The Use of PEG-rHuMGDF in Platelet Apheresis

Transcription

1 The Use of PEG-rHuMGDF in Platelet Apheresis DAVID J. KUTER Massachusetts General Hospital Cancer Center, Harvard Medical School, Boston, Massachusetts, USA Key Words. Thrombopoietin PEG-rHuMGDF Thrombocytopenia Blood platelets Platelet apheresis Platelet transfusion ABSTRACT Platelet transfusions are increasingly being used to treat thrombocytopenic conditions ranging from aplastic anemia to that caused by cancer chemotherapy. Although historically whole-blood transfusions were the primary source of platelets for transfusion, random donor platelet concentrates and single-donor apheresis platelets are currently the only products used. The use of these products in the United States varies widely for different medical conditions; for example, surgical patients receive random donor platelet concentrates much more commonly than single-donor apheresis products, while the opposite is true for hematology/oncology patients. The past decade has seen a great change in the type of platelet product prescribed. Whereas random donor platelet concentrates were mostly used in the past, over 60% of the platelets transfused are now obtained from donors by apheresis. A crucial variable in the ability to collect platelets by apheresis is the donor platelet count. With the recent availability of thrombopoietin, there has been considerable interest in using this hematopoietic growth factor to stimulate platelet production in donors. Preliminary studies with the administration to platelet donors of one of the thrombopoietic growth factors, PEG-rHuMGDF, have demonstrated a marked increase in the apheresis yield and no side effects. The PEG-rHuMGDF-mobilized platelets were effective upon transfusion. Whether stimulation of platelet production in donors with thrombopoietic growth factors will become a widely accepted method will depend largely on the safety of this approach for the donor as well as on a number of lesser issues which concern the recipient and blood center. Stem Cells 1998;16(suppl2): INTRODUCTION For over thirty years, platelet transfusions have been used in a wide variety of thrombocytopenic conditions ranging from aplastic anemia to bone marrow transplantation. Platelets for transfusion may be obtained from three sources: fresh whole blood, random donor platelet concentrates, and single-donor apheresis platelets. Furthermore, these platelet products may also be HLA-typed. Although fresh whole blood has rarely been used in recent years to provide platelets for transfusion, fresh whole blood was historically the first product used for platelet transfusion. Indeed, the first transfusion of whole blood into a human, that from an animal into a man, was done in 1667 by Jean-Buptiste Denis, a physician on the staff of King Louis XIV. He transfused the blood of a healthy sheep into a weak and febrile boy who subsequently improved [I]. This was the first demonstration of the beneficial effects Thrombopoietin: From Molecule to Medicine STEM CELLS 1998;16(suppl2): OAlphaMed Press. All rights reserved.

2 232 Use of PEG-rHuMGDF in Platelet Apheresis of red blood cells, but since the transfused blood also contained platelets, it was also unknowingly the first platelet transfusion. At the same time in England, Richard Lower described an account of an experimental transfusion practiced upon a man in London in which whole blood of a lamb, presumably containing platelets, was transfused into a man that is a little frantic with some improvement in his condition [2]. Although no comments were made as to whether any hemostatic benefit ensued, several other patients were transfused during this era in both England and France with the blood from dogs, lambs, calves, or cats. Unfortunately, complications from the transfusion of whole-animal blood into humans occurred, and transfusions were subsequently banned by both the English and French authorities. Transfusion therapy essentially died out or was not reported in the medical literature for 150 years until the early nineteenth century when James Blundell, an obstetrician and lecturer in physiology at Guys Hospital, reported the first transfusions of human blood into human recipients. He had recognized the difficulties involved in transfusing animal blood into humans and was motivated by the hemorrhagic death of many of his young patients. After first experimenting with the transfusion of blood by syringe into animals, in 1818 he performed the first transfusion of human whole blood into a patient who presumably had gastric cancer and subsequently had some temporary improvement of his weakness [3]. These experiments ultimately led Blundell over the next ten years to devise several devices for transfusion, one a syringe device called the impellor [4] and another which allowed direct transfusion from the donor to the recipient called the gravitator [5]. The above experiments were performed by transfusing whole and often partially clotted blood into humans. Platelets had not yet been discovered. Although there are various reports throughout the nineteenth century of improved hemostatic function after the transfusion of whole blood, it wasn t until the pioneering work of Osler and Wright that the blood platelet was described and its key role in hemostasis identified. From 1902 until 1910, James Homer Wright working at the Massachusetts General Hospital described the origin and function of platelets. In his 1902 report [6], he unequivocally identified the platelet using his Wright stain, and in 1906 [7] and