- Email: [email protected]
EFFECT OF HUMAN ERYTHROPOIETIN DERIVED FROM RECOMBINANT DNA ON THE ANAEMIA OF PATIENTS MAINTAINED BY CHRONIC HAEMODIALYSIS
Saturday 22 November 1986 EFFECT OF HUMAN ERYTHROPOIETIN DERIVED FROM RECOMBINANT DNA ON THE ANAEMIA OF PATIENTS MAINTAINED BY CHRONIC HAEMODIALYSIS CHRISTOPHER G. WINEARLS1
MARTIN J. PIPPARD3 MICHAEL R. DOWNING4
DESMOND O. OLIVER2 CECIL REID3 P. MARY COTES3
Department of Medicine, Royal Postgraduate Medical School, London W12 OHS;1 Renal Unit, Churchill Hospital, Oxford,2 Section of Haematology, Clinical Research Centre, Harrow, Middlesex, UK;3 and Amgen, Thousand Oaks, California, USA4
Summary
Ten patients with end-stage renal failure and anaemia (mean haemoglobin 6·1 g/dl,
range 4·6-8·8 g/dl) on thrice-weekly haemodialysis were treated with human erythropoietin derived from recombinant DNA (rHuEPO). This was given as an intravenous bolus after each dialysis in rising doses within the range 3-192 IU/kg. All patients showed increases in reticulocyte numbers and haemoglobin concentration and after the first week of treatment none of the four previously transfusion-dependent patients needed further transfusions. In nine patients treated for 12 weeks haemoglobin rose to a mean of 10·3 g/dl, range 9·5 to 12·8 g/dl. Thereafter the dose of erythropoietin was adjusted to avoid a further rise in haemoglobin. During treatment one patient had an episode of hypertensive encephalopathy and two had clotting in their arteriovenous fistulas (complete in one). rHuEPO is an effective treatment for the anaemia of end-stage renal failure but longer-term observations are needed on the consequences of increasing the haematocrit.
Introduction ANAEMIA is an almost invariable feature of chronic renal failure and is particularly severe in anephric patients treated by long-term maintenance haemodialysis.l Many of these
patients require regular blood transfusions with their attendant risks of hepatitis, iron overload, and sensitisation
histocompatibility antigens (which reduces the chances of successful renal transplantation). The factors that may contribute to this anaemia include shortened red-cell survival, marrow suppression by "uraemic toxins", repeated blood loss on dialysis, and aluminium toxicity.2 However, the critical defect is probably inadequate secretion of erythropoietin.3 This is the major hormonal regulator of red-cell production and is produced principally but not
to
solely by the kidney.4Thus anaemia is most severe in anephric patients who have inappropriately low erythropoietin concentrations6 and least severe in patients with adult polycystic kidney disease in whom renal mass, if not function, is preserved.7 Furthermore, patients whose non-functioning, previously shrunken kidneys develop "acquired cystic disease" frequently show a striking rise in haemog10bin8>9 as well as increased serum erythropoietin concentrations1O despite an unchanged uraemic state. Lately, Eschbach and colleaguesll have shown that daily infusions of erythropoietin-rich plasma can completely reverse anaemia in uraemic sheep. However, the routine administration of erythropoietin for the treatment of anaemia in patients with renal failure has hitherto been impossible because there is no source from which native human erythropoietin can be extracted in a sufficient quantity for therapeutic use. The gene coding for human erythropoietin has now been cloned12>13 and expressed in cultured mammalian cells." Recombinant human erythropoietin (rHuEPO), purified from the cell-conditioned medium, has been prepared in sufficient quantities for clinical use. We describe here the results of replacement therapy with erythropoietin in anaemic patients with end-stage renal disease maintained by haemodialysis. Patients and Methods This study was approved by the local ethical committees of the three participating institutions. Subjects were ten clinically stable patients, aged 26-65 years, on thrice-weekly maintenance haemodialysis because of chronic renal failure. Four were anephric and four were receiving regular blood transfusions. They were anaemic (packed cell volume less than 30 %) and had little chance of receiving a renal transplant, no additional cause for anaemia, and no major underlying disease. All -gave informed consent to participation in the trial. Patients without evidence of iron overload due to multiple blood transfusions were receiving oral iron (65-180 mg daily, except patient 3, 65 mg thrice weekly). The rHuEP013 used in the study was purified to greater than 98% homogeneity from the cell-conditioned medium and then formulated in a buffered saline solution. It had a specific activity of 129 000 IU/mg ofhorIIlone.14 Complete sequence analysis of highly purified human urinary erythropoietin1s and rHuEPO showed that
the
two
hormones have identical aminoacid sequences, and
on
carbohydrate structure analysis they have very similar oligosaccharide chains. In addition, the biological activity and immunological properties of the human urinary and recombinant hormones have not been distinguishable.14,16-18 The manufacturer’s production protocols and samples of each lot of rHuEPO used in the study were examined at the National Institute for Biological Standards and Control, London. 8517
1176 The clinical study protocol was a non-randomised, escalatingdose regimen designed to investigate the efficacy and safety of rHuEPO. The material was administered thrice weekly as an intravenous bolus at the end of each routine dialysis treatment. After two weeks of placebo treatment during which each patient received 1 ml of normal saline instead of rHuEPO, the patients were admitted for administration of the first dose and monitoring of vital signs (oral temperature, blood pressure, pulse, and respiration rates) for 24 h. Thereafter each dose was administered at the completion of dialysis and vital signs were measured before and 30 min after injection of the rHuEPO. Blood counts were done with a Coulter counter and reticulocytes were counted by standard methods. A biochemical profile including information on iron status was performed weekly. The dose range studied was 12 to 96 or 192 IU/kg body weight, except for patient 1 in whom it was 3-192 IU/kg thrice weekly. Doses were increased stepwise by doubling at each increment. In the range 3-24 IU/kg each dose was administered for one week only and at 48 and 96 IU/kg for at least two weeks. The dose was increased until there was a haemoglobin response, defined as an increase of 2 g/dl or more above the mean concentration during the placebo period. In patients whose course was complicated by transfusion, response was defined as an increase of 2 g/dl or more above the highest concentration induced by transfusion. Thereafter the dose of rHuEPO was adjusted to maintain haemoglobin at 11-13 g/dl. This upper limit was selected in the hope of avoiding thrombotic events. For assessment of changes in haemoglobin concentration, reticulocyte counts, and mean corpuscular volume (MCV), values compared (and reported here) were the means of three measurements before the second and third doses of each week reviewed and before the first dose of the following week.
