Granulocyte and Erythropoietic Stimulating Proteins After High-Dose Chemotherapy for Myeloma

REVIEW Granulocyte and erythropoietic stimulating proteins after high-dose chemotherapy for myeloma

TJ Littlewood and GP Collins

Department of Haematology, John Radcliffe Hospital, Oxford, UK

High-dose chemotherapy is an established treatment for of patients with myeloma6 but only a small number of patients with myeloma. In randomized trials it has been people are young and fit enough to undergo such a shown to prolong disease-free survival by around 1year procedure. In order to succeed, such high-dose therapy compared to patients receiving chemotherapy alone. strategies must be accompanied by effective use of Physically and psychologically high-dose therapy takes supportive measures to prevent and treat common compli- its toll on the patient who may be in hospital for around cations of therapy. To this end, colony stimulating factors 3 weeks and take some weeks or months to convalesce (CSFs) and erythropoietin (Epo) have been enjoying after discharge. Granulocyte colony stimulating factors increasing popularity in a number of different contexts and erythropoietic stimulating agents will speed neutrophil and their use to support high-dose procedures will be and red cell recovery, respectively, when used at an reviewed here. appropriate time after the high-dose chemotherapy. The clinical value of these laboratory findings is uncertain and the role of these agents after high-dose chemotherapy Colony stimulating factors remains a subject for debate. Bone Marrow Transplantation (2007) 40, 1147–1155; Granulocyte colony stimulating factor (G-CSF) was doi:10.1038/sj.bmt.1705845; published online 10 September 2007 identified as a substance which specifically induced the Keywords: erythropoietin; G-CSF; high-dose chemotherapy formation of neutrophil colonies when added to human or mouse bone marrow cells in culture.7 Granulocyte–macrophage stimulating factor (GM-CSF), as its name suggests, was found to increase the production of eosinophils, Introduction macrophages and dendritic cells in addition to granulocytes. G-CSF is a glycoprotein of molecular weight 19 kDa The treatment of myeloma has been transformed in recent and is produced by a number of cell types including years by a combination of novel therapies such as endothelial cells. In the mid 1980s, the gene encoding thalidomide and the increasingly widespread use of high- G-CSF was cloned and located on chromosome 17q which dose procedures such as autologous and, to a lesser extent, opened the door to the production of recombinant human allogeneic stem cell transplants. The latter trend is G-CSF (rhG-CSF). Subsequent structural analyses of the illustrated by an EBMT study showing that just a few protein itself have shown the existence of four antiparallel hundred haematopoietic stem cell transplants were per- alpha-helices which cluster together to form a helical formed per year for myeloma in Europe in 1990 whereas in bundle with the resulting tertiary structure conferring 2002 there were over 4000.1 Average life expectancy for specificity of binding to the G-CSF receptor.8 The action patients with myeloma is typically around 3–5 years and and specificity of G-CSF is mediated through its specific most patients will die from the disease despite many transmembrane receptor which is expressed on cells from changes to treatment during the last 20 years. High-dose the myeloblast to the mature neutrophil with the highest chemotherapy with peripheral blood stem cell support number of receptors on the mature neutrophil.9 G-CSF has has been shown to prolong life expectancy compared been implicated in a number of physiological processes, to standard chemotherapy in some2,3 but not all studies.4 including the basal production of granulocytes by the bone A recent study has also suggested that a tandem transplant marrow and the emergency response of the bone marrow to protocol may be superior to a single transplant for certain increase production of granulocytes in the setting of acute patients.5 Allogeneic transplant may cure around 20–30% infection. Indeed, mice which have low endogenous G-CSF display a marked neutropenia in combination with a failure to mobilize granulocytes during infections.10 These effects Correspondence: Dr TJ Littlewood, Department of Haematology, John are thought to be primarily due to maturation and Radcliffe Hospital, Headley Way, Headington, Oxford, OX3 9DU, UK. E-mail: [email protected] proliferation signals being delivered to myeloid precursors Received 6 February 2007; revised 30 May 2007; accepted 23 July 2007; in the bone marrow. In addition to its effects on the published online 10 September 2007 bone marrow granulocyte precursors, G-CSF also has G-CSF and erythropoietin after high-dose chemotherapy TJ Littlewood and GP Collins 1148 effects on peripheral mature neutrophils. These include study by Colby et al.25 (comparing day þ 1 with day þ 4) effects on degranulation,11 adhesion,12 phagocytosis13 and showing a significant benefit for early therapy, although delaying apoptosis.14 Of crucial importance in the setting of this study was retrospective. Not surprisingly, delayed stem cell transplantation, G-CSF has also been implicated therapy was associated with a significant cost saving and in the mobilization of immature haematopoietic precursors this finding really should be subjected to a more rigorous, from the bone marrow into the peripheral circulation large randomized study. In summary, the clear efficacy of although this topic will not be dealt with further in this rhG-CSF with respect to a number of different parameters review. has led the American Society of Oncology (ASCO) to consider primary prophylaxis with this agent during high- dose therapy with autologous stem cell support as the Efficacy of CSFs as infection prophylaxis during high-dose standard of care.26 procedures

A number of randomized, controlled clinical trials have CSFs as treatment for therapy-induced febrile neutropenia been performed on the use of rhG-CSF in primary prophylaxis during high-dose procedures for a number of An alternative to use of G-CSF therapy as primary haematological cancers and solid tumours, and a recent prophylaxis is to administer these agents alongside large meta-analysis has been performed.15 The majority of antibiotics to patients who have developed a febrile illness the studies show a reduction in documented infections, in whilst neutropenic. There are fewer studies assessing this days of parenteral antibiotic use and in time to discharge. approach and the data are mainly in the setting of solid However, no statistically significant reduction in either tumour patients receiving chemotherapy. Results from overall mortality or infection-related mortality has been some of the rather small trials addressing the use of G- demonstrated. For example, one of the more recent, larger CSF or GM-CSF in this situation were also conflicting with studies which was randomized and double-blinded com- some suggesting an improvement in time to recovery of pared lenograstim at a dose of 150 mg/m2 per day from day fever27,28 while others failed to show a benefit.29 Clark þ 1 after autologous PBSC infusion with placebo. Leno- et al.30 undertook a meta-analysis of randomized-con- grastim