again in 1910 [8], described the fact that platelets came from megakaryocytes and were involved in blood clotting, and that their absence was associated with bleeding. In 1909, also at the Massachusetts General Hospital, William Duke was the first to identify the relationship of blood platelets to hemorrhagic disease when he reported three cases of hemorrhagic disease that responded to the transfusion of whole blood [9]. As was common at the time, these transfusions were administered by a direct arteriovenous connection between the donor and the recipient. In one of these cases, a young Armenian tailor with hemorrhage, a very low platelet count, and an elevated bleeding time was transfused with blood from an Armenian friend of the same age. His platelet count rose from 6,000 to 123,000, and the bleeding time dropped from 90 min to 3 min with no change in the other coagulation parameters. The patient s bleeding stopped, and clinical improvement was noted. This was the first unequivocal demonstration that platelets contained in whole blood might provide significant benefit for thrombocytopenic individuals. Two other cases with less striking results were also reported by Duke at this time. SOURCES OF PLATELETS FOR TRANSFUSION Fortunately, since the pioneering work of Duke and Wright, transfusions have become somewhat easier to administer and no longer require direct arteriovenous connections from donor to recipient. With the advent of anticoagulation, blood fractionation, and adequate storage conditions [ 101, several platelet products are widely available for transfusion into thrombocytopenic patients. At the present time, whole blood is rarely used for platelet transfusion. Platelets for transfusion are of two basic types: random donor platelet concentrate units and single-donor apheresis platelet units. Random donor platelet concentrate units are obtained from a single unit of donated whole blood. Following fractionation by one of several different methods, these products usually contain greater than 5.5 x 10 platelets in a volume of ml of plasma. Usually, five to six such units are pooled and administered to a thrombocytopenic adult recipient.

3 Kuter 233 Single-donor apheresis platelet units are obtained by connecting the donor to a cell separator machine. The machine processes whole blood using citrate anticoagulation for approximately 60 to 100 min, red blood cells and plasma being returned to the donor and the single-donor platelet product collected into a volume of approximately ml of plasma. Such products usually contain greater than 3 x 10 platelets and are equivalent to at least five to six units of random donor platelet concentrate units. Both of these platelet products may be HLA-typed, but for economic reasons, the single-donor apheresis platelet units are usually the only products so tested. As described in greater detail below, single-donor apheresis platelet units are being increasingly used in the United States. At the present time, approximately 60% of platelet transfusions administered in the United States are obtained from donors who have undergone this procedure. Although the superiority of single-donor apheresis platelets over pooled random donor platelet concentrates has not yet been adequately demonstrated, potential advantages to the recipient include exposure to fewer donors, fewer infectious complications, standardization of product, easier HLA typing, and irradiation of the product. The platelet apheresis procedure itself has a number of variables which are important. These include the efficiency of the cell separator, the duration of the apheresis procedure, the donor size, gender, and hematocrit, and finally, the preapheresis platelet count. Of these, the preapheresis platelet count is probably the most critical parameter in predicting the yield of a procedure. Donors having platelet counts lower than 150 x lo9 are usually excluded from donating. In some institutions, donors with high but normal platelet counts are prized for their ability to provide especially large apheresis products. PLATELET USAGE IN THE UNITED STATES IN 1995 An analysis of blood product use in the United States was recently undertaken (D.J. Kuter, 1997, unpublished) using the data directly reported from approximately 1.7 million inpatient hospital admissions from a sample of hospitals for 1993, 1994, and Platelet usage in this sample was collected using the International Classification of Clinical Services (ICCS) codes for platelet transfusions and analyzed by the Diagnostic Related Group (DRG) of the recipient, the type and amount of platelet product transfused. The data from this representative sample of hospitals were then extrapolated for nationwide usage. This approach with this sample of hospitals has been previously validated for other clinical situations and is comparable to that provided by the National Hospital Discharge Survey (NHDS). Figure 1 shows the likelihood that patients with various categories of diseases will receive any platelet transfusion during a hospital admission. As expected, most patients undergoing bone marrow 100% 80% 60% 40% 20% 0 % Figure 1. The likelihood of receiving an inpatient platelet transfusion in As described in the text, using a nationwide estimate ofplatelet usage for all inpatient admissions in the United States for 1995, the percentage of inpatients in each DRG classification receiving any platelet transfusion was calculated. The individual DRG classifications for the inpatient admissions were combined into ten general disease groupings.

