Background Cost-effectiveness of treatments in rheumatoid arthritis (RA) is of growing importance thus, the ideal is to have a tool that help us to select patients most likely to respond to biotechnological drugs.

Objectives To assess the relationship between predicted disease states and the observed amount of improvement applying the remission and low disease activity (LDA) matrix tool presented by MSD as a poster at EULAR 2015¹ to our patients receiving golimumab (GLM).

Methods We identified RA patients who had done GLM as first biological therapy in our hospital. Demographic and clinical data at baseline and at 6 months were collected. The calculation of remission or LDA rates at 6 months of GLM treatment was made based on the proposed prediction matrix tool. Patients were divided into 2 groups by their predicted remission rate at 6 months: lowest chance if <30% and greatest if ≥30%. Thereafter, we compared the amount of improvement in DAS28-4v, tender joint count (TJC), HAQ and Patient Global Assessment of Disease Activity Visual Analog Scale (PGA-VAS) scores observed in each group during 6 months of treatment with GLM.

Results At our hospital, 239 patients with RA are under biological therapy. Of these, 20 had GLM as the first biological agent. Most patients were females (n=16, 80%) and mean age (SD) at start of biological therapy was 50 (11.8) years. Most were non-smoking (n=15, 78.9%). Only 2 patients had rheumatoid factor and anti–citrullinated protein antibodies negative and mean disease duration was 9.5 (5.6) years. The mean of DAS28 4v was 5.71 (1.16). At 6 months we found that only one patient had inactive disease, the remaining showed moderate (n=12) or high (n=4) disease activity; 6 patients did not have EULAR response. When we applied the matrix tool, we found that in these cases only three and one case had less than 20% predicted rate of remission and LDA, respectively. We observe that patients with lowest chance of remission had greater improvement in DAS28-4v, TJC, HAQ, and PGA-VAS scores than those with highest chance, during 6 months of treatment: 1.80 vs 0.54, 8.13 vs 0.27, 0.44 vs 0.26 and 19.63 vs 15.91, respectively (p<0.05 only for DAS28-4v and TJC).

We found 9 cases (45%) of drug discontinuations at 6 months, 8 of which for primary failure and 1 for adverse event (at 1 month). Of the patients that remained under GLM at 6 months, 9 remained beyond 12 months of treatment, 4 of which have achieved remission.

Conclusions The application of the remission and LDA matrix tool in our sample showed contradictory results with regard to the predictive ability of remission or LDA rate at 6 months. On the other hand, our results reinforce that EULAR response may be a more appropriate goal of treatment than LDA or remission in a subgroup of patients.

1. Vastesaeger N, et al. EULAR 2015 (SAT0360)

Disclosure of Interest None declared