells or plasma cells [ 17 , 18 ], therapeutic B cell depletion with rituximab leaves these B cell populations intact [ 19 ].

Rituximab was recently approved by the FDA for use in combination with methotrexate to reduce signs and symptoms in adult patients with moderately to severely active RA who have had an inadequate response to one or more tumour necrosis factor antagonist therapies. A randomized, double-blind Phase II trial in patients with active RA despite ongoing treatment with methotrexate confirmed the benefit of rituximab therapy in this population [ 19 ]. A single course of two rituximab infusions, alone or in combination with cyclophosphamide [the inclusion of cyclophosphamide induction in one arm of this study was based on the successful treatment of NHL with rituximab in combination with CHOP (cyclophosphamide, hydroxydoxorubicin, vincristine and prednisolone [ 12 ])] or continued methotrexate, resulted in a significant reduction in disease activity according to American College of Rheumatology (ACR) and European League Against Rheumatism (EULAR) response criteria at 24 weeks compared with placebo+ methotrexate. This differential benefit was sustained at 48 weeks in some patients. Moreover, significant improvements in physical function, as measured by the Health Assessment Questionnaire Disability Index (HAQ-DI), were evident at week 24 [ 20 ].

To characterize these sustained responses further, this report examines clinically meaningful improvements in physical function and the magnitude of treatment effects in components of ACR and EULAR response criteria observed in the subset of patients who demonstrated sustained clinical benefit following the single course of rituximab.

Methods

Patients and study design

Adult rheumatoid factor seropositive patients, with RA diagnosed by revised 1987 American Rheumatism Association criteria [ 21 ], who had failed 1–5 disease-modifying anti-rheumatic drugs (DMARDs) and had active disease despite ongoing treatment with methotrexate (≥10 mg/week) for ≥16 weeks, were enrolled into the study. To be eligible, patients had to have a swollen joint count (SJC) ≥8, a tender joint count (TJC) ≥8 and at least two of the following: an elevated C-reactive protein (CRP) level (≥1.5 mg/dl) and/or erythrocyte sedimentation rate (ESR) (≥30 mm/h), and/or morning stiffness ≥45 min. Patients were randomized in a double-blind fashion to one of the four treatment groups: continuing oral methotrexate (≥10 mg/week)+ placebo rituximab; rituximab alone [1000 mg intravenous (iv) infusion days 1 and 15]; rituximab+ cyclophosphamide (750 mg iv on days 3 and 17); or rituximab+ continuing methotrexate. All patients received methylprednisolone 100 mg iv before infusions (rituximab or placebo) and oral prednisone for 2 weeks after the first infusion (total prednisone dose 510 mg).

The primary end-point of the study was the proportion of patients at week 24 with an ACR50 response, defined as ≥50% improvements from baseline in TJC and SJC, and three of the five components of patient pain [by visual analogue scale (VAS) score], patient assessment of global disease activity (VAS), physical function (HAQ-DI), physician global assessment of disease activity (VAS) and ESR or CRP [ 22 ]. Secondary outcome measures included EULAR responses based on ‘good’ and ‘moderate’ improvements in Disease Activity Scores (DAS) derived from TJC, SJC, patient assessment of global disease activity and ESR or CRP [ 23–25 ].

Patients continued protocol participation for up to 2 yrs without further treatment with rituximab, or until they and their treating physicians considered that the initial clinical benefit had lapsed and a repeat course of rituximab or alternative therapy was indicated. Those randomized to methotrexate-containing arms continued methotrexate, and patients and investigators remained blinded to treatment assignments during the 2-yr follow-up period. The study was approved by the institutional review board or the ethics committee at each study site. All patients gave written informed consent.

Study assessments

ACR criteria were evaluated at screening, baseline (day 1), week 12, every 4 weeks to the primary end-point at 24 weeks, and every 8 weeks thereafter. Physical function was assessed using the HAQ-DI [ 26 ], a 20-item questionnaire that scores performance of physical activities and activities of daily living—arising, dressing, eating, walking, hygiene, reaching and gripping—on a scale from 0= without difficulty to 3= unable to perform without assistance. The minimal clinically important difference (MCID) in HAQ-DI reflecting a meaningful improvement in physical function is a decrease of ≥0.22, derived from correlations with global assessments of disease activity in multiple randomized controlled trials and longitudinal case series [ 3 , 27–32 ].

Statistical considerations

Descriptive statistics were used to compare baseline demographic and disease parameters in the initial intent-to-treat (ITT) population with those completing either 1 or 2 yrs of protocol participation. ACR responses were calculated using the ITT population. Patients with insufficient data to calculate an ACR response and patients who withdrew from protocol participation were classified as non-responders. In this study, a reduction in HAQ-DI of ≥0.25 from baseline was used to determine the proportion of patients with improvements in HAQ-DI that met or exceeded the MCID at each time point.

