C h a p t e r 5 1

Bone Marrow Failure Syndromes : An Overview

PK Muthukumaraswamy
Professor of Medical Oncology,
Madurai Medical College & Government Rajaji Hospital, Madurai, Tamil Nadu

Definition
Bone marrow failure may be simply defined as pancytopenia (anemia, leukopenia, and
thrombocytopenia, sometimes in various combinations) resulting from deficient hematopoiesis, (as
against the cytopenias arising from peripheral destruction). This concept of bone marrow failure is
imprecise but convenient.1,2 The classification shown in Table 1 is all-inclusive, and attempts to group
the syndromes according to the mechanism of the bone marrow failure and to distinguish the inherited
or congenital syndromes from the acquired. Occasionally, bone marrow failure syndromes manifest as
single lineage disorders (Table 2). As will be apparent from the tables, the term bone marrow failure is
rather vague and inclusive and awaits redefinition with more precise understanding of pathophysiologic
processes.
One approach to a more precise definition of the bone marrow failure syndromes is to consider only
the hypoproliferative states associated with bone marrow damage as ‘pure’ bone marrow failure.
Hypoproliferative anemias are characterized by normocytic and normochromic red cells and an
inappropriately low reticulocyte response (reticulocyte index < 2.5); and includes early iron deficiency
(before hypochromic microcytic red cells develop), acute and chronic inflammation (including many
malignancies), renal disease, and hypometabolic states such as protein malnutrition, all of these being
characterized by an abnormal erythropoietin response.
The hypoproliferative anemias associated with marrow damage include aplastic anemia,
myelodysplasia, pure red cell aplasia, and myelofibrosis. Anemia in these disorders is not a solitary or
even the major finding; bone marrow failure state being a much more apt descriptor. This concept is
not perfect (it excludes the proliferative myelodysplasias), and is open to challenge.
The pathogenesis of most of these disorders is unknown and their separation depends mainly upon
morphological criteria: the fatty bone marrow of aplastic anemia, the disordered hematopoiesis of the
myelodysplasias, and the fibrosis of myelofibrosis.
While practical distinction among these syndromes is clear in typical cases, they are sometimes so
interrelated that the differential diagnosis may be arbitrary and one diagnosis may appear to evolve
into another. This metastable nature confirms the important pathophysiologic relationship among
372 CME 2004
Table 1 : Classification of Bone Marrow Failure

Diseases
Pathogenesis
Acquired Congenital
Hematopoietic stem-cell failure Acquired aplastic anaemia Fanconi anaemia
Dyskeratosis congenita
Hematopoietic failure during Pure red-cell aplasia Diamond-Blackfan anaemia
differentiation Amegakaryocytic thrombocytopenia Thrombocytopenia with absent
Chronic acquired neutropenia radii
Kostmann’s Syndrome
Congenital dyserythropoietic
anaemias

Proliferative dysplasias with Myelodysplastic syndromes

abnormal differentiation
Abnormal environment Proliferative dysplasias with fibrosis Osteopetrosis
Myelofibrosis
Infiltrations Leukaemias / lymphomas
Lipid-storage disease
Amyloid

Infections HIV
Dengue fever
Parvovirus B19
Bone marrow necrosis

Table 2 : Bone Marrow Failure Affecting Single Cell Lines

Disease Cause Example

Inherited Diamond-Blackfan syndrome
With thymoma
Idiopathic / autoimmune With other autoimmune disorders
Isolated
Red cell aplasia Diphenylhydantoin, chlorpropamide, chloramphenicol,
Drug induced
D- penicillamine
Aplastic crisis in haemolytic anaemia (Parvovirus)
Virus induced
Transient erythroblastopenia in childhood
Riboflavin deficiency Experimental
Inherited Kostmann syndrome (Infantile genetic agranulocytosis)
Scwachman-Diamond syndrome (With pancreatic insufficiency)

Neutropenia Idiopathic / autoimmune Cyclical neutropenia

Drug induced Thiazides, semisynthetic penicillins
Virus induced Rubella
With total absence of radii
Congenital
Amegakaryocytosis Isolated
Acquired Variant of aplastic anaemia

these syndromes, in their sharing of immune-mediated mechanisms of marrow destruction and some
element of genomic instability resulting in a higher rate of malignant transformation.
This article will exclude detailed discussion myelodysplasias, as they are in many ways, a distinctive
Bone Marrow Failure Syndromes: An Overview 373
entity. Instead, the focus will be on some unique pathophysiologic mechanisms leading to bone
marrow failure, and on the general principles of management of these diverse entities.

