SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. Acknowledgements
  9. REFERENCES

Objective

To assess the efficacy and safety of R788 (fostamatinib disodium), an inhibitor of spleen tyrosine kinase (Syk), in patients with active rheumatoid arthritis (RA) that did not respond to biologic therapies.

Methods

A total of 219 patients with active RA in whom treatment with biologic agents had failed were enrolled in a 3-month multicenter, randomized, double-blind, placebo-controlled trial of R788. The primary end point was the percentage of patients who met the American College of Rheumatology 20% improvement criteria (achieved an ACR20 response) at month 3. Secondary end points included changes in inflammation and damage, as assessed by magnetic resonance imaging (MRI), and changes in the Disease Activity Score.

Results

The ACR20 response in the R788 100 mg twice daily group was 38%, versus 37% in the placebo group, at month 3. No significant differences were achieved in the ACR20, ACR50, or ACR70 response levels at 3 months. There were differences between the groups from baseline to month 3 in the secondary end points C-reactive protein (CRP) level and synovitis score on MRI. There were baseline differences in steroid use, prior biologic use, and synovitis score on MRI between the R788 group and the placebo group that may have affected the outcomes. A high placebo response rate was seen in this trial, and exploratory analysis suggested that this may in part have been driven by patients who entered the trial with an elevated erythrocyte sedimentation rate but normal CRP level.

Conclusion

Our findings indicate that there were no differences in the primary end point between the R788 and placebo groups. Differences were observed between the R788 and placebo groups in secondary end points, particularly in those patients who entered the study with an elevated CRP level.

The treatment of rheumatoid arthritis (RA) has traditionally been based on the use of conventional disease-modifying antirheumatic drugs (DMARDs) and, since the late 1990s, on newer, biologically based therapies that inhibit cytokine activity, block T cell–mediated costimulation, or modify B cell biology. These therapies have been effective in RA and have slowed disease progression, particularly when combined with methotrexate (MTX). However, response rates, based on the American College of Rheumatology 20% improvement criteria (ACR20 response) (1), in RA patients in whom one or more biologic approaches have failed have not exceeded 50% (2–5).

Alternative therapies, including the blockade of spleen tyrosine kinase (Syk), are being explored. Syk is an intracellular cytoplasmic tyrosine kinase that is an important mediator of immunoreceptor signaling in macrophages, neutrophils, mast cells, and B cells. In collagen-induced arthritis models and in cell culture studies of human synoviocytes, the Syk inhibitor R788 or R406 (the active metabolite of R788) demonstrated potent antiinflammatory activity, suggesting a role for Syk inhibitors in the treatment of RA (6, 7).

The first trial of the Syk inhibitor R788 in RA was a phase II, randomized, placebo-controlled trial of 189 patients with active RA (8). Three dosages of R788 (50 mg, 100 mg, and 150 mg twice daily) were compared with placebo when added to background MTX. A significant improvement was seen in the two highest dosage groups, 100 mg and 150 mg, as compared with the placebo and 50 mg dosage groups at 12 weeks, with >65% of the patients in the 100 mg and 150 mg dosage arms achieving an ACR20 response as compared with 38% of those in the placebo arm and 32% in the 50 mg dosage arm. The significant AEs were diarrhea, neutropenia, and elevation in blood pressure (BP), which were all dose related (8).

Because of the continued unmet need of patients in whom biologic treatments have failed, we studied R788 in this population with refractory RA. In addition to the standard arthritis activity measurements, magnetic resonance imaging (MRI) was incorporated into the study design to develop a preliminary understanding of the effects of Syk inhibition on synovium and bone.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. Acknowledgements
  9. REFERENCES

Patients and study design.

This was a multicenter, randomized, double-blind, placebo-controlled trial conducted at 49 sites in 7 countries (Belgium, Colombia, France, Germany, Italy, Peru, and the US). Patients met the ACR criteria for RA (9), had been diagnosed as having RA for at least 12 months, and had disease that was currently not responding to or had previously failed to respond to treatment with a biologic agent, such as an inhibitor of tumor necrosis factor, anakinra, abatacept, and/or rituximab, at an approved label dose for >3 months. These treatments were designated as failures on the basis of lack of efficacy, tolerability, or safety. Active RA was defined as ≥6 swollen joints (28-joint count), ≥6 tender joints (28-joint count), and at least 1 of the following: erythrocyte sedimentation rate (ESR) greater than the upper limit of normal (ULN) for the local laboratory or C-reactive protein (CRP) level greater than the ULN for the central reference laboratory (>8.0 mg/liter). DMARDs, including MTX, leflunomide, sulfasalazine, chloroquine, hydroxychloroquine, nonsteroidal antiinflammatory drugs, and oral corticosteroids (prednisone ≤10 mg/day or equivalent) were permitted if dosages were stable for 30 days prior to randomization. The MTX dosage must have been stable for 3 months and folic acid or folinic acid dosages stable for 6 weeks. A washout period for biologic agents was required. Anakinra, etanercept, abatacept, and adalimumab required a 30-day washout, infliximab required a 60-day washout, and rituximab required a 180-day washout and normal CD19 count prior to day 1.