Secondary analyses included mean and median changes relative to baseline in HAQ-DI at weeks 12–104 without imputation. The significance of the change from baseline was determined by analysis of variance using all available data.

To characterize the relative magnitude of change in outcome measures and enable comparisons across individual parameters, standard effect sizes (SES) for changes from baseline in each component of the ACR and EULAR response criteria were calculated at weeks 24, 48, 72 and 104 for each treatment group, using available data. SES were calculated using pooled standard deviations using the formula:  

formula
Large magnitudes of effect are defined by values ≥0.8, small effect sizes by values ≥0.2 and medium effects by intermediate values (approximately 0.5) [ 33 ].

Results

Patient disposition and characteristics

One hundred and sixty-one patients were randomized, all of whom received at least one dose of study medication (rituximab/placebo), with at least one follow-up assessment after baseline. Patients were randomized as follows: 40 in each of the placebo+ methotrexate, rituximab monotherapy, and rituximab+ methotrexate groups, and 41 in the rituximab+ cyclophosphamide group. Median doses of methotrexate at study entry were 12.5–15 mg/week across the three groups.

The disposition of patients in each treatment group over the 104 weeks of follow-up is summarized in Fig. 1 . At all time points, the proportion of patients continuing protocol participation was greatest in the rituximab+ methotrexate group—98%, 95% and 70% at weeks 24, 48 and 72, respectively. At 104 weeks, 45% of patients in this treatment group remained in protocol follow-up compared with 15%, 10% and 22% in the placebo+ methotrexate, rituximab monotherapy and rituximab+ cyclophosphamide groups, respectively. Protocol discontinuation due to lack of efficacy by week 104 was lowest in the rituximab+ methotrexate group. The most frequent reason for protocol discontinuation across the groups (categorized as ‘other’ in Fig. 1 ) was related to a return of symptoms requiring a further course of rituximab in a separate protocol. All patients in the placebo+ methotrexate group who discontinued protocol participation for this reason did so within 48 weeks. In contrast, the majority of discontinuations for this reason in each of the rituximab treatment groups occurred between weeks 48 and 104, with the lowest number in the rituximab+ methotrexate group.

F ig . 1.

Disposition of patients over 2 yrs. *The majority of protocol discontinuations classified as ‘other’ were owing to the requirement for retreatment with rituximab under a separate protocol; § includes 1 death due to pneumonia.

F ig . 1.

Disposition of patients over 2 yrs. *The majority of protocol discontinuations classified as ‘other’ were owing to the requirement for retreatment with rituximab under a separate protocol; § includes 1 death due to pneumonia.

Small sample sizes at 104 weeks in the methotrexate+ placebo, rituximab alone, and rituximab+ cyclophosphamide groups ( Fig. 1 ) precluded accurate comparisons at this time point across treatments and compared with the rituximab+ methotrexate group.

Baseline demographics and disease characteristics were comparable across treatment groups ( Table 1 ). Mean HAQ-DI scores (1.8–2.0) at baseline indicated, overall, that patients across all groups had ‘much difficulty’ in performing routine tasks before the start of the study. Baseline demographic and disease characteristics in cohorts completing follow-up at 1 yr [ n = 128 (79.5%)] and 2 yrs [ n = 37 (23.0%)] were generally comparable with the baseline ITT population ( n = 161) and, although baseline SJC and TJC values were lower in patients who completed 2 yrs on protocol, patient-reported measures—patient global assessment, pain and HAQ-DI—were similar across the ITT and follow-up subpopulations at this time point.

T able 1.