Incidence
The myelodysplastic syndrome is the most common cause of primary bone marrow failure, with an
incidence of 5 per 100,000 population per year in those aged 50 to 59 years, which rises to 15 / 100,000
for the age group of 60 – 69 years, and to 50 / 100,000 in the ages 70 – 79 years. Myelodysplasias
do occur in childhood also, albeit rarely. In the west, 15 – 20 % of myelodysplastic syndromes are
therapy-related. Paroxysmal nocturnal hemoglobinuria is reported to have a prevalence of 1 / 100,000
population. Acquired aplastic anemia has a three-to five-fold higher incidence in Asia than in the west.
Applying strict diagnostic criteria, the International Aplastic Anemia Study has reported an incidence of
0.2 / 100,000 per year in Europe and the United States. Idiopathic myelofibrosis is seen in about 0.5
persons / 100,000 population per year. The inherited / congenital bone marrow failure syndromes are
exceedingly rare.

Some Representative Bone Marrow Failure Syndromes

Acquired Aplastic Anemia
Acquired aplastic anemia is in many ways, the paradigm of bone marrow failure syndromes. The
condition is simply defined as peripheral blood pancytopenia and a hypocellular bone marrow.
Depending upon the presumptive etiology, the following forms are recognized: (1) Inevitable aplastic
anemia, which follows exposure to cytotoxic drugs or irradiation in a dose – related ‘inevitable’ manner;
(2) Idiosyncratic aplastic anemia, which occurs unexpectedly in rare individuals exposed to drugs /
chemicals such as NSAIDs, thyrostatics, psychotropics, anticonvulsants, antidiabetics, chloramphenicol,
sulfonamides, antimalarials, certain cardiovascular agents, benzene, aromatic hydrocarbons, heavy
metals etc; without a clear dose-response relationship; (3) Idiopathic aplastic anemia, which represents
most cases of this disorder; (4) Infections causing aplastic anemia, most commonly a hepatitis (non-A,
non-B, non-C, non-G), Infectious mononucleosis, Epstein-Barr virus infection, Parvovirus B19 infection
etc.; (5) Immune aplastic anemia, which is usually of short duration and sometimes associated with
circulating antibodies, as seen in systemic lupus erythematosus, eosinophilic fasciitis and transfusion-
associated graft-versus-host-disease; and (6) Malignant aplastic anemia, which is transiently seen in the
course of the evolution of acute leukemias and myelodysplasias.
The typical patient is a young and previously well individual who presents with symptomatic anemia
and mild bleeding diathesis and is found to have peripheral blood pancytopenia and a hypocellular
marrow replaced by fat. Although global bone marrow failure is common, the diminished marrow
function may affect the three hematopoietic lineages in a non-uniform manner, with odd combinations
of bicytopenias and even monocytopenias. The degree of bone marrow hypocellularty may be variable
and does not always correlate with the peripheral blood counts. Even an apparently typical idiopathic
aplastic anemia may vary in its clinical presentation and course, from a fulminant illness with major
infections and bleeding, to an indolent process manageable by transfusions alone.

Idiopathic Myelofibrosis (Agnogenic Myeloid Metaplasia)

Idiopathic myelofibrosis, otherwise called as agnogenic myeloid metaplasia or myelofibrosis with
myeloid metaplasia is a clonal disorder of a multipotent hematopoietic progenitor cell.3 Its etiology
is unknown. The disease is characterised by marrow fibrosis, myeloid metaplasia, splenomegaly and
extramedullary hematopoiesis. The usual presentation is with asymptomatic splenomegaly. These
patients develop progressive anemia. Exuberant extramedullary hematopoiesis can lead to ascites,
pulmonary hypertension, pericardial tamponade, spinal cord compression, skin nodules etc. Idiopathic
myelofibrosis progresses to inexorable marrow failure and transfusion dependent anemia. The patients
eventually die of cardiac failure or serious infections. About 10 % terminate in an acute leukemia. There
374 CME 2004
are no specific diagnostic features and the diagnosis is established by exclusion. Teardrop red cells and
nucleated red cells in the peripheral smear and pathologic evidence of extramedullary hematopoiesis
are important diagnostic clues.