Results
During treatment all ten patients showed an increase in haemoglobin concentration (figure). Within eleven weeks of starting treatment, all patients fulfilled our criterion for a response. In nine patients treated for twelve weeks, haemoglobin concentration increased to a mean of 10-3 g/dl, range 9 5-128 g/dl. Four patients received blood transfusions during the placebo period or the first week of treatment (figure, lower panel) but none has needed further transfusion despite a requirement during the preceding year ranging from 17 to 31units per month (table). In all ten patients, pretreatment reticulocyte counts were 31 (SD 22) x 109/1-ie, inappropriately low for the degree of anaemia. After two weeks at 48 IU/kg body weight, reticulocyte counts reached 75 (SD 27) x 109/1 and an increase of dose to 96 IU/kg was associated with a further rise to 109 (SD 35) x 109/1. None of the patients had evidence of iron-deficient erythropoiesis at the start of the study, as judged by the
Haemoglobin concentration and duration of
preceding entry;
tafter 10 weeks’
with
Upper panel, patients not transfused during the course of the study (2 = 0, 3=,7=0,8=,9= A, and 10 = A).Lower panel, patients transfused (to -amaximum haemoglobin concentration T) during the pretreatment placebo period or during the first week of treatment (1= 0,4= 8, 50, and 6 = . The baseline haemoglobin concentration is the mean of all observations made during the placebo period with exclusion of values complicated by increase from transfusion. In the case of subject 1, in whom treatment was started at 3 IU/kg body weight, values shown at 4, 8, and 9 weeks’treatment are those achieved after 6,10, and 11 weeks. The cumulative dose after 4 weeks’ treatment was 396 IU/kg (subject 1,423). For each subject the cumulative dose in IU/kg for the total duration of treatment plotted is shown in parentheses.
PATIENT DETAILS AND IRON STATUS
*Units of blood transfused during year TIBC = total iron binding capacity.
treatment
rHuEPO.
treatment
including 1 week each of 3 and 6 TU/kg (thrice weekly).
1177
alarming, but in retrospect blood pressure had been poorly controlled. The rise in haematocrit would probably have increased peripheral resistance and thereby raised the blood pressure still further. In no other patient was a rise in blood pressure seen that could be related directly to
MCV and mean corpuscular haemoglobin. Five patients were iron-loaded (serum ferritin 924-5195 ug/1), four because of previous blood transfusions and one because of previous intravenous iron supplements, and the remaining five patients had normal or low iron stores (serum ferritin 26-273 ug/1). Serum ferritin tended to fall during therapy (table). However, even in those with normal or reduced iron stores (who were taking oral iron supplements) there was no evidence that lack of iron prevented the haemoglobin response to the dose schedule of erythropoietin administered (figure). There was a rise in MCV in the five iron-loaded patients from a mean of 90 (range 86-94) fl before treatment to 101 (range 96-105) fl during treatment. In the patients who were not iron-loaded, MCV was unchanged with pre and post treatment ranges 82-91 and 82-95 fl, respectively. There were no changes in other haematological indices, and in particular no increases in
was
with rHuEPO or its consequences. Careful blood pressure control will be needed if is haemoglobin to be increased by treatment and there may be a case for restricting the rate of rise in haemoglobin. It is noteworthy that, before the start of treatment, five patients had normal or reduced iron stores; indeed, in one of these (patient 8) the initial serum ferritin was only 26 ug/1. In addition, after eight weeks of treatment the serum ferritin was less than 50 f.lg/1 in four of these five patients, consistent with the exhaustion of the iron stores’9 as iron was mobilised for haemoglobin synthesis. These patients’ response to treatment was therefore presumably dependent at least partly upon iron supplied from their daily oral iron supplements. That an increase in haemoglobin occurred in the presence of limited iron stores provides further evidence that, despite renal failure, iron absorption can be enhanced in response to increased demand for iron.20,21 Nonetheless, our data do not exclude the possibility that the decreased availability of iron may, in some patients, limit the rate or magnitude of response to treatment. Our results confirm the importance of erythropoietin deficiency in the pathogenesis of anaemia in chronic renal failure. On treatment with rHuEPO, some patients have achieved near-normal haemoglobin concentrations despite continued exposure to the metabolic effects of chronic renal failure and routine haemodialysis. Thus the anaemia of endstage renal failure is corrected by treatment with rHuEPO but the long-term effects of near-normal haemoglobin concentrations from maintenance therapy remain to be evaluated. treatment
attention
platelet counts. An improved sense of wellbeing was described by nine of ten patients and eight noted increased exercise tolerance. Four experienced several episodes of aching in the limbs and pelvis 2 h after injection; this was accompanied by a feeling of cold and then sweating but no fever. The episodes lasted no longer than 12 h and the patients regarded this occasional adverse effect as trivial in relation to the benefit felt from treatment. Patient 1 had an episode of hypertensive encephalopathy at a time when her haemoglobin was 93 g/dl; blood pressure, usually 170/90 mm Hg, rose to 220/140 mm Hg. The rHuEPO was stopped and within three weeks the haemoglobin had fallen to the pretreatment value. After control of the blood pressure, treatment with rHuEPO was restarted at 96 IU/kg thrice weekly and again there was an increase in haemoglobin and reticulocyte counts. Complete and partial clotting of arteriovenous fistulas occurred in two patients when haemoglobin was 9 and 13 g/dl. Discussion out