Exclusion criteria included active infection or untreated latent infection, presence of hepatitis B surface antigen or hepatitis C, a history of cancer (other than nonmelanomatous skin cancers) within the previous 5 years, uncontrolled hypertension, or other major medical conditions. Patients with a serum alanine aminotransferase (ALT) level >1.2 times the ULN, hemoglobin <10 gm/dl, platelet count <125,000/mm3, or creatinine level greater than the ULN were also excluded.

Study protocol.

Two hundred nineteen patients were randomly assigned in a 2:1 ratio to 1 of 2 oral dosing groups: R788 100 mg twice daily or placebo twice daily. Efficacy and safety were evaluated over 3 months. A trained independent joint assessor (who was blinded with regard to laboratory results and treatment assignment) was responsible for performing joint counts and completing the global assessment of disease activity. MRI of the hands and wrists was performed at baseline and at month 3. One hand and wrist of each patient was imaged using a 1.5T whole-body MRI scanner. Standardized pulse sequences, including coronal STIR and 3-dimensional gradient-echo scans, were acquired before and after intravenous injection of gadolinium diethylenetriaminepentaacetic acid (0.1 mmoles/kg). Image acquisition was centrally quality controlled, and images were centrally read by 2 independent radiologists who were blinded with regard to visit sequence and treatment. Interreader agreement, as measured by intraclass correlation coefficient, was excellent (erosion score 0.87, osteitis score 0.80, and synovitis score 0.72). The study was conducted in accordance with the ethical principles of the Declaration of Helsinki and was approved by the Institutional Review Board of each study center. The authors had full access to the data and certify the completeness and veracity of the data and the data analysis.

Measures of clinical efficacy.

The primary end point was the ACR20 response at month 3 (1). Secondary measures included the ACR50 and ACR70 responses, improvements in individual ACR components and the Disease Activity Score in 28 joints (DAS28) (10), and changes from baseline to month 3 in radiologic/structural responses, as assessed by the Outcome Measures in Rheumatology Clinical Trials (OMERACT) RA MRI Scoring (RAMRIS) method (11).

Safety measures.

Patients were monitored at weeks 1, 2, 4, 6, 8, and 12 for changes in vital signs, clinical and laboratory findings, adverse events (AE), and serious AEs (SAEs).

The study drug dose was reduced by 50% in a blinded manner if the patient developed ALT elevations >3–5 times the ULN, absolute neutrophil counts <1,500 mm3, unacceptable gastrointestinal (GI) symptoms (nausea, vomiting, or diarrhea), or a sustained increase in BP. All vascular events or potential vascular events were submitted for a blinded adjudication.

Statistical analysis.

The primary end point was the ACR20 response rate at month 3. Patients who withdrew prior to month 3 were considered to be nonresponders. Differences between treatment groups were compared using Pearson's chi-square test. ACR20, ACR50, and ACR70 responses were compared between the R788 treatment group and the placebo group using a Pearson's chi-square test. Time plots of the proportion of patients who were ACR20, ACR50, and ACR70 responders were provided by treatment group. Changes from baseline in swollen and tender joint counts and CRP level were compared using the stratified Wilcoxon rank sum test, with geographic region as strata between the R788 treatment group and placebo group. Patients from the intent-to-treat population were studied in a post hoc analysis of time to first ACR20 response using a Cox proportional hazards model. Patients in whom an ACR20 response was never achieved were censored at their last patient visit. Covariates that were chosen for consideration in the model were study treatment (R788 100 mg twice daily versus placebo), use of prednisone during screening, number of prior biologic agents used for treatment of RA, and baseline synovitis score on MRI. The number of prior biologic medications was categorized into 2 groups (1 prior biologic versus >1 prior biologic). Two-way interaction terms were considered for treatment by prior biologic categorization, treatment by synovitis score, and prior biologic by synovitis score. Using the parameters from the Cox regression model, predicted times to an ACR20 response were generated by inserting values for the baseline RAMRIS synovitis score to explore how different values affected the time to response. The 25th, 50th, and 75th percentiles for the synovitis distribution were chosen to show curves for patients. SAS procedure PHREG, version 9.1 was used to perform the analyses.

