NEFROLOGIA. Vol. XIV. Núm. 3. 1994 Individual differences in the response to recombinant human erythropoietin therapy Lj. Djukanovic, C. K. Clemons*, V. Leiaic', A. Radmilovic, M. Milosavljevic', M. Gajic y V. PavlovicKentera** Department of Nephrology, University Clinical Center, Beograd, Yugoslavia. * Lawrence Berkeley Laboratory, University of California, Berkeley, California, USA. ** Institute for Medical Research, Beograd, Yugoslavia. DIFERENCIAS INDIVIDUALES EN LA RESPUESTA A LA ERITROPOYETINA H U M A N A RECOMBINANTE RESUMEN Se administró eritropoyetina por vía subcutánea a 20 enfermos en prediálisis y 21 en hemodiálisis. La anemia mejoró en todos los casos, si bien se observaron importantes diferencias individuales en cuanto al ritmo de ascenso de la hemoglobina. Del análisis de los factores potencialmente reponsables de las diferencias individuales no se encontraron diferencias entre ambos grupos de enfermos. Ni la edad, peso, ni los niveles séricos de creatinina, hierro, ferritina, transferrina o los índices de saturación de la transferrina tuvieron influencia significativa sobre el ritmo de incremento de la hemoglobina. Se encontró una correlación negativa estadísticamente significativa entre los niveles basales de hemoglobina y eritropoyetina y el ritmo de incremento de la hemoglobina. La respuesta al tratamiento con eritropoyetina dependió también de la nefropatía de base. Los enfermos con glomerulonefritis y nefropatía de los Balcanes respondieron mejor al tratamiento con eritropoyetina que aquellos con nefropatías tubulointersticiales distintas de la nefropatía de los Balcanes. Esta mejor respuesta al tratamiento con eritropoyetina en enfermos con determinados procesos está de acuerdo con la correlación negativa previamente establecida entre los niveles iniciales de hemoglobina y eritropoyetina y la respuesta al tratamiento. Palabras clave: Eritropoyetina humana recombinante. Enfermos en predialisis. Enfermos en hemodialialis. Respuesta terapéutica. Recibido: 5-III-93. En versión definitiva: 20.XII-93 Aceptado: 28-XII-93. Correspondencia: Prof. Ljubica Djukanovic' Department of Nephrology. University Clinical Center. Pasterova 2. 11000 Beograd. Yugoslavia. 316 DIFERENCIAS INDIVIDUALES EN LA RESPUESTA A LA EPO SUMMARY R e c o m b i n a n t human erythropoietin (rHuEpo) was administered subcutaneously to 20 predialysis and 2 1 hemodialysis (HD) patients. Anemia improved in all patients but individua/ differences in the rate of hemoglobin (Hb) increase were noted. The analysis of the factors possibly responsible for the individual differences in the response to rHuEpo therapy revealed the same factors correlating with Hb inc r e a s e rate in both predialysis and HD patients. Patient age, body weight, serum c r e a t i n i n e , iron, ferritin, transferin levels and transferin saturation had no significant influente on the rate of Hb increase. Statistically significant negative correlat i o n between initial Hb and erythropoietin levels (Epo) and Hb increase rate was found. The response to rHuEpo therapy also depended on the underlying kidney disease. Patients with glomerulonephritis (GN) and Balkan endemic nephropathy (BEN) responded better to rHuEpo therapy than those with tubulointerstitial nephropathy other than BEN. This better response to rHuEpo therapy in patients with GN and BEN is in accordance with already established negative correlation between the initial both Hb and Epo levels and response to rHuEpo therapy. Key words: Recombinanf human eryfhropoiefín. Predialysis patienfs. Hemodialysis pafienfs. Therapy response. Introduction Human recombinant erythropoietin (rHuEpo) is widely used for correcting anemia of chronic renal failure (CRF) patients. Initial studies have been carried out in patients on regular hemodialysis 1-3, but successful results in predialysis CRF patients 4-7, CAPD patients 8 as well as in patients with kidney graft failure 9 were reported. Anemia improvement was reported in majority of patients treated with rHuEpo. Nevertheless, individual differences in the rate and degree of the improvement were noted but not entirely elucidated 3,6,7,10,11. The principal objective of this study was to analyze the factors possibly responsible for the individual differences in the response to rHuEpo therapy. Patients and methods Patients The study comprised 20 patients with CRF not yet