We are grateful to the staff of laboratories at all three centres for additional blood and reticulocyte counts and determinations of serum iron and transferrin, to the haemodialysis nurses for extra observations, and to Mr R. C. Tam (Clinical Research Centre) for estimates of ferritin. We thank Dr A. J. Rees and Dr D. N. S. Kerr for allowing us to study patients under their care and for advice and encouragement; Dr Nowell Stebbing for help in the early stages of the study; and Dr D. R. Bangham of the National Institute for Biological Standards and Control for help and encouragement. Ortho-Cilag Pharmaceutical Ltd provided the rHuEPO, which was manufactured by
The findings reported here are of major importance to patients who require long-term dialysis. It is clear that a thrice-weekly dose regimen is effective for the correction of
the anaemia of renal failure but this may not be the most appropriate or economical long-term schedule. Furthermore, the dose of rHuEPO needed to maintain a nearnormal concentration of haemoglobin and its relation to normal production of the hormone has yet to be established. For example, most patients showed a rise in reticulocyte count at a dose of 24 IU/kg thrice weekly but the subsequent increases required by our treatment schedule made it impossible to know whether this small dose would, if continued, itself have induced an adequate therapeutic response. The benefits of erythropoietin treatment revealed in this
preliminary study are an improvement in patient wellbeing and elimination of the need for blood transfusions. We foresee the possibility that treatment with rHuEPO combined with venesection may permit the correction of serious existing iron overload in patients with renal failure. There are, however, potential risks in raising the haemoglobin concentration and blood viscosity in a group of patients prone to accelerated vascular disease. We recorded two episodes of thrombosis of vascular access, one minor and one complete. Long-term observation of patients treated with rHuEPO will be required to assess the possibility of an increased incidence of thrombotic events. The episode of hypertensive encephalopathy in patient 1
to
Amgen and is being jointly developed by Amgen, Thousand Oaks, California, USA, by Cilag Schaffhausen, Switzerland, by Kirin, Tokyo, Japan, and by Ortho Pharmaceutical Corporation, Raritan, New Jersey, USA. We thank Dr W. L. Jefferson and Mr R. C. Hill of Ortho-Cilag Pharmaceutical Ltd for continuing help. .
Correspondence should be addressed to C. G. W., Department of Medicine, Royal Postgraduate Medical School, London W12 OHS. REFERENCES
JW. Hematologic problems of dialysis patients. In: Drukker W, Parsons FM, Maher JF, eds. Replacement of renal function by dialysis. Boston: Martinus Nijhoff, 1983: 630-45. 2. Eschbach JW, Adamson JW. Anemia of end-stage renal disease (ESRD). Kidney Int 1. Eschbach
1985; 28: 1-5. 3. Adamson JW, Eschbach JW, Finch CA. The kidney and erythropoiesis. Am J Med 1968; 44: 725-33. 4. Jacobson LO, Goldwasser E, Fned W, Plzak L. Role of the kidney in erythropoiesis. Nature 1957; 179: 633-34. 5. Erslev AJ. Erythropoietin function of the kidney. In: Wesson LG, ed. Physiology of the human kidney. New York: Grune and Stratton, 1969: 521-34. 6. Caro J, Brown S, Miller O, Murray T, Erslev AJ. Erythropoietin levels in uremic nephric and anephric patients. J Lab Clin Med 1979; 93: 449-58. 7. Chandra M, Miller ME, Garcia JF, Mossey RT, McVicar M. Serum immunoreactive erythropoietin levels in patients with polycystic kidney disease as compared with other hemodialysis patients. Nephron 1985; 39: 26-29. 8. Goldsmith HJ, Ahmad R, Raichura N, Lal SM, McConnell CA, Gould DA, Gyde OHB, Green J. Association between rising haemoglobin concentration and renal cyst formation in patients on long term regular haemodialysis treatment. Proc EDTA 1982; 19: 313-18.
1178
EFFECT OF DUODENAL ULCER SURGERY AND ENTEROGASTRIC REFLUX ON GAMPYLOBACTER PYLORIDIS H.
J. O’CONNOR1 J. I. WYATT2
M. F. DIXON2 A. T. R. AXON1 E. P. DEWAR3
D. C. WARD3
D. JOHNSTON3 Gastroenterology Unit1 and University Departments of Pathology2 and Surgery,3 General Infirmary at Leeds, Leeds LSI 3EX
Summary
To
assess
the effect of duodenal ulcer
Campylobacter pyloridis gastric biopsies were done and fasting bile acid concentrations in gastric aspirates were measured in 35 patients with active duodenal ulceration and 54 who had undergone surgery at some time. Biopsy specimens were assessed blind for the presence of C pyloridis and scored for severity of reflux gastritis by the use of a histological grading system. Among patients who had undergone highly selective vagotomy the proportion who were C pyloridis-positive was similar to that in the unoperated group, but among those who had undergone Billroth I partial gastrectomy, Billroth II partial surgery
on
gastrectomy, or truncal vagotomy and gastroenterostomy it The absence of C was significantly lower (p < 0·001). correlated pyloridis strongly (p<0·001) with high reflux scores and increased bile acid concentrations in the stomach. Reflux scores and bile acid concentrations were significantly higher (p<0·01) after Billroth I and Billroth II partial gastrectomies and truncal vagotomy and gastroenterostomy than in the active duodenal ulcer or highly selective vagotomy groups. There was a highly significant correlation (p<0·001) between reflux scores and bile acid concentrations. These results suggest that reflux may disrupt mucus and thus cause the death of campylobacters that live beneath it. They also suggest that reflux may produce a reflux-specific gastritis. Highly selective vagotomy may protect against these changes in the gastric mucosa.
