Background/Purpose: Considering the increased cardiovascular risk in RA patients and the influence of biologics, especially tocilizumab (TCZ) on the lipid profile, we aimed to investigate the differential effect of biologics on cardiovascular risk-associated biomarkers.

Methods: This ancillary study used serum samples at inclusion (M0) and 6 months (M6) of active RA patients inadequate responders to a 1^st iTNF participating to the randomized ROC (Rotation or Change of Biotherapy, NCT01000441) trial, which compared EULAR response at M6 between a 2^nd iTNF and another biologic (IV TCZ, abatacept or rituximab). At baseline and M6, we assessed lipoprotein-related apoproteins (ApoA₁ and ApoB₁₀₀ reflecting mainly high density lipoprotein (HDL) and low density lipoprotein (LDL), respectively) and cardio-metabolic markers: lipoprotein(a) (Lp(a)) and adipokines (leptin and adiponectin) for the calculation of leptin/adiponectin ratio (LAR), that estimates insulin resistance. We compared the M0-M6 change of each marker between the 2^nd iTNF group and the other biologic group and between the 2^nd iTNF group and TCZ. We also compared this M0-M6 change according to the EULAR response in each group of treatment.

Results: From the 300 patients included in the ROC trial, 203 were tested due to the availability of the serum samples: 96 in the 2^nd iTNF group and 107 in the other biologic group (47 TCZ, 26 abatacept and 34 rituximab). None of the measured markers deteriorated between M0-M6 in the two groups. Conversely, ApoA₁ level significantly increased in the 2^nd iTNF group (+0.008 ±0.22g/L mean ±standard deviation, p<0.001) as well as in the other biologic group (+0.06 ±0.29 g/L, p<0.001). Lp(a) level only decreased in the other biologic group (-0.02 ±0.09 g/L, p=0,02), mainly due to TCZ (-0,06 ±0.08 g/L, p<0,001). TCZ also improved ApoA1 (+0.11 ±0.32 g/L, p <0.001) and adiponectin (+ 0,48 ±1.38 mg/L, p=0.02) levels. Compared to the 2^nd iTNF group, Lp(a) and LAR decreased more in the other biologic group and Lp(a) and ApoA1 levels were improved in the TCZ group (Table 1). The improvement in the biomarkers level was mainly due to the EULAR responders in each group of treatment. Thus, Apo1 level increased with 2^nd iTNF (+ 0.14 g/L difference between responders and non-responders, p=0,001), with other biologic (difference +0.1 g/L, p = 0.04) and with TCZ (difference +0.3 g/L, p=0.03).

Conclusion: In active RA patients in inadequate response to a 1^st iTNF, cardiovascular risk-associated biomarkers did not deteriorate whatever the 2^nd line of treatment and even improved in responders. Lp(a) and ApoA1 levels, both involved in atherogenesis, improved with a 2^nd iTNF as well as with other biologics, but more with TCZ. This improvement seems driven by EULAR responders arguing for the control of RA as a major factor for cardiovascular risk improvement.