Platelet Transfusion Requirements During Autologous Peripheral Blood Progenitor Cell Transplantation Correlate with the Pretransplant Platelet Count

Home , Autotransplantation

Platelet transfusion requirements during autologous peripheral blood progenitor cell transplantation correlate with the pretransplant platelet count

BJ Bolwell, M Goormastic, S Andresen, B Overmoyer, B Pohlman and M Kalaycio

Department of Hematology and Medical Oncology, at the Cleveland Clinic Foundation, Cleveland, OH, USA

Summary: after the infusion of autologous marrow.1,2 The use of primed peripheral blood progenitor cells (PBPC) has The use of primed peripheral blood progenitor cells resulted in more rapid platelet engraftment, with most trials (PBPC) has improved platelet engraftment following reporting platelet recovery in 14–18 days.3–8 Relatively few autologous bone marrow/PBPC transplantation series specifically report the number of platelet transfusions (ABMT). The thrombocytopenia associated with ABMT required during autologous transplantation when using generally lasts 14–18 days, and is associated with vari- PBPC. At our own institution, a median of five platelet able platelet transfusion requirements. Little, if any, transfusions are required during PBPC autologous trans- data exist examining prognostic parameters for platelet plantation. transfusion requirements during autologous transplan- Many studies have documented that both neutrophil and tation. We retrospectively examined 286 consecutive platelet engraftment correlates with the number of CD34+ patients undergoing autologous transplantation from 1 cells infused.9–12 However, little data exist examining other January l994 to 1 June l996 with respect to platelet potential prognostic parameters for platelet engraftment, or engraftment and platelet transfusion requirements. One for the number of platelet transfusion events required dur- hundred and fifty four patients were transplanted for ing autologous transplantation. Such parameters are neces- breast cancer (54%), 72 for non-Hodgkin’s lymphoma sary as new platelet stimulating agents, such as thrombopo- (25%), 35 for Hodgkin’s disease (12%), 13 for acute leu- ietin, are being incorporated into clinical trials. Anecdotal kemia (5%), eight for myeloma (3%), and four for other data from our institution suggested that patients beginning malignancy (1%). The median age was 44. All patients PBPC autologous transplant with a low platelet count received cytokine priming, usually with G-CSF, for the required more platelet transfusions during the transplant procurement of PBPC. The median number of CD34+ than did those patients beginning with a normal platelet cells collected was 4.3 × 106/kg. All patients received a count. To study this issue, we retrospectively examined 286 chemotherapeutic preparative regimen and all received consecutive patients undergoing autologous progenitor cell an autologous transplant using PBPC alone. The transplantation at our institution from 1 January l994 to median time to a platelet count of 20 × 109/l was 13 days. 1 June l996 to analyze whether the platelet count at the Patients beginning the transplant with a less than nor- time of transplantation correlated with either platelet mal platelet count (less than 150 × 109/l) engrafted in 17 engraftment, or the number of transfusion events required days, and received a median number of seven platelet during the transplant. transfusions, as compared with platelet engraftment of 12 days, and four platelet transfusions, for patients beginning the transplant with a normal platelet count Materials and methods (P = 0.001). Both groups of patients received an equiva- + lent dose of CD34 cells. We conclude that thrombocyto- Patient characteristics penia at the initiation of autologous transplantation is associated with increased platelet transfusion require- From 1 January l994 to 1 June l996, 286 consecutive adult ments, independent of the dose of CD34+ cells infused. patients underwent high-dose chemotherapy with autolog- Keywords: autologous marrow transplantation; stem ous progenitor cell transplantation at the Cleveland Clinic cells; platelet transfusions Foundation. Patient characteristics are shown in Table 1.