Characteristics of patients at study entry

Baseline mean (±  s.d .)  Placebo + MTX RTX alone RTX+ CTX RTX+ MTX 
(A) Demographic characteristics 
Patients, n (%)  40 (100) 40 (100) 41 (100) 40 (100) 
Gender (% female) 80.0 72.5 82.9 75.0 
Age (yrs) 53.7 (11.2) 53.5 (10.2) 52.9 (9.9) 53.5 (11.9) 
Duration of RA (yrs) 11.0 (7.1) 9.3 (5.5) 9.8 (6.1) 11.5 (7.3) 
DMARDs failed ( n ) (other than MTX)  2.6 (1.3) 2.5 (1.6) 2.6 (1.4) 2.5 (1.4) 
(B) Disease characteristics 
SJC ( n )  18.6 (9.6) 20.5 (11.2) 19.4 (10.2) 22.7 (12.7) 
TJC ( n )  32.0 (12.8) 33.8 (14.8) 32.7 (13.8) 32.3 (15.9) 
MD global 66.7 (15.7) 66.7 (16.2) 67.4 (14.5) 65.6 (14.0) 
Pt global 65.4 (20.3) 67.8 (22.8) 63.4 (20.5) 61.4 (20.4) 
Pt pain assessment 62.6 (16.1) 62.0 (20.2) 57.5 (20.0) 54.6 (17.8) 
(VAS, mm)     
HAQ-DI 2.0 (0.5) 2.0 (0.6) 1.8 (0.7) 1.8 (0.6) 
ESR (mm/h) 51.5 (31.8) 46.9 (22.9) 54.9 (28.6) 53.3 (23.4) 
CRP (mg/dl) 3.19 (42.7) 2.56 (2.20) 3.98 (4.08) 2.88 (3.18) 
DAS28 6.9 (0.7) 6.8 (1.0) 6.9 (0.8) 6.8 (0.9) 
Baseline mean (±  s.d .)  Placebo + MTX RTX alone RTX+ CTX RTX+ MTX 
(A) Demographic characteristics 
Patients, n (%)  40 (100) 40 (100) 41 (100) 40 (100) 
Gender (% female) 80.0 72.5 82.9 75.0 
Age (yrs) 53.7 (11.2) 53.5 (10.2) 52.9 (9.9) 53.5 (11.9) 
Duration of RA (yrs) 11.0 (7.1) 9.3 (5.5) 9.8 (6.1) 11.5 (7.3) 
DMARDs failed ( n ) (other than MTX)  2.6 (1.3) 2.5 (1.6) 2.6 (1.4) 2.5 (1.4) 
(B) Disease characteristics 
SJC ( n )  18.6 (9.6) 20.5 (11.2) 19.4 (10.2) 22.7 (12.7) 
TJC ( n )  32.0 (12.8) 33.8 (14.8) 32.7 (13.8) 32.3 (15.9) 
MD global 66.7 (15.7) 66.7 (16.2) 67.4 (14.5) 65.6 (14.0) 
Pt global 65.4 (20.3) 67.8 (22.8) 63.4 (20.5) 61.4 (20.4) 
Pt pain assessment 62.6 (16.1) 62.0 (20.2) 57.5 (20.0) 54.6 (17.8) 
(VAS, mm)     
HAQ-DI 2.0 (0.5) 2.0 (0.6) 1.8 (0.7) 1.8 (0.6) 
ESR (mm/h) 51.5 (31.8) 46.9 (22.9) 54.9 (28.6) 53.3 (23.4) 
CRP (mg/dl) 3.19 (42.7) 2.56 (2.20) 3.98 (4.08) 2.88 (3.18) 
DAS28 6.9 (0.7) 6.8 (1.0) 6.9 (0.8) 6.8 (0.9) 

CRP, C-reactive protein; CTX, cyclophosphamide; DAS, Disease Activity Score defined according to European League Against Rheumatism (EULAR) criteria; DMARDs, disease-modifying anti-rheumatic drugs; ESR, erythrocyte sedimentation rate; HAQ-DI, Health Assessment Questionnaire Disability Index; MD global, physician global assessment of disease activity; MTX, methotrexate; Pt global, patient global assessment of disease activity; RA, rheumatoid arthritis; RTX, rituximab; s.d ., standard deviation; SJC, swollen joint count; TJC, tender joint count; VAS, visual analogue scale.

Physician-reported measures: SJC, TJC and MD global; patient-reported measures: Pt global, Pt Pain, HAQ-DI; acute-phase reactants: ESR and CRP.

Clinical efficacy, tolerability and safety

Compared with placebo+ methotrexate, the proportion of patients achieving ACR20, ACR50 and ACR70 was higher with rituximab treatment. Sustained responses over time were greatest in those receiving rituximab+ methotrexate [ 19 , 34 ]. At week 104, more patients treated with rituximab+ methotrexate achieved ACR20, ACR50 and ACR70 responses, compared with the other groups ( Table 2 ). Moreover, at week 104, a greater proportion of patients in the rituximab+ methotrexate group (39%) achieved a EULAR response compared with the placebo + methotrexate (16%), rituximab monotherapy (16%) and rituximab+ cyclophosphamide (24%) groups. There were no differences in the occurrence of adverse events that led to withdrawal, serious adverse events or infections in the rituximab treatment groups compared with the placebo+ methotrexate group [ 34 ].

T able 2.

Proportion of patients with ACR responses a at weeks 24 and 48 (ITT population)

