Aplastic anemia is a disease in which the bone marrow, and the blood stem cells that reside there, are damaged. This causes a deficiency of all three blood cell types (pancytopenia): red blood cells (anemia), white blood cells (leukopenia), and platelets (thrombocytopenia).

Until the 1970s, severe aplastic anemia (SAA) was almost uniformly fata, but in the early 21st century most patients can be effectively treated and can expect long-term survival.


The pathophysiology responsible for marrow cell destruction and peripheral blood pancytopenia has itself been inferred from the results of treatment in humans, with substantial in vitro and animal model support.

The success of HSCT in restoring hematopoiesis in SAA patients implicated a deficiency of HSCs. Hematologic improvement after immunosuppressive therapy (IST) implicated the immune system in destruction of marrow stem and progenitor cells.

Genetics influences both the immune response and its effects on the hematopoietic compartment. There are histocompatibility gene associations with SAA, and some cytokine genes may be more readily activated in patients because of differences in their regulation, as suggested by polymorphisms in promoter regions.

An inability to repair telomeres and to maintain the marrow’s regenerative capacity, resulting from mutations in the complex of genes responsible for telomere elongation, has been linked to patients with familial or apparently acquired SAA, with or without the typical physical stigmata of constitutional aplastic anemia.

Approximately 5% to 10% of patients with SAA have a preceding seronegative hepatitis. However, most patients do not have a history of identifiable chemical, infectious, or medical drug exposure before onset of pancytopenia.

The antigen(s) inciting the aberrant immune response have not been identified in SAA. Furthermore, the current simple mechanistic outline may be supplemented in the future with better understanding of now theoretical possibilities, suggested by provocative murine models.

Management of Acquired Aplastic Anemia

Table 1. Criteria for SAA

For patients with moderate aplastic anemia, as defined by lack of blood count criteria for SAA, observation is often appropriate, especially when they do not require transfusion support. The criteria for SAA is in Table 1. Many of these patients may have stable blood counts for years, but in some pancytopenia worsens over time.

Once the disease progresses and the patients meets the criteria for SAA or become transfusion-dependent, treatment is always required. The therapeutic strategies for SAA include immediate supportive care, transplantation (HSCT), and immunosuppressive thearpy (IST), in which HSCT and IST are called definitive treatment.

Supportive care initiates and the definitive treatment should be started as soon as possible, since prolonged delay until initiation of primary treatment for SAA is not generally desirable and can result in serious complications before definitive therapy. Watchful waiting, especially if neutropenia is profound, can be harmful and is not appropriate once a diagnosis of SAA is confirmed. To realize this goal, it is prudent to rapidly assess whether matched sibling donors exist in the family for any patients younger than 40 years of age. Why the upper limit of age is 40 years is related to the risk of GVHD and we will discuss further below.

Immediate Supportive Care

Symptoms related to anemia and thrombocytopenia can be readily corrected with transfusions. Broad spectrum parenteral antibiotics are indicated when fever or documented infection occurs in the presence of severe neutropenia (< 0.5 × 109/L). As we have discussed in the blog “Platelet transfusion for patients with cancer (part two)“, prophylatic platelet transfusion is necessary when needed, and the threshold in these patients is 10 × 109/L.

Of note, transfusion of red cells aims to alleviate symptoms of anemia, not simply to target a specific hemoglobin threshold. Adequate red blood cell transfusions in symptomatic patients should not be deferred because of fear of iron accumulation or to reduce the risk of alloimmunization. However, in some guidelines it is recommended that starting iron chelation therapy in those patients who have received 20-25 units of packed RBCs or who have a serum ferritin level of >1000 μg/L (see post “The Management of Myelodysplastic Syndromes“).

In evaluating febrile neutropenic patients, simple chest X-ray is of limited value, so routine CT imaging of the sinus and chest is a preferred approach which followed by nasal endoscopy, bronchoscopy, and biospy for microbiologic confirmation when indicated. If fungal infection is suspected or neutropenic fever persists for more than several days despite broad-spectrum antimicrobials, empiric antifungal therapy should include drugs active against Aspergillus sp, as this pathogen has remained the most common fungal isolate in SAA patients for the past 20 years.

supportive measures alone, growth factors, androgens, or cyclosporine (CsA) are not definitive therapies. Patients should not be subject to initial trials of G-CSF or erythropoietin. Corticosteroids are of unproven benefit and inferior in efficacy to conventional immunosuppression regimens, but they are more toxic and should not be used as therapy in SAA. It is very unfortunate when a patient with SAA presents for transplant or IST but already has life-threatening fungal infection because of weeks or months of exposure to corticosteroids.

Figure 1. Algorithm for Initial Management of SAA

Algorithm for initial management of SAA


For malignancies, GVHD may offer graft-versus-tumor-benefits. For example, in AML the presence of GVHD is associated with reduced relapse of AML. However, in SAA GVHD is unequivocally to be avoided since the presence of GVHD in SAA definitely decreases survival and long-term quality of life.

