Dendritic cell, monocyte, B and NK lymphoid deficiency defines the lost lineages of a new GATA-2 dependent myelodysplastic syndrome

A novel immunodeficiency syndrome has recently been defined in young adults. Known variously as dendritic cell, monocyte, B and NK lymphoid (DCML deficiency),1 ‘autosomal dominant and sporadic monocytopenia’2 or ‘MonoMAC’ (monocytopenia with Mycobacterium avium complex),3 it is characterized by a composite mononuclear cell deficiency, atypical mycobacterial and viral infection, and progression to myelodysplasia and leukemia. Strikingly, a number of patients also develop pulmonary alveolar proteinosis (PAP). Patients typically present in their 3 or 4 decade, although historical blood counts may show pre-existing monocytopenia for ten years or more. A substantial number of patients develop autoimmune phenomena such as erythema nodosum/panniculitis or arthritis. Solid tumors are common, notably carcinoma in situ associated with chronic infection with human papilloma virus. Up to a third of cases have a family history with apparent autosomal dominant inheritance. Left untreated, many will die from infection, leukemia or respiratory failure. Loss of dendritic cells (DCs), monocytes and B/NK lymphoid cells is universal, hence the term DCML deficiency to describe this syndrome. Infection with Mycobacterium avium complex, alluded to in the term ‘MonoMAC’, is seen in the majority of patients but not all.

DCML deficiency/MonoMAC was first described by Steven Holland and colleagues at the National Institutes of Health, in a paper published in the journal Blood in 2010.2 A smaller group of subjects were also described earlier this year in a report highlighting the loss of dendritic cells and specific deletion of bone marrow progenitor subsets. This study also demonstrated a marked increase in Flt-3 ligand and loss of regulatory T cells (Tregs).1 As this article went to press, the genetic basis was solved in two papers which are shortly to be published in the journal Blood. GATA-2 mutation has been found in nearly all cases, first in a candidate gene study from the Holland group4 and second by exome sequencing of 4 unrelated patients.5 In this issue of the journal, Calvo et al. from the Holland group take a more in depth look at the hematologic features as bone marrow failure progresses, providing evidence that DCML deficiency/MonoMAC inevitably becomes a more recognizable yet distinct form of myelodysplasia (MDS).3 A model of the pathogenesis of this disorder incorporating these new findings is presented in Figure 1.

Through this careful description, Calvo and colleagues show clearly that advanced DCML deficiency overlaps with MDS; 61%–78% of their patients had an additional cytopenia and nearly half (39%) were pancytopenic. More than half (59%) also had cytogenetic abnormalities, most commonly monosomy 7/7q deletion, trisomy 8 and trisomy 1. The study subjects are necessarily a more advanced group of patients than previously reported as the inclusion criteria were defined as both a diagnosis of ‘MonoMAC’ and the development of “additional cytopenia(s) typically associated with MDS”. The combination of cytopenias and cytogenetics gave 15 of 18 patients intermediate to high scores according to the MDS International Prognostic Scoring System (IPSS).

It is worth emphasizing at this point that DCML deficiency patients may have significant risk to their health before additional cytopenias evolve.2 Disseminated mycobacterial infection, PAP and carcinoma in situ may all occur prior to the onset of an obvious myelodysplastic syndrome, underscoring the fact that immunodeficiency is a cardinal feature of the disorder. In the series of 4 subjects described in 2011, one had mild anemia (10.3 g/dL) and one mild thrombocytopenia (123×10/L), yet 2 required hematopoietic stem cell transplantation (one for refractory mycobacterial infection, one for PAP) and a third died of influenza H1N1 prior to transplantation.1 In addition, many other factors may contribute to mild cytopenias, including: atypical mycobacterial infection, respiratory failure, autoimmune thrombocytopenia and hemolysis, and drug therapy.

Table 1.Diagnostic criteria for DCML deficiency/MonoMAC.

The second finding from the current study is that although DCML deficiency can progress to a form of MDS as defined by standard criteria, it has some very particular and unusual features that make it distinct from typical MDS. The main diagnostic features are summarized in Table 1. Important clues are often present in patients’ case history such as a much younger age at presentation and longstanding problems with warts or poorly characterized autoimmune phenomena. Atypical infection, respiratory failure, cytopenias and family history are much stronger warnings that this is not typical MDS.

The most obvious laboratory finding is, of course, the loss of mononuclear cells: the ‘lost lineages’ of a failing marrow that are not recognized by the standard definitions of ‘trilineage dysplasia’. In particular, it should be easy to identify monocytopenia from the usual monocytosis of sporadic MDS. It is important to note that moncytopenia may be the only deficit to be seen from automated counting because the loss of B cells and NK cells is masked by preservation of peripheral T cells. DCML deficiency has now earned its place in the textbook alongside hairy cell leukemia, as a differential diagnosis of monocytopenia. Vinh et al. previously observed that the monocyte deficit is usually much more profound in DCML deficiency than hairy cell leukemia (0.01×10/L compared with 0.07×10/L). It is also vital to recognize that patients may present to a number of different medical specialities, including genito-urinary medicine, infectious disease, respiratory medicine and rheumatology, and that failure to recognize monocytopenia in this context is likely to result in a missed diagnosis.