4 ~~ 234 Use of PEG-rHuMGDF in Platelet Apheresis Oncology 9% Figure 2. The 1995 inpatient use of all platelet products by disease grouping. As described in Figure I, the national estimate of total platelet usage (random donor platelet concentrates, single donor apheresis platelet units, and HLA-matched single donor apheresis platelet units) for all inpatient admissions in the United States for 1995 was calculated for each recipient s DRG classification. The individual DRG classifications for the inpatient admissions were combined into ten general disease groupings and the relative use of all platelet products (measured in single-donor apheresis unit equivalerits ) in these groups calculated. A single-donor apheresis unit equivcilent is cipproximately 3 X 10 platelets; six random-donor platelet concentrates equal one single-donor upheresis unit equivalent. :ardiovascular 30% I \ Surgi Oncology R% Trauma 3% Leukemia 3 Yo BMT 1 % lematolc SY 8% ID 10% Liver 2 Yo Figure 3. The 1995 inpatient use of random donor platelet concentrates by disease grouping. The use of random-donor platelet concentrates was unalyzed as described in Figure 2 for the same ten general disease groupings. transplantation receive platelet transfusions, as do about half of the patients treated for leukemia. Cardiovascular surgical patients, especially those requiring cardiac bypass, need platelet transfusions nearly 20% of the time. Hematology patients suffering from aplastic anemia, myelodysplasia, and idiopathic thrombocytopenic purpura, but not leukemia are transfused during approximately 13% of the admissions, while oncology patients receive transfusions during fewer than 5% of admissions. However, if the total inpatient use of platelet products by these same disease categories is analyzed (Fig. 2) it is readily apparent that while bone marrow transplantation patients are frequent recipients of platelet transfusions, they accounted for only 6% of the total platelets used in Together, oncology, leukemia, bone marrow transplantation, and general hematology patients consumed 39% of the platelets used in the United States in Cardiovascular surgery, general surgery, and trauma surgery patients accounted for 42% of the platelets transfused while a wide variety of other conditions (including liver disease, infectious disease, and respiratory failure patients) accounted for the remaining 19%. There was a great difference in the type of platelet product (random donor platelet concentrates, single donor apheresis platelets, and HLA-matched single donor apheresis platelets) used by these different disease categories (Figs. 3-5). Several general findings emerge. The first is that random donor platelet concentrates (Fig. 3) are used much less by hematology/oncology (including leukemia

5 Kuter 235 and BMT) patients than by surgical patients (cardiovascular, general, and trauma). Only 24% of the random donor platelet concentrates were used by hematology/oncology patients, whereas 54% were used by the surgical patients in Second, single-donor apheresis platelets (Fig. 4) are used much more often in hematologyloncology patients than in surgical patients. Fiftyone percent of the single-donor platelets were used by hematologyloncology patients, but only 32% were used by surgical patients Third HLA-matched single-donor apheresis Platelets (Fig. 5) are used to a much Cardiovascular 16% 1%,? IU 9% Oncology I 0% Hematology 10% Figure 4. The 1995 inpatient use of single donor apheresis platelets by disease grouping. The use of single-donor apheresis platelets was analyzed as described in Figure 2 for the same ten general disease groupings. greater extent in the hematology/oncology patients, accounting for 50% of the use, than in the surgical patients, who used only 20% of this product. Finally, although the total use of platelets from 1992 to 1995 remained virtually constant, with approximately 1,000,000 single-donor apheresis unit equivalents used, there has been a striking increase in the use of single-donor apheresis platelets in the United States. In 1995, 58% of the platelets transfused were obtained from single-donor apheresis products (+/- HLA matching) versus 42% from random-donor platelet concentrates (Fig. 6). With this increasing reliance upon single-donor apheresis platelets, attempts to increase the yield from apheresis procedures have received considerable attention. Efforts to improve platelet yields include: increasing the efficiency of the cell separator, reducing leukocytes during rather than after apheresis, selecting donors with high normal platelet counts, and lengthening the duration of the apheresis procedure. With the recent identification of thrombopoietic growth factors, there is much interest in using these growth factors to stimulate platelet production in apheresis donors. Since the donor s platelet count is the critical parameter in determining the yield from apheresis procedures, any significant increase in it may markedly increase the apheresis platelet yield. 1% in Leukemia 26% Figure 5. The 1995 inpatient use of HLA-matched single-donor apheresis platelets by disease grouping. The use of HU-matched single-donor apheresis platelets was analyzed as described in Figure 2 for the same ten general disease groupings.

