British lournal o/ Haernatology. 199 1, 77. 41 9-42 3 ADONIS 0007104891000685 The treatment of anaemia in the myelodysplastic syndromes with recombinant human erythropoietin DAVID BOWEN, DOMINIC CULLIGAN AND ALLAN JACOBS Department of Haematology, University of Wales College of Medicine, Cardifl Received 29 June 1990; accepted for publication 11 November I990 Summary. Recombinant human erythropoietin was admini- stered subcutaneously to 10 patients with myelodysplasia (MDS) who had haemoglobin concentrations less than 10 g/dl. in an attempt to relieve their anaemia. Doses of 60 units/kg/d rising to 90 units/kg/d were given over a maximum period of 16 weeks. Two out of 10 patients showed a steady rise in haemo- globin concentration during treatment. One patient with refractory anaemia had a sustained rise from 9.9 g/dl to 11.3 g/dl. and one patient with refractory anaemia with excess blasts (RAEB)had a rise from 9.5 g/dl to 11.4 g/dl but then relapsed with the development of an iron deficient state. Serum concentrations of immunoreactive EPO varied con- siderably between patients, but both responders had rela- tively low baseline levels. Both responders were also new diagnoses and had received no red cell transfusions. The criteria for response to recombi- nant human erythropoietin therapy, as well as the indica- tions for therapy remain to be clarified. The myelodysplastic syndromes (MDS) are a heterogeneous group of disorders characterized by clonal haemopoietic proliferation and the production of cells with abnormalities of maturation and function. In most cases progression leads to anaemia, neutropenia or thrombocytopenia, often in com- bination (Jacobs & Clark, 1986). Present management strategies are limited to supportive care for peripheral cytopenias, and the use of cytotoxic or differentiating agents in an attempt to modify the abnormal proliferation and maturation. In exceptional cases bone marrow transplanta- tion may be indicated. There is significant morbidity from anaemia, infection and bleeding, with death eventually resulting from overt bone marrow failure or leukaemic transformation. Recombinant human erythropoietin (rhEPO) is now well established as treatment for the anaemia of chronic renal disease in pre-dialysis patients (Lim et al. 1989) and patients on haemodialysis or continuous ambulatory peritoneal dialy- sis (CAPD) (Winearls et al. 1986; Eschbach et al. 1987; Adamson & Eschbach. 1989). Therapy is usually admini- stered intravenously to patients receiving haemodialysis. However, the subcutaneous route was investigated to pro- vide greater convenience for patients on CAPD. Lower doses of EPO can be used subcutaneously than intravenously with an equal erythropoietic stimulus (Hughes et al, 1990; Correspondence: Dr D. Bowen. Department of Haematology. Univer- sity of Wales College of Medicine, Heath Park, Cardiff CF4 4XN. Bommer et al. 1988). although the pharmacokinetic profiles of EPO administered by the two routes are quite different (MacDougaIl et al. 1989a). Studies of therapy with rhEPO in other anaemias are now in progress. Preliminary results suggest that some patients with chronic anaemias of rheumatoid arthritis (Means et al, 1989), disseminated malignancy (Henry et al. 1989), mul- tiple myeloma (Ludwig et al. 1990), and anaemia associated with Zidovudine therapy for AIDS (Fischl et al, 1990) will respond to rhEPO therapy and will tolerate doses as high as 4500 u/kg/week with no side effects (Fischl et al. 1990). The following study reports on subcutaneous rhEPO therapy in 10 patients with MDS. PATIENTS 10 adult patients, five male and five female aged 23-79 years (mean 64 years) were studied. All had MDS as defined by the FAB classification (Bennett et al. 1982): four patients with refractory anaemia (RA). four with sideroblastic anaemia (SA). and two with refractory anaemia with excess blasts (RAEB). All patients had a steady state haemoglobin con- centration < 10 g/dl or were transfusion dependent (defined as >600 ml transfused in the previous 6 months). 4/10 patients received red cell transfusions monthly, 3/10 less frequently, and 3/10 had received no transfusions. 3/10 patients were studied within 6 months of diagnosis. The remainder had been followed as part of our cohort of MDS 419
The treatment of anaemia in the myelodysplastic syndromes with recombinant human erythropoietin
DAVID BOWEN, DOMINIC CULLIGAN A N D ALLAN JACOBS
Department of Haematology, University of Wales College of Medicine, Cardifl
Received 29 June 1990; accepted for publication 11 November I990
Summary. Recombinant human erythropoietin was admini- stered subcutaneously to 10 patients with myelodysplasia (MDS) who had haemoglobin concentrations less than 10 g/dl. in an attempt to relieve their anaemia. Doses of 60 units/kg/d rising to 90 units/kg/d were given over a maximum period of 16 weeks.