PBPC collection and processing Platelet engraftment following autologous bone marrow All patients received cytokine priming for PBPC collection, transplantation (ABMT) historically occurred 4–5 weeks with G-CSF 5–10 ␮g/kg per day subcutaneously for 8–10 days, with PBPC collection beginning on day +5. Total × 6 + Correspondence: Dr BJ Bolwell, Cleveland Clinic Foundation, 9500 yield of PBPC collected was at least 2 10 CD34 cell/kg. Euclid Avenue, Cleveland, OH 44195, USA Prior to 1 March 1995, an acceptable alternative to the total + Received 30 December 1996; accepted 22 May 1997 CD34 count was 10 × 108 mononuclear cell/kg. The Platelet transfusion during PBPC autologous transplant BJ Bolwell et al 460 Table 1 Patient characteristics

Group A Group B Total (low platelet count) (normal platelet count)

No. of patients 74 212 Age (median) 44 44 44 Karnofsky status (pretransplant) % 90 90 90 Months from diagnosis to transplant (mean) 25 27 26 Preparative regimen (%) Bu-Cy-Vp 41 (55) 77 (36) 118 (41) Bu-Cy 9 (12) 36 (17) 45 (16) Ctx-Carbo-TT 18 (25) 76 (36) 94 (38) Ltx-Plat-BCNU 4 (5) 21 (10) 25 (9) Other 2 (3) 2 (1) 4 (1) Diagnosis (%) Breast cancer 29 (39) 125 (59) 154 (54) Non-Hodgkin’s lymphoma 23 (31) 49 (23) 72 (25) Hodgkin’s disease 10 (13) 25 (12) 35 (12) Acute leukemia 8 (11) 5 (2) 13 (5) Multiple myeloma 2 (3) 6 (3) 8 (3) Other 2 (3) 2 (1) 4 (1) Disease status of transplant (%) CR 46 (62) 131 (62) 177 (62) PR 26 (35) 76 (30) 102 (36) Refractory 2 (3) 5 (2) 7 (2)

Bu-Cy-VP = busulfan, cyclophosphamide, etoposide; Bu-Cy = busulfan, cyclophosphamide; Ctx-Carbo-TT = cyclophosphamide, carboplatin, thiotepa; Ctx-Plat-BCNU = cyclophosphamide, cisplatin, BCNU.

patients were harvested on a COBE Spectra leukapheresis generally for testicular cancer. G-CSF was routinely given machine (COBE, Denver, CO, USA). at a dose of 5 ␮g/kg per day beginning 4 h after the infusion of PBPC. G-CSF was discontinued after patients reached 9 3 + 1.0 × 10 /l neutrophils/mm on 2 consecutive days. CD34 analysis Peripheral blood progenitor cells obtained by apheresis Supportive care were collected into ACD-A anticoagulant. An aliquot of each collection was counted and a minimum of 20 × 106 Patients received prophylactic antibiotics consisting of oral cells were submitted for flow cytometric evaluation. For acyclovir, ciprofloxacin, and low-dose Amphotericin B (0.2 each sample, 250 000 events were required as list mode mg/kg), beginning on the day after PBPC infusion. Patients data using a Cell Quest (BDIS, San Jose, CA, USA) experiencing persistent fevers had their antibiotic coverage software document which had been specifically tailored and broadened. Patients received red blood cell transfusions to stored for this purpose. Total CD34+ cells were calculated maintain a hemoglobin greater than 9 gm/dl. Platelet trans- by multiplying the percentage of CD34+-positive cells in fusions were given for platelet counts less than 10 × 109/l the flow cytometric analysis by the total number of or for clinical bleeding. A platelet transfusion event was nucleated cells in the apheresis product. This number was either with single donor infusions, or pooled platelets divided by the patient’s actual weight to give the total (pooled platelets generally consisted of six pooled units CD34+ cells per kilogram. of platelets).

