High-Dose Cyclophosphamide in the Treatment of Children with Acquired Severe Aplastic Anemia Refractory to Conventional Immunosuppression.

Severe aplastic anemia (SAA), an illness characterized by the depletion of hematopoietic precursors within bone marrow leading to pancytopenia, has an overall mortality rate of >75% if left untreated. For children with human leukocyte antigen (HLA)-identical sibling donors, the treatment of choice for acquired SAA is allogeneic bone marrow transplant (BMT) with long-term survival of 70%–85%. For the 80% of children who lack suitable bone marrow donors, the standard treatment is immunosuppression with anti-thymocyte globulin (ATG), cyclosporine A (CSA), and often hematopoietic growth factors. Large series utilizing this immunosuppressive therapy have demonstrated a complete response rate of 60%–80%.

When a patient does not have an HLA-identical sibling and fails to respond to conventional immunosuppression, options for further treatment are limited. One option is to pursue a matched unrelated donor (MUD) BMT, but matched donors are not available for all patients and long-term survival ranges between 20%– 60%. Treatment with high-dose cyclophosphamide is another option for refractory SAA patients that appears promising but has not been extensively studied in children.

In 1999, a group of pediatric hematology centers joined together to study the use of high-dose cyclophosphamide for the treatment of children with acquired SAA not eligible for BMT and refractory to immunosuppression (Pediatric Aplastic Anemia Cooperative Trial #2). The goals of this study were to determine the response rate and toxicity associated with high-dose cyclophosphamide in children with refractory SAA. Between 10/1/99 and 6/3/02, 6 patients from 6 different centers were enrolled and received 4 days of intravenous cyclophosphamide (45 mg/kg/day), MESNA, and GM-CSF (250 mcg/m²/day SC) post-cyclophosphamide. The patient population consisted of 4 males and 2 females ranging in age from 2 to 18 years. The interval between diagnosis of SAA and cyclophosphamide treatment ranged from 8 to 88 months. All had failed to respond to earlier treatment with ATG and CSA.

Twelve months following cyclophosphamide, there was 1 complete response (transfusion independent and normal blood counts), 1 partial response (transfusion independent but with moderate pancytopenia), 2 infectious deaths without recovery of blood counts, and 1 failure to respond. One patient was removed from the study before response could be assessed. High-dose cyclophosphamide in this population was associated with significant toxicity as 2 patients developed disseminated fungal infections within 30 days of starting therapy leading to death at day 18 in one. Another patient experienced grade 3 gingivitis/stomatitis lasting about 30 days.

In summary, high-dose cyclophosphamide can lead to marrow recovery in some children with refractory SAA but is associated with the potential for life-threatening infectious complications, especially with fungus.