Two out of 10 patients showed a steady rise in haemo- globin concentration during treatment. One patient with refractory anaemia had a sustained rise from 9.9 g/dl to 11.3 g/dl. and one patient with refractory anaemia with
excess blasts (RAEB) had a rise from 9.5 g/dl to 11.4 g/dl but then relapsed with the development of an iron deficient state. Serum concentrations of immunoreactive EPO varied con- siderably between patients, but both responders had rela- tively low baseline levels.
Both responders were also new diagnoses and had received no red cell transfusions. The criteria for response to recombi- nant human erythropoietin therapy, as well as the indica- tions for therapy remain to be clarified.
The myelodysplastic syndromes (MDS) are a heterogeneous group of disorders characterized by clonal haemopoietic proliferation and the production of cells with abnormalities of maturation and function. In most cases progression leads to anaemia, neutropenia or thrombocytopenia, often in com- bination (Jacobs & Clark, 1986). Present management strategies are limited to supportive care for peripheral cytopenias, and the use of cytotoxic or differentiating agents in an attempt to modify the abnormal proliferation and maturation. In exceptional cases bone marrow transplanta- tion may be indicated. There is significant morbidity from anaemia, infection and bleeding, with death eventually resulting from overt bone marrow failure or leukaemic transformation.
Recombinant human erythropoietin (rhEPO) is now well established as treatment for the anaemia of chronic renal disease in pre-dialysis patients (Lim et al. 1989) and patients on haemodialysis or continuous ambulatory peritoneal dialy- sis (CAPD) (Winearls et al. 1986; Eschbach et al. 1987; Adamson & Eschbach. 1989). Therapy is usually admini- stered intravenously to patients receiving haemodialysis. However, the subcutaneous route was investigated to pro- vide greater convenience for patients on CAPD. Lower doses of EPO can be used subcutaneously than intravenously with an equal erythropoietic stimulus (Hughes et al, 1990;
Correspondence: Dr D. Bowen. Department of Haematology. Univer- sity of Wales College of Medicine, Heath Park, Cardiff CF4 4XN.
Bommer et al. 1988). although the pharmacokinetic profiles of EPO administered by the two routes are quite different (MacDougaIl et al. 1989a).
Studies of therapy with rhEPO in other anaemias are now in progress. Preliminary results suggest that some patients with chronic anaemias of rheumatoid arthritis (Means et al, 1989), disseminated malignancy (Henry et al. 1989), mul- tiple myeloma (Ludwig et al. 1990), and anaemia associated with Zidovudine therapy for AIDS (Fischl et al, 1990) will respond to rhEPO therapy and will tolerate doses as high as 4500 u/kg/week with no side effects (Fischl et al. 1990).
The following study reports on subcutaneous rhEPO therapy in 10 patients with MDS.
PATIENTS
10 adult patients, five male and five female aged 23-79 years (mean 64 years) were studied. All had MDS as defined by the FAB classification (Bennett et al. 1982): four patients with refractory anaemia (RA). four with sideroblastic anaemia (SA). and two with refractory anaemia with excess blasts (RAEB). All patients had a steady state haemoglobin con- centration < 10 g/dl or were transfusion dependent (defined as >600 ml transfused in the previous 6 months). 4/10 patients received red cell transfusions monthly, 3/10 less frequently, and 3/10 had received no transfusions. 3/10 patients were studied within 6 months of diagnosis. The remainder had been followed as part of our cohort of MDS
419
420 D. Bowen, D. Culligan and A. Jacobs
v'
patients for at least a year prior to study. Informed consent was obtained and the study approved by the South Glam- organ Joint Ethics Committee.
