Background Rheumatoid factor (RF) and anti-citrulinated protein antibodies (ACPA) are used as diagnostic tools, but may also be used as prognostic factors or as predictors of response to therapy, as these biomarkers have been associated with better clinical responses to some bDMARDs.
Objectives To examine whether seropositivity has a similar impact on drug discontinuation of different bDMARDs with a non-anti-TNF mode of action (non-aTNF bDMARDs).
Methods This is a pooled analysis of 10 observational European RA registries (FR, CZ, DK, NO, PT, RO, ES, SE, CH, NL). Inclusion criteria were a diagnosis of RA, initiation of treatment with abatacept (ABA), rituximab (RTX) or tocilizumab (TCZ) and available information on RF and/or ACPA status. The exposure of interest was seropositivity, which was defined as positive if RF or ACPA was positive and negative if both were negative. The primary endpoint was overall drug discontinuation, defined as the period between treatment initiation and treatment discontinuation. Because national differences may constitute a potential confounder, we only included national registries with information for all 3 bDMARD and pooled data across registries only after excluding significant effect modification by country. Drug discontinuation was analyzed using a Cox proportional hazard model, including drug, seropositivity, and their interaction, adjusting for age, gender, disease duration, baseline DAS28, concomitant synthetic DMARD (sDMARDs), number of previous sDMARDs and bDMARDs, and stratifying by country and calendar year.
Results We found no effect modification by country, allowing us to pool data from 12040 patients (Table). In crude analyses, seropositivity was associated with a lower drug discontinuation with all 3 bDMARD (p-value interaction 0.22), with a hazard ratio for seropositive vs. seronegative (HR) 0.89 (95% CI: 0.82–0.97). In adjusted analyses, seropositivity remained associated with a lower drug discontinuation, but the effect differed by drug (p interaction 0.01): ABA: HR for seropositive vs. seronegative: 0.76 (95% CI 0.66–0.88), RTX: 0.88, (95% CI: 0.70–1.10), and TCZ: 1.08, (95% CI: 0.89–1.31). Two-by-two drug to drug comparisons showed that the effect of seropositivity differed between ABA and TCZ (p=0.004), but not between ABA and RTX (p=0.29), or between TCZ and RTX (p=0.16). Other factors associated with discontinuation were higher baseline disease activity and more previous bDMARD.
Conclusions Data from this pooled european registry analysis suggests that seropositivity is associated with lower drug discontinuation of non-aTNF bDMARDs. This effect differed between drugs and was significant for ABA, but not for TCZ or RTX. The impact of seropositivity on other measures of effectiveness still needs to be investigated.
Disclosure of Interest A. Finckh Grant/research support from: BMS, Speakers bureau: Abbvie, Roche, Pfizer, UCB, D. Courvoisier: None declared, J.-E. Gottenberg: None declared, X. Mariette: None declared, J. Morel: None declared, S. A. Bergstra: None declared, V. Hernandez: None declared, C. Codreanu Grant/research support from: Abbvie, Pfizer, Roche, Consultant for: MSD, Pfizer, Roche, Paid instructor for: Abbvie, Bristol-Myers Squibb, Janssen, MSD, Pfizer, Roche and UCB, T. K. Kvien: None declared, M. J. Santos: None declared, K. Pavelka: None declared, M. Hetland Grant/research support from: AbbVie, BMS, MSD, Biogen, UCB, Pfizer, Eli Lilly, Roche, Speakers bureau: Orion, K. Chatzidionysiou Consultant for: AbbVie, Pfizer, Eli Lilly, UCB, Roche, J. Askling Grant/research support from: Abbvie, UCB, Lilly, Janssen, Samsung, Pfizer, MSD, Roche, C. Turesson Grant/research support from: Abbvie, Pfizer, Roche, Consultant for: MSD, Pfizer, Roche, Paid instructor for: Abbvie, Bristol-Myers Squibb, Janssen, MSD, Pfizer, Roche and UCB, C. Gabay: None declared, R. Van Vollenhoven: None declared