Use of G-CSF in Patients with Severe Aplastic Anemia Treated with ATG and Cyclosporine Increases Neutrophils and Decreases Infection Rates and Hospitalization Days but Does Not Improve Long-Term Outcome: Results of a Prospective, Randomized Clinical Trial of the EBMT.

Immunosuppression (IS) with ATG and Cyclosporine (CSA) is the treatment of choice for patients with severe aplastic anemia (AA) not eligible for hematopoietic stem cell transplantation. The role of additional growth factors such as G-CSF on long-term outcome is still a matter of debate. The Aplastic Anemia Working Party of the EBMT conducted a randomized controlled study comparing IS with ATG and CSA with or without G-CSF and with or without early retreatment at day 120 in case of nonresponse. The study was designed to identify a difference of 10% for overall survival (OS) and event free survival (EFS). Three hundred forty subjects were to be enrolled. From 2002 to 2008, 205 patients were randomized, but 13 had to be excluded (1 incorrect diagnosis; 12 no follow-up data); 192 were evaluable for analysis (95 with G-CSF; 97 without G-CSF). The median age of the patients was 46 years (2-81), 95 (49%) were males, 69 (36%) had very severe AA. There was no difference between both groups in respect of age, sex, severity of the AA, and number of transfusions before treatment. Horse ATG (Lymphoglobulin) was administered (15mg/kg/BW/d ×5 days), and CSA given orally at a dose of 5mg/kg/d. Patients randomized to receive G-CSF were given glycosylated rHuG-CSF from day 8 to 120 (150mg/m²/d, sc).

OS at 6 years was 75%, it was 82% for patients with severe AA and 66% for patients with very severe AA (P=0.001). Survival decreased with increasing age from 100% (age <20years) and 92% (20-40 years) to 71% (40-60 years) and 56% (>60 years) (P<0.001). There was no difference in OS (P=0.64) and EFS, defined by death, need for transplantation, relapse and non response as events (P=0.358) between the study arms at 6 years. This was the case for the entire cohort as well as when stratified according to age and AA severity (Table).

The median number of neutrophils was significantly higher between day 30 and 240, in the treatment group but this difference did not persist to day 360, when G-CSF was stopped. There were fewer infections (36% no G-CSF; 24% with G-CSF; P=0.006), and less days of hospitalization during the first 90 days (P=0.03) in the group of patients with G-CSF, mainly for patients with very severe AA. During the study period 44 patients died. The most common cause of death was infection (55%).There was no difference in deaths and causes of death between both treatment groups. However, there were more late deaths (>3 years) in the G-CSF group (P=0.01). There was no difference in respect of response rates between both groups. Overall 73% of patients with and 62% without G-CSF did respond to IS (P=0.488). Response rates at days 30, 60, 90 120, 150, 180, 240 and 360 were similar between the treatment groups. Fifty seven patients did not respond to first-line therapy, and 31 patients relapsed during the first year of treatment, without any difference between both groups.

In conclusion, G-CSF given together with standard IS increases neutrophil counts, and decreases rate of infections and days of hospitalization, mainly in patients with very severe AA. G-CSF does not improve long-term outcome and has no impact on OS, EFS, remission, death and relapse rates. G-CSF may decrease infection risks when used as an adjunct to IS therapy. This has to be weighed against possibly higher risks of MDS/AML, as suggested by previous studies.

Schrezenmeier:Alexion: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding.