Bone Marrow Transplantation, (1997) 19, 771–775  1997 Stockton Press All rights reserved 0268–3369/97 $12.00

Serum thrombopoietin levels in patients undergoing autologous peripheral blood stem cell transplantation

C Shimazaki1, T Inaba2, H Uchiyama1, T Sumikuma1, T Kikuta1, H Hirai1, Y Sudo1, N Yamagata1, E Ashihara1, H Goto1, S Murakami3, H Haruyama3, N Fujita2 and M Nakagawa1

1Second Department of Medicine; 2Department of Clinical Laboratory Medicine, Kyoto Prefectural University of Medicine; and 3Department of Medicine, Shakaihoken Kyoto Hospital, Kyoto, Japan

Summary: gated in patients receiving BMT and PBSCT.5–11 A strong inverse correlation between serum granulocyte Recently, the ligand for c-mpl has been cloned and colony-stimulating factor (G-CSF) level and absolute neu- initial studies have shown it to be the regulatory trophil count has been demonstrated,5–9 which suggests that factor, thrombopoietin (TPO). To elucidate the role of G-CSF plays a critical role in the reconstitution of myeloid TPO in the reconstitution of megakaryopoiesis and lineage progenitors after stem cell transplantation. How- platelet production after stem cell transplantation, we ever, of the growth factors studied, including -3 measured serum TPO levels in nine patients undergoing (IL-3), interleukin-6 (IL-6) and inhibitory factor autologous peripheral blood stem cell transplantation (LIF), none have been demonstrated to correlate with plate- (PBSCT) and in healthy volunteers. Serum TPO levels let recovery after stem cell transplantation.6,9–11 A possible significantly correlated with the degree of peripheral exception was reported by Testa et al11 who showed that and a strong inverse correlation peak IL-6 levels are directly correlated with the peak of between serum TPO level and platelet count was platelet recovery. observed (r =−0.700, P Ͻ 0.001). Serum TPO levels Recently, the ligand for c-mpl has been cloned; in vitro began to rise as the platelet count decreased after and in vivo studies have shown that it stimulates both . TPO levels peaked at over 25.00 megakaryocytopoiesis and platelet production, suggesting fmoles/ml between days 0 and 10; TPO levels then that it is the long-sought platelet regulatory factor, throm- decreased gradually as the platelet count began to rise. bopoietin (TPO) itself.12–15 To elucidate the role of TPO One patient with multiple myeloma received purified in megakaryopoiesis and platelet recovery after stem cell CD34+ peripheral blood stem cells. No difference was transplantation, we measured serum TPO levels in patients observed in the kinetics of serum TPO levels between undergoing autologous PBSCT using a recently established unfractionated and purified PBSCT. These observations sensitive sandwich enzyme-linked immunosorbent assay suggest that TPO plays a critical role in the reconsti- (ELISA).16 tution of megakaryopoiesis and platelet production after PBSCT. Keywords: thrombopoietin; PBSCT; platelet Materials and methods