The age of the patient with SAA is a crucial element for the decision to transplantaton. Generally, patient older than 40 years of age are not recommended to undergo transplantation since the correlation of increasing age with the increased risk of GVHD and therefore the significant morbidity and mortality of this complication is apparent. A study of more than 1300 SAA patients who were transplanted from 1991 to 2004 showed that survival at 5 years for patients younger than 20 years of age was 82%, for those 20 to 40 years of age 72%, and for those older than 50 years of age closer to 50%. Rates of GVHD increased with age, accounting for much of the decreased survival in older patients and the long-term morbidity.

The stem cell could be collected from either bone marrow or peripheral. The stem cell source also is important for survival and long-term quality of life. In a retrospective analysis, the rate of chronic GVHD was greater with peripheral blood (27%) compared with bone marrow stem cell grafts (12%) in patients younger than 20 years of age. In a subsequent retrospective analysis, similar higher rates of chronic GVHD were observed for patients of all ages undergoing HSCT with peripheral blood compared with bone marrow derived stem cell grafts.

For unrelated donor transplants, bone marrow source of stem cells was associated with lower rates of acute GVHD (31%) compared with peripheral blood-derived CD34+ cells (48%), which not only means bone marrow source is preferred than peripheral blood source but also matched sibling donor is preferred than unrelated donor.

The donor for HSCT consists of matched sibling donor and alternative donor. Again matched sibling donor is preferred since experience from larger cohorts reported in the last 5 years from the United States, Japan, Korea, and Europe suggests that the outcome with unrelated donor HSCT is still not as favorable as that of a matched sibling donor. Meanwhile, practically, identification of a matched unrelated donor and coordination with a transplant center usually takes several months, and delaying definitive IST while conducting a serach for nonfamily donor may be dangerous.

Prospective trials using umbilical cord HSCT in SAA are limited to smaller case series, which do show encouraging results. However, experience from larger cohorts in retrospective analyses indicate that overall survival is not as favorable as in pilots, at approximately 40% at 2 to 3 years. Graft rejection and poor immune reconstitution continue limit the success of umbilical cord HSCT.

Immunosuppressive Therapy

For patients older than 40 years of age, or yonger than 40 years of age but without matched sibling donor, IST is the first alternative therapeutic approach other than unrelated donor HSCT.

For patient older than 40 years of age, if initial IST fails (no response at 6 months, we define refractory SAA as blood counts still fulfilling criteria for severe pancytopenia 6 months after initiation of IST), we consider matched sibling HSCT first if he/she has a matched sibling donor. For patients younder than 40 years of age but without matched sibling donor, if initial IST fails, we conisider matched unrelated donor HSCT first if he has a histocompatible donor.

Whereas if the patient has no histocompatible donor or don’t suitable for HSCT we consider to repeat the IST therapy with rabbit ATG and CsA or with Alemtuzumab alone. If the patient remains no response 6 months after the second IST therapy, we consider HSCT (mismatched unrelated, haploidentical, or umbilical cord) or non-HSCT approach (see in Figure 1).

Standard initial IST is horse ATG (anti-thymocyte globulin) and CsA (cyclosporine), which produces hematologic recovery in 60% to 70% of cases and excellent long-term survival among responders, as shown in several large prospective studies in the United States, Europe, and Japan.

The mechanism of action by which polyclonal antilymphocyte preparations suppress immune responses is not fully understood. Possible mechanisms by which Thymoglobulin may induce immunosuppression in vivo include: T-cell clearance from the circulation and modulation of T-cell activation, homing, and cytotoxic activities. Thymoglobulin includes antibodies against T-cell markers such as CD2, CD3, CD4, CD8, CD11a, CD18, CD25, CD44, CD45, HLA-DR, HLA Class I heavy chains, and ß2 micro-globulin. In vitro, thymoglobulin (concentrations >0.1 mg/mL) mediates T-cell suppressive effects via inhibition of proliferative responses to several mitogens. In patients, T-cell depletion is usually observed within a day from initiating Thymoglobulin therapy. Thymoglobulin has not been shown to be effective for treating antibody (humoral) mediated rejections.

Study showed the addition of CsA to ATG increased the hematologic response rate, however, other agents such as mycophenolate mofetil, growth factors, or sirolimus to horse ATG/CsA did not improve rates of response, relapse, or clonal evolution. A more lymphocytotoxic agent rabbit ATG, has been successful in salvaging patients with refractory or relapsed SAA after initial horse ATG. But in recent large, randomized controlled study, hematologic response to rabbit ATG (37%) was about half that observed with standard horse ATG (68%), with inferior survival noted in the rabbit ATG arm. Therefore horse ATG remains the most effective regimen for first-line IST therapy regimen of SAA.

Therapy Response Evaluation

We use simple definition for hematologic response: no longer meeting blood count criteria for SAA, which closely correlates with transfusion independence and long-term survival. Hematologic improvement is not to be expected for 2 to 3 months after ATG. The majority of response (90%) occur within the first 3 months, with fewer patients responding between 3 and 6 months or after.

After ATG therapy patients with SAA needed to long-term follow-up, we will discuss this scope in another post. Coming soon.