treatment was associated with a reduction in the trolled trials in an effort to provide clarity. A clear and number of patients experiencing at least one infectious expected reduction in duration of neutropenia was ob- episode from 86/94 patients to 66/98 patients.16 served as was the duration of hospitalization. Although no A significant reduction in the average number of days effect on overall survival was demonstrated, a marginally of intravenous antibiotics was observed (from 10 days to significant reduction in infection-related mortality was 8 days) along with a significant reduction in the time observed. Interestingly, it appeared on sub-group analysis to neutrophil recovery (defined as achieving an absolute that the groups most likely to benefit from this approach neutrophil count of 40.5 Â 109/l) from 15 to 11 days. Mean were those patients with haematological malignancies. The hospital stay was reduced from 17 to 15 days. The observed results of this and one other31 meta-analysis have prompted improvement in a number of clinical parameters has led ASCO to recommend adjunctive G-CSF therapy for some investigators to conclude that there may be an patients with febrile neutropenia and a clinical risk factor economic benefit of using rhG-CSF as primary prophy- for the development of complications, such as expected laxis. One of the early reports by Luce et al.17 concluded prolonged and profound neutropenia, old age, pneumonia, that the initial average cost of an autologous bone marrow hypotension and multi-organ dysfunction.26 transplant for lymphoid cancer fell by 21% in patients treated with rhGM-CSF largely due to a reduction in the length of hospital stay. Subsequent analyses of the cost- CSFs and graft-versus-host disease effectiveness of rhCSFs (both GM-CSF and G-CSF) in this setting sometimes agreed with this early report18,19 whereas One serious concern that has arisen with the use of CSFs in others have drawn an opposite conclusion.20,21 It might be the allogeneic stem cell transplant setting is the possibility argued that since the introduction of better tolerated but of a higher rate of graft-versus-host disease which is one of significantly more expensive antifungal therapy, the cost the major causes of morbidity and mortality following such benefit for rhG-CSF is likely to become more pronounced a procedure. Indeed, the ASCO guidelines specifically with time. However, one of the more recent cost analyses state that CSF use after autologous, but not allogeneic, concluded that there was a cost saving associated with the transplantation is the current standard of care.26 This placebo group.22 It is therefore currently premature to recommendation is largely based on the conclusions of a conclude that primary prophylaxis of CSFs is associated European study which looked at the outcome of 1789 with a significant financial saving.15 The cost of G-CSF patients with acute leukaemia receiving bone marrow may be reduced by optimization of the dosing schedule. transplantation and 434 patients receiving PBSC trans- To this end, several studies have compared the efficacy of plantation.32 In the bone marrow transplantation group, G-CSF when started later than day þ 1 after stem cell G-CSF therapy within the first 14 days after marrow infusion. Unfortunately, results are conflicting with studies infusion was associated with a statistically significant by Bence-Bruckler et al.23 (comparing day þ 1 with day increase in incidence of grades II–IV acute GvHD (from þ 7) and Bolwell et al.24 (comparing day 0 with day þ 3 39 to 50%), in incidence of chronic GvHD (relative risk with day þ 5) showing no difference in efficacy whereas a 1.29) and in transplant-related mortality (relative risk 1.79).

Bone Marrow Transplantation G-CSF and erythropoietin after high-dose chemotherapy TJ Littlewood and GP Collins 1149 There was no effect on the leukaemia relapse rate but there there is insufficient evidence to recommend one form of was a reduction in overall survival and leukaemia-free rhG-CSF over another. survival. Interestingly, no such effect was seen in the PBSC A more recent compound to have come onto the market group although the numbers were smaller. The observed is a pegylated form of filgrastim called pegfilgrastim which increase in the acute GvHD incidence has also been shown maintains biological activity yet has a substantially longer by other groups.33,34 However, the studies which demon- half-life compared with filgrastim due to decreased renal strate these results are all retrospective analyses which are excretion.46,47 A phase II trial studied 38 patients with susceptible to bias. Indeed, the recent meta-analysis by myeloma or lymphoma undergoing high-dose therapy with Dekker et al.15 which included only prospective, rando- autologous stem cell support who were all given a single mized-controlled trials did not find any significant associa- subcutaneous (sc) injection of 6 mg pegfilgrastim on day tion between CSF use and acute or chronic GvHD, þ 1 following stem cell infusion.48 Neutrophil engraftment treatment-related mortality or overall survival. The study was compared with historical controls and shown to be inclusion criteria means that the result obtained should be similar to that achieved with filgrastim yet significantly free of the potential confounding factors which are so shorter than in patients who did not receive CSF support. difficult to control for in retrospective analyses. Publication A subsequent prospective, randomized study comparing bias is an issue for meta-analyses but this tends to produce pegfilgrastim administered on day þ 1 after the stem cell a falsely positive result whereas in this case the result was infusion with filgrastim administered daily commencing on negative. Another earlier meta-analysis by Ho et al.35 came day þ 5, in myeloma patients receiving high-dose melpha- to a similar conclusion although data were included from lan and autologous stem cell transplants showed no prospective and retrospective studies. In short, results from statistical difference between time to neutrophil engraft- some retrospective and prospective trials are conflicting. ment and platelet engraftment.49 However, in the pegfil- Although the results from prospective trials are more grastim group there was a significantly greater white cell robust, the total number of patients included in these count on day þ 15 (6.0 Â 109/l compared with 2.7 Â 109/l) studies is much lower than that in the single retrospective together with a reduction in the incidence of febrile study by Ringden et al.32 In fact, the largest study of neutropenia from 100 to 61% and a reduction in the allogeneic patients included in the meta-analysis by Dekker duration of febrile neutropenia from 4 days to 1.5 days. et al. contained just over 100 patients.36 To clarify the issue The total number of patients randomized were only 37, so a therefore, a large multicentre, prospective, randomized trial larger study is clearly needed in order to confirm this is required. intriguing result. At present, it is too early to clearly define a role for pegfilgrastim in high-dose therapy protocols. GM-CSF is rarely used in high-dose therapy procedures. Which CSF to use A randomized study comparing the CD34 þ cell mobilizing efficacy of filgrastim, lenograstim and