Immune Pathophysiology of Aplastic Anemia and Immunosuppressive Treatments

National Heart, Lung, And Blood Institute

Investigators

Linked publications, trials & patents

Abstract

In aplastic anemia (AA), the bone marrow is replaced by fat, blood cells are not adequately produced, and anemia, bleeding or infection result. AA is a disease of mainly young persons and, when severe, untreated is almost invariably fatal. AA historically was associated in the clinic to a diversity of putative exposures: chemicals s, in particular benzene; medical drugs; pregnancy; seronegative hepatitis; and select immune system diseases. The traditional nosology of AA based on clinical associations has been replaced with a simpler classification based on pathophysiology: damage due to physicochemical insults; constitutional syndromes such as telomere biology disorders, Fanconi anemia and others; and immune aplastic anemia as a distinct disease. We have developed and publicly posted a simple program from machine learning to discriminate between immune and genetic AA based on standard laboratory parameters—and telomere length of leukocytes. The serendipitous observation of autologous hematopoietic recovery post-marrow transplantation led to the positing of an immune pathophysiology. Anti-thymocyte globulin (ATG) produces hematologic recovery in most patients, and immunosuppressive therapies (IST) are the most frequent treatment of marrow failure worldwide. Research laboratory data have also revealed abnormalities of the immune system: lymphocyte populations that induce apoptosis in hematopoietic target cells by the Fas-mediated pathway, and oligoclones of effector T cells which express type 1 cytokines, especially gamma-interferon. Eltrombopag, a thrombopoietin mimetic, was shown by us to be effective in improving clinical outcomes, leading to its approval by the FDA for use in both refractory and treatment-nave severe AA. The Hematology Branch has been a global leader in both scientific research and medical studies of AA pathophysiology and treatment. Additionally, our collaborative and complementary efforts have advanced the field: the confirmatory RACE trial was led by former members of the Branch, and we continue to participate in the i4MDS international consortium, headed by a close collaborator at Kings College in London. In clinical work, our protocol for severe, treatment-nave AA purposely was extended to collect more clinical data and ancillary research laboratory samples, both aimed in particular at in depth understanding of alterations in hematopoiesis and the immune response and long-term outcomes before- and after-IST. Our large cohort is unique in providing accurate estimates of rates of response, relapse, and evolution to myeloid malignancies, important as the triple regimen is now standard-of-care worldwide. For the NIH cohort, for which accrual has now terminated, there have been no significant changes in rates of relapse and malignant evolution from the original publication. Prediction of the development of MDS and AML, the most serious late complications of aplastic anemia, has been a major focus. High risk clonal evolution occurs in about 10% of patients overall. We have defined two distinct patterns of malignant clonal progression, early (<1 year post-therapy) and late (>2 years). Early evolution is almost entirely driven by chromosome abnormalities, distinctively chromosome 7. When sufficient cells can be recovered from hypocellular marrow specimens obtained at diagnosis, 7- is detectable at low levels. Late evolution, which disappointingly can occur long after good hematologic remissions, is dominated instead by clones of cells somatically mutated in know myeloid malignancy genes, Utilizing single cell DNA genomics of serial samples, we have established high risk patterns, The presence of singly mutated clones is not predictive, but acquisition by ASXL1-mutated clones of additional spliceasome gene mutations, and less frequently RUNX1 mutations, is highly prognostic of cancer development. Conversely, patients other have a very low risk of evolution. In the research laboratory, advanced single cell methodologies have been applied to bone marrow obtained per protocol before and after IST. We utilized single cell RNA sequencing (scRNAseq), time-of-flight flow cytometry (CYToF), and proteomics. Cytotoxic lymphocytes in marrow are increased in AA and markedly reduced by triple therapy, but they are not eliminated; single cell techniques based on sequencing and flow cytometry are concordant in identification of unsuspected subsets of CD8 cells that are distinct to marrow failure and vary with treatment. The appearance of novel clones with T cell receptor sequences