AB1142 DERMAL CELLULAR SENESCENCE AND ENDOTHELIAL-TO-MESENCHYMAL TRANSITION IN PATIENTS WITH SYSTEMIC SCLEROSIS TREATED WITH CYCLOPHOSPHAMIDE OR AUTOLOGOUS HAEMATOPOIETIC STEM CELL TRANSPLANTATION

Background:

Systemic sclerosis (SSc) is a systemic autoimmune disease characterised by an interplay of inflammation, vasculopathy and fibrosis. In our previous cross-sectional study, a correlation between fibroblast senescence, endothelial to mesenchymal transition (EndMT), inflammation and fibrosis in skin biopsies of SSc patients was demonstrated¹. However, how these processes respond to different treatments is still unclear.

Objectives:

To evaluate the response of dermal cellular senescence and EndMT in SSc patients undergoing cyclophosphamide (CYC) and autologous haematopoietic stem cell transplantation (HSCT).

Methods:

We analysed the forearm skin biopsies from the ASTIS trial (a randomised control trial in diffused cutaneous SSc patients undergoing HSCT or CYC)². Sections of skin biopsies at baseline and 6 months were stained with haematoxylin and eosin stain (HE), Masson's Trichrome stain (MTC), P16, P21, connective tissue growth factor (CTGF), urokinase plasminogen activator receptor (uPAR), double staining of α-Smooth muscle actin (α-SMA)-CD31 and α-SMA-ERG. Dermal fibrosis (scored from 1 to 3) and inflammation (scored absent, weak and strong; 0 to 2) were semiquantified in HE and MTC. P16, P21, uPAR and CTGF stained sections were semiquantified in fibroblasts and endothelium (scored 0 to 2) and digitally analysed with colour deconvolution and pixel classification. Cellular senescence was observed with the abundance of P21 and P16, excluding Ki-67 on cellular nuclei. The percentage of EndMT vessels was evaluated in the co-localisation of CD31 and α-SMA in immunofluorescent double-stained (α-SMA-CD31 positive) sections and by an enclosure of immunohistochemical ERG-positive endothelial cell nuclei by α-SMA stained cytoplasm (α-SMA-ERG positive). Vascularity was measured in the amount of vessel sections in the dermis (/mm2). The difference in continuous variables between groups was determined using the Wilcoxon rank-sum test and the Wilcoxon signed-rank test, as appropriate. Correlations between ordinal and/or continuous variables were examined using Spearman's correlation.

Results:

Fourteen pairs of skin biopsies were analysed, of which eight pairs were from patients undergoing CYC and six from HSCT. The mean age was 46±10 years. The mean change in modified Rodnan skin score (mRSS) was -5±6 in patients undergoing CYC and -13±4 in HSCT (p 0.028). There were no significant changes in dermal fibrotic and inflammatory scores in both CYC and HSCT groups. With a spatial approach in superficial and deep dermis, the mean change in abundance of P21 on vessels was -0.25±0.75 in patients undergoing HSCT and 0.44±0.89 in CYC (p 0.049). The mean change in α-SMA-CD31 positive percentage was -4.65±2.44 in patients undergoing HSCT and -2.02±2.03 in CYC (p 0.043). The percentage pixel of uPAR decreased after treatment (p 0.005), and the decrease was more pronounced in skin non-responders (mRSS decrease < 25%, p 0.022). The correlation between changes in pathological markers is shown in Figure 1. Change in dermal inflammation was associated with change in CTGF, especially on fibroblasts. Although change in mRSS did not correlate with fibrotic score, it correlated with change in EndMT (α-SMA-CD31 positive, r 0.547 p 0.043) and reversely correlated with uPAR on fibroblasts (r -0.645 p 0.012) and vessels (r -0.651 p 0.012). The change in mRSS also correlated with baseline vascularity (r 0.573 p 0.032) and uPAR score on vessels (r 0.571 p 0.033).

Conclusion:

The forearm dermal fibrosis did not change significantly despite improving clinical mRSS at six months; however, changes in pathological pathophysiologies and vascular response can be observed. Vascular responses, including EndMT as well as P21 and uPAR on vessels, differed between treatment arms. The change in dermal fibrosis was associated with the change in EndMT and uPAR on fibroblasts. The abundance of uPAR may reflect the activity of tissue remodelling.

REFERENCES:

[1] YH Chiu, J Spierings, JM van Laar, et al. Clin Exp Rheumatol. 2023;41(8):1612-7. [2] JM van Laar, D Farge, JK Sont, et al. JAMA. 2014;311(24):2490-8.

Acknowledgements:

NIL.

Disclosure of Interests:

Yu-Hsiang Chiu: None declared, Marijke van Dijk: None declared, Roel Goldschmeding: None declared, Jeska K. de Vries-Bouwstra J.K. de Vries-Bouwstra received research grants from Galapagos, Janssen, Roche; and honoraria from Boehringer Ingelheim, Janssen and Abbvie; all payments were made to the institution LUMC., Jacob M. van Laar J. M. van Laar has received honoraria from Abbvie, Arxx Tx, Galapagos, Gesyntha, Leadiant, Roche, and research grants from Astra Zeneca, MSD and Roche., Julia Spierings A research grant from Boehringer.