  Placebo+ MTX ( n = 40)   RTX alone ( n = 40)   RTX+ CTX ( n = 41)   RTX+ MTX ( n = 40)  
Week 24 
    ACR20 15 (38)  26 (65) *  31 (76) **  29 (73) ** 
    ACR50 b 5 (13) 13 (33)  17 (41) **  17 (43) ** 
    ACR70 2 (5) 6 (15) 6 (15)  9 (23) * 
Week 48 
    ACR20 8 (20) 12 (30)  19 (46) *  27 (68) ** 
    ACR50 2 (5) 5 (13)  10 (24) *  14 (35) ** 
    ACR70 3 (8) 4 (10)  6 (15) * 
Week 104 
    ACR20 5 (13) 3 (8) 5 (13) 13 (33) 
    ACR50 4 (10) 3 (8) 4 (10) 8 (20) 
    ACR70 3 (8) 1 (3) 3 (8) 4 (10) 
  Placebo+ MTX ( n = 40)   RTX alone ( n = 40)   RTX+ CTX ( n = 41)   RTX+ MTX ( n = 40)  
Week 24 
    ACR20 15 (38)  26 (65) *  31 (76) **  29 (73) ** 
    ACR50 b 5 (13) 13 (33)  17 (41) **  17 (43) ** 
    ACR70 2 (5) 6 (15) 6 (15)  9 (23) * 
Week 48 
    ACR20 8 (20) 12 (30)  19 (46) *  27 (68) ** 
    ACR50 2 (5) 5 (13)  10 (24) *  14 (35) ** 
    ACR70 3 (8) 4 (10)  6 (15) * 
Week 104 
    ACR20 5 (13) 3 (8) 5 (13) 13 (33) 
    ACR50 4 (10) 3 (8) 4 (10) 8 (20) 
    ACR70 3 (8) 1 (3) 3 (8) 4 (10) 

*P < 0.05, ** P < 0.01 based on Fisher's exact test for comparisons with the placebo+ MTX group.

a Patients withdrawn/retreated or with insufficient data to calculate an ACR response are classified as non-responders.

b Note that the primary end-point of the study was ACR50 at week 24; all other P -values are exploratory.

ACR, American College of Rheumatology; CTX, cyclophosphamide; MTX, methotrexate; RTX, rituximab.

Physical function

Mean changes from baseline in HAQ-DI were statistically significant in all arms, and were generally higher in active-treatment groups than with placebo+ methotrexate ( Fig. 2 ). Improvements were maximal at 24 weeks and sustained thereafter in the rituximab+ methotrexate group. At week 72, the mean change from baseline in HAQ-DI with rituximab+ methotrexate ( n = 28) was –0.6 ( P < 0.0001), compared with –0.3 ( P ≤ 0.05) with placebo+ methotrexate ( n = 15), –0.6 ( P = 0.01) with rituximab monotherapy ( n = 17) and –0.1 ( P = 0.05) with rituximab+ cyclophosphamide ( n = 22). At week 104, higher patient retention with rituximab+ methotrexate treatment (48% compared with 15%, 10% and 22% in placebo+ methotrexate, rituximab monotherapy and rituximab+ cyclophosphamide, respectively) precluded accurate between-group comparisons.

F ig . 2.

Change from baseline in HAQ-DI. Negative values indicate change from baseline: aP ≥ 0.0001; bP = 0.0002; cP ≥ 0.005; dP = 0.01; eP ≥ 0.05; baseline mean ( s.d .) HAQ-DI values were: placebo+ MTX: 2.0 (0.54); RTX monotherapy: 2.0 (0.54) ( n = 40); RTX+ CTX: 1.8 (0.68) ( n = 41); and RTX+ MTX: 1.8 (0.57) ( n = 40). Data at week 104 are not presented owing to low numbers in the MTX ( n = 6), RTX ( n = 4) and RTX+ CTX ( n = 9) treatment groups, which preclude accurate comparisons. CTX, cyclophosphamide; HAQ-DI, Health Assessment Questionnaire Disability Index; MCID, minimal clinically important difference; MTX, methotrexate; PLA, placebo; RTX, rituximab.

F ig . 2.

Change from baseline in HAQ-DI. Negative values indicate change from baseline: aP ≥ 0.0001; bP = 0.0002; cP ≥ 0.005; dP = 0.01; eP ≥ 0.05; baseline mean ( s.d .) HAQ-DI values were: placebo+ MTX: 2.0 (0.54); RTX monotherapy: 2.0 (0.54) ( n = 40); RTX+ CTX: 1.8 (0.68) ( n = 41); and RTX+ MTX: 1.8 (0.57) ( n = 40). Data at week 104 are not presented owing to low numbers in the MTX ( n = 6), RTX ( n = 4) and RTX+ CTX ( n = 9) treatment groups, which preclude accurate comparisons. CTX, cyclophosphamide; HAQ-DI, Health Assessment Questionnaire Disability Index; MCID, minimal clinically important difference; MTX, methotrexate; PLA, placebo; RTX, rituximab.

At week 24, the proportions of patients reporting improvements in HAQ-DI that met or exceeded MCID (a reduction of ≥0.25) were higher in active-treatment groups (68%, 59% and 63% with rituximab monotherapy, rituximab + cyclophosphamide and rituximab+ methotrexate, respectively) than with placebo+ methotrexate (45%) ( Figs 2 and 3 ). At weeks 48 and 72, the proportions of patients with clinically meaningful improvements in physical function were greater in the rituximab+ methotrexate group (68% and 48%, respectively) than in the placebo+ methotrexate (28% and 18%), rituximab monotherapy (43% and 18%) and rituximab+ cyclophosphamide (39% and 22%) groups. This difference in favour of the rituximab+ methotrexate group persisted at week 104, notwithstanding the reduced number of patients continuing protocol participation.

