This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
High dose chemotherapy followed by autologous hematopoietic stem cell transplantation (HSCT) is the mainstay treatment for multiple myeloma and has been shown to improve patient survival. During the treatment, collection of sufficient HSC is important for successful engraftment and hematopoietic recovery. The optimal strategy for stem cell mobilization has yet been determined and regimens incorporating chemotherapy and granulocyte-colony stimulating factor (G-CSF) or G-CSF alone are widely used. More recently, novel mobilization agent, namely plerixafor that antagonizes the CXCR4 (CXC chemokine receptor 4) and SDF1 (stromal-cell-derived factor-1Î±) interaction, has been approved by the FDA as salvage agent after G-CSF mobilization failure. In this retrospective study, the mobilization and HSCT outcomes of 28 myeloma patients who underwent different mobilization regimens were compared: cyclophosphamide (CY) plus G-CSF (CY group), G-CSF alone (G-CSF group) and G-CSF + plerixafor (P group). Mobilization in the CY group resulted in significantly higher HSC yield (9.46Â±1.33 x 106/kg) and few apheresis sessions (1.17Â±0.09) compared with those in the G-CSF (2.68Â±0.28 x 106/kg, p<0.001; 2Â±0.00 sessions, p<0.001) and P groups (2.57Â±0.24 x 106/kg, p=0.001; 2Â±0.32 sessions, p=0.002). However, the CY group resulted in neutropenia and thrombocytopenia at nadir (0.15Â±0.05 x 109/L and 76.4Â±9.23 x 109/L) and a need for hospital stay for 13Â±0.26 days during mobilization. These were not present in the G-CSF and P groups. Despite the difference in HSC yield, there was no difference in absolute neutrophil (CY: 10Â±0.19 vs G-CSF: 10.8Â±0.2 vs P: 10, p=0.062) and platelet engraftment (CY 15.1Â±1.41 vs G-CSF 18.8Â±4.57 vs P 14.5Â±0.65, p=0.319) as well as hospital stay (CY: 22.72Â±1.71 vs G-CSF: 20.2Â±1.77 vs P: 24.5Â±1.5, p=0.209) between the three groups. The results suggested that G-CSF alone could be used for mobilization in patients with multiple myeloma if there could be back up by plerixafor to achieve adequate HSC harvest. The risk and morbidity associated with CY mobilization would have to be balanced with the drug cost of plerixafor. The long-term difference in disease control in these three groups needs to be determined.
High dose chemotherapy and autologous hematopoietic stem cell transplantation (HSCT) is the mainstay of treatment for multiple myeloma (MM) and relapsed cases of Non-Hodgkins' Lymphoma (NHL) and Hodgkin's lymphoma (HL) (Pusic et al. 2008). Direct harvest of bone marrow (BM) used to be the major source of HSC for transplantation as it is the site where HSC normally reside. In the last two decades, peripheral blood stem cell (PBSC) is emerging as an important stem cell source in autologous transplantation and it avoids the risk of general anaesthesia and the trauma associated with bone marrow aspiration (Leung & Kwong, 2010). Furthermore, PBSCT is associated with a more rapid hematologic reconstitution and mobilized progenitor cells are less contaminated with tumour cells (Alegre et al. 1997). In fact, autologous PBSCT has now become the mainstay treatment of multiple myeloma, leading to better patient survival and minimal toxicity. In the following sections, an overview of the concept of stem cells, the various mobilization regimens, myeloma biology and the predicting factors of PBSC yield will be described.
1.2 Concept of hematopoietic stem cells and progenitors
In human adult, blood formation (hematopoiesis) occurs in the bone marrow and it arises from a small population of cells, known as the hematopioetic stem cells (HSC) that are replicatively quiescent but are capable of continuous self-renewal and differentiation into progenitors and functional blood cells of different lineages. Most important, HSC are capable of restoring hematopoiesis when transplanted into hosts (patients) who have received high dose chemotherapy and irradiation that would have caused fatal bone marrow aplasia without the transplant. The characteristics of human HSC have not been fully defined. Early studies in primates demonstrated that HSC expresses CD34+ (Berenson et al. 1988). However, CD34+ cells are heterogenous and the true HSC may be enriched in a more immunophenotypically restricted population i.e. CD34+CD38-Lin- based on xenogeneic transplantation models. For most clinical transplantation protocols, CD34+ remained to the standard phenotypical markers for the enumeration of HSC. Unlike HSC, the hematopoietic progenitors are defined by their proliferative potential (clonogenic potential) in semi-solid culture medium in which individual progenitors (colony forming unit, CFU) form progenies that would be clustered as a colony. Progenitors that are restricted to myeloid differentiation will form colonies of granulocytes (G) and macrophage (M) are known as CFU-GM. (fig. 1).