The sample size was based on the ACR20 response rate at month 3 comparing the R788 100 mg twice daily treatment group and the placebo group. Assuming an ACR20 response rate in the placebo group of 38%, 140 patients in the active treatment group and 70 patients in the placebo group provided 85% power to detect a difference of 22% or more between the active and placebo treatment groups, based on the use of a 2-sided test at a 0.05 significance level.

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. Acknowledgements
  9. REFERENCES

Randomization and withdrawals.

A total of 219 patients were enrolled, with 146 patients receiving R788 at 100 mg twice daily and 73 receiving placebo twice daily. Of the patients receiving placebo, 14% failed to complete the study, compared with 15% of the patients receiving R788. The predominant reason for withdrawal in the placebo group was lack of efficacy (11%); in the R788 group, the main reasons were AEs and lack of efficacy (6% and 3%, respectively).

Baseline demographic characteristics.

Baseline demographic features and disease activity are summarized in Table 1. The population was 80% women, had a mean age of 56 years, and had a mean disease duration of >11 years. A higher proportion of patients in the R788 group were receiving prednisone (42% with a mean daily dosage of 8.6 mg versus 31% with a mean daily dosage of 7.2 mg in the placebo group) and had disease that failed to respond to ≥3 biologic therapies (21% versus 12% in the placebo group). The mean erosion, osteitis, and synovitis OMERACT RAMRIS scores at baseline were all considerably higher in the R788 group than in the placebo group despite similar levels of clinical severity between the groups.

Table 1. Baseline demographic and disease characteristics*
 Placebo (n = 73)R788 100 mg twice daily (n = 146)
  • *

    Except where indicated otherwise, values are the mean ± SD. CRP = C-reactive protein; RF = rheumatoid factor; TNF = tumor necrosis factor; ESR = erythrocyte sedimentation rate; HAQ DI = Health Assessment Questionnaire disability index; DAS28 = Disease Activity Score in 28 joints; RAMRIS = Rheumatoid Arthritis Magnetic Resonance Imaging Scoring.

Age, mean (range) years56 (22–79)56 (18–82)
% female80.880.1
% white78.156.8
% Hispanic16.437.7
Disease duration, mean years11.312.3
% with elevated CRP69.969.2
% RF positive85.582.4
% taking prednisone (<10 mg/day)31.542.5
% with 1 or 2 prior biologic agents87.778.8
% with ≥3 prior biologic agents12.321.3
No. unresponsive to TNF inhibitors60143
No. unresponsive to >1 TNF inhibitor1732
No. unresponsive to abatacept1733
No. unresponsive to rituximab1325
No. of swollen joints (28 assessed)12.7 ± 5.212.3 ± 5.1
No. of tender joints (28 assessed)16.1 ± 6.715.7 ± 6.7
CRP, mg/liter16.52 ± 20.619.96 ± 27.37
ESR, mm/hour40.7 ± 20.043.1 ± 23.5
HAQ DI (scale 0–3)1.69 ± 0.601.69 ± 0.70
DAS286.19 ± 0.846.23 ± 0.87
RAMRIS synovitis score5.69 ± 4.097.46 ± 4.67
RAMRIS osteitis score6.26 ± 7.548.40 ± 9.13
RAMRIS erosion score13.89 ± 13.8319.32 ± 17.06

Reductions in study drug dosage.

Two patients (3%) in the placebo group had their drug dosage reduced, compared with 21 patients (14%) in the R788 group. The most common reasons for a reduction in the dosage of R788 were GI symptoms (5%) and BP elevations (5%), followed by serum ALT elevations (3%) and transient neutropenia (2%). Of the 21 patients receiving R788 whose dosage was reduced, 17 patients completed the study at the lower dosage of 100 mg per day; 9 of those patients who received a reduced dosage of R788 achieved an ACR20 response at month 3. Four patients failed to complete the study due to GI symptoms (n = 2) and BP elevations (n = 2). One patient in the placebo group withdrew from the study due to BP elevations.

Clinical efficacy.

Improvement of signs and symptoms.

No significant differences were seen between the R788 group and the placebo group in the ACR20, ACR50, or ACR70 response or change from baseline in DAS28 at month 3 (Table 2). ACR20, ACR50, and ACR70 responses were achieved in 38%, 22%, and 9%, respectively, of the patients in the R788 group versus 37%, 12%, and 5%, respectively, of the patients in the placebo group. The mean change in DAS28 was −1.62 in the R788 group and −1.27 in the placebo group. Although significant differences in ACR20 responses were noted at week 6 (P = 0.003), an increasing placebo response at months 2 and 3 resulted in the failure to distinguish differences between the R788 and placebo groups at later time points (data not shown). Changes in individual components in the ACR responder index are shown in Table 3. Notwithstanding the above, significant changes from baseline in CRP level and ESR at all time points were achieved in the R788 100 mg twice daily group versus the placebo group (P = 0.003 and P = 0.004 at month 3, respectively) (Table 3).