Paroxysmal Nocturnal Hemoglobinuria

Paroxysmal nocturnal hemoglobinuria is an acquired primary bone marrow disorder that not only
affects the red cell lineage but also affects the platelet, leukocyte, and pluripotent hematopoietic stem
cell lines. It is believed to be a disorder of stem cells of a clonal nature and can arise from or evolve
into other dysplastic bone marrow diseases, including aplastic anemia, sideroblastic anemia, and
myelofibrosis. The disease is caused by a specific genetic mutation in the PIG gene located in the short
arm of the X chromosome (see below).
Paroxysmal nocturnal hemoglobinuria was initially thought to consist of paroxysms of intravascular
hemolysis causing nocturnal hemoglobinuria. It is now recognized that chronic intravascular hemolysis
is the more frequent clinical finding. Rarely, Paroxysmal nocturnal hemoglobinuria may also evolve
into acute leukemia. The three major pathologic processes of the disease are hemolysis, thrombosis
and pancytopenia.4 These processes do not always manifest concurrently. It is not unusual to see
either hemolysis or thrombosis dominate the clinical picture. About a third of patients with paroxysmal
nocturnal hemoglobinuria will have received a formal diagnosis of aplastic anemia before the
condition was recognized. Another third are first diagnosed as paroxysmal nocturnal hemoglobinuria
with hemolysis and / or thrombosis, and they eventually manifest aplastic anemia during the ensuing
years and decades. The remaining third of ‘hemolytic’ paroxysmal nocturnal hemoglobinuria patients
manifest bone marrow failure only in laboratory assays of hematopoiesis.
Hemolytic paroxysms will lead to jaundice. At these times dark urine may be apparent. Thrombotic
complications usually involve the splanchnic, hepatic or cerebral venous systems. The ‘aplastic’
patient will often manifest symptomatic anemia, neutropenia causing infectious episodes and
thrombocytopenia leading to mucocutaneous bleeds.
Ham’s test (lysis of the patient’s RBCs after complement activation by acidification or sucrose) has been
the mainstay for the diagnosis of paroxysmal nocturnal hemoglobinuria. Today, it is best established by
flow cytometric documentation of granulocytes deficient in CD 55 and CD 59.

Fanconi’s Anemia
Fanconi’s anemia is an autosomal recessive disorder resulting from several different genetic defects.
The disease is manifested by a progressive pancytopenia that is in no way different from acquired
aplastic anemia. However these patients show congenital developmental anomalies, increased
risk of malignancy and laboratory evidence of increased chromosomal fragility. These patients
characteristically have short stature, café au lait spots, skeletal abnormalities and other congenital
anomalies (Table 3). Fanconi’s anemia patients are susceptible to both hematological and solid organ
malignancy. Fanconi’s anemia is usually grouped with the inherited cancer-prone syndromes such as
ataxia telangiectasia, Bloom’s syndrome, and xeroderma pigmentosum. For some of these disorders,
the genetic defect has been identified and correlated with a defect in DNA repair. For others, including
Fanconi’s anemia, the relationship to DNA repair is inferential.

Pathophysiology
Acquired Aplastic Anemia
Bone marrow failure results from a severe damage to the hematopoietic stem cell compartment. This
is reflected in the replacement of the bone marrow by fat as seen in magnetic resonance imaging
of the bone marrow of the vertebral column and in the morphology of the biopsy specimen. Flow
Bone Marrow Failure Syndromes: An Overview 375
cytometry shows cells bearing the CD34 antigen in the bone marrow (most of the committed progenitor
cells and stem cells) to be greatly diminished. In vitro assays of bone marrow function, by cultures in
semisolid media to enumerate cells with lineage commitment, show a severe reduction in the number
of hematopoietic progenitor cells. These assays suggest that the stem cell pool is reduced to < 1%
of normal at the time of presentation. Qualitative abnormalities such as limited number of functional
stem cell clones and shortened telomere length, reflecting the shrunken state of hematopoiesis are also
seen.
Altered drug metabolism may be a likely mechanism for the genesis of idiosyncratic aplastic anemia.
The metabolic pathways of many drugs and chemicals involve enzymatic degradation to highly
reactive electrophilic compounds which are toxic to cellular macromolecules. Excessive generation
of such toxic intermediates or the failure to detoxify them may be genetically determined and would
become apparent only on specific drug challenge.
Acquired aplastic anemia appears to result from the destruction of hematopoietic cells by complex but
ultimately similar immunologic mechanisms incited by chemicals / drugs / viruses. In this immunologic
mechanism, aplastic anemia shares pathophysiologic features with other organ-specific autoimmune
diseases for which initiating antigens have been proposed: myelin basic protein for multiple sclerosis,
islet cell proteins for type 1 diabetes mellitus, keratin for uveitis etc. The classic animal model for
immunologically-mediated bone marrow destruction is runt disease, which is produced by infusion of
donor lymphocytes that recognize host cells as foreign, but are themselves not rejected by the host. In
humans, transfusion-associated graft-versus-host-disease causes severe and often fatal bone marrow
destruction by the same mechnism.
The initial antigen recognition event in aplastic anemia that leads to the breached tolerance and
subsequent cascade of immune reactions against the hematopoietic cells is unknown. However the
close association of a few histocompatibility types with aplastic anemia (most commonly HLA-DR2;
other class II antigens such as DR4 and Dpw3; class I loci such as A2, B14, Cw7; HLA-B8 with
posthepatitic aplasia) appears to point to the central role of CD4+ lymphocytes to the process. It is
likely that antigens derived from infection with a virus, exposure to drugs or chemicals, or neoantigens
from a somatic genetic event are processed by the immune system and leads to T cell activation against
the specific offending antigen, and probably more importantly, also against normal cellular antigens

