The myelodysplastic syndromes (MDS) are a heterogeneous group of bone marrow failure syndromes collectively characterized by ineffective haematopoiesis resulting in peripheral blood cytopenias, most commonly anaemia. MDS have a tendency to evolve over time, and there is a risk of progression to acute myeloid leukaemia (AML), which is arbitrarily defined by the World Health Organization (WHO) as the presence of 20% or more immature myeloid “blast” cells in the blood or bone marrow.
Although some investigators feel that MDS should not be considered a form of cancer because some cases remain stable for many years, and a few patients have a pathophysiology dominated by autoimmune suppression of hematopoietic progenitor cells and resembling aplastic anaemia, MDS are classified as myeloid neoplasms by the WHO and other organizations. Many of the somatic gene mutations that are recurrent in MDS are also present in AML and other myeloid neoplasms, and there are typically one or just a few mutant clones in the bone marrow contributing to most of the marrow cellularity.
The minimal diagnostic criteria for MDS as defined by the WHO include the presence of a meaningful and persistent cytopenia (i.e., haemoglobin <11 g/dL, absolute neutrophil count <1500/mm, or platelet count <100,000/mm 3 ) together with at least one of the following features: (1) greater than 5% blasts in the bone marrow, (2) an abnormal marrow karyotype with an acquired chromosome anomaly that is consistent with the diagnosis of MDS, (3) other evidence of clonal haematopoiesis besides marrow karyotype (e.g., abnormal result using a fluorescence in situ hybridization [FISH] panel of probes directed at common MDS-associated chromosomal abnormalities), or (4) greater than 10% cells in any given hematopoietic linage that appear dysplastic (i.e., morphologically abnormal, described further below) in a patient in whom other causes of dysplastic cell morphology such as nutritional deficiency have been ruled out. Patients sometimes have persistent cytopenias and do not quite meet these diagnostic criteria, yet no other diagnosis is apparent after detailed investigation. These disorders are sometimes termed idiopathic cytopenias of undetermined significance (ICUS), and such patients should be followed over time because they have a risk of developing overt MDS.
The incidence and prevalence of MDS have been difficult to estimate accurately, in part because of varying definitions and imprecise terminology historically, and also because until recently, MDS cases were not captured by most global cancer registries. The U.S. National Cancer Institute’s Survey, Epidemiology, and End Results (SEER) data indicate that 10,000 to 12,000 new cases are diagnosed in the United States each year. However, analysis of insurance claims data, including Medicare claims, suggests that the actual incidence of MDS is much higher—at least 40,000 new U.S. cases per year, several times higher than the incidence of AML. 3 In addition, many geriatric patients who have cytopenias of uncertain aetiology and may have MDS are incompletely evaluated (e.g., they do not undergo bone marrow aspiration) and never receive an MDS diagnosis.
The primary risk factor for development of MDS is aging, with disease a consequence of cumulative acquisition of somatic mutations in marrow stem cells. The rate at which MDS is diagnosed is slightly increased in workers in certain industries such as the petroleum industry and agricultural sector, probably owing to exposure to genotoxic hydrocarbons, accelerating acquisition of such mutations. There is a slight male predominance overall, perhaps representing patterns of occupational exposure. In Asia and Eastern Europe, MDS is typically diagnosed at younger ages than in the West; in the United States, the median age at which MDS is diagnosed is 71 years, whereas in China it is closer to 50 years.
Individuals who have been exposed to ionizing radiation or certain types of cytotoxic chemotherapy (e.g., alkylating agents such as chlorambucil or mephalan, or topoisomerase II inhibitors such as etoposide or doxorubicin) have an increased risk of developing MDS subsequently. The peak incidence of MDS diagnosis after alkylating agent or radiation exposure is 5 to 10 years later, whereas after a topoisomerase inhibitor exposure, it is 1 to 3 years. The risk never goes away entirely; even 50 years after the Hiroshima and Nagasaki atomic bomb events in Japan, radiation-exposed individuals continued to be diagnosed with MDS at a higher rate than the unexposed population. Patients with a chemotherapy or radiation exposure history are said to have therapy-related MDS (t-MDS) or “secondary” MDS.