Table 2. Response rates at month 3*
 Placebo (n = 73)R788 100 mg twice daily (n = 146)P
  • *

    Except where indicated otherwise, values are the percent of patients. ACR20 = American College of Rheumatology 20% improvement criteria; DAS28 = Disease Activity Score in 28 joints.

ACR2037380.84
ACR5012220.09
ACR70590.37
DAS28 <2.61012
DAS28 <3.21822
DAS28, mean ± SD change from baseline−1.27 ± 1.7−1.62 ± 1.40.15
Table 3. Change in signs and symptoms of rheumatoid arthritis from baseline to month 3*
 Placebo (n = 73)R788 100 mg twice daily (n = 146)
  • *

    Values are the range. HAQ = Health Assessment Questionnaire; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate.

Tender joint count (range 0–28)−5.9, 8.8−6.7, 7.3
Swollen joint count (range 0–28)−4.5, 7.4−5.2, 5.8
Patient's assessment of pain (range 0–100)−13.3, 31.9−17.2, 32.4
Patient's global disease assessment (range 0–100)−11.7, 33.7−18.1, 31.6
Physician global disease assessment (range 0–100)−20.3, 26.5−26.0, 28.2
HAQ (range 0–3)−0.25, 0.56−0.34, 0.59
CRP, mg/liter3.62, 18.35−6.87, 29.77
ESR, mm/hour1.41, 25.2−8.9, 21.3
Improvement on MRI.

At month 3, R788 improved disease progression on MRI, as measured by the mean change in the synovitis score from baseline to month 3 (−0.5 in the R788 group versus +0.4 in the placebo group [P = 0.038]) and the mean change in the osteitis score from baseline to month 3 (−0.2 in the R788 group versus +1.2 in the placebo group [P = 0.058]) (Figure 1). However, both groups showed progression of bone erosion, with no statistically significant difference between the groups in terms of the mean change in the bone erosion score from baseline to month 3 (0.8 in the R788 group versus 0.9 in the placebo group [P = 0.616]) (Figure 1).

thumbnail image

Figure 1. Mean change from baseline to month 3 in synovitis, osteitis, and erosion scores in patients receiving placebo (shaded bars) and patients receiving R788 100 mg twice daily (solid bars). RAMRIS = Rheumatoid Arthritis Magnetic Resonance Imaging Scoring.

Download figure to PowerPoint

Exploratory analysis.

In an attempt to understand the discrepancy between the weak impact of R788 on signs and symptoms and its notable effect on the objective measures of CRP level and the synovitis and osteitis scores on MRI, additional exploratory analyses were performed. An unusual cluster of patients from two sites with high enrollment numbers entered the study with high ESRs but normal CRP levels. Furthermore, there was a high placebo response rate observed in patients from these same sites. We subsequently assessed clinical outcomes according to whether the patient qualified for the study by CRP level or ESR.

Although the proportion of patients who qualified with an elevated CRP level was balanced between the R788 and placebo groups (69% versus 70%, respectively), the ACR responses of those patients who qualified by CRP level and those who qualified by ESR were notably different. Twenty-six percent of the patients receiving placebo and 42% of the patients receiving R788 who qualified with an elevated CRP level at baseline or screening had an ACR20 response at month 3 (P = 0.051). In contrast, 64% of patients in the placebo group who qualified for entry into the study with an elevated ESR and a normal CRP level achieved an ACR20 response at month 3 compared with 32% of the patients in the R788 group (Figure 2). Moreover, the mean change from baseline in the DAS28 was significantly greater at month 3 in the R788 group (−1.67) than in the placebo group (−0.82) in those patients who qualified for the study with an elevated CRP level (P = 0.002).

thumbnail image

Figure 2. Percentages of patients receiving placebo (shaded bars) or R788 100 mg twice daily (solid bars) who met the American College of Rheumatology 20% improvement criteria (ACR20), met the ACR50 criteria, had a Disease Activity Score in 28 joints (DAS) of <3.2, or had a DAS of <2.6 at month 3, stratified according to the presence of an elevated C-reactive protein (CRP) level and a normal or elevated erythrocyte sedimentation rate (ESR) at baseline (CRP qualifying) or the presence of an elevated ESR but normal CRP level at baseline (ESR qualifying). P values are for R788 versus placebo. NS = not significant.

