Article Text

Download PDFPDF

Concise report
Effect of adalimumab on joint disease: features of patients with psoriatic arthritis detected by magnetic resonance imaging
  1. A P Anandarajah1,
  2. P Ory2,3,
  3. D Salonen4,
  4. C Feng1,
  5. R L Wong5,
  6. C T Ritchlin1
  1. 1
    Clinical Immunology Research Center, Allergy, Immunology and Rheumatology Research Division, University of Rochester Medical Center, Rochester, New York, USA
  2. 2
    University of Washington, Seattle, Washington, USA
  3. 3
    Highline Community Hospital, Burien, Washington, USA
  4. 4
    University of Toronto, Toronto, Ontario, Canada
  5. 5
    Abbott Laboratories, Parsippany, New Jersey, USA
  1. Correspondence to Dr A P Anandarajah, Clinical Immunology Research Center, Allergy, Immunology and Rheumatology Research Division, University of Rochester Medical Center, 601 Elmwood Avenue, PO Box 695, Rochester, NY 14642, USA; allen_anandarajah{at}


Background: Bone marrow oedema (BMO), synovitis, effusion and joint erosion on magnetic resonance imaging (MRI) may be used as outcome measures in psoriatic arthritis (PsA).

Objective: To assess the impact of adalimumab on BMO, synovitis, effusion and erosions in PsA, as measured by MRI.

Methods: Fifteen patients with active PsA (⩾3 tender and ⩾3 swollen joints) were enrolled in an open-label pilot study. Each received adalimumab subcutaneously every other week for 24 weeks. MRI was obtained at baseline and 24 weeks.

Results: MRI was available for 11 patients, pre and post-therapy. BMO and effusion scores improved markedly after 24 weeks of adalimumab, while no significant change was noted in erosion score. An unanticipated finding, however, was the lack of improvement in the MRI synovitis score.

Conclusions: Improvement in BMO and unchanged erosion scores may explain the “anti-erosive” effects of adalimumab in PsA. Persistence of BMO and synovitis on MRI suggests ongoing disease activity and supports the continuation of long-term anti-TNF therapy.

View Full Text

Statistics from

Psoriatic arthritis (PsA) is a chronic, inflammatory arthritis affecting 10–30% of patients with psoriasis.1 Joint destruction, manifested as erosions on plain radiographs, is frequently noted within the first 2 years of diagnosis of PsA.2 3 4 Magnetic resonance imaging (MRI) provides direct visualisation of bone and soft tissues, has greater sensitivity than radiography for the detection of bone erosions and can detect bone marrow oedema (BMO).5 6 BMO, a common finding in rheumatoid arthritis (RA) that predicts subsequent radiographic damage at 6 years,7 8 9 is present as early as 9 weeks after symptom onset and is also a predictor of physical function at 6 years.10 11 BMO and synovitis, as assessed by MRI, have been used as outcome measures in RA clinical trials.12 13 14 Although the role of MRI to assess treatment response in PsA has not yet been established, it has great potential, because one can visualise not only bone and cartilage, but also entheses and soft tissues.15 16

Studies have demonstrated that adalimumab is an effective treatment for PsA.17 18 19 20 Based on the knowledge that PsA can affect a broad array of joint structures and that MRI can measure improvement after anti-TNF therapy in ankylosing spondylitis and RA,21 22 we hypothesised that adalimumab therapy would lessen BMO, synovitis, effusion, joint erosions and soft tissue inflammation, as assessed by MRI. Therefore, we designed a pilot study to assess the impact of adalimumab therapy on MRI endpoints.


Study design and patients

This study was a 24-week, single-centre, open-label trial, and was approved by the Institutional Review Board at the University of Rochester, NY, Medical Center. All patients provided written consent. Patients with PsA based on the Moll and Wright criteria23 between the ages of 18 and 65 years were eligible. Patients had to demonstrate active disease, defined as greater than or equal to three tender and three swollen joints. Patients with a history of infections, severe or progressive medical illness, or demyelinating lesions, and those who had been treated with other anti-TNF agents or other biological agents were excluded. The 15 patients selected for the study were given subcutaneous injections of adalimumab 40 mg every other week for 24 weeks.

