Sub-Analysis of ELF Score Biomarkers Components Indicates a Specific Correlation with Different Organ Involvement in Systemic Sclerosis.

Background/Purpose: A recent large multicenter study has identified an algorithm, known as Enhanced Liver Fibrosis (ELF), by combining the serum concentrations of amino-terminal propeptide of procollagen type III (PIIINP), tissue inhibitor of matrix metalloproteinase-1 (TIMP-1) and hyaluronic acid (HA) in a weighted average developed to match liver fibrosis pathology scoring The algorithm has been shown to predict liver related outcomes in patients with chronic liver diseases and recently it has been shown to correlate with several measures of fibrosis in SSc including modified Rodnan SkinScore, presence of ILD, DLCO as well as age, disease activity and different aspects of disease severity. The aim of this study was to compare the performance of ELF with its single components in correlating with the different clinical and instrumental variables in SSc, to determine whether any of the three biomarkers could have a specific predictive value as surrogate outcome measure in SSc. Methods: The serum concentrations of the three biomarkers were analysed in 129 SSc patients employing the same platform used to calculate the ELF score (siemens, advia centaur). All patients were investigated for clinical and serological subset, disease duration, skin and internal organ involvement, HAQ-DI, disease severity and activity. Correlations were calculated using Spearman correlation test. Mann-Whitney test was used to perform comparison between groups. Statistical analysis was performed using GraphPrism software. Results: Median, correlation coefficient and statistical significance of ELF and its single analytes are summarised in table 1. All three components of ELF showed a similar strong correlation with mRSS and HAQ-DI, confirming a strong predictive value on skin involvement. Interestingly, the concentration of HA was the only parameter correlating with Age, muscle severity and Heart severity, whereas it did not correlate with DLCO% or lung severity. In this regard the biomarker with better performance on Lung involvement was TIMP-1, which showed a strong correlation with DLCO and lung Severity. Furthermore SSc patients with interstitial lung disease (ILD) showed significant higher levels of TIMP-1 (P0.0136) and TIMP-1 was the only biomarker to correlate with the EScSG-Activity Index. On the contrary, PIIINP was the only one to correlate with Joint and kidney severity. Coefficient correlation (rho) between ELF score and single serum markers with clinical parameters in 129 SSc patients ELF score PIIINP (ng/mL) TIMP-1 (ng/mL) HA (ng/mL) Serum values (median, range) 8.84, 6.49–.10.84 rho 6.25, 2.63–.33.06 rho 215.3, 88.5–.531.2 rho 41.53, 4.69–.236.4 rho Age 0.34*** 0.05 0.11 0.42*** mRSS 0.26** 0.30*** 0.33*** 0.19* DLCO, absolute value 0.26** 0.1 0.28** 0.20* DLCO % 0.06 0.05 0.20* 0.02 Skin_sev 0.34*** 0.34*** 0.37*** 0.20* Join/tendon_sev 0.26** 0.25** 0.13 0.11 Muscle_sev 0.34*** 0.17 0.08 0.26** GI_sev 0.17* 0.15 0.03 0.09 Lung_sev 0.01 0.01 0.18* 0.11 Heart_sev 0.16 0.08 0.09 0.21* Kidney_sev 0.16 0.23** 0.001 0.05 EScSG-AI 0.15 0.09 0.20* 0.08 HAQ-DI 0.32*** 0.25** 0.31*** 0.22* *P0.05; **P0.01; ***P0.001 Conclusion: Subanalysis of the single serum markers included in the ELF score algorithm suggests that the different biomarkers may function as surrogate outcome measure of specific organ involvement in SSc. In this regard, longitudinal studies confirming the predictive value and sensitivity to change over time of the single biomarkers may pave the way to develop specific algorithms tailored to carry the maximum predictive value on specific organ involvement in SSc.