Download figure to PowerPoint

In addition, a post hoc analysis of time to first ACR20 response was performed using a Cox proportional hazards model. Of the independent variables, treatment, prior biologic categorization, baseline synovitis score, and treatment by baseline synovitis score were statistically significant predictors of time to ACR20 response (P < 0.05). For the complete data set, the sample statistics for the baseline synovitis score were (mean ± SD) 6.84 ± 4.54 and n = 185. Using the final model including independent variables for treatment, prior biologic categorization, baseline synovitis score, and synovitis score by treatment interaction, we inserted the synovitis score from the 25th, 50th, and 75th percentiles into the predicted model to explore how the different baseline synovitis scores affected the predicted time to ACR20 response. These values were considered to be adequate to show the various time-to-response curves for patients in the middle 50% of the synovitis score distribution. Kaplan-Meier curves of the time to ACR20 response based on these baseline synovitis percentiles are shown in Figure 3.

thumbnail image

Figure 3. Cox regression predictions of time to an American College of Rheumatology 20% improvement criteria (ACR20) response (survival estimate) using the 25th percentile (top), 50th percentile (middle), and 75th percentile (bottom) of the baseline Rheumatoid Arthritis Magnetic Resonance Imaging Scoring system synovitis score. The model included treatment, number of prior failed biologic agents (prior bio), baseline synovitis score, and treatment–synovitis score interaction.

Download figure to PowerPoint

At the 25th percentile of synovitis scores (score of 3; low synovitis), the hazard ratio (R788 over placebo) was 0.97, which indicated that patients receiving placebo were 1.025 times more likely to achieve an ACR20 response than those receiving active drug. However, at the 50th percentile (score of 6; greater synovitis), the hazard ratio was 1.344, indicating the benefit of R788 over placebo at that value. Furthermore, at the 75th percentile (score of 9.5), the hazard ratio was 1.955, reflecting that patients with higher levels of synovitis were almost twice as likely to respond to R788 than to placebo (Figure 3). Also, patients who had received only 1 prior biologic agent were 1.51 times more likely to achieve an ACR20 response compared with those who had received >1 prior biologic agent, adjusted for study treatment assignment and baseline synovitis score.

Safety.

Overall, 10 patients (4.6%) withdrew from the study due to AEs, 1 (1%) in the placebo group and 9 (6%) in the R788 group. The most common AEs that resulted in withdrawal from the R788 group were nausea and diarrhea. A total of 14 patients (6%) experienced an SAE. The most common SAEs were related to infections (3 in the R788 and 1 in the placebo group). There were no opportunistic infections. One patient in the placebo group died of septic shock.

Overall, 77% of the patients in the R788 group and 52% of the patients in the placebo group had ≥1 AE during the study. The most common AEs in the overall safety population were diarrhea and headache. Diarrhea occurred in 17 patients (12%) in the R788 group and 5 (7%) in the placebo group. Headache occurred in 14 patients (10%) in the R788 group and 2 (3%) in the placebo group.

Clinically significant abnormalities in individual laboratory tests.

Treatment with R788 produced mild increases in serum transaminase (ALT and aspartate aminotransferase [AST]) levels. Four patients (3%) in the R788 group and no patients in the placebo group had ALT values greater than 3 times the ULN at any time during the study. Of the 4 patients in the R788 group who had an ALT value greater than 3 times the ULN, 1 withdrew from the study (due to lack of efficacy), and 3 completed the study on a reduced R788 dosage (100 mg per day) without return of ALT or AST elevations.

Absolute neutrophil count <1,500/mm3 at any time during the study was seen in 3 patients (2%) in the R788 group and no patients in the placebo group. In all cases, the absolute neutrophil count returned to >1,500 mm3 within 3–21 days after interrupting or reducing the R788 dosage. No infections were associated with neutropenia. There were no significant effects on serum lipids or renal function.

The frequency of BP ≥140/90 mm Hg was greater in patients treated with R788 than in patients treated with placebo. Increases in systolic and diastolic BP were more pronounced in patients presenting with hypertension at screening or baseline. Overall, the mean increase from baseline in systolic BP (placebo adjusted) for patients receiving R788 was ∼5 mm Hg for the total population (those with and without preexisting hypertension) and 3 mm Hg for those patients without pre-existing hypertension. In all cases, elevated BP was responsive to antihypertensive medications. Of the subset of patients without preexisting hypertension, 1 patient (1%) in the placebo group, and 4 (6%) in the R788 group were treated with an antihypertensive medication.