molgramostim (GM- An obvious question for the use of CSF support during CSF) showed no statistically significant differences between high-dose procedures is which agent to use. Early in vitro the three compounds.42 However, only one study has studies suggested that the glycosylated form of G-CSF assessed its use during subsequent high-dose therapy. (lenograstim) had more biological activity than filgras- Legros et al.50 performed a randomized, placebo-controlled tim.37,38 Studies have also been reported which tested the trial and showed that the use of GM-CSF had no effect on potency of CSF in healthy volunteers to mobilize CD34 þ the time to neutrophil engraftment or on the incidence of progenitor cells and they too suggested that lenograstim documented infections. In fact, the placebo group actually was more potent than filgrastim.39,40 However, no differ- had significantly fewer days of fever. There was a ence between the drugs has been identified in clinical suggestion that GM-CSF may result in more rapid platelet studies. Certainly when CSFs are used to mobilize stem engraftment and reduced severity of mucositis51,52 but this cells prior to autologous stem cell transplantation, the was not born out in subsequent clinical trials.42,50 The weight of evidence suggests there is no significant difference demonstrated lack of benefit over placebo means that in the potency of filgrastim, lenograstim and the GM-CSF current formulations of GM-CSF are unlikely to be used preparation molgramostim.41,42 Fewer studies have com- widely in high-dose therapy programmes. pared these agents as primary prophylaxis during autologous or allogeneic stem cell transplantation. Bonig et al.43 performed the first prospective randomized comparison of Long-term safety of CSFs filgrastim with lenograstim as primary prophylaxis during chemotherapy in paediatric patients. This trial has only There has been intense scrutiny over the safety of G-CSF small numbers in each group (only 11) and showed no therapy in large part due to its frequent use in healthy stem difference in a number of different laboratory and clinical cell donors as part of an allogeneic transplant. A report outcomes. A retrospective comparison of patients under- which generated concern was recently published by the going PBSC transplantation following high-dose therapy Research on Adverse Drug Events and Reports project actually showed a more rapid neutrophil recovery and which identified 2 out of 200 healthy donors who went on platelet recovery along with a reduced duration of to develop acute myeloid leukaemia some years after antibiotic therapy and a shorter hospital stay with having received G-CSF as part of stem cell mobilization filgrastim.44 A more recent retrospective analysis however for a sibling allogeneic transplant regimen.53 It is impos- showed no difference between the two agents.45 At present, sible to demonstrate a causative role for the G-CSF in these

Bone Marrow Transplantation G-CSF and erythropoietin after high-dose chemotherapy TJ Littlewood and GP Collins 1150 cases, particularly in view of the described increased risk of genes are modulated by STAT5 but one of particular acute leukaemia in first degree relatives.54 More reassur- importance is Bcl-x. The level of this molecule in erythroid ingly, the National Marrow Donor Programme in America progenitors is increased by the action of Epo and it is has not reported any cases of leukaemia in more than 4000 thought to protect cells from undergoing apoptosis.66 volunteer, unrelated stem cell donors to date.55 Of more Along with bone pain, renal impairment and an concern to patients receiving G-CSF as part of a stem cell increased incidence of infection, anaemia is a common transplant regimen are recent reports of case-control complication of myeloma. More specifically, anaemia is studies linking use of G-CSF with an increased risk of present in approximately 50% of patients at the time of developing subsequent myelodysplastic syndrome/ diagnosis and most of the remainder will develop anaemia AML.56,57 The relative risk ranged from 2 to 4 although at some point during the course of their illness.67 The the absolute risk remained low. These studies do highlight anaemia may have many causes but the most common the fact that such patients merit prolonged follow up. explanations include the cytokine driven anaemia of chronic disease (AOCD) plus the myelotoxic impact of treatment with chemo and/or radiotherapy.68 Cytokines Introduction to erythropoietin such as tumour necrosis factor-a, g-interferon and inter- leukin-1 contribute to the AOCD by shortening red cell Erythropoietin (Epo) is a heavily glycosylated glycoprotein survival, blunting erythropoiesis both directly and indir- hormone encoded by a gene which is located on chromo- ectly (by reducing Epo production) and by inhibiting the some 7q11–22.58 The resulting 165 amino acid glycoprotein availability of tissue iron.68 Hepcidin, a peptide produced in acts to promote erythroid proliferation, differentiation and the liver, increases in patients with AOCD and has been survival with a resulting increase in the mass of circulating shown to be a key player in reducing iron absorption by gut red cells.59 The major site of Epo production is the renal enterocytes and blocking iron release from tissue macro- peritubular cell although in adult life 10% is produced by phages.69 Recently, it was suggested that malignant plasma the liver. The liver plays a more important role in Epo cells expressing Fas ligand and TRAIL (TNF-related synthesis during fetal life, the switch to predominant renal apoptosis inducing ligand) was a major pathogenetic production occurring at about 30 weeks of gestation.60 mechanism for anaemia in patients with myeloma.70 RHuEpo is synthesized by Chinese Hamster Ovary cells Anaemic patients with myeloma may benefit from treat- and has the identical 165 amino acid sequence to ment with recombinant Epo which can improve the endogenous circulating human Epo.61 The molecule con- haemoglobin concentration, decrease transfusion need tains two disulphide bonds between residues 7 and 161 and and improve quality of life.71 Because of the success of 29 and 33, and the disruption of these bonds results in recombinant Epo in treating anaemia in myeloma patients instability of the molecule along with reduced biological undergoing standard chemotherapy,72,73 a number of activity. A total of 60% of the mass of RHuEpo is protein studies were designed to determine its effectiveness in whereas 40% is carbohydrate, and the glycosylation status patients receiving high-dose therapy with stem cell support. of the Epo molecule is also important for determining certain biological properties. The three N-linked sugars (at positions 24, 38 and 83) are necessary for persistence of the Erythropoietin after high-dose chemotherapy with molecule within the circulation. Removal of these sugars autologous marrow or peripheral blood stem cell support produces a biologically active Epo molecule but with a much reduced circulating half-life.62 There is in addition The majority of patients are anaemic on admission to one O-linked sugar (at position 126) which does not appear hospital for high-dose treatment as a consequence of their to be of functional importance. underlying disease and its treatment. A