6 236 Use of PEG-rHuMGDF in Platelet Apheresis S/D 56% H LA 2 Yo PC 42% Figure 6. The relative use of three different platelet products in the United States in As described in Figure 2, the total amount (measured in single-donor apheresis unit equivalents ) of platelets used in the United States in 1995 was estimated, and the relative amount of single-donor apheresis platelet units (S/D), HLA-matched single-donor apheresis platelet units (HLA), and random-donor platelet concentrates (PC) calculated. The total amount of platelet products used in 1995 was estimated as 1,023,612 units (measured in single-donor apheresis unit equivalents. ) NEW THROMBOPOIETIC GROWTH FACTORS Over the past four years, a number of hematopoietic growth factors with thrombopoietic potential have been described (Table 1). Interleukin 3 (IL-3) and IL-6 have modest thrombopoietic potential but are too toxic for use in a normal donor population. IL-11 (oprelvekin, NEUMEGA@) has recently been licensed in the United States for use in the prevention of thrombocytopenia secondary to chemotherapy. In the nonchemotherapy setting, it also stimulates platelet production with a peak platelet count approximately 10 to 14 days later and a maximum increase of the platelet count of approximately twofold [ 11, 121. Although the toxicity profile of IL-11 is acceptable for use in oncology patients, its universal production of anemia secondary to intravascular fluid shifts, 59% rate of edema or other fluid-retention side effects, and the 12% rate of atrial arrhythmias make it unsuitable for use in normal platelet donors. The recent discovery of thrombopoietin [ 13-17] and the introduction of several recombinant thrombopoietin molecules may provide a suitable thrombopoietic growth factor for use in a normal platelet donor population. Currently, there are two molecules under intensive investigation [ 18, 191. The first, recombinant human thrombopoietin, is a full-length, glycosylated protein which is Table 1. Thrombopoietis grov, h lactori identical to the native molecule, throm- A MPL ligands: bopoietin, and is produced by Genentech (South San Francisco, CA). The second mol- Endogenous thrombopoietin (TPO) ecule, pegylated recombinant human rhutpo megakaryocyte growth and development PEG-rHuMGDF factor (PEG-rHuMGDF), is a truncated mol- Promegapoietin (TPOOL-3 fusion protein) ecule which contains the first 163 amino TPO mimetic peptide acids of thrombopoietin and is not glycosylated but is coupled on the amino-terminus A IL-3 to polyethylene glycol and is manufactured A IL-6 by Amgen (Thousand Oaks, CA). Both of A IL-ll these recombinant molecules have a half-life A Synthokine (synthetic IL-3) in the circulation of approximately 30 to 40 h and are potent stimulators of platelet pro- A Myelopoietin (IL-36-CSF fusion protein) duction in both animal models and in A Progenipoietin G (Flt-31G-CSF fusion protein) humans. Both produce a rise in the platelet count with no effect on the hematocrit or

7 Kuter 237 white blood cell count. Furthermore, in extensive testing in cancer patients, these molecules have been shown to be virtually free of any significant toxicity [20, 211. Neither has caused any acute-phase reaction responses such as those seen with IL-11, nor has thrombosis, marrow fibrosis, or stimulation of tumor growth been seen. Both of these are potentially effective growth factors for stimulating platelet production in normal platelet donors with the goal of increasing the yield at apheresis. To date, only PEG-rHuMGDF has been tested for this purpose. ADMINISTRATION OF PEG-RHuMGDF TO NORMAL HUMAN VOLUNTEERS AND PLATELET APHERESIS DONORS To determine whether PEG-rHuMGDF might be a safe and effective means of increasing the platelet count, Tomita recently reported the results of a randomized dose escalation study in which normal volunteers were administered PEG-rHuMGDF or placebo [22]. Normal healthy subjects between 18 and 50 years of age were screened for the absence of thrombotic risk factors, including prior myocardial infarction, deep venous thrombosis, or pulmonary embolism and randomized to receive either placebo or PEGrHuMGDF. The blinded study drug was administered as a single S.C. injection on day 1. Blood samples were taken daily on days 1 through 14, three times between days 15 and 21, and on day 28 in order to measure changes in the platelet count. Doses of 1 pgkg and 3 pgkg were studied sequentially compared to placebo. Preliminary results from these studies showed that the baseline platelet counts of all the subjects were the same, ranging from 231 x 109/1 in the placebo group to 231 x 109/1 and 253 x 109/1 in the 1 pgkg and 3 pgkg PEG-rHuMGDF recipients, respectively. The median maximum platelet count in those receiving placebo injections was 258 x 10y/l and in those receiving PEG-rHuMGDF was 364 x 109/1 or 524 x 109/1, depending upon whether they received the 1 or 3 pgkg dose, respectively. This represents a 1.1-fold increase in the platelet count in the placebo-injected donors and a or 2.36-fold increase in those who received the 1 pgkg or 3 pgkg dose of PEG-rHuMGDF. No adverse events were reported in these normal volunteers, and these studies suggested that a clinically effective single dose of 3 pgkg of PEG-rHuMGDF increased the platelet count with a peak on days 10 through 14. Comparable results have recently been reported in preliminary studies using PEG-rHuMGDF in normal platelet apheresis donors [23, 241. Using a randomized, placebo-controlled, blinded, crossover, sequential