Patients High-dose chemotherapy followed by peripheral blood Nine patients with various types of malignancies were stem cell transplantation (PBSCT) has become an option evaluated (Table 1). There were seven males and two for the treatment of high-risk hematological malignancies females, age range 18 to 67 years. Diagnosis included non- and solid tumors.1–3 Following myeloablative chemo- Hodgkin’s lymphoma (NHL) (four), acute myelogenous therapy, the recovery of hematopoiesis is dependent on leukemia (AML) (two), acute lymphoblastic leukemia stem cell self-renewal capacity and differentiation to (ALL) (one), multiple myeloma (MM) (one) and small cell lineage-committed progenitors, which undergo further dif- lung cancer (SCLC) (one). ferentiation and then maturation to morphologically recog- nizable precursor cells and terminal cells circulating in per- ipheral blood.4 Hematopoietic growth factors are thought Transplant procedures to play a role in the engraftment of stem cells following The details of peripheral blood stem cell (PBSC) collection bone marrow transplantation (BMT) or PBSCT. Thus far, and PBSCT procedures have been previously described.17 serum levels of several growth factors have been investi- Briefly, patients were treated with a regimen of high-dose cytosine arabinoside (Ara-C) (12 g/m2), high-dose cyclo- phosphamide (CY) (4 g/m2), or high-dose etoposide (VP- Correspondence: Dr C Shimazaki, Second Department of Medicine, Kyoto 2 Prefectural University of Medicine, 465 Kawaramachi-Hirokoji, Kami- 16) (2 g/m ) followed by subcutaneous administration of gyoku, Kyoto 602, Japan recombinant human granulocyte colony-stimulating factor Received 15 July 1996; accepted 30 November 1996 (rhG-CSF) (filgrastim) 50 ␮g/m2. PBSC were collected Serum TPO levels after PBSCT and BMT C Shimazaki et al 772 during leukocyte recovery using a CS3000 blood cell separ- microplate reader (International Reagents Corporation, ator (Fenwal, Deerfield, IL, USA), and stored at −130°C. Kobe, Japan). The absorbance of each sample was sub- In the MM patient, CD34+ cells were separated from total tracted from that of each sample incubated with TN1. The PBSC using an Isolex system (Baxter Healthcare Immuno- average value of each standard or high TPO serum sample therapy Division, Round Lake, IL, USA). The pre-trans- was subtracted from that of the blank for the standard. The plant conditioning regimen for AML consisted of busulfan sample concentration was calculated by regression analysis (BU) (16 mg/kg) and CY (120 mg/kg); that for NHL con- for the standard curve. sisted of CY (120 mg/kg), VP-16 (1500 mg/m2) and rani- This ELISA system was highly sensitive and specific for mustin (MCNU) (500 mg/m2); that for MM consisted of human TPO. The lower limit of detection was 0.09 melphalan (L-PAM) (200 mg/m2); and that for SCLC con- fmoles/ml in serum and no significant cross-reactivity to sisted of CY (3000 mg/m2), VP-16 (1000 mg/m2) and car- other was observed. boplatin (CBDCA) (1200 mg/m2). PBSC were infused on day 0. All patients received rhG-CSF at a dose of 50 ␮g/m2 Statistical analysis subcutaneously from day 1 until the neutrophil count exceeded 0.5 × 109/l. A software program, StatView 4.0 (Abacus Concepts, Berkeley, CA, USA), was used for statistical analyses. The relationship between hematologic data and serum TPO lev- Serum samples els was evaluated by Student’s t-test. A P value of Ͻ0.05 After informed consent had been obtained, blood samples was accepted as statistically significant. Correlation stat- were drawn from the nine patients undergoing PBSCT. istics were determined by Pearson’s correlation coefficient. Samples were collected on day −8, and every 2–4 days thereafter until platelet recovery. Serum was separated by centrifugation shortly after collection and was frozen at Results −80°C until assayed. Control subjects Measurement of serum level of TPO Serum TPO levels obtained from five healthy volunteers were 0.41 Ϯ 0.23 fmoles/ml (mean Ϯ s.d.). TPO levels in serum were measured by a sandwich enzyme-linked immunosorbent assay (ELISA).16 Each well in a 96-well flat-bottomed microtiter plate (Maxisorp, PBSCT patients Nunc, Roskilde, Denmark) was coated at 4°C overnight Platelet recovery in the nine patients receiving PBSCT was with 100 ␮l of TN1 (mouse anti-human TPO monoclonal rapid except for the two patients with AML and the median antibody) (Kirin Pharmaceutical Research Laboratory, time to reach 50 × 109/l was 16 days (Table 1). Gunma, Japan) at a concentration of 10 ␮g/ml in 50 mm The relationship between the kinetics of serum TPO and carbonate buffer (pH 9.4). After washing twice with 20 mm platelet counts in the blood of seven patients with the fastest Tris-HCl containing 0.5 m NaCl and 0.1% NaN (pH 7.5) 3 platelet recovery is shown in Figure 1. Serum TPO levels (TBS), 200 ␮l of a blocking reagent (Super Block in TBS; began to rise as the platelet count decreased after high- Pierce, Rockford, IL, USA) was added to the wells and dose chemotherapy. The TPO level peaked at over 25.00 incubated at room temperature for 30 min. After the block- fmoles/ml between days 0 and 10 and levels then gradually ing reagent was aspirated, 100 ␮l of recombinant human (rh) TPO standard, serum test sample or blank were added to each well and reacted with the coated TN1 at room tem- perature overnight. After washing with 20 mm Tris-HCl 300 40