9. Ratcliffe PJ, Dunnill MS, Oliver DO. Clinical importance of acquired cystic disease of the kidney in patients undergoing dialysis. Br Med J 1983; 287: 1855-58. 10. Shalhoub RJ, Rajan U, Kim VV, Goldwasser E, Kark JA, Antoniou LD. Erythrocytosis in patients on long term dialysis. Ann Intern Med 1982; 97: 686-90.
11. Eschbach JW, Mladenovic J, Garcia JF, Wahl PW, Adamson JW. The anemia of chronic renal failure in sheep, response to erythropoietin-rich plasma in vivo. J Clin Invest 1984; 74: 434-41. 12. Jacobs K, Shoemaker C, Rudersdorf R, et al. Isolation and characterization of genomic and cDNA clones of human erythropoietin. Nature 1985; 313: 806-10. 13. Lin F-K, Suggs S, Lin C-H, et al. Cloning and expression of the human erythropoietin gene. Proc Natl Acad Sci USA 1985; 82: 7580-84. 14. Browne JK, Cohen AM, Egrie JC, et al. Erythropoietin: gene cloning, protein structure and biological properties. Cold Spring Harbor Symp Quant Biol (in press). 15. Lai P-H, Everett R, Wang FF, Arakawa T, Goldwasser E. Structural characterization of human erythropoietin. J Biol Chem 1986; 261: 3116-21. 16. Egrie JC, Browne JK, Lai P, Lin F-K. Characterization of recombinant monkey and human erythropoietin. In: Stamatoyanopoulos G, Nienhuis AW, eds. Experimental approaches for the study of hemoglobin switching. New York: Alan R. Liss, 1985: 339-50. 17. Egrie JC, Strickland TW, Lane J, et al. Characterization and biological effects of recombinant human erythropoietin. Immutobiology (in press). 18. Egrie JC, Cotes PM, Lane J, Gaines Das RE, Tam RC. Development of valid radioimmunoassays for erythropoietin using recombinant EPO as tracer and immunogen. Blood 1985; 66: 149a. 19. Milman N, Christensen TE, Visfeldt J. Diagnostic efficiency of various laboratory tests in the assessment of bone marrow iron stores in patients with chronic uraemia. Stand j Haematol 1981; 26: 257-64. 20. Gokal R, Millard PR, Weatherall DJ, Callender STE, Ledingham JGG, Oliver DO. Iron metabolism in haemodialysis patients. Quart J Med 1979; 48: 369-91. 21. Eschbach JW, Cook JD, Scribner BH, Finch CA. Iron balance in hemodialysis patients. Ann Intern Med 1977; 87: 710-13.
Introduction
THE presence of Campylobacter pyloridis on gastric mucosal biopsies is strongly associated with peptic ulcer diseased This is particularly so for duodenal ulceration (DU), where virtually all patients have C pyloridis-positive antral gastritis.2 Although there are data on the effects of medical therapy for peptic ulcer on C pyloridis,3 the fate of this organism after peptic ulcer surgery has not been established. We have explored the effect of different types of gastric surgery on C pyloridis by studying a large series of DU patients, in some instances preoperatively and in others after a conservative type of operation (highly selective vagotomy-HSV) or after operations that remove or bypass the pylorus.
Subjects and Methods Patients The 89 patients could be divided into five groups (table)-one with active duodenal ulcers ("untreated" DU group) and four consisting of patients who had been treated surgically in the past for DU. None of the postoperative patients had recurrent ulcers. All patients gave informed written consent for the study.
Sites of Biopsies and Sampling
Techniques
All the patients underwent upper gastrointestinal endoscopy after an overnight fast. At endoscopy, at least two gastric mucosal biopsies were taken from within 5 cm of the pylorus from untreated patients and those who had undergone HSV and from within 5 cm of the stoma from patients in the Biltroth I partial gastrectomy (BIPG), Billroth II partial gastrectomy (BIIPG), and truncal vagotomy and gastroenterostomy (TVGE) groups. In the intact stomach the presence of organisms in the antrum affected by gastritis is almost invariably accompanied by the existence of organisms in biopsy specimens of the body of the stomach even when the body mucosa looks normal 4 We therefore consider it valid to compare the prevalence of C pyloridis in the pre-pyloric mucosa of the intact stomach with the peristomal mucosa of the postoperative stomach. In 71 patients (80%) fasting gastric contents were aspirated for measurement of total bile acid concentration (BAC). In 25 patients gastric aspirates (5-10 ml) were obtained at endoscopy, by the use of a sterile ’Teflon’ cannula passed down the biopsy channel of the endoscope whereas in the other 46 patients contents were aspirated through a size 12 nasogastric tube positioned in the stomach by the water recovery method5 after a further overnight fast.