Therapystopped 4 Therapy recommenced
METHODS
Recombinant human W O (Boehringer Mannheim Gmbh) was administered once daily subcutaneously for 6 d per week. The initial dose was 60 units/kg/d rising to 90 units/kg/d if there was no response in the first 6 weeks. A response was defined as a rise in haemoglobin concentration of > 1 g/dl and/or a reduction in the average monthly transfusion requirement during the treatment period compared to the 16 weeks pre-study. Treatment was stopped if the haemo- globin concentration rose to > 11 g/dl for 2 consecutive weeks and failing this, at 16 weeks therapy.
Weekly full blood counts, reticulocyte counts (Sysmex R- 1000: Sysmex-Toa Electronics Company) and blood pres- sure measurements were performed, and samples taken monthly for biochemistry profile, liver function tests, periph- eral blood progenitor assays (granulocyte/macrophage col- ony forming unit, CFU-GM: erythroid burst-forming unit, BFU-E), serum ferritin. serum iron and total iron binding capacity. and serum EPO concentration. Serum was stored at - 3OoC and batched for P O radioimmunoassay (Eckhardt et al, 1988). Bone marrow morphology, cytogenetics and red cell mass were assessed before and after therapy. All post therapy investigations were performed the day after treat- ment was stopped.
RESULTS
Nine out of 10 patients showed no change in haemoglobin concentration or transfusion requirements at the end of
Table I. Pre- and post-treatment haemoglobin concentration (Hb). peripheral blood erythroid progenitors (BFU-E), and transfusion requirements
Mean units Hb BFU-E transfused (g/dl)* (/ml blood) per month
* Haemoglobin concentrations are the lowest pre-transfusion values immediately prior to and after therapy.
Hb (gldl)
12 1 Retics x 10 (g/l)
r 70
0 4 8 12 16
Week of therapy
Fig 1. Haemoglobin concentration (0) and reticulocytes (A) in patient 9 treated with rhEPO therapy.
Hb (g/dl) Retics
131
16 weeks treatment (Table I). Two patients showed a steady increase in haemoglobin concentration above baseline (Figs 1 and 2). Neither of these two were previously transfusion dependent. Patient 9 (RAEB) had a rise in haemoglobin concentration from 9.5 g/dl to 11.4 g/dl at 6 weeks, but failed to maintain this, and thereafter declined steadily to pre- treatment values at 16 weeks despite an increase in r h P O dosage. Patient 10 (RA) showed a rise in haemoglobin concentration from 9.9 g/dl to 11.3 g/dl at week 5 and 11.2 g/dl at week 6, and therapy on study was therefore stopped. This was however recommenced on a 'compassion- ate need' basis at week 8. Mean reticulocyte counts of the group were not significantly changed on therapy. However, in patient 9 a clear reticulocytosis was seen initially, returning to basal level by week 4 despite a continuing rise in haemoglobin concentration (Fig 1). Unfortunately a pre- treatment reticuloycte measurement was not available for patient 10, but post treatment a fall from week 1 to week 5 was observed in the presence of a continuously rising haemoglobin concentration (Fig 2).
Red cell mass was measured before and at the cessation of therapy and in no case was there an unequivocal change.
Treatment of Anaemia with rhEPO in MDS 42 1 Table 11. Serum erythropoietin concentration in patients with MDS on rhEPO therapy
Mean serum immunoreactive EPO concentration pre-treat- ment was 555 mU/ml (range 47-2350 mU/ml). There was little variation in serum EPO through the course of the study except in relation to blood transfusion (Table 11). although mean serum EPO at the end of the study had risen to 1123 mll/ml (range 78-3522 mU/ml). The two responding patients had concentrations of 4 9 and 199 mU/ml res- pectively before treatment and the highest concentration recorded in these patients were 9 4 and 382 mU/ml. Peri- pheral blood CFU-GM and BFU-E did not change significantly on treatment although the two responders showed a small rise in BFU-E at week 4 compared with pre-treatment values (data not shown). Erythroid progenitor numbers were still however subnormal in these two (Table I). Mean leucocyte count and mean platelet count did not alter significantly on therapy.