dose escalation design, the safety and efficacy of PEG-rHuMGDF in the normal platelet donor setting were tested by Goodnough and colleagues [23]. Participants were healthy platelet apheresis donors who had donated twice in the preceding year, were 18 to 50 years old and did not have prior thrombotic risk factors, including those due to smoking, prior cardiovascular disease, obesity, or oral contraceptives. Donors were studied over two 28-day cycles. The blinded study drug was administered at a dose of 0, 1, or 3 pgkg on day 1 of each 28-day cycle. Platelet counts were measured on days 1, 6, 9, 12, and 15 with platelet apheresis performed on day 15 using a standardized protocol. There was a dose-dependent rise in the platelet count of those who received PEG-rHuMGDF with peak platelet counts on days 10 to 14 of 366 x 109/1 or 599 x 109/1 in those who received the 1 pgkg (n = 23) or 3 pgkg (n = 22) injections, respectively, compared to a platelet count of 225 x 109/1 in the 65 placebo cycles. The increased platelet count translated directly into an increased platelet yield at apheresis with the placebo recipients providing 3.7 x 10" platelets versus 5.6 x 10" or 11.0 x 10" platelets for those who received the 1 pgkg or 3 pgkg injections. There were no adverse events in any of the donors. In a companion study [24], the platelet product obtained from these PEG-rHuMGDF- or placebo-stimulated donors was transfused into cancer patients with chemotherapy-induced grade IV thrombocytopenia (platelet count less than or equal to 25 x 109/1). A total of 90 apheresis products were transfused. The apheresis products were transfused in their entirety and were not split into separate units. Fifty-one apheresis products obtained from placebo-injected individuals were transfused. These products contained a median of 3.7 x 10" platelets and were transfused into individuals with a baseline platelet count of

8 238 Use of PEG-rHuMGDF in Platelet Apheresis Day pr 11 I Figure 7. The platelet count response to normal versus PEG-rHuMGDFmobilized platelets transfused into a thrombocytopenic recipient. A 60-year-old woman underwent a matched, related allogeneic bone marrow transplant for refractory acute myeloid leukemia. The conditioning regimen was begun on day 1, and bone murrow was infused on day I7 (open triangle). Platelet transfusions were all single-donor apheresis units, CMV-free, and irradiated, and were administered (tj whenever the platelet count fell below 20 X 10y/l. Of these, the patient received four single-donor apheresis units (0) from study donors that had been treated with either placebo (transfused on days 13.8, 26.7, and 38.1) or 3 pg/!ig PEG-rHuMGDF (transfused on day 39.6). 13 x 109/k they produced a median platelet increase of 11 x 109A in these recipients. In contrast, 18 platelet products were transfused from donors injected with 1 pgkg of PEG-rHuMGDF. These contained a median of 5.7 x 10" platelets and were transfused into recipients whose baseline platelet count was I x I 0~11, resulting in a median increase in platelet count of 24 x IO'/l. The product from 21 donors injected with the 3 pgkg dose of PEG-rHuMGDF contained a median of 11 x 10" platelets and was transfused into recipients who had a baseline platelet count of I1 x 109/l. The transfusion resulted in a median rise in platelet count of 43.5 x 10y/l. When these platelet increments were expressed as the h corrected count increment (CCI), both PEG-rHuMGDF-mobilized platelet products demonstrated a 33% greater rise than the placebo-mobilized product. None of the recipients had any unexpected adverse event, and the instance of transfusion reactions was the same in all three groups. Figure 7 illustrates the effect of one of these PEG-rHuMGDF-mobilized platelet transfusions in a patient undergoing bone marrow transplantation. This patient had multiple transfusions, all with single-donor apheresis products and most from donors not injected with either placebo or PEGrHuMGDF. However, three of the platelet products (containing 3-4 x 10" platelets) were obtained from donors who had been treated with placebo injections and one (containing 11 x 10" platelets) was from a donor who had received a 3 pg/kg injection. The platelet count rise with the platelets from the three placebo-injected donors was comparable to that from the 20 other non-study transfusions (absolute platelet increment: 23 _t 9 x 109/1 versus 20.8 f 11 x 109/l). The transfusion of platelets from the PEG-rHuMGDF donor produced an absolute platelet count increment of 114 x 109/1 (platelet count rose from 17 x 109/1 to 131 x 109/1) but the CCI at h was no different from the other transfusions. In this and most of the other recipients of PEG-rHuMGDF-mobilized platelets, the rise in platelet count after transfusion was closely related to the dose of platelets administered, and there was a suggestion that the need for subsequent platelet transfusions was delayed when the PEG-rHuMGDFmobilized platelets were transfused. These preliminary results suggest that PEG-rHuMGDF is apparently safe and effective in stimulating normal volunteers and normal apheresis donors to increase their rate of platelet production. The peak platelet count appears to occur on days 10 through 14. The platelet yield after apheresis was directly related to the pre-apheresis platelet count. The PEG-rHuMGDF-mobilized platelet product was effective upon transfusion, giving corrected count increments which were at least comparable to those of the standard apheresis product.