requiring dialysis (predialysis patients) and 21 patients with end-stage renal disease on maintenance hemodialysis (HD patients). The predialysis patients group consisted of 13 women and 7 men with stable serum creatinine levels between 315 to 835 µmol/l. The diagnosis of renal disease was established by clinical and laboratory criteria and renal biopsy in patients with glomerulonephritis (GN). Primary renal disease was: mesangiocapillary CN in 2, focal segmental glomerulosclerosis (FSG) in 2, Balkan endemic nephro- pathy (BEN) in 5, reflux nephropathy in 2, pyelonephritis in 2, polycystyc kidney disease in 4 and hypertensive nephrosclerosis in 3. Twenty-one HD patients, 14 women and 7 men, on regular hemodialysis for 13 to 210 months (71 .6 f 58.8) were dialyzed three times a week for 4 hours using cuprophan membrane dialyrers. Diagnosis of their renal disease included GN in 5 (mesangiocapillary 1, FSG 3, membranous l), BEN in 4, reflux nephropathy in 3, pyelonephritis in 3, hypertensive nephrosclerosis in 2, nephrolithiasis in 2, while 2 patients were anephric. All patients suffered anemia with hemoglobin (Hb) level less than 82 g/l at the outset of the rHuEpo therapy. The exclusion criteria for the study were: anemia due to causes other than CRF, uncontrolled hypertension, uncontrolled diabetes mellitus, abnormal liver function, pregnancy, therapy known to affect erythropoiesis, folate, B12 or iron deficiency. The study was approved by the Ethics Commitee of the University Clinical Center. Written informed consent was obtained from each patient. Study Design The rHuEpo used was kindly given by Cilag AG lnternational and it was subcutaneously administered. The starting dose of 50 U/kg body weight (bw) rHuEpo three times a week was maintained for four weeks. As the target Hb level of 100-120 g/l had not been achieved at the end of four-week period, the dose was increased by 25 U/kg bw. When the target Hb level was achieved the rHuEpo dose was regu317 Lj. DJUKANOVIC y cols. l a r l y adjusted to maintain a Hb level of about 100 g/l. The patients were followed up for 3 to 26 months. The dietary protein intake was limited to 0,6 g/kg/day in predialysis patients, while normal protein intake was recommended to HD patients. Iron supplementation was given to patients with normal or lou serum iron level or ferritin levels 100 µg/l. Antihypertensive therapy was adjusted to keep blood pressure normal. A complete blood count was done once a week during the first eight-weeks and once a fortnight later. Blood chemistry including urea, creatinine, electrolytes, liver function test, serum iron, transferin, transferin saturation, total iron binding capacity and serum ferritin was performed once a month. Serum erythropoietin levels (Epo) were measured by radioimmunoassay at L.awrence Berkeley Laboratory( Berkeley, California in 15 patients at the outset of the study. The control value of serum Epo in healthy nonanemic volunteers was 17.8 * (SD) 4.2 mU/ml 12. Intact parathormon (PTH) plasma levels were determined by radio-immunodssay (Nichols Institute Diagnostics, normal range 10-53 pg/ml). Statistical analyses pneumonia in eighth week of therapy wich caused anemia and renal function deterioration. Otherwise, the improvement of anemia due to rHuEpo therapy had no effect either on predialysis serum creatinine levels or on CRF progression expressed by the slope of regression line generated by plotting 1/serum creatinine against time (table I). Table I. Effects of rHuEpo therapy on anemia and serum creatinine levels Patients Predialysis Number. 20 Age , years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 7 . 6 + 13.3 rHuEpo dose ,U/kg/wk: maximal l . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 0 5 . 3 * 49.0 . - maintenance. . . . . . . . . . . . . . . . . . . . . . . 70.6 ? 20.3 Hemoglobine g/l: , 74.2 + 5.2 initial -at the end 104.6+13.1 No of weeks untill Hht 7.1 f 2.1 Hb increase, g/l/wk 4.7 f 1.9 s-creatinine umol/l: , 508 f 143 initial - at the end 690 f 366 slop 1/Cr VS time: prior therapy -0.111 f 0.99 -during therapy .