F ig . 3.

Proportion of patients with improvements from baseline in HAQ-DI meeting or exceeding MCID (defined as decreases from baseline score ≥ 0.25. Observed data are used at each interval; missing values were classified as ‘not improved’) over 104 weeks. CTX, cyclophosphamide; HAQ-DI, Health Assessment Questionnaire Disability Index; MTX, methotrexate; RTX, rituximab.

F ig . 3.

Proportion of patients with improvements from baseline in HAQ-DI meeting or exceeding MCID (defined as decreases from baseline score ≥ 0.25. Observed data are used at each interval; missing values were classified as ‘not improved’) over 104 weeks. CTX, cyclophosphamide; HAQ-DI, Health Assessment Questionnaire Disability Index; MTX, methotrexate; RTX, rituximab.

SES in physician- and patient-reported measures, as well as ESR and CRP, reflected different sensitivities to change both across treatment groups and across intervals from baseline. Data are presented at 24, 48 and 72 weeks. Overall, SES with active treatment were large, confirming the statistical significance of ACR and EULAR responses in this trial, despite sample sizes of 40–41 patients/treatment group. At all time points, SES with placebo treatment were intermediate in value (with the exception of TJC and physician global assessment of disease activity). At week 24, excluding CRP, SES in all rituximab treatment groups ranged from 0.9 to 2.8 compared with 0.6–1.1 with placebo+ methotrexate. The greatest SES were evident with rituximab+ cyclophosphamide treatment across both physician- and patient-reported components. At week 48, SES for these components were largest in the rituximab+ methotrexate group (ranging from 1.0 to 2.4). At week 72, the largest SES were seen in four of the eight components assessed in both the rituximab monotherapy and rituximab+ methotrexate treatment groups. Larger SES for patient-reported measures were consistently evident over 24, 48 and 72 weeks of treatment with rituximab+ methotrexate treatment relative to each of the other active treatment groups ( Fig. 4 a–d).

F ig . 4.

Standard effect sizes (SES) over time for changes from baseline in components of ACR response criteria. SES comparisons at week 104 are omitted owing to low numbers of patients in the placebo+ methotrexate (a), rituximab monotherapy (b) and rituximab+ cyclophosphamide (c) groups compared with the rituximab+ methotrexate group at this time point. CRP, C-reactive protein; CTX, cyclophosphamide; ESR, erythrocyte sedimentation rate; HAQ-DI, Health Assessment Questionnaire Disability Index; MTX, methotrexate; Phys global, physician global assessment of disease activity; Pt global, patient global assessment of disease activity; RTX, rituximab; SJC, swollen joint count; TJC, tender joint count; VAS, visual analogue scale.

F ig . 4.

Standard effect sizes (SES) over time for changes from baseline in components of ACR response criteria. SES comparisons at week 104 are omitted owing to low numbers of patients in the placebo+ methotrexate (a), rituximab monotherapy (b) and rituximab+ cyclophosphamide (c) groups compared with the rituximab+ methotrexate group at this time point. CRP, C-reactive protein; CTX, cyclophosphamide; ESR, erythrocyte sedimentation rate; HAQ-DI, Health Assessment Questionnaire Disability Index; MTX, methotrexate; Phys global, physician global assessment of disease activity; Pt global, patient global assessment of disease activity; RTX, rituximab; SJC, swollen joint count; TJC, tender joint count; VAS, visual analogue scale.

Discussion

As published, ACR and EULAR responses were highest with rituximab+ methotrexate and rituximab+ cyclophosphamide treatment at the 24-week primary end-point of this trial, and with rituximab+ methotrexate at 48 weeks. Continued protocol participation was highest at 48 weeks in this latter treatment group [ 19 , 34 ]. This report analysed long-term responses in patients remaining in protocol follow-up after a single treatment course of rituximab with or without methotrexate background therapy compared with placebo+ continued methotrexate. High attrition rates between 1 and 2 yrs in this blinded protocol follow-up are not surprising as patients exited when further treatment was indicated; low patient numbers in the placebo+ methotrexate and rituximab monotherapy treatment groups (15% and 10%, respectively) confound between-group comparisons at 2 yrs. The rituximab+ methotrexate population had the highest completion rates at 1 and 2 yrs—95% and 45%, respectively—compared with 83% and 22% with rituximab+ cyclophosphamide. Furthermore, at 2 yrs rituximab+ methotrexate provided higher rates of clinical improvement at each level of ACR response than the other groups. These data suggest that background therapy with methotrexate resulted in more sustained clinical responses following treatment with rituximab in RA.