9 Kuter 239 IMPLICATIONS FOR TRANSFUSION MEDICINE Although the impact of the thrombopoietic growth factors on the national need for platelets cannot yet be estimated, preliminary studies suggest that the introduction of thrombopoietic growth factors will probably not markedly alter the need for platelet products in the future [25, 261. Application of these growth factors to the surgical population would be particularly difficult to do, and since this group consumes 40% of the total product, demand by this group should remain strong. Furthermore, recent studies in bone marrow transplantation and acute leukemia, prime consumers of platelet products in oncology, have shown no particular benefit of either recombinant thrombopoietin or PEG-rHuMGDF in reducing the need for platelet transfusions [ 18, 19, 25, 261. Patients with chronic thrombocytopenic disorders have not yet been shown to benefit from administration of thrombopoietic growth factors and indeed in several situations, such as the thrombocytopenia with absent radii syndrome, these growth factors are unlikely to confer benefit because the thrombopoietin receptor-signaling mechanism is defective [27]. Although thrombopoietic growth factors will probably be successful in preventing secondary thrombocytopenia in patients undergoing nonmyeloablative chemotherapy [21,22,28], the overall need for platelet transfusions in this group of patients is relatively small. What seems promising is the use of thrombopoietic growth factors such as recombinant thrombopoietin or PEG-rHuMGDF in transfusion medicine. A number of areas of potential use have been identified [ 18, 19, 25, 261; these include stimulating peripheral blood progenitor cells in donors for bone marrow transplantation [21], expanding ex vivo cord blood or peripheral blood progenitor cells [29], and stimulating normal platelet apheresis donors [23, 24, 301. As noted above, considerable progress has recently been made in demonstrating the utility of PEG-rHuMGDF in stimulating normal platelet donors [ Although the numbers of platelet donors treated so far are certainly small, it appears that PEG-rHuMGDF is effective in increasing the yield of platelet apheresis, a procedure which in the United States and several other countries accounts for a large portion of the platelet product produced. There is a direct relationship between pre-apheresis platelet count and the apheresis platelet yield. There are potential advantages of this approach for the blood center, the donor and the platelet recipient. For the blood center, routine use of this approach could improve the supply and reduce the cost of apheresis platelet production by increasing the number of multiple products produced, reducing the number of substandard (<3 x 10 platelets) products, and allowing marginal donors (platelet counts <I50 x 109/1) to contribute. In special situations, such as rare HLA types or directed donors, use of PEGrHuMGDF may be vital to providing adequate platelet support for some patients. Coupled with cryopreservation, PEG-rHuMGDF stimulation may allow highly alloimmunized patients to collect large numbers of platelets prior to undergoing a subsequent cycle of chemotherapy. Also, since only one-third of the mobilized platelets are routinely removed at any one time by the apheresis, use of PEG-rHuMGDF could allow multiple platelet harvests over a four- or five-day period. For the donor, the use of PEG-rHuMGDF may allow shorter apheresis run times, reduce suboptimal collections, and increase the gratification of providing platelets (something often underestimated in assessing platelet donor satisfaction). For the recipient, the use of PEG-rHuMGDF-mobilized platelets presents several possible advantages related to both the number of platelets transfused and their quality. First, although not yet sub jected to rigorous human testing, platelet survival studies have demonstrated that circulating platelet survival rises with the platelet count. Conceivably, the provision of a larger platelet dose would increase survival and reduce the need for subsequent transfusions. Second, by providing platelets that are on average younger and more hemostatically active, recipients would have longer survival of platelets in the circulation, improved hemostasis, and reduced need for subsequent transfusions. As suggested by the preliminary data presented above, transfusion of platelets mobilized with PEG-rHuMGDF produced an improved CCI. These inducements to use PEG-rHuMGDF in platelet apheresis must be balanced by potential concerns that affect the donor, blood center, and recipient. First and foremost among these is the