Table 3 : Abnormalities Associated with Fanconi’s Anaemia

Condition Patients affected (%)
Short stature 80
Hyperpigmentation of the skin 75
Malformation of the skeleton
Microsomy - 60 %
Aplasia or hypoplasia of the thumb - 50 % 66
Microcephaly - 40 %
Reduced number of carpal bones - 30 %
Aplasia or hypoplasia of the radii - 17 %
Syndactyly - 15 %
Strabismus (in males) 30
Malformation of kidneys 28
Cryptorchidism 20
Mental retardation 17
Deafness 7

376 CME 2004

which are excessively / aberrantly expressed in the hematopoietic cells. The rarity of occurrence of
aplastic anemia despite common exposures to drugs / chemicals / viruses suggests that genetically
determined features of the immune response can convert this ‘normal’ and usually transient process to
a pathophysiologic and persisting autoimmune response.
Central to these late events in the immunosupression of hematopoiesis are the cytokines interferon-γ,
interleukin-2 and tumor necrosis factor, which are secreted into the marrow microenvironment by the
activated immune system, and act as very potent supressors of hematopoiesis by effects on the mitotic
cell cycle. These also induce the expression of the Fas receptor on CD34+ hematopoietic progenitor
cells, leading to their apoptotic cell death through the triggering of multiple intracellular signal
transduction pathways. Interferon-γ and tumor necrosis factor also increase nitric oxide synthase and
nitric oxide production by marrow cells, adding to the immune-mediated cytotoxicity and elimination
of the hematopoietic cells.

Idiopathic Myelofibrosis (Agnogenic Myeloid Metaplasia)

The primary defect in agnogenic myeloid metaplasia is an abnormal clone at the stem cell level
that gives rise to all blood cells including T and B-lymphocytes. The marrow fibrosis in agnogenic
myeloid metaplasia is not of clonal origin and represents a secondary reaction of the stromal cells,
an exaggeration of the normal marrow fibrous pattern. Bone marrow fibroblasts from patients with
agnogenic myeloid metaplasia have physical characteristics similar to fibroblasts from normal marrows.
The reversibility of myelofibrosis with suppression or extinction of the malignant clone after prolonged
chemotherapy or allogeneic bone marrow transplantation provides further evidence that the fibrotic
process is a consequence of the hematological malignancy. In agnogenic myeloid metaplasia there
is a 10- to 20-fold increase in circulating progenitor cells as opposed to a threefold increase in other
myeloproliferative disorders that are also associated with myelofibrosis.
The bone marrow stroma in agnogenic myeloid metaplasia contains increased amounts of total collagen.
In the early hypercellular phase of the disease, collagen is mostly soluble. With disease progression,
the collagen becomes polymeric and insoluble due to extensive cross-linking. With immunohistologic
techniques, both interstitial and basement membrane collagens types I, III, IV, and V are significantly
increased. In advanced disease there is also an increased deposition of the glycoproteins, fibronectin,
tenascin, and vitronectin. Ultimately, membranous or appositional new bone formation can occur,
leading to osteosclerosis.
The observation that collagen surrounds areas where abnormal megakaryocytes are clustered focused
research on the identification of possible growth factors released from megakaryocytes and platelets.5,6
Platelet derived growth factor, transforming growth factor- β and epidermal growth factor all of which
are contained in the alpha granules of platelets, stimulate bone marrow fibroblast proliferation.
Transforming growth factor-β promotes the formation and accumulation of extracellular matrix in
the bone marrow by enhancing gene expression for collagens type I, III, IV, and flbronectin and also
by decreasing the synthesis of various collagenase-like enzymes that degrade collagen. Furthermore,
transforming growth factor -β controls other growth factors by stimulating platelet derived growth factor
production by bone marrow endothelial stromal cells and by influencing the mitogenic activity of both
platelet derived growth factor and epidermal growth factor. In agnogenic myeloid metaplasia there
are changes in marrow vascularity, with dilation of the marrow sinusoids, endothelial hyperplasia, and
neovascularisation. This effect may be also mediated by transforming growth factor-β, which stimulates
angiogenesis.
The mechanism of extramedullary hematopoiesis in spleen, liver, and sometimes lymph tissues is
unknown. However, the marrow fibrosis is associated with displacement of hematopoietic progenitor
cells into the blood. Circulating stem cells likely then seed other tissues of the reticuloendothelial system
Bone Marrow Failure Syndromes: An Overview 377
and establish extramedullary sites of hematopoiesis. In turn, the same process causing fibrosis in the
marrow leads to fibrosis in the spleen. The spleen eventually can become massively and irreversibly
enlarged.
Anemia in agnogenic myeloid metaplasia is multifactorial. Reduced red blood cell production is
part of the myeloproliferative disorders. This is complicated by dyserythropoiesis and ineffective
extramedullary erythropoiesis. There is also a hemolytic component to the anemia due to a
reduced life span of the dyserythropoietic red blood cells. Rarely, a direct antiglobulin test
positive autoimmune hemolytic anemia occurs. Splenic pooling of blood aggravates the anemia
and sequestration of platelets contributes to the thrombocytopenia. Patients have an increased
tendency to bleed because of thrombasthenia and thrombocytopenia and from esophageal varies
and peptic ulceration.