Although familial MDS is rare, germline gene mutations in several genes, including RUNX1 and GATA2, predispose to MDS. RUNX1 mutations are associated with a prodrome of thrombocytopenia that can be mistaken for immune thrombocytopenic purpura, whereas GATA2 mutations are associated with a history of mycobacterial infections and monocytopenia.
MDS are clonal disorders, meaning that one or several stem or progenitor cells in the marrow that bear somatic gene mutations expand and proliferate at the expense of healthy stem cells and come to dominate the bone marrow. These clones can then acquire additional mutations that give rise to subclones that contribute to progression of disease. The process of somatic mutation acquisition can be accelerated through exposure to radiation or DNA-damaging chemicals, as noted earlier.
Marrow cellularity is usually normal or increased in MDS, but excessive apoptosis (i.e., programmed cell death) of hematopoietic cells within the bone marrow accounts for the peripheral blood cytopenias. Excessive apoptosis within the marrow characterizes earlier MDS, but as the disease progresses toward AML and one clone begins to dominate, intramedullary apoptosis may decrease. In addition, hematopoietic differentiation, which is disordered but still can proceed in earlier stages of the disease, may become completely impaired late in the disease, such that most developing hematopoietic cells are arrested at the myeloblast stage. If 20% or more of the cells in the marrow or blood are blast cells, this is considered AML; MDS by definition is associated with less than 20% blast cells.
Large gains and losses of chromosomal material within diseased cells, including monosomies, trisomies, and large chromosomal deletions, are detectable by routine metaphase karyotyping present in one half of patients with de novo MDS and more than 80% of patients with t-MDS. Newer genetic techniques such as array-based comparative genomic hybridization may uncover cryptic chromosomal deletions in patients with MDS who have normal karyotype results.
More than 25 genes are now known to be recurrently mutated in MDS. No one mutation dominates; MDS are associated with considerable genetic heterogeneity. These genes and the probable function of the proteins they encode are listed below. More than 90% of patients have at least one detectable clonal somatic mutation, and most patients with MDS have several such mutations detectable in their marrow cells. Mutations affecting epigenetic patterning (e.g., DNA methylation) and chromatin conformation (e.g., histone-DNA interactions) are common in MDS, and this may be why agents that affect DNA methylation, such as azacytidine and decitabine, are effective in some patients. In addition, mutations affecting pre-mRNA splicing are common, especially in patients with MDS subtype refractory anaemia with ring sideroblasts, but the specific genes that are misspliced and give rise to an MDS phenotype are unclear.
GENE MUTATIONS ASSOCIATED WITH MDS AND FUNCTIONAL CLASS OF THE ASSOCIATED PROTEIN
|RECURRENTLY MUTATED GENES||MUTATION FREQUENCY IN MDS||ADDITIONAL NOTES|
|SPLICEOSOME COMPONENTS (ALTER PRE-mRNA SPLICING)|
|SF3B1||20-25%||Strongly associated with presence of ring sideroblasts (60-80% of RARS cases), often co-occurs with DNMT3A * mutations|
|U2AF1 *||5-10%||Often co-occurs with del(20q)|
|SRSF2||5-10%||More frequent in CMML (25-30%), often co-occurs with RUNX1 * or ASXL1 * mutations|
|U2AF2, SFRA1, PRPF40B, SF1||<2% each|
|EPIGENETIC PATTERN & CHROMATIN CONFORMATION MODIFIERS|
|TET2||20%||More frequent in CMML (≈40%)|
|DNMT3A *||10-15%||Associated with poor outcome after transplant|
|SETBP1||5-10%||More common in CMML; germline mutations cause Schinzel–Giedion syndrome|
|ASXL1 *||10-20%||More frequent in CMML (≈40%)|