-0.103 f 0,95 Hemodialysis 21 43.6 i 12.8 196.1 f 52.2 66.1 + 32.8 64.7 f 9.2* 101.3+9.8 8.2 f 2.5 4.6 + 1.6 1013*203 1069 i 265 The rate of Hb increase during initial phase of rHuEpo therapy was calculated as follows: Hb increase = Hbt - HbO/number of weeks until Hb, were Hbt is the first Hb value in the target Hb interval (100-120 g/l), HbO the Hb value at the outset of the study. CRF progression prior to and during rHuEpo therapy was expressed by the slope of the regression line obtained by plotting the reciprocals of serum creatinine against time for these two periods 13. The review periods were four to six months prior to therapy and three to 18 months under therapy. The number of observations per patients for each period ranged from 4 to 19. The data were analyred using linear regression analysis, analysis of variance and Student t-test. Results The results presented in table I show that HD patients had more severe anemia at outset of the study than predialysis patients. Treatment with rHuEpo improved anemia in all patients, but individual differences in the rate and the degree of Hb increase as well as in the rHuEpo doses necessary for reaching and maintaining the target Hb level were noted in both predialysis and HD patients. The target Hb levels reached in the fifth to twelfth week of therapy in all but two predialysis patients. In one patients iron deficiency due to patient irregular use of iron supplements was discovered. Another patient suffered 318 rHuEp odos e maximal-dose e necessary fo r reachinggtarget H b level (Hbt); H b increase rate was calculatedas dcscribe din Methods .* p < 0.01 , campared with predialysis patients. The rate of Hb increase during the initial phase of rHuEpo therapy widely varied from 1.5 to 7.4 g/l/ week. The coefficients of correlation between the rate of Hb increase and the factors wich might influence the response to rHuEpo therapy are presented in table II. The patient with iron deficiency and that Table II. Coefficients of correlation between the rate of hemoglobin increase and the relevant variables variables patients Predialysis spatients Hemodialysis s r Body weight. .............. H b , initial l . . . . . . . . . . . . . . . . . . Rt, initial .. . . . . . . . . . . . . . . . . . . R t , maximal l . . . . . . . . . . . . . . . s - H u E p o dose * .......... r - c r e a t i n i n e e. . . . . . . . . . . . . . . . s - i r o n" . . . . . . . . . . . . . . . . . . . . . . . . . s - T I B C ........................ 5.transferin ..-............... transferin nsaturatio n.. . s-ferritin n. . . . . . . . . . . . . . . . . . . . . PT H ........................... 0.088 0.084 -0.492 0.339 0.496 -0.499 -0.249 0.195 0.142 0.058 0.255 0.286 0.332 P 0.364 0.371 0.019 0.090 0.021 0.018 0.160 0.219 0.303 0.409 0.323 0.125 0.090 r 0.098 0.105 -0.471 0.335 0.507 -0.423 -0.354 0.335 0.196 0.102 0.212 0.335 0.350 P 0.672 0.650 0.031 0.148 0.019 0.063 0.118 0.148 0.42 1 0.661 0.355 0.148 0.119 Rt: retirulocyte. * maxima, rHuEpo weekl ydos enecessar yo r rearhin g rgett Hb level. f ta DIFERENCIAS INDIVIDUALES EN LA RESPUESTA A LA EPO Table III. Mean (SD) serum iron, ferritin and transferin levels and intact PTH plasma levels in patients groups with different underlying kidney disease at the outset 1and at the end 2 of the study GN 1 16.3 (4.1) 184 (51) 2.1 (0.4) 92 (361 Variables s-iron, ,umol/l s-ferritin pg/l s-transferin, g/l BEN 2 12.9 (1.8)* 163 (20) 2.0 (0.3) TIN 2 16.1 (8.1) 189 (21) 1.9 (0.3) -. 1 19.7 (8.0) 196 (28) 1.9 (0.4) 89 (33; 1 19.5 (9.1) 216 (90) 2.0 (0.4) 101 (39) 2 15.3 (8.4) 181 (67) 1.9 (0.4) _ PTH, pg/ml GN: elomeruloneohritis: BEN: Balkan endemic ner~hronathv: TIN: tubulointerstitial nephropñthy othel thsn BEN. with pneumonia were excluded from this analysis. The results presented showed statistically significant negative correlation between initial Hb level and Hb increase rate in both patient groups examined. No significant correlation between age, body weight or s e r u r n creatinine, PTH, iron, total iron binding capacity, ferritin, transferin level, transferin saturation and the rate of Hb increase w a s noted. Searching futher for factors influencing different response to rHuEpo therapy we compared the Hb in crease rate for paticnt groups formed according to underlying kidney disease (fig. 1). Only the groups consisting of more than five patients were compared. Hb increase rate was significantly higher for the patients with GN and BEN than for patients with tubulointerstitial nephropathy other than BEN (TIN) (reflux nephropathy and pyelonephritis). At the same time patients with GN and BEN had significantly lower initial Hb levels than patients