Early MRI Measures Independently Predict 1- and 2- year X-ray Progression: Results from a Large Clinical Trial

Background/Purpose:   Early predictors of progression in structural joint damage in RA are lacking. We evaluated if early MRI measures of inflammation and erosion at baseline, 12 and 24wks could predict subsequent progression in structural damage as measured by standard x-ray at 1 and 2 yrs of follow-up among 256 pts from GO-BEFORE, a large randomized trial of golimumab + MTX vs MTX alone in RA pts who were MTX-naïve.

Methods: Methods and results of the original trial have been published ¹. MRIs (contrast-enhanced; 1.5T) of the wrist and the 2^nd-5^th metacarpophalangeal joints of the dominant hand at baseline and wks 12, 24, 52, and 104 were obtained. MRIs were scored by 2 independent, blinded readers using the RA MRI Scoring (RAMRIS) system. X-rays (hands, wrists, forefeet at baseline, wk52, and104) were scored by 2 other, blinded readers using vdHS system.  X-ray progression was defined as a change in vdHS score > 0.5 as it was in the original trial.  MRI synovitis and bone edema scores were evaluated as continuous variables (per unit difference or change). Change in RAMRIS bone erosion scores was highly skewed, and was therefore dichotomized at >0.5. Multivariable logistic regression was used to determine if baseline and early measures of change in component RAMRIS scores predicted x-ray progression independent of clinical disease activity [DAS28(CRP)], change in DAS28(CRP), age, sex, baseline vdHS score, and treatment group.

Results: Higher baseline synovitis scores and less improvement in synovitis over the first 24 wks of follow up were both significantly and independently associated with a greater risk of x-ray progression at 1- and 2yrs (Table). Higher baseline bone edema and less improvement in bone edema were independently associated with a greater risk of x-ray progression at 1yr, and tended to be associated with progression at 2yrs.  An increase in RAMRIS bone erosion score >0.5 at wk 24 significantly predicted x-ray progression at 1- and 2 yrs. Baseline and wk12 changes in MRI scores all significantly predicted x-ray progression at wk 52 (all p<0.05), and tended to be associated with x-ray progression at wk 104 (p= 0.004-0.2).

Conclusion: Early MRI measures at 12 and 24 wks independently predict x-ray changes at 1 and 2yrs of follow-up. These data support the use of MRI in clinical trials for early identification of pts with (OR who will develop) structural joint damage progression during follow up. This has implications for clinical trial design.

Table:  Multivariable-adjusted risk of x-ray progression at 1- and 2 years of follow-up based on early MRI measures at 24 wks (per 1 unit difference or change in respective RAMRIS score).

*adjusted for age, sex, DAS28(CRP) at baseline, change in DAS28(CRP) over the first 24wks, vdHS score at baseline, and treatment group.

1.  Østergaard M, Emery P, Conaghan PG, et al. Arthritis Rheum 2011; 63(12):  3712-3722.