Active Immune Thrombocytopenia (ITP) Disease Is Characterised By a Reduced Treg:CD8 Effector T Cell Ratio Which Is Modulated By Thrombopoietin-Receptor Agonists (TPO-RA)

Background: Immune thrombocytopenia (ITP) is an autoimmune disorder characterized by isolated low platelet count and a skewed proinflammatory Th1/Th17 profile. However, little is known about the involvement of CD8⁺ cytotoxic T cells in ITP pathophysiology and whether they are regulated by regulatory T cells (Treg). Immunosuppressive therapy has been the mainstay treatment in ITP. More recently, Thrombopoietin receptor agonists (TPO-RA); Romiplostim (Romi) and Eltrombopag (EPAG), have been increasingly used to stimulate megakaryocytopoiesis to produce more platelets. TPO-RAs are reported to induce complete remission in up to 30% of cases, with limited understanding of their impact on the immune system. Here we describe changes in T cell subsets in patients with ITP: how these changes are affected by disease activity and how TPO-RA may induce remission through modulating the immune system.

Methods: Multi-color flow cytometric panels were designed to characterize peripheral blood T cell subsets, including CD8⁺ T cell and Treg subsets, phenotypically as well as functionally through intracellular cytokine expression. To determine whether CD8⁺ cells were platelet specific, an IFNγ ELISpot assay was performed using platelets from a healthy donor and PBMC from both HC and patients. Forty patients with ITP were included: 13 were on Romi, 11 on EPAG and 16 on no treatment at the time of analysis. Of these 40 patients, 15 patients had active disease (AD) (platelet-count less than 30 x 10⁹/L) and 25 had stable disease (SD) (> 30 x 10⁹/L). These patients were compared with 26 age and gender-matched healthy controls (HC). Data were presented as median values; Mann Whitney U and Kruskal Wallis tests were used with Dunn's multiple comparisons correction; a P value of < 0.05 was considered significant.

Results: CD4/CD8 T cell ratio was significantly lower in patients compared to HC [1.77 vs. 3.97; P value < 0.001]. CD45RA⁺CD62L^- Terminally-differentiated CD8⁺ T cells were significantly higher in patients compared to HC [66.3% vs. 8.56%; P value < 0.001]. This finding was more prominent in AD patients than those with SD [66% vs. 44.4%; P value < 0.05]. This effector population is polyfunctional, expressing high levels of proinflammatory cytokines including TNFα, IFNγ and Granzyme B when compared to HC [P value < 0.05]. Additionally, this population lacks the exhaustion markers PD-1 and Tim-3. Furthermore, these cells were reactive to platelets showing higher IFNγ-producing cells when co-cultured with platelets.

CD3⁺CD4⁺CD25^hiCD127^lo Treg frequency did not differ between patients and HC [P value>0. 05]. Treg functionality was preserved in these patients; no important changes were observed in their capacity to express interlukin2 intracellularly, nor in the cell surface receptor (CD25) [P value > 0. 05]. However, Tregs were significantly lower in AD patients with compared to SD patients [2.32% vs. 4.46%; P value < 0.05]. Treg function is interactive with other T cell subsets and depends on its relative abundance in relation to other subsets; The Treg/effector CD8⁺ T cell ratio was significantly lower in patients compared to HC [0.06 vs. 0.16; P value < 0.01] and was also significantly lower in AD compared to SD patients [0.03 vs. 0.09 ; P value < 0.05].

EPAG-treated patients had a significantly lower effector CD8⁺ T cells compared to Romi-treated patients [42.4% vs 76.8%; P value <0.01]. Although EPAG did not have a direct effect on the frequency of Treg, the Treg/effector CD8⁺ T cell ratio was significantly lower than that in patients on Romi [0.04 vs. 0.11; P value < 0.05] because of the CD8^+.T cell difference. The Treg/effector CD8⁺ T cell ratio in EPAG-treated patients was comparable to the ratio of HC [0.11 vs. 0.16; P value > 0.05].

Conclusion: While Th1/Th2 cell ratio is often considered as driving ITP, these results demonstrate the involvement of cytokine secreting effector CD8⁺ T cells in the disease pathogenesis. The imbalance in immune tolerance is also highlighted in the form of a significant reduction in the Treg/effector CD8 T cell ratio. The differences seen in T cell subsets between EPAG- and Romi-treated patients suggest the potential for an additional, differential immunomodulatory effect between the two agents which is currently being explored.

Acknowledgment: The authors wish to thank Prof James B Bussel for his insightful comments and support of this work.