Erythropoietin response in anaemic patients with multiple myeloma and other lymphoid malignancies I'.nfiltr ating the bone marrow -
0
0
Ariad S, Clifford D, Penfold G, MacPhail AP, Bezwoda WR. Erythropoietin response in anaemic patients with multiple myeloma and other lymphoid malignancies infiltrating the bone marrow. Eur J Haematol 1992: 49: 59-62.
I
Abstract: Immunoreactive erythropoietin levels were measured in 42 patients with lymphoid malignancies with anaemia and bone marrow involvement. Results were compared to a control group of 16 patients suffering from anaemia due to other causes. Significant inverse correlations between serum erythropoietin level and haemoglobin concentration were shown for the patients with lymphoid malignancies and also for the control subjects. Overall, the erythropoietin levels of patients with lymphoid malignancies with bone marrow infiltration and with normal renal function did not differ significantly from erythropoietin levels of the anaemic controls. We conclude that anaemia in patients with lymphoproliferative disorders with bone marrow infiltration and normal renal function is caused primarily by a diminished/inadequate response to erythropoietin at the level of the target cell.
Introduction
Lymphoid tumours are frequently associated with anaemia, the causes of which are complex and include displacement of erythropoietic elements by tumour infiltration of the marrow, immune mechanisms and/or hypersplenism resulting in diminished red cell survival, alteration of plasma volume with dilutional anaemias (in patients with paraproteinaemia) or the effects of mediators such as TNF and y-interferon (1) as well as other lymphokines which, amongst other actions, impair internal iron kinetics giving rise to the so-called anaemia of chronic disease. Whatever the relative contribution of each of these factors, the occurrence of anaemia per se indicates that erythropoiesis is insufficient to maintain haemoglobin concentration at normal levels. Normal erythropoiesis is regulated primarily by erythropoietin (EPO), which acts on committed erythroid progenitors (2, 3). In iron-deficiency anaemia, elevated EPO levels are found with a linear-inverse correlation between haemoglobin concentration and EPO. In renal failure EPO levels are uniformly low (4) whereas in rheumatoid arthritis ( 5 ) and HIV infection (6) the EPO response is inadequate, but EPO levels continue to be inversely related to haemoglobin concentrations. More variable results have been found in patients with malignancy and anaemia.
S. Ariad', D. Clifford*, G. Penfold', A. P. MacPhail and W. R. Bezwoda
'
'
'
Department of Medicine and the South African Institute of Medical Research, University of the Witwatersrand Medical School, Parktown, Johannesburg, South Africa
Key words: anaemia - multiple myeloma bone marrow - erythropoietin Correspondence: W.R. Bezwoda, Haematology/Oncology Unit, Department of Medicine, University of the Witwatersrand Medical School, 7 York Road, Parktown 2193, Johannesburg, South Africa Accepted for publication 4 May 1992
Leukaemic patients with bone marrow infiltration and anaemia have been shown to have relatively high levels of EPO (7). A recent investigation has, however, demonstrated a decreased EPO response in patients with non-haematologic malignancies (8). The erythropoietin response to anaemia in patients with lymphoid tumours infiltrating the marrow has not, as yet, been adequately determined despite a recent report demonstrating that the anaemia occurring in patients with multiple myeloma may respond to treatment with recombinant EPO (9). It is unclear whether these patients have an inadequate EPO response for the degree of anaemia or whether there is a diminished proliferative response of erythropoietic cells to normal levels of EPO or both. In an attempt to answer this question, we studied circulating EPO in anaemic patients with multiple myeloma and other lymphoid tumours with bone marrow infiltration. Methods
Forty-two patients suffering from lymphoid malignancies with bone marrow infiltration and anaemia attending the Haematology Unit of the Johannesburg Hospital were studied between September and November, 1990. All patients were ambulatory, capable of self care, haemodynamically stable and with no respiratory disorder or active infection. The pa-
59