recent survey of the Two Epos, epoetin-a and epoetin-b are currently Oxford myeloma patients admitted for high-dose treatment available plus darbepoetin, a longer acting derivative. A showed that 76% were anaemic with a haemoglobin of number of erythropoietic agents are in development o12.0 g dlÀ1. Several relatively small trials of Epo treat- including epoetin delta and CERA (continuous erythro- ment after high-dose chemotherapy and autologous trans- poietin receptor activator). As with other glycoprotein plant have been performed. Until recently, all these studies hormones, Epo exists as a mixture of different isoforms reported no benefit in terms of haemoglobin concentrations which differ mainly in their glycosylation pattern.63 Epo and a reduction in transfusion need from Epo. Perhaps the binds to the dimeric erythropoietin receptor (EpoR) which best of these trials was the randomized, placebo-controlled is a member of the cytokine receptor superfamily. Receptor trial reported by Link et al.74 One-hundred and fourteen binding induces a conformational change which has the patients with haematological malignancies were rando- effect of bringing the cytoplasmic tails of the two subunits mized to rHuEpo or placebo by continuous IV infusion into closer proximity.64 This enables the EpoR-associated after high-dose chemotherapy and autologous transplant. tyrosine kinase Jak2 (a member of the Janus kinase family) Treatment was continued until transfusion independence or to phosphorylate tyrosine residues which reside in the until day þ 41. There was no advantage for the patients cytoplasmic domain of the partner subunit. Once phos- treated with rHuEpo in terms of haemoglobin concentra- phorylated, signal transducer and activator of transcription tions and time to transfusion independence. 5 (STAT5) can interact with the EpoR via Src-homology More encouraging results have been reported since then 2 domains, resulting in subsequent translocation to the by a number of groups who have investigated treatment nucleus and modulation of gene expression.65 A number of with rHuEpo at different time points both before and after

Bone Marrow Transplantation G-CSF and erythropoietin after high-dose chemotherapy TJ Littlewood and GP Collins 1151 high-dose chemotherapy. Martino et al.75 treated 22 independence was 19 days in the treatment group versus patients with myeloma with rHuEpo by daily sc injection 27 days in the placebo group. There was no impact on the starting 30 days before high-dose treatment. A total of 40 number of red cell units transfused between the two groups similar patients who did not receive rHuEpo were selected between day 0 and 20 but after this time the treatment as historical controls. Although there was no significant group required fewer transfusions. Other investigators have difference between the haemoglobin concentrations of the also reported benefit for treatment with rHuEpo after two groups at the time of transplant, the treatment group allogeneic transplant.78,79 Exploiting the same argument had a significantly better nadir haemoglobin (median 10.0 as in the autologous setting, Baron et al.80 showed that versus 7.6 g/dl) and fewer than 20% of the treatment group treatment with rHuEpo starting at day þ 30 or beyond, in required blood transfusions compared to more than half of patients transplanted with allogeneic peripheral blood stem the control patients. cells, resulted in transfusion independence within 2 weeks in Baron et al.76 have shown that after high-dose chemo- 12/13 patients versus 50% of patients in an historical therapy serum Epo concentrations are higher than expected control group. The mean Hb concentrations were signifi- for the degree of anaemia for 1–3 weeks. During this time cantly higher in the treatment compared to the control the marrow erythroid activity is suppressed from the groups from day þ 60 post-transplant. In this study, the chemotherapy and therefore it is not surprising that investigators also showed that serum Epo concentrations treatment with rHuEpo during the first 3–4 weeks after were normal or high until day þ 28 and then became high-dose treatment is not effective. They hypothesized that inappropriately low for at least the next 6 months. The starting rHuEpo 4 weeks after high-dose treatment, at a persistently lower Epo concentrations in allograft com- time when serum Epo levels were normal or sub-optimal pared to autograft patients may be in part attributable to for the degree of anaemia, but erythroid activity was renal damage induced by immunosuppressive agents such recovering, would be advantageous. They treated 41 as cyclosporine. patients (rHuEpo 500 U/kg per week by sc injection) with myeloma or lymphoma and compared them to an historical control group. In the treatment group, 87% achieved the Safety target haemoglobin of 13.0 g/dl within 12 weeks of starting rHuEpo compared with 14% in the controls. No patient in In the studies described above Epo was safe with no the treatment group required transfusion after rHuEpo was increased risk of hypertension or venous thrombosis. started compared to 8/45 patients in the control group. During the last few years, two major concerns have been Encouraged by these results, the same group has reported a raised about erythropoetic agents. Firstly, in patients with small randomized trial of rHuEpo starting on day þ 30 renal failure Casadevall et al.81 described the development versus no treatment in 20 patients (of whom 13 had of pure red cell aplasia caused by the development of anti- myeloma) undergoing high-dose chemotherapy with per- Epo antibodies. This rare complication seemed to be ipheral blood stem cell support.77 At day 30, the mean confined to patients with renal failure treated with one of serum Epo concentration was sub-optimal for the degree of the agents (epoetin-a). Changes in manufacturing practice anaemia in the treatment and control groups of patients. were undertaken and the complication has largely dis- All the patients in the treatment group responded with a appeared. greater than 2.0 g/dl rise in Hb versus a 28% response in the Secondly, there remains doubt about the possible adverse control group. impact of the erythropoetic agents on survival in patients A total of 90% of patients in the rHuEpo group achieved with cancerous illnesses. A number of studies in head and a haemoglobin 412.0 by day 100, whereas only 40% of neck cancer, breast cancer and non-small cell lung cancer patients achieved this target in the control group. This have suggested a survival disadvantage for patients treated small randomized study provides some more interesting with Epo compared to placebo.82 The subject remains data supporting the investigators’ hypothesis that delaying widely debated as many other studies show either a survival rHuEpo after high-dose chemotherapy until red cell advantage or no disadvantage between treatment and production is recovering, and endogenous Epo concentra- control groups.83 tions are sub-optimal results in speedier red cell recovery The trials in myeloma described above have not shown than in a control group. However, in this study, there was any concerns about the survival of the Epo compared to the no difference in transfusion need between the treatment control treatment arms. and control groups.