Transplant protocol Definitions Ninety-nine percent of patients were treated with one of Platelet engraftment was defined as a platelet count of four chemotherapy (only) preparative regimens: cyclophos- 20 × 109/l independent of platelet transfusions. A platelet phamide, thiotepa, carboplatin (cyclophosphamide 1500 mg transfusion ‘event’ was any infusion of platelets, either per m2 daily × 4 days; thiotepa 125 mg per m2 daily × 4 pooled or a single donor. days; carboplatin 200 mg per m2 daily × 4 days); busulfan, For the purposes of this study, the platelet count was cyclophosphamide, VP-16, (busulfan 14 mg/kg, etoposide deemed normal (150 × 109/l or greater), or abnormal, on 50 mg/kg, cyclophosphamide 120 mg/kg); busulfan and the day of initiation of high-dose chemotherapy. Thus, all cyclophosphamide (busulfan 16 mg/kg, cyclophosphamide patients had already received cytokine priming and the col- 120 mg/kg); cyclophosphamide, cisplatin, BCNU lection of peripheral progenitor cells, at the time of evalu- (cyclophosphamide 1875 mg/m2 per day × 3 days; cisplatin ation of a normal or abnormal platelet count. The apheresis 55 mg/m2 per day × 3 days; BCNU 600 mg per m2). Four to collect CD34+ cells occurred 5–10 days prior to the patients (1%) received other chemotherapeutic regimens, initiation of the preparative regimen. Platelet transfusion during PBPC autologous transplant BJ Bolwell et al 461 Table 2 Prior chemotherapy and PBPC yield did group B. A multivariate analysis was then performed to examine factors related to days to engraftment. Age, sex, Group A Group B Total prior exposure to radiation therapy, number of courses of (low platelet (normal platelet prior chemotherapy, preparative regimen, diagnosis, CD34+ count) count) cell dose, and group A vs group B were included in the multiple regression model. This multivariate analysis found No. of patients 74 212 286 that the most important variable correlating with the time No. of courses of prior chemotherapy (%) + = 1 33 (44) 96 (46) 129 (45) to platelet engraftment was CD34 cell dose (P 0.004). 2 29 (39) 84 (40) 113 (40) The number of courses of prior chemotherapy was also stat- 3 or greater 12 (17) 31 (14) 43 (15) istically significant, with those having increasing numbers CD34+ cells/kg 4.54 4.18 of prior chemotherapy requiring more days to engraftment (mean) (range (range (P = 0.04). The difference between the two treatment 0.54–22.1) 0.1–18.2) groups was also statistically significant in this multivariate analysis, with patients presenting with thrombocytopenia (group A) being a significant factor in terms of platelet engraftment (P = 0.02). Statistical analysis Group A patients also required more platelet transfusions A Wilcoxon rank sum test was used when engraftment time than did group B, as shown in Table 3 (P = 0.001). We then (days) for CD34+ cell dose was compared for two groups. performed a multivariate analysis of variables affecting the A Kruskal–Wallis test was used if more than two groups number of transfusions required during the transplant, were being compared. A P value of 0.05 or less was con- including CD34+ dose, underlying diagnosis, sex, age, prior sidered to be statistically significant. A forward stepwise exposure to radiation therapy, number of courses of prior regression was used to see what factors were independently chemotherapy, and preparative regimen in the multiple associated with the number of transfusions and days to regression model. Group A vs group B patients was a sig- platelet engraftment. A significance level of 0.05 was used nificant factor in this model (P = 0.001), as was age at as the entry criterion. transplant, with older age resulting in a higher platelet transfusion requirement (P = 0.04). A box plot of the number of platelet transfusions by platelet count is shown in Results Figure 1. We next examined the results of engraftment and trans- Seventy-four patients began the transplant with an abnor- fusion requirements comparing those patients with a plate- mally low platelet count (defined as less than 150 × 109/l), let count of less than 100, with patients with a platelet count and 212 patients began the transplant with a normal platelet of 100–149. Only 28 patients began the transplant with a count. The number of courses of prior chemotherapy, as platelet count of less than 100; the median number of trans- well as the number of CD34+ cells harvested, is shown in fusions was 7.5, and the median time of platelet Table 2. engraftment was l9 days. These were not statistically sig- Table 3 shows engraftment and transfusion results for nificant differences when compared with patients entering both treatment groups. Patients presenting with thrombocy- the transplant with a platelet count of 100–149, requiring topenia at the time of transplant experienced a longer time a median six transfusion events, and experiencing platelet to platelet engraftment, as compared with patients engraftment in 16 days. We further analyzed patients with presenting with a normal platelet count. This was true a normal platelet count, comparing those with a platelet despite the fact that group A patients actually received a count of 150–250 thousand (n = 152) with those with a higher number of CD34+ cells per kilogram infused than platelet count greater than 250 × 109/l (n = 60). The time to engraftment was identical between these two groups (median 12 days and mean 17 days). The group with a Table 3 Engraftment and transfusion results platelet count of at least 250 × 109/l required a mean of 4.4 platelet transfusion events (median = 3) compared to 5.9 Group A Group B for those with a count 150–250 × 109/l (median = 4), but (low platelet (normal platelet = count) count) this difference did not reach statistical significance (P 0.15). Initial platelet count (mean) 74 212 Both groups tolerated autologous transplantation equally Engraftment (days) well. There was only one death within 30 days, that being PMN 0.5 × 109/l (mean) 12 10.5 of a patient with a normal platelet count at the initiation Platelet 20 × 109/l at transplant. There was no difference in the incidence of (mean) 24 17 infections, fevers, or overt clinical bleeding. P = 0.001 (median) 17 12 Platelet transfusion events (mean) 9 5 Discussion P = 0.001 (median) 7 4 The use of PBPC, with or without the use of autologous bone marrow, has dramatically shortened platelet Platelet transfusion during PBPC autologous transplant BJ Bolwell et al 462 Table 4 Summary of studies reporting platelet transfusions during ABM/PBPC autotransplantation

