Imaging Biomarker Based Patient Stratification: Initial Data and Validation in Four Most Common Knee Arthritic Diseases

Olga Kubassova¹, Mikael Boesen², Adam Taylor³, Robert Riis⁴, Lars Hornum⁵, Henning Bliddal², Christine Ballegaard⁶ and Else Marie Bartels⁴, ¹R&D, IAG, Image Analysis Group, London, United Kingdom, ²Department of Rheumatology, The Parker Institute, Copenhagen University Hospital at Frederiksberg, Frederiksberg, Denmark, ³IAG, Image Analysis Group, London, United Kingdom, ⁴The Parker Institute, Copenhagen University Hospital at Frederiksberg, Frederiksberg, Denmark, ⁵Novo Nordisk A/S, Måløv, Denmark, ⁶The Parker Institute, Copenhagen University Hospital, Bispebjerg and Frederiksberg, Copenhagen F, Denmark

Meeting: ACR/ARHP Annual Meeting 2018

Keywords: IL-6, Knee and MRI

Session Information

Date: Sunday, October 21, 2018

Title: Imaging of Rheumatic Diseases Poster I: MRI

Session Type: ACR Poster Session A

Session Time: 9:00AM-11:00AM

Background/Purpose:

Biomarker science has advanced to aid in distinguishing between different forms of arthritis: inflammatory arthritides such as rheumatoid arthritis (RA) and psoriatic arthritis (PsA) and OA. Biomarkers are also used to assess disease activity. Diagnostic serum biomarkers such as rheumatoid factor (RF) and cyclic-citrullinated peptide (CCP) and assays of disease activity such as C-reactive protein (CRP), and multi-biomarker assays have utility but lack complete sensitivity and specificity. Increasingly quantitative imaging biomarkers may fill an important gap in disease identification and assessment.

Objectives: 1) To investigate the association between imaging measures of inflammation in the synovium of the knee joint and systemic levels of CRP in patients with RA, PsA and OA. 2) Investigate how imaging and clinical markers correlate to IL-6 levels from joint fluid in different patient cohorts.

Methods:

38 patients with a flare of pain in the knee were recruited. 12 were diagnosed with RF positive (+) RA, 6 with RF negative (-) RA, 6 PsA, and 14 OA, according to ACR/EULAR criteria. CRP in blood and IL-6 levels from joint fluid were determined. Patients underwent MRI, including Dynamic Contrast Enhanced (DCE)-MRI exam prior to an ultrasound-guided arthrocentesis. MRI were scored for synovitis [1] and DCE-MRI were quantified using Dynamic Enhanced MRI Quantification (DEMRIQ) method, extracting the volume of enhancing voxels (Nvoxel), Initial Rate of Enhancement (IRE), Maximum Enhancement (ME). Inflammation was quantified as IRExNvoxels and MExNvoxels [2]. Correlation between all clinical scores and all imaging parameters was done using Spearman rho, with significance levels of p<0.05.

Results:

The imaging markers of perfusion in the synovium of the knee (MExNvoxels and IRExNvoxels) were the only imaging measures, which showed a very high association with CRP in both RF+ RA (r=0.92 / 0.97, p<0.05) and PsA patients (0.93 / 0.99, p<0.05), whereas all other imaging markers of inflammation showed no statistical association with blood levels of CRP in these diseases. We found no association between CRP and any imaging assessed scores of inflammation in either RF- RA or OA. In addition, only RF+ RA patients showed a positive moderate to high association between MExNvoxels and IL-6 (r=0.66, p<0.05) in the knee joint aspirate.

Conclusion:

Quantitative imaging and blood biomarkers of inflammation, such as DCE-MRI parameters and CRP, appear to relate differently to each other in the four most common knee arthritic diseases, RF + RA, RF- RA, PsA and OA. DCE-MRI may have specific utility in differentiating these conditions and their disease activity.

References:

[1] Guermazi A et al.; MRI-based semiquantitative scoring of joint pathology in osteoarthritis. Nat Rev Rheumatol. 2013 Apr;9(4):236–251

[2] Kubassova O et al.; A computer-aided detection system for rheumatoid arthritis MRI data interpretation and quantification of synovial activity. European journal of radiology. Eur J Radiol. 2010 Jun;74(3):e67-72.

Disclosure: O. Kubassova, None; M. Boesen, None; A. Taylor, None; R. Riis, None; L. Hornum, None; H. Bliddal, None; C. Ballegaard, None; E. M. Bartels, None.

To cite this abstract in AMA style:

Kubassova O, Boesen M, Taylor A, Riis R, Hornum L, Bliddal H, Ballegaard C, Bartels EM. Imaging Biomarker Based Patient Stratification: Initial Data and Validation in Four Most Common Knee Arthritic Diseases [abstract]. Arthritis Rheumatol. 2018; 70 (suppl 9). https://acrabstracts.org/abstract/imaging-biomarker-based-patient-stratification-initial-data-and-validation-in-four-most-common-knee-arthritic-diseases/. Accessed .

ACR Meeting Abstracts - https://acrabstracts.org/abstract/imaging-biomarker-based-patient-stratification-initial-data-and-validation-in-four-most-common-knee-arthritic-diseases/