Conclusions Erythropoietin after allogeneic bone marrow or peripheral blood stem cell transplant Due to the large number of studies performed and more than one published meta-analysis, it is clear that G-CSF use Link et al.74 conducted a study in which 215 patients as primary prophylaxis during autologous PBSC trans- scheduled for allogeneic bone marrow transplant using plantation in myeloma and other tumours confers several myeloablative conditioning were randomized to rHuEpo beneﬁts such as reduced antibiotic usage and time to (150 IU kgÀ1 per day by continuous intravenous infusion discharge and this is reﬂected in the recommendations from day 0 to transfusion independence or a maximum issued by both ASCO and the British Committee for of 41 days) or placebo. The median time to red cell Standards in Haematology.26,84 It should be emphasized

Bone Marrow Transplantation G-CSF and erythropoietin after high-dose chemotherapy TJ Littlewood and GP Collins 1152 however that a survival advantage has not been shown and Spanish cooperative group PETHEMA. Blood 2005; 106: so some centres do not routinely administer such agents. 3755–3759. The evidence for a worse outcome for patients undergoing 5 Attal M, Harousseau JL, Facon T, Guilhot F, Doyen C, allogeneic bone marrow transplantation treated with Fuzibet JG et al. Single versus double autologous stem-cell G-CSF is, in the opinion of the authors, far from conclusive transplantation for multiple myeloma. N Engl J Med 2003; and further studies are urgently needed to address this 349: 2495–2502. 6 Crawley C, Iacobelli S, Bjorkstrand B, Apperley JF, Nieder- issue. Pegfilgrastim is an interesting new agent and may wieser D, Gahrton G. Reduced intensity conditioning for well come to be used increasingly as primary infection myeloma: lower non-relapse mortality but higher relapse rates prophylaxis during high-dose procedures although compared to myeloablative conditioning. Blood 2007; 109: larger studies are required to establish its role. Certainly 3588–3594. patients are likely to prefer its single dose to the sometimes 7 Nicola NA, Metcalf D, Matsumoto M, Johnson GR. arduous daily dosing regimes of the conventional G-CSF Purification of a factor inducing differentiation in murine formulations. myelomonocytic leukemia cells. Identification as granulocyte The problem in interpreting all of the Epo studies is that colony-stimulating factor. J Biol Chem 1983; 258: 9017–9023. there are too few patients included. If, however, they are 8 Welte K, Gabrilove J, Bronchud MH, Platzer E, Morstyn G. Blood taken at face value, the suggestion is that using recombi- Filgrastim (r-metHuG-CSF): the first 10 years. 1996; 88: 1907–1929. nant Epo before and/or starting around 30 days after high- 9 Avalos BR. Molecular analysis of the granulocyte colony- dose chemotherapy with stem cell support will significantly stimulating factor receptor. Blood 1996; 88: 761–777. increase the haemoglobin concentration and decrease 10 Lieschke GJ, Grail D, Hodgson G, Metcalf D, Stanley E, transfusion need. There are no data on quality of life Cheers C et al. Mice lacking granulocyte colony-stimulating and no information about whether anaemia correction factor have chronic neutropenia, granulocyte and macrophage could have any impact on life expectancy or the duration progenitor cell deficiency, and impaired neutrophil mobiliza- of response. Is any of this important when you remember tion. Blood 1994; 84: 1737–1746. that high-dose treatment is just one aspect in the total 11 Xu S, Hoglund M, Venge P. The effect of granulocyte colony- care package for people with myeloma? Well, in 2005, stimulating factor (G-CSF) on the degranulation of secondary in vivo nearly 6000 people in Europe with myeloma underwent granule proteins from human neutrophils may be 85 indirect. Br J Haematol 1996; 93: 558–568. high-dose treatment with autologous stem cell support. If 12 Chakraborty A, Hentzen ER, Seo SM, Smith CW. treatment with Epo decreases transfusion requirements by Granulocyte colony-stimulating factor promotes adhesion of just one unit per patient then this would be an important neutrophils. Am J Physiol Cell Physiol 2003; 284: C103–C110. reduction in the total amount of red cell units across the 13 Villalta F, Kierszenbaum F. Effects of human colony- continent. The real reduction may be greater than this. stimulating factor on the uptake and destruction of a Whether such a reduction in use would be deemed cost- pathogenic parasite (Trypanosoma cruzi) by human neutro- effective will depend on the health economy of each phils. J Immunol 1986; 137: 1703–1707. individual country; but if the availability of red cells for 14 Begley CG, Lopez AF, Nicola NA, Warren DJ, Vadas MA, transfusion declines further any strategy to reduce use will Sanderson CJ et al. Purified colony-stimulating factors in be welcomed. The reported studies seem to indicate that enhance the survival of human neutrophils and eosinophils vitro: a rapid and sensitive microassay for colony-stimulating the timing of Epo use is probably of great importance in factors. Blood 1986; 68: 162–166. obtaining the maximum benefit; but much larger trials are 15 Dekker A, Bulley S, Beyene J, Dupuis LL, Doyle JJ, Sung L. needed before recombinant Epo can be unequivocally Meta-analysis of randomized controlled trials of prophylactic recommended in this field. granulocyte colony-stimulating factor and granulocyte- macrophage colony-stimulating factor after autologous and allogeneic stem cell transplantation. J Clin Oncol 2006; 24: 5207–5215. References 16 Schmitz N, Ljungman P, Cordonnier C, Kempf C, Linkesch W, Alegre A et al. Lenograstim after autologous peripheral 1 Gratwohl A, Baldomero H, Passweg J, Frassoni F, Nieder- blood progenitor cell transplantation: results of a double- wieser D, Schmitz N et al. Hematopoietic stem cell blind, randomized trial. Bone Marrow Transplant 2004; 34: transplantation for hematological malignancies in Europe. 