Abstract
Background: Immune aplastic anemia (IAA) is a T-cell mediated disease manifesting as pancytopenia associated with increased risks of infection, bleeding complications and transfusion dependence. Platelet transfusion is the most common cause of human leukocyte antigen (HLA) class I alloimmunization. A preemptive threshold for platelet transfusion is not as well defined as for packed red blood cells. Previously, we have reported an adverse correlation between HLA class I alloimmunization and overall survival (OS) in IAA but without an explanation for this observation. We hypothesized that the HLA class I antibodies (Ab) might modulate autoimmunity via complement mediated immune activation. Platelet sensitization results from IgG alloantibody mediated opsonization, C1q associated membrane attack complex formation, de-glycosylation and activation of CD32a/ Fc-gamma-IIa receptor pathways. C3a and C5a are potent anaphylatoxins, which recruit inflammatory cells, activate phagocytotic cells, release granule-based enzymes and generate of oxidants. Increased cytokines IL-6, IL-8, TNF-α, IFN-gamma, and increased cytotoxic CD8 activity would modulate innate immune function and potentially worsen IAA. Utilizing single cell RNA sequencing (scRNA-seq), we assessed transcriptomic differences between samples from patients with and without alloimmunization.
Methods: Samples were obtained from patients on clinical trial NCT01623167 (horse- ATG, cyclosporine and eltrombopag). ScRNA-seq was performed on pre-IST treatment bone marrow mononuclear cells (BMMNCs) using the 10x Genomics Single Cell Immune Profiling Solution v 1.1 (Chromium Single Cell 5' Reagent Kit v1 and v1.1). Library preparation, sequencing, data processing, and quality control were done following previously published pipeline. Differentially expressed genes were identified with the FindMarkers function in Seurat, by comparing gene expression in one cell subset with expression in all others. Gene Set Enrichment Analysis (GSEA) was done based on fold changes of all detected genes.
Results: Twenty patients with AA (8 HLA class I alloimmunized and 12 non-alloimmunized) had scRNA-seq performed. Six in the HLA alloimmunized and 7 in non-alloimmunized group had very severe AA. Median ages for HLA class I alloimmunized and non-alloimmunized patients were 35 years (range 11-64) and 36.5 years (range 13-73). The alloimmunized group had more females (n=5) compared to non-alloimmunized (n=4). Patient cohort was diverse (3 White, 8 Blacks, 8 Hispanics and 1 Asian). Presence of a PNH clone was more prevalent in the alloimmunized group, 50% (n=4) vs. 33%(n=4).
Cell population counts were similar between the groups for CD4+, CD8+, Pro-B, B-plasma, natural killer (NK) cells, monocytes and hematopoietic stem and progenitor cell (HSPC). However, BMMNCs from sensitized patients, including CD4+, CD8+ T and NK cells, exhibited upregulated gene-expression associated with interferon gamma (IFN) response, tumor necrosis factor alpha (TNF-α) via nuclear factor-κB (NF-κB), and complement. The top upregulated genes were TNAIP3, NFKBIA, CCL5, JUN, IRF7, CD69 and CD74; these have been associated with chemotaxis, recruiting a variety of leukocytes to inflammatory sites, and T cell maturation, consequently modulating immune responses.
Similarly, inflammatory and complement activation pathways were upregulated in the differential expressed genes analysis of myeloid cell subpopulations. Proteins encoded by upregulated genes in HLA alloimmunized SAA were involved in interferon gamma response, TNF alpha signaling via NF-κB, and complement pathway exhibited functional interactions.
Conclusion: The coexistence of upregulated complement and immune pathways in HLA alloimmunized patients indicates a possible association. The phlogogenic C3a and C5a may be responsible for cascading immune activation and worsened disease pathophysiology providing a plausible explanation for the clinical finding of reduced OS association with HLA class I alloimmunization.
No relevant conflicts of interest to declare.