containing 0.5 m NaCl, 0.05% Tween 20 and 0.1% NaN3 (T-TBS), 100 ␮l of biotinylated anti-rhTPO at a concen- 250 30

tration of 500 ng/ml in dilution buffer (T-TBS containing /l) 1% bovine serum albumin and 2% PEG 6000, pH 7.5) was 9 200 10 added to each well and incubated at room temperature for × 150 20 3 h. After washing with T-TBS, 100 ␮l of alkaline phospha- tase-conjugated streptavidin (1 mU/ml in dilution buffer;

100 TPO (fmoles/ml)

Boehringer Mannheim, Mannheim, Germany) was added to Platelets ( 10 each well, then incubated at room temperature for 1 h. The 50 color was developed using an amplification system (Gibco BRL, Gaithersburg, MD, USA). After washing with T- 0 0 TBS, 50 ␮l of substrate solution was added to each well, –10 –5 0 5 10 15 20 and incubated at room temperature for 40 min. Then 50 ␮l Days after PBSCT of amplifier solution (NADPH-amplification system; Gibco) was added to each well and incubated at room tem- Figure 1 Relationship between serum TPO level (¼) and absolute plate- let count (ș) in seven patients undergoing peripheral blood stem cell trans- perature for 30 min. Color development was stopped by plantation (PBSCT). Serum TPO levels rose immediately after PBSCT, adding 50 ␮l of 0.3 m H2SO4. The color intensity was meas- reaching a peak between days 0 and 10. TPO levels decreased as the ured using the A630 nm subtracted from the A492 nm on a platelet counts rose. Error bars indicate the standard error of the mean. Serum TPO levels after PBSCT and BMT C Shimazaki et al 773 50 Discussion

40 This study demonstrated a significant correlation between the degree of thrombocytopenia post-transplantation and endogenous TPO production. A strong inverse correlation 30 between serum TPO levels and circulating platelet counts was observed in patients undergoing PBSCT and purified + 20 CD34 PBSCT. These observations suggest that TPO plays a critical role in the reconstitution of megakaryocytopoiesis TPO (fmoles/ml) and platelet production after stem cell transplantation. 10 To date, various cytokines including IL-3,18 IL-6,19 IL- 1120 and LIF21 have been shown to stimulate the growth of 0 megakaryocytic progenitors. However, in vitro studies have 0 100 200 300 400 500 shown that the effects of these cytokines are less potent than those of TPO, and no direct correlation has been Platelets (×109/l) observed between the levels of these cytokines and platelet Figure 2 Inverse correlation between serum level of TPO and absolute recovery after stem cell transplantation.6,9,10 Recently, platelet count (r =−0.70, P Ͻ 0.001). Kuter and Rosenberg22 have determined the relationship between blood levels of TPO and changes in the circulating platelet mass in rabbits treated with busulfan. As the plate- let mass declined, levels of TPO increased inversely and decreased as the platelet count began to rise (Figure 1). proportionally and peaked during the platelet nadir. Bern- Serum TPO levels significantly correlated with the degree stein et al23 reported a significant inverse relationship of peripheral thrombocytopenia and an inverse correlation between platelet counts and plasma TPO levels but not was observed between serum levels of TPO and the absol- those of IL-3, IL-11 or LIF during the course of chemo- ute platelet count (r =−0.70, P Ͻ 0.001) (Figure 2). In the therapy in patients with AML. In addition, Nichol et al24 patient with myeloma who received purified CD34+ PBSC, demonstrated that the platelet nadir was always associated the times required for neutrophil and platelet recovery were with the peak of serum TPO which returned towards base- similar to those for unfractionated PBSCT; the times to line as platelet counts recovered in patients receiving reach 0.5 × 109/l neutrophils and 50 × 109/l platelets were PBSCT with or without bone marrow. These observations, 12 and 18 days, respectively. In this patient, the serum TPO together with ours, suggest that TPO is a major growth level reached a significantly elevated level on day 0, and factor for platelets. Two mechanisms have been postulated high levels continued until the platelet count began to rise. concerning the regulation of serum TPO levels. One poss- The kinetics of serum TPO levels were similar in patients ible mechanism of TPO clearance from serum is receptor- receiving unfractionated PBSCT. In the two patients with mediated degradation of TPO by and plate- AML, who showed delayed platelet recovery, serum TPO lets.22,25,26 Based on the finding that platelets bear receptors levels rose immediately after PBSCT and high levels con- for TPO, Kuter and Rosenberg22 have proposed that TPO tinued until the platelet count began to rise. expression is constant and that serum levels are con-

