Distinct Biomarkers Enrich For Clinical Response To Tocilizumab (Tcz) And Adalimumab (Ada) In Adacta: A Head-To-Head Monotherapy Study In Patients (Pts) With Methotrexate (Mtx)-Ir Rheumatoid Arthritis (Ra)

Background RA pts treated with TCZ showed superior clinical benefit vs pts treated with ADA in the ADACTA study1; however, it is unclear whether distinct RA subpopulations responded preferentially to anti-IL6R vs anti-TNFα therapy. Objectives To evaluate biomarker-defined pt subpopulations and their response to TCZ and ADA monotherapy. Methods ADACTA was a randomised, double-blind, 24-wk study in pts with RA of ≥6-mo duration and DAS28 >5.1 who were MTX intolerant or for whom continued treatment with MTX was considered ineffective or inappropriate. Pts received TCZ 8 mg/kg IV every 4 wks (+ placebo [PBO] ADA) or ADA 40 mg SC every 2 wks (+ PBO TCZ) for 24 wks as monotherapy. Serum biomarkers previously identified in TCZ or ADA studies—including CXCL13, CXCL10, CCL2, CCL13, IL8, IL6, IL2, IL10, TNFα, IFNγ, GM-CSF, IL12p70, IL1β, sICAM1, sCD25, MMP1, MMP3 and MMP9—were measured using quantitative immunoassay methods. Baseline (BL) levels (wk 0, n=203) were assessed against ACR20/50/70 response and ΔDAS28-ESR at wk 24 using local regression. Biomarkers were determined to enrich for clinical response if both efficacy endpoints showed improvement. Post-dose changes were assessed at wks 4 and 24 for biomarkers demonstrating enrichment for response. Statistical assessment for post-dose changes was performed using the Wilcoxon rank sum test comparing absolute biomarker levels at wk 0 vs 24, without correction for multiple comparisons. Post-dose changes vs efficacy compared wk 24 levels in ACR50 responders [R] vs non-responders [NR], except sICAM1 wk 4 as indicated. Results Analysis of BL biomarker levels vs efficacy at wk 24 revealed that pts with BL levels above the median for MMP3, CXCL13 or CXCL10 had increased ACR50 response rates and decreased DAS28-ESR scores after TCZ treatment vs pts with BL levels below the median; higher BL levels of these biomarkers did not enrich for ADA response (Table). Conversely, BL levels above the median for sICAM1 were associated with enhanced ADA but not TCZ clinical response. TCZ reduced post-dose levels of MMP3 in all comers (wk 0 vs 24, p<0.0001), while levels decreased with ADA, mainly in ACR50 responders (R vs NR, p<0.05). Reduced CXCL13 levels were observed in all comers with either TCZ (wk 0 vs 24, p<0.0001) or ADA (wk 0 vs 24, p<0.0001), but the CXCL13 decrease correlated with ACR50 response only with ADA (R vs NR, p<0.05). CXCL10 also decreased with ACR50 response only with ADA (R vs NR, p<0.05), though BL CXCL10 levels enriched for TCZ response. Circulating sICAM1 levels decreased with efficacy with TCZ (R vs NR, wk 4, p<0.05), while BL levels enriched for ADA clinical response. Image/graph Conclusions Distinct patterns of biomarkers enriching for response suggest that TCZ and ADA may preferentially demonstrate efficacy in different subpopulations of RA pts. Higher BL levels of distinct biomarkers (MMP3, CXCL13 or CXCL10 for TCZ; sICAM1 for ADA) may enrich for clinical response. Furthermore, post-dose data suggest BL biomarker levels may enrich for efficacy independently of modulation by treatment. References Gabay C et al. Lancet. In press Disclosure of Interest A. Herman Employee of: Genentech, a member of the Roche group, D. Musselman Employee of: Genentech, a member of the Roche group, S. Fischer Employee of: Genentech, a member of the Roche group, A. Setiadi Employee of: Genentech, a member of the Roche group, C. Gabay Grant/research support from: Roche, Consultant for: Roche, Abbvie, Pfizer, UCB, BMS, MSD, A. Kavanaugh Grant/research support from: Roche, Abbott, Amgen, UCB, BMS, Pfizer, Janssen, M. Klearman Employee of: Genentech, a member of the Roche group, A. Song Shareholder of: Roche, Employee of: Genentech, a member of the Roche group, M. Townsend Shareholder of: Genentech, a member of the Roche group, Employee of: Genentech, a member of the Roche group