(Rhtpo) After Autologous Bone Marrow Transplantation: a Phase I Pharmacokinetic and Pharmacodynamic Study

Home , Oprelvekin

Bone Marrow Transplantation (2001) 27, 261–268  2001 Nature Publishing Group All rights reserved 0268–3369/01 $15.00 www.nature.com/bmt Haematopoietic growth factors Recombinant human thrombopoietin (rhTPO) after autologous bone marrow transplantation: a phase I pharmacokinetic and pharmacodynamic study

SN Wolff1, R Herzig2, J Lynch3, SG Ericson3, JP Greer1, R Stein1, S Goodman1, MC Benyunes4, M Ashby4, DV Jones Jr4 and J Fay5

1Vanderbilt University, Nashville, TN; 2University of Louisville, Louisville, KY; 3University of West Virginia, Morgantown, WV; 4Genentech, Inc., South San Francisco, CA; and 5Baylor University Medical Center, Dallas, TX, USA

Summary: Thrombocytopenia following myelotoxic therapy is a common problem and can lead to substantial morbidity and Thrombocytopenia following myelotoxic therapy is a mortality when severe (Ͻ10 000–20 000/␮l). Other detri- common problem and when severe (Ͻ20 000/␮l) can mental events such as transfusion reaction, alloimmuniz- lead to severe morbidity and mortality. Thrombopoietin ation, infection, the necessity of frequent outpatient follow- (TPO) is a naturally occurring glycosylated peptide up and the time to return to normal quality of life may also which stimulates the differentiation of bone marrow depend on the duration of severe thrombocytopenia. stem cells into megakaryocyte progenitor cells, induces Thrombocytopenia-associated bleeding is treated with the expression of megakaryocyte differentiation mark- transfusions of platelets obtained from volunteer blood ers, promotes megakaryocyte proliferation, polyploidiz- donors.1,2 There is a lack of uniformity about what platelet ation and, ultimately, the formation of increased num- count warrants prophylactic transfusion. Current practice is bers of platelets in the circulation. TPO has now been to maintain the platelet count above 10 000–20 000/␮l, produced by recombinant technology and has entered especially with active bleeding or during periods of treat- clinical trials. This open label phase I study was ment-related morbidity. However, recent studies have sug- designed to determine the safety, tolerance and pharma- gested that many patients will be able to tolerate the lower cokinetics of recombinant thrombopoietin (rhTPO) platelet threshold for transfusion or can be transfused only when administered to patients after undergoing high- when febrile or actively bleeding.3–6 Patients requiring dose chemotherapy followed by autologous bone mar- invasive procedures must sustain substantially higher row transplantation. rhTPO was administered intra- platelet levels. venously by bolus injection at doses ranging from 0.3 Patients undergoing dose-intensive or myeloablative to 4.8 ␮g/kg/day every 3 days to 30 patients and anti-neoplastic therapy will universally develop severe and 0.6 ␮g/kg daily to three patients. rhTPO was begun the sustained thrombocytopenia.7 Autologous hematopoietic day after marrow infusion and continued until platelet stem cell transplantation as an adjunct to dose-intensive recovery to Ͼ20 000/␮l. G-CSF was concomitantly therapy, while hastening recovery and ultimately restoring administered to promote myeloid recovery. Serious hematopoiesis, is still associated with profound thrombocy- adverse events or neutralizing antibodies to rhTPO topenia.8–10 The duration of severe thrombocytopenia is the were not observed during the study. Median platelet shortest with the transplantation of peripherally derived recovery after ABMT was 19 days (range, 11–41). stem cells (PBSC) mobilized by chemotherapy and/or Neither the dose nor the schedule of rhTPO appeared cytokine(s). The median time from PBSC transplantation to have any impact upon the time course of platelet (PBSCT) to platelet recovery is in the order of 10–13 days recovery. In this phase I study, rhTPO was found to be compared to more than 1 week longer with stem cells well tolerated without the development of neutralizing directly harvested from marrow.11 Stem cells which are antibodies and without compromising neutrophil recov- manipulated ex vivo are associated with more prolonged ery. Platelet recovery was similar for all doses studied thrombocytopenia compared with unmanipulated cells from warranting further evaluation in phase II and III trials the same source.12 The number of stem cells inversely designed to test for platelet recovery efficacy. Bone Mar- affects the duration of thrombocytopenia.13,14 This relation- row Transplantation (2001) 27, 261–268. ship is important since approximately one quarter of Keywords: thrombocytopenia; rhTPO; platelet recovery patients who undergo autologous mobilization of peripheral stem cells mobilize low numbers of stem cells