Paroxysmal Nocturnal Hemoglobinuria

A major clue to the etiology of this disease is provided by the recent finding that patients have a
somatic mutation for a protein (glycosylphosphatidylinositol class A) that controls the formation of the
phosphatidylinositol anchor of several membrane proteins, including complement control proteins.
At least 19 proteins are attached to blood cells by a glycosylphosphatidylinositol anchor. Amongst this
glycosylphosphatidylinositol class of proteins are those termed class A, which lack the ability to transfer
glucosamine to phosphatidylinositol. The gene responsible for the defect in class A cells and paroxysmal
nocturnal hemoglobinuria is phosphatidylinositol glycan A (PIG-A). The PIG-A gene is located on the
short arm of the X chromosome and codes for a 54-kd protein. The protein is presumed to be an N
- acetylglucosamine transferase, and over 100 somatic mutations spread over the entire coding region
have been identified. The mutations, mostly deletions or insertions, generally cause a stop codon that
results in premature termination of the peptide to yield a truncated protein. These truncated proteins
may be either non-functional, partially functional, or unstable. The ultimate result of a non-functional
PIG-A gene product is a PNH III cell, and a partially functional gene product results in a PNH II cell.
Patients with paroxysmal nocturnal hemoglobinuria have an unusual sensitivity of their erythrocytes to
the lytic action of complement, due to the lack of two membrane-anchored proteins, decay accelerating
factor (DAF / CD 55) and a membrane inhibitor of reactive lysis (MIRL / CD 59).7,8 Activation of
complement by either the classic or alternative pathway results in the deposition of larger numbers of
C3 molecules on the PNH blood cell surface than on normal cells. The result is greater activation of
the terminal complement components C5 to C9, which causes more cell lysis than occurs with normal
cells. Furthermore, type II PNH cells are more effectively damaged by the C5b-C9 complex generated
on the erythrocyte surface because the C5b-C9 lytic complex penetrates PNH cell membranes more
efficiently than normal cell membranes.
The granulocytes and platelets of paroxysmal nocturnal hemoglobinuria patients also lack the DAF
and MIRL proteins. However, excessive activation of complement on the platelet surface does not lead
to platelet lysis, but indirectly stimulates platelet aggregation and hypercoagulability. This probably
accounts for the thrombotic tendency in these patients.
Many patients with paroxysmal nocturnal hemoglobinuria have several populations of abnormal
erythrocytes. The complement lysis sensitivity test, which examines the susceptibility of antibody-
sensitized erythrocytes to complement-mediated lysis, can be used to define the various cell
populations. PNH type II cells have a moderate increase in susceptibility to complement attack. These
erythrocytes appear to have markedly decreased levels of the complement control protein decay-
accelerating factor, but they do not have the membrane deficit that leads to sensitivity to attack by the
C5b-C9 complex. PNH type III cells are highly susceptible to complement attack. They appear to lack
phosphatidylinositol-linked control proteins completely.
378 CME 2004
Although pancytopenia mimicking acquired aplastic anemia is seen clinically in about two-thirds of
all patients with paroxysmal nocturnal hemoglobinuria, its exact relationship to the genetic defect of
paroxysmal nocturnal hemoglobinuria is not well understood. Today, there is an emerging consensus
that the aplastic anemia / subtle bone marrow failure seen in patients with paroxysmal nocturnal
hemoglobinuria is the primary disease process, and that the pathophysiology of this aplastic process
in some way confers a relative survival advantage on the PNH stem cells. In this scenario, a PIG-A
mutation would render GPI-AP deficient cells resistant to the cytotoxic autoimmune attack, enabling
them to emerge and undergo a clonal expansion. Even though the ‘survival advantage’ hypothesis
may explain all the aspects of this intriguing disease, a formal proof of this theory is still lacking.