Author/Ref. Source of cells Transfusion trigger Platelet source Platelet engraftment No. of platelet (platelet count (days) transfusions × 109/l)

Chao4 PBPC Ϯ ABM 20 Single donor 13 5 Peters5 PBPC + ABM 25 Single donor Not reported 9.5–30 Huan17 PBPC + ABM 20 Single or random 16 4 Demirer6 PBPC + ABM Not reported Not specified 14 25 Brice8 PBPC alone Not reported Not specified 17 4 days of transfusion Haynes7 PBPC alone 10 Not specified 13 12 Pettengell19 PBPC alone 20 Not specified 10 24–39 Faucher8 PBPC alone 20 Not specified 14 2

PBPC = peripheral blood progenitor cells; ABM = autologous bone marrow.

70 28 46 86 66 41 19 bleeding. Additionally, different transplant centers 65 employed different criteria to initiate a platelet transfusion, 60 × 9 55 with the platelet trigger values ranging 10–25 10 /l. 50 A summary of studies reporting platelet transfusion 45 requirements during autologous transplantation with PBPC 40 support is shown in Table 4. While days to platelet 35 engraftment are fairly consistent over these series, there is 30 25 a wide range in the number of platelet transfusions 20 reported. This may reflect the fact that a platelet transfusion No. of transfusions 15 event may consist of pooled platelets, which is made up of 10 7.5 6 four to eight individual units. Different institutional guide- 5 44 33 0 lines concerning platelet transfusions also clearly contribute <100 100–149 150–199 200–249 250–299 300+ to these discrepancies. We found a difference in the number of platelet trans- Platelets fusion events, as well as time to platelet engraftment, based Figure 1 Number of transfusions by platelet count. on the platelet count at the initiation of the preparative regimen. From an engraftment perspective, the number of CD34+ cells was the most important variable in the multi- engraftment after autologous transplantation. Historically, variate analysis. However, the pretransplant platelet count a period of 4–5 weeks was required to achieve a platelet was also a statistically important variable within this multi- count of 20 × 109/l after ABMT. The use of the hematopo- variate model. Additionally, patients presenting with throm- ietic growth factors after autologous marrow transplantation bocytopenia required more transfusions than did those had been shown to enhance neutrophil engraftment, but not presenting with a normal platelet count. The latter point is to affect platelet engraftment.13–16 Numerous reports have possibly intuitive, as those presenting with thrombocyto- demonstrated that primed PBPC have dramatically short- penia frequently experience a more rapid initial platelet ened time of platelet engraftment, with most reports nadir, and thus the initiation of platelet transfusions began reporting platelet independence in 12–18 days following sooner in their transplant course. However, the explanation infusion of autologous progenitor cells, with or without for prolonged platelet engraftment is less obvious and any autologous marrow.3–8 explanation for this phenomenon is conjectural. It is poss- Many studies have shown that engraftment of neutrophils ible that patients presenting with thrombocytopenia might and platelets correlates with the number of progenitor cells have a more damaged bone marrow microenvironment infused. The number of circulating progenitor cells is rou- from prior chemotherapy and/or radiation therapy. Bone tinely measured by quantifying CD34+ cells with multi- marrow microenvironment differences might potentially color flow cytometry. Transplant centers frequently