Table 1 Clinical data and serum TPO level after peripheral blood stem cell transplantation

Case Age/Sex Diagnosis No. of cells infused TPO Days to recovery peak value MNC CFU-GM(fmoles/ml) ANC Ͼ 0.5 Platelets Ͼ 50 (×108/kg) (×105/kg) (×109/l) (×109/l)

1 67/M SCLC 1.1 3.7 27.00 9 10 2 42/F NHL 4.0 15.2 59.00 9 9 3 18/M NHL 2.3 5.6 31.50 17 12 4 20/M NHL 3.0 21.0 48.25 9 15 5 47/F NHL 2.6 22.4 34.81 11 16 6 34/M AML 2.5 6.4 43.00 14 40 7 40/M AML 1.9 3.0 35.75 11 63 8 36/M ALL 3.2 14.6 40.62 10 18 9a 31/M MM 1.7b 1.4 27.25 12 18

M = male; F = female; SCLC = small cell lung cancer; NHL = non-Hodgkin’s lymphoma; AML = acute myelogenous leukemia; MM = multiple myeloma; ALL = acute lymphoblastic leukemia; MNC = mononuclear cells; CFU-GM = colony forming units-granulocyte macrophage; TPO = thrombopoietin; ANC = absolute neutrophil count. aPurified CD34+ PBSCT. bNumber of CD34+ cells (×106/kg). Serum TPO levels after PBSCT and BMT C Shimazaki et al 774 trolled by the platelet mass through uptake and metabolism. 9 Uchiyama H, Shimazaki C, Fujita N et al. Kinetics of serum Higher platelet counts would result in increased catabolism cytokines in adults undergoing peripheral blood progenitor of TPO, leading to a lower serum TPO level. Similar recep- cell transplantation. Br J Haematol 1994; 88: 639–642. tor-mediated clearance mechanisms have been proposed for 10 Baiocchi G, Scambia G, Benedetti P et al. Autologous stem G-CSF,27,28 M-CSF29 and GM-CSF.30 Another potential cell transplantation: sequential production of hematopoietic cytokines underlying granulocyte recovery. Cancer Res 1993; mechanism is feedback regulation at the level of gene 31 32 53: 1297–1303. expression, as seen with . McCarty et al 11 Testa U, Martucci R, Rutella S et al. Autologous stem cell have reported that TPO mRNA cannot be detected in the transplantation: release of early and late acting growth factors bone marrow and spleen of monkeys during steady state relates with hematopoietic ablation and recovery. Blood 1994; but does appear during a thrombocytopenic state. No data 84: 3532–3539. have been reported on the site of TPO production in 12 de Sauvage FJ, Haas PE, Spencer SD et al. Stimulation of humans. Further study is required to elucidate these hypo- megakaryopoiesis and thrombopoiesis by the c-Mpl ligand. theses. Nature 1994; 369: 533–565. In this study, no differences in the kinetics of serum TPO 13 Kaushansky K, Lok S, Holly RD et al. Promotion of mega- levels were observed between patients undergoing PBSCT karyocyte progenitor expansion and differentiation by the c-Mpl ligand thrombopoietin. Nature 1994; 369: 568–571. and purified CD34+ PBSCT although the number of + 14 Bartley TD, Bogenberger J, Hunt P et al. Identification and patients receiving purified CD34 PBSCT is small. As for cloning of a growth and development factor the kinetics of G-CSF, the increase in serum G-CSF levels that is a ligand for the receptor Mpl. Cell 1994; 77: in PBSCT patients starts immediately following graft 1117–1124. infusion partly because endogenous G-CSF