Calvo and colleagues present very convincing additional diagnostic findings that separate DCML deficiency from typical MDS. In most cases, megakaryocytes are dysplastic, even in the absence of thrombocytopenia, and there is a high rate of fibrosis (73%). Bone marrow cellularity tends to be normal or low (89%), even when additional cytopenias have begun to evolve; it is only the very advanced leukemic phases in which cellularity finally begins to increase and become more like typical proliferative MDS. In the description of DCML deficiency, the elevation of Flt-3 ligand was highlighted.1 Although this has been described in aplastic anemia,6 it is not a typical feature of MDS and might become a useful measurement. The progenitor cell compartment was also characterized in detail showing selective loss of the multi-lymphoid progenitor (MLP), a recently described pluripotent cell which appropriately gives rise to DC, monocytes and lymphoid cells.7 Granulocyte macrophage progenitors (GMP) are also depleted but are present in sufficient numbers to sustain granulopoiesis in the early phase of the disease. All these observations point to the mechanism of pathogenesis being an intrinsic stem cell defect that preferentially affects mononuclear cell production before progressing to global hematopoietic failure. This mechanism has now been confirmed by the discovery of GATA-2 mutations which are likely to compromise stem cell survival and are also known to affect the formation of GMP, monocytes, B cells and NK cells in murine models of haplo-insufficiency.8,9

Figure 1.Evolution of DCML deficiency/MonoMAC. Schematic representation of the evolution of DCM deficiency beginning with an asymptomatic phase of cellular deficiency gradually evolving probably in the first decade of life. Infections begin in the second and third decade followed by increasing risk of carcinoma related to HPV, respiratory complications and finally overt leukemic transformation. BM cellularity declines and Flt-3 ligand probably rises in an inverse relationship until the proliferative phase of MDS occurs. This coincides with the acquisition of cytogenetic abnormalities and onset of cytopenias. MLP: multi-lymphoid progenitor; GMP: granulocyte-macrophage progenitor; HSC: hematopoietic stem cell; MEK megakaryoctye/erythroid progenitor.

The evaluation of lung disease in DCML deficiency is also very important. Mycobacterial infection may promote the pathogenesis, yet also obscure the diagnosis, of underlying PAP. The critical test in DCML deficiency is that GM-CSF antibodies are not detected and there is no response to administration of GM-CSF, unlike most cases of acquired PAP.10 Physicians evaluating respiratory problems define the PAP of DCML deficiency as ‘secondary’ to bone marrow failure, although this is not as straightforward as it appears.11 Unexpectedly, tissue macrophages, including alveolar macrophages are present in normal numbers, despite the profound monocytopenia.1,2 There is speculation about whether alveolar macrophages survive for very long periods, proliferate locally or are even derived from sources other than monocytes, but the fact remains that the resident cells in DCML deficiency must be defective in function rather than simply depleted in number. In support of this, there is evidence that GATA-2 plays a role in phagocytosis.12 Case reports of secondary PAP in the context of monocytopenia or MDS have been known for a number of years and were very well reviewed in the original paper on DCML deficiency by Vinh et al.2

The new study by Calvo et al. is very informative about the later stages of bone marrow failure in DCML deficiency but also provides a framework for conceptualizing the entire evolution of the syndrome. Most patients have survived childhood vaccination and viral illness without incident and at presentation have specific anti-viral titres and normal levels of immunoglobulin. Calvo et al. also show plasma cells in the bone marrow and highlight the relative preservation of peripheral T cells, although there are interesting abnormalities of phenotype in both. Together, these observations suggest that immunity was intact in infancy allowing some B- and T-cell memory to form. Indeed, in the original case series, normal blood counts were found in one child in infancy, prior to the onset of atypical infection.2

Unravelling the genetic basis of this disorder is important to provide confirmatory testing for families with affected individuals and links DCML deficiency to a group of hereditary leukemias described in a third independent study.13 In terms of therapy, the NIH group has already led the way, showing that hematopoietic stem cell transplantation is potentially curative, and can be performed even on patients with advanced respiratory failure, so rapid is the correction of the alveolar macrophage defect. It is very likely that many more cases exist, probably a pedigree or a number of sporadic cases in every major city. The discovery and successful treatment of these patients will depend upon the recognition of monocytopenia as the vital clue in revealing the many guises of DCML deficiency.

Footnotes

• Venetia Bigley is George Walker Fellow in Haematology at Newcastle University and Honorary Consultant Haematologist at Newcastle upon Tyne Hospitals, UK. Matthew Collin is Professor of Hematology at Newcastle University and Honorary Consultant Hematologist at Newcastle upon Tyne Hospitals, UK.
• Related Original Article on page 1221
• Financial and other disclosures provided by the author using the ICMJE (www.icmje.org) Uniform Format for Disclosure of Competing Interests are available with the full text of this paper at www.haematologica.org.