with TIN. No significant difference in serum iron, ferritin, transferin levels as well as serum PTH levels was found between the three etiologically different groups (table III). In fig. 1 the serum Epo levels for patients with different kidney disease are also presented. The results indicated that patients with BEN and GN, who respoded faster to rHuEpo therapy, had lower initial serum Epo levels than patients with TIN. That directed us to examine the relationship hetween initial serum Epo l e v e l and the rate of Hb increase during rHuEpo ther a p y At the outset of the study serum Epo level was measured in 15 patients including 12 presented in fig 1 . The regression analysis, which compared all these Epa leveIs to Hb increase rate during rHuEpo ther a p y , revealed significant negative Iinear correlation between these two variabies (fig. 2). 0 I 10 20 30 40 5õ Epo. mU$ Fig 2.-Relationship between the rate of hemoglobin i n c r e a s e during rHuEpo therapy (Hbi ) and serum erythropoietin levels (Epo) at the outset of the study. o: predialysis patienh; l : hemodialysis patients 3ñ2 6 0 6 t r BEN l 0 s 2 4 " CN TIN CN BEN TIN CN BEN TIN Fig 1 .-The rate of hemoglobin increase during rHuEpo therapy (Hbi), initial hemoglobin Epo levels in predialysis (0) and hemodialysis (0) patients with different kidney disease. values (Hbo) and initial serum immunoreactive 319 Lj. DJUKANOVIC y cols. Discusión The correlation of anemia in predialysis and HD patients with rHuEpo therapy found here was expected as confirmed in earlier clinical studies 1-7. The increase Hb level in the response to rHuEpo therapy has been shown to be dose dependent 4.5.14. However, individual differences in the response to rHuEpo therapy was indicated in predialysis patients 6,7,11 as had previously been in HD patients 3,10,15. This study has focused individual variations in the response to rHuEpo therapy. So far, several studies have reported the causes of resistance to rHuEpo therapy but only a few have discussed the factors which might influente rHuEpo therapy response. As the cause of resistance to rHuEpo therapy iron and vitamine deficiency 16, 17, aluminium intoxication 18, hyperparathyroidism, acute or chronic inflammation 7 were reported. In two of our patients a poor response to rHuEDo therapy could be explained by iron deficiency or acute infection. In the remaining patients neither one of the causes of resistance to rHuEpo therapy was known. Niether one of predialysis patients was treated with aluminium hydroxid. Aluminium intoxication was not excluded in our HD patients but the dose of aluminium hydroxid in our patients was never higher than 1.6 gr/day. Therefore, aluminium intoxication most probably was not of influente on rHuEpo therapy response. Nevertheless, the significant individual differences in the Hb increase rate were noted. The statistical analysis revealed that the same factors correlated with the Hb increase rate during rHuEpo therapy in both predialysis and HD patìents. Individual variations were not age, body weight or renal failure degree dependent. Absence of correlation between serum iron, ferritin, transferin level and transferin saturation at the outset of the study and the rate of Hb increase under rHuEpo therapy revealed that individual variations were not the results of different iron status in our patients. In contrast to the other authors who found no correlation between underlying kidney disease and the response to rHuEpo therapy 6,10, our patients with GN and BEN responded better to rHuEpo therapy than those with TIN other than BEN. The difference in rHuEpo therapy response in patients with different underlying kidney disease was not caused by different iron status or degree of secondary hyperparathyroidism. In our previous studies patients with GN and BEN with advanced renal failure and those on maintenance HD had lower Hb and serum Epo levels than patients with pyelonephritis 20. In the present study patients with GN and BEN had significantly lower initial Hb level than patients with TIN other than BEN. The better response to rHuEpo therapy in patients with GN and BEN was in accordance with the 320 finding of significant negative correlation between initial Hb level and the response to rHuEpo therapy found here for the whole group examined. This correlation between initial Hb levels and rHuEpo therapy response