necessitating multiple leukaphereses. In fact, some patients or even nor- Correspondence: Dr SN Wolff, Bone Marrow Transplant Program, 2617 mal allogeneic donors do not mobilize at all, or mobilize TVC, Vanderbilt University, Nashville, TN 37232–5505, USA so poorly, that they require direct bone marrow harvesting Received 1 May 2000; accepted 13 September 2000 to assure an adequate graft.15,16 rhTPO after auto-BMT SN Wolff et al 262 Thrombopoietin (TPO) is a naturally occurring glycosyl- organ function (bone marrow function: WBC у3000/␮l, ated peptide which stimulates the differentiation of bone absolute neutrophil count у1500/␮l, platelet count between marrow stem cells into megakaryocyte progenitor cells, 100 000/␮l and 500 000/␮l, and a minimum marrow mono- induces the expression of megakaryocyte differentiation nuclear cell harvest of 2 × 108/kg for transplantation; hep- markers, promotes megakaryocyte proliferation and poly- atic function: SGOT (AST)/SGPT (ALT) Ͻ3 × normal ploidization and, ultimately, the formation of increased and bilirubin Ͻ1.5 × normal; renal function: creatinine numbers of platelets in the circulation.17–19 Additionally, Ͻ1.5 × normal; cardiac function: ejection fraction of у50% there is enhanced proliferation of early erythroid and (±5%) by echocardiogram or MUGA scan; pulmonary myeloid progenitors.20 Both the full length intact molecule function: (DLCO) of Ͼ50% of predicted. A target of (TPO 1–332) and peptide fragments derived from the 2 × 108/kg total nucleated cells was the minimum target for amino terminus (TPO 1–153) are capable of differentiating adequacy of marrow collection which was the standard of progenitor cells into megakaryocytes and increasing the care at the time of the trial. Marrow CD34+ cell content platelet count following in vivo administration.21 TPO is was measured but not used as an eligibility criterion. the most effective cytokine to hasten recovery from Patients were deemed ineligible if there was any history of thrombocytopenia in in vitro and in vivo models of severe a functional platelet disorder such as idiopathic thrombo- thrombocytopenia.22–24 cytopenic purpura; a history of ischemic, thrombotic, Recombinant human thrombopoietin (rhTPO; Genen- or embolic phenomena; an uncontrollable cardiac tech, South San Francisco, CA, USA; Pharmacia and dysrhythmia; prior therapy with mitomycin C or a Upjohn, Bridgewater, CT, USA) has been entered into a nitrosourea; three or more prior chemotherapy regimens; broad array of clinical trials in the setting of myelosup- prior myeloablative therapy with ABMT or PBSCT; a his- pressive and myeloablative cytotoxic therapy. Other forms tory of malignant disease within the central nervous system of the molecule have also been evaluated in clinical or the bone marrow; use of any marrow purging; or any studies.25–29 experimental agents within the 4 weeks prior to beginning The phase I study we are now reporting was conducted this trial. to evaluate the safety, pharmacokinetics and potential activity of rhTPO in reducing the time to platelet transfusion independence after high-dose chemotherapy with Study design and rhTPO administration autologous bone marrow transplantation (ABMT) in women with breast cancer. Issues planned for evaluation This was a multicenter phase I, non-randomized, open- by this study included the impact of rhTPO on G-CSF label, dose escalation trial of multiple intravenous doses of enhanced neutrophil recovery, pharmacokinetic evaluation rhTPO in patients following ABMT. The study was IRB of dose and schedule, formation of neutralizing antibody approved at each treatment center and all patients gave against rhTPO and other safety issues. ABMT (as opposed informed consent. Patients were assigned to six dosing regi- to PBSCT) was used to facilitate the observation of the mens of rhTPO (Table 1). All patients received their mar- effect of TPO on the duration of thrombocytopenia since row reinfusion on day 0, and began rhTPO 1 day later (day platelet recovery after ABMT is more prolonged compared 1). No intra-patient dose-escalation was permitted. rhTPO, with PBSCT and since no adverse impact upon overall out- based on actual baseline body weight, was administered come was anticipated using bone marrow. It was hoped that intravenously by bolus injection every 3 days for a this phase I study would also suggest a dose of rhTPO to maximum of 10 injections (dose kevels 1–4 and 6), or daily for a maximum of 21 injections (dose level 5), or until the be used in more definitive phase II/III studies to test for Ͼ ␮ platelet recovery efficacy. untransfused platelet count was 75 000/ l for at least two consecutive measurements separated by at least 24 h. The study period consisted of the 35 days following the initial rhTPO administration with follow-up 1 month later. Methods

