Anemia is one of the main clinical problems for patients with low-risk myelodysplastic syndromes (LR-MDS). The evolution of the disease results in transfusion dependence and iron overload. The addition of erythropoietin (EPO) to the in vitro culture of erythroid colonies leads to an increased formation of colonies, particularly when it is associated with other hematopoietic factors. Recombinant human erythropoietin (rHEPO, epoetin alfa) and darbepoetin alfa, both prescribed alone or in combination with granulocytic colony stimulating factor (G-CSF), have been extensively shown to improve erythropoiesis and reduce transfusion requirements in patients with anemia and MDS. The Scandinavian group validated a predictive response scheme for erythropoietic agents in which serum EPO (sEPO) levels and transfusion requirements predict response to these drugs1,2 (B).

A retrospective study evaluated MDS patients who were submitted to red blood cell (RBC) transfusions [hemoglobin (Hb) <8.5g/dL]. Of 192 patients, 83 patients received rHEPO treatment for at least 12 weeks, and of these 24 patients (28.9%) achieved an erythroid response according to the International Working Group (IWG) 2006 criteria. The response to rHEPO was significantly associated with lower levels of endogenous EPO, lower percentage of blasts in bone marrow and less transfusion requirements. The median duration of response was 17 months. There was no difference between treated and untreated patients in terms of progression to acute myeloid leukemia (AML) and other causes of death (infections, bleeding, secondary tumors, cardiovascular and thromboembolic events) and no death could be attributed to rHEPO. No difference was found in the median overall survival (OS) between patients who received and those who did not receive treatment (36 vs. 38 months, respectively). However, a longer OS was found in responders as compared to non-responders (mean 42 vs. 31 months, respectively; p-value <0.009) and the response to rHEPO remained an independent prognostic value for OS in the multivariate analysis3 (B).

In a prospective study to evaluate the efficacy of fixed doses of epoetin alfa, 55 patients with a median age of 78 years, low or intermediate-1 International Prognostic Scoring System (IPSS) risk and Hb <10g/dL received subcutaneous rHEPO (40000IU) once weekly for at least 3 months. After a 12-week period, 65.5% of patients achieved an erythroid response (IWG 2006 criteria). Considering the transfusion requirement, rHEPO led to improvements in anemia in 27 of the 33 patients without prior transfusions and in nine of the 22 patients with prior transfusions. Serum EPO levels <200mU/mL were associated with a better response rate to rHEPO treatment. There was a trend toward a higher response rate in patients with lower risk (by IPSS and the World Health Organization Classification-Based Prognostic Scoring System). Treatment was well tolerated with minor side effects being observed in 7.2% of patients; discontinuation of treatment was required in only one patient due to intractable pruritus4 (B).

In a biosimilar study of epoetin alfa, 24 over 65-year-old patients with LR-MDS were treated with 40000IU once a week for 12 weeks. No major side effects were observed. Sixteen patients (66.6%) achieved erythroid response and all of them except for one became transfusion-independent for at least 3 months. Responders had significantly higher levels of Hb and less transfusion requirements than non-responders. There was also a positive correlation between the Hb levels and the quality of life (QoL) of these patients [Functional Assessment of Cancer Therapy-Anemia score (FACT-An): p-value <0.003; Mini Mental State Examination (MMSE) evaluation: p-value <0.01]5 (B).

Sixty-two patients, mostly with low and intermediate-1 IPSS risk MDS and endogenous EPO levels <500mU/mL, received darbepoetin 300mg subcutaneously once weekly (considered equivalent of 20000U of recombinant epoetin alfa three times per week). Forty-four patients (71%) presented erythroid response. There was no response in the platelet or neutrophil lineages. The median time from the onset of treatment to response was 4 weeks. In 11 patients, the treatment had to be discontinued temporarily due to Hb levels above 14g/dL. Eight of the 13 patients who had been unsuccessfully treated with recombinant EPO alfa or beta responded to darbepoetin. Ten of the patients who did not respond to darbepoetin alone received G-CSF for 12 weeks with only two responding. The overall response rate was 74%6 (B).

Thirty-seven patients (23 transfusion-dependent) with low IPSS risk and anemia received darbepoetin (150μg) subcutaneously once a week. Erythroid responses were detected in 40.5% patients with a stable Hb level >9.5g/dL being seen in most after 7–22 months of treatment. Serum EPO levels <100mU/mL, no excess of blasts in the bone marrow, low or no RBC transfusion requirements and hypoplastic bone marrow were confirmed as predictors of erythroid response in multivariate analyses. No significant side effects were reported during the study7 (B).

Ten studies (647 patients) were included in a recently published meta-analysis to evaluate the efficacy and safety of darbepoetin. The erythroid response rate varied from 38 to 72%; the response duration was from 12 to more than 51 months. Better responses were observed in patients with sEPO <100mU/mL and in patients not previously treated with erythropoiesis-stimulating agents. Higher hemoglobin levels, higher doses, transfusion independence and lower risk disease were associated with better response rates8 (A).

