Darbepoetin-alpha for the anaemia of myelofibrosis with myeloid metaplasia


Francisco Cervantes, MD, Haematology Department, Hospital Clínic, Villarroel 170, 08036 Barcelona, Spain. E-mail: fcervan@clinic.ub.es


Darbepoetin-α, a novel hyperglycosylated erythropoiesis-stimulating protein, was administered to 20 patients with myelofibrosis with myeloid metaplasia and anaemia. The initial weekly dose, 150 μg, was increased to 300 μg when no response was observed after 4–8 weeks. Eight patients (40%) responded to treatment, including six complete and two partial responses, and five maintained their response at a median follow-up of 12 months (range 4–22). Univariate analysis indicated that older age was the only factor associated with a favourable response to treatment (P = 0.006). None of the patients with appropriate serum erythropoietin levels responded. Treatment was usually well tolerated.

Recombinant human erythropoietin (rHuEPO) can be useful in patients with myelofibrosis with myeloid metaplasia (MMM) and anaemia with inappropriate erythropoietin (Epo) serum levels (Cervantes et al, 2004). Darbepoetin-α, a novel hyperglycosylated erythropoiesis-stimulating protein, has greater in vivo activity and longer half-life than rHuEPO, allowing less frequent administration (Egrie & Brown, 2001). It is effective in the treatment of anaemia caused by renal failure (Toto et al, 2004), cancer (Pirker, 2004), chemotherapy (Waltzman, 2004) and myelodysplasia (Musto et al, 2005), but the experience in MMM is limited (Nguyen et al, 2003).

The present study was aimed at assessing the efficacy of darbepoetin-α in the anaemia of MMM.

Patients and methods

Patient characteristics and diagnostic criteria

From March 2004 to December 2005, darbepoetin-α (Aranesp, Amgen laboratories) was given to 20 MMM patients with Hb ≤10 g/dl (13 males, seven females; median age: 68 years, range 31–93 years). At treatment start, mean Hb was 8.8 ± 1.1 g/dl and eight patients had transfusion requirements. Median time between diagnosis and treatment was 24 months (range 1–190 months). Median serum Epo level was 48 U/l (range 3–1288 U/l). MMM followed essential thrombocythaemia in four patients and one had post-polycythaemic myelofibrosis. MMM diagnosis was based on standard criteria (Barosi et al, 1999).

Treatment schedule, response criteria and toxicity assessment

Based on our results with rHuEPO (Cervantes et al, 2004), the general rule was to administer darbepoetin-α as first-choice treatment to patients with inadequate Epo levels (<125 U/l) and danazol to the remainder. However, three patients received darbepoetin-α after failing to danazol despite having adequate Epo levels.

Following oral informed consent according to the guidelines of the local Ethical Committee, patients received a weekly subcutaneous dose of 150 μg of darbepoetin-α, which was increased to 300 μg when no Hb increase was noted at 4 weeks or a partial response had not been attained at 8 weeks. Treatment was discontinued in case of failure at 3 months. Oral iron supplements were given whenever inadequate iron status, as assessed by serum ferritin and soluble transferrin receptor levels, was observed. Packed red blood cells were transfused when the Hb dropped below 8.0 g/dl or the anaemia was symptomatic.

The response criteria were those of the European myelofibrosis network (Barosi et al, 2005). Complete response was considered as Hb normalisation (≥12 g/dl) and partial response as a Hb increase ≥2.0 g/dl in non-transfusion-dependent patients or a ≥50% reduction in transfusion requirements. The remaining situations were considered as failures.

Standard haematological and biochemical parameters were monitored and the possible adverse effects recorded.

Statistical methods

The capacity of several pre-treatment variables [including age, gender, lapse time between diagnosis and treatment, transfusion requirement, Hb value, white blood cell (WBC) and platelet counts, and Epo levels] to predict a response to darbepoetin-α was analysed using the χ2-test and the Student's t-test.


Darbepoetin-α was the first treatment option in six patients, while the remainder had received other therapies (hydroxyurea and danazol, seven cases each; anagrelide, n = 2; and splenectomy, n = 1). Some patients had received more than one therapeutic modality before darbepoetin-α, but either the previous treatment had been stopped due to failure or darbepoetin-α added because of inadequate response.