|EZH2 *||6%||More frequent in CMML (≈12%)|
|KDM6A||<2%||Rare in MDS, more frequent in CMML|
|IDH1, IDH2||<2%||More frequent in AML than MDS|
|ATRX||Rare||Associated with acquired α-thalassemia (haemoglobin H inclusions in RBCs)|
|RUNX1 *||10-15%||Germline mutations cause familial platelet disorder with propensity to AML (FPD-AML)|
|ETV6 *||<5%||Translocations common in AML; formerly known as TEL|
|GATA2||Rare||Germline mutations cause familial MDS with monocytopenia|
|CEBPA||Rare||More frequently mutated in AML, germline mutations cause familial AML without preceding dysplasia|
|WT1||Rare||More frequently mutated in AML|
|TP53 *||5-10%||Associated with very poor prognosis; associated with t-MDS and with complex karyotype|
|TYROSINE KINASE SIGNALING||As a group, these genes are more often mutated in myeloid malignancies with a proliferative component|
|NRAS * , KRAS, BRAF||5-10% collectively||NRAS mutations more common in CMML and AML; first mutations described in MDS (1987)|
|JAK2||<5%||More frequent in RARS with thrombocytosis (RARS-T) (50%)|
|CBL, CBLB||<5%||More frequently mutated in CMML than MDS|
|FLT3, MPL, KIT||<1%||MPL mutated in (5%) of RARS-T, FLT3 and KIT mutations are rare in MDS and much more common in AML|
|NF1||<1%||Germline mutations cause neurofibromatosis type 1|
|PTPN11||<1%||Rare in MDS, more frequent in JMML (30%)|
|NPM1||<2%||Much more frequently mutated in AML|
|Cohesins (RAD21, STAG1, STAG2, SMC3, SMC1A)||Rare|
* Those mutations that appear to be independently associated with poor outcomes in MDS are denoted by an asterisk. AML = acute myelogenous leukaemia; CMMS = chronic myelomonocytic leukaemia; JMML = juvenile myelomonocytic leukaemia; RARS = refractory anaemia with ring sideroblasts; t-MDS = therapy-related MDS.
A subset of patients with pathologic findings that resemble MDS will have autoimmune suppression of haematopoiesis, which may be driven by a clonal T-lymphocyte population. Patients are more likely to have an immune component to their marrow failure if they are younger than 55 years and have a hypocellular marrow similar to that seen in aplastic anaemia, a normal chromosome pattern or trisomy 8 rather than a more complex karyotype, a paroxysmal nocturnal haemoglobinuria cell population detectable by flow cytometry, and HLA-DR15 tissue type. Such patients may respond to immunosuppressive therapy with anti–T-cell therapies.
Most patients with MDS come to medical attention because they have anaemia (which is present in 95% cases and is often macrocytic); one half of patients also have neutropenia, thrombocytopenia, or both at the time of diagnosis. The peripheral blood and bone marrow exhibit characteristic “dysplastic” cell morphologic abnormalities. In the peripheral blood, these include hypogranular neutrophils, hypolobated neutrophils such as the two-lobed pseudo-Pelger-Hüet cell, platelet size and granularity abnormalities, poorly haemoglobinised red cells, and circulating early myeloid cells such as small numbers of myeloblasts. Maturation abnormalities in the bone marrow include megaloblastoid erythroid maturation, multi-nucleated erythroid cells, ring sideroblasts (i.e., erythroid precursors with abnormal peri-nuclear iron-containing mitochondria visible with Perls’ Prussian blue reaction), nuclear karyorrhexis, myeloid lineage abnormalities such as left-shifted myelopoiesis and the presence of hypogranular white cells, and megakaryocytic abnormalities such as micromegakaryocytes, very large hypernucleated megakaryocytes, or hypolobated megakaryocytes. A mild increase in reticulin fibrosis is common in MDS, but severe reticulin fibrosis or collagen fibrosis are rare. Extramedullary haematopoiesis is uncommon in MDS, and the presence of splenomegaly or hepatomegaly should prompt consideration of another cause.