is derived from 15 Kato T, Ogami K, Shimada Y et al. Purification and charac- an increased number of infused monocytes present in PBSC terization of thrombopoietin. J Biochem 1995; 118: 229–236. grafts.33 In contrast, TPO has been reported to be produced 16 Tahara T, Usuki K, Sato H et al. A sensitive sandwich ELISA by and but not monocytes.32,34 These reports for measuring thrombopoietin in human serum: serum throm- are consistent with our observations that no kinetic differ- bopoietin levels in healthy volunteers and in patients with ences in serum TPO levels were demonstrated between hematopoietic disorders. Br J Haematol 1996; 93: 783–788. + 17 Shimazaki C, Oku N, Ashihara E et al. Collection of peri- PBSCT and purified CD34 PBSCT. pheral blood stem cells mobilized by high-dose Ara-C plus The present study demonstrates that TPO is an important VP-16 or aclarubicin followed by recombinant human gra- cytokine for the reconstitution of megakaryocytic lineage nulocyte colony-stimulating factor. Bone Marrow Transplant cells after stem cell transplantation. Considering the poten- 1992; 10: 341–346. tial clinical application of this cytokine in patients undergo- 18 Teramura M, Katahira J, Hoshino S et al. Clonal growth of ing stem cell transplantation,35 further study is warranted human megakaryocyte progenitor in serum free cultures; to clarify the mechanisms regulating TPO in humans. effect of recombinant human . Exp Hematol 1988; 16: 843–848. 19 Lotem J, Shabo Y, Sachs L. Regulation of megakaryocyte References development by interleukin-6. Blood 1989; 74: 1545–1551. 20 Bruno E, Briddell RA, Cooper RJ, Hoffman R. Effects of 1 Kessinger A, Armitage JO. The evolving role of autologous recombinant interleukin 11 on human megakaryocyte progeni- peripheral stem cell transplantation following high-dose ther- tor cells. Exp Hematol 1991; 19: 378–381. apy for malignancies. Blood 1991; 77: 211–213. 21 Metcalf D, Hilton D, Nicola NA. Leukemia inhibitory factor 2 Gale RP, Henon P, Juttner C. Blood stem cell transplants can potentiate murine megakaryocyte production in vitro. come of age. Bone Marrow Transplant 1992; 9: 151–155. Blood 1991; 77: 2150–2153. 3 Demuynck H, Delforge M, Zachee P et al. An update on peri- 22 Kuter DJ, Rosenberg RD. The reciprocal relationship of pheral blood progenitor cell transplantation. Ann Hematol thrombopoietin (c-Mpl ligand) to changes in the platelet mass 1995; 71: 29–33. during busulfan-induced thrombocytopenia in the rabbit. 4 Gordon MY, Greaves MF. Physiological mechanisms of stem Blood 1995; 85: 2720–2730. cell regulation in bone marrow transplantation and hemato- 23 Bernstein SH, Baer MR, Lawrence D et al. Serial determi- poiesis. Bone Marrow Transplant 1989; 4: 335–338. nation of thrombopoietin (TPO), IL-3, IL-6, IL-11 and leuke- 5 Cairo MS, Suen Y, Sender L. Circulating granulocyte colony- mia inhibitory factor (LIF) levels in patients undergoing stimulating factor (G-CSF) levels after allogeneic and autolog- chemotherapy for acute myeloid leukemia (AML). Blood ous bone marrow transplantation: endogenous G-CSF pro- 1995; 86 (Suppl. 