Patient-reported physical function, measured by mean changes from baseline in HAQ-DI and the proportion of patients with clinically meaningful changes, confirmed the benefit of active therapy, as did SES for components of ACR and EULAR response criteria. Importantly, these analyses demonstrated that rituximab+ methotrexate treatment resulted in consistent, sustained improvements over time, as demonstrated by higher ACR and EULAR responses at week 48 and lower protocol dropout rates over 2 yrs. The apparently high proportion of patients (45%) receiving placebo+ methotrexate who had improved HAQ-DI scores may be a result of the short (16-week) duration of required background methotrexate treatment; clinical evidence suggests that maximal responses to methotrexate therapy are observed at or after 6 months [ 35 ]. This result may also reflect the high doses of glucocorticoids administered to all patients in the study.

Statistically significant changes from baseline in randomized controlled trials may not necessarily be clinically meaningful or easily translated into everyday practice. Although MCID may represent a ‘minimum’ perceived benefit, recent trials have illustrated its importance, especially in patients with longstanding active RA (despite ongoing methotrexate treatment) and moderate impairment in performance of day-to-day physical activities (as in this study). Improvements in physical function over time in the present trial compare favourably with those reported for other biologic agents in similar patient populations. Patients with longstanding disease receiving infliximab+ methotrexate over 2 yrs reported significantly greater reductions in HAQ-DI than those randomized to placebo+ methotrexate [ 36 ]. In DMARD-resistant patients receiving adalimumab+ background methotrexate, mean reductions in HAQ-DI of 0.59–0.61 at 52 weeks were statistically superior to that seen with placebo+ methotrexate (–0.25) [ 37 ]. In both trials, comparable benefits in physical function required regular therapy rather than a single treatment course of two infusions over 2 weeks, as is possible with rituximab. Given the strong association between physical functioning and joint damage in the RA patients with 8- to 15-yrs disease duration, sustained improvements in physical function observed after a single course of rituximab+ methotrexate in this protocol population may be of particular clinical significance [ 1 ].

Statistically significant improvements in physical function in this trial were clinically meaningful and associated with large SES. Effect size calculations essentially confirm statistical power retrospectively. In addition to being particularly useful for meta-analyses, SES also indicate whether statistically significant findings are robust. In this protocol, SES were large in all active-treatment groups—exceeding 0.8 in most components over 24–72 weeks vs placebo+ methotrexate, where most values were <0.8—and compare favourably with SES reported from other controlled trials in RA [ 2 , 38 ]. SES were largest in the rituximab+ cyclophosphamide group at week 24 (but not thereafter), reflecting early clinical responses in this population that were not sustained by ACR criteria or improvements in physical function. Using a conservative non-responder imputation (where all dropouts were assigned as non-responders), SES with rituximab+ methotrexate treatment remained large at weeks 48 and 72, and were most consistent for patient-reported measures across all time points.

In this study, SES (calculated for each group using pooled standard deviations to characterize the relative magnitude of change for an outcome measure) facilitated comparisons across treatment groups by individual parameters. SES for changes from baseline confirmed that HAQ-DI positively discriminated between active and placebo treatment, and consistently identified sustained responses with rituximab+ methotrexate at weeks 48 and 72, as did patient and physician assessments of global disease activity, in addition to ESR and CRP. Taken together, these analyses support the robustness of statistically significant results with this combination treatment—reflected in more clinically meaningful improvements in physical function and continued protocol participation in 45% of patients without the need for a further course of treatment with rituximab over 24 months.

In summary, this study has shown that a substantial proportion of RA patients who had inadequately responded to multiple DMARDs experienced clinically meaningful and robust improvements in physical function for up to 2 yrs following a single course of rituximab with continuing methotrexate. Moreover, these data demonstrate concordance between clinical response and patient-reported outcomes in this group of patients.

graphic

Acknowledgements

Funding to pay the Open Access publication charges for this article was provided by …

Writing assistance for this manuscript was provided by Genentech, Inc.

VS serves as a consultant to Genentech and Roche and serves on advisory boards for Roche. PE has provided expert advice and received speaker's fees from Roche. PL is an employee of Roche products Ltd, UK. MY, NFL and SA are employees of Genentech, and hold stock in the company. AB-G and KP have no conflicts of interest to disclose.