10 240 Use of PEG-rHuMGDF in Platelet Apheresis demonstration of safety for the donor. Although the preliminary studies mentioned above described the absence of side effects of PEG-rHuMGDF in normal volunteers and platelet apheresis donors [22-241, this experience in a relatively small number of highly selected donors with single exposures to the growth factor needs to be confirmed in a much larger cohort of normal individuals with multiple exposures. Fortunately, in the cancer population where many more patients have received this molecule (some having multiple injections), no significant toxicities have yet been detected [20, 21, 281. Potential side effects include those associated with a high platelet count, such as thrombosis, as well as those related to the injection of a reconibinant molecule, such as antibody formation and allergic response. In contrast to the hematology/oncology patients, the frequency of adverse events which would be tolerated in a normal platelet donor population is difficult to assess and certainly must await the results of extensive study in normal individuals. For the blood center, numerous concerns also arise. Aside from the obvious issue of donor safety and liability, the blood center will need to be more medical in administering drugs and educating and monitoring donors undergoing PEG-rHuMGDF stimulation. At a practical level, technical issues of the apheresis procedure itself as well as issues of product storage and platelet viability need to be addressed. Some cell separators lack algorithms for collecting product from donors with platelet counts elevated above the normal range, and the efficiency of the on-line leukoreduction procedures has not yet been assessed at these higher donor platelet counts. Furthermore, when large numbers of platelets are harvested, the average storage concentration rises above the standard 1.4 X 109/ml and may affect platelet viability during storage. Finally, there are also potential concerns for the recipient. If larger numbers of platelets are to be transfused, there is a greater plasma volume administered which may affect patients with tenuous fluid balance. Also, providing more platelets with greater hemostatic potential might result in increased thrombotic events for the recipients. Clotted intravenous catheters may be one relatively minor outcome of this, but concern must be raised as to whether recipients might experience more arterial or venous thrombosis at a higher platelet count stimulated by concurrent antibiotics or chemotherapy. It is conceivable that a low platelet count in the setting of, say, bone marrow transplantation might actually be partially protective of hepatic veno-occlusive disease. CONCLUSIONS Since the first transfusions of human blood over 170 years ago, the utility of platelet transfusions in a number of thrombocytopenic conditions has been well demonstrated. Although a number of thrombopoietic growth factors have recently been identified, it is unlikely that they will have a major impact on the need for platelet transfusions in the future. Therefore, the need for platelet transfusions will persist, and since this product is increasingly being generated by apheresis technology, the provision of single-donor platelet products will continue to be a major challenge for blood centers. The recent availability of effective and presumably safe thrombopoietic growth factors such as recombinant thrombopoietin and PEG-rHuMGDF may increase the platelet yield from donors. Whether the use of these thrombopoietic growth factors will achieve widespread acceptance in platelet apheresis will primarily depend on the demonstration of the safety of these growth factors in the platelet donor. Furthermore, an opportunity now exists to test a number of issues regarding the optimal platelet dose. One of these is the hypothesis that transfusing a larger number of platelets might be more beneficial to a recipient than multiple small transfusions of the same total number of platelets. Another is to determine whether stimulated donors might provide younger platelets, which might have a prolonged circulating life span, increased hemostatic ability. and be a better transfusion product. These are all the subject of ongoing studies, and hopefully, the results will be available within the next several years. ACKNOWLEDGMENTS This work was supported by NIH Grant HL54838.