Statistical analyses

The paired t test was used to compare the change in MRI scores from baseline to 24 weeks. The McNemar test was employed to compare the joint feature changes shown in table 1. All analyses were based on 11 patients with both baseline and follow-up data available. Statistical analyses were performed using SAS software version 9.1.

Table 1

Joint features on MRI at baseline and following 24 weeks of adalimumab therapy

Outcome measures

Clinical assessments

Tender (68) and swollen (66) joint counts and the health assessment questionnaire were completed for all patients. The American College of Rheumatology (ACR 20, ACR50 and ACR70) response rates were calculated for week 24.

MRI assessments

The most active wrist or knee joint on clinical examination was selected for MRI evaluation. A gadolinium-enhanced MRI was obtained for all patients, but only 11 of 15 patients had a repeat scan at week 24 available for comparison. MRI studies were performed using a 1.5-T superconductive magnet (GE Signa, Milwaukee, Wisconsin, USA). All patients were examined in at least one orthogonal plane using fast T2-weighted spin echo pulse sequences with fat suppression, three-dimensional gradient recall echo and gadolinium-enhanced images with fat suppression. Field of view varied by the joint imaged. Two musculoskeletal radiologists, blinded to clinical scores and the chronology of scans, interpreted the MRI findings. The presence of BMO and bone erosions was assessed at 13 sites of each wrist and three sites of each knee. Synovitis and effusions were assessed at three sites of the wrists and were determined to be present or absent for knee joints. Each image was also scored for BMO (0–5), erosions (0–5), synovitis (0–3) and effusions (0–5). Scoring for wrists was based on the RA magnetic resonance image scoring system developed by the outcomes measures in rheumatology.24 A consensus score was calculated, before and post-therapy, for each of these features.


All patients in this study were white, with mean disease durations of 14.7 and 5.9 years, respectively, for a diagnosis of psoriasis and PsA, respectively.

Treatment response

Significant improvements versus baseline in numbers of tender (p = 0.001) and swollen joint counts (p<0.001) after 24 weeks of adalimumab were noted. ACR20, ACR50, and ACR70 responses were achieved by 67%, 40% and 20% of patients, respectively. A significant improvement in the median health assessment questionnaire score was also observed (1.6–0.6, p<0.05).

Magnetic resonance imaging

A total of 11 patients (nine wrist and two knee joints) had baseline and follow-up MRI scans.

Bone marrow oedema

At baseline, BMO was detected in seven (six wrists, one knee) of 11 images. Eighty-seven sites were assessable for the presence of BMO (81 wrists, six knees). Twenty-one sites (19 wrists, three knees) were noted to have BMO (24% of assessed sites). An improvement in BMO was observed at 13 sites (all wrists), three were worse (two wrists, one knee), and no change was noted at five sites (four wrists, one knee, see table 1). The total BMO score decreased from 37 at baseline to 13 (65% improvement) at week 24 (table 2). In addition, fig 1 illustrates BMO improvement for one patient following 24 weeks of adalimumab therapy.

Figure 1

Bone marrow oedema (BMO) and soft-tissue inflammation at baseline and after 24 weeks of adalimumab therapy demonstrating an improvement in BMO and soft-tissue inflammation.

Table 2

MRI scores at baseline and following 24 weeks of adalimumab therapy

Joint erosions

Nine of the 11 patients (eight wrists, one knee) had erosions on MRI. A total of 113 of 143 possible sites was assessable. Of these, 35 sites had erosions (34 wrists, one knee). Erosion scores improved at one site (wrist), deteriorated at four sites (all wrists), and did not change at 30 (29 wrists, one knee) (table 1). The joint erosion score increased from 55 to 70 for these nine patients at 24 weeks (table 2).


Ten of 11 patients (eight wrists, two knees) had detectable synovitis. A total of 28 sites (26 wrists and two knees) of the possible 29 sites was assessable for synovitis. Of these, 19 sites had synovitis (17 wrists, two knees). Synovitis scores improved for nine (eight wrists, one knee), deteriorated for seven (six wrists, one knee) and remained the same in three sites (all wrists) (table 1). The synovitis score decreased slightly, from 30 to 29, at 24 weeks (table 2).