Histological Assessment of Gastric Mucosal Biopsy Specimens were oriented on filter paper and fixed in formol saline. Paraffm-processed sections immediately were cut at three levels and stained by haematoxylin and eosin and alcian-blue pH 25/PAS. A further section was taken from each set of biopsy specimens and stained by the undifferentiated Giemsa method.2 These sections were examined "blind" by J. I. W. for the presence of C pyloridis.
The
biopsy specimens
PATIENT GROUPS
MARTIN J. PIPPARD3 MICHAEL R. DOWNING4
DESMOND O. OLIVER2 CECIL REID3 P. MARY COTES3
Department of Medicine, Royal Postgraduate Medical School, London W12 OHS;1 Renal Unit, Churchill Hospital, Oxford,2 Section of Haematology, Clinical Research Centre, Harrow, Middlesex, UK;3 and Amgen, Thousand Oaks, California, USA4
Summary
Ten patients with end-stage renal failure and anaemia (mean haemoglobin 6·1 g/dl,
range 4·6-8·8 g/dl) on thrice-weekly haemodialysis were treated with human erythropoietin derived from recombinant DNA (rHuEPO). This was given as an intravenous bolus after each dialysis in rising doses within the range 3-192 IU/kg. All patients showed increases in reticulocyte numbers and haemoglobin concentration and after the first week of treatment none of the four previously transfusion-dependent patients needed further transfusions. In nine patients treated for 12 weeks haemoglobin rose to a mean of 10·3 g/dl, range 9·5 to 12·8 g/dl. Thereafter the dose of erythropoietin was adjusted to avoid a further rise in haemoglobin. During treatment one patient had an episode of hypertensive encephalopathy and two had clotting in their arteriovenous fistulas (complete in one). rHuEPO is an effective treatment for the anaemia of end-stage renal failure but longer-term observations are needed on the consequences of increasing the haematocrit.
Introduction ANAEMIA is an almost invariable feature of chronic renal failure and is particularly severe in anephric patients treated by long-term maintenance haemodialysis.l Many of these
patients require regular blood transfusions with their attendant risks of hepatitis, iron overload, and sensitisation
histocompatibility antigens (which reduces the chances of successful renal transplantation). The factors that may contribute to this anaemia include shortened red-cell survival, marrow suppression by "uraemic toxins", repeated blood loss on dialysis, and aluminium toxicity.2 However, the critical defect is probably inadequate secretion of erythropoietin.3 This is the major hormonal regulator of red-cell production and is produced principally but not
to
solely by the kidney.4Thus anaemia is most severe in anephric patients who have inappropriately low erythropoietin concentrations6 and least severe in patients with adult polycystic kidney disease in whom renal mass, if not function, is preserved.7 Furthermore, patients whose non-functioning, previously shrunken kidneys develop "acquired cystic disease" frequently show a striking rise in haemog10bin8>9 as well as increased serum erythropoietin concentrations1O despite an unchanged uraemic state. Lately, Eschbach and colleaguesll have shown that daily infusions of erythropoietin-rich plasma can completely reverse anaemia in uraemic sheep. However, the routine administration of erythropoietin for the treatment of anaemia in patients with renal failure has hitherto been impossible because there is no source from which native human erythropoietin can be extracted in a sufficient quantity for therapeutic use. The gene coding for human erythropoietin has now been cloned12>13 and expressed in cultured mammalian cells." Recombinant human erythropoietin (rHuEPO), purified from the cell-conditioned medium, has been prepared in sufficient quantities for clinical use. We describe here the results of replacement therapy with erythropoietin in anaemic patients with end-stage renal disease maintained by haemodialysis. Patients and Methods This study was approved by the local ethical committees of the three participating institutions. Subjects were ten clinically stable patients, aged 26-65 years, on thrice-weekly maintenance haemodialysis because of chronic renal failure. Four were anephric and four were receiving regular blood transfusions. They were anaemic (packed cell volume less than 30 %) and had little chance of receiving a renal transplant, no additional cause for anaemia, and no major underlying disease. All -gave informed consent to participation in the trial. Patients without evidence of iron overload due to multiple blood transfusions were receiving oral iron (65-180 mg daily, except patient 3, 65 mg thrice weekly). The rHuEP013 used in the study was purified to greater than 98% homogeneity from the cell-conditioned medium and then formulated in a buffered saline solution. It had a specific activity of 129 000 IU/mg ofhorIIlone.14 Complete sequence analysis of highly purified human urinary erythropoietin1s and rHuEPO showed that
the
two
hormones have identical aminoacid sequences, and
on
carbohydrate structure analysis they have very similar oligosaccharide chains. In addition, the biological activity and immunological properties of the human urinary and recombinant hormones have not been distinguishable.14,16-18 The manufacturer’s production protocols and samples of each lot of rHuEPO used in the study were examined at the National Institute for Biological Standards and Control, London. 8517
1176 The clinical study protocol was a non-randomised, escalatingdose regimen designed to investigate the efficacy and safety of rHuEPO. The material was administered thrice weekly as an intravenous bolus at the end of each routine dialysis treatment. After two weeks of placebo treatment during which each patient received 1 ml of normal saline instead of rHuEPO, the patients were admitted for administration of the first dose and monitoring of vital signs (oral temperature, blood pressure, pulse, and respiration rates) for 24 h. Thereafter each dose was administered at the completion of dialysis and vital signs were measured before and 30 min after injection of the rHuEPO. Blood counts were done with a Coulter counter and reticulocytes were counted by standard methods. A biochemical profile including information on iron status was performed weekly. The dose range studied was 12 to 96 or 192 IU/kg body weight, except for patient 1 in whom it was 3-192 IU/kg thrice weekly. Doses were increased stepwise by doubling at each increment. In the range 3-24 IU/kg each dose was administered for one week only and at 48 and 96 IU/kg for at least two weeks. The dose was increased until there was a haemoglobin response, defined as an increase of 2 g/dl or more above the mean concentration during the placebo period. In patients whose course was complicated by transfusion, response was defined as an increase of 2 g/dl or more above the highest concentration induced by transfusion. Thereafter the dose of rHuEPO was adjusted to maintain haemoglobin at 11-13 g/dl. This upper limit was selected in the hope of avoiding thrombotic events. For assessment of changes in haemoglobin concentration, reticulocyte counts, and mean corpuscular volume (MCV), values compared (and reported here) were the means of three measurements before the second and third doses of each week reviewed and before the first dose of the following week.
