Page 96 - 2019_03-Haematologica-web

S. Park et al.
A
B
Figure 2. Duration of response to erythropoiesis-stimulating agents. Kaplan Meier curves showing duration of response (in months) according to (A) GDF-15 level (P=0.0008) and (B) hepcidin:ferritin ratio (P=0.01). Hepcidin:ferritin >9: red curve; hepcidin:ferritin ≤9: blue curve.
Discussion
The findings of this study suggest that epoetin zeta is as effective as other ESA3,4 in this group of patients with lower-risk MDS with relatively favorable characteristics (only 28% of them were transfusion-dependent and their median sEPO level was relatively low), with a HI-E response rate of 47.6% and a median duration of response of 26.1 months. Safety findings were consistent with the known safety profiles of epoetin alfa and darbepoetin.20,21 The two major results of our study are that a low hep- cidin:ferritin ratio and a high RED score were predictive of lower HI-E, while GDF-15 >2000 pg/mL and hepcidin:fer- ritin <9 could be predictive factors of shorter response duration.
To the best of our knowledge, this is the first time that iron homeostasis parameters and dyserythropoiesis have been observed to have a predictive value with regards to response to ESA.
Hepcidin, a 25-amino acid peptide, is produced mainly by hepatocytes and secreted into the plasma. This peptide lowers the amount of iron in the serum by inhibiting iron export by ferroportin, a membrane-bound cellular iron exporter present on macrophages and at the basolateral site of enterocytes, which release iron into the circulation. Hepcidin is suppressed by increased erythropoietic iron
Table 3. Multivariate analysis of predictors of duration of response to erythropoiesis-stimulating agents.
Variables Threshold values
Hepcidin:ferritin (x100) ≤9 >9 (ref)
HR on duration P of response ESA
(95% CI)
2.46 [1.23-5.30] 0.01 1.00
GDF-15 (pg/mL) >2000
3.47 [1.74-7.21]
1.00
0.0004
IPSS Intermediate
Low (ref)
1.73 [0.90-3.33] 0.09
1.00
≤2000 (ref)
GDF-15: growth-differentiation factor-15; IPSS: International Prognosis Scoring System.
demand and is upregulated in the presence of increased iron levels or elevated body iron stores.22,23 In this study, a tendency to lower hepcidin:ferritin ratios was found in RARS patients. Theoretically, hepcidin suppression should lead to an increase of iron transport across the cell mem- brane and, therefore, greater delivery of iron to transferrin for the needs of increased erythropoiesis, but as MDS patients have ineffective erythropoiesis, hepcidin suppres- sion actually results in iron overload and a potential toxic- ity to erythropoiesis with the Fenton reaction leading to an excess of reactive oxygen species particularly in RARS. This suggests that erythropoiesis and iron homeostasis are tightly related and that dyserythropoiesis, as measured by the RED score, influences iron equilibrium by suppressing hepcidin and increasing iron absorption.
The erythroid mediator GDF-15, produced in response to erythropoietin by erythroblasts, has been described as a potential suppressor of hepcidin expression.15 However, the correlation between GDF-15 and hepcidin levels was poor in studies in phlebotomized mice and in MDS patients.24,25 Recently, the role of GDF-15 in the regulation of hepcidin production in physiological and pathological settings has been challenged by the discovery of erythro- ferrone, which has been demonstrated to be the real sup- pressor of hepcidin.26 GDF-15 levels are elevated in tha- lassemia and congenital dyserythropoietic anemia, two disorders with a high degree of ineffective erythro- poiesis15,16,27 and correlate with the severity of anemia. GDF-15 level is, therefore, more the reflection of erythroid precursor activity and a useful indicator of ineffective ery- thropoiesis.
Taken altogether, hepcidin suppression by increased erythropoietic activity appears to underlie the develop- ment of abnormal iron overload in anemia, independently of RBC transfusions and is strongly correlated with worse erythroid response to ESA.
Hepcidin, routinely available in some countries, can be measured by mass spectrometry and international labora- tories are working to uniform hepcidin analysis.28,29 This assay could also be used for MDS studies. To conclude, our results could lead to the use of hepcidin measurements to predict response to ESA. Patients with a low hepcidin:ferritin ratio, high RED score and high GDF-15 level may require other treatments than ESA alone. These treatments should aim at improving the use of iron in MDS. We recently showed potential positive effects of low doses of deferasirox in MDS patients. It is postulated that this drug quenches levels of reactive oxygen species, leading to activation of nuclear factor-κB in erythroblasts,
502
haematologica | 2019; 104(3)