P67 Prevalence of Metabolic Dysfunction-Associated Steatotic Liver Disease and the diagnostic accuracy of non-invasive biomarkers for liver fibrosis in patients with psoriasis

Abstract Psoriasis has been associated with an increased risk of liver fibrosis. Recent studies indicate that metabolic dysfunction-associated steatotic liver disease (MASLD) might be the underlying cause. Furthermore, no clear consensus exists on how to monitor patients with psoriasis for liver fibrosis. Therefore, we assessed the prevalence of MASLD and examined the diagnostic accuracy of non-invasive biomarkers for liver fibrosis. We included patients with psoriasis from a prospective observational cohort study at a university hospital. The prevalence of MASLD was assessed using Fibroscan with hepatic steatosis defined as a controlled attenuation parameter of >250 dB/m and significant/advanced fibrosis as a median liver stiffness measurement (LSM) ≥8 kPa. Non-invasive biomarkers for fibrosis included, fibrosis-4 index (FIB-4, increased if ≥1.3), N-terminal propeptide of collagen type III (PIIINP, increased if >11.0 µg/L for ≤40 years and >9.5 µg/L for >40 years), and the enhanced liver fibrosis (ELF) test (increased if ≥9.8). Analyses were parametric or non-parametric depending on the distribution of data. We enrolled 403 patients (median age 53 years, 54% men). The mean body mass index was 27.8 (standard deviation ± 6.0). In the self-reported questionnaire, 5% of the patients reported consuming >16 units of alcohol per week, 31% reported having hypertension, 28% dyslipidaemia, and 6% Type 2 diabetes. Most patients (75%) received systemic treatment (conventional or biologics), while 25% only used topicals. The patients were generally well-treated with a median psoriasis area and severity index (PASI) of 2.6 (range, 0–41). Based on the FibroScan, 40% had hepatic steatosis (mean CAP 245, ±55 dB/m). The proportion with possible fibrosis was 5% based on FibroScan (median LSM, 4.9 kPa; range, 1.8–20.6), 24% based on FIB-4 (median, 0.89; range, 0.19–8.11), 20% based on ELF (mean, 9.0, ±0.98), and 14% based on PIIINP (median, 6.9 µg/L; range, 2.3–23.1). Pearson’s correlation coefficient between liver fibrosis and the non-invasive biomarkers was highest for FIB-4 (r = 0.57), with corresponding values of r = 0.40 for ELF and r = 0.30 for PIIINP. Furthermore, with Fibroscan as the gold standard, FIB-4 and ELF had better diagnostic accuracy than PIIINP for diagnosing liver fibrosis with an area under the receiver operating characteristic curve of 0.92 (95% confidence interval, 0.87–0.97) for FIB-4, 0.90 (0.83–0.97) for ELF, and 0.77 (0.66–0.87) for PIIINP. In conclusion, 40% of the patients had liver steatosis and 5% displayed liver fibrosis as determined by Fibroscan. FIB-4 and ELF tests demonstrated superior diagnostic accuracy compared with PIIINP.