Ariad et al. tients were divided into two groups. The first group comprised 26 patients with multiple myeloma. Sixteen patients had IgG myeloma, 7 IgA myeloma and 3 had light chain myeloma. Other patient details including renal function status are shown in Table 1. The second group included 16 patients with nonHodgkins lymphoma who all had bone marrow infiltration and anaemia. Seven patients had follicular lymphomas (either low or intermediate grade, according the Working Formulation). Another 7 patients had lymphoproliferative disorders with IgM paraproteinaemia with or without Bence Jones proteinuria, with a specific diagnosis of Waldenstroms macroglobulinaemia in 5. Two patients suffered from diffuse high grade lymphomas of B-cell phenotype. All but 2 of the above patients had normal renal function. The control group consisted of 16 anaemic males and females. Eleven of the 16 patients had irondeficiency anaemia. All 11 were females of childbearing age. Gastrointestinal malignancy as a cause for the iron deficiency had been excluded. The diagnosis of pure iron deficiency was confirmed by the findings of serum ferritin levels < 10 pg/I, serum iron levels < 50 pmol/l, serum transferrin concentration > 3.6 mg/l and transferrin saturation < 10%. The other 5 patients had agnogenic myeloid metaplasia with myelofibrosis as the cause for their anaemia. These 5 patients had more severe anaemia and were receiving regular blood transfusions although patients with an unstable haemodynamic state or suspected acute blood loss were excluded from the control group. The myelofibrotic patients were generally older than the iron-deficient patients, with an age range which was more comparable to that of the study group. All control patients had normal renal function. Blood samples were collected between 08.0010.00, allowed to clot and the serum separated and frozen at - 20°C until assayed. Immunoreactive EPO was measured by means of a commercially available iminunoradiometric assay, EPO-TRAC (Incstar Shelwater, Minnesota, USA). This assay uses radiolabelled recombinant EPO as the antigen and a monospecific, polyvalent rabbit antiserum to recombinant human EPO. The intraassay coefficients of variation for low, medium and high standards were 8.7, 9.5 and 6.2% respectively, and the between-assay variations for the same samples were 14.1, 10.6 and 10.3% respectively. Haemoglobin estimations were performed using a Technicon (H 1) autoanalyser. Serum iron, transferrin and percentage saturation was measured according to the ICSH panel recommendations. Ferritin estimations were by means of an enzyme linked immunoassay previously reported. EPO values were significantly skewed, which was
60
corrected by log transformation. Erythropoietin levels were analysed in relation to haemoglobin concentration. Statistical analyses included t tests, both paired and unpaired, and analysis of variance and co-variance. All p-values reported are 2-sided. The study was approved by the Committee of Ethics of Human Research of the University of the Witwatersrand. Results
Serum immunoreactive EPO levels did not differ significantly when anaemic control subjects and patients with myeloma or other lymphoproliferative diseases were compared (Table 1). Mean haemoglobin levels of the control and investigational groups were also similar. Patients with renal dysfunction did, however, have significantly lower EPO levels than either the control group or the investigational group with normal renal function. Analysis of various haematologic indices, biochemical investigations and immunoreactive EPO levels showed significant inverse correlations between log EPO levels and both haemoglobin and haematocrit values among the patients with multiple myeloma, those with other lymphoproliferative disorders, and among the anaemic control group. When analysed separately, there were no significant differences with regard to the EPO response in relation to Table 1. lmmunoreactive EPO levels in patients with myeloma, other lymphoproliferative disorders and anemic controls
Log EPO
(Uilt
Hb concentration (g/dl) MeanfSD
No of patients
Geometric Mean SD range
Anemic controls
16
90.4 19.8-411.7
10.3f 1.4
Multiple myeloma All patients
26
95.3 39.9-226.9 120.2 52.4-295.4 56.2 26.4-1 19.6
1l.Otl.6
Normal renal function
18
Impaired renal function
8
Lymphoproliferative disorder All patients
16
With IgM paraproteinemia
7
Without paraproteinemia
9
Normal renal function
14
All 6-lymphoid malignancy
42
Normal renal function
26
140.3 23.7-829.8 95.4 12.8-712.8 188.4 36.8-963.8 186.2 33.3-1039.9 110.6 30.6-399.0 145.8 40.2-529.6
11.1+1.8 10.8k1.4 9.6t 1.7 9.8f1.3 10.3+1.6 10.6k1.4 10.2+ 1.5 ll.O+l.6
Erythropoietin response in anaemia Table 2. Plasma immunoreactive erythropoietin levels and hematologic and biochemical indices: correlation analysis Hb g/dl
HTC ratio
Paraprotein gil
Creatinine pmolil
MCV fl
Multiple myeloma Log EPO -0.42 lUil p
0
0
Ariad S, Clifford D, Penfold G, MacPhail AP, Bezwoda WR. Erythropoietin response in anaemic patients with multiple myeloma and other lymphoid malignancies infiltrating the bone marrow. Eur J Haematol 1992: 49: 59-62.