Objectives

The primary objectives of this phase I study were to deter- Table 1 Dose levels of rhTPO mine the safety, tolerance and pharmacokinetics of intravenously (i.v.) administered rhTPO in patients following Dose level rhTPO Dose regimen Number No. doses ABMT. The secondary objectives were to evaluate biologic ␮g/kg/day patients rhTPO per activity by the time to platelet recovery and transfusion patient median independence and to determine the maximally tolerated or (range) biologically active dose in the absence of dose-limiting toxicity. 1 0.3 q3 days × 10 7 7 (2–11) 2 0.6 q3 days × 10 7 7 (3–17) 3 1.2 q3 days × 10 7 7 (3–11) Patient eligibility 4 2.4 q3 days × 10 3 7 (3–16) 5 0.6 daily × 21 3 19 (19–21) Patients with advanced breast cancer (stage III/IV) were 6 4.8 q3 days × 10 6 7.5 (4–20) eligible if they had a Zubrod performance status Ͻ2, were Total 33 age у18 years old, and had adequate hematopoietic and

Bone Marrow Transplantation rhTPO after auto-BMT SN Wolff et al 263 Cytotoxic treatment rhTPO antibody assay and serum rhTPO measurements After harvest and cryopreservation (using standard Serum samples were screened for the presence of anti- methodology) of unmanipulated autologous bone marrow, rhTPO antibodies using a panel of four assays: ELISA patients were treated with cyclophosphamide 180 mg/kg based on full-length rhTPO molecule; ELISA based on a (60 mg/kg/day for 3 consecutive days), thiotepa 900 mg/m2 truncated rhTPO molecule; ELISA based on blocking the (300 mg/m2 for 3 consecutive days) with or without car- binding of rhTPO to the c-mpl receptor; bioassay based on boplatin 600 mg/m2, followed in 3 days by infusion of rap- inhibition of an rhTPO-dependent cell line.33 Two baseline idly thawed marrow cells (day 0). All patients received G- samples were collected prior to infusion of rhTPO followed CSF subcutaneously at a dose of 5 ␮g/kg beginning on by weekly sampling until day 35 and then 1 month later day 1 until the absolute neutrophil count was maintained (day 60). Serum rhTPO concentrations were measured by Ͼ500/␮l for 2 consecutive days. Platelets (single donor or an ELISA method using the recombinant c-mpl receptor as random pooled) were transfused for a morning platelet a coat with lower limit of detection of 0.078 ng/ml. count Ͻ20 000/␮l or whenever clinically indicated for Samples were obtained just prior to the first infusion and active bleeding. At the time of this study, this platelet at the following time points after administration: 5 min, 30 transfusion trigger was the standard of practice. min, 1 h, 2 h, 4 h, 6 h, 8 h, 12 h, 18 h, 24 h and 48 h. On the 4th, 7th, 10th, 13th, 16th, 19th, 22nd, 25th and 28th day of administration samples were obtained prior to and Statistical analysis and definitions 5 min after rhTPO administration. The study was prospectively designed to use the Jon- kheere–Terpstra test for a decreasing trend in time to platelet transfusion independence with increasing dose.30,31 Test Results statistics were compared to quartiles of the tabled exact null 32 distribution. These nonparametric tests were appropriate Patients for the study endpoint with a small sample size and mini- mal censoring. Dose and schedule cohorts were planned to A total of 33 female patients were enrolled. The median consist of approximately six subjects per dosing level for age was 48.2 years (range 24–61) with 29 Caucasians, two a total of 36 subjects for the regimens. This sample size was African-Americans, and two Hispanics enrolled. Twenty- chosen to achieve adequate power (Ͼ80%) for the dose- seven patients had a PS of 0, and six had a PS of 1. The response trend test in the case of a strong decreasing trend median number of months since diagnosis was 9.0 (range, in time to platelet recovery (linear decline in median days 3.6–124). Seventeen patients had received two prior regi- from 23 at 0.3 mg/kg to 13 at 4.8 mg/kg) and a constant or mens of chemotherapy, while 16 had received only one decreasing variance with increasing dose level. For end- prior regimen; nine patients had also received one prior points such as occurrence of a specific adverse event or course of radiotherapy. Twenty-three of the patients were formation of antibodies to rhTPO, this sample size guaran- treated with cyclophosphamide, thiotepa and carboplatin teed at least 80% power to detect one or more events among and the remaining 10 received cyclophosphamide and the six subjects if the true event probability is у24%. Tox- thiotepa. The median number (range) of marrow harvested icity was graded using the NCI Common Toxicity Criteria total nucleated cells × 108/kg and CD34 × 106/kg was 2.25 with serious adverse events prospectively defined as grade (0.20–193.7) and 2.82 (0.00–13.4), respectively. Compari- IV or V. son between treatment groups (ANOVA) of the TNC and To facilitate comparison with other trials, platelet recov- CD34+ content of the harvest marrow was not significantly ery was defined as the first day of 3 consecutive days of different (P = 0.06 and P = 0.59, respectively). an untransfused platelet count у20 000/␮l (definition 1). and more rigorously as the first of у2 consecutive days of untransfused platelet counts у20 000/␮l with the second rhTPO dosing count greater than or equal to the preceding value (definition 2). Subjects not achieving platelet recovery by The first subject began dosing on 7 May, 1996 and the last the end of the study were censored on the day of the last subject completed dosing on 4 March, 1997. Seven patients measured platelet count. Neutrophil recovery was defined were entered on each of the first three dose levels, three as the first of 2 consecutive days with an absolute neutro- patients on each of the next two levels and six on the high- phil count у500/␮l. est dose level (Table 1). This variance from the scheduled six patients per level was due to the concomitant entrance of patients at multiple centers during the early part of the rhTPO trial and the desire to treat patients at the highest dose level during the later part of the study. A total of 225 doses of rhTPO was provided by Genentech Inc as a sterile liquid rhTPO were delivered on the every third day schedule with ready for parenteral administration in 3 ml glass vials in a median of seven doses/patient administered with the preservative-free buffer. rhTPO was shipped on wet ice, exception of the 4.8 ␮g/kg dose level, where a median of and stored refrigerated at 2–8°C. Sites were instructed to 7.5 doses/patient were administered (Table 1). A total of avoid freezing, vigorous agitation, exposure to bright light 59 doses were administered on the 0.6 ␮g/kg daily sched- or using the vials beyond the stamped expiration date. ule, with a median of 19 doses/patient.