The decline in haemoglobin concentration in patient 9 corresponded with a fall in ferritin. serum iron, and per cent transferrin saturation to iron deficient levels (Table 111). Patient 10 also showed a fall in serum ferritin during
treatment, with a more modest reduction in serum iron and transferrin saturation. None of the non-responders were iron deficient pre-treatment and none showed a fall in serum iron or ferritin on treatment.
No adverse effects of rhEPO therapy were observed and no changes in renal or hepatic function or mean arterial blood pressure were seen. There was no disease progression on therapy as assessed by blood counts, bone marrow morpho- logy, or cytogenetics. One of our responders (patient 10) had a normal karyotype throughout study, but the other (patient 9) showed a complex karyotype [10/20 cells with +8, -7. -12, -16. -17, -22 , +dic(12;22)(p11.1:p12). +der(l7)t(7:17)(pl1.2:pl1.2), del (5) (q15q33). 18p+] which also remained unchanged on therapy. It is interesting to note that a complex karyotypic abnormality does not appear to preclude a response to rhEPO in MDS patients.
Patient 9 has transformed to acute myeloid leukaemia 6 months post study and patient 10 remains on rhEPO therapy at 5000 units twice per week subcutaneously with iron supplementation to maintain her haemoglobin concen-
422 D. Bowen, D. Culligan and A. Jacobs tration at > 13 g/dl (Fig 1). Red cell mass has now increased inpatient 10 from 13*5ml/kgpre-treatmentto 19.3ml/kgat week 28 of therapy.
DISCUSSION Subcutaneous therapy is clearly an effective treatment for the anaemia of MDS in selected patients and was convenient and well tolerated in our study. The effkacy of intravenous rhEPO therapy in MDS has already been demonstrated (Hirashima et al. 1989). Despite the lower bioavailability of rhEP0 administered subcutaneosly, it is thought to provide a more sustained stimulus to erythropoiesis than the peaks and troughs of the intravenous route (Bommer et al, 1988). The erythropoietic response (as measured by the erythron trans- ferrin uptake) to subcutaneous rhEPO, is similar to that of intravenous therapy in patients with the anaemia of chronic renal failure (Hughes et al, 1989).
Only two small studies of rhEPO therapy in MDS patients have previously been reported. Urabe et al(1988) found an increase in bone marrow erythroblasts and reticulocytes in 3/4 patients studied, with an increase in haematocrit in only 1/4. Hirashima et al(l989) found 3/8 patients responded to treatment as measured by a reduction in transfusion require- ment in two and an increase in haemoglobin concentration in one. Higher doses of rhEPO were administered than in our study, and the intravenous route three times weekly was preferred. Basal serum EPO concentration was higher than in our responders and interestingly a marked fall in serum EPO concentration was observed following an increase in haemo- globin concentration in one patient. A similar fall in serum EPO concentration has been described following an erythroid response to recombinant granulocyte-macrophage colony- stimulating factor in MDS patients (Vadhan-Raj et al, 1990).
Serum erythropoietin levels in MDS are variable between patients at similar haemoglobin concentrations (Jacobs et al, 1989). It is not therefore surprising to 6nd variable responses to rhEPO therapy amongst MDS patients. In our study one patient had a sustained increase in haemoglobin concentra- tion and left the study early. The other responding patient developed overt iron deficiency and failed to sustain his elevated haemoglobin concentration. Functional iron defi- ciency as a cause of non response to r W O is well recorded in renal anaemia (Eschbach et al, 1987; Machugall et al, 1989b). Non-responders in our study had transferrin satura- tions of > 50% and serum ferritin >400 pg/l prior to therapy. Iron deficiency was therefore unlikely to have contributed to their failure to respond.
