Using tissue microarray to detect inflammasome signaling components that contribute to the pathogenesis of myelodysplastic syndrome.

Abstract

7044 Background: Myelodysplastic syndromes (MDS) are characterized by aberrant maturation, ineffective hematopoiesis, cytopenia, and progression to acute myeloid leukemia. MDS pathogenesis is multifactorial and potentially linked to constitutive innate immune stimulation converging upon the NLRP3 inflammasome to induce pyroptosis, a caspase-1 dependent cell death. Inflammasome assembly is initiated by both cell-extrinsic stimuli including S100A9, the TLR4 and CD33 ligand, and cell-intrinsic danger signals licensing caspase-1 which activates IL1b and beta-catenin resulting in cell death and cellular proliferation leading to maturation and differentiation blocks. Further, EYA2 has been suggested to be an inflammasome activator, whereas cPLA2 has been suggested to be an inhibitor. The purpose of this study is to determine whether immunohistochemistry (IHC) may be utilized to assess expression of inflammasome components. Methods: An IRB protocol was approved prior to initiating this study. We retrospectively identified 43 low risk MDS patients. A tissue microarray (TMA) was constructed utilizing MDS bone marrow biopsy samples (2-3 representative cores per sample). IL-1, S100A9, EYA2, cPLA2, beta-catenin, and TLR4 expression were assessed by IHC after validation of each antibody. IHC expression was scored independently by two hematopathologists by calculating scores (product of staining intensity x percent expression). IHC expression was compared using Spearman correlation estimate. Demographic and clinical data were collected and correlated with IHC expression using Kruskal-Wallis test, Spearman correlation, and Logrank test. Results: Patients were median 72 years of age, 67% men and included 47% MDS-MLD, 35% MDS-RS, 14% MDS-SLD, 2% MDS del5q and 2% MDS-U. IL-1 expression correlated with beta-catenin expression, r = 0.42, 95% CI 0.115 to 0.658 (p = 0.007). There was a trend towards significance between IL-1 and cPLA2, r = 0.30 (p = 0.067); S100A9 and cPLA2, r = 0.31 (p = 0.052); and S100A9 and EYA2, r = 0.31 (p = 0.057). Percentage EYA2 expression correlated with blast count, r = 0.425 (p = 0.008). The IHC expression of these antigens did not correlate with WHO MDS subclassification, IPSS, R-IPSS, disease progression, or survival (p > 0.05). Conclusions: IHC staining of inflammasome activators using TMA may allow better characterization of molecular pathways contributing the MDS pathogenesis. A correlation was seen between expression of antigens known to be increased downstream of NLRP3 inflammasome activation. Furthermore, increased expression of EYA2 correlated with blast count. A future study will compare expression patterns between normal, low risk MDS and high risk MDS samples and correlate these findings with clinical outcome data to further elucidate the pathogenesis of MDS and identify potential targetable markers for novel therapeutic strategies.