Fanconi’s Anemia
Fanconi’s anemia is the best-defined inherited bone marrow failure disorder. The first step in the
understanding of the disease was the recognition that the patients’ cells and chromosomes are sensitive
to certain chromosome-damaging (clastogenic) agents such as diepoxybutane and mitomycin C.
This disease defining property was successfully used to segregate Fanconi’s anemia into several
heterogenous groups and then to identify the genes involved.
It is now known that Fanconi’s anemia is an autosomal recessive disorder caused by defects in at least
eight distinct genes: FANCA, B, C, D1, D2, E, F and G. To date, six of these genes (for A, C, D2, E,
F, and G) have been cloned. The principal cellular consequence of these different genetic defects is
hypersensitivity to DNA damage, particularly to interstrand DNA crosslinks.9,10 The FA proteins appear
to constitute a multi-protein pathway and / or a multimer complex whose precise biochemical functions
remain unknown. Five of the FA proteins (FANCA, C, E, F and G) are known to interact in a nuclear
complex. FANCD2 is a downstream component of the FA pathway. This component is ubiquitinated
in response to DNA damage and translocates to nuclear foci containing BRCA1. As BRCA1 (a breast
cancer susceptibility gene) plays an important role in DNA repair, it is inferred that the ubiquitinated
FANCD2 also has a role in DNA repair functions. Although FANCB and FANCD1 have not yet been
cloned, it is likely that FANCB is part of the nuclear complex mentioned above; and FANCD1 acts
downstream of FANCD2. Together, these data suggest that the FA pathway functions primarily as a
DNA damage response system, although its exact role (direct involvement in DNA repair versus an
indirect, facilitating role) has not yet been defined.
The FA pathway is likely to play a critical role as a caretaker of genomic integrity in hematopoietic stem
cells. Clarifying the molecular basis of this disease may provide new insights into the pathogenesis of
bone marrow failure syndromes and myeloid malignancies.

Laboratory Diagnosis
Laboratory diagnosis of the bone marrow failure syndromes includes general morphology, bone
marrow aspirates and biopsies. Clinical suspicion of particular diagnoses can often be substantiated
by laboratory tests of varying specificity. Specific diagnoses require unique tests that are only available
for a few of the diagnoses.11 Some of them have already been referred to above. The most useful
is chromosome breakage in the diagnosis of Fanconi’s anemia. In this disorder, gene mutation
analysis or mapping will become the gold standard when all of the involved genes have been cloned.
Diamond-Blackfan anemia usually has elevated red-cell ADA, and the genetic abnormality may map
to chromosome 19 (19q13). Decreased serum trypsinogen or other evidence of exocrine pancreatic
insufficiency may provide laboratory proof of Scwachman-Diamond syndrome. Cartilage-hair
hypoplasia syndrome is substantiated when absent central pigment in hair is found and when it is
mapped to chromosome 9 (9p21-p13). The diagnosis of dyskeratosis congenita remains clinical at
this time, although linkage to the X chromosome (Xq28) and skewed maternal X inactivation may

Bone Marrow Failure Syndromes: An Overview 379

be helpful in some families. Kostmann’s syndrome is diagnosed in patients who have congenital
nonimmune severe neutropenia.

Management Principles12
The management of bone marrow failure includes supportive care of the pancytopenic patient,
and more definitive therapies that can reverse the underlying marrow failure. Foremost amongst
these definitive measures are immunosuppressive therapies for acquired aplastic anemia, and bone
marrow transplantation. However, these definitive measures are not considered here owing to space
constraints.

