Background: Liver dysfunction, particularly Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD), is a recognized extra-articular manifestation of both Psoriatic Arthritis (PsA) and Rheumatoid Arthritis (RA), contributing significantly to the burden of comorbidities in these patients. Although the underlying mechanisms are not fully elucidated, emerging evidence points to a complex interplay between chronic inflammation, metabolic dysfunction, treatments, and immune-mediated pathways. Despite its high prevalence, the early identification of MASLD remains challenging due to the suboptimal sensitivity and specificity of currently available non-invasive diagnostic indices, particularly in patients with systemic inflammatory diseases. Therefore, there is an urgent need to identify novel diagnostic biomarkers that can better capture liver dysfunction in this specific population, facilitating earlier intervention and personalized management.

Objectives: 1) to evaluate the presence and severity of hepatic steatosis and fibrosis in a cohort of patients with rheumatoid arthritis (RA) and psoriatic arthritis (PsA), without obesity, using FibroScan and the Attenuated Coefficient Parameter (CAP); 2) to identify clinical factors and inflammatory factors associated with the presence of MASLD in these diseases and 3) to develop a machine learning-based predictive model integrating clinical and proteomic data to accurately differentiate patients with hepatic steatosis.

Methods: A descriptive, observational, cross-sectional study was conducted in a cohort of 99 patients (49 with RA and 50 with PsA), without obesity, matched by age and sex, alongside a control group of 26 healthy donors. Clinical parameters, liver disease indices, and inflammatory biomarkers were evaluated. Liver stiffness and hepatic fat infiltration were assessed using FibroScan and Controlled Attenuation Parameter (CAP). Additionally, serum levels of 276 proteins related to inflammation, organ damage, and cardiovascular disease were quantified using Olink Proximity Extension Assay technology. A machine learning approach was applied to identify diagnostic biomarkers of hepatic steatosis. Feature selection was performed using the Boruta algorithm. To ensure the generalizability and reliability of the predictive model, Leave-One-Out Cross-Validation (LOOCV) was employed for both training and validation.

Results: RA and PsA patients showed low disease activity scores (DAS28=2.83±1.43 and DAPSA=11.80±7.36, respectively), with a disease duration of 4.63±4.80 years for RA and 9.58±6.64 years for PsA. Hepatic steatosis was detected in 45.8% of PsA patients and 62.8% of RA patients, while hepatic fibrosis was present in 12% of PsA patients and 4.7% of RA patients. In the overall cohort of patients with inflammatory arthritis (IA), hepatic steatosis was significantly associated with metabolic alterations, including elevated levels of glucose, triglycerides, and apolipoprotein B, along with increased abdominal circumference. Additionally, significantly higher ALT levels and lower alkaline phosphatase levels were observed in IA-MASLD patients. Among the liver risk indices analyzed, Fatty Liver Index (FLI) demonstrated the highest accuracy in identifying patients with steatosis (AUC=0.78; p<0.0001). Proteomic analysis revealed significant alterations in the levels of five proteins (Tissue Plasminogen Activator (T-PA), CUB Domain-Containing Protein 1 (CDCP-1), Growth Differentiation Factor 15 (GDF-15), Fibroblast Growth Factor 21 (FGF-21), and Insulin-like Growth Factor Binding Protein 7 (IGFBP-7)) in patients with steatosis compared to those without. Notably, T-PA and FGF-21 levels showed significant correlations with clinical and metabolic parameters. A predictive model integrating three protein ratios (based on six proteins) and a clinical characteristic was developed, achieving high accuracy in differentiating patients with and without steatosis (AUC=0.94), with a sensitivity of 93% and a specificity of 78%.

Conclusion: 1) Patients with RA and PsA, even in the absence of obesity, exhibit a high prevalence of hepatic steatosis, which is strongly associated with metabolic alterations, including elevated levels of CRP, triglycerides, and insulin; 2) the presence of hepatic steatosis in these patients is characterized by a distinct serum protein signature, suggesting that specific proteins may serve as potential biomarkers for liver dysfunction in the context of inflammatory arthritis and 3) Our findings support the integration of proteomic data with clinical variables as a novel and effective strategy for the early identification and risk stratification of patients with hepatic steatosis, particularly in populations with chronic inflammatory diseases.

REFERENCES: NIL.

Acknowledgements: Projects “PI20/00079” funded by Instituto de Salud Carlos III (ISCIII) and co-funded by the European Union. Project “PID2023-152503OB” funded by the Minister of Science, Innovation and Universities co-financed by the European Union.

Disclosure of Interests: None declared.

© The Authors 2025. This abstract is an open access article published in Annals of Rheumatic Diseases under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ). Neither EULAR nor the publisher make any representation as to the accuracy of the content. The authors are solely responsible for the content in their abstract including accuracy of the facts, statements, results, conclusion, citing resources etc.