Results
During treatment all ten patients showed an increase in haemoglobin concentration (figure). Within eleven weeks of starting treatment, all patients fulfilled our criterion for a response. In nine patients treated for twelve weeks, haemoglobin concentration increased to a mean of 10-3 g/dl, range 9 5-128 g/dl. Four patients received blood transfusions during the placebo period or the first week of treatment (figure, lower panel) but none has needed further transfusion despite a requirement during the preceding year ranging from 17 to 31units per month (table). In all ten patients, pretreatment reticulocyte counts were 31 (SD 22) x 109/1-ie, inappropriately low for the degree of anaemia. After two weeks at 48 IU/kg body weight, reticulocyte counts reached 75 (SD 27) x 109/1 and an increase of dose to 96 IU/kg was associated with a further rise to 109 (SD 35) x 109/1. None of the patients had evidence of iron-deficient erythropoiesis at the start of the study, as judged by the
Haemoglobin concentration and duration of
preceding entry;
tafter 10 weeks’
with
Upper panel, patients not transfused during the course of the study (2 = 0, 3=,7=0,8=,9= A, and 10 = A).Lower panel, patients transfused (to -amaximum haemoglobin concentration T) during the pretreatment placebo period or during the first week of treatment (1= 0,4= 8, 50, and 6 = . The baseline haemoglobin concentration is the mean of all observations made during the placebo period with exclusion of values complicated by increase from transfusion. In the case of subject 1, in whom treatment was started at 3 IU/kg body weight, values shown at 4, 8, and 9 weeks’treatment are those achieved after 6,10, and 11 weeks. The cumulative dose after 4 weeks’ treatment was 396 IU/kg (subject 1,423). For each subject the cumulative dose in IU/kg for the total duration of treatment plotted is shown in parentheses.
PATIENT DETAILS AND IRON STATUS
*Units of blood transfused during year TIBC = total iron binding capacity.
treatment
rHuEPO.
treatment
including 1 week each of 3 and 6 TU/kg (thrice weekly).
1177
alarming, but in retrospect blood pressure had been poorly controlled. The rise in haematocrit would probably have increased peripheral resistance and thereby raised the blood pressure still further. In no other patient was a rise in blood pressure seen that could be related directly to
MCV and mean corpuscular haemoglobin. Five patients were iron-loaded (serum ferritin 924-5195 ug/1), four because of previous blood transfusions and one because of previous intravenous iron supplements, and the remaining five patients had normal or low iron stores (serum ferritin 26-273 ug/1). Serum ferritin tended to fall during therapy (table). However, even in those with normal or reduced iron stores (who were taking oral iron supplements) there was no evidence that lack of iron prevented the haemoglobin response to the dose schedule of erythropoietin administered (figure). There was a rise in MCV in the five iron-loaded patients from a mean of 90 (range 86-94) fl before treatment to 101 (range 96-105) fl during treatment. In the patients who were not iron-loaded, MCV was unchanged with pre and post treatment ranges 82-91 and 82-95 fl, respectively. There were no changes in other haematological indices, and in particular no increases in
was
with rHuEPO or its consequences. Careful blood pressure control will be needed if is haemoglobin to be increased by treatment and there may be a case for restricting the rate of rise in haemoglobin. It is noteworthy that, before the start of treatment, five patients had normal or reduced iron stores; indeed, in one of these (patient 8) the initial serum ferritin was only 26 ug/1. In addition, after eight weeks of treatment the serum ferritin was less than 50 f.lg/1 in four of these five patients, consistent with the exhaustion of the iron stores’9 as iron was mobilised for haemoglobin synthesis. These patients’ response to treatment was therefore presumably dependent at least partly upon iron supplied from their daily oral iron supplements. That an increase in haemoglobin occurred in the presence of limited iron stores provides further evidence that, despite renal failure, iron absorption can be enhanced in response to increased demand for iron.20,21 Nonetheless, our data do not exclude the possibility that the decreased availability of iron may, in some patients, limit the rate or magnitude of response to treatment. Our results confirm the importance of erythropoietin deficiency in the pathogenesis of anaemia in chronic renal failure. On treatment with rHuEPO, some patients have achieved near-normal haemoglobin concentrations despite continued exposure to the metabolic effects of chronic renal failure and routine haemodialysis. Thus the anaemia of endstage renal failure is corrected by treatment with rHuEPO but the long-term effects of near-normal haemoglobin concentrations from maintenance therapy remain to be evaluated. treatment
attention
platelet counts. An improved sense of wellbeing was described by nine of ten patients and eight noted increased exercise tolerance. Four experienced several episodes of aching in the limbs and pelvis 2 h after injection; this was accompanied by a feeling of cold and then sweating but no fever. The episodes lasted no longer than 12 h and the patients regarded this occasional adverse effect as trivial in relation to the benefit felt from treatment. Patient 1 had an episode of hypertensive encephalopathy at a time when her haemoglobin was 93 g/dl; blood pressure, usually 170/90 mm Hg, rose to 220/140 mm Hg. The rHuEPO was stopped and within three weeks the haemoglobin had fallen to the pretreatment value. After control of the blood pressure, treatment with rHuEPO was restarted at 96 IU/kg thrice weekly and again there was an increase in haemoglobin and reticulocyte counts. Complete and partial clotting of arteriovenous fistulas occurred in two patients when haemoglobin was 9 and 13 g/dl. Discussion out
