Rheumatoid Arthritis Highlights

Jon Giles, M.D., MPH

Abstract #597: Rates and Predictors of Influenza, Pneumococcal, and Herpes Zoster Vaccination in Adult Patients with Rheumatic Diseases

Authors:

Wolfe and Michaud; Omaha, Nebraska, USA

Background:

Owing to comorbidities and chronic immunosuppression, most patients with rheumatic diseases are appropriate to receive yearly influenza vaccination and periodic pneumococcal vaccination.  In addition, vaccination against varicella zoster is commercially available.  However, how frequently these are administered, and the predictors of receiving vaccination, have not been systematically explored.

Methods:

Self-report of immunization against influenza (yearly), pneumococcus (within 5 years), and zoster (ever) in patients with RA, SLE, fibromyalgia, and other non-inflammatory rheumatic disorders participating in the National Data Bank for Rheumatic Diseases was explored.  Longitudinal data analysis methods (specifically, Generalized Estimating Equations (GEE)) were used to explore demographic and disease-related predictors of immunization among the RA group.

Results:

A total of 25,559 observations from 10,381 individual patients were analyzed. The pooled rate of influenza vaccination was 41% for patients less than 40 years of age.  Within this age group, influenza vaccination was highest for the SLE (approximately 50%) and RA (approximately 40%) groups, and lower for the fibromyalgia and non-inflammatory disease groups.  Influenza vaccination increased with age for all groups.  The pooled influenza vaccination rate was 70% for patients over the age of 65 years, with rates similar between the four disease groups.  For pneumococcal vaccination, the rate of vaccination within the past 5 years was 24% for patients under 40 and increased to 66% for those over 65.  The rate of zoster vaccination was 9% in the group over the age of 65.

Higher vaccination rates were associated with demographics (education, income, non-minority status) and comorbid conditions (diabetes, cardiovascular disease, and pulmonary disease.  RA patients using prednisone were more likely to receive influenza vaccination, and those receiving biologics were more likely to receive influenza and pneumococcal vaccination.

Conclusions:

Immunization rates for RA and SLE patients are not at the levels recommended, particularly for younger patients.

Editorial Comment:

The recent H1N1 virus outbreak has highlighted the importance of vaccination, particularly for the immuncompromised rheumatic disease patient.   While overall vaccination rates in the general population are much less than universal, it is somewhat disappointing to see that rates in RA and SLE groups are not much better than those of the general population.  Within the context of this study, it is encouraging at least to observe that among younger patients, that influenza vaccination rates are highest for the SLE and RA groups, suggesting that some emphasis on vaccination relating to immunocompromise is occurring.  The catch up in vaccination rates for the older non-rheumatic disease patients is likely related to public health campaigns aimed at the general population.  The use of zoster vaccine in immunosuppressed patients remains controversial, so it is no surprise that rates of zoster vaccination are overall low.

Abstract #599: Prevalence and Determinants of Vitamin D Insufficiency/Deficiency in U.S. Veterans with Rheumatoid Arthritis

Authors:

Kerr, Richards, Mikuls, Sabahl, Reimold, Cannon, Johnson, and Caplan; Veterans Affairs Health System, USA

Background:

In addition to well known effects on bone metabolism, vitamin D also participates in immune function.  Thus, individuals with low levels of vitamin D may be at a higher risk for developing systemic autoimmune diseases, such as RA, having more disease related inflammation, and/or having more severe manifestations of disease.  This abstract reports associations of vitamin D status with RA disease characteristics, comorbid conditions, and mortality among veterans receiving care through the U.S. Veterans Administration Health System.

Methods:

Vitamin D status was explored among enrollees in the Veterans Affairs RA Registry (VARA), with a 25-hydroxy vitamin D (25OH-D) level of <30ng/mL defined as insufficient and <20ng/mL defined as deficient.  The associations of demographics, body mass index, and RA characterisitics with vitamin D status, and the subsequent association of vitamin D status with mortality, were explored.