Median follow-up from darbepoetin-α start was 12 months (range 4–22 months). Eight patients (40%) showed a favourable response to treatment, including six complete and two partial responses (Table I). One of the eight responders had post-thrombocythaemic and other post-polycythaemic myelofibrosis. Median time to response was 2 months (range 1–3 months). In three patients the response followed a dose increase at 4 weeks. Three responders lost response at 2, 2.5 and 12 months, respectively, and five have maintained the response at 5, 6, 10, 12 and 14 months of achievement. In the latter patients, the maintenance dose was 150 μg weekly or every other week (two cases each) and 300 μg weekly (n = 1). At univariate analysis, older age was the only factor associated with a favourable response (P = 0.006, Table II). No patient with inappropriate serum Epo levels responded, whereas three of eight with transfusion requirements did. Treatment was well tolerated. One patient had a moderate increase in splenomegaly.

Table I.   Main clinico-haematological features in eight patients with myelofibrosis with myeloid metaplasia who responded to darbepoetin-α treatment.
Patient ×109/lAge (years)GenderPRBC pre-/ post-treatmentHb, g/dl, pre-/ post-treatmentPlatelets pre-/ post-treatment
  1. M, male; F, female; PRBC, number of packed red blood cells transfused per month; Hb, haemoglobin concentration.

Table II.   Univariate study of the pre-treatment variables associated with a favourable response to darbepoetin-α in 20 patients with myelofibrosis with myeloid metaplasia.
  1. M, male; F, female; WBC white blood cells; NS, not significant.

  2. *Mean ± SD.

Age (years)*75 ± 960 ± 120.006
Gender (M/F)6/27/5NS
Time diagnosis-treatment (months)33 ± 2253 ± 59NS
Transfusion requirements (yes/total)3/85/12NS
Hb (g/dl) concentration*9.2 ± 18.3 ± 1.7NS
Serum Epo (U/l)*32 ± 3296 ± 356NS
WBC count (×109/l)11 ± 139 ± 5NS
Platelet count (×109/l)212 ± 118237 ± 199NS

Among four additional patients that were switched from rHuEpo to darbepoetin-α, the two patients that were in response at the time of therapy switch maintained the response, while the two that failed to respond to rHuEpo also failed to respond to darbepoetin-α.


The anaemia of myelofibrosis is traditionally treated with androgens (Cervantes et al, 2005) or splenectomy in selected patients (Mesa & Tefferi, 2001). rHuEPO is effective in a proportion of patients, usually those with inappropriate Epo levels (Cervantes et al, 2004).

Darbepoetin-α is a novel hyperglycosylated erythropoiesis-stimulating protein with greater in vivo activity than rHuEPO, generated by recombinant DNA technology to permit delayed administration (Egrie & Brown, 2001). Its efficacy and safety have been demonstrated in the anaemia of renal failure (Toto et al, 2004), cancer (Pirker, 2004), chemotherapy (Waltzman, 2004) and myelodysplasia (Musto et al, 2005). Information in MMM concerns only two patients submitted to haemopoietic stem cell transplantation (Nguyen et al, 2003).

In the present study, darbepoetin-α produced favourable responses in 40% of the cases with anaemia of MMM, and was maintained in 20% of cases. This is in line with the response rate (45%) obtained with rHuEPO (Cervantes et al, 2004). It must be pointed out, however, that both studies involved a selected cohort of patients, most of whom had inappropriate Epo levels. Three responding patients needed an increased dose of darbepoetin-α to achieve the response. Older age was associated with favourable response, although the small size of the series precludes definitive conclusions. In keeping with the results of rHuEpo treatment (Cervantes et al, 2004), no patient with adequate Epo levels responded to darbepoetin-α. Moreover, among those patients that were switched from rHuEpo to darbepoetin-α, the responses were maintained in those responding to rHuEpo, while none of the rHuEpo non-responders responded to darbepoetin-α. Treatment was well tolerated.

In conclusion, darbepoetin-α is effective in the anaemia of MMM with inappropriate Epo levels. However, since treatment of the anaemia of MMM remains unsatisfactory, newer therapies, such as the immunomodulatory drugs or allogeneic stem cell transplantation, need to be attempted (Cervantes, 2005).