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. Acknowledgements
  9. REFERENCES

In this multicenter, placebo-controlled trial, we studied a novel small-molecule inhibitor of Syk plus background DMARDs in the treatment of patients with active RA in whom biologic agents had failed. There were no meaningful differences seen in the primary end point between the treatment arm (R788 100 mg twice daily) and the placebo arm. At 12 weeks, the proportion of patients in whom an ACR20 response was achieved was 38% in the R788 arm and 37% in the placebo arm. However, secondary and exploratory analyses suggested a potential response in the R788 treatment arm in this difficult-to-treat population.

R788 had previously been studied in RA in a phase II trial in patients with disease that failed to respond adequately to MTX (8). In that study there appeared to be a dose-related increase in efficacy as well as AEs. While the 150 mg twice daily dosage appeared to have robust efficacy, the side-effect profile, including a dose-dependent increase in hypertension and GI side effects, appeared less than ideal for long-term administration in RA. The 100 mg twice daily dosage did appear to potentially be the most effective and tolerated dosage for the treatment of RA. Because of the continued unmet medical need in patients in whom treatment with biologic agents has failed, we studied the effects of treatment with 100 mg of R788 twice daily in this population. Furthermore, in order to develop an understanding of the effects of Syk inhibition on synovium and bone, we included MRI scores obtained at baseline and at 12 weeks using the RAMRIS system.

With the 100 mg twice daily dosage, few serious side effects were seen, and there were only a limited number of issues with regard to hypertension and GI side effects. The SAE rate over 12 weeks in patients taking background DMARDs was 6% in R788 and 1% in the placebo arm. Serious infections were seen in 3 patients treated with R788 and 1 patient treated with placebo. No malignancies were seen, and only 1 death occurred (in the placebo group). The study agent was well tolerated, with discontinuation rates of 15% in the R788 group versus 14% in the placebo group.

Hypertension was a potential side effect of concern. The study drug dosage was reduced for 7 patients in the R788 arm and 1 patient in the placebo arm. Among all patients in the study, 17% of the patients treated with R788 and 8% of the patients treated with placebo either changed their antihypertensive medication or started a new one. The majority of BP elevations were seen in patients who had a history of hypertension, were hypertensive at baseline, or were taking antihypertensive agents at baseline. Among patients who were normotensive at baseline and had no history of hypertension, little change in mean BP was seen (2.6 mm Hg in the patients treated with R788 versus 0.4 mm Hg in the patients treated with placebo).

GI issues similarly have been a focus of concern with this agent. Seven patients in the R788 group and 1 patient in the placebo group received reduced dosages due to these symptoms. Similarly, 4 patients in the R788 group and 0 patients in the placebo group received reduced dosages due to an elevation of ALT or AST levels greater than 3 times the ULN.

With the tolerability and safety profile within an acceptable range, the main problem in this study was the apparent lack of efficacy. It is possible that R788, which seems to be effective in patients who have an inadequate response to MTX, does not work as well in patients with an inadequate response to biologic agents. Certainly, the inability to discriminate a difference in ACR response between R788 and placebo in this difficult-to-treat population calls into question the potential for clinical applicability in this population. However, there were some factors that may have contributed to the inability to discern a difference in the proportions of responders.

Despite randomization, it appears that patients in the R788 arm had both more active disease and more refractory disease. The R788 arm had more patients in whom treatment with a greater number of biologic agents had failed, greater prednisone use, and significantly higher scores for erosion, synovitis, and osteitis on baseline MRIs. Unfortunately, patients were not stratified by the number of prior biologic exposures, which might have helped better distribute patients with more refractory disease. These baseline differences may have played a role in the inability to distinguish a difference between arms in the primary outcome. Results from a post hoc Cox proportional hazards model suggested that patients with greater synovitis were statistically more likely to benefit from R788. Also, patients who had received only 1 prior biologic drug were 1.51 times more likely to achieve an ACR20 response than were those who had received >1 prior biologic drug when adjusted for study treatment assignment and baseline synovitis score.

Furthermore, differences in responses were observed between patients who entered the study under different inclusion requirements. The entry criteria required the presence of both 6 tender joints and 6 swollen joints. However, patients were also required to have either an elevated CRP level or an elevated ESR. For those patients who qualified via an elevated CRP level, exploratory analysis suggested a meaningful difference between those randomized to R788 and those randomized to placebo, with an ACR20 response at 12 weeks of 42% versus 26%, respectively. If, however, patients qualified with a normal CRP level and an elevated ESR, the ACR20 response at 12 weeks was 32% in the R788 arm and 64% in the placebo arm. This latter group of patients who qualified via ESR comprised >30% of the intent-to-treat population, came from 2 sites, and had a high negative impact on the overall efficacy. Patients qualifying based on ESR, which was measured locally rather than at a central laboratory, might have been subject to less rigorous laboratory standards and may have disproportionately affected the results of the study.