Fig. 1 Human Haemopoiesis
1.3 Mobilization regimen
1.3.1 Chemotherapy and HGFs
The concept of PBSC mobilization arose from the phenomenal observations by McCredie et al. (1971) and Richman et al. (1976) that high dose chemotherapy significantly increased circulating stem cells at a time that coincides with the early neutrophil recovery from the nadir. Since then, the use of chemotherapeutic agents, particularly cyclophosphamide together with hematopoietic growth factors (HGF), has been widely used in mobilization protocol. The yield of PBSC could be further enhanced by increasing the dose of chemotherapy (Lie et al. 1996). Using chemotherapy as a mobilization agent has a theoretical benefit of in-vivo tumor purging prior to PBSC collection and resulted in higher cell yield and few apheresis sessions (Gertz et al. 2009). However, it is associated with neutropenia and thrombocytopenia, resulting in morbidity and occasional mortality. Generally, PBSC can be collected when the circulating CD34+ cells exceed 20 x 106/L as neutrophil counts begin to rise from nadir. However, the optimal time of PBSC mobilization could be variable because recovery from cytopenic phase after chemotherapy was unpredictable (To et al. 1997).
1.3.2 Hemopoietic growth factors (HGFs)
Another mobilization regime is by using hemopoietic growth factors (HGFs) alone. Granulocyte-colony stimulating factor (G-CSF, Filgrastim) and granulocyte-macrophages colony stimulating factor (GM-CSF, Sargarmostim) are two commonly use mobilizing agents. Socinski et al. (1988) and Duhrsen et al. (1988) showed that circulating CFU-GM increased remarkably by administration of GM-CSF and G-CSF. This formed the basis of using haemopoietic growth factors (HGFs) alone in stem cell mobilization.
Advantage of using HGFs avoid cytotoxic side effect caused by chemotherapy. Administraion of HGFs is also applicabe to mobilizing normal healthy donors in allogeneic HSCT. A 40- to 80-fold increase of circulating CFU-GM over steady-state level after 4-5 days of HGFs administration had been observed. In a comparative studies of mobilization kinetics between G-CSF and GM-CSF in same dose (5Âµg/kg/day), the yield of CD34+ cells in GM-CSF arm was 3-folded lower than G-CSF and more side effects had been observed in GM-CSF (Bensinger et al. 2009; Fischmeister et al. 1999). Hence, GM-CSF is rarely used nowadays. However, for patients who failed mobilization by G-CSF, a combination of G-CSF with GM-CSF might be efficacious for re-mobilization (Cottler-Fox et al. 2003).
However, the optimal dose of G-CSF and GM-CSF have not been deteremined. The dose of G-CSF ranges from 10Î¼g/kg/day to high dose as high as 32Î¼g/kg/day. Normally, G-CSF was administrated at 10Î¼g/kg/day for 4 days before the start of apheresis and continued until the end of apheresis session (Bensinger et al. 2009). Kroger et al.(2000) suggested that a significant improvement in CD34+ cell yield if G-CSF was administrated subcutaneously twice daily in an interval of 12 hours at 5Î¼g/kg. For those patients who failed primary mobilization, adequate number of stem cells might be collected in re-mobilization by administrated at high dose G-CSF (32Î¼g/kg/day) (Gazitt et al, 1999).