We are grateful to the staff of laboratories at all three centres for additional blood and reticulocyte counts and determinations of serum iron and transferrin, to the haemodialysis nurses for extra observations, and to Mr R. C. Tam (Clinical Research Centre) for estimates of ferritin. We thank Dr A. J. Rees and Dr D. N. S. Kerr for allowing us to study patients under their care and for advice and encouragement; Dr Nowell Stebbing for help in the early stages of the study; and Dr D. R. Bangham of the National Institute for Biological Standards and Control for help and encouragement. Ortho-Cilag Pharmaceutical Ltd provided the rHuEPO, which was manufactured by
The findings reported here are of major importance to patients who require long-term dialysis. It is clear that a thrice-weekly dose regimen is effective for the correction of
the anaemia of renal failure but this may not be the most appropriate or economical long-term schedule. Furthermore, the dose of rHuEPO needed to maintain a nearnormal concentration of haemoglobin and its relation to normal production of the hormone has yet to be established. For example, most patients showed a rise in reticulocyte count at a dose of 24 IU/kg thrice weekly but the subsequent increases required by our treatment schedule made it impossible to know whether this small dose would, if continued, itself have induced an adequate therapeutic response. The benefits of erythropoietin treatment revealed in this
preliminary study are an improvement in patient wellbeing and elimination of the need for blood transfusions. We foresee the possibility that treatment with rHuEPO combined with venesection may permit the correction of serious existing iron overload in patients with renal failure. There are, however, potential risks in raising the haemoglobin concentration and blood viscosity in a group of patients prone to accelerated vascular disease. We recorded two episodes of thrombosis of vascular access, one minor and one complete. Long-term observation of patients treated with rHuEPO will be required to assess the possibility of an increased incidence of thrombotic events. The episode of hypertensive encephalopathy in patient 1
to
Amgen and is being jointly developed by Amgen, Thousand Oaks, California, USA, by Cilag Schaffhausen, Switzerland, by Kirin, Tokyo, Japan, and by Ortho Pharmaceutical Corporation, Raritan, New Jersey, USA. We thank Dr W. L. Jefferson and Mr R. C. Hill of Ortho-Cilag Pharmaceutical Ltd for continuing help. .
Correspondence should be addressed to C. G. W., Department of Medicine, Royal Postgraduate Medical School, London W12 OHS. REFERENCES
JW. Hematologic problems of dialysis patients. In: Drukker W, Parsons FM, Maher JF, eds. Replacement of renal function by dialysis. Boston: Martinus Nijhoff, 1983: 630-45. 2. Eschbach JW, Adamson JW. Anemia of end-stage renal disease (ESRD). Kidney Int 1. Eschbach
1985; 28: 1-5. 3. Adamson JW, Eschbach JW, Finch CA. The kidney and erythropoiesis. Am J Med 1968; 44: 725-33. 4. Jacobson LO, Goldwasser E, Fned W, Plzak L. Role of the kidney in erythropoiesis. Nature 1957; 179: 633-34. 5. Erslev AJ. Erythropoietin function of the kidney. In: Wesson LG, ed. Physiology of the human kidney. New York: Grune and Stratton, 1969: 521-34. 6. Caro J, Brown S, Miller O, Murray T, Erslev AJ. Erythropoietin levels in uremic nephric and anephric patients. J Lab Clin Med 1979; 93: 449-58. 7. Chandra M, Miller ME, Garcia JF, Mossey RT, McVicar M. Serum immunoreactive erythropoietin levels in patients with polycystic kidney disease as compared with other hemodialysis patients. Nephron 1985; 39: 26-29. 8. Goldsmith HJ, Ahmad R, Raichura N, Lal SM, McConnell CA, Gould DA, Gyde OHB, Green J. Association between rising haemoglobin concentration and renal cyst formation in patients on long term regular haemodialysis treatment. Proc EDTA 1982; 19: 313-18.
1178
EFFECT OF DUODENAL ULCER SURGERY AND ENTEROGASTRIC REFLUX ON GAMPYLOBACTER PYLORIDIS H.
J. O’CONNOR1 J. I. WYATT2
M. F. DIXON2 A. T. R. AXON1 E. P. DEWAR3
D. C. WARD3
D. JOHNSTON3 Gastroenterology Unit1 and University Departments of Pathology2 and Surgery,3 General Infirmary at Leeds, Leeds LSI 3EX
Summary
To
assess
the effect of duodenal ulcer
Campylobacter pyloridis gastric biopsies were done and fasting bile acid concentrations in gastric aspirates were measured in 35 patients with active duodenal ulceration and 54 who had undergone surgery at some time. Biopsy specimens were assessed blind for the presence of C pyloridis and scored for severity of reflux gastritis by the use of a histological grading system. Among patients who had undergone highly selective vagotomy the proportion who were C pyloridis-positive was similar to that in the unoperated group, but among those who had undergone Billroth I partial gastrectomy, Billroth II partial surgery
on
gastrectomy, or truncal vagotomy and gastroenterostomy it The absence of C was significantly lower (p < 0·001). correlated pyloridis strongly (p<0·001) with high reflux scores and increased bile acid concentrations in the stomach. Reflux scores and bile acid concentrations were significantly higher (p<0·01) after Billroth I and Billroth II partial gastrectomies and truncal vagotomy and gastroenterostomy than in the active duodenal ulcer or highly selective vagotomy groups. There was a highly significant correlation (p<0·001) between reflux scores and bile acid concentrations. These results suggest that reflux may disrupt mucus and thus cause the death of campylobacters that live beneath it. They also suggest that reflux may produce a reflux-specific gastritis. Highly selective vagotomy may protect against these changes in the gastric mucosa.