I
Abstract: Immunoreactive erythropoietin levels were measured in 42 patients with lymphoid malignancies with anaemia and bone marrow involvement. Results were compared to a control group of 16 patients suffering from anaemia due to other causes. Significant inverse correlations between serum erythropoietin level and haemoglobin concentration were shown for the patients with lymphoid malignancies and also for the control subjects. Overall, the erythropoietin levels of patients with lymphoid malignancies with bone marrow infiltration and with normal renal function did not differ significantly from erythropoietin levels of the anaemic controls. We conclude that anaemia in patients with lymphoproliferative disorders with bone marrow infiltration and normal renal function is caused primarily by a diminished/inadequate response to erythropoietin at the level of the target cell.
Introduction
Lymphoid tumours are frequently associated with anaemia, the causes of which are complex and include displacement of erythropoietic elements by tumour infiltration of the marrow, immune mechanisms and/or hypersplenism resulting in diminished red cell survival, alteration of plasma volume with dilutional anaemias (in patients with paraproteinaemia) or the effects of mediators such as TNF and y-interferon (1) as well as other lymphokines which, amongst other actions, impair internal iron kinetics giving rise to the so-called anaemia of chronic disease. Whatever the relative contribution of each of these factors, the occurrence of anaemia per se indicates that erythropoiesis is insufficient to maintain haemoglobin concentration at normal levels. Normal erythropoiesis is regulated primarily by erythropoietin (EPO), which acts on committed erythroid progenitors (2, 3). In iron-deficiency anaemia, elevated EPO levels are found with a linear-inverse correlation between haemoglobin concentration and EPO. In renal failure EPO levels are uniformly low (4) whereas in rheumatoid arthritis ( 5 ) and HIV infection (6) the EPO response is inadequate, but EPO levels continue to be inversely related to haemoglobin concentrations. More variable results have been found in patients with malignancy and anaemia.
S. Ariad', D. Clifford*, G. Penfold', A. P. MacPhail and W. R. Bezwoda
'
'
'
Department of Medicine and the South African Institute of Medical Research, University of the Witwatersrand Medical School, Parktown, Johannesburg, South Africa
Key words: anaemia - multiple myeloma bone marrow - erythropoietin Correspondence: W.R. Bezwoda, Haematology/Oncology Unit, Department of Medicine, University of the Witwatersrand Medical School, 7 York Road, Parktown 2193, Johannesburg, South Africa Accepted for publication 4 May 1992
Leukaemic patients with bone marrow infiltration and anaemia have been shown to have relatively high levels of EPO (7). A recent investigation has, however, demonstrated a decreased EPO response in patients with non-haematologic malignancies (8). The erythropoietin response to anaemia in patients with lymphoid tumours infiltrating the marrow has not, as yet, been adequately determined despite a recent report demonstrating that the anaemia occurring in patients with multiple myeloma may respond to treatment with recombinant EPO (9). It is unclear whether these patients have an inadequate EPO response for the degree of anaemia or whether there is a diminished proliferative response of erythropoietic cells to normal levels of EPO or both. In an attempt to answer this question, we studied circulating EPO in anaemic patients with multiple myeloma and other lymphoid tumours with bone marrow infiltration. Methods
Forty-two patients suffering from lymphoid malignancies with bone marrow infiltration and anaemia attending the Haematology Unit of the Johannesburg Hospital were studied between September and November, 1990. All patients were ambulatory, capable of self care, haemodynamically stable and with no respiratory disorder or active infection. The pa-
59