Plerixafor (MozobilÂ®; Genzyme Corp., Cambridge, MA, USA) was licensed in 2008 by the United States Food and Drug Administration for use with G-CSF to mobilize HSC for autologous transplantation in patients with NHL and MM. For patients who failed mobilization by conventional regimens, Plerixafor showed a significant improvement in stem cell yield. Plerixafor is a novel agent that binds specifically to CXCR4 receptor in bone marrow stroma and interferes with the formation of SDF-1Î±/CXCR4 complex between BM stromal cells and HSC, leading to the release of the latter into the blood stream. Phase III clinical trial (DiPersio et al. 2009) showed that 71.6% myeloma patients achieved â‰¥ 6 x 106 CD34+ cells/kg cells in less than 2 apheresis after administration of plerixafor with G-CSF whereas only 34.4% achieved under G-CSF alone. In the same study, the successful rate in collecting â‰¥ 2 x 106 CD34+ cells/kg in â‰¤ 4 apheresis sessions after addition of plerixafor versus G-CSF alone were 95.3% and 88.3% respectively. The number of circulating CD34+ cells increased 4.8-fold after plerixafor while only increased 1.7-fold with G-CSF alone. Patients normally received plerixafor on Day 4 12 hours prior to apheresis on Day 5. Fowler et al. (2009) showed that 90% patients receiving combination of G-CSF plus plerixafor collected sufficient stem cells (range 2.5 - 6.2 x 106/kg) for autologous stem cell transplantation in their second attempts who have failed mobilization previously.
The optimal mobilization regimens for PBSC collection has not been determined. Previous studies have demonstrated that the lack of hematopoietic reconstitutions by PBSC grafts collected by either chemomobilization or G-CSF alone (Gertz et al. 2009; Narayanasami et al. 2001). Dingli et al. (2006) also argued about the in vivo purging effects of intermediate dose CY (3g/m2) chemomobilization and demonstrated a lack of difference in complete response rate of PBSCT when compared to G-CSF alone regimen.
1.4 Multiple Myeloma (MM)
Multiple Myeloma (MM) comprises about 10-15% of hematologic neoplasm and 1% of all cancers (Kyle & Rajkumar, 2008). It is characterized by the presence of abnormal plasma cell in the bone marrow and the skeleton, resulting in secretion of monoclonal immunoglobulins or light chains. MM is generally classified by the specific monoclonal proteins secreted from the plasma cells. It is a group heterogeneous diseases with overall survivals varying from few years to decades. The majority of patients are elderly with a median age of onset of about 60-70 years old. Clinically, the patients typically present with anaemia and bone pain resulting from infiltration of bone marrow and skeleton by neoplastic plasma cells. Patients with severe diseases can also present with renal failure and hyperviscosity. A number of staging systems are available for prognostication, including the Durie Salmon (DS) and the International Staging System (ISS), taking into consideration of abnormal biochemical tests, the degree of anaemia, skeletal destruction and renal dysfunction.
The optimal treatment of MM has not been defined. In the past, elderly patients were treated with steroid and alkylating agents like melphalan and less than 5% patients can achieve complete remission. Occasional young patients were treated with chemotherapy followed by allogeneic bone marrow transplantation with the possibility of a cure. However, this approach was associated with significant transplant-related mortality (TRM) of as high as 30-40%. As a result, the treatment has been unsatisfactory (Barlogie et al. 2004, Kyle and Rajkumar, 2009). Recent advances in the pathogenesis of MM have led to novel therapy for this disease. This includes immunomodulatory agents like thalidomide and lenolidomide and proteosome inhibitor bortezomib (VelcadeÂ®). These agents are used alone or in combination with steroid and chemotherapy, resulting in higher complete remission rate (Table 1). Despite these treatments, the diseases frequently relapses and further consolidation with autologous stem cell transplantation is now considered as the standard treatment, with TRM of less than 5% and a possibility of prolonging survival in this group of patients. Collection of adequate autologous PBSC for transplantation becomes a prerequisite for successful MM treatment. The impact of these novel therapeutic agents on PBSC yield is presently unclear.
Table 1. Summary of results associated with CR (Complete response) and OS (overall survival) rate using novel therapeutic agent for induction therapy of multiple myeloma
Palumbo et al. 2008.
Facon et al. 2007.
Harousseau et al. 2010.
Cavo et al. 2010.
Luwig et al. 2011.
Abbreviaions: MPT, melphalan, prednisone, and thalidomide; MP, melphalan and prenisone; VD, Velcade and dexamethasone; VAD, vincristine, doxorubicine, and dexamethasone; VTD, bortezomib, thalidomide, and dexamethasone; TD, thalidomide, dexamethasone; PAD, bortezomib, doxorubicin, and dexamethasone.
