Hepcidin and GDF15 Levels during the First 2 Years Follow-up in Patients with Low and Int-1 Risk Myelodysplastic Syndromes (MDS) from the European Leukemianet MDS Registry

Background: The EUMDS registry is a prospective observational registry to collect data on lower risk MDS. 17 Countries and 133 centers are participating. We analyzed serum from 101 patients for ferritin, hepcidin, growth differentiation factor 15 (GDF15) and C-reactive protein (CRP) at six-month intervals in order to evaluate temporal changes in iron metabolism.

Objective: To explore hepcidin and GDF15 levels over time in lower risk MDS patients and their relation with WHO2001 subtype, transfusion history and conventional iron parameters.

Results: The median age of the study population was 73 years (range 44-95 years). The majority was male: 64%. Distribution according to WHO2001 MDS subtype was RCMD (41%), RARS (25%), RA (14%), RAEB (11%), 5q-syndrome (6%) and RCMD-RS (3%). Table 1 shows iron parameters at registration, 1 year and 2 years follow-up both in transfusion-dependent (TD) and transfusion-independent (TI) patients and stratified according to MDS subtypes: RS (RARS/RCMD-RS) or MDS Other (RA/RCMD/RAEB/5q-syndrome). Serum ferritin was increased in TD patients with a median concentration at registration of 550µg/L and at 2 years 818µg/L, compared to TI patients (median <250µg/L, all time points). During follow-up ferritin was most elevated in patients who received >10 red blood cell (RBC) units: median at registration 1482µg/L - 2 years 1970µg/L. Ferritin correlated significantly with hepcidin (r=0.57; p<0.001) as well as CRP: r=0.27, p<0.001. Median CRP was within reference range (<10mg/L) for both TD and TI patients at registration and during follow-up, but mainly TD patients had elevated CRP levels >50mg/L. Median serum hepcidin levels were elevated in TD patients at registration and remained elevated during follow-up, especially in patients with >10 RBC units transfused (median 27.4nmol/l at registration, 12.8nmol/l at 2 years). Remarkable fluctuation in hepcidin levels occurred in patients with elevated hepcidin during follow-up. Even in the longitudinal cohorts hepcidin fluctuated considerably, maybe due to the interval between the previous transfusion and the measurement of hepcidin or due to diurnal fluctuation. Hepcidin was lowest in MDS RS TI patients and showed a tendency to decrease over time from a median level of 4.4nmol/l at registration to 2.4nmol/l after 2 years, associated with ineffective erythropoiesis. This is supported by the high median GDF15 in these patients. Lowest GDF15 was found in TD patients with ‘MDS Other’ associated with transfusional load. The number of transfused RBC units did not affect the median GDF15 levels.

Conclusions: Hepcidin levels were influenced by RBC transfusion history, but hepcidin levels appear to decrease over time in the RS subtype only. Interestingly, increase in hepcidin after transfusions was already visible early in follow-up, depending on the transfusional load and erythropoietic activity of the bone marrow. GDF15 concentration appeared to be most related to MDS subtype, functioning as a marker of ineffective erythropoiesis.