Results:

Among the 1,160 VARA enrollees studied, 90% were male and 77% were Caucasian.  The average age was 64 years.  Vitamin D insufficiency was detected in 84% and deficiency in 45%.  Low 25OH-D level was significantly associated with non-Caucasian ethnicity, younger age, higher BMI, and anti-CCP seropositivity.  These associations remained significant after adjusting for demographics and season.  Other measures of RA disease activity and severity, including markers of systemic inflammation, were not associated with vitamin D status in crude or adjusted analyses.  Vitamin D deficiency/insufficiency was not significantly associated with higher mortality over an average follow-up time of 1.9 years.

Conclusions:

Low vitamin D levels were common among elderly U.S. veterans.  However, low levels were not associated with RA disease activity, severity, or mortality.

Editorial Comment:

This is one of the largest studies to date exploring the potential links between vitamin D status and disease outcomes in RA patients.  From these findings alone, one might be tempted to conclude that vitamin D supplementation may not affect RA disease related outcomes or mortality.  However, caution should used when interpreting the apparent lack of association, considering that vitamin D status was measured in this cohort at one time only in patients with varying disease durations, and that the average follow-up time was relatively short.  Accordingly, an interventional trial is needed to definitively conclude whether vitamin D supplementation leads to less inflammation or better outcomes in RA patients.

Abstract #1190: Does Statin Use Protect Against or Modify RA Disease Activity at Diagnosis?

Authors:

Holmqvist, Gallais, Wedren, Klareskog, Alfredsson, Askling; Stockholm, Sweden

Background:

The HMG-CoA reductase inhibitors (statins) have anti-inflammatory properties and reduce disease activity in RA patients.  However, whether their use reduces the risk of developing RA has not been investigated.

Methods:

A case-control design was used to explore statin use in the 5 years prior to RA diagnosis between incident RA cases in the Swedish Epidemiological Investigation of Arthritis (EIRA) compared to randomly selected age, gender, and geographically matched controls.  The association of statin use and disease activity at diagnosis was explored among EIRA participants followed longitudinally in the Swedish Rheumatology Register.

Results:

There were 1,973 incident RA cases compared to 2,230 matched controls.  Among these, statin use in the 5 years preceding diagnosis was observed in 59 cases and 62 controls, resulting in an odds ratio for statin exposure of 1.0 (95% CI 0.7-1.5) for case vs. controls after adjusting for age, gender, residence, body mass index, and smoking.  These associations were not different in groups stratified by the presence/absence of anti-CCP antibodies.

There were 1,468 incident RA cases followed longitudinally.  Among those with low baseline disease activity (DAS28 <5.1; n=601), 18 were treated with statins at diagnosis.  Among those with high baseline disease activity (DAS28 ≥5.1; n=867), 34 were treated with statins at diagnosis.  The resulting odds ratio for statin treatment was 1.0 (95% CI 0.5 – 1.8) after adjusting for age, gender, residence, smoking, and ischemic heart disease.  These associations were not different in groups stratified by the presence/absence of anti-CCP antibodies.

Conclusions:

The use of statins prior to RA diagnosis was not associated with a reduction in incident RA.  In addition, RA disease activity at presentation did not differ by statin use.

Editorial Comment:

The study of pre-RA diagnosis risk factors is challenging, but well served by the study design employed here.  The lack of a protective effect of statins could be related to a several factors.  For one, statins may provide no immunologic protection against the development of RA, nor reduce initial disease activity.  Other possibilities are various biases related to the study design, including information bias (limited data on dose and compliance with statin therapy) and confounding by indication (since the allocation of statins to individuals with hypercholesterolemia may be related to the outcome).  Also, the use of statins is quite low in the overall group, reflecting the age distribution of incident cases of RA.