Secondary analyses focused on objective markers of response. Mean changes in CRP level between the active and placebo arms demonstrated clinically and statistically significant differences. MRI analysis similarly demonstrated marked differences in mean changes in synovitis and osteitis scores between the active and placebo arms. However, while erosion scores trended in favor of active treatment, there were no statistically significant differences seen. Possible explanations for this discrepancy include the relatively short duration of the study, the large differences in baseline scores, and the lack of a significant impact of the therapy on erosive change.

Recent studies have verified that osteitis on MRI represents intraosseous inflammation that leads to local bone erosion (12, 13). Accordingly, decreased synovitis and osteitis might be expected to result in subsequently decreased progression of bone erosion and joint destruction. However, how quickly and linearly changes in osteitis or even synovitis translate into changes in bone erosion is not known. Additionally, how quickly synovitis and osteitis responded to R788 in this study is not known. Even if the suppression had occurred early enough to manifest changes in erosion, because of the markedly greater severity of osteitis and synovitis in the R788 group than in the placebo group at baseline, osteitis and synovitis were still 9% and 15% higher, respectively, in the R788 group at 12 weeks, potentially masking some of the effect on bone erosion.

Data from a second concurrent phase IIb trial (14) have provided some support for the belief that R788 may yet play a role in the treatment of patients in whom biologic therapy has previously failed. In that phase II study, 457 patients were randomly assigned in a 2:2:1:1 ratio into 1 of the following 4 oral dosing groups: R788 100 mg twice daily, R788 150 mg once daily, placebo twice daily, and placebo once daily. The efficacy and tolerability of 2 dosages of R788 were compared with placebo at 24 weeks in patients who had active RA despite therapy with MTX. Dosages of 100 mg twice daily and 150 mg once daily were significantly superior to placebo at month 6 (ACR20 responses of 66% and 57% versus 35%, respectively; P < 0.01) (12). Interestingly, a limited number of patients in whom treatment with biologic agents had previously failed (14%) were eligible to enroll in the study. Forty-three percent of these patients in the 100 mg twice daily group achieved an ACR20 response, as compared to 14% in the placebo group (14).

Our study was fraught with seemingly contradictory data. Overall ACR responses failed to demonstrate benefit, suggesting a potential lack of benefit of R788 in this difficult-to-treat population. Alternatively, other factors, such as differences in baseline disease activity, 2:1 randomization, and failure to stratify based on the number of prior treatment failures with biologic agents may have compromised the primary end point. Elevated placebo responses driven from two sites with patients qualifying with abnormal ESRs but normal CRP levels appeared to impact the assessment of clinical benefit. There was objective evidence of some benefit of R788 in reduction in CRP levels, improvements on MRI, and secondary analysis of ACR response and DAS28 response of patients who had entered based on an elevated CRP. Furthermore, safety results appeared to suggest that 100 mg twice daily of R788 was a tolerable dosage for long-term administration in RA. There appears to be sufficient grounds to again study R788 in this patient population with refractory disease, with some important lessons learned from this phase II study.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. Acknowledgements
  9. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Genovese had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Genovese, Kavanaugh, Weinblatt, Peterfy, DiCarlo, Grossbard, Magilavy.

Acquisition of data. Genovese, Kavanaugh, Peterfy, DiCarlo, White, O'Brien, Magilavy.

Analysis and interpretation of data. Genovese, Kavanaugh, Weinblatt, Peterfy, DiCarlo, White, Grossbard, Magilavy.

ROLE OF THE STUDY SPONSOR

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. Acknowledgements
  9. REFERENCES

Rigel was responsible for the creation of the study design in close collaboration with Drs. Genovese, Kavanaugh, Weinblatt, DiCarlo, and Peterfy. Rigel oversaw the data collection and analysis, as well as data interpretation. Rigel actively participated in the review and editing of the manuscript. Rigel approved the content of the submitted manuscript prior to its submission for publication. Publication of the manuscript was not contingent on Rigel having an opportunity to review it prior to its submission.

