The Mirage of Stability: Challenges of Predicting Disease Activity Over Time in Systemic Lupus Erythematosus (SLE)
Author(s)
Peter Gal, MSc1, Darren Talbot, PhD2, david Isenberg, MD3, A E Clarke, MD4, joshua a. ray, BS, MSc5.
1Visible Analytics Ltd, Budapest, Hungary, 2Viatris Innovation, Allschwil, Switzerland, 3University College London, London, United Kingdom, 4University of Calgary, Calgary, AB, Canada, 5Viatris Innovation, Basel, Switzerland.
1Visible Analytics Ltd, Budapest, Hungary, 2Viatris Innovation, Allschwil, Switzerland, 3University College London, London, United Kingdom, 4University of Calgary, Calgary, AB, Canada, 5Viatris Innovation, Basel, Switzerland.
OBJECTIVES: Published predictive equations evaluate the natural history of SLE, using projected Annual Mean Systemic Lupus Erythematosus Disease Activity Index (AMS) score which measures disease activity. We aimed to compare them, assess their clinical face validity and explore their suitability for cost-effectiveness models (CEMs).
METHODS: A targeted literature review was conducted for predictive equations based on large databases with long-follow-up in North America. Disease activity trajectories were simulated using each equation for both a hypothetical population and representative patient profiles. Resulting AMS trajectories were reviewed by clinical and health economics experts during an advisory board to assess alignment with clinical plausibility, clinical expectations and applicability in CEMs.
RESULTS: Four predictive equations (NICE TA397, Watson et al. 2015, Clarke et al. 2021, Touma et al. 2024) were identified. All equations produced similar AMS trajectories: steep decline in years 3-5 and a plateau around year 10, indicating low disease activity. Only minor differences were observed, relating to the speed of decline and the plateau level. Experts noted that while these patterns may be reasonable at the population level, they do not reflect individual variability, the relapsing-remitting nature of SLE, or the role of flares. The inclusion of age as a covariate was also questioned, due to the inclusion of time since diagnosis and limited long term follow-up data (10 yrs). It is considered clinically plausible that disease activity reduces over time.
CONCLUSIONS: Existing predictive equations of AMS offer a consistent, but simplified, view of SLE disease activity progression. Expert feedback highlighted the importance of individual variability, clinical plausibility and their ability to capture key features of SLE impacting disease progression. Further research should focus on developing improved predictive equations designed to better capture aspects of SLE disease activity that affect long-term outcomes, with a specific focus on their inclusion into CEMs.
METHODS: A targeted literature review was conducted for predictive equations based on large databases with long-follow-up in North America. Disease activity trajectories were simulated using each equation for both a hypothetical population and representative patient profiles. Resulting AMS trajectories were reviewed by clinical and health economics experts during an advisory board to assess alignment with clinical plausibility, clinical expectations and applicability in CEMs.
RESULTS: Four predictive equations (NICE TA397, Watson et al. 2015, Clarke et al. 2021, Touma et al. 2024) were identified. All equations produced similar AMS trajectories: steep decline in years 3-5 and a plateau around year 10, indicating low disease activity. Only minor differences were observed, relating to the speed of decline and the plateau level. Experts noted that while these patterns may be reasonable at the population level, they do not reflect individual variability, the relapsing-remitting nature of SLE, or the role of flares. The inclusion of age as a covariate was also questioned, due to the inclusion of time since diagnosis and limited long term follow-up data (10 yrs). It is considered clinically plausible that disease activity reduces over time.
CONCLUSIONS: Existing predictive equations of AMS offer a consistent, but simplified, view of SLE disease activity progression. Expert feedback highlighted the importance of individual variability, clinical plausibility and their ability to capture key features of SLE impacting disease progression. Further research should focus on developing improved predictive equations designed to better capture aspects of SLE disease activity that affect long-term outcomes, with a specific focus on their inclusion into CEMs.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
CO245
Topic
Clinical Outcomes, Economic Evaluation, Study Approaches
Topic Subcategory
Clinical Outcomes Assessment
Disease
Systemic Disorders/Conditions (Anesthesia, Auto-Immune Disorders (n.e.c.), Hematological Disorders (non-oncologic), Pain)