The natural history of MDS is highly variable. Approximately half of patients will die of complications of the cytopenias, particularly infection due to both neutropenia and to functional neutrophil defects (i.e., impaired bactericidal activity due to hypogranularity). Among lethal infections in MDS, pneumonia and bacteraemia due to gastrointestinal or urinary infections are the most common. Bleeding is the second leading common cause of death. In addition, anaemia may exacerbate cardiovascular or cerebrovascular disease. Approximately 25% of patients will progress to AML, which is usually a fatal complication. Finally, because MDS is often diagnosed in elderly patients or in those with a history of cancer, unrelated causes are the chief contributor to death in at least one third of patients.
The clinical course in MDS is dominated by complications of the cytopenia, including mucosal bleeding, recurrent infections, and exertional dyspnoea and fatigue. Fatigue correlates poorly with the degree of anaemia and may be due in part to cytokine release by the abnormal clonal cells. Paraneoplastic manifestations including neutrophilic dermatosis, inflammatory arthritis, and other rheumatologic syndromes occur in 10 to 15% of patients. Repeated transfusions of red cells can lead to transfusional hemosiderosis and organ dysfunction, including hepatic and cardiac abnormalities. The frequency with which actual clinical complications of iron overload occur, as well as the importance of chelation therapy in MDS, are controversial topics.
The diagnosis of MDS is usually suspected when the patient is discovered to have cytopenias, especially macrocytic anaemia in an older person that is not due to B 12 or folate deficiency, medication effect (e.g., methotrexate or azathioprine), or alcohol abuse. The blood count abnormality may be an incidental finding, but most patients with MDS are symptomatic with fatigue, dyspnoea, and other cytopenia-associated symptoms.
A bone marrow aspirate is required to make the diagnosis, and bone marrow core (trephine) biopsy is also useful and can provide information about the marrow architecture and overall marrow cellularity. 9 Metaphase cytogenetics is essential, but FISH is only necessary if karyotyping fails because of an unsuccessful aspiration. FISH abnormalities are rarely detected in patients for whom at least 20 metaphases can be counted during karyotyping.
Not all that is dysplastic is MDS, and it is important to consider other potential causes for cytopenias and morphologic abnormalities. The differential diagnosis includes nonclonal disorders such as B 12 or folate deficiency, copper deficiency, iron deficiency, HIV infection, medication exposure (especially cytotoxic chemotherapy agents), or an immune disorder such as T-cell large granular lymphocyte (T-LGL) leukaemia. In patients with isolated sideroblastic anaemia, it is important to rule out congenital sideroblastic anaemias, including those due to germline mutations in ALAS2 , which can present late in life.
In addition, MDS must be separated from aplastic anaemia, which can be oligoclonal but is not typically associated with Pelger-Hüet cells or other morphologic abnormalities, and the chromosome karyotype in aplastic anaemia is usually normal. MDS can also be mistaken for AML or a myeloproliferative neoplasm (MPN) such as primary myelofibrosis. Cases in which MDS features, such as cytopenias, overlap with MPN features, such as leukocytosis and splenomegaly, include chronic myelomonocytic leukaemia (CMML). The WHO categorizes MDS/MPN overlap syndromes separately from MDS, and these cases have a unique mutational spectrum.
Several different MDS subtypes of varying risk of progression to leukaemia are recognized by the WHO. If a patient has cytopenias but does not have characteristic MDS-associated morphologic abnormalities, yet has a typical MDS-associated marrow cytogenetic abnormality such as loss of chromosome 7 or deletion of the long arm of chromosome 5, the patient can be considered to have “unclassifiable MDS” (MDS-U) and does have a risk of progression.
Given the heterogeneity of MDS and the wide spectrum of functional status and comorbidities present in the typically older population with MDS, treatment must be individualized. Currently, lack of a specific well-characterized molecular target in most patients with MDS precludes a true biologically based approach to therapy. Algorithms for treatment are based primarily on the patient’s prognosis, as discussed further below.
