Background: Systemic Lupus Erythematosus (SLE) is clinically and immunologically heterogeneous with a variable response to treatment. MASTERPLANS is an MRC-funded consortium that seeks to identify immunophenotypic subgroups of patients that predict response to therapy. Autoantibody profiles can differentiate subgroups of patients and have potential to predict response to treatment. Objectives: To determine whether known and novel autoantibodies are associated with response to rituximab (RTX), and analyse the association between these antibodies and disease involvement in various organ systems. Methods: Serum was obtained from 224 SLE patients in the BILAG Biologics Registry who received rituximab according to NHS England criteria (2). Patients were recruited if they were starting a first cycle of rituximab for active SLE (BILAG A or 2xBILAG B) despite previous cyclophosphamide or mycophenolate mofetil. Evidence of any single organ system involvement previous or current was taken as having a BILAG score of A-D but not E. Disease activity was measured using BILAG-2004. Clinical response was defined as improvement by >=1 grade in active BILAG-2004 systems with no worsening in other systems. Autoantibodies were measured by immunoprecipitation of proteins by sera from 35 S-labelled K562 cell lines, followed by SDS-PAGE separation and autoradiography. Autoantibodies not able to be detected by this technique (anti-Ro52, anti-dsDNA and aCL) were measured by ELISA. Autoantibody data was analysed in IBM SPSS and GraphPad Prism v8.2. Association between autoantibodies and RTX response was analysed using binary logistic interaction terms and Pearson’s Chi-Square test. Results: Of the 224 patients (201 female, 23 male, median age 40 years) the most common system involvement from the 9 BILAG domains was musculoskeletal (164 patients) and the least ophthalmic (11 patients). Patients with anti-Ro52 and anti-U1RNP/Sm had more frequent involvement of mucocutaneous ( p <0.036, p <0.012) and musculoskeletal domains ( p <0.015 for U1RNP) respectively. There were 136 patients with sufficient data to define as either responders (n=67) or non-responders (n=69) to RTX at 6 months. RTX responders had a higher frequency of anti-U1RNP/Sm compared to non-responders (Figure 1). Further Pearson’s Chi-Square analysis showed a significant association between presence of anti-U1RNP/Sm and better response to RTX ( p <0.018). Conclusion: Our findings suggest that the presence of U1RNP/Sm autoantibodies in a cohort of patients who have received treatment with RTX is associated with more frequent musculoskeletal and mucocutaneous involvement and predicts a more favourable response to treatment. Acknowledgments : Funded by a grant from the Medical Research Council, grant number MR/M01665X/1. BILAG BR has been funded by unrestricted educational donations from Roche, GSK and LUPUS UK. Part-funded by a grant from LUPUS UK. Disclosure of Interests: : Danyang Li: None declared, Hui Lu: None declared, Juliet Dunphy: None declared, Theresa Smith: None declared, Edward Vital Grant/research support from: AstraZeneca, Roche/Genentech, and Sandoz, Consultant of: AstraZeneca, GSK, Roche/Genentech, and Sandoz, Speakers bureau: Becton Dickinson and GSK, Ian N. Bruce Grant/research support from: Genzyme Sanofi, GSK, and UCB, Consultant of: Eli Lilly, AstraZeneca, UCB, Iltoo, and Merck Serono, Speakers bureau: UCB, Neil McHugh: None declared