Abstract #1160: Moderate to Severe Adult Periodontitis Increases the Risk of Rheumatoid Arthritis in Non-Smokers and is Associated with Elevated ACPA Titers: The ARIC Study

Authors:

Molitor, Alonso, Wener, Michalowicz, Beck, Gersuk, Buckner, Folsom; Minneapolis, Seattle, Chapel Hill, USA

Background:

There is a potential link between periodontal disease, smoking, and epithelial modification of proteins (e.g. citrullination) that is a plausible mechanism for the initiation phases preceding the development of RA, particularly for the development of RA characterized by the presence of antibodies directed against citrullinated proteins (i.e. anti-CCP antibodies).

Methods:

Incident and prevalent RA were explored among participants enrolled in the Atherosclerosis Risk in Communities (ARIC) study who had detailed periodontal exams.  The associations of periodontal disease (classified as none or mild (less than 10% of teeth with attachment loss ≥ 3mm); moderate (10-30% of teeth with attachment loss ≥ 3mm); or severe (>30% of teeth with attachment loss ≥ 3mm) with incident and prevalent RA, according to smoking, and shared epitope/autoantibody status (HLA DR4 alleles, RF, and anti- CCP antibodies) were explored using multivariate Cox proportional hazards modeling.  Hospital records were used to determine incident and prevalent RA cases.

Results:

Among the 6,931 ARIC participants with dental exams, 6,661 records were available to search for RA cases.  Among these, there were 45 RA cases, 33 of which were incident cases.  Among the 33 incident cases of RA, 27 had moderate to severe periodontitis and 6 had none or mild periodontitis.  The odds of prevalent RA was 5-fold higher for never smokers with moderate to severe periodontitis compared to never smokers with no or mild periodontitis (OR 5.3 (95% CI 1.2-23.8)) after adjusting for age, gender, and race.  The adjusted risk of incident RA was more than 2-fold higher for never smokers with moderate to severe periodontitis compared to never smokers with no or mild periodontitis (OR 2.6 (95% CI 1.0-6.4)).  However, none of these associations was significant in current or former smokers.

Mean anti-CCP antibody levels with higher in RA cases with periodontitis compared to those with no or mold periodontitis (223 vs. 8 units, respectively; p=0.04).  The majority of anti-CCP positive RA cases were smokers with concomitant moderate to severe periodontitis (11/13 or 85%; p=0.007).

Conclusions:

Moderate to severe periodontitis was associated with a higher risk of developing RA among non-smokers, but not smokers, and was associated with higher titers of anti-CCP antibodies.

Editorial Comment:

The link between periodontal disease, smoking, and the generation of an autoantibody response that may predate the development of RA is compelling.  Although this investigation suggests a mechanistic link, there are several issues that make it less than definitive.  For one, the lack of an association in smokers is counter to the mechanistic hypothesis.  Second, despite having periodontal data on a very large cohort, the incidence of RA was still quite low.  Because the rate was lower than what would expect for a group with the demographic characteristics of ARIC, then there is the possibility that incident cases of RA were not included, potentially biasing the reported findings.  However, this misclassification would likely tend to bias the findings towards the null hypothesis.

Abstract #1161: The Mechanisms Underlying Arthritogenicity of Human Anti-Citrulline Antibodies

Authors:

Britnell, Bel, and Cairns; London, Ontario, Canada

Background:

Antibodies against citrullinated peptides circulate in many, but not all, RA patients.  In addition, articular tissues in RA patients contain abundant citrullinated proteins.  However, it is unclear if or how a biologic interaction between articular citrullinated proteins and anti-CCP antibodies mediates articular inflammation in damage in RA patients.

Methods:

An IgG anti-citrulline antibody was affinity purified against a citrullinated peptide similar to CCP2 from 10 RA patients seropositive for anti-CCP antibodies.  The antibody was administered into the peritoneal cavity of normal mice and mice genetically lacking the Fcγ IIB receptor (FcγRIIB -/-).  The mice were monitored for the development of inflammatory arthritis, and articular tissues were explored for inflammation, citrullination, and complement deposition using immunohistochemistry.