Infections
Neutropenia increases the individual patient’s susceptibility to serious bacterial and fungal infections.
The degree and duration of neutropenia correlates with the number of infectious episodes. In leukemic
children, 10% of days at granulocyte levels above 1500/µL were associated with proven infection, but
this number rose to 20% with granulocyte counts of 500 to 1000/µL; to 35% with 100 to 500/µL, and
to 50% with neutrophil counts of less than 100/µL. Although this subset of patients are not directly
comparable with the patients of bone marrow failure (leukemic patients suffer gross disruptions of
the integumentary barriers to infection, their neutropenia is part of a complex of problems including
immunosupression, but their period of neutropenia induced by the cytotoxic chemotherapy is relatively
shorter!), the longer duration of neutropenia in patients with bone marrow failure makes them even
more prone than leukemic patients to life-threatening infections.
Some infections in neutropenic patients can be prevented. Most important are simple measures
directed at increasing general hygiene. A hospital ward’s physical surroundings should be well
maintained to reduce nosocomial infection. Dental hygiene should be perfected and dental sources of
infection removed with antibiotic and platelet prophylaxis. Nystatin oral rinses can prevent candidiasis.
“Routine” rectal examinations are more likely to be harmful than helpful. Taking of capillary blood
samples from fingertips and ear lobes must be avoided. Early attention to the signs of infection,
especially after discharge from hospital, can avert the catastrophic complications of initially minor
infectious episodes.
Total protection environments, often in combination with nonabsorbable antibiotics or selective gut
decontamination to preserve some anaerobic bacteria, have been advocated, with a view to preventing
infections. However, these are not obviously superior to simpler methods of isolation and compulsive
hand washing. Sterile diets, avoidance of fresh fruits and vegetables, no flowers in the room and similar
prescriptions are of unproven practical value.
Initiation of prompt and aggressive empiric antibiotic therapy is the key to managing neutropenic sepsis
successfully. The cardinal rule is: if the absolute neutrophil count is less than 500/µL and infection is
suspected, immediately begin broad-spectrum, parenteral antibacterial therapy. Any such regimen will
require modification based on results of cultures, new symptoms or signs, or a deteriorating clinical
course. In general, bacteremia is present in only 20% of febrile neutropenic episodes, and in only
about 40% can a microbiologic cause or localizing physical findings be identified. In patients who
remain febrile despite adequate antibacterial therapy for more than 7 days, antifungal therapy should
be instituted. Early aggressive treatment of neutropenic patients can reverse fungal disease.

Red Cell Transfusions

Correction of anemia in patients with bone marrow failure inevitably leads to considerable improvement
in the quality of life, and is an important aspect of their management. With acclimation, fit patients are
usually not symptomatic at hemoglobin concentrations of more than 7 g/dL. Thus, the maintenance of
hemoglobin around 8 g/dL is a reasonable therapeutic goal. Patients with underlying cardiovascular
380 CME 2004
disease should probably be maintained at a higher level of hemoglobin, around 9 g/dL. Iron chelation
is indicated for transfusion-dependant patients with chronic anemia who have a reasonable expectation
of survival. Perhaps in contrast to alloimmunisation by platelet transfusions, there is a relatively low
frequency (about 10%) of alloimmunisation due to packed red blood cell transfusions.

Platelet Transfusions
Platelet transfusions are indicated in the treatment of bleeds caused by thrombocytopenia, usually
when less than 20,000 / dL. Platelet concentrates are readily available and safe to administer. Other
than cost and convenience, the major problem related to platelet transfusions is the development of
a refractory state in the recipient caused by alloimmunisation. The life span of the transfused platelets
in the circulation is dramatically shortened by host antibodies directed against HLA class I antigens.
Alloimmunisation is suggested by poor platelet count recovery at the 1-hour post-transfusion platelet
count, and is confirmed by finding specific HLA antibodies in serum. Refractoriness can often be
overcome by selection of HLA-matched donors. Nevertheless, 5% to 40% of perfectly HLA-matched
transfusions fail. Alloimmunization can be prevented by the use of single-donor platelets rather than
pooled platelets. Physical methods of leukocyte depletion by filtration, gamma radiation or ultraviolet
treatment is equally efficacious.
Correction of the platelet count by transfusion almost always alleviates the minor mucocutaneous
bleeding common in thrombocytopenic patients. Major bleeding from the gastrointestinal and
genitourinary tracts is usually not due to thrombocytopenia alone, and other causes for massive
hemorrhage should always be sought. The bleeding time improves after erythrocyte transfusion in
patients with anemia. Therefore, the treatment of serious hemorrhage should include correction of
severe anemia as red blood cell transfusions may lessen bleeding symptoms.
The role of prophylactic platelet transfusion is controversial. The primary indication for platelet
prophylaxis is to avoid intracranial hemorrhage, but the risk of this complication in the chronically
thrombocytopenic patient is low. Prophylactic platelet transfusions have not been shown to alter
patient survival.

Granulocyte Transfusions
Harvesting of granulocytes from the peripheral blood became feasible in the early 1970s, but controlled
trials failed to demonstrate significant improvement in the survival of patients who received such
transfusions. In addition, this type of transfusion is expensive and associated with serious toxicities:
severe febrile reactions, pulmonary capillary leak syndrome, increased risk of cytomegalovirus
transmission, and alloimmunization. However, a meta-analysis of the published studies suggested that
granulocyte transfusions were beneficial in patients with sepsis unresponsive to antibiotics if: adequate
cell numbers were administered (2 to 3 X 1010 cells daily) to compatible recipients who were unlikely
to have imminent improvement in marrow function. In this context, cytokines offer a new strategy for
increasing the efficiency of donor harvests. Administration of G-CSF to normal donors can greatly
increase the yield of leukapheresis without adverse effects on the volunteers. Such large numbers of
neutrophils can be obtained as to allow dramatic increases in the absolute blood levels of granulocytes
in neutropenic recipients.