1): 46 (Abstr. 170). duction correlates with myeloid engraftment. Blood 1992; 79: 24 Nichol JL, Hokom MM, Hornkohl A et al. Megakaryocyte 1869–1873. growth and development factor analyses of in vitro effects on 6 Kawano Y, Takaue Y, Saito S et al. Granulocyte colony- human megakaryopoiesis and endogenous serum levels during stimulating factor (CSF), macrophage-CSF, granulocyte– chemotherapy-induced thrombocytopenia. J Clin Invest 1995; macrophage CSF, interleukin-3, and interleukin-6 levels in 95: 2973–2978. sera from children undergoing blood stem cell autografts. 25 Fielder PJ, Gurney AL, Stefanich E et al. Regulation of throm- Blood 1993; 81: 856–860. bopoietin levels by c-mpl-mediated binding to platelets. Blood 7 Miksitis K, Beyer J, Siegert W. Serum concentration of G- 1996; 87: 2154–2161. CSF during high-dose chemotherapy with autologous stem 26 Stoffel R, Wiestner A, Skoda RC. Thrombopoietin in throm- cell rescue. Bone Marrow Transplant 1993; 11: 375–377. bocytopenic mice: evidence against regulation at the mRNA 8 Haas R, Gericke G, Witt B, Cayeux S, Hunstein W. Increased level and for a direct regulatory role of platelets. Blood 1996; serum levels of granulocyte colony-stimulating factor after 87: 567–573. autologous bone marrow or blood stem cell transplantation. 27 Shimazaki C, Uchiyama T, Fujita N et al. Serum levels of Exp Hematol 1993; 21: 109–113. endogenous and exogenous granulocyte colony-stimulating Serum TPO levels after PBSCT and BMT C Shimazaki et al 775 factor after autologous blood stem cell transplantation. Exp 32 McCarty JM, Sprugel KH, Fox NE et al. Murine thrombo- Hematol 1995; 23: 1497–1502. poietin mRNA levels are modulated by platelet count. Blood 28 Layton JE, Hockman H, Sheridan WP, Morstyn G. Evidence 1995; 86: 3668–3675. for a novel in vivo control of granulopoiesis: mature cell- 33 Takamatsu Y, Akashi K, Harada M et al. Cytokine production related control of a regulatory . Blood 1989; 74: by peripheral blood monocytes and T cells during haemo- 1303–1307. poietic recovery after intensive chemotherapy. Br J Haematol 29 Bartocci A, Mastrogiannis DS, Migliorati G et al. Macrophage 1993; 83: 21–27. specifically regulate the concentration of their own growth fac- 34 Shimada Y, Kato T, Ogami K et al. Production of thrombo- tor in the circulation. Proc Natl Acad Sci USA 1987; 84: poietin (TPO) by rat hepatocytes and hepatoma cell line. Exp 6179–6183. Hematol 1995; 23: 1388–1396. 30 Cebon JS, Bury RW, Lieschke GJ, Morstyn G. The effect of 35 Molineux G, Hartley C, McElroy P et al. Megakaryocyte dose and route of administration on the pharmacokinetics of growth and development factor accelerates platelet recovery granulocyte–macrophage colony-stimulating factor. Eur J in peripheral blood progenitor cell transplant recipients. Blood Cancer 1990; 26: 1064–1069. 1996; 88: 366–376. 31 Goldberg MA, Gaut CC, Bunn HF. Erythropoietin mRNA levels are governed by both the rate of gene transcription and post-transcriptional events. Blood 1991; 77: 271–276.