A meta-analysis involving rHEPO-treated MDS patients (22 studies; 925 patients) or patients who received darbepoetin alfa (8 studies; 389 patients) evaluated erythroid response using the IWG criteria (rHEPO: 584 patients; Darbepoetin: 389 patients). Univariate analysis showed a better erythroid response for patients with refractory anemia or refractory anemia with ring sideroblasts (RA/RARS; p-value <0.001), low sEPO levels and the use of a fixed-dose regimen as opposed to a weight-based regimen. There was no difference in erythroid response rates between rHEPO and darbepoetin (57.6 vs. 59.4%, respectively; p-value=0.828)9 (A).

Anemic patients with LR-MDS were randomized to receive rHEPO (30000IU weekly; n=15) or the same dosage of rHEPO plus G-CSF (300μg twice weekly; n=15) for at least 8 weeks. An erythroid response was observed in 40% of patients treated with r-HEPO alone compared to 73.3% of patients receiving the combined regimen, but there was no statistical difference between the groups. Taking into consideration the need for transfusions, rHEPO was effective in improving anemia in 55.5% of previously non-transfused patients and in 16.6% of previously transfused patients, whereas the combined treatment induced a favorable response in 80 and 60% non-transfused and previously transfused patients, respectively. No relevant side effects were reported in the two groups. After 4 months of treatment, the response rate was 33% for rHEPO and 62.5% for rHEPO plus G-CSF (p-value=0.032). QoL was assessed using the FACT-An questionnaire at the beginning of treatment and after 8 and 16 weeks. There was an apparent correlation between erythroid response and QoL10 (B).

In a study conducted by the French group, 95 patients with LR-MDS received darbepoetin alfa (500μg) subcutaneously every 2 weeks. For non-responders at 12 weeks, filgrastim was added to the treatment (300μg twice weekly; this dose was subsequently adjusted to maintain white cell count between 5.0 and 10.0×109L–1) and maintained for an extra 12 weeks. Erythroid response according to the IWG criteria at 12 weeks was achieved in 48% of patients. The addition of filgrastim in non-responders increased the overall response rate to 56% at 24 weeks. Improvement in the QoL assessed by FACT-An was seen in responders, but no difference was detected in physical performance. With a mean follow-up of 52 months, mean response duration was not reached and the 3-year cumulative incidence of AML and OS were 14.5 and 70%, respectively. The sEPO level was the only independent predictor of response at 12 weeks with the most discriminant cutoff value being 100mU/mL11 (B).

Immunosuppressive therapy, which includes antithymocyte globulin, cyclosporin A and alemtuzumab, has been used based on an observation of abnormal T-cell function in patients with LR-MDS, increased levels of proinflammatory cytokines and apoptosis of bone marrow cells. The National Comprehensive Cancer Network (NCCN) guidelines list five characteristics that may indicate the possibility of response to immunosuppressants: IPSS score <1, age <60 years, hypocellular bone marrow, presence of the paroxysmal nocturnal hemoglobinuria (PNH) clone and expression of HLA-DR15.

In a retrospective study, 29 patients with LR-MDS were treated with cyclosporin A alone (2–6mg/kg/day to maintain blood concentration at 150–250ng/mL) or associated with antithymocyte globulin (n=5). Eight patients (27.6%) had complete remission and nine (31.0%) achieved some hematological improvement resulting in an overall response rate of 58.6%. The median OS was 8.6 years with OS at 5 and 10 years of 74.5 and 48.3%, respectively. OS rate of responders at 5 and 10 years was 100 and 72.7%, respectively, was better than in non-responders (41.7 and 15.6%, respectively; p-value <0.001)12 (B).

Seventy-one patients with LR-MDS received either antithymocyte globulin therapy for 4 days (dose 4mg/kg/day) followed by cyclosporin A (3–5mg/kg/day) for 3 months or cyclosporin A alone. The total hematological response rate according to the IWG 2006 criteria was 77.5% (55/71 cases) with 11 complete responses. Of the responders, 77% had an erythroid response with the median increase in Hb being 3.6g/dL. Of the 71 cases, 60 (84.5%) were alive at the end of the 24-month follow-up period. During this follow-up, 69 patients remained without increased blast cell counts or leukemic transformation13 (B).

Thrombocytopenia occurs in 37–65% of individuals with MDS and can result in hemorrhagic complications, including mortality in around 24% of cases. In addition, many of the agents used to treat MDS can cause or exacerbate thrombocytopenia. Romiplostim is an agent that binds and activates thrombopoietin receptors, stimulating platelet formation through a mechanism similar to that of endogenous thrombopoietin.