Patient 9 responded to rhEPO therapy with an increased erythroid output as measured by a transient reticulocytosis and rise in haemoglobin concentration. This reticulocytosis is too prolonged to be due simply to shift from the marrow reticulocyte population into the circulation. A reticulocytosis in patient 10 could not be c o n h e d due to the lack of pre- treatment data. Despite this measured early increase in erythroid output, neither patient showed a significant in- crease in circulating erythroid progenitors. Red cell mass post treatment was unchanged in both responders perhaps due to the development of functional iron deficiency and subsequent
fall in haemoglobin in patient 9 by the end of the study. and to the early measurement (week 5) of red cell mass after a rapid response necessitating study termination in patient 10. Subsequently an increase in red cell mass has been docu- mented in patient 10 as stated above.
Although a study with small numbers does not enable identification of those patients who are likely to respond best to therapy, the two responding patients were notable for a number of reasons. Both patients (a) had never received red cell transfusion, (b) had relatively low serum ferritin ( < 200 pg/l) pre-treatment, (c) were newly diagnosed, (d) showed detectable circulating BFU-E, and (e) had serum EPO concen- tration pre-treatment < 200 mU/ml.
In at least some patients with myelodysplasia EPO sensitive progenitors are presumably present which are capable of responding to stimulation by modest doses of rhEP0 with an increase in haemoglobin concentration. Although a consis- tent increase in circulating BFU-E numbers on treatment could not be demonstrated, it is interesting that the two responding patients were among the 5/10 patients with detectable basal peripheral blood BFU-E and indeed only one patient had higher pre-treatment values, massively raised at 699 BFU-E/ ml blood. Detectable bone marrow CFU-E pre- treatment was a common feature of responders in another similar study (Hirashima et al, 1989). It may be that our two responders, at a relatively early stage of the disease process, retained residual normal progenitors yet to be inhibited or suppressed by expansion of the abnormal MDS clone, or that progenitors from the abnormal clone retain EPO responsive- ness.
Basal serum EPO concentration was relatively low in our two responders and it is notable that responders to rhEPO therapy for the anaemia of myeloma (Ludwig et al. 1990). and AIDS patients on zidovudine (Fischl et al, 1990) had significantly lower basal serum EPO than non responders. We cannot. however, assume that serum immunoreactive EPO levels necessarily reflect biological activity, and the effect of the relatively modest increment in serum EPO through our study in the responding patients at the level of the WO responsive progenitors is unknown. A larger study is clearly required to further clarify features predictive of response to rhEP0 therapy in MDS.
ACKNOWLEDGMENTS
We would like to thank Dr Freda Houghton (MCP Phar- maceuticals) for supplying Recombinant Human Erythro- poietin, and Drs Ehmer and Neubert (Boehringer Mannheim) for the serum EPO assays.
REFERENCES
Adamson, J.W. & Eschbach, J.W. (1989) Management of the anaemia of chronic renal failure with recombinant erythropoietin. Quarlerlu Iournal of Medicine N.S. 73, 1093-1 101.
Bennett. J.M.. Catovsky. D.. Flandrin, G.. Galton. D.A.G., Gralnick, H.R. & Sultan, C. (1982) Proposals for the classification of the myelodysplastic syndromes. British lournal of Haematology, 5 1, 189-199.
Treatment of Anaemia with rhEPO in MDS 423 Bommer. J., Ritz. E.. Weinrich. T.. Bommer G.. Ziegler. T. (1988)
Subcutaneous erythropoietin. Lancet. ii, 406. Eckardt. K-U. . Kurtz. A.. Hirth. P.. Scigalla. P.. Wieczorek. L. & Bauer,
C. (1 988) Evaluation of the stability of human erythropoietin in samples for radioimmunoassay. Klinische WochenschriJt, 66,
Eschbach. J.W.. Egrie. J.C.. Downing, M.R.. Browne. J.K. & Adamson, J.W. (1987) Correction of the anaemia of end-stage renal disease with recombinant human erythropoietin. New England Iournal of Medicine. 316, 73-78.
Fischl. M.. Galpin. J.E.. Levine. J.D.. Groopman. J.E.. Henry. D.H.. Kennedy, P.. Miles, S.. Robbins. W.. Starrett. B.. Zalusky. R., Abels, R.I.. Tsai. H.C. & Rudnick. S.A. (1990) Recombinant human erythropoietin for patients with AIDS-treated with Zidovudine. New England /ournu1 of Medicine. 322, 1488-1493.