955–962. Leukemia 2003; 17: 941–959. 17 Luce BR, Singer JW, Weschler JM, Buckner CD, 2 Attal M, Harousseau JL, Stoppa AM, Sotto JJ, Fuzibet JG, Sheingold SH, Shannon-Dorcy K et al. Recombinant Rossi JF et al. A prospective, randomized trial of autologous human granulocyte-macrophage colony-stimulating factor bone marrow transplantation and chemotherapy in multiple after autologous bone marrow transplantation for lymphoid myeloma. Intergroupe Francais du Myelome. N Engl J Med cancer: an economic analysis of a randomised, double- 1996; 335: 91–97. blind, placebo-controlled trial. Pharmacoeconomics 1994; 6: 3 Child JA, Morgan GJ, Davies FE, Owen RG, Bell SE, 42–48. Hawkins K et al. High-dose chemotherapy with hematopoietic 18 Lee SM, Radford JA, Dobson L, Huq T, Ryder WD, stem-cell rescue for multiple myeloma. N Engl J Med 2003; Pettengell R et al. Recombinant human granulocyte colony- 348: 1875–1883. stimulating factor (filgrastim) following high-dose chemother- 4 Blade J, Rosinol L, Sureda A, Ribera JM, Diaz-Mediavilla J, apy and peripheral blood progenitor cell rescue in high-grade Garcia-Larana J et al. High-dose therapy intensification non-Hodgkin’s lymphoma: clinical benefits at no extra cost. Br compared with continued standard chemotherapy in multiple J Cancer 1998; 77: 1294–1299. myeloma patients responding to the initial chemotherapy: 19 Brice P, Godin S, Libert O, Marolleau JP, Makki J, Extra JM long-term results from a prospective randomized trial from the et al. Effect of lenograstim on the cost of autologous bone

Bone Marrow Transplantation G-CSF and erythropoietin after high-dose chemotherapy TJ Littlewood and GP Collins 1153 marrow transplantation. A preliminary communication. Pharmaco- 33 Remberger M, Naseh N, Aschan J, Barkholt L, LeBlanc K, economics 1995; 7: 238–241. Svennberg P et al. G-CSF given after haematopoietic stem cell 20 Ojeda E, Garcia-Bustos J, Aguado M, Arrieta R, Quevedo E, transplantation using HLA-identical sibling donors is asso- Yuste VJ et al. A prospective randomized trial of granulocyte ciated to a higher incidence of acute GVHD II–IV. Bone colony-stimulating factor therapy after autologous blood stem Marrow Transplant 2003; 32: 217–223. cell transplantation in adults. Bone Marrow Transplant 1999; 34 Eapen M, Horowitz MM, Klein JP, Champlin RE, Loberiza Jr 24: 601–607. FR, Ringden O et al. Higher mortality after allogeneic 21 Przepiorka D, Smith TL, Folloder J, Anderlini P, Chan KW, peripheral-blood transplantation compared with bone marrow Korbling M et al. Controlled trial of filgrastim for acceleration in children and adolescents: the Histocompatibility and of neutrophil recovery after allogeneic blood stem cell Alternate Stem Cell Source Working Committee of the transplantation from human leukocyte antigen-matched International Bone Marrow Transplant Registry. J Clin Oncol related donors. Blood 2001; 97: 3405–3410. 2004; 22: 4872–4880. 22 Gonzalez-Vicent M, Madero L, Sevilla J, Ramirez M, Diaz 35 Ho VT, Mirza NQ, Junco Dd D, Okamura T, Przepiorka D. MA. A prospective randomized study of clinical and economic The effect of hematopoietic growth factors on the risk of graft- consequences of using G-CSF following autologous peripheral vs-host disease after allogeneic hematopoietic stem cell blood progenitor cell (PBPC) transplantation in children. Bone transplantation: a meta-analysis. Bone Marrow Transplant Marrow Transplant 2004; 34: 1077–1081. 2003; 32: 771–775. 23 Bence-Bruckler I, Bredeson C, Atkins H, McDiarmid S, 36 Nemunaitis J, Rosenfeld CS, Ash R, Freedman MH, Deeg HJ, Hamelin L, Hopkins H et al. A randomized trial of Appelbaum F et al. Phase III randomized, double-blind granulocyte colony-stimulating factor (Neupogen) starting placebo-controlled trial of rhGM-CSF following allogeneic day 1 vs day 7 post-autologous stem cell transplantation. bone marrow transplantation. Bone Marrow Transplant 1995; Bone Marrow Transplant 1998; 22: 965–969. 15: 949–954. 24 Bolwell BJ, Pohlman B, Andresen S, Kalaycio M, Goormastic 37 Nissen C. Glycosylation of recombinant human granulocyte M, Wise K et al. Delayed G-CSF after autologous progenitor colony stimulating factor: implications for stability and cell transplantation: a prospective randomized trial. Bone potency. Eur J Cancer 1994; 30A (Suppl 3): S12–S14. Marrow Transplant 1998; 21: 369–373. 38 Pedrazzoli P, Gibelli N, Pavesi L, Preti P, Piolini M, Bertolini 25 Colby C, McAfee SL, Finkelstein DM, Spitzer TR. Early vs F et al. Effects of glycosylated and non-glycosylated G-CSFs, delayed administration of G-CSF following autologous alone and in combination with other cytokines, on the growth peripheral blood stem cell transplantation. Bone Marrow of human progenitor cells. Anticancer Res 1996; 16: 1781–1785. Transplant 1998; 21: 1005–1010. 39 Fischer JC, Frick M, Wassmuth R, Platz A, Punzel M, Wernet 26 Smith TJ, Khatcheressian J, Lyman GH, Ozer H, Armitage P. Superior mobilisation of haematopoietic progenitor cells JO, Balducci L et al. 2006 update of recommendations for the with glycosylated G-CSF in male but not female unrelated use of white blood cell growth factors: an evidence-based stem cell donors. Br J Haematol 2005; 130: 740–746. clinical practice guideline. J Clin Oncol 2006; 24: 3187–3205. 