Twenty-nine patients with MDS receiving decitabine were randomized to receive either romiplostim (750μg; n=15) or placebo (n=15). The group receiving romiplostim did not present a reduction in major bleeding events compared to the Placebo Group but presented a reduction in the need for transfusions and in the number of transfused units. The drug was well tolerated and only one patient in each group progressed to AML14 (A).

The results of the first phase of a blinded, randomized, controlled phase II study on the use of eltrombopag vs. placebo for LR-MDS patients with thrombocytopenia were recently published. Ninety patients were analyzed with 59 being allocated to the Eltrombopag Group. The median follow-up to evaluate the platelet response was 11 weeks with platelet response being reported in 47% of the Eltrombopag Group vs. 3% in the Placebo Group (p-value=0.0017). There were more bleeding events in the Placebo Group (42 vs. 14%; p-value=0.0025). Progression to AML or disease progression occurred in 12% of the Eltrombopag Group vs. 15% of the Placebo Group (p-value=0.81)15 (B).

Hypomethylating agents [azacitidine (AZA) or decitabine (DAC)] have a clear role in the treatment of high-risk MDS. These drugs have also been used in patients of lower risk who are transfusion-dependent and refractory to erythropoiesis-stimulating agents. In a retrospective study of the Italian group, 74 patients with LR-MDS, 84% of whom were transfusion dependent, were treated with different AZA regimens giving an overall response rate of 45.9% with hematologic improvement in 20.3%. Responders had a better OS of 94% vs. 54% at 2.5 years (p-value <0.0014). Grade 3 or 4 adverse events included myelosuppression (21.6%) and infection (6.8%)22 (B).

Three prospective studies have been published on the use of hypomethylating agents in LR-MDS, two with AZA23,24 (B) and one with DAC25 (B). The hematological response rates ranged from 10 to 60%. In the Nordic study with 30 patients, 20% became transfusion independent with a short median response and high complication rate24 (B). A randomized phase II study was published as an abstract. The trial compared AZA (75mg/m2) for 5 days (n=20) vs. supportive care (n=20) in patients with LR-MDS who did not have the del(5q) and had no response or were refractory to erythropoiesis-stimulating agents. Using the IWG 2006 criteria, a higher erythroid response rate was observed in patients in the AZA arm (43.7 vs. 5.5%). Moreover, transfusion independence was observed in 31% of cases of the AZA arm with 53% of adverse events (myelotoxicity)26 (B).

Iron overload in patients with MDS is due to recurrent transfusions, increased iron absorption and ineffective erythropoiesis. Iron overload leads to organ damage, particularly the liver, heart and endocrine glands. Patients with transfusion-dependent MDS have lower OS and leukemia-free survival.

The use of deferasirox in the clinical practice was investigated in a prospective, multicenter, observational, non-interventional study. Data of 123 patients with MDS who had not received iron chelators were compared with 44 patients who had received iron chelators. A reduction in serum ferritin levels was seen in treatment-naive patients, but the decrease was not significant in those previously treated with chelators. Side effects were documented in 34.1% of the patients in the treatment-naive group and in 40.9% of the previously chelated patients. Most of the side effects were diarrhea, nausea, increased serum creatinine levels and rash of low to moderate severity27 (B).

One hundred and seventy-three patients with low- to intermediate-risk MDS who had received at least 20 units of RBC and who had serum ferritin levels >1000μg/L were selected to receive an initial dose of 20mg/kg/day of deferasirox with subsequent doses being readjusted to a maximum of 40mg/kg/day based on the serum ferritin level, patient weight and serum creatinine. Efficiency was assessed by comparing ferritin levels at baseline and at 1, 2 and 3 years. In the first year, mean ferritin fell by 23.2% (n=91), by 36.6% at the end of the second year and by 36.5% at the end of the third year. Of the 173 patients, 51 (28%) presented hematological improvements according to the IWG 2006 criteria. During the 3-year course, side effects, death and problems to administer the drug were responsible for a 79.7% rate of discontinued treatment. The main adverse effects were related to gastrointestinal symptoms28 (B).

Patients with LR-MDS (n=33) and HR-MDS/AML (n=27) and serum iron levels of at least 1000ng/mL were treated with deferasirox at a dose of 20mg/kg/day for at least 1 year. An absolute reduction in ferritin levels compared to baseline values was obtained in the patients with LR-MDS. The most common side effects were related to the gastrointestinal tract29 (B).

No study was eligible for inclusion in a systematic review of randomized clinical trials conducted to evaluate the use of deferasirox in patients with MDS30 (A). However, a meta-analysis of eight observational studies with different types of chelators involving 1562 participants (median participants per study: 153; range: 78–534) showed a trend toward longer OS in patients who took chelators particularly those with LR-MDS. On average, patients who received chelators survived 5 years longer than those who did not use this drug31 (B).