Permissive Conditions for Evolution of PNH Clones Are Characterized by Overproduction of IFN-γ by Clonal CD4 and CD8 T Cells, Fas-L by CTLs, and Promoted by Immunogenetic Background

The association between immune-mediated aplastic anemia (AA) and paroxysmal nocturnal hemoglobinuria (PNH) has been well documented, yet the related immune pathophysiology remains ill-defined. Response of AA patients with PNH clones to immunosuppressive agents such as anti-thymocyte globulin (ATG) and cyclosporine A provides clinical evidence for the involvement of the immune system in the evolution of PNH. The similar immunobiology of PNH and AA is also exemplified by the overrepresentation of HLA-DR*15 in AA, AA/PNH and PNH. To discern the relationship between T cell responses and effector mechanisms we applied a battery of immune tests to patients with these rare diseases. First, using T cell receptor Vβ flow cytometry in a cohort of patients with AA (N=42), AA/PNH (N=15), or PNH (N=18), we identified cytotoxic CD8 T cell (CTL) clonal expansions in 28/42 (66.7%), 9/15 (60%), and 7/18 (38.8%) patients, respectively. CD4 T cell expansions were present in 6/42 of AA (14.3%), 2/15 (13.3%) of AA/PNH, and 2/18 (11.1%) of PNH patients. We then studied whether expanded clones were associated with production of inflammatory cytokines; across the entire cohort, patients with clonal CD8 Vβ expansions demonstrated a significantly increased proportion of IFN-γ producing T cells as well as elevated levels of circulating Fas-L when compared to patients without clonal skewing (p=.032 CD4⁺IFN-γ⁺, p=.008 CD8⁺IFN-γ⁺, p=.097 sFAS-L). Even more pronounced was the increase in the proportion of IFN-γ producing CD4 T cells in patients with clonal CD4 Vβ expansions (p=.010). Furthermore, while a strong trend toward increased sFAS-L as detected by ELISA was found in patients with CD8 Vβ skewing vs. those without, patients with pronounced CD4 expansions did not produce elevated Fas-L levels, consistent with different effector mechanisms employed by CD4 vs. CD8 T cells. Based on these results we hypothesized that the presence of PNH clones will be associated with activation of immune effector mechanisms. Linear regression analysis of the size of the PNH clone vs. proportion of IFN-γ CTLs displayed a positive correlation that nearly reached statistical significance at α=0.05 (p=.067). A high proportion of CD4 IFN-γ cells (defined by a value above 95% mean confidence intervals of controls) was also associated with the presence of PNH (p=.048). Genetic analysis revealed further clues as to the increased propensity of patients with AA and PNH clones to produce elevated levels of IFN-γ; the hypersecretor genotype T/T for IFN-γ was over-represented in AA (28% vs. 10% in controls, p=.02) and correlated with presence of a PNH clone (35% vs. 14%, p=.01). An essential role of T cells in generating permissive conditions for the evolution of PNH clones is also supported by the immunogenetic relationship of PNH to HLA-DR*15, a relationship which was confirmed in our population of patients: phenotype frequency of HLA DR*15 was 42.8% AA, 40% AA/PNH, 27.8% PNH vs. 17.2% in control group. When HLA DR*15 positive and DR15 negative patients were compared, those with DR*15 displayed a strong trend toward increased proportion of CD8⁺IFN-γ producing cells (p=.094), previously shown to be elevated in patients with PNH clones. Our results reveal insights into the nature of permissive conditions involving oligoclonal T cell responses, oversecretion of proinflammatory cytokines, and immunogenetic background which together may promote the expansion of PNH clones. Conversely, it remains possible that the cytotoxic milieu may be a result of an immune response directed against intrinsically abnormal PNH clones.