Background Tocilizumab (TCZ), a recombinant humanised antibody targeting soluble and membrane IL-6 receptor, is commonly used in rheumatoid arthritis (RA) patients refractory to tumour necrosis factor inhibitors, demonstrating 60–80% effectiveness. Clinical parameters such as years of disease prior to TCZ treatment, naïve for biological therapy (BT naïve), baseline Disease Activity Score 28 (DAS28) and Health Assessment Questionnaire (HAQ), have been associated with response to TCZ, although with conflicting results. Identification of clinical predictors of response may lead to a better selection of BT alternatives in DMARDs refractory RA patients.

Purpose To assess the effectiveness of TCZ in RA patients and the influence of clinical parameters.

Material and methods This was a retrospective cohort study. Linear or logistic regression models were applied to evaluate the influence of clinical parameters (baseline DAS28, baseline HAQ, BT naïve, years of disease prior to TCZ treatment, age at TCZ start, concomitant DMARDs and corticosteroids, baseline CRP and ESR) on TCZ effectiveness, measured according to relative percentage of variation in DAS28, and EULAR response (responders vs non-responders), after 18 months of therapy in RA patients.

Results 61 patients (83.6% women; 53.4±12.6 years) were investigated, with mean disease duration of 10 (7–18) years and 8 (3–13.5) years of disease evolution before TCZ therapy. Only 22 patients were naïve for BT (22/61; 36.1%). Baseline DAS28 and HAQ were 5.6±1.15 and 1.66±0.66, respectively. EULAR response was 88.5% (54/61), and relative percentage of DAS28 variation was −58.9% (−68.8, −44.7) at 18 months. The decrease in relative percentage of DAS28 variation (R2=0.229) was higher in patients with higher baseline DAS28 (coef: −7.98; 95% CI −13.1, −2.8; p=0.03), lower baseline HAQ (coef: 14.9; 95% CI 6.02, 23.8; p=0.01) and BT naïve (coef= −11.5, 95% CI −22.3, −8.0; p=0.036). EULAR response was more frequent in patients with higher baseline DAS28 (OR 3; 95% CI 1.2, 8.3; p=0.048), shorter time of disease prior to TCZ treatment (OR 0.8; 95% CI 1.02–1.5, p=0.026) and lower number of BT failures (OR 0.31; 95% CI 0.11–0.99; p=0.016).

Conclusion TCZ effectiveness in RA patients after 18 months of therapy was >88% (EULAR), with an approximated 60% of relative reduction in DAS28. High baseline DAS28, low baseline HAQ, BT naïve patients and shorter time of disease prior to TCZ treatment have been identified as predictors of better response to TCZ therapy. Hence TCZ should become the first option of BT in refractory DMARDs RA patients.

No conflict of interest