Background:

• Severe aplastic anemia (SAA) can lead to problems with bone marrow health and result in low blood cell counts, which require frequent transfusions. Standard initial treatment for SAA involves injections of antithymocyte globulin (ATG) plus cyclosporine (CsA). Patients with SAA who do not respond to initial treatment with ATG (refractory) have a high risk of dying without additional treatment. In these cases, for those who do not have a matched bone marrow transplant donor there is no well-defined standard therapy. In our experience with patients who do not respond to horse ATG + CsA, only about one-third of patients who are re-treated with rabbit ATG + CsA improve. Experience with cyclophosphamide in the treatment of refractory severe aplastic anemia suggests that this drug is able to improve blood counts in about 50% of cases. However, the cyclophosphamide regimen has been associated with a significant infection risk (mostly caused by fungus) in studies conducted over 10 years ago due to the lowering of the white blood cell levels.
• Better antibiotic drugs against fungus have been developed and are widely used to treat patients who have low white blood cell counts and are at risk of developing infections. In SAA patients in particular, these newer antibiotics have had a large impact in preventing and treating fungus infections. Researchers are revisiting the use of cyclophosphamide at lower doses to minimize its side effects given in combination with another immune suppressant, fludarabine.

Objectives:

- To determine the safety and effectiveness of the combination of fludarabine plus cyclophosphamide in treating severe aplastic anemia that has not responded to initial treatments.

Severe aplastic anemia (SAA) is a life-threatening bone marrow failure disorder characterized by pancytopenia and a hypocellular bone marrow. Allogeneic hematopoietic stem cell transplantation (HSCT) offers the opportunity for cure in 70 percent of patients, but most patients are not suitable candidates for this treatment modality due to advanced age, comorbidities or lack of a histocompatible donor. For these patients, comparable long-term survival is attainable with immunosuppressive treatment (IST) with anti-thymocyte globulin (ATG) and cyclosporine (CsA). However, approximately 1/3 of patients do not show blood count improvement after ATG/CsA and are considered to have refractory disease. Furthermore, analysis of our own extensive clinical data suggests that poor blood count responses to a single course of ATG (nonrobust responders), even when transfusion-independence is achieved, predicts a markedly worse prognosis compared to those who achieve a robust hematologic improvement (protocol 90-H-0146) .

In patients who are refractory to horse ATG (h-ATG) and do not have a histocompatible sibling, alternative donor (AD) HSCT or a repeat course of IST are options. Registry data suggests that outcome for AD HSCT in SAA is not as favorable compared to a matched sibling HSCT with long-term survival at about 40-50 percent and a higher risk of graft-versus-host disease. However, in recent smaller retrospective studies survival after AD HSCT in children rivals that of a sibling transplant when an unrelated donor who matches in 10 human leukocyte antigen (HLA) loci (matched unrelated donor) is available. With repeat IST, response rates with rabbit anti-thymocyte has varied from 22 percent up to 77 percent. Our experience in refractory SAA (protocol 03-H-0249) is that rabbit ATG + CsA and alemtuzumab are equally efficacious as salvage therapies, with a response rate of about 30 percent for each treatment. For the 20-30 percent of patients who remain refractory after 2 courses of treatment, further courses of IST have been of limited value with responses observed only occasionally. In addition, efforts to improve initial IST in treatment-na ve patients (addition of mycophenolate mofetil and sirolimus) have not yielded promising regimens with activity in SAA (protocols 00-H-0032, 03-H-0193, and 06-H-0034). Therefore, novel regimens are needed to improve outcomes in SAA for those without a histocompatible sibling, which encompass the majority of patients with this disease.