employ explain a difference in the speed of platelet recovery. It a threshold number of CD34+ cells/kg to determine might be possible that different patient populations have adequate collection of progenitor cells. While many studies different levels of circulating megakaryocytic stimulating have reported time to platelet engraftment when using cytokines, both pre- and post-transplantation. Immunologic PBPC, relatively few series have reported specific data of abnormalities resulting in thrombocytopenia pre-BMT number of platelet transfusions required during autologous might persist post-BMT, resulting in prolonged time to transplantation. The reported data are complicated by the engraftment, as well as decreased platelet life span of trans- fact that there is a lack of a standardization of what defines fused platelets during the period of thrombocytopenia. The a platelet transfusion, as some centers exclusively utilize two patient groups behaved identically clinically, in our single donor platelets, while others prefer to transfuse study, so we do not think that our results are due to the pooled platelets. The utility of platelet transfusions is thrombocytopenic group having more infections, bleeding, clearly affected by clinical variables, such as fevers and or other readily clinically apparent variables. Platelet transfusion during PBPC autologous transplant BJ Bolwell et al 463 While the precise pathophysiology may be unclear, we transplantation: parameters affecting bone marrow do believe that we have demonstrated that the platelet count engraftment. Bone Marrow Transplant 1993; 12: 609–614. at the initiation of autologous bone marrow/progenitor cell 10 Siena S, Bregni M, Brando B et al. Flow cytometry for clinical transplantation is a variable of potential importance that estimation of circulating hematopoietic progenitors for auto- should be analyzed when reporting engraftment and trans- logous transplantation in cancer patients. Blood 1991; 77: 400–409. fusion requirements for this procedure. New cytokines, 11 Passos-Coelho JL, Hayden GB, Davis JM et al. Predictive fac- such as thrombopoietin, have recently been described to tors for peripheral-blood progenitor-cell collections using a 20–22 stimulate megakaryocyte growth and development. single large-volume leukapheresis after cyclophosphamide and Studies are beginning to examine the utility of such new granulocyte–macrophage colony-stimulating factor mobiliz- cytokines in autologous transplantation. We feel that such ation. J Clin Oncol 1995; 13: 705–714. clinical trials examining platelet transfusion requirements 12 Haas R, Wit B, Mohle H et al. Sustained long-term hematopo- and platelet engraftment during autologous transplantation iesis after myeloablative therapy with peripheral blood pro- should consider utilizing the pretransplant platelet count as genitor cell support. Blood 1995; 85: 3754–3761. a clinical variable of potential importance. 13 Brandt SJ, Peters WP, Atwater SK et al. Effect of recombinant human granulocyte–macrophage colony-stimulating factor on hematopoietic reconstitution after high-dose chemotherapy and autologous bone marrow transplantation. New Engl J Med References 1988; 318: 869–876. 14 Gorin NC, Coiffier B, Hayat M et al. Recombinant human 1 Appelbaum FR. Marrow transplantation for hematologic granulocyte–macrophage colony-stimulating factor after high- malignancies: a brief review of current status and future pros- dose chemotherapy and autologous bone marrow transplan- pects. Semin Hematol 1988; 25 (Suppl. 