Transfusion-Dependency Is the Most Important Prognostic Factor for Survival in 1000 Newly Diagnosed MDS Patients with Low- and Intermediate-1 Risk MDS in the European LeukemiaNet MDS Registry

Abstract

Abstract 2775 Background:The ELN EUMDS registry collects information about demographics and disease-management of newly diagnosed low- and intermediate-1 risk MDS patients. Objective:To define the prognostic impact of transfusion-dependency and serum ferritin levels. Results:From April 2008 until December 2010, 1000 patients have been registered in fourteen European countries. The median age is 74 years (range 18–95), 60% are male. The WHO subgroups are RCMD (35.8%), RARS (18.2%), RA (17.6%), RAEB-1 (11.6%), del5q (6.2%), MDS-U (3.1%) and RAEB-2 (0.4%). IPSS score (n=935) is 0 in 48.8%, 0.5 in 31.2% and 1 in 13.5%. At time of registration 19% of the patients had started MDS specific treatment, increasing to 46% at 18 months of follow-up, usually consisting of erythroid-stimulating agents (ESA). At registration 29% of the patients were transfusion-dependent, which remained stable during follow-up. The number of transfusion-dependent patients with ESA treatment decreased from 58% to 27% at 18 months of follow-up.Overall survival at 18 months of follow-up was 89%. The mortality rate in transfusion-independent and transfusion-dependent patients at 18 months of follow-up was 5% and 21%, respectively (P<0.0001). Prognostic value of the IPSS score was confirmed with Cox regression analysis: the hazard ratio (HR) with IPSS score 0.5 was 2.20 (95% Confidence Intervals (CI); 1.40–3.48) and with IPSS score 1.0 was 3.08 (95% CI; 1.78–5.31) respectively, adjusted for age, sex, country and WHO category. For overall survival based on ferritin status, patients were divided in three groups according to their ferritin levels, based on expected clinical significance; group 1. ≤300 μg/L, group 2. >300 <1000 μg/L and group 3. ≥1000 μg/L. The mortality rate according to serum ferritin at registration in these groups was 6%, 14% and 23%, respectively, with HRs 1.80 (95%CI 1.06–3.03) and 3.21 (95% CI 1.66–6.20), including adjustment for transfusion status and number of transfusions. To define the relationship between serum ferritin and transfusion-dependency, the serum ferritin levels in the three groups were compared in transfusion-independent and transfusion-dependent patients. The mortality rate according to serum ferritin at registration in transfusion-independent patients was 3%, 6% and 3%, respectively (HR 2.17 and 0.95 respectively). The mortality rate according to serum ferritin at registration in transfusion-dependent patients was 14%, 21% and 35%, respectively (HR 1.69 and 3.93 respectively) (Table 1; Graph 1). To define the prognostic value of transfusion burden, patients were divided in groups based on transfusion status: 1. No transfusions, 2. Transfusions <20 units, 3. Transfusion >20 units. The mortality rate in these groups was 5%, 18% and 30%, respectively. The adjusted HRs were 3.67 (95% CI; 2.28–5.91) and 5.50 (95% CI; 3.18–9.52), respectively (Table 1).Table 1:Overall survival at 18 months adjusted for age, sex, country, WHO classification, cytogenetics, cytopenias and % blastsTotalDiedOverall survival HR (95% CI)1Log rank testN1000114Transfusions2:No573261Yes427884.11 (2.61–6.46)P<0.0001Transfusion-independentFerritin at registration (μg/L):1. ≤300239612. >300 & <100015192.17 (0.75–6.26)3. ≥10003210.95 (0.11–8.08)p=0.24Transfusion-dependent2Ferritin at registration (μg/L):1. ≤3001181712. >300 & <1000153321.69 (0.92–3.09)3. ≥100054193.93 (1.98–7.78)P<0.0001Transfusion status:1. No Transfusions5732612. Transfusions <20 units3323573.67 (2.28–5.91)3. Transfusions >20 units3104315.50 (3.18–9.52)P<0.00011Hazard Ratio (HR) (95% Confidence Intervals (CI))2At least one red blood cell transfusion recorded at 0,6,12,18 months of follow-up3Total number of transfused units from registration until 18 months of follow-up [Display omitted] Conclusion:After 18 months of follow-up serum ferritin levels appear to be an important prognostic factor for transfusion-dependent patients only. However, transfusion burden is the most important prognostic factor for survival in patients with more than 20 units transfused compared to non-transfused patients. An increased risk was also apparent for moderately transfused patients (<20 units) so direct toxicity of transfusions (toxic iron radicals?), even at a relatively low iron load, might have an important adverse impact on survival. Disclosures:Fenaux:Celgene: Honoraria, Research Funding; Novartis: Honoraria, Research Funding; Janssen Cilag: Honoraria, Research Funding; Roche: Honoraria, Research Funding; Amgen: Honoraria, Research Funding; Glaxo Smith Kline: Honoraria, Research Funding; Merck: Honoraria, Research Funding; Cephalon: Honoraria, Research Funding. Germing:Celgene: Consultancy, Research Funding.