contrasts with the results reported previously in hemodialysis 10 and predialysis 11 patients. Only Kleinman et al 6 reported the most rapid increase in hematocrit in two patients with the most severe anemia. The higher rate of Hb increase in patients with lower initial serum Epo levels same as lower Hb values seems to be important but can not be explained by the results presented here. It could be suggested though that is due to the difference in sensitivity to Epo of erythroid precursors in the bone marrow of these patients. References 1. W i n e a r l s CG, Oliver DO, Pippard MJ, et al: Effect of human e r y h t r o p o i e t i n derived from recombinant DNA on the anemia o f patients maintained by chronic haemodialysis. Lancet 11:1175-1178, 1986. 2 E s c h b a c h JW, Dowing MR, Egrie JC, et al: USA multicentric c l i n i c a l trial with recombinant human erythropoietin (Amgen) r e s u l t s in hemodialysis oatients. Contrib Neohrol 76:160-165, 1989. 3. Suzuki M, Hirosawa Y, Hiroshim K, et al: Dose-finding, doub l e blind, clinical trial of recombinant human erythropoietin (Chugai) in Japonese patients with end-stage renal disease. Contrib Nephrol 76:179-192, 1989. 4. Frenken LAM, Verberckmoes R, Michielsen P and Koene RAP: Efficacy and tolerance of treatment with recombinant human e r y t h r o p o i e t i n in chronic renal failure (pre-dialysis) patients. Nephrol Dial Transplant 41782-786, 1989 . 5. Lim VS, Degowin RI-, Zavala D, et al: Recombinant human e r y t h r o p o i e t i n treatment in predialysis patients. Ann Int Med 11 0:108-114, 1989. 6. Kleinman KS, Schweitzer SU, Perdue ST, et al: The use of recombinant human erythropoietin in correction of anemia in predialysis patients and its effect on renal function: a double b l i n d , placebo-controlled trial. Am J Kidney Dis 14:486-495, 1989. 7. Lim VS, Fangman J, Flanigan MJ, et al: Effect of recombinant human erythropoietin on renal function in humans. Kidney Int 37:131-136, 1990. 8. Cheng JKP, Cy C, Chan MK, et al: Correction of anemia in patients on continuous ambulatory peritoneal dialysis with subc u t a n e o u s recombinant eryhtropoietin twice a week: a longterm study. Clin Nephrol 35:207-212, 1991. 9. Y o s h i m u r a N, Oka T, Ohmori Y and Aikawa I: Effects of recombinant human erythiopoietin on the anemia of renal transp l a n t recipients with chronic rejection. Transplantation 527529,1989. 10. Sundal E and Kaeser U: Correction of anemia of chronic renal f a i l u r e with recombinant human erythropoietin: safety and efficacy of one years treatment in a European multicentric study o f 150 haemodialysis-dependent patients. N e p h r o l D i a l Transplant 4:979-987, 1989. 11 Stone WJ, Graber SE, Krantz SB, et al: Treatment of the anem i a of predialysis patients with recombinant human erythrop o i e t i n : a randomized, placebo-controlled trial. Am J Med Sci 296:171-179,1988. DIFERENCIAS INDIVIDUALES EN LA RESPUESTA A LA EPO 12. García JF, Sherwood J and Goldwasser E: Radioimmunoassay of erythropoietin. Blood Cells 5:405-419, 1979. 13. Mitch WE, Walser M, Buffington GA and Leman JI: A simple method of estimating progression of chronic renal failure. Lancet 11:1326- 1328, 1976. 14. Teehan BP: For use US Recombinant Human Eryhtropoietin Predialysis Study Croup: Double-blind, placebo controlled study of the therapeutic use of recombinant human erythropoietin for anemia associated with chronic renal failure in predialysis patients. Am J Kidney Dis 18:50-59, 1991 . 15. Casati S, Passerini P, Campise MR, et al: Benefits and risks of protracted treatment with human recombinant erythropoietin in patients having hemodialysis. Brit Med J 295:117-120, 1987. 16. MacDougall IC, Hutton RD, Cavill J, et al: Poor response to treatment of renal anemia with erythropoietin corrected by iron gen intravenously. Brit Med J 299:157-158, 1989. 17. Stivelman J: Resistence to recombinant human erythropoietin therapy. A real clinical entity? Seminars Nephrology (Suppl): 8- 11, 1989. 18. Casati S, Castelnovo C, Campise M y Ponticelli C: Aluminium interferente in the treatment of hemodialysis patients with recombinant human erythropoietin. Nephrol Dial Transplant 5:441-443, 1990. 2 0 . Pavlovic-Kentera V, Clemons CK, Trbojevic S, et al: Eryhtropoietin and anemia in the progression of Balkan endemic nephropathy and other renal diseases. Nephron 54:139143, 1990. 321