Association between ethnicity and initial response to tnf inhibitors in people with rheumatoid arthritis: results from the british society for rheumatology biologics register for rheumatoid arthritis (bsrbr-ra)

Background Little is known about the association between ethnicity and response to TNF inhibitors (TNFi) in people with rheumatoid arthritis (RA). Objectives This study examines the association between self-reported ethnicity and DAS28 response after 6 months of treatment in patients starting their first TNFi using data from the British Society for Rheumatology Biologics Register for RA. Methods 14133 RA patients with self-reported ethnicity starting their first TNFi were included. Due to a very low proportion of non-white patients, ethnicity was divided into white or non-white for analysis. Outcomes included the change in disease activity using DAS28, the proportion of patients who achieved DAS28 remission and EULAR response at month 6. Adjusted regression models appropriate to outcome were used to compare between the two groups. Multiple imputation was used to account for missing data. Results Of 14133 patients starting TNFi with recorded ethnicity, only 607 (4.3%) recorded themselves as non-white (389 Asian, 134 Black, 57 Mixed ethnicity and 27 other) (Table 1). At start of TNFi, non-white patients were younger (non-white vs. white: mean 51 vs. 57 years; p<0.001), with shorter disease duration (median 8 vs. 10 years; p<0.001), higher proportion of females (86% vs. 76%; p<0.001) and fewer current cigarette smokers (9% vs. 20%; p<0.001). Non-white patients had lesser improvement in DAS28 at month 6 (adjusted regression coefficient (95% confidence interval (95%CI)): 0.3 (0.04-0.5)). However, using the white patients as a reference, non-white patients were not associated with the achievement of DAS28 remission (adjusted odds ratio (95%CI): 0.7 (0.5-1.1)) or EULAR response (aOR (95%CI): 0.8 (0.7-1.0)) at month 6. White Non-white p N 13515 607 Age 57.0±12.0 51.4±12.3 <0.001 Female (% ) 10262 (75.9%) 520 (85.7%) <0.001 Ethnicity (% ) White 13515 (100%) - Black - 134 (22.1%) Asian - 389 (64.1%) Mixed ethnicity - 57 (9.4%) Other - 27 (4.4%) Disease duration (years ) 10 (4-18) 8 (4-15) <0.001 RF positive (% ) 8341 (61.7%) 390 (67.5%) 0.08 DAS28 6.4±1.1 6.4±1.1 0.4 Tender joints 15 (10-21) 15 (9-21) 0.6 Swollen joints 10 (6-14) 9 (6-13) <0.001 VAS 75 (61-86) 75 (61-88) 0.9 ESR 35 (19-58) 41 (25-65) <0.001 CRP 24 (10-53) 19 (8-41) 0.005 HAQ 1.9±0.6 1.9±0.7 0.4 Biologic treatment (% ) Adalimumab 4107 (30.4%) 165 (27.2%) 0.1 Certolizumab 875 (6.5%) 50 (8.2%) Etanercept 5322 (39.4%) 238 (39.2%) Infliximab 3211 (23.8%) 154 (25.4%) Concomitant csDMARDs (% ) 7595 (56.2%) 340 (56%) 0.9 Methotrexate (%) 5512 (40.8%) 243 (40.0%) 0.7 Concomitant glucocorticoids (% ) 5017 (37.1%) 199 (32.8%) 0.03 Current cigarette smoking (% ) 2698 (20.0%) 57 (9.4%) <0.001 Response to first-line TNFi at month 6 DAS28 at month 6 4.7±1.8 5.0±1.8 0.02 ΔDAS28* 1 [reference] 0.3 [0.04-0.5] 0.02 DAS28 remission (%) 13% 11% 0.1 Odds of DAS remission* 1 [reference] 0.7 [0.5-1.1] 0.1 Good EULAR response (%) 18.2% 16.5% Moderate EULAR response (%) 33.7% 33.2% Odds of achieving higher EULAR response* 1 [reference] 0.8 [0.7-1.0] Data is expressed as mean±SD, median (interquartile range), number (percentage), adjusted regression coefficient [95% confidence interval] or adjusted odds ratio [95% confidence interval]. *Adjusted for baseline age, gender, disease duration, concomitant csDMARDs, concomitant steroid, rheumatoid factor status, cigarette smoking status, DAS28 and HAQ Conclusion Recruitment to the BSRBR-RA of non-white patients was exceptionally low compared to expected population distribution – the reasons for this are not immediately evident. Among those recruited, there were no significant differences observed in initial treatment response although non-white patients were younger, more likely to be female, had shorter disease duration and less likely to be smokers. Further evaluation into both ways to increase representativeness in our national RA treatment register and influence of ethnic differences on longer term outcomes with biologic therapies is warranted. Reference NIL. Acknowledgements: NIL. Disclosure of Interests Man Fung Tsoi: None declared, Rajinder Singh Andev Speakers bureau: Speaker fees from Novartis, Lianne Kearsley-Fleet: None declared, Kath Watson: None declared, Shirish Dubey Speakers bureau: Speaker fees from Janssen, Consultant of: Honoraria from Boehringer Ingelheim, Kanta Kumar: None declared, Arumugam Moorthy Speakers bureau: Novartis and Galapagos, Consultant of: Honoraria from Lilly and UCB, Monica Gupta: None declared, Adewale Adebajo: None declared, Kimme Hyrich Consultant of: Honoraria from Abbvie, Grant/research support from: Pfizer and BMS, and is supported by the NIHR Manchester Biomedical Research Centre.