The current limitations of IST in SAA are: 1) the majority of the responses observed following initial h-ATG/CsA are partial with only a few patients achieving normal blood counts; 2) 1/3 of patients are refractory to initial h-ATG/CsA; 3) response rate in these refractory patients who are retreated is only 30 percent; 4) hematologic relapses occur in 35 percent of responders following initial response to h-ATG/CsA; 5) among relapsed patients chronic use of CsA is not infrequent which often leads to toxicities from the long term exposure to this drug (especially in older patients); and 6) clonal evolution is still observed in 10-15% of patients. Towards the goal of addressing these limitations we are proposing a novel regimen of fludarabine (Flu) plus cyclophosphamide (Cy) in SAA patients refractory to horse ATG/CsA. The Hematology Branch has considerable experience with Flu/Cy as part of the condition regimen in allotransplantation protocols (protocols 99-H-0050, 97-H-0196, 99-H-0064, 99-H-0050, 97-H-0196, 02-H-0111, 01-H0162, 03-H-0192, 04-H-0112, 06-H-0248, 07-H-0136). In addition, this regimen has been incorporated into the NCI's Surgery Branch preparative regimen for autologous HSCT prior to infusion of tumor infiltrating lymphocytes. Flu/Cy is well tolerated and a potent immunosuppressive regimen that is not myeloablative. Therefore, we propose to investigate Flu/Cy to address the current limitations of IST in SAA.

The main objective of this study is to assess the safety and efficacy of Flu/Cy in refractory SAA. The primary endpoint will be hematologic response, defined as no longer meeting criteria for SAA, at 6 months. Secondary endpoints are relapse, robustness of hematologic recovery at 6 months, response at 3 months and 12 months, survival, clonal evolution to paroxysmal nocturnal hemoglobinuria (PNH), myelodysplasia and acute leukemia. The primary endpoint will be changes absolute neutrophil count, platelet count, and reticulocyte count at 6 months. Secondary endpoints will include time to relapse, changes in cytogenetics, and time to death.

Eligibility:

- Individuals at least 2 years of age who have severe aplastic anemia that has not improved after treatment with horse ATG or both horse and rabbit ATG.

Design:

• After initial screening, medical history, and blood tests, participants will be admitted to the inpatient unit at the National Institutes of Health Clinical Center.
• Participants will receive 2 days of cyclophosphamide, followed by 5 days of fludarabine.
• Participants will also receive antibiotics and other drugs to protect against bacterial, fungal, and viral infections. Participants will take these drugs regularly until their white blood cell counts improve.
• After discharge from the clinical center, participants will have follow-up evaluations at 3 months, 6 months, and annually for 5 years. Evaluations will include blood samples and periodic bone marrow biopsies.

-INCLUSION CRITERIA:

1. Severe aplastic anemia characterized by:

   Bone marrow cellularity < 30 percent (excluding lymphocytes)

   AND

   At least two of the following:

   • Absolute neutrophil count < 500/ microL
   • Platelet count < 20,000/ microL
   • Absolute reticulocyte count < 60,000/ microL

2. Failure to respond to an initial course of h-ATG/CsA at least 3 months post-treatment or a suboptimal response to initial h-ATG/CsA defined by both platelet and reticulocyte count < 50,000 /microL at 3 months post-treatment

   OR

3. Refractory SAA unresponsive to both horse and rabbit ATG-based regimens

4. Age greater than or equal to 2 years old

5. Weight greater than or equal to 12 kg

EXCLUSION CRITERIA:

1. Diagnosis of Fanconi anemia
2. Cardiac ejection fraction < 30 percent (evaluated by ECHO)
3. Evidence of a clonal hematologic bone marrow disorder on cytogenetics. Patients with the presence of trisomy 8, loss of Y or del(20q) will not be excluded in the absence of dysplastic changes in the marrow. Patients with very severe neutropenia (ANC < 200 /microL) will not be excluded initially if cytogenetics are not available or pending. If evidence of a clonal disorder is later identified, the patient will go off study)
4. Prior immunosuppressive therapy with high dose Cy
5. Infection not adequately controlled with appropriate therapy
6. Serologic evidence of HIV infection
7. Moribund status or concurrent hepatic, renal, cardiac, neurologic, pulmonary, infectious, or metabolic disease of such severity that it would preclude the patient's ability to tolerate protocol therapy, or that death within 30 days is likely
8. Subjects with cancer who are on active chemotherapeutic treatment or who take drugs with hematological effects
9. Current pregnancy or unwillingness to take oral contraceptives or refrain from pregnancy if of childbearing potential
10. Not able to understand the investigational nature of the study or to give informed consent

6 months