11 Kuter 24 1 REFERENCES I Denis J. A letter concerning a new way of curing sundry diseases by transfusion of blood. Philos Trans R Soc Lond 1667;2: Lower R. An account of the experiment of transfusion practised upon a man in London. Philos Trans R Soc Lond 1667;2: Blundell J. Some accounts of a case of obstinate vomiting in which an attempt was made to prolong life by the injection of blood into the veins. Med Chir Trans 1819;10: Jones HW, Mackmull G. The influence of James Blundell on the development of blood transfusion. Ann Med Hist 1928;10: Blundell J. Observations on the transfusion of blood. Lancet 1828;2: Wright JH. Differential staining of blood films and malarid parasites. J Medical Res 1902;7: Wright JH. The origin and nature of blood plates. Boston Medical and Surgical Journal 1906; Wright JH. The histogenesis of the blood platelets J Morph 1910;21: Kato T, Ogami K, Shimada Y et al. Purification and characterization of thrombopoietin. J Biochern 199S;I 19: Duke WW. The relationship of blood platelets to hemorrhagic disease: description of a method for determining the bleeding time and coagulation time. JAMA 1910;55: Gottlieb AM. A Pictorial History of Blood Practices and Transfusion. St. Joseph, Michigan: Imperial Publishing Services, Du X, Williams DA. Interleukin-11: review of molecular, cell biology, and clinical use. Blood 1997;89: Tepler I, Elias L, Smith JW et al. A randomized placebo-controlled trial of recombinant human interleukin-1 1 in cancer patients with severe thrornbocytopenia due to chemotherapy. Blood 1996;87: Lok S, Kaushansky K, Holly RD et al. Cloning and expression of murine thrombopoietin cdna and stimulation of platelet production in vivo. Nature 1994i de Sauvage FJ, Hass PE, Spencer SD et al. Stimulation of megakaryocytopoiesis and thrombopoiesis by the c-mpl ligand. Nature 1994; 369: Bartley TD, Bogenberger J, Hunt P et al. Identification and cloning of a megakaryocyte growth and development factor that is a ligand for the cytokine receptor Mpl. Cell 1994; 77: Kuter DJ, Beeler DL, Rosenberg RD. The purification of megapoietin: a physiological regulator of megakaryocyte growth and platelet production. ProcNatl AcadSciUSA 1994;91: Kuter DJ. In vivo effects of MPL ligand administration and emerging clinical applications for the MPL ligands. Curr Opin Hematol 1997; 4: Sheridan WP, Kuter DJ. Mpl ligand: mechanism of action and clinical trials. Curr Opin Hematol 1997;4: Fanucci M, Glaspy J, Crawford J et al. Effects of polyethylene glycol-conjugated recombinant human megakaryocyte growth and development factor on platelet counts after chemotherapy for lung cancer. New Engl I Med 1997;336: Basser RL, Rasko JEJ, Clarke K et al. Randomized, blinded, placebo-controlled phase I trial of pegylated recombinant human megakaryocyte growth and development factor with filgrastim after dose-intensive chemotherapy in patients with advanced cancer. Blood 1997;89: Tomita D, Petrarca M, Paine T et al. Effect of a single dose of pegylated human recombinant megakaryocyte growth and development factor (PEG-rHuMGDF) on platelet counts: implications for platelet apheresis. Transfusion 1997; 37(suppl 1):2S. 23 Goodnough LT, DiPersio J, McCullough J et al. Pegylated recombinant human megakaryocyte growth and development factor (PEG-rHuMGDF) increases platelet (PLT) count (CT) and apheresis yields of normal PLT donors: initial results. Transfusion 1997;37(suppl 1):266S. 24 Kuter D, McCullough J, Romo JD et al. Treatment of platelet (PLT) donors with pegylated recombinant human megakaryocyte growth and development factor (PEG-rHuMGDF) increases circulating PLT counts (CTS) and PLT apheresis yields and increases platelet increments in recipients of PLT transfusions. Blood 1997;90(suppl):579a. 25 Kuter DJ. Thrombopoietin: biology and clinical applications. The Oncologist 1996; 1 : Kuter DJ. Thrombopoietins and thrombopoiesis: a clinical perspective. Vox Sang 1998; (in press).

12 242 Use of PEG-rHuMGDF in Platelet Apheresis 27 Ballmaier M, Schulze H, Straub G et al. Thrombopoietin in patients with congenital thrombocytopenia and absent radii: elevated serum levels, normal receptor expression, but defective reactivity to thrombopoietin. Blood I 997;90: Vadhan-Raj S, Verschraegen C, McGarry L et al. Recombinant human thrombopoietin (rhtpo) attenuates high-dose carboplatin (C)- induced thrombocytopenia in patients with gynecological malignancy. Blood 1997; 90( suppl): 580a. 29 Piacibello W, Sanavio F, Garetto L et al. Extensive amplification and self-renewal of human primitive hematopoietic stem cells from cord blood. Blood 1997;89: Kuter DJ, Hunt P, Sheridan WP et al., eds. Thrombopoiesis and Thrombopoietins: Molecular, Cellular, Preclinical, and Clinical Biology. Totowa, New Jersey: Humana Press, 1997.