Bone Marrow Transplantation rhTPO after auto-BMT SN Wolff et al 264 Toxicity Table 2. The peak rhTPO levels ranged from a 10-fold to a 90-fold increase over the baseline. Trough rhTPO and All patients experienced adverse events within the spectrum interdose trough levels were 2.3–10 times greater than base- of toxicities that are observed after myeloablative therapy. line levels (Figure 1). Maximum mean concentration (Cmax) No pattern or unanticipated toxicity was perceived to be increased with dose except for the 0.6 ␮g/kg every third related to the administration of rhTPO. A total of 14 serious day schedule. Cmax was observed at the first sampling (5 adverse events were observed in 10 patients; none was min after bolus administration) in most subjects, and time deemed by the investigators to be possibly or probably of maximum concentration (Tmax) ranged from 5 min to related to rhTPO administration. There was one death on 4 h. With daily dosing, significant accumulation of TPO study, a 52-year-old female died of multi-system organ fail- was not observed although modest increase from trough ure on study day 20 secondary to toxicities of the prepara- levels were noted (Figure 2). Systemic clearance (Cl) esti- tive regimen, having received seven doses of rhTPO. The mates ranged from 0.18 to 4.9 ml/h/kg and appeared to overall death rate during the study period was 3% (95% increase with increasing rhTPO dose until reaching a pla- confidence interval of 0.5–15%). One patient experienced teau at 2.4–4.8 ᒊg/kg. The terminal half-life (t ) ranged a subdural hematoma on study day 16 with a platelet count . ␮ from 21.4 to 637 h; mean values for those subjects receiv- of 25 000/ l and was treated with platelet transfusions fol- ing 0.3 or 0.6 ␮g/kg were largely inflated by two patients lowed by surgical evacuation and recovered. One patient with t values of Ͼ600 and Ͼ200 h. At doses у1.2 ␮g/kg, experienced a perceived delayed neutrophil recovery (day . the mean t. was approximately 38 h. The area under the 28) which was felt to be secondary to a human herpes virus concentration × time curve extrapolated to infinity (AUC ) 6 infection. No patient experienced any thrombotic event inf was estimated using 72 h (AUC0–72) of observations, but or developed veno-occlusive disease of the liver. since the decline in rhTPO concentration was not monoex- In addition to the patient expiring from cytotoxic regi- ponential, these results should be interpreted with caution. men-related toxicity, one other patient did not complete full In patients receiving 0.6 ␮g/kg daily for up to 21 doses, rhTPO dosing, This patient discontinued study drug after ± ± mean Cmax was 13.0 ( 4.11) and 36.5 ( 8.37) ng/ml at day seven doses due to an anxiety event. During the study and 1 and day 19, respectively. As shown in Figure 2, rhTPO follow-up period, five patients experienced eight non-seri- concentrations did not appear to have reached a plateau by ous events regarded as possibly or probably related to the last rhTPO administration (day 19 or 21). rhTPO administration. One episode of asthenia was categorized as severe grade III, the remaining events were graded as mild or moderate grade I–II. These included rash (n = 3), Hematologic recovery back pain, chills, myalgias, and vasodilatation with facial = All 33 patients in this trial developed severe thrombocytop- flushing (n 1 each). None of these side-effects had any enia (Ͻ20 000/␮l) and received at least two platelet trans- influence upon subsequent dose administration. There were fusions. The median number of platelet transfusions per no clinically significant change in vital signs or findings on patient was 7 (range, 2–20) without any apparent effect of physical examination related to rhTPO administration. rhTPO dose or schedule (Table 3). The median last day of Trend analysis was performed at the completion of this platelet transfusion was day 16 (range, 8–37) across all study for the following parameters: platelet and neutrophil doses and schedules. The median day to platelet recovery counts, hemoglobin, AST, ALT, total bilirubin, serum crea- by definition 1 was 19 (range, 11–41). Recovery of platelet tinine, and blood urea nitrogen (BUN). No significant count by this definition by rhTPO dose is demonstrated in trends in the occurrence, extent, or distribution of abnormal Figure 3. The median day to platelet recovery by the more values (ie values outside of normal ranges) were apparent stringent definition 2 was 18 (range, 9–41). Recovery of for any of the laboratory parameters. platelet count by this definition by rhTPO dose is demon- Three patients were noted to have activity in the ELISA strated in Figure 4. No dose-related or schedule effect on based on full-length rhTPO either at screening (prior to the time to platelet recovery was apparent. Subjects receiv- receiving any rhTPO) or at some later time point. No ing 2.4 ␮g/kg every third day of rhTPO had the shortest patient at any time point was positive in the ELISA assay platelet recovery times with respect to all definitions. Sub- based on truncated rhTPO, the ELISA c-mpl blocking jects receiving 1.2 ␮g/kg every third day of rhTPO had assay, or in the bioassay which detects neutralizing anti- some of the longest recovery times with respect to all body. definitions. The median day to platelet recovery to у50 000/␮l and у ␮ rhTPO pharmacokinetics 100 000/ l were 23 (range, 12–62) days and 26 (range, 15–62+) days, respectively. There were suggestions of fas- Prior to rhTPO administration, measurable TPO levels were ter recoveries to у50 000/␮l and у100 000/␮l favoring found in all patients. The mean baseline TPO level was higher doses and the daily dosing schedule (3.5–5 day dif- 0.96 ng/ml (range, 0.14–2.36 ng/ml). Mean baseline TPO ference to у50 000/␮l and 8-day difference у100 000/␮l), concentration for each dosing level was 1.01 (±0.68), 0.80 although these did not reach statistical significance with (±0.44), 1.07 (±0.37), 0.77 (±0.57) and 1.17 (±0.64) ng/ml, P = 0.41 and P = 0.16, respectively. respectively for patients receiving 0.3, 0.6, 1.2, 2.4 and Sixteen patients received marrow transplants with low 4.8 ␮g/kg every third day and 0.77 (±0.62) ng/ml for sub- CD34+ cells numbers (median 1.3 × 106 CD34+/kg, range, jects receiving 0.6 ␮g/kg daily. 0–2.5) and therefore were felt to be at risk for delayed Selected pharmacokinetic parameters are noted in recovery. However, there was no delay in count recovery