When choosing a potential therapy for a patient with MDS, clinicians should consider the patient’s age, comorbidity and functional status, disease biology (currently based on assessment of bone marrow and peripheral blood findings plus cytogenetic analysis), and pace of evolution of the disease. Some patients with MDS have mild cytopenias and minimal or no symptoms, and they can reasonably be observed without therapy. For higher-risk patients (e.g., International Prognosis Scoring System [IPSS] intermediate 2 or high risk), immediate therapy is usually indicated because patients are usually severely cytopenic and have a life expectancy of less than 2 years. For lower-risk patients (IPSS low and intermediate 1 risk) a more cautious approach is appropriate. The only curative modality for MDS is allogeneic stem cell transplantation, so this option should be considered in developing a therapeutic plan if the patient is young enough and lacks other major medical problems.
Hematopoietic Growth Factors and Other Approaches to Cytopenias
Because most patients with MDS are anaemic, erythropoiesis-stimulating agents (ESA) including epoetin and darbepoetin are often used, even though these are not specifically U.S. Food and Drug Administration (FDA) approved for treatment of MDS. The benefits of ESAs are generally modest and of limited duration, and no prospective data confirm that they prolong survival. Patients with lower baseline serum erythropoietin levels (<500 U/L, and especially <100 U/L) are more likely to respond to ESAs than patients with higher baseline serum erythropoietin levels. An inappropriately low erythropoietin response to a given degree of anaemia is more likely in older patients because of the higher frequency of renal insufficiency. If no response is observed after 2 to 3 months of ESA therapy, the trial should be concluded. For patients with refractory anaemia with ring sideroblasts (RARS), the addition of granulocyte colony-stimulating factor (G-CSF) to the ESA may potentiate the erythropoietic response. A few patients will benefit from treatment with androgens, but the risk of prostate enlargement in men and other complications such as hepatotoxicity must be considered.
Although myeloid growth factors such as G-CSF can ameliorate the neutropenia often seen in MDS patients, such an approach (probably not leukaemogenic, as was once feared) is not likely to reduce the risk of infections or provide other clinical benefit, probably because of the neutrophil dysfunction that may occur in addition to neutropenia. There is no clear role for prophylactic antibiotics.
The thrombopoiesis-stimulating agents romiplostim and eltrombopag have been approved to treat immune thrombocytopenic purpura. These agents have undergone limited study in MDS and are not FDA approved in MDS. They probably should not be used in higher-risk patients with excess blasts, because they could potentially promote leukaemogenesis, but a therapeutic trial could be justified in a patient with lower-risk disease with severe thrombocytopenia or bleeding. Platelet transfusions should be used judiciously in MDS patients owing to the risk of alloimmunization, with prophylactic transfusion appropriate only in those with platelet counts less than 5000 to 10,000/mm3. Patients with mucosal bleeding and refractory thrombocytopenia may benefit from treatment with antifibrinolytic agents such as aminocaproic acid.
Iron Chelation Therapy
The multiple red blood cell (RBC) transfusions needed by some patients with MDS may cause iron overload and tissue injury, but the frequency with which this complication occurs with clinical significance is unclear. Patients with higher serum ferritin levels fare less well than patients with lower values, both in the transplant and nontransplant settings. However, whether the ferritin is simply a marker of inflammation or more advanced disease, or whether the iron deposition in marrow, liver, heart, and pancreas directly results in this adverse outcome, remains uncertain. That uncertainty coupled with the rarity of reported deaths in MDS secondary to iron overload, as well as the toxicity of available chelators, makes the role of iron chelation therapy with agents such as deferasirox or deferoxamine in this disease quite controversial. Although it is reasonable to recommend a trial of iron chelation therapy in patients with lower-risk disease and high transfusional burdens (e.g., >20 to 40 RBC units transfused), there are no prospective data yet to suggest a clear-cut benefit in this setting. Retrospective data suggest that chelated patients live longer than nonchelated patients, but these studies are confounded by patient selection factors.
As described earlier, some patients with MDS display a pathophysiology that seems similar to aplastic anaemia, including T-lymphocyte–driven autoimmune attack against hematopoietic cells. Such patients may respond to anti–T-cell therapies, including antithymocyte globulin and calcineurin inhibitors such as cyclosporine or tacrolimus. However, although younger patients with normal karyotype, trisomy 8, paroxysmal nocturnal haemoglobinuria clones, or HLA-DR15 status seem to have a higher likelihood of benefitting, there are currently no good predictive strategies for selecting patients for immunosuppressive therapy.
