Background Tocilizumab (TCZ) is a well-established biologic therapy in the treatment of rheumatoid arthritis (RA). There is limited data on imaging and synovial tissue histology changes.

Objectives To evaluate level of response as defined by power Doppler (PD) ultrasound (US) and synovial tissue histology changes.

Methods Patients with RA, inefficacy to minimum one DMARD +/− TNFi, with DAS28≥3.2 and knee synovitis amenable to synovial biopsy were recruited to this open-label 48-week study. Patients randomised to TCZ/methotrexate (MTX) for 48 weeks or placebo (PBO)/MTX for the first 16 weeks followed by TCZ/MTX until week 48. Clinical and US hand, wrist and knee assessments with US-guided knee synovial biopsy at baseline (BL), weeks 12 and 48 (biopsy optional). US was scored accrording to OMERACT 0–3 grey scale and PD synovitis scoring system. Synovial tissue was assessed for synovial inflammatory infiltrate, stromal cell density, synovial lining on 0–3 scale and overall synovitis (0–9 scale) determined.

Results 15 patients recruited: 12 (80%) females; 9 received TCZ/MTX, 6 PBO/MTX. 2 patients withdrew at each arm, one due to TCZ infusion reaction. Week 16: 33% (3/9) TCZ/MTX achieved DAS28ESR-rem vs 0 PBO+MTX, latter remaining in moderate/high disease activity. Week 48: 92% (12/13) whole group in DAS28ESR-rem. US response: 38% (3/8) TCZ/MTX group who had BL PD synovitis in hand/wrist had absence of PD synovitis at week 16 vs none PBO/MTX group. All patients with abnormal BL PD in knees (median (IQR) PD score of 2 (0–9)) had improved week 48 PD score (median (IQR) PD score of 0 (0–1)). 13 patients (8 TCZ/MTX, 5 PBO/MTX) had synovial biopsies obtained weeks 0 & 12; 21/26 samples (80%) samples useable. No difference between pre- and week 12 synovitis score in both groups. Median (IQR) total synovitis score at BL and after week 12 respectively of 3 (2.75, 4.25) and 3 (2.5–4) in the TCZ/MTX group vs 6 (4,7) and 6 (4,6) in PBO/MTX group. BL total synovitis score did not predict early or late response.

Conclusions TCZ/MTX was associated with significant clinical and imaging improvement compared to MTX alone. An absence of change in synovial infiltrate with TCZ/MTX at 12 weeks suggests a different mechanism for response compared to other anti-cytokine therapies such as the TNFi. Further histochemistry analysis and investigation may potentially determine mechanism and indicators of response on a tissue level.

Disclosure of Interest None declared