Median Overall survival (OS)
45 months vs 47.6 months (p = 0.79)
51.6 months vs 33.2 months (p = 0.027)
3 years OS, 81.4% vs 77.4% (p = 0.572)
3 years OS, 86% vs 84% (p = 0.3)
Superior in PAD arm
Complete response (CR)(%)
15.6% vs 3.7%
13% vs 2%
39.5% vs 22.5%
58% vs 41%
49% vs 34%
Number of patients
129 vs 126
125 vs 196
240 vs 242
236 vs 238
371 vs 373
MPT vs MP
VD vs VAD
VTD vs TD
PAD vs VAD
1.5 Determination of haematopoietic recovery
Successful hematologic reconstitution after autologous PBSCT depends on the dose of stem cells infused. A minimum number of CD34+ cell dose at 2 x 106 cells/kg recipient body weight is needed to ensure rapid haematological reconstitution. In the pre-CD34+ era, CFU-GM clonogenic assay of the graft facilitates estimation of granulopoiesis reconstitution after PBSCT. An estimation of 10-50 x 104 CFU-GM/kg are considered optimal for haematological reconstitution (To et al. 1997). However, CFU-GM assay is labour intensive and can only provide information about myeloid progenitors but not true HSC. On the other hand, enumeration of CD34+ by flow cytometry provides quantitative measurement of stem cell in the graft. Research findings demonstrated that there is a significant correlation between CD34+ cells and CFU-GM (Gianni et al, 1989; Jansen et al, 2007; To et al, 1986).
1.6 Research question and hypothesis
In this present study, I attempted to address the question if G-CSF alone can practically replace chemomobilization for patients with MM during PBSC mobilization in a cohort of patients who have been treated with thalidomide and bortezomib containing regimens. The use of plerixafor in patients who might fail G-CSF mobilization will also be evaluated. The pros and cons of these mobilization regimens will be compared.
Patients and Methods
Consecutive MM patients who underwent autologous stem cell transplantation in Queen Mary Hospital, Hong Kong, between 2009 - 2010 were studied. Clinical data including the hospital stay for mobilization and PBSCT, CD34+ cells, CFU-GM yield as well as hematological recovery after PBSCT were analyzed. Autologous PBSC were cryopreserved and re-infused to patients according to the standard procedures in the Bone Marrow Transplantation Centre, Queen Mary Hospital. Absolute neutrophil engraftment was defined when neutrophil count reaches consistently above 0.5 x 109/L for at least 3 days. Platelet engraftment is defined when platelet count rises to above 50 x 109/L unsupported by transfusion.
2.2 Mobilization and collection of PBSC
Patients were divided into 3 cohorts. In the chemomobilization (CY-G-CSF) group, patients received a dose of cyclophosphamide (3g/m2) on day 1 of mobilization followed by G-CSF (5Âµg/kg G-CSF administrated at every 12 hours) administration on day 2 onwards. PB CD34+ cells were monitored everyday after day 10. In the G-CSF group, patients were given G-CSF alone (5Âµg/kg G-CSF administrated every 12 hours) for four days. Plerixafor (0.24 mg/kg) was given subcutaneously on day 4 evening in patients (G-CSF-P) in whom the treating physicians considered likely to fail mobilization based on low PB CD34+ cells or insufficient PBSC harvested with G-CSF alone. CD34+ count in peripheral blood and PBSC product were enumerated by flow cytometry (Cytomics FC500, Beckman Coulter, USA) under ISHAGE guideline. Circulating CD34+ cells was enumerated on day 8 in the CY-G-CSF, and day 5 in the G-CSF and G-CSF-P group. PBSC collection by apheresis will begin when peripheral blood CD34+ rose above 10/Âµl and G-CSF would be continued until completion of PBSC collection. Apheresis was performed by continuous-flow blood cell separator (COBE Spectra, Gambro BCT, Lakewood, CO, USA). Optimal CD34+ dose was defined as 4 x 106/kg and minimal dose was 2 x 106/kg of recipient body weight.
Multiple Myeloma Patient underwent autologous stem cell transplantation
Figure 2. Study algorithm
Continuous G-CSF support and administered plerixafor (10 hours before upcoming apheresis) until â‰¥ 2 x 106/kg CD34+ cells were collected
2 x 5Âµg/kg/day G-CSF
Sign of decreasing in
PBSC CD34+ cells
Sign of decreasing in
PB circulating CD34+ cells
PB CD34