Early reports showing improvements in anaemia in some MDS patients during thalidomide therapy prompted a trial of the immunomodulatory lenalidomide in patients with lower-risk disease. An initial trial and a large phase II trial that followed clearly showed that patients with loss of chromosome 5q (5q−), either alone or in conjunction with other chromosomal abnormalities, had high response rates to lenalidomide, including a nearly 70% transfusion independence rate and better than 30% cytogenetic normalization rate, lasting a median of more than 2 years. Therefore, in the United States, lenalidomide has become the treatment of choice for such 5q− lower-risk MDS patients. Lenalidomide has also been used in higher-risk MDS and AML with 5q− chromosomal abnormalities, but it is less effective. A large randomized trial of two doses of lenalidomide versus placebo shows that AML progression was not increased in lower-risk 5q− MDS with lenalidomide therapy and that responses to a 10-mg dose were more frequent than to a 5-mg dose. In patients with lower-risk MDS without a chromosome 5 abnormality, the response rate to lenalidomide is approximately 25%, which is about the same as one might expect with hypomethylating agents; responses last a median of 8 to 9 months. Important adverse effects of lenalidomide include cytopenias, diarrhoea, and rash.
DNA Hypomethylating Agents
Low doses of the nucleoside analogue cytarabine (Ara-C) were used in the 1980s as cytoreductive and so-called differentiating therapy in patients with MDS, with limited effect. Use of low-dose Ara-C has largely been supplanted by azacytidine and decitabine, two azanucleoside analogues that irreversibly inhibit the enzyme DNA methyltransferase, thereby reducing the cytosine methylation “epigenetic” status of DNA and altering gene expression. Whether or not this putative epigenetic mechanism of action accounts for the response to these agents is unproven, but hypomethylating agents have become the mainstay of therapy for most patients with higher-risk MDS, as well as for patients with lower-risk MDS who are refractory to other therapies.
A randomized trial showed that the use of azacytidine at a dose of 75 mg/m 2subcutaneously daily for 7 days every month (given until disease progression or toxicity), compared to observation, resulted in a delay in time to progression to AML and improved quality of life. Subsequently, an international multicentre trial randomized patients to receive either 7-day azacytidine or conventional care therapies (i.e., doctor’s choice of either AML induction chemotherapy, low-dose Ara-C, or supportive care alone) and found a 9-month survival prolongation (from 15 to 24 months) in patients receiving azacytidine, compared with the control arm. These results made azacytidine the standard of care for patients with higher-risk MDS. Cytopenias and gastrointestinal upset are the most common adverse effects of this class of drugs.
Decitabine is also an active agent in patients with MDS. However, a randomized trial of decitabine compared to observation failed to show survival benefit, probably because of a suboptimal dose and schedule of decitabine.
The complete and partial response rates associated with these drugs are low, but at least half of the patients who receive a hypomethylating agent do experience an improvement in at least one cytopenia. Despite these relatively modest responses, a survival benefit is seen even in patients who do not have a complete response. The major problem is that these agents are not curative, and once they fail, the patient’s life expectancy is less than 6 months. There are numerous therapies in development that are being tested, but at present there is no standard of care for patients whose disease worsens while undergoing therapy with azacytidine or decitabine.
Stem Cell Transplantation
Because allogeneic stem cell transplantation is potentially curative, this modality should be considered in all patients with higher-risk MDS who are approximately 75 years old or younger and lack major comorbid conditions. Reduced-intensity conditioning stem cell transplantation is feasible in adults up to that age, whereas myelosuppressive stem cell transplants are generally reserved for those younger than approximately age 55. Studies considering both types of stem cell transplants suggest that patients with higher-risk disease should be referred immediately for stem cell transplantation if feasible. The optimal conditioning regimen is unclear.