Results:

In FcγRIIB-/- mice, intraperitoneal administration of human anti-citrulline antibody induced an inflammatory arthritis, with dose dependent synovial proliferation and inflammatory cell infiltrate on immunohistochemistry.  Co-staining for citrulline and C3 revealed complement deposition in area staining for citrulline.  Non-immunized FcγRIIB-/- mice also demonstrated synovial citrullination.  Antibodies against human IgG were not arthritogenic. In contrast, wild type mice did not demonstrate synovial citrullination, nor did they develop arthritis when immunized with anti-citrulline antibody unless intraarticular citrullinated fibrinogen was administered.

Conclusions:

Binding of human anti-citrulline antibodies to intraarticular citrulline induced complement deposition, inflammation, synovial proliferation, and arthritis in mice.

Editorial Comment:

These are a very elegant series of experiments that help to define the relationship between a potential pathogenic antigen/antibody response.  Whether these experimental relationships in mice are similar in humans remains to be answered.  It is not known from these experiments why the FcγRIIB-/- mice spontaneously demonstrate synovial citrullination, an essential step for the development of arthritis in this system.  In addition, how antibody binding induces an inflammatory response is also not known.  The fact that not all RA patients have antibodies against citrullinated proteins likely means that this mechanism, if present in humans, is not the only process driving synovial inflammation.

Abstract #1163: Citrullinated Fibrinogen Stimulates TNF Release Via TLR-4 and Citrullinated Fibrinogen Immune Complexes Co-Stimulate through TLR-4 and Fc Gamma Receptor

Authors:

Sokolove, Zhao, and Robinson; Palo Alto, California, USA

Background:

The immune effector mechanisms linking an antigen/antibody interaction between synovial citrullinated proteins and citrulline specific antibodies with downstream articular inflammation and damage in RA have not been fully elucidated.  This abstract reports the innate immune signaling response of TLR-4 and Fc receptors when exposed to immune complexes containing modified and unmodified fibrinogen.

Methods:

Murine peritoneal macrophages from wild type and TLR-4 deficient mice were treated with increasing concentrations of native and citrullinated fibrinogen, lipopolysacharride (LPS), citrullination buffer, or immune complexes produced by combining native or citrullinated fibrinogen with rabbit anti-human polyclonal anti-fibrinogen antibody and analyzed for TNF expression.

Results:

Dose dependent stimulation of TNF expression was noted  on incubation of murine macrophages with murine native and citrullinated fibrinogen; however, citrullinated fibrinogen was 10 times more potent in stimulating macrophage TNF production.  The effect  of native and citrullinated fibrinogen on TNF production was negated in mice lacking TLR-4.  A similar effect was noted when murine macrophages were exposed to human native and citrullinated fibrinogen over a threshold level of 3.1 μg/mL.  Immune complexes consisting of antibody and citrullinated fibrinogen also stimulated macrophage TNF production, and effect that was abrogated in TLR-4 deficient mice and in mice lacking MyD88 (a protein that mediates downstream inflammatory activation upon signaling through TLR-4).

Conclusions:

Post-translational citrullination of fibrinogen may increase the potency of the peptide to induce inflammation through innate immune signaling mechanisms.

Editorial Comment:

Understanding the mechanisms involved in the initiation and propagation of articular inflammation may be keys to unlocking the processes at the very core of RA.  These experiments point to several potential key steps in which thresholds for articular inflammation may be surpasses: the transition from native to citrullinated articular peptides and the formation of immune complexes containing these citrullinated peptides and citrulline-specific antibody.  As with the previous abstract (1161), the fact that anti-citrulline antibodies are not present in all RA patients means that the mechanisms identified by this abstract are not necessarily the sole cause of articular inflammation in RA.  However, the study does identify TLR-4 as a interesting potential therapeutic target in RA.

Abstract #1164 Dosage of the HLA Shared Epitope Alleles in Relationship to Anti-Citrullinated Protein Antibodies and Radiological Damage in Rheumatoid Arthritis

Authors:

Scherer, van der Woude, Syversen, van der Linden, Lie, Huizinga, van der Heijde, van der Helm-van Mil, Kvein, and Toes; Berlin, Germany

Background:

Specific genetic variants coding the binding region of MHC class II (known as the shared epitope (SE) are associated with the development of RA, the presence of anti-CCP antibodies, and the severity of articular damage in RA patients.  Anti-CCP antibodies in RA patients are directed against a variety of citrullinated protein epitopes in the joint that could mediate the known association between anti-CCP antibodies and erosive damage.

Methods:

Two cohorts with genotyping for SE and longitudinal radiographic assessment of articular damage, scored using the modified Sharp method, were studied.  In the first, the Leiden Early Arthritis cohort (n=150 anti-CCP seropositive patients) had serum reactivity to five citrullinated assessed using ELISA.  The combined associations of the specific anti-citrullinated peptide antibodies, the presence of SE, and longitudinal radiographic damage were explored within the Leiden cohort.  The association of anti-CCP antibody seropositivity, SE, and radiographic damage was further explored among 238 patients enrolled in the Norwegian EURODISS cohort.

Results:

Among the five citrullinated antigens studied (vimentin 1-16, vimentin 59-74, fibrinogen α27-34, fibrinogen β36-52, and enolase 5-20) only vimentin 59-74 and enolase 5-20 were significantly associated with the presence of SE in a dose dependent manner.  Specifically, 79% of patients with two SE alleles demonstrated serum vimentin 59-74 activity, compared to 16% of SE-/- patients, and 40% of SE+/- patients (p<0.001) and 45% of patients with two SE alleles demonstrated serum enolase 5-20 activity compared to 13% of SE-/- patients and 16% of SE+/- patients (p=0.006).  However, the presence of neither of these peptides (nor the other three not associated with SE) was associated with radiographic progression.  In both cohorts, SE was not predictive of radiographic progression after stratifying for the presence of anti-CCP.

Conclusions:

Although SE alleles were associated with specific citrullinated peptides in RA patients, these peptides were not associated with radiographic progression in a cohort of early RA patients.  Further, SE was not predictive of radiographic progression over and above its association with anti-CCP seropositivity.

Editorial Comment:

Studying the fine specificity anti-CCP activity in RA its relationship to disease related outcomes may lead to insights into disease pathogenesis and potentially guide therapeutics.  Although reactivity to none of the citrullinated peptides studied in this investigation was associated with radiographic damage, this does not rule out the possibility that reactivity to other unstudied antigens might mediate the association.  Indeed, fine specificity of anti-CCP may be a moving target in RA patients due to epitope spreading, making the process of identifying the causative antigens, if they exist, like finding a needle in a haystack.  Thus, a more global approach to identifying antigens may be a more appropriate method than choosing and testing a few specific candidates at a time.

Abstract #1248: Changes in the Rates of Joint Surgery Among Patients with Rheumatoid Arthritis in California, 1983-2007

Authors:

Louie and Ward; Bethesda, Maryland, USA

Background:

The temporal trend over the past decade for earlier, more aggressive treatment of RA aimed at tight control of disease activity should result in the development of fewer disease related outcomes, such as the need for joint reconstructive surgery.  However, not all reports have indicated a corresponding decline in joint surgery.  This abstract investigates rates of several categories of joint surgery commonly performed in RA patients in California between 1983 and 2007.

Methods:

Serial cross-sectional analyses of a population based sample of California residents with RA was conducted exploring temporal trends in the yearly rates of total knee, hip, ankle, and wrist arthroplasty, and wrist and ankle arthrodesis from 1983-2007.

Results:

More than 100,000 RA patients were included.  For all procedures, surgery rates had peaked by 1997, with significantly lower rates noted after 1998.  For both age groups (40-59, and ≥ 60 years of age), the rate of total knee replacement increased slightly for the period 2003-2007 compared to the preceding period 1998-2002.  However, the rates of the other surgical interventions (including total hip replacement) continued to decline during the 2003-2007 period compared to the preceding period.  These trends were mirrored in the general (non-RA population).  Stratification by gender was not associated with a difference in trends.

