Impact of Genetic Polymorphisms on Immune Response and Clinical Features in MDS.

The pathogenesis of cytopenias in myelodysplastic syndrome (MDS) is not fully explained by evolution of the clonal displacement of normal hematopoiesis. The quality of immune surveillance may contribute to generalized hematopoietic inhibition. E.g., certain cases may present with hypocellular marrow reminiscent of idiopathic aplastic anemia (AA). Immunogenetic background resulting from predisposing complex traits can influence the quality of immune response and shape clinical features of MDS. Immunogenetic factors include: KIR and KIR-ligand (KIR-L) genotype, as well as cytokine/receptor single nucleotide polymorphisms (SNP). We genotyped 130 patients with MDS (32 RA/RAMCD, 24 RARS/ RAMCD-RS, 51 RAEB/sAML, 23 MDS/MPD and CMML). Controls included current (n=87) and historical controls. We studied HLA type, KIR, KIR-L genotypes, and immunomodulatory SNPs, including: IL-1α (−889 T/C), IL-1R (−1970 C/T), IL-1RA (mspa111100 T/C), IL-4RA (+ 190 G/A), IL-1β (−511 C/T, +3962 T/C), IL-6 (−174 C/G, nt565 G/A), IL-10 (−1082 G/A, −819 C/T), IL-12 (−1188 C/A), TGF-β (+10 C/T, +25 G/C), TGF-βR2 (+358 A), INF-γ (+874 A/T), TNF-α (−308 G/A, −238 G/A), CTLA-4 (exon 1, +49 A/G), FcγIIIR (+559 G/T) as well as SNPs in CD45 gene (exon 4 +77 C/G, +138 A/G). In all MDS, no difference in the frequency of KIR genotype constellations was identified. However, higher frequency of stimulatory KIR2DS2 in low grade MDS (72% vs 48% in controls, p=.01) and 2DS5 in hypocellular MDS (62% vs 26%, p=.02) was found. When KIR-L C1/C2 genotype was studied, no significant difference in haplotype frequency was observed compared to controls. However, an increased incidence of C2/C2 was found in high grade MDS (70% vs 24%, p=.003). Analysis of the resulting KIR/KIR-L combination demonstrated increased frequency of inhibitory KIR2DL3/C1 mismatch (34% vs 16%, p=.04) in the MDS cohort consistent with higher cytotoxicity. A similar observation was made when we compared advanced and low-grade MDS (70% vs 23%, p=.02). A higher proportion of 2DS2/C1 mismatches (50% vs 10%, p=.006) in advanced MDS could reflect lesser degree of immune surveillance facilitating clonal expansion. No significant difference in MDS cohort, nor in any of subgroups compared to control and each other, was found for the SNPs in IL-4Rα, IL-12, IL-1β, IL-6, TGF-βR2, IL-10, IL-12, IL-1α, IL-1R, FcγIIIR, and CD45. However, when we examined the frequency of INF-γ and TGF-β genotypes, increased frequency of INF-γ AT variant (62% vs 33% in controls, p=.015) and a decrease frequency of TGF-β genotype T/T-G/G or T/C-G/G (53% vs 74%, p=.04) were found, consistent with high secretor phenotype. In addition, higher incidence of G/G phenotype for the CTLA-4 gene (p=.0001) was found in the MDS cohort and in low grade MDS (p=.003). Heightened immune response could result in similar hematopoietic suppression as observed in AA, and lead to hypoplasia. Consequently, we subdivided all MDS patients according to marrow cellularity and found that hypoplastic variant of MDS (n=10) was characterized by higher prevalence of A/T phenotype of INF-γ gene (70% vs 33%, p=.02), similar to a cohort of AA patients (n=26, 50% vs 33%, p=.05). An analogous observation was also made for G/A genotype of TNF-α. In sum, our findings demonstrate that a large number of immunogenetic predisposition factors associated with more risk immune response may exist in MDS, which may influence certain clinical features and phenotype.