Treatment of anemia in myelodysplastic syndromes with Granulocyte Colony-Stimulating Factor and erythropoetin Response and impact on survival in a long-term follow-up of 129 patients

Jadersten, M.; Montgomery, S. M.; Astermark, J.; Carlsson, M.; Dahl, I. M. rie; Dybedahl, I.; Engstrom, L. M. ria; Grimfors, G.; Hesse Sundin, E.; Hjorth, M.; Juliusson, G.; Linder, O.; Luthman, M.; Nilsson Ehle, H.; Oberg, G. J.; Porwit Mcdonald, A.; Samuelsson, J.; Tangen, J.

Treatment of anemia in patients with low-risk myelodysplastic syndromes (MDS) with erythropoetin (Epo) and Granulocyte Colony-Stimulating Factor (G-CSF) may normalize or improve hemoglobin levels and improve quality of life in around 40% of patients. This study is the first to report long-term outcome of a large cohort of MDS patients treated with G-CSF+Epo. 129 MDS patients included in three Nordic MDS Group studies between 1990 and 1999 were followed up regarding survival, evolution to acute myeloid leukemia (AML), response duration and optimal dosing regimens. Inclusion criteria were hemoglobin below 100 g/l or transfusion dependent anemia. The third study excluded patients who belonged to the poor predictive group for response according to a validated decision model (Hellstrom-Lindberg, Br JH 2003). The effect of treatment on long-term outcome was estimated through a comparison with untreated patients from the International Prognostic Scoring System (IPSS) Database for MDS (Greenberg, Blood 1997). In the current study, the majority of patients had low-risk MDS; 30 refractory anemia (RA), 41 RA with ringed sideroblasts (RARS), and 58 RA with excess blasts (RAEB), according to French-American-British (FAB) classification. 118 patients were risk-stratified according to the IPSS; 34 low, 57 intermediate-1 (INT-1), 22 intermediate-2 (INT-2), and 5 high. Forty-three, 45, and 16 patients had good, intermediate, and poor probability of response, respectively, according to the predictive model. The median follow-up time was 30 months (range 2-142 months). The median survival was 30 months from start of study. The overall erythroid response-rate was 39% (RA 39%, RARS 50%, RAEB 31%); 22% complete remission (CR) and 17% partial remission (PR). The median duration of response in patients receiving maintenance therapy was 23 months (range 3-116 months). Responses were more durable in RA+RARS than in RAEB (median 28 vs. 12 months, p=0.04), and in CR compared to PR, (median 29 vs. 12 months, p=0.006). More than 20% of responses had a duration of >4 years. To make a valid comparison of the overall survival from diagnosis between the treated and the IPSS patients, a cohort of treated patients with short time from diagnosis to start of study was selected. The odds ratio for risk of death for the patients treated <3 months from diagnosis was 1.1 (0.7-1.6 within a 95% C.I.). However, this cohort had worse prognostic factors than patients treated >3 months from diagnosis. The time to 25% of patients evolving to AML did not differ between IPSS and the treated cohort (p=0.90). Only 1 of 20 patients responding >2 years developed AML. The lowest effective maintenance dose for Epo ranged from 5 000 to 50 000 U/week (median 30 000 U), and for G-CSF from 0 to 900 mg/week (median 225mg). RA patients required lower doses of Epo than RARS and RAEB. RAEB patients required higher G-CSF doses than RA and RARS. We conclude that G-CSF+Epo may induce complete and durable responses in a substantial proportion of low-risk MDS-patients with anemia, and thereby increase patients' quality of life, with no negative effect on long-term outcome.