Conclusions:

Temporal declines in major joint replacement and reconstructive surgery in RA indicate improved outcomes for RA patients

Editorial Comment:

These are encouraging findings, and certainly confirm what has been observed in clinical care.  That several prior reports did not indicate a decline in orthopedic surgery may have reflected a lag between improvement in disease management and a change in surgery rates.  Are these changes solely due to general trends in RA management?  Although these data cannot confirm this, the observation that similar trends in joint s surgery decline were noted in the non-RA population may suggest that other factors may have contributed.  Another interesting finding is the leveling out of total knee replacement rates for the most recent years, suggesting that the nadir of knee replacement rates may have been reached.

Abstract #LB2: Treatment of RA with an Oral Syk Kinase Inhibitor: A 6 Month Randomized Placebo Controlled Phase 2b Study in Patients with Active RA on Chronic Methotrexate

Authors:

Weinblatt, Kavanaugh, Genovese, Grossbard, Magilavy; Rigel Pharma, California, USA

Background:

Modulation of intracellular immune signaling is a potential therapeutic target for the treatment of RA and other rheumatic diseases.  A prior published Phase II investigation with the oral Syk kinase inhibitor, R788, showed promise for this agent (see).  Differing from the previous investigation, a higher dosing option was studied (100 mg twice daily) with outcomes followed longer, out to 24 weeks.

Methods:

Patients with active RA despite treatment with methotrexate were randomized to one of three groups: R788 150 mg once daily; R788 100 mg twice daily; or placebo.  Assessments were performed in double-blind fashion on a background of stable methotrexate for all participants, and stable NSAIDS and/or low dose glucocorticoids for those entering the trial on these agents.  The primary endpoint was the proportion obtaining an ACR20 response at week 24.  Secondary endpoints were the ACR50 and 70 responses, DAS remission, and change in fatigue, physical function, and quality of life (measured using the SF-36).  Safety data were collected throughout and compared between groups.

Results:

A total of 153, 152, and 152 patients were enrolled into the placebo, R788 150 mg QD, and R788 100 mg BID groups, respectively.  The average age of the combined groups was 53 years, with the majority being female (85%).  Caucasians made up only 43% of the groups.  The average disease duration was more than 9 years.  Disease activity was high, with an average DAS28 at baseline of 6.2.  Patient characteristics did not differ according to treatment allocation.  Among the placebo group, 79% completed the 24 weeks of the study compared to 83% and 86% of the active treatment groups.

Primary and selected secondary responses at 24 weeks were significantly greater in the active treatment groups compared to placebo:

 

Placebo

R788 150 mg QD

R788 100 mg BID

ACR20

35%

57%

66%

ACR50

19%

32%

43%

ACR70

10%

14%

28%

DAS Remission

7%

21%

31%

Serious Adverse Events

n=7

n=5

n=13

Diarrhea

3%

12%

19%

Neutropenia

1%

3%

3%

ALT elevation

2%

4%

4%

Additionally, improvements in fatigue, HAQ, and quality of life scores were higher in the active treatment groups compared to placebo.  Drug responses were rapid, with significant differences from placebo as early as the second week.  Most responders were identified within the first 6-8 weeks of use.

There were heterogeneities in response noted.  Those without any prior exposure to biologics had more robust responses.  The were higher responses noted in patients recruited from South America and Eastern Europe; however, differences between active treatment and placebo were similar.

Serious adverse events were not significantly higher in the active treatment groups, and no opportunistic infections were observed.  The most common adverse event was diarrhea, which was dose dependent.  Non-emergent increases in blood pressure were observed in the active treatment groups (an average of 3mm of systolic and diastolic blood pressure) that were observed only in the subgroup with a history of hypertension.

Conclusions:

Treatment with the oral Syk kinase inhibitor R788 was associated with a dose-related improvement in RA disease activity that was superior to placebo in RA patients failing methotrexate monotherapy.  No new safety signals were detected.