References

1
Scott
DL
Smith
C
Kingsley
G
Joint damage and disability in rheumatoid arthritis: an updated systematic review
Clin Exp Rheumatol
 , 
2003
, vol. 
21
 (pg. 
S20
-
27
)
2
Tugwell
P
Wells
G
Strand
V
, et al.  . 
Clinical improvement as reflected in measures of function and health-related quality of life following treatment with leflunomide compared with methotrexate in patients with rheumatoid arthritis: sensitivity and relative efficiency to detect a treatment effect in a twelve-month placebo-controlled trial comparing leflunomide with methotrexate. Leflunomide Rheumatoid Arthritis Investigators Group
Arthritis Rheum
 , 
2000
, vol. 
43
 (pg. 
506
-
14
Erratum: Arthritis Rheum 2000;43:2615–6.
3
Strand
V
Cohen
S
Crawford
B
Smolen
JS
Scott
DL
Patient-reported outcomes better discriminate active treatment from placebo in randomized controlled trials in rheumatoid arthritis
Rheumatology
 , 
2004
, vol. 
43
 (pg. 
640
-
7
)
4
Cohen
SB
Strand
V
Aguilar
D
Ofman
JJ
Patient- versus physician-reported outcomes in rheumatoid arthritis patients treated with recombinant interleukin-1 receptor antagonist (anakinra) therapy
Rheumatology
 , 
2004
, vol. 
43
 (pg. 
704
-
11
)
5
Shanahan
JC
Moreland
LW
Carter
RH
Upcoming biologic agents for the treatment of rheumatic diseases
Curr Opin Rheumatol
 , 
2003
, vol. 
15
 (pg. 
226
-
36
)
6
Mpofu
S
Fatima
F
Moots
RJ
Anti-TNF-α therapies: they are all the same (aren't they?)
Rheumatology
 , 
2005
, vol. 
44
 (pg. 
271
-
3
)
7
Gause
A
Berek
C
Role of B cells in the pathogenesis of rheumatoid arthritis: potential implications for treatment
BioDrugs
 , 
2001
, vol. 
15
 (pg. 
73
-
9
)
8
Dörner
T
Burmester
GR
The role of B cells in rheumatoid arthritis: mechanisms and therapeutic targets
Curr Opin Rheumatol
 , 
2003
, vol. 
15
 (pg. 
246
-
52
)
9
Edwards
JCW
Cambridge
G
Abrahams
VM
Do self-perpetuating B lymphocytes drive human autoimmune disease?
Immunology
 , 
1999
, vol. 
97
 (pg. 
188
-
96
)
10
Takemura
S
Klimiuk
PA
Braun
A
Goronzy
JJ
Weyand
CM
T cell activation in rheumatoid synovium is B cell dependent
J Immunol
 , 
2001
, vol. 
167
 (pg. 
1072
-
80
)
11
Weyand
CM
Goronzy
JJ
Ectopic germinal center formation in rheumatoid arthritis
Ann NY Acad Sci
 , 
2003
, vol. 
987
 (pg. 
140
-
9
)
12
McLaughlin
P
Grillo-Lopez
AJ
Link
BK
, et al.  . 
Rituximab chimeric anti-CD20 monoclonal antibody therapy for relapsed indolent lymphoma: half of patients respond to a four-dose treatment program
J Clin Oncol
 , 
1998
, vol. 
16
 (pg. 
2825
-
33
)
13
Anderson
DR
Grillo-Lopez
A
Varns
C
Chambers
KS
Hanna
N
Targeted anti-cancer therapy using rituximab, a chimeric anti-CD20 antibody (IDEC-C2B8) in the treatment of non-Hodgkin's B-cell lymphoma
Biochem Soc Trans
 , 
1997
, vol. 
25
 (pg. 
705
-
8
)
14
Clynes
RA
Towers
TL
Presta
LG
Ravetch
JV
Inhibitory Fc receptors modulate in vivo cytotoxicity against tumor targets
Nat Med
 , 
2000
, vol. 
6
 (pg. 
443
-
6
)
15
Gong
Q
Ou
Q
Ye
S
, et al.  . 
Importance of cellular microenvironment and circulatory dynamics in B cell immunotherapy
J Immunol
 , 
2005
, vol. 
174
 (pg. 
817
-
26
)
16
Reff
ME
Carner
K
Chambers
KS
, et al.  . 
Depletion of B cells in vivo by chimeric mouse human monoclonal antibody to CD20
Blood
 , 
1994
, vol. 
83
 (pg. 
433
-
45
)
17
Roitt
I
Bostoff
J
Male
D
Immunology
 , 
1989
2nd
London
Gower Medical Publishing
(pg. 
14.1
-
14.10
Chapter 14
18
Sell
S
Max
EE
Immunology, immunopathology and immunity
 , 
2001:
6th
Washington, USA
ASM Press
(pg. 
101
-
49
Chapter 4
19
Edwards
JC
Szczepanski
L
Szechinski
J
, et al.  . 
Efficacy of B-cell-targeted therapy with rituximab in patients with rheumatoid arthritis
N Engl J Med
 , 
2004
, vol. 
350
 (pg. 
2572
-
81
)
20
Pavelka
K
Nahir
AM
Edwards
JCW
Hessey
E
Saidabadi
N
Shaw
T
Improvement in patient-reported outcomes with rituximab in patients with rheumatoid arthritis
Ann Rheum Dis
 , 
2004
, vol. 
50
 