Relapses are common after stem cell transplantation. It is uncertain whether there is benefit from the administration of so-called bridging therapy with azacytidine or decitabine before a planned transplant in an effort to reduce the leukemic burden. Because most patients with MDS are older and uncommonly have a younger sibling who is a suitable stem cell donor, azacytidine or decitabine may be useful to stabilize a patient prior to transplant while an unrelated donor search is ongoing. Although the long-term outcome of patients undergoing stem cell transplant varies widely according to disease and host features, approximately one third of patients with MDS can expect to be long-term disease-free survivors after a stem cell transplant at a cost of 20 to 30% in treatment-related mortality, a figure that may be dropping with time.
For higher-risk patients, if feasible, urgent transplantation should be the goal, either directly or after exposure to a hypomethylating agent to decrease marrow blasts. If transplant is not the goal, prolonged administration of hypomethylating agents is generally indicated. In lower-risk patients who have a chromosome 5 abnormality, the immunomodulatory agent lenalidomide is the treatment of choice. For other lower-risk patients, supportive care alone, hematopoietic growth factors, immunosuppressive therapy, immunomodulatory therapy, or hypomethylating agents can be considered, depending on the clinical situation.
WORLD HEALTH ORGANIZATION (WHO) SUBTYPES OF MYELODYSPLASTIC SYNDROME
|MDS SUBTYPE||DYSPLASIA||BLAST COUNT||OTHER FINDINGS|
|Refractory cytopenia with unilineage dysplasia (RCUD)||· • Present in only 1 lineage (>10% of cells in that lineage):
o • Erythroid (subtype: refractory anaemia)
o • Myeloid (subtype: refractory neutropenia)
o • Megakaryocytic (subtype: refractory thrombocytopenia)
|· •Peripheral blood: <1%
· • Bone marrow: <5%
|· • Cases with bilineage cytopenias may be included in this category, but marrow dysplasia must be limited to 1 lineage
· • Ring sideroblasts represent <15% of erythroid precursors
|Refractory anaemia with ring sideroblasts (RARS)||· • Present in the erythroid lineage (>10% of erythroid precursors)||· •Peripheral blood: none
· • Bone marrow: <5%
|· • Anaemia (normocytic/macrocytic)
· • Ring sideroblasts comprise ≥ 15% of erythroid precursors
|Refractory cytopenia with multilineage dysplasia (RCMD)||· • Present in 2 or more lineages (>10% of cells in each affected lineage)||· •Peripheral blood: <1%
· • Bone marrow: <5%
|· • One or more cytopenias
· • No Auer rods
· • May or may not have ≥ 15% ring sideroblasts
|Refractory anaemia with excess blasts-1 (RAEB-1)||· • Present in 1 or more lineages (>10% of cells in each affected lineage)||· •Peripheral blood: <5%
· • Bone marrow: 5-9%
|· • One or more cytopenias
· • No Auer rods
· • 2-4% blasts in the peripheral blood → RAEB-1
|Refractory anaemia with excess blasts-2 (RAEB-2)||· • Present in 1 or more lineages (>10% of cells in each affected lineage)||· •Peripheral blood: 5-19%
· • Bone marrow: 10-19%
|· • One or more cytopenias
· • Presence of Auer rods for any blast count <20% → RAEB-2
|MDS with isolated deletion of chromosome 5q (5q− syndrome)||· • Increased numbers of megakaryocytes, many small and with hypolobated/nonlobated nuclei
· • Dysplasia in other lineages less common
|· •Peripheral blood: no or rare blasts (<1%)
· • Bone marrow: <5%
|· • Anaemia (often macrocytic) with or without other cytopenias/thrombocytosis
· • No Auer rods
· • Interstitial or terminal deletion of the long arm of chromosome 5
|MDS, unclassifiable (MDS-U)||· • Unequivocal, but present in <10% of cells of one or more lineages||· •Peripheral blood: ≤1%
· • Bone marrow: <5%
|· • May progress to a specific MDS
· • Can also include cases otherwise classified as RCUD or RCMD but with 1% blasts in peripheral blood; RCUD but with pancytopenia
· • Cases with an MDS-associated chromosome abnormality (other than loss of Y chromosome or trisomy 8 or deletion of chromosome 20) but without dysplasia can be included in this category
At present there is no clear way MDS can be prevented, with the exception of avoiding known precipitants (e.g., using lower doses of radiotherapy or avoiding radiation altogether, and avoiding exposure to alkylating agents or topoisomerase II inhibitors). In the future, it is likely that emerging clonal haematopoiesis following treatment for cancer or arising de novo may be recognized at an early stage. Initiation of immune-based therapies designed to break the lymphocyte tolerance of such abnormal clones may become useful.