40 Hoglund M, Smedmyr B, Bengtsson M, Totterman TH, 27 Ravaud A, Chevreau C, Cany L, Houyau P, Dohollou N, Cour-Chabernaud V, Yver A et al. Mobilization of CD34+ Roche H et al. Granulocyte-macrophage colony-stimulating cells by glycosylated and nonglycosylated G-CSF in healthy factor in patients with neutropenic fever is potent after low- volunteers—a comparative study. Eur J Haematol 1997; 59: risk but not after high-risk neutropenic chemotherapy regi- 177–183. mens: results of a randomized phase III trial. J Clin Oncol 41 Lefrere F, Bernard M, Audat F, Cavazzana-Calvo M, 1998; 16: 2930–2936. Belanger C, Hermine O et al. Comparison of lenograstim vs 28 Mayordomo JI, Rivera F, Diaz-Puente MT, Lianes P, filgrastim administration following chemotherapy for periph- Colomer R, Lopez-Brea M et al. Improving treatment eral blood stem cell (PBSC) collection: a retrospective study of of chemotherapy-induced neutropenic fever by administration 126 patients. Leuk Lymphoma 1999; 35: 501–505. of colony-stimulating factors. J Natl Cancer Inst 1995; 87: 42 Kopf B, De Giorgi U, Vertogen B, Monti G, Molinari A, Turci 803–808. D et al. A randomized study comparing filgrastim versus 29 Vellenga E, Uyl-de Groot CA, de Wit R, Keizer HJ, lenograstim versus molgramostim plus chemotherapy for Lowenberg B, ten Haaft MA et al. Randomized placebo- peripheral blood progenitor cell mobilization. Bone Marrow controlled trial of granulocyte-macrophage colony-stimulating Transplant 2006; 38: 407–412. factor in patients with chemotherapy-related febrile neutro- 43 Bonig H, Silbermann S, Weller S, Kirschke R, Korholz D, penia. J Clin Oncol 1996; 14: 619–627. Janssen G et al. Glycosylated vs non-glycosylated granulocyte 30 Clark OA, Lyman GH, Castro AA, Clark LG, Djulbegovic B. colony-stimulating factor (G-CSF)—results of a prospective Colony-stimulating factors for chemotherapy-induced febrile randomised monocentre study. Bone Marrow Transplant 2001; neutropenia: a meta-analysis of randomized controlled trials. 28: 259–264. J Clin Oncol 2005; 23: 4198–4214. 44 Kim IH, Park SK, Suh OK, Oh JM. Comparison of 31 Berghmans T, Paesmans M, Lafitte JJ, Mascaux C, lenograstim and filgrastim on haematological effects after Meert AP, Jacquy C et al. Therapeutic use of granulocyte autologous peripheral blood stem cell transplantation and granulocyte-macrophage colony-stimulating factors in with high-dose chemotherapy. Curr Med Res Opin 2003; 19: febrile neutropenic cancer patients. A systematic review of 753–759. the literature with meta-analysis. Support Care Cancer 2002; 45 Huttmann A, Schirsafi K, Seeber S, Bojko P. Comparison of 10: 181–188. lenograstim and filgrastim: effects on blood cell recovery after 32 Ringden O, Labopin M, Gorin NC, Le Blanc K, Rocha V, high-dose chemotherapy and autologous peripheral blood Gluckman E et al. Treatment with granulocyte colony- stem cell transplantation. J Cancer Res Clin Oncol 2005; 131: stimulating factor after allogeneic bone marrow transplant- 152–156. ation for acute leukemia increases the risk of graft-versus-host 46 Molineux G, Kinstler O, Briddell B, Hartley C, McElroy P, disease and death: a study from the Acute Leukemia Working Kerzic P et al. A new form of filgrastim with sustained Party of the European Group for Blood and Marrow duration in vivo and enhanced ability to mobilize PBPC in both Transplantation. J Clin Oncol 2004; 22: 416–423. mice and humans. Exp Hematol 1999; 27: 1724–1734.

Bone Marrow Transplantation G-CSF and erythropoietin after high-dose chemotherapy TJ Littlewood and GP Collins 1154 47 Molineux G. The design and development of pegfilgrastim 64 Remy I, Wilson IA, Michnick SW. Erythropoietin receptor (PEG-rmetHuG-CSF, Neulasta). Curr Pharm Des 2004; 10: activation by a ligand-induced conformation change. Science 1235–1244. 1999; 283: 990–993. 48 Jagasia MH, Greer JP, Morgan DS, Mineishi S, Kassim AA, 65 Cheung JY, Miller BA. Molecular mechanisms of erythro- Ruffner KL et al. Pegfilgrastim after high-dose chemo- poietin signaling. Nephron 2001; 87: 215–222. therapy and autologous peripheral blood stem cell trans- 66 Silva M, Benito A, Sanz C, Prosper F, Ekhterae D, Nunez G plant: phase II study. Bone Marrow Transplant 2005; 35: et al. Erythropoietin can induce the expression of bcl-x(L) 1165–1169. through Stat5 in erythropoietin-dependent progenitor cell 49 Martino M, Pratico G, Messina G, Irrera G, Massara E, lines. J Biol Chem 1999; 274: 22165–22169. Messina G et al. Pegfilgrastim compared with filgrastim after 67 Birgegard G, Gascon P, Ludwig H. Evaluation of anaemia in high-dose melphalan and autologous hematopoietic peripheral patients with multiple myeloma and lymphoma: findings of the blood stem cell transplantation in multiple myeloma patients. European Cancer Anaemia Survey. Eur J Haematol 2006; 77: Eur J Haematol 2006; 77: 410–415. 378–386. 50 Legros M, Fleury J, Bay JO, Choufi B, Basile M, Condat P 68 Ludwig H, Pohl G, Osterborg A. Anemia in multiple et al. rhGM-CSF vs placebo following rhGM-CSF-mobilized myeloma. Clin Adv Hematol Oncol 2004; 2: 233–241. PBPC transplantation: a phase III double-blind randomized 69 Ganz T. Hepcidin and its role in regulating systemic iron trial. Bone Marrow Transplant 1997; 19: 209–213. metabolism. Hematology Am Soc Hematol Educ Program 51 Gianni AM, Siena S, Bregni M, Tarella C, Stern AC, Pileri A 2006, 29–35. et al. Granulocyte-macrophage colony-stimulating factor to 70 Silvestris F, Cafforio P, Tucci M, Dammacco F. Negative harvest circulating haemopoietic stem cells for autotransplan- regulation of erythroblast maturation by Fas-L(+)/ tation. Lancet 1989; 2: 580–585. TRAIL(+) highly malignant plasma cells: a major patho- 52 Bregni M, Siena S, Di Nicola M, Dodero A, Peccatori F, genetic mechanism of anemia in multiple myeloma. Blood Ravagnani F et al. Comparative