9. Ratcliffe PJ, Dunnill MS, Oliver DO. Clinical importance of acquired cystic disease of the kidney in patients undergoing dialysis. Br Med J 1983; 287: 1855-58. 10. Shalhoub RJ, Rajan U, Kim VV, Goldwasser E, Kark JA, Antoniou LD. Erythrocytosis in patients on long term dialysis. Ann Intern Med 1982; 97: 686-90.
11. Eschbach JW, Mladenovic J, Garcia JF, Wahl PW, Adamson JW. The anemia of chronic renal failure in sheep, response to erythropoietin-rich plasma in vivo. J Clin Invest 1984; 74: 434-41. 12. Jacobs K, Shoemaker C, Rudersdorf R, et al. Isolation and characterization of genomic and cDNA clones of human erythropoietin. Nature 1985; 313: 806-10. 13. Lin F-K, Suggs S, Lin C-H, et al. Cloning and expression of the human erythropoietin gene. Proc Natl Acad Sci USA 1985; 82: 7580-84. 14. Browne JK, Cohen AM, Egrie JC, et al. Erythropoietin: gene cloning, protein structure and biological properties. Cold Spring Harbor Symp Quant Biol (in press). 15. Lai P-H, Everett R, Wang FF, Arakawa T, Goldwasser E. Structural characterization of human erythropoietin. J Biol Chem 1986; 261: 3116-21. 16. Egrie JC, Browne JK, Lai P, Lin F-K. Characterization of recombinant monkey and human erythropoietin. In: Stamatoyanopoulos G, Nienhuis AW, eds. Experimental approaches for the study of hemoglobin switching. New York: Alan R. Liss, 1985: 339-50. 17. Egrie JC, Strickland TW, Lane J, et al. Characterization and biological effects of recombinant human erythropoietin. Immutobiology (in press). 18. Egrie JC, Cotes PM, Lane J, Gaines Das RE, Tam RC. Development of valid radioimmunoassays for erythropoietin using recombinant EPO as tracer and immunogen. Blood 1985; 66: 149a. 19. Milman N, Christensen TE, Visfeldt J. Diagnostic efficiency of various laboratory tests in the assessment of bone marrow iron stores in patients with chronic uraemia. Stand j Haematol 1981; 26: 257-64. 20. Gokal R, Millard PR, Weatherall DJ, Callender STE, Ledingham JGG, Oliver DO. Iron metabolism in haemodialysis patients. Quart J Med 1979; 48: 369-91. 21. Eschbach JW, Cook JD, Scribner BH, Finch CA. Iron balance in hemodialysis patients. Ann Intern Med 1977; 87: 710-13.
Introduction
THE presence of Campylobacter pyloridis on gastric mucosal biopsies is strongly associated with peptic ulcer diseased This is particularly so for duodenal ulceration (DU), where virtually all patients have C pyloridis-positive antral gastritis.2 Although there are data on the effects of medical therapy for peptic ulcer on C pyloridis,3 the fate of this organism after peptic ulcer surgery has not been established. We have explored the effect of different types of gastric surgery on C pyloridis by studying a large series of DU patients, in some instances preoperatively and in others after a conservative type of operation (highly selective vagotomy-HSV) or after operations that remove or bypass the pylorus.
Subjects and Methods Patients The 89 patients could be divided into five groups (table)-one with active duodenal ulcers ("untreated" DU group) and four consisting of patients who had been treated surgically in the past for DU. None of the postoperative patients had recurrent ulcers. All patients gave informed written consent for the study.
Sites of Biopsies and Sampling
Techniques
All the patients underwent upper gastrointestinal endoscopy after an overnight fast. At endoscopy, at least two gastric mucosal biopsies were taken from within 5 cm of the pylorus from untreated patients and those who had undergone HSV and from within 5 cm of the stoma from patients in the Biltroth I partial gastrectomy (BIPG), Billroth II partial gastrectomy (BIIPG), and truncal vagotomy and gastroenterostomy (TVGE) groups. In the intact stomach the presence of organisms in the antrum affected by gastritis is almost invariably accompanied by the existence of organisms in biopsy specimens of the body of the stomach even when the body mucosa looks normal 4 We therefore consider it valid to compare the prevalence of C pyloridis in the pre-pyloric mucosa of the intact stomach with the peristomal mucosa of the postoperative stomach. In 71 patients (80%) fasting gastric contents were aspirated for measurement of total bile acid concentration (BAC). In 25 patients gastric aspirates (5-10 ml) were obtained at endoscopy, by the use of a sterile ’Teflon’ cannula passed down the biopsy channel of the endoscope whereas in the other 46 patients contents were aspirated through a size 12 nasogastric tube positioned in the stomach by the water recovery method5 after a further overnight fast.
Histological Assessment of Gastric Mucosal Biopsy Specimens were oriented on filter paper and fixed in formol saline. Paraffm-processed sections immediately were cut at three levels and stained by haematoxylin and eosin and alcian-blue pH 25/PAS. A further section was taken from each set of biopsy specimens and stained by the undifferentiated Giemsa method.2 These sections were examined "blind" by J. I. W. for the presence of C pyloridis.
The
biopsy specimens
PATIENT GROUPS