Background: Biomarkers to predict response to rituximab include plasmablasts and, in the current MASTERPLANS consortium, Sm/U1RNP antibodies and high expression of IFN Score B (a subset of interferon-stimulated genes that predict more clinical outcomes than a classic interferon signature). The relationships amongst these biomarkers and their association with response to conventional therapies are less well described. Objectives: To analyse the inter-relationships amongst immune biomarkers in two independent SLE cohorts in association with disease activity and stage of therapeutic pathway. Methods: CONVAS is a cohort of unselected SLE patients; data available include current and historic disease activity, use of biologic therapy, flow cytometry, gene expression (IFN Score A and IFN Score B), and immunoprecipitation for autoantibodies (n=91). BILAG-BR is a British registry study for SLE patients commencing biologics; data available include current and historic disease activity, gene expression (IFN Score A and IFN Score B) and immunoprecipitation for autoantibodies (n=112). In both cohorts, biologics were only prescribed to patients with active disease (BILAG 1 x A or 2 x B) and failure of either cyclophosphamide or mycophenolate. Given the mixture of continuous and categorical variables, data were clustered using Gower distance and Partitioning Around Medioids. K was chosen using silhouette coefficient and clusters visualised with t-Distributed Stochastic Neighbor Embedding (t-SNE). Results: There were 6 clusters. In rituximab-naïve patients: Sm/U1RNP+, Ro60-, highest IFN Score A, low CD4 + T cells, low NK cells, high plasmablasts Sm/U1RNP-, Ro60+, medium IFN Score A, low CD4 + T cells, high NK cells, high plasmablasts Sm/U1RNP-, Ro60-, lowest IFN Score A, high CD4 + T cells, low NK cells, low plasmablasts Other antibody subtypes and flow cytometric markers did not improve the accuracy of clustering. In rituximab-treated patients, 3 equivalent clusters for antibody subtypes and IFN Score A were observed but differentiated due to flow cytometry findings, as expected after rituximab treatment. Overall, the patients in the cluster defined by Sm/U1RNP antibodies and high IFN Score A were notable for a higher rate of prior disease activity in the renal, neurological and general BILAG domains (Table 1). Table 1 : Clinical features in unselected SLE patients (CONVAS) System affected (ever ) Sm/U1RNP & high IFN Score A (n=27 ) Other (n=92 ) p value General 14/27 (52%) 24/92 (26%) 0.02 Mucocutaneous 23/27 (85%) 73/92 (79%) 0.50 Neuro 10/27 (37%) 17/92 (19%) 0.04 MSK 25/27 (93%) 83/92 (90%) 0.71 Cardiorespiratory 9/27 (33%) 20/92 (22%) 0.22 Renal 12/27 (44%) 15/92 (16%) 0.005 Haematology 25/27 (93%) 67/92 (73%) 0.03 Analysis of autoantibody status and interferon scores only in BILAG-BR confirmed similar clustering. Across both cohorts, the prevalence of the Sm/U1RNP and high IFN Score A cluster was associated with inadequate response to conventional immunosuppressive treatment (Table 2). Table 2 : Prevalence according to stage of therapy Treatment group Sm/U1RNP & high IFN Score A Other p value Antimalarial or conventional IS-treated (CONVAS) (n=90) 16/90 (17.8%) 74/90 (82%) 0.02 Conventional IS inadequate response, Previous rituximab (CONVAS) (n=38) 14/38 (36.8%) 24/38 (63.2%) Conventional IS inadequate response, starting rituximab (BILAG-BR) (n=163) 51/163 (31.2%) 112/163 (68.7%) N/A Conclusion: A cluster of 23% of unselected SLE patients had more severe immune abnormalities, more severe clinical disease activity and were less likely to be maintained on conventional therapies, with twice as many requiring biologic therapy. Other data in MASTERPLANS have demonstrated that Sm/U1RNP antibodies and IFN Scores predict better response to rituximab. This subgroup of patients may therefore be more appropriate for first-line biologic therapy. Disclosure of Interests: Antonios Psarras: None declared, Danyang Li: None declared, Adewonuola Alase: None declared, Zoe Wigston: None declared, Ian Bruce Grant/research support from: Genzyme, Sanofi, GSK, UCB, Consultant of: Eli Lilly, AstraZeneca, Iltoo, Merck Serono, Neil McHugh: None declared, Edward Vital Grant/research support from: AstraZeneca, Roche/Genentech, and Sandoz, Consultant of: AstraZeneca, GSK, Roche/Genentech, and Sandoz, Speakers bureau: Becton Dickinson and GSK
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)