Bone Marrow Transplantation rhTPO after auto-BMT SN Wolff et al 265 Table 2 Selected pharmacokinetic parameters

␮ Dose ( g/kg q3d) Cmax (ng/ml) Cl (ml/h/kg) Vss (ml/kg) t. (h) AUC0–72 (ng/ml/h) AUCinf (ng/ml/h)

0.3 12.9 ± 6.1 0.81 ± 0.345 64.0 ± 47.0 124 ± 227 235 ± 78 532 ± 505 0.6 11.6 ± 5.0 1.16 ± 0.433 98.7 ± 46.7 70.7 ± 58.6 345 ± 171 596 ± 250 1.2 29.8 ± 6.2 1.46 ± 0.477 71.4 ± 23.5 38.1 ± 12.1 654 ± 172 900 ± 286 2.4 39.6 ± 20.2 3.22 ± 1.500 155 ± 99.5 37.1 ± 7.3 669 ± 289 844 ± 319 4.8 93.0 ± 26.1 2.09 ± 0.290 102 ± 22.2 37.9 ± 7.4 1775 ± 261 2337 ± 304

Cmax = maximum concentration; Cl = systemic clearance; VSS = volume of distribution at steady state; t. = terminal half-life; AUC0–72 = area under the × = × concentration time curve from 0 to 72 h after dosing; AUCinf area under the concentration time curve extrapolated to inﬁnity.

1.0 100 30 0.8 10 0.6

1 0.3 m g/kg 0.6 m g/kg 0.4 1.2 m g/kg 2.4 m g/kg rhTPO q3d×10 @0.3 m g/kg (n = 7) 4.8 m g/kg rhTPO q3d×10 @0.6 m g/kg (n = 7) 0.0781 0.2 rhTPO q3d×10 @1.2 m g/kg (n = 7) rhTPO q3d×10 @2.4 m g/kg (n = 3) rhTPO q3d×10 @4.8 m g/kg (n = 6)

Mean serum rhTPO (ng/ml) rhTPO qd×21 @0.6 m g/kg (n = 3) Platelet recovery probability 048 96 144 192 240 288 336 384 432 480 528 576 624 0.0 Hours after first rhTPO infusion 0 10 20 30 40 50 60

Figure 1 Mean serum rhTPO levels in (ng/ml) measured by the rhTPO Days post-transplant MPL assay in hours after first rhTPO infusion categorized by every third day varying dosing regimen. Figure 3 Day after marrow infusion to recovery of unsupported platelet count (Ͼ20 000/␮l) by rhTPO dosing regimen with recovery defined as the first day of 3 consecutive days of an untransfused platelet count Ͼ20 000/␮l (definition 1). 100 30 1.0 10

0.8 1 0.6 m g/kg 0.6

0.0781 0.4 Mean serum rhTPO (ng/ml) 0 48 96 144 192 240 288 336 384 432 480 528 576 624 rhTPO q3d×10 @0.3 m g/kg (n = 7) rhTPO q3d×10 @0.6 m g/kg (n = 7) 0.2 rhTPO q3d×10 @1.2 m g/kg (n = 7) Hours after first rhTPO infusion rhTPO q3d×10 @2.4 m g/kg (n = 3) rhTPO q3d×10 @4.8 m g/kg (n = 6) rhTPO qd×21 @0.6 m g/kg (n = 3) Figure 2 Mean serum rhTPO levels in (ng/ml) measured by the rhTPO Platelet recovery probability 0.0 MPL assay in hours after ﬁrst rhTPO infusion by daily dosing regimen 0 10 20 30 40 50 60 of 0.6 ␮g/kg. Days post-transplant

Figure 4 Day after marrow infusion to recovery of unsupported platelet Table 3 Results of dose levels of rhTPO on platelet recovery to count (Ͼ20 000/␮l) by rhTPO dosing regimen with recovery defined as Ͼ20 000/␮l the first of у2 consecutive days of an untransfused platelet count Ͼ20 000/␮l with the second count being greater than or equal to the pre- Dose level Days to last Number of Day platelet Day platelet ceding value (definition 2). (␮g/kg/day) platelet platelet recovery recovery transfusion events/patient definition 1 definition 2 median median median median for this subset, in whom the median day to platelet recovery (range) (range) (range) (range) to у20 000/␮l (stringent definition 2) was day 15 (range, 8–37). 0.3 q 3 day 15 (8–19) 7 (2–11) 19 (11–22) 16 (9–22) The median time to neutrophil recovery to у500/␮l and 0.6 q 3 day 16 (11–37) 5 (3–17) 18 (13–38) 18 (12–38) у1500/␮l was 12 days (range, 10–28) and 14 days (range, 1.2 q 3 day 19 (12–20) 7 (3–11) 22 (15–23) 20 (13–27) 2.4 q 3 day 13 (9–32) 6 (3–16) 15 (12–41) 14 (10–41) 11–31), respectively. There was no evidence of either a 0.6 daily 14 (13–20) 7 (4–10) 17 (15–21) 16 (15–21) dose-dependent enhancement or delay of neutrophil recov- 4.8 q 3 day 18 (15–27) 11.5 (4–20) 19 (16–38) 19 (13–28) ery across any of the rhTPO dose groups. For each patient, a median of four red blood cell transfusions were adminis-