Ariad et al. tients were divided into two groups. The first group comprised 26 patients with multiple myeloma. Sixteen patients had IgG myeloma, 7 IgA myeloma and 3 had light chain myeloma. Other patient details including renal function status are shown in Table 1. The second group included 16 patients with nonHodgkins lymphoma who all had bone marrow infiltration and anaemia. Seven patients had follicular lymphomas (either low or intermediate grade, according the Working Formulation). Another 7 patients had lymphoproliferative disorders with IgM paraproteinaemia with or without Bence Jones proteinuria, with a specific diagnosis of Waldenstroms macroglobulinaemia in 5. Two patients suffered from diffuse high grade lymphomas of B-cell phenotype. All but 2 of the above patients had normal renal function. The control group consisted of 16 anaemic males and females. Eleven of the 16 patients had irondeficiency anaemia. All 11 were females of childbearing age. Gastrointestinal malignancy as a cause for the iron deficiency had been excluded. The diagnosis of pure iron deficiency was confirmed by the findings of serum ferritin levels < 10 pg/I, serum iron levels < 50 pmol/l, serum transferrin concentration > 3.6 mg/l and transferrin saturation < 10%. The other 5 patients had agnogenic myeloid metaplasia with myelofibrosis as the cause for their anaemia. These 5 patients had more severe anaemia and were receiving regular blood transfusions although patients with an unstable haemodynamic state or suspected acute blood loss were excluded from the control group. The myelofibrotic patients were generally older than the iron-deficient patients, with an age range which was more comparable to that of the study group. All control patients had normal renal function. Blood samples were collected between 08.0010.00, allowed to clot and the serum separated and frozen at - 20°C until assayed. Immunoreactive EPO was measured by means of a commercially available iminunoradiometric assay, EPO-TRAC (Incstar Shelwater, Minnesota, USA). This assay uses radiolabelled recombinant EPO as the antigen and a monospecific, polyvalent rabbit antiserum to recombinant human EPO. The intraassay coefficients of variation for low, medium and high standards were 8.7, 9.5 and 6.2% respectively, and the between-assay variations for the same samples were 14.1, 10.6 and 10.3% respectively. Haemoglobin estimations were performed using a Technicon (H 1) autoanalyser. Serum iron, transferrin and percentage saturation was measured according to the ICSH panel recommendations. Ferritin estimations were by means of an enzyme linked immunoassay previously reported. EPO values were significantly skewed, which was
60
corrected by log transformation. Erythropoietin levels were analysed in relation to haemoglobin concentration. Statistical analyses included t tests, both paired and unpaired, and analysis of variance and co-variance. All p-values reported are 2-sided. The study was approved by the Committee of Ethics of Human Research of the University of the Witwatersrand. Results
Serum immunoreactive EPO levels did not differ significantly when anaemic control subjects and patients with myeloma or other lymphoproliferative diseases were compared (Table 1). Mean haemoglobin levels of the control and investigational groups were also similar. Patients with renal dysfunction did, however, have significantly lower EPO levels than either the control group or the investigational group with normal renal function. Analysis of various haematologic indices, biochemical investigations and immunoreactive EPO levels showed significant inverse correlations between log EPO levels and both haemoglobin and haematocrit values among the patients with multiple myeloma, those with other lymphoproliferative disorders, and among the anaemic control group. When analysed separately, there were no significant differences with regard to the EPO response in relation to Table 1. lmmunoreactive EPO levels in patients with myeloma, other lymphoproliferative disorders and anemic controls
Log EPO
(Uilt
Hb concentration (g/dl) MeanfSD
No of patients
Geometric Mean SD range
Anemic controls
16
90.4 19.8-411.7
10.3f 1.4
Multiple myeloma All patients
26
95.3 39.9-226.9 120.2 52.4-295.4 56.2 26.4-1 19.6
1l.Otl.6
Normal renal function
18
Impaired renal function
8
Lymphoproliferative disorder All patients
16
With IgM paraproteinemia
7
Without paraproteinemia
9
Normal renal function
14
All 6-lymphoid malignancy
42
Normal renal function
26
140.3 23.7-829.8 95.4 12.8-712.8 188.4 36.8-963.8 186.2 33.3-1039.9 110.6 30.6-399.0 145.8 40.2-529.6
11.1+1.8 10.8k1.4 9.6t 1.7 9.8f1.3 10.3+1.6 10.6k1.4 10.2+ 1.5 ll.O+l.6
Erythropoietin response in anaemia Table 2. Plasma immunoreactive erythropoietin levels and hematologic and biochemical indices: correlation analysis Hb g/dl
HTC ratio
Paraprotein gil
Creatinine pmolil
MCV fl
Multiple myeloma Log EPO -0.42 lUil p