Effusion was present in 10 of 11 patients. Twenty-six of a possible 29 sites were assessed. Twenty-two of these sites had an effusion (20 wrists, two knees). Effusion scores improved for 11 (nine wrists and two knees), deteriorated for one (wrist) and was unchanged for 10 sites (all wrists) (table 1). The total effusion score decreased from 36 to 20 after 24 weeks of therapy with adalimumab (table 2).


We found that PsA patients treated with adalimumab demonstrated significant improvements in clinical outcome measures for joint disease at 24 weeks.25 Analyses of MRI revealed a marked improvement in the extent of BMO and effusion following adalimumab therapy. Persistent BMO was detected, however, in most patients after 6 months of treatment. An unanticipated finding was that the overall synovitis score on MRI did not change with therapy, although synovitis did improve at nine of the 19 MRI locations evaluated. Collectively, these findings suggest that BMO may be a better biomarker for treatment response than synovitis. In addition, the results in this small open-label pilot study indicate that disease activity may persist, even in patients who respond clinically to adalimumb.

Recent studies in patients with RA have demonstrated that areas with the greatest BMO signal correlate with areas of intense cellular infiltrates on histology.7 In addition to reflecting disease activity, BMO can predict radiographic progression and joint damage.9 11 26 27 In other studies that applied three-dimensional volumetric analysis, we demonstrated that etanercept therapy was associated with a reduction in the extent of BMO in PsA patients.28 Persistent oedema was noted nonetheless, even in patients with a marked clinical response after 24 weeks of etanercept therapy. These findings are similar to results from our current study, in which, despite a marked reduction in the overall extent of BMO, only four of seven patients had improvements in the overall extent of BMO. Furthermore, none of the patients achieved complete resolution. In contrast, Marzo-Ortega et al29 noted a dramatic improvement in BMO after four infusions of infliximab, with a complete resolution of BMO in seven of nine patients, as measured after 14 weeks of therapy.

Synovitis, on MRI of joints, is a frequent finding in patients with spondyloarthropathies.16 30 Our study detected no overall change in synovitis scores. The extent of synovitis decreased for five of 10 patients, and increased for four patients. Our findings are in contrast to results of other MRI studies in PsA, which showed reductions in synovial volume following therapy with anti-TNF agents.29 31 One potential explanation is that a longer duration of anti-TNF therapy may be necessary to detect significantly decreased synovitis in PsA. Interestingly, a recent study of RA patients treated with infliximab plus methotrexate revealed persistent synovitis on ultrasound examination in patients in clinical remission.32

Joint erosions and effusions are frequent findings in patients with PsA and often appear early in the natural course of the disease.2 3 4 Studies have also demonstrated a marked reduction in joint signs and symptoms and inhibition of radiographic progression following treatment with adalimumab.17 18 19 20 In our study, five of nine patients with baseline erosions demonstrated worsening of erosion scores. However, these scores increased for only five of 35 sites overall (14%) and did not change for 29 locations (83%). The lack of a placebo arm, however, limits any definitive conclusions from these data. Effusion was observed in almost all the patients. The marked improvement in the extent of effusion following adalimumab therapy may explain the decline in swollen joint scores.

In this pilot study, we also observed reductions in the extent of BMO and effusion scores on MRI following 24 weeks of adalimumab therapy. The lack of change in the extent of synovitis, even for patients with improvements in all clinical markers, was an unanticipated finding. We conclude that whereas BMO is a good marker of clinical response in PsA, synovitis measured on static MRI may be a poor marker of clinical response. These findings need to be confirmed in a larger study with a validated MRI scoring method for psoriatic joints.


The authors wish to thank Michael A Nissen, ELS, of Abbott Laboratories, for his editorial assistance in the development of the manuscript. This support was funded by Abbott.


View Abstract


  • Competing interests Declared. APA, CF, CTR, PO and DS are study investigators for Abbott Laboratories. RLW is an Abbott employee.

  • Ethics approval This study was approved by the Institutional Review Board at the University of Rochester, NY, Medical Center.

  • Patient consent Obtained.

  • Provenance and Peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.