3): 16–22. tation with unpurged and purged marrow in non-Hodgkin’s 2 Bolwell BJ. Autologous bone marrow transplantation for Lymphoma: a double-blind placebo-controlled trial. Blood Hodgkin’s disease and non-Hodgkin’s lymphoma. Semin 1992; 80: 1149–1157. Hematol 1994; 21 (Suppl. 7): 86–95. 15 Link H, Boogaerts MA, Carella AM et al. A controlled trial 3 Bolwell BJ, Goormastic M, Yanssens T et al. Comparison of of recombinant human granulocyte–macrophage colony- G-CSF with GM-CSF for mobilizing peripheral blood pro- stimulating factor after total body irradiation, high-dose genitor cells and for enhancing marrow recovery after autolog- chemotherapy, and autologous bone marrow transplantation ous bone marrow transplant. Bone Marrow Transplant 1994; for acute lymphoblastic leukemia or malignant lymphoma. 14: 913–918. Blood 1992; 80: 2188–2195. 4 Chao NJ, Schriber JR, Grimes K et al. Granulocyte colony- 16 Nemunaitis J, Rabinowe SN, Singer JW et al. Recombinant stimulation factor ‘mobilized’ peripheral blood progenitor granulocyte–macrophage colony-stimulating factor after auto- cells accelerate granulocyte and platelet recovery after high- logous bone marrow transplantation for lymphoid cancer. New dose chemotherapy. Blood 1993; 81: 2031–2035. Engl J Med 1991; 324: 1773–1778. 5 Peters WP, Rosner G, Ross M et al. Comparative effects of 17 Huan SD, Hester J, Spitzer G et al. Influence of mobilized granulocyte–macrophage colony-stimulating factor (GM-CSF) peripheral blood cells on the hematopoietic recovery by auto- and granulocyte colony-stimulating factor (G-CSF) on prim- logous marrow and recombinant human granulocyte– ing peripheral blood progenitor cells for use with autologous macrophage colony-stimulating factor after high-dose cyclo- bone marrow after high-dose chemotherapy. Blood 1993; 81: phosphamide, etoposide, and cisplatin. Blood 1992; 79: 1709–1719. 3388–3393. 6 Demirer T, Buckner CD, Appelbaum FR et al. Rapid 18 Brice P, Divine M, Marolleau JP et al. Comparison of auto- engraftment after autologous transplantation utilizing marrow grafting using mobilized peripheral blood stem cells with and and recombinant granulocyte colony-stimulating factor- without granulocyte colony-stimulating factor in malignant mobilized peripheral blood stem cells in patients with acute lymphomas. Bone Marrow Transplant 1994; 14: 51–55. myelogenous leukemia. Bone Marrow Transplant 1995; 15: 19 Pettengell R, Morgenstern GR, Woll PJ et al. Peripheral blood 915–922. progenitor cell transplantation in lymphoma leukemia using a 7 Haynes A, Hunter A, McQuaker G et al. Engraftment single apheresis. Blood 1993; 82: 3770–3777. characteristics of peripheral blood stem cells mobilized with 20 Kaushansky K. Thombopoietin: the primary regulator of plate- cyclophosphamide and the delayed addition of G-CSF. Bone let production. Blood 1995; 86: 419–431. Marrow Transplant 1995; 16: 359–363. 21 Farese AM, Hunt P, Boone T, MacVittie TJ. Recombinant 8 Faucher C, le Corroller AG, Blaise D et al. Comparison of human megakaryocyte growth and development factor stimu- G-CSF-primed peripheral blood progenitor cells and bone lates thrombocytopoiesis in normal nonhuman primates. Blood marrow auto transplantation: clinical assessment and cost- 1995; 86: 54–59. effectiveness. Bone Marrow Transplant 1994; 14: 895–901. 22 Banu, N, Wang J, Deng B et al. Modulation of megakaryocy- 9 Bolwell BJ, Fishleder A, Andresen SW et al. G-CSF primed topoiesis by thrombopoietin: the c-Mpl Ligand. Blood 1995; peripheral blood progenitor cells in autologous bone marrow 86: 1331–1338.