Pos1122 Two-Variable Lupus Arthritis and Musculoskeletal Disease Activity Score (Lamda2): A Novel Tool to Measure Arthritis Disease Activity in People With Systemic Lupus Erythematosus

Background: The majority of people with systemic lupus erythematosus (SLE) recruited to clinical trials have joint and/or skin involvement. There is an unmet need for a dedicated arthritis disease activity measure. Existing global measures such as SLEDAI-2K and BILAG do not capture arthritis in sufficient detail to be sensitive to change. We previously proposed a preliminary index of musculoskeletal disease activity: the LAMDA. Objectives: To derive a novel composite arthritis disease activity measure to capture musculoskeletal (MSK) disease activity in people with SLE from a pool of existing arthritis-specific variables, including 28-joint counts. Methods: SLE patients with physician-diagnosed inflammatory joint pain requiring treatment (intramuscular methylprednisolone 120mg) were recruited to the USEFUL observational study1. Clinical assessments, patient-reported outcomes, bilateral hand reference range 0.5-2.0). Wilcoxon signed rank tests at 2-sided alpha=0.05, Holm-corrected, were used to compare variables pre- and post-treatment in USEFUL. Results: Characteristics of the dataset have been described before1. The earlier version of LAMDA included 66 swollen joint count, participant MSK pain VAS, physician MSK disease activity VAS and erythrocyte sedimentation rate. Using 28 joint counts and replacing ESR with IgG in the candidate variable list, only SJC28 and physician MSK DA VAS were retained in the final model. In the baseline data, LAMDA2 was correlated with total GSPD at rho=0.60 (n=133); correlations with other variables were substantive, except participant MSK DA VAS (rho=0.12). In those with longitudinal data (n=121), all arthritis-specific measures decreased at week 6; effect size r (Wilcoxon z-score/√N) for LAMDA2 (0.37) was greater than SLEDAI-K or BILAG (both 0.32), and similar to total GSPD (0.38). Amongst those in whom SLEDAI-2K arthritis score remained unchanged between baseline and week 6 (76/121; 63%), LAMDA2, TJC28, EMS VAS, physician VAS and total GSPD improved on average, with medium effect sizes (Table 1; Figure 1). In this group with no change in arthritis according to SLEDAI-2K, some participants' SJC28 improved by 18 joints whilst others worsened by 5 joints, although the majority did not change (Table 1; Figure 1). Effect sizes were small for SJC28 and participant MSK DA VAS, and in this smaller subset the data were consistent with the null hypothesis of no change in these two variables. Amongst those in whom BILAG MSK remained unchanged between baseline and week 6 (59/121; 49%), the results were similar. Conclusion: Existing SLE measures are limited in their scope to show changes in musculoskeletal disease activity; around half of the participants treated with steroid in the USEFUL study showed no change at week 6 in the SLEDAI arthritis and BILAG MSK scores. However, a novel two-variable disease activity measure, LAMDA2, derived against ultrasound-detected inflammation in the joints of the hands and wrists at baseline, was able to detect improvement in these apparently static groups. LAMDA2 requires further validation in larger, multicentre studies and double-blind comparative trials but may improve the ability of SLE trials to detect differences in treatments. REFERENCES: 1 DOI: 10.1093/rheumatology/keab288. Acknowledgements: Funding from LupusUK Disclosure of Interests: Edward M. Vital AstraZeneca, Novartis, Ostuka, AstraZeneca, UCB, AstraZeneca, Novartis, UCB, Otsuka, Abbvie, Merck, Lilly, Pfizer, AstraZeneca, Sandoz, Elizabeth Hensor: None declared, Khaled Mahmoud: None declared.