Bone Marrow Transplantation

Allogeneic bone marrow transplantation from a histocompatible matched sibling is curative therapy
in the majority of aplastic patients who undergo the procedure. Recently, cytokine-primed peripheral
blood has also served as a source of hematopoietic stem cells. and the convenience of this method is
likely to lead to it being the major method of stem cell collection. The results of bone marrow transplants
should broadly hold good for peripheral blood stem cell transplants also.
Bone Marrow Failure Syndromes: An Overview 381
In a similar manner, the hematological consequences of Fanconi’s anemia can be effectively
treated by complete replacement of patient stem cells by those from a histocompatible donor.
Unfortunately, allogeneic stem cell transplantation is currently limited to patients with an
unaffected matched sibling donor. Transplantation from alternative donors, while successful in
selected cases, is associated with a high risk of graft failure and must be carefully considered in
terms of risk and benefit for each individual. For those patients with Fanconi’s anemia lacking an
appropriate donor, new therapies need to be devised. In this context, gene therapy strategies hold
promise.

Prognosis
The great majority of aplastic anemia patients who are treated with allogenic bone marrow transplant
or immunosupression are cured / have significant amelioration. In contrast, only 20% of patients with
severe aplastic anemia who are treated with transfusions alone survive beyond one year.
The median survival of patients with myelodysplasia varies from many years (sideroblastic anemia,
5q- syndrome), to a few months (refractory anemia with excess blasts, monsomy 7 syndrome). The
majority dies of the complications of pancytopenia, and not due o leukemic transformation. Therapy-
related myelodysplasia patients usually progress within a few months to acute myeloid leukemia, which
is unresponsive to therapy.
Idiopathic myelofibrosis patients have a median survival of about 5 years, with around 20% surviving
beyond a decade. Cytogenetic abnormalities and marked cytopenias are adverse prognostic features.
Leukemic transformation occurs in about 10% of patients.
Pure red cell aplasia is compatible with long life, and about 50% of those with acquired red cell aplasia
can be cured with immunosupression. Patients of congenital anemias who are treated with transfusions
and iron chelation often survive for decades.

References
1. Young NS, (ed). Bone Marrow Failure Syndromes. Philadelphia, WB Saunders. 2000.
2. Gordon-Smith EC. Aplastic anemia and other causes of bone marrow failure. In Weatherall DJ, Ledingham JGG,
Warrell DA, (eds). Oxford textbook of medicine, 3rd ed. Oxford: Oxford university press. 1996;3441-49.
3. Jacobson R, Salo A, Fialkow P. Agnogenic myeloid metaplasia: a clonal proliferation of hematopoietic stem cells
with a secondary myelofibrosis. Blood 1978;51:189.
4. Hillmen P, Lewis SM, Bessler M. Natural history of paroxysmal nocturnal hemoglobinuria. N Engl J Med
1995;333:1253-58.
5. Groopman J. The pathogenesis of myelofibrosis in myeloproliferative disorders. Ann Intern Med 1980;92:857.
6. Kimura A, Katoh O, Kuramoto A. Effect of platelet derived growth factor, epidermal growth factor and transforming
growth factor B on the growth of human marrow fibroblasts. Br J Haematol 1988;69:9.
7. Rosti V. The molecular basis of paroxysmal nocturnal hemoglobinuria. Haematologica 2000;85:82-7.
8. Boccuni P, Del Vecchio L, Di Noto R, Rotoli B. Glycosylphosphatidylinositol (GPI)-anchored molecules and the
pathogenesis of paroxysmal nocturnal hemoglobinuria. Crit Rev Oncol Hematol 2000;33:25-43
9. Grompe M, D’Andrea A. Fanconi’s anemia and DNA repair. Hum Mol Genet 2001;10:2253-9.
10. Yamashita T, Nakahata T. Current knowledge on the pathophysiology of Fanconi anemia: from genes to
phenotypes. Int J Hematol 2001;74:33-41.
11. Alter BP. Bone marrow failure syndromes. Clin Lab Med 1999; 1:113-33
12. Young NS. Supportive treatment of aplastic anemia. In Young NS, Alter BP, (eds). Aplastic anemia, Acquired and
Inherited. Philadelphia: WB Saunders, 1994; pp 201-15.

382 CME 2004