Editorial Comment:

Comparing these results to those previously published, this larger, longer-term study confirms the efficacy of this novel immunomodulator over placebo.  The advantages of this agent are its apparent rapidity of effect (within the first few weeks) and its oral administration.  The responses are somewhat lower than what was observed in the 3 month study with lower dosing, with a comparable placebo response noted in both studies.  Even so, the responses are comparable to those observed with other biologics in similar populations.  The higher overall responses (placebo and treatment) noted in the non-U.S. sites is of interest that, while not biasing the comparative findings of the study, makes it problematic to compare the magnitude of treatment effect from this study to other biologic agents.

Abstract #LB3: An Oral Syk Kinase Inhibitor in the Treatment of RA: A 3 Month Randomized Placebo Controlled Phase 2 Study in Patients with Active RA who had Failed Biologic Agents

Authors:

Genovese, Kavanaugh, Peterfy, Magilavy; Rigel Pharma, California, USA

Background: Previous studies have confirmed the superior efficacy over placebo of the oral Syk kinase inhibitor R788 in RA patients with active disease who have failed non-biologic DMARDs (see and LB2).  Responses in patients failing biologics have not been previously reported.

Methods:

In this Phase II study, patients with active RA on stable background non-biologic DMARDs who had failed any previous biologic DMARD due to inefficacy or intolerance were randomized to receive R788 100 mg twice daily or placebo.  The primary endpoint was the proportion obtaining an ACR20 response at week 12.  Secondary endpoints were the ACR50 and 70 responses, DAS remission.  Hand and wrist MRIs were performed at baseline and 12 weeks, with change in synovitis and erosions at 12 weeks as an additional secondary endpoint.  Safety data were collected throughout and compared between groups.

Results:

A total of 146 patients were randomized to R788 and 73 to placebo.  Of these, 85% completed the study to 12 weeks, a proportion that was equal between the groups.  The average age of the combined groups was 56 years, with the majority being female (80%).  Caucasians made up approximately 60% of the groups.  The average disease duration was more than 11.5 years.  Randomization was not successful, as there were imbalances between the groups in the proportion of subjects using prednisone and according to the types of prior biologics.

Primary and most secondary responses at 12 weeks were not significantly greater in the active treatment group compared to placebo:

 

Placebo

R788 100 mg BID

ACR20

37%

38%

ACR50

12%

22%

ACR70

5%

9%

DAS Remission

10%

12%

Change in MRI Synovitis

0.35

-0.52 (p = 0.038)

Change in MRI Erosions

0.94

0.78

Change in MRI Osteitis

1.17

-0.19 (p = 0.058)

Serious Adverse Events

n=1

n=3

Neutropenia

n=0

n=3

Discontinued due to AE

3%

14%

As noted in LB2, placebo responses were higher in non-U.S. sites.  There were differences in response according to those with elevated CRP vs. elevated ESR at baseline in post hoc subgroup analyses.

Conclusions:

R788 was not superior to placebo in the primary outcome of ACR response at 12 weeks in active RA patients with a previous inadequate response to biologic DMARDs

Editorial Comment:

There are a number of potential explanations for the difference in efficacy observed between this study in biologic inadequate responders and others with the same agent in methotrexate non-responders.  One possibility is that a lack of response to biologics identifies a group of RA patients that categorically will not respond to modulation of intracellular tyrosine kinase signaling.  However, a more likely explanation relates to problems with the execution of the study, with unsuccessful randomization and subgroup heterogeneities noted.  Regarding the unsuccessful randomization, a lack of balance on measured characteristics may also reflect imbalances in unmeasured characteristics, any of which may account for the lack of effect observed.  The improvements in MRI parameters are of interest, but these short term changes cannot cancel out the lack of superiority in clinical response.  The post hoc subgroup differences in response according to whether the patients had elevated ESR or CRP are unconvincing.

AddThis Social Bookmark Button