Pos1154 pre-treatment differentially expressed metagenes characterize systemic lupus patients who subsequently achieve clinical response to belimumab

Background Belimumab, which targets the B Cell activator and survival factor BLyS, is an approved treatment for systemic lupus (SLE) and has demonstrated efficacy in multiple clinical trials around the world. Nevertheless, like other SLE treatments, belimumab can have disappointing results in a significant proportion of this heterogenous population, a problem further complicated by the unpredictable impact of various combination treatments used with belimumab in trials and clinical practice. Objectives Examination of gene expression variables in SLE patients prior to belimumab treatment with limited background medications in order to distinguish those who subsequently respond from those who do not. Methods A prospective open-label study of belimumab was conducted in 24 SLE patients with active but non-organ threatening disease, who were required to withdraw other immune suppressants (only antimalarials or low dose steroids allowed). At baseline, patients were given optional steroid injections for temporary relief, then followed for six months on intravenous belimumab (10 mg/kg). The primary endpoint was time to treatment failure compared to historical controls from the BOLD study which followed the same protocol but without treatment after initial steroids [1] . Response was also evaluated at six months by the SLE Responder Index (SRI-4) or the BILAG-based Combined Lupus Assessment (BICLA). Before and during treatment, Pax Gene tubes were collected, and RNA sequenced, using the same methods for healthy control samples. To maximally discriminate clinical responders from non-responders by baseline gene expression patterns in a small, heterogenous population, principal component analysis was conducted with modules (metagenes) comprised of differentially expressed genes (DEGs). Results Of 24 patients entering the study, 20 completed 6 months of treatment. The mean survival time without flare, treatment change, or study withdrawal was 18.4 weeks (CI 15.3-21.5) compared to the 41 BOLD participants with mean survival 9.829 weeks (CI 0.999-7.871) (p<0.001 by log rank test). At 6 months, 14 patients who received belimumab (58%) and 1 from the BOLD study (2.4%) remained free of flare. The SRI-4 was met by 7 patients in this study (29.2%), the BICLA by 9 (37.5%). SRI-4 response was used as the basis for modeling two metagenes comprised of the 25 most upregulated and 25 most downregulated genes at baseline in belimumab responders as compared to non-responders. There was no overlap between metagene scores of responders’ vs non-responders (Figure 1). After 3 months of treatment with belimumab all metagene scores corrected towards levels in healthy controls. The most discriminatory DEGs included protein transcription factors and long non-coding RNA (transcription regulators). Conclusion When background immune suppressants are excluded, belimumab still achieves SRI 4 or BICLA response for some patients, and most do not flare for at least six months. A composite metagene model has been identified as a potential predictor of belimumab response. Confirmation studies should be prospectively conducted. Reference [1]Merrill, Arth Rheum 2017 69:1257 Acknowledgements: NIL. Disclosure of Interests Joan T Merrill: None declared, Miles Smith: None declared, Kevin Thomas: None declared, Carla Guthridge: None declared, Nicolas Dominguez: None declared, Susan Macwana: None declared, Wade DeJager: None declared, Stan Kamp: None declared, Cristina Arriens Speakers bureau: AstraZeneca and Aurinia, Consultant of: Advisor or review panel: AstraZeneca, Aurinia, Bristol Myers Squibb, GlaxoSmithKline, and Kezar, Grant/research support from: AstraZeneca and Bristol Myers Squibb, Bridget Parrish: None declared, Judith A. James: None declared.