Because the natural history of MDS varies widely among patients, several prognostic tools have been derived to help clinicians distinguish patients with a high risk of progression to AML and death from cytopenias within a few months, from those patients whose disease is likely to be more indolent and stable for several years. Among these tools is the 1997 International Prognosis Scoring System (IPSS), which assessed patients’ risk by scoring the number of cytopenias, the chromosome pattern (some anomalies are associated with a better prognosis than others), and the proportion of blast cells in the marrow. In 2012, a revised version of the IPSS (IPSS-R) was published; it included a broader range of cytogenetic abnormalities than the original IPSS and weighted abnormal cytogenetics more heavily. The IPSS-R defines five different subgroups of MDS with varying risks of death and progression to AML.
2012 REVISED INTERNATIONAL PROGNOSTIC SCORING (IPSS-R) FOR MYELODYSPLASTIC SYNDROMES
|RISK GROUP||INCLUDED KARYOTYPES||MEDIAN SURVIVAL, YEARS||25% OF PATIENTS TO AML, YEARS||PROPORTION OF PATIENTS IN THIS GROUP|
|Very good||del(11q), −Y||5.4||N/R||4%|
|Good||Normal, del(20q), del(5q) alone or with 1 other anomaly, del(12p)||4.8||9.4||72%|
|Intermediate||+8, del(7q), i(17q), +19, any other single or double abnormality not listed||2.7||2.5||13%|
|Poor||Abnormal 3q, −7, double abnormality include –7/del(7q), complex with 3 abnormalities||1.5||1.7||4%|
|Very poor||Complex, with >3 abnormalities||0.7||0.7||7%|
|PARAMETER||CATEGORIES AND ASSOCIATED SCORES|
|Cytogenic risk group||Very good||Good||Intermediate||Poor||Very poor|
|Marrow blast proportion||≤2%||>2-<5%||5-10%||>10%|
|Haemoglobin||≥10 g/dL||8-<10 g/dL||<8 g/dL|
|Absolute neutrophil||≥0.8 × 10 9/L||<0.8 × 10 9 /L|
|Platelet count||≥100 × 10 9/L||50 -100 × 10 9L||<50 × 10 9/L|
|Possible range of summed scores: 0-10.|
|RISK GROUP||POINTS||% PATIENTS ( n = 7012; AML DATA ON 6485)||MEDIAN SURVIVAL, YEARS||MEDIAN SURVIVAL FOR PATIENTS UNDER 60 YEARS||TIME UNTIL 25% OF PATIENTS DEVELOP AML, YEARS|
|Very low||0-1.5||19%||8.8||Not reached||Not reached|
Top panel, Cytogenetic (chromosome) classification used in IPSS-R. Middle panel, Table for calculation of IPSS-R risk group. Bottom panel, Outcomes by risk group. Cytogenic (chromosome) classification used in IPSS-R.
Comorbid conditions also influence prognosis independently of the MDS. Other relevant factors that are not included in current prognostic models include serum ferritin and lactate dehydrogenase levels, aberrant expression of certain cell surface markers on myeloid cells detected by flow cytometry (e.g., CD5 or CD56 on myeloid cells), rarer cytogenetic abnormalities not included in the IPSS, and most importantly, the presence of certain molecular abnormalities. Several molecular changes, including mutations in EZH2 , TP53 , ASXL1 , RUNX1 , and ETV6 are associated with a poorer prognosis than would have been predicted by the IPSS; in patients with lower-risk disease, only EZH2 retains independent prognostic value. Additionally, the presence of either a TP53 or a DNMT3A mutation predicts for a worse outcome after stem cell transplant.