Suppl 1
(pg. 
289
-
90
)
21
Lawrence
RC
Helmick
CG
Arnett
FC
, et al.  . 
Estimates of the prevalence of arthritis and selected musculoskeletal diseases in the United States
Arthritis Rheum
 , 
1998
, vol. 
41
 (pg. 
778
-
99
)
22
Felson
DT
Anderson
JJ
Boers
M
, et al.  . 
The American College of Rheumatology preliminary core set of disease activity measures for rheumatoid arthritis clinical trials
Arthritis Rheum
 , 
1993
, vol. 
36
 (pg. 
729
-
40
)
23
Prevoo
ML
van't Hof
MA
Kuper
HH
van Leeuwen
MA
van de Putte
LB
van Rield
PL
Modified disease activity scores that include twenty-eight joint counts: development and validation in a prospective longitudinal study of patients with rheumatoid arthritis
Arthritis Rheum
 , 
1995
, vol. 
38
 (pg. 
44
-
8
)
24
van Gestel
AM
Anderson
JJ
van Riel
PL
, et al.  . 
ACR and EULAR improvement criteria have comparable validity in rheumatoid arthritis trials
J Rheumatol
 , 
1999
, vol. 
26
 (pg. 
705
-
11
)
25
van Riel
PL
Provisional guidelines for measuring disease activity in clinical trials in rheumatoid arthritis
Br J Rheumatol
 , 
1992
, vol. 
31
 (pg. 
793
-
4
)
26
Bruce
B
Fries
JF
The Stanford Health Questionnaire: a review of its history, issues, progress, and documentation
J Rheumatol
 , 
2003
, vol. 
30
 (pg. 
167
-
78
)
27
Kosinski
M
Zhao
SZ
Dedhiya
S
Osterhaus
JT
Ware
JE
Jr
Determining minimally important changes in generic and disease-specific health-related quality of life questionnaires in clinical trials of rheumatoid arthritis
Arthritis Rheum
 , 
2000
, vol. 
43
 (pg. 
1478
-
87
)
28
Krishnan
E
Sokka
T
Hakkinen
A
Hubert
H
Hannonen
P
Normative values for the Health Assessment Questionnaire disability index: benchmarking disability in the general population
Arthritis Rheum
 , 
2004
, vol. 
50
 (pg. 
953
-
60
)
29
Redelmeier
DA
Lorig
K
Assessing the clinical importance of symptomatic improvements. An illustration in rheumatology
Arch Intern Med
 , 
1993
, vol. 
153
 (pg. 
1337
-
42
)
30
Strand V, Scott DL, Emery P, Kalden JR, Smolen JS, Cannon GW, for the Leflunomide Rheumatoid Arthritis Investigators Groups
Physical function and health-related quality of life: analysis of 2-year data from randomized, controlled studies of leflunomide, sulfasalazine, or methotrexate in patients with active rheumatoid arthritis
J Rheumatol
 , 
2005
, vol. 
32
 (pg. 
590
-
601
)
31
Wells
GA
Tugwell
P
Kraag
GR
Baker
PR
Groh
J
Redelmeier
DA
Minimum important difference between patients with rheumatoid arthritis: the patient's perspective
J Rheumatol
 , 
1993
, vol. 
20
 (pg. 
557
-
60
)
32
Zhao
SZ
Feichtner
JI
Tindall
EA
, et al.  . 
Evaluation of health-related quality of life of rheumatoid arthritis patients treated with celecoxib
Arthritis Care Res
 , 
2000
, vol. 
13
 (pg. 
112
-
21
)
33
Cohen
J
Statistical power analysis for the behavioral sciences, 2nd edn. NJ: Lawrence Erlbaum
 , 
1988
34
Emery
P
Sheeran
T
Lehane
PB
Saiedabadi
N
Shaw
TM
Efficacy and safety of rituximab at 2 years following a single treatment in patients with active rheumatoid arthritis
Arthritis Rheum
 , 
2004
, vol. 
50
 
Suppl 9
pg. 
S659
 
35
Strand
V
Counterpoint from the trenches; a pragmatic approach to therapeutic trials in rheumatoid arthritis
Arthritis Rheum
 , 
2004
, vol. 
50
 (pg. 
1344
-
50
)
36
Maini
RN
Breedveld
FC
Kalden
JR
, et al.  . 
Sustained improvement over two years in physical function, structural damage, and signs and symptoms among patients with rheumatoid arthritis treated with infliximab and methotrexate
Arthritis Rheum
 , 
2004
, vol. 
50
 (pg. 
1051
-
65
)
37
Keystone
EC
Kavanaugh
AF
Sharp
JT
, et al.  . 
Radiographic, clinical, and functional outcomes of treatment with adalimumab (a human anti-tumor necrosis factor monoclonal antibody) in patients with active rheumatoid arthritis receiving concomitant methotrexate therapy: a randomized, placebo-controlled, 52-week trial
Arthritis Rheum
 , 
2004
, vol. 
50
 (pg. 
1400
-
11
)
38
Buchbinder
R
Bombardier
C
Yeung
M
Tugwell
P
Which outcome measures should be used in rheumatoid arthritis clinical trials? Clinical and quality of life measures’ responsiveness to treatment in a randomized controlled trial
Arthritis Rheum
 , 
1995
, vol. 
38
 (pg. 
727
-
35
)
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Comments

0 Comments