Acknowledgements

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. Acknowledgements
  9. REFERENCES

We thank Hisham Al-Ugaily, MD (Safety Monitor, Kendle, Inc.), the members of our Data Safety Monitoring Board (Lee Simon, MD, Gary Firestein, MD, and Robert Makuch, PhD), Daria Faustino (Statistical Programmer, Kendle, Inc.), Vanessa Reyna (Clinical Operations, Kendle Inc.), Anne Ponugoti (Clinical Operations, Rigel, Inc.), Michael Sterba (Clinical Operations, Rigel, Inc.), and all of the patients and study investigators and their staff for their support of this study.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. ROLE OF THE STUDY SPONSOR
  8. Acknowledgements
  9. REFERENCES
  • 1
    Felson DT, Anderson JJ, Boers M, Bombardier C, Furst D, Goldsmith C, et al. American College of Rheumatology preliminary definition of improvement in rheumatoid arthritis. Arthritis Rheum 1995; 38: 72735.
  • 2
    Genovese MC, Becker JC, Schiff M, Luggen M, Sherrer Y, Kremer J, et al. Abatacept for rheumatoid arthritis refractory to tumor necrosis factor α inhibition. N Engl J Med 2005; 353: 111423.
  • 3
    Cohen SB, Emery P, Greenwald MW, Dougados M, Furie RA, Genovese MC, et al. Rituximab for rheumatoid arthritis refractory to anti–tumor necrosis factor therapy: results of a multicenter, randomized, double-blind, placebo-controlled, phase III trial evaluating primary efficacy and safety at twenty-four weeks. Arthritis Rheum 2006; 54: 2793806.
  • 4
    Emery P, Keystone E, Tony HP, Cantagrel A, van Vollenhoven R, Sanchez A, et al. IL-6 receptor inhibition with tocilizumab improves treatment outcomes in patients with rheumatoid arthritis refractory to anti–tumour necrosis factor biologicals: results from a 24-week multicentre randomised placebo-controlled trial [published erratum appears in Ann Rheum Dis 2009;68:296]. Ann Rheum Dis 2008; 67: 151623.
  • 5
    Smolen JS, Kay J, Doyle MK, Landewe R, Matteson EL, Wollenhaupt J, et al, GO-AFTER Study Investigators. Golimumab in patients with active rheumatoid arthritis after treatment with tumour necrosis factor alpha inhibitors (GO-AFTER study): a multicentre, randomised, double-blind, placebo-controlled, phase III trial [published erratum appears in Lancet 2009;374:1422]. Lancet 2009; 374: 21021.
  • 6
    Cha HS, Boyle DL, Inoue T, Schoot R, Tak PP, Pine P, et al. A novel spleen tyrosine kinase inhibitor blocks c-Jun N-terminal kinase-mediated gene expression in synoviocytes. J Pharmacol Exp Ther 2006; 317: 5718.
  • 7
    Pine P, Chang B, Schoettler N, Banquerigo M, Wang S, Lau A, et al. Inflammation and bone erosion are suppressed in models of rheumatoid arthritis following treatment with a novel Syk inhibitor. Clin Immunol 2007; 124: 24457.
  • 8
    Weinblatt ME, Kavanaugh A, Burgos-Vargas R, Dikranian AH, Medrano-Ramirez G, Morales-Torres JL, et al. Treatment of rheumatoid arthritis with a Syk kinase inhibitor: a twelve-week, randomized, placebo-controlled trial. Arthritis Rheum 2008; 58: 330918.
  • 9
    Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988; 31: 31524.
  • 10
    Prevoo ML, van 't Hof MA, Kuper HH, van Leeuwen MA, van de Putte LB, van Riel 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; 38: 448.
  • 11
    Ostergaard M, Peterfy C, Conaghan P, McQueen F, Bird P, Ejbjerg B, et al. OMERACT Rheumatoid Arthritis Magnetic Resonance Imaging Studies. Core set of MRI acquisitions, joint pathology definitions, and the OMERACT RA-MRI scoring system. J Rheumatol 2003; 30: 13856.
  • 12
    McQueen FM, Gao A, Ostergaard M, King A, Shalley G, Robinson E, et al. High-grade MRI bone oedema is common within the surgical field in rheumatoid arthritis patients undergoing joint replacement and is associated with osteitis in subchondral bone. Ann Rheum Dis 2007; 66: 15817.
  • 13
    Hetland ML, Ejbjerg B, Horslev-Petersen K, Jacobsen S, Vestergaard A, Jurik AG, et al, and the CIMESTRA Study Group. MRI bone oedema is the strongest predictor of subsequent radiographic progression in early rheumatoid arthritis: results from a 2-year randomised controlled trial (CIMESTRA). Ann Rheum Dis 2009; 68: 38490.
  • 14
    Weinblatt ME, Kavanaugh A, Genovese MC, Musser TK, Grossbard EB, Magilavy DB. An oral spleen tyrosine kinase (Syk) inhibitor for rheumatoid arthritis. N Engl J Med 2010; 363: 130312.