Bone Marrow Transplantation rhTPO after auto-BMT SN Wolff et al 266 tered (range, 2–18). Similarly, the number of red blood cell power to detect inter-dose differences. Alternatively, larger transfusion was not related to the dose or administration doses of rhTPO or modification of administration method schedule of rhTPO. or schedule could still prove effective. Inference about further rhTPO dose augmentation to increase response may be extrapolated from the pharmacokinetic data. Our study Discussion observed high peak (post-dose) and trough levels (prior to the next dose) compared with baseline with similar levels In this report, a c-mpl ligand (rhTPO) was studied after measured in patients with severe thrombocytopenia due to ABMT. The primary objective of this phase I trial was to diminished production.36–38 These observations demonstrat- evaluate the safety and tolerance of escalating doses of ing substantial rhTPO levels make it unlikely that even rhTPO in patients undergoing ABMT for breast cancer con- larger doses or schedule modification would be more ben- comitantly receiving the myeloid growth factor, G-CSF. eficial. Definitive efficacy will require phase II and III stud- Although there were preliminary data suggesting that ies designed to primarily test for platelet recovery. Which rhTPO was safe in combination with G-CSF, synergistic dose of rhTPO one would consider for future trials is not toxicities were a concern with the use of these lineage-spe- clearly defined since our trial did not demonstrate toxicity cific cytokines.9,34 In this study, rhTPO was shown to be or any pronounced dose effect. Considering the above facts, safe, since all severe adverse events were those anticipated a dose in the mid-range, such as 2.4 ␮g/kg, would be in patients undergoing high-dose therapy followed by reasonable for further evaluation. ABMT. One patient developed severe asthenia felt possibly Interleukin 11 (oprelvekin, Neumega; Genentics Insti- related to the study drug, though this event is also common tute, Cambridge, MA, USA) has been approved by the US place after high-dose therapy. Therefore, substantial side- Food and Drug Administration for the secondary prophy- effects were not observed and no dose-limiting toxicity was laxis of thrombocytopenia associated with myelosuppress- noted. Neutralizing antibodies to rhTPO were not detected ive chemotherapy. Its use has been modest because of in any patient at any time point. observed side-effects.39 The effects of interleukin 11 on In our trial, all doses of rhTPO resulted in an equivalent thrombocytopenia after hematopoietic stem cell transplan- time course of platelet and neutrophil recovery. Since there tation have also been published.40 In that report, 80 women was no evidence of an adverse effect on neutrophil recov- with breast cancer were randomized to placebo or one of ery, with the exception of one patient who may have dem- two doses of oprelvekin after ABMT and PBSCT. The onstrated human herpes virus 6-related pancytopenia, the median time to platelet transfusion independence was simi- potential of stem cell competition by one cytokine over lar between the three groups, though there were trends to another was excluded. fewer platelet transfusion events and to a lower incidence Without a concurrent control group, the absolute effect of failure to engraft platelets by day 30 in the oprelvekin- on platelet recovery can only be surmised by comparison treated groups. with historical controls. Bernstein et al11 in a large retro- The most potent cytokine for platelet recovery in pre- spective study of thrombocytopenia after transplantation, clinical trials is c-MPL ligand.41 Thus far, only one other reported in a small cohort of patients that the time to plate- report of a c-mpl ligand, pegylated megakaryocyte growth let recovery after ABMT as 40 days (95% confidence inter- and development factor (Peg-MGDF; Amgen, Thousand val of 27–49). That report included all types of diseases Oaks, CA, USA) has reported results after ABMT.35 Bever- undergoing ABMT, but still should be valid for comparison idge randomized 50 patients to either placebo or 5 or since disease does not directly correlate with hematopoietic 10 ␮g/kg/day of Peg-MGDF and noted a 6-day reduction recovery after ABMT. In our data, using a similar stringent in the time to platelet recovery (from 24 to 18 days) definition, the overall median time to platelet recovery was between placebo and the higher dose level, as well as a 7- 18 days. While such comparison must be made very cau- day decrease in the duration of profound thrombocytopenia tiously, platelet recovery in this study, appeared to be more and a 45% decrease in platelet transfusion events (though rapid when compared with ABMT without rhTPO. The the latter two did not reach statistical significance). It is results in our trial are also quite similar to those reported currently accepted that the day of platelet recovery after with the pegylated megakaryocyte growth and develop- stem cell transplantation is related to the dose of CD34+ ment factor.35 cells/kg body weight infused, with recoveries tending to be In the report of Bernstein et al,11 the recovery of platelets more rapid with cell doses exceeding 2.5 × 106 CD34+/kg.11 after PBSCT was 13 days (95% confidence interval of 12– In that report, the median CD34+ cell count in the marrow 13). Our study in recipients of ABMT with rhTPO reports harvest was 1.6 × 106/kg. In our study, patients with low a recovery which compares favorably, although is still CD34+ cell counts had similar and rapid recovery compared delayed. Without direct comparison, this must also be to the entire group. viewed very cautiously considering the variable count In summary, rhTPO in combination with G-CSF was increment after platelet transfusion potentially obscuring found to be safe and well-tolerated in patients undergoing the earliest time to platelet recovery, as well as small differ- myeloablative therapy followed by ABMT rescue. No evi- ences in time intervals. dence of adverse interaction with G-CSF (stem cell steal) The fact that there was no apparent dose-response effect was noted. When compared with historical ABMT controls, in this small phase I study may indicate that platelet recov- rhTPO at all studied doses and schedules, may have ery was not impacted by rhTPO or that at all doses resulted enhanced the recovery from thrombocytopenia and in equivalent thrombopoietic activity with inadequate approached that achieved after PBSCT, although no con-