effects of granulocyte- 2002; 99: 1305–1313. macrophage colony-stimulating factor and granulocyte col- 71 Ludwig H, Rai K, Blade J, Dammacco F, Degos L, Itri L et al. ony-stimulating factor after high-dose cyclophosphamide Management of disease-related anemia in patients with multi- cancer therapy. J Clin Oncol 1996; 14: 628–635. ple myeloma or chronic lymphocytic leukemia: epoetin 53 Bennett CL, Evens AM, Andritsos LA, Balasubramanian L, treatment recommendations. Hematol J 2002; 3: 121–130. Mai M, Fisher MJ et al. Haematological malignancies 72 Cazzola M, Beguin Y, Kloczko J, Spicka I, Coiffier B. Once- developing in previously healthy individuals who received weekly epoetin beta is highly effective in treating anaemic patients haematopoietic growth factors: report from the Research on with lymphoproliferative malignancy and defective endogenous Adverse Drug Events and Reports (RADAR) project. Br erythropoietin production. Br J Haematol 2003; 122: 386–393. J Haematol 2006; 135: 642–650. 73 Dammacco F, Castoldi G, Rodjer S. Efficacy of epoetin alfa in 54 Rauscher GH, Sandler DP, Poole C, Pankow J, Mitchell B, the treatment of anaemia of multiple myeloma. Br J Haematol Bloomfield CD et al. Family history of cancer and incidence 2001; 113: 172–179. of acute leukemia in adults. Am J Epidemiol 2002; 156: 74 Link H, Boogaerts MA, Fauser AA, Slavin S, Reiffers J, Gorin 517–526. NC et al. A controlled trial of recombinant human erythropoietin 55 Confer DL, Miller JP. Long-term safety of filgrastim (rhG- after bone marrow transplantation. Blood 1994; 84: 3327–3335. CSF) administration. Br J Haematol 2007; 137: 77–78; author 75 Martino M, Oliva E, Console G, Stelitano C, Fujo M, Messina reply 79–80. G et al. Administration of recombinant human erythropoietin 56 Le Deley MC, Suzan F, Cutuli B, Delaloge S, Shamsaldin A, alpha before autologous stem cell transplantation reduces Linassier C et al. Anthracyclines, mitoxantrone, radiotherapy, transfusion requirement in multiple myeloma patients. Support and granulocyte colony-stimulating factor: risk factors for Care Cancer 2005; 13: 182–187. leukemia and myelodysplastic syndrome after breast cancer. 76 Baron F, Frere P, Fillet G, Beguin Y. Recombinant human J Clin Oncol 2007; 25: 292–300. erythropoietin therapy is very effective after an autologous 57 Hershman D, Neugut AI, Jacobson JS, Wang J, Tsai WY, peripheral blood stem cell transplant when started soon after McBride R et al. Acute myeloid leukemia or myelodysplastic engraftment. Clin Cancer Res 2003; 9: 5566–5572. syndrome following use of granulocyte colony-stimulating 77 Vanstraelen G, Baron F, Frere P, Hafraoui K, Fillet G, Beguin factors during breast cancer adjuvant chemotherapy. J Natl Y. Efficacy of recombinant human erythropoietin therapy Cancer Inst 2007; 99: 196–205. started one month after autologous peripheral blood stem cell 58 Law ML, Cai GY, Lin FK, Wei Q, Huang SZ, Hartz JH et al. transplantation. Haematologica 2005; 90: 1269–1270. Chromosomal assignment of the human erythropoietin gene 78 Biggs JC, Atkinson KA, Booker V, Concannon A, Dart GW, and its DNA polymorphism. Proc Natl Acad Sci USA 1986; Dodds A et al. Prospective randomised double-blind trial of 83: 6920–6924. the in vivo use of recombinant human erythropoietin in bone 59 Lacombe C, Mayeux P. Biology of erythropoietin. Haemato- marrow transplantation from HLA-identical sibling donors. logica 1998; 83: 724–732. The Australian Bone Marrow Transplant Study Group. Bone 60 Kendall RG. Erythropoietin. Clin Lab Haematol 2001; 23: Marrow Transplant 1995; 15: 129–134. 71–80. 79 Ivanov V, Faucher C, Mohty M, Bilger K, Ladaique P, Sainty 61 Inoue N, Takeuchi M, Ohashi H, Suzuki T. The production of D et al. Early administration of recombinant erythropoietin recombinant human erythropoietin. Biotechnol Annu Rev 1995; improves hemoglobin recovery after reduced intensity condi- 1: 297–313. tioned allogeneic stem cell transplantation. Bone Marrow 62 Romanowski RR, Sytkowski AJ. The molecular structure of Transplant 2005; 36: 901–906. human erythropoietin. Hematol Oncol Clin North Am 1994; 8: 80 Baron F, Frere P, Beguin Y. Once weekly recombinant human 885–894. erythropoietin therapy is very efficient after allogeneic 63 Storring PL, Tiplady RJ, Gaines Das RE, Stenning BE, peripheral blood stem cell transplantation when started soon Lamikanra A, Rafferty B et al. Epoetin alfa and beta differ in after engraftment. Haematologica 2003; 88: 718–720. their erythropoietin isoform compositions and biological 81 Casadevall N, Nataf J, Viron B, Kolta A, Kiladjian JJ, properties. Br J Haematol 1998; 100: 79–89. Martin-Dupont P et al. Pure red cell aplasia and

Bone Marrow Transplantation G-CSF and erythropoietin after high-dose chemotherapy TJ Littlewood and GP Collins 1155 anti-erythropoietin antibodies in patients treated with recom- 84 Pagliuca A, Carrington PA, Pettengell R, Tule S, Keidan J. binant erythropoietin. N Eng J Med 2002; 346: 469–475. Guidelines on the use of colony-stimulating factors in 82 Crawford J. Erythropoietin: high proﬁle, high scrutiny. J Clin haematological malignancies. Br J Haematol 2003; 123: 22–33. Oncol 2007; 25: 1021–1023. 85 Gratwohl A, Baldomero H, Frauendorfer K, Urbano-Ispizua 83 Bohlius J, Wilson J, Seidenﬁeld J, Piper M, Schwarzer G, A, Niederwieser D. Results of the EBMT activity survey 2005 Sandercock J et al. Recombinant human erythropoietins and on haematopoietic stem cell transplantation: focus on increas- cancer patients: updated meta-analysis of 57 studies including ing use of unrelated donors. Bone Marrow Transplant 2007; 39: 9353 patients. J Natl Cancer Inst 2006; 98: 708–714. 71–87.

Bone Marrow Transplantation