Henry, D.H.. Rudnick. S.A. & Bryant, E. (1989) Preliminary report of two double blind placebo controlled studies using recombinant human erythropoietin in the anaemia associated with cancer. Blood. 74, Suppl. 1. 6a.
Hirashima. K.. Bessho. M.. Susaki. L., Miyazawa. K. & Nagashima, M. (1989) Improvement of anaemia by intravenous injection of recombinant erythropoietin in patients with myelodysplastic syndrome and aplastic anaemia. Experimental Haematology. 1 7, 385.
Hughes, R.T.. Cotes, P.M.. Oliver, D.O.. Pippard. M.J.. Royston. J.P., Stevens, J.M.. Strong, C.A.. Tam, R.C. & Winearls. C.G. (1989) Correction of the anaemia of chronic renal failure with erythro- poietin (r-HuEPO): pharmacokinetic studies during haemodialysis and C.A.P.D. Contributions to Nephrology. 76, 122-1 30.
Hughes, R.T.. Cotes, P.M., Pippard. M.J.. Stevens, J.M.. Oliver, D.O.. Winearls. C.G. & Royston, J.P. (1990) Subcutaneous admini- stration of recombinant human erythropoietin to subjects on continuous ambulatory peritoneal dialysis: an erythro-kinetic assessment. British journal of Haematolog~. 75. 268-273.
Jacobs, A. &Clark, R. (1986) Pathogenesis and clinical variations in
241-245.
the myelodysplastic syndrome. Clinics in Haematology. 1 5,
jacobs. A.. Janowska-Wieczorek. A.. Caro. J.. Bowen. D.T. & Lewis. T. ( 1 989) Circulating erythropoietin in patients with myelodys- plasia. British Journal of Haematology. 73. 36-39.
Lim. V.S.. DeGowin. R.L.. Zavala, D.. Kirchner, P.T.. Abels. R., Perry, P. & Fangman. J. (1989) Recombinant human erythropoietin therapy in predialysis patients. Annals of Internal Medicine. 110.
Ludwig, H., Fritz. E.. Kotzmann. H.. Hocker, A.. Gisslinger, H. & Barnes. U. (1990) Erythropoietin treatment of anemia associated with multiple myeloma. New England Journal of Medicine. 322,
Machugall. I.C., Roberts. D.E.. Neubert. P.. Dharmasena. A.D.. Coles. G.A. &Williams. J.D. (1989a) Pharmacokinetics of recombi- nant human erythropoietin in patients on continuous ambulatory peritoneal dialysis. Lancet. i, 425-427.
Machugall, I.C.. Hutton, R.D., Cavill. I.. Coles. G. & Williams, J. (1 989b) Poor response to treatment of renal anaemia corrected by iron given intravenously. British Medical Journal. 299, 157-1 58.
Means, R.T.. Olsen. N.J.. Krantz. S.B., Dessypris. E.N.. Graber. S.E.. Stone, W.J.. ONeil. V.L. & Pincus. T. (1989) Treatment of the anaemia of rheumatoid arthritis with recombinant human erythropoietin: clinical and in vitro studies. Arthritis and Rheuma-
Urabe. A., Kitamura, T. & Takaku, F. (1988) Clinical application of erythropoietin. Acta Haematologica (japonica). 51, 13 55-1 359.
Vadhan-Raj, S.. Hittleman. W.N.. Lepe-Zuniga, J.. Gutterman. J.U. & Broxmeyer. H.E. (1 990) Regulation of endogenous erythropoietin levels in anemia associated with myelodysplastic syndromes. New England journal ofMedicine, 75, 1749-1 750.
Winearls. C.G.. Oliver, D.O.. Pippard. M.J.. Reid, C.. Downing. M.R. & Cotes, P.M. ( 1 986) EfTect of human erythropoietin derived from recombinant DNA on the anaemia of patients maintained by chronic haemodialysis. Lancet. ii, 1 1 75-1 178.