Bone Marrow Transplantation rhTPO after auto-BMT SN Wolff et al 267 clusions about efficacy can be made. Further investigations 16 Tricot G, Jagannath S, Vesole D et al. Peripheral blood stem with larger numbers of patients will be necessary to evalu- cell transplants for multiple myeloma: identification of favor- ate efficacy and define a suitable dose for further study. able variables for rapid engraftment in 225 patients. Blood Nevertheless, this trial demonstrates the safety of rhTPO in 1995; 85: 588–596. the clinical setting of high-dose chemotherapy followed by 17 Alexander WS, Begley CG. Thrombopoietin in vitro and in vivo. Cytokines Cell Mol Ther 1998; 4: 25–34. ABMT for patients with breast cancer. 18 Baatout S. Thrombopoietin. A review. Haemostasis 1997; 27: 1–8. 19 Cohen-Solal K, Debili N, Vainchenker W, Wendling F. References Thrombopoietin (Mpl ligand) and the regulation of platelet production. Eur Cytokine Netw 1997; 8: 311–314. 1 Fetscher S, Mertelsmann R. Supportive care in hematological 20 Kaushansky K. Thrombopoietin and the hematopoietic stem malignancies: hematopoietic growth factors, infections, trans- cell. Blood 1998; 92: 1–3. fusion therapy. Curr Opin Hematol 1999; 6: 262–273. 21 Brereton ML, Adams JA, Briggs M, Liu Yin JA. The in vitro 2 Webb IJ, Anderson KC. Risks, costs, and alternatives to plate- effect of pegylated recombinant human megakaryocyte growth let transfusions. Leuk Lymphoma 1999; 34: 71–84. and development factor (PEG rHuMGDF) on megakaryopo- 3 Gmur J, Burger J, Schanz U et al. Safety of stringent prophy- iesis in patients with aplastic anaemia. Br J Haematol 1999; lactic platelet transfusion policy for patients with acute leu- 104: 119–126. kaemia. Lancet 1991; 338: 1223–1226. 22 Neelis KJ, Qingliang L, Thomas GR et al. Prevention of 4 Heckman KD, Weiner GJ, Davis CS et al. Randomized study thrombocytopenia by thrombopoietin in myelosuppressed rhe- of prophylactic platelet transfusion threshold during induction sus monkeys accompanied by prominent erythropoietic stimu- therapy for adult acute leukemia: 10 000/microl versus lation and iron depletion. Blood 1997; 90: 58–63. 20 000/microl. J Clin Oncol 1997; 15: 1143–1149. 23 Harker LA. Physiology and clinical applications of platelet 5 Rebulla P, Finazzi G, Marangoni F et al. A multicenter ran- growth factors. Curr Opin Hematol 1999; 6: 127–134. domized study of the threshold for prophylactic platelet trans- 24 Harker LA, Marzec UM, Novembre F et al. Treatment of fusions in adults with acute myeloid leukemia. Gruppo Itali- thrombocytopenia in chimpanzees infected with human ano Malattie Ematologiche Maligne dell’Adulto. New Engl J immunodeficiency virus by pegylated recombinant human Med 1997; 337: 1870–1875. megakaryocyte growth and development factor. Blood 1998; 6 Wandt H, Frank M, Ehninger G et al. Safety and cost effec- 91: 4427–4433. tiveness of a 10 × 109/l trigger for prophylactic platelet trans- 25 Basser RL, Rasko JE, Clarke K et al. Randomized, blinded, fusions compared with the traditional 20 × 109/l trigger: a pro- placebo-controlled phase I trial of pegylated recombinant spective comparative trial in 105 patients with acute myeloid human megakaryocyte growth and development factor with leukemia. Blood 1998; 91: 3601–3606. filgrastim after dose-intensive chemotherapy in patients with 7 Blay JY, Le Cesne A, Mermet C et al. A risk model for throm- advanced cancer (published erratum appears in Blood 1997; bocytopenia requiring platelet transfusion after cytotoxic 90: 2513). Blood 1997; 89: 3118–3128. chemotherapy. Blood 1998; 92: 405–410. 26 Basser RL, Rasko JE, Clarke K et al. Thrombopoietic effects 8 Sheridan WP, Begley CG, Juttner CA et al. Effect of periph- of pegylated recombinant human megakaryocyte growth and eral-blood progenitor cells mobilised by filgrastim (G-CSF) development factor (PEG-rHuMGDF) in patients with on platelet recovery after high-dose chemotherapy. Lancet advanced cancer. Lancet 1996; 348: 1279–1281. 1992; 339: 640–644. 27 Abraham R, Basser RL. Megakaryocyte growth and develop- 9 Nash R, Gooley T, Davis C, Appelbaum F. The problem of ment factor: a review of early clinical studies. Oncologist thrombocytopenia after hematopoietic stem cell transplan- 1997; 2: 311–318. tation. Stem Cells 1996; 14 (Suppl. 1): 261–273. 28 O’Malley CJ, Rasko JE, Basser RL et al. Administration of 10 Hassan H, Zander AR. Thrombocytopenia after high-dose pegylated recombinant human megakaryocyte growth and chemotherapy and autologous stem cell transplantation: an development factor to humans stimulates the production of unresolved problem and possible approaches to resolve it. J functional platelets that show no evidence of in vivo acti- Hematother 1996; 5: 407–414. vation. Blood 1996; 88: 3288–3298. 11 Bernstein S, Nademanee A, Vose J et al. A multicenter study 29 Fanucchi M, Glaspy J, Crawford J et al. Effects of polyethyl- of platelet recovery and utilization in patients after myeloabl- ene glycol-conjugated recombinant human megakaryocyte ative therapy and hematopoietic stem cell transplantation. growth and development factor on platelet counts after chemo- Blood 1998; 91: 3509–3517. therapy for lung cancer (see comments). New Engl J Med 12 Smith B, Jones R, Piantidosi S et al. Autologous bone marrow 1997; 336: 404–409. transplantation (ABMT) using 4hydroperoxycyclophospham- 30 Joncheere A. A distribution-free k-sample test against ordered ide (4HC) purging for acute myeloid leukemia beyond first alternatives. Biometrika 1954; 41: 133–145. remission: a ten year experience. Blood 1998; 92 (Suppl. 1): 31 Terpstra T. The asymptotic normality and consistency of Ken- 323A. dall’s test against trend when ties are present in one ranking. 13 Bensinger W, Appelbaum F, Rowley S et al. Factors that Indagationes Mathamaticae 1952; 14: 327–333. influence collection and engraftment of autologous peripheral- 32 Hollander M, Wolfe DA. Nonparametric Statistical Methods. blood stem cells. J Clin Oncol 1995; 13: 2547–2555. Wiley: New York. 1973. 14 Pecora AL, Preti RA, Gleim GW et al. CD34+CD33− cells 33 Lowe J, Peterson M, Ramsey D et al. A novel bioassay for influence days to engraftment and transfusion requirements in analysis of neutralizing anti-thrombopoietin antibodies. Blood autologous blood stem-cell recipients. J Clin Oncol 1998; 16: 1996; 88 (Suppl. 1): A1397. 2093–2104. 34 Neelis KJ, Hartong SC, Egeland T et al. The efficacy of sin- 15 Brown RA, Adkins D, Goodnough LT et al. Factors that gle-dose administration of thrombopoietin with coadminis- influence the collection and engraftment of allogeneic periph- tration of either granulocyte/macrophage or granulocyte col- eral-blood stem cells in patients with hematologic malig- ony-stimulating factor in myelosuppressed rhesus monkeys. nancies. J Clin Oncol 1997; 15: 3067–3074. Blood 1997; 90: 2565–2573.