similar to those of initiating clones, suggesting that they also target hematopoietic stem and progenitor cells, carries a poor prognosis. Epitope spreading of the adaptive immune response may be an ongoing response to new antigens derived from destruction of marrow cells. Conversely, stem cells are markedly reduced at disease presentation, and short term (6 months) recovery, even in hematologic complete responders, occurs from expansion of early progenitor cells, with q continued marked stem cell deficit. Comprehensive depiction of disease bone marrow shows great complexity in the aberrant immune response. While cytotoxic lymphocytes dominate the picture, regulatory T cells, natural killer cells and B cells are not normal and there are marked interactions among these cells as imputed from ligand-receptor pairing in silico. T cell receptor usage, aberrant signaling pathways and clone size changes are concordant with our data in large granular lymphocytosis. Our current clinical protocol for treatment-naïve severe aplastic anemia tests early intervention using low risk oral therapy consisting of cyclosporine and eltrombobag, initiated before admission to the Clinical Center. The aim is to institute immunosuppression to prevent stem cell loss to lymphocyte cytotoxicity and stem cell stimulation to rapidly iprove hematopoietic function Nearly 50 patients have been enrolled on both the original protocol and an extension arm. early. The primary end points are safety of the regimen, as measured by drug toxicity, diagnostic errors, and patient compliance. A single instance of misdiagnosis occurred, in which a patient died; his presentation was unusual in several aspects, including delayed referral to treatment and presentation of late telomere biology disorder without family history, other apparent organ involvement, and hyperacuity of onset of pancytopenia. The overall response rate for early intervention equivalent to the standard regimen; notably neutrophil recovery, critical to patient survival, is more rapid than with conventional IST. The rate of evolution to myeloid malignancy has remained low, restricted to a single case of early high risk progression with chromosome 7 aneuploidy. Our ruxolitinib protocol, testing this jakinib in relapsed/refractory severe aplastic anemia, hypoplastic myelodysplastic syndrome, pure red cell aplasia, moderate aplastic anemia, and large granular lymphocytosis, continues to accrue. Our relapse prevention trial, assessing rapamycin (sirolimus) as a lymphocytotoxic, tolerogenic agent has been terminated and results are being analyzed. Protocols in development include participation in a multicenter test of regulatory T cells derived from umbilical cord in refractory patients, and substation of a romiplostim-like thrombopoietin agonist to substitute eltrombopag. We have developed a widely adopted mouse model of immune marrow failure, in which lymphocytes (without contaminating hematopoietic cells) produce severe aplastic anemia when infused into recipients across major and minor H2 antigen barriers. We previously reported the therapeutic efficacy of a commercial Jak1/Jak2 inhibitor, ruxolitinib, in our murine models. Ruxolitinib is effective even days after cell infusion. We have systematically tested other jakinibs, and only baricitinib also showed efficacy comparable to ruxolitinib broadly active agents that target both Jak1 and Jak2 may be preferred. In other experiments, having observed in our model the persistence of megakaryocytes in marrow despite severe peripheral thrombocytopenia, we characterized the functionality of these precursors. A method to enrich for megakaryocytes was developed for this purpose, and they were studied in a variety of assays of morphology, phenotype, transcriptional program, and functional inhibition of progenitor colony formation in vitro. We established that, in the setting of provoked immune destruction of hematopoiesis, megakaryocytes acquire or are selected for their innate immune role rather than for platelet production typical of the unstressed steady state. A role for innate immunity was also inferred from separate experiments examining the effect of CSF-1R inhibition on the immune pathophysiology. Hematopoiesis, immune cell populations, and gene expression were assessed by flow cytometry, cytokine analysis, and single-cell RNA sequencing. CSF-1R inhibition with the small molecule PLX3397, unusually, worsened marrow failure in recipients, leading to early mortality accompanied by enhanced inflammation and proinflammatory M1 macrophage polarization. These data are consistent with our prior results for myeloid-derived suppressor cells as regulators of the immune response in the rodent model.