Bone Marrow Transplantation rhTPO after auto-BMT SN Wolff et al 268 35 Beveridge R, Schuster M, Waller E et al. Randomized, dou- 39 Tepler I, Elias L, Smith JW II et al. A randomized placebo- ble-blind, placebo-controlled trial of pegylated recombinant controlled trial of recombinant human interleukin-11 in cancer human megakaryocyte growth and development factor (Peg- patients with severe thrombocytopenia due to chemotherapy. MGDF) in breast cancer patients (pts) following autologous Blood 1996; 87: 3607–3614. bone marrow transplantation (ABMT). Blood 1997; 90: 40 Vredenburgh JJ, Hussein A, Fisher D et al. A randomized trial (Suppl. 1): 580a. of recombinant human interleukin-11 following autologous 36 Shimazaki C, Inaba T, Uchiyama H et al. Serum thrombopoietin bone marrow transplantation with peripheral blood progenitor levels in patients undergoing autologous peripheral blood stem cell cell support in patients with breast cancer. Biol Blood Marrow transplantation. Bone Marrow Transplant 1997; 19: 771–775. Transplant 1998; 4: 134–141. 37 Porcelijn L, Folman CC, Bossers B et al. The diagnostic value 41 Vadhan-Raj S, Murray LJ, Bueso-Ramos C et al. Stimulation of thrombopoietin level measurements in thrombocytopenia. of megakaryocyte and platelet production by a single dose of Thromb Haemost 1998; 79: 1101–1105. recombinant human thrombopoietin in patients with cancer. 38 Cerutti A, Custodi P, Duranti M et al. Thrombopoietin levels Ann Intern Med 1997; 126: 673–681. in patients with primary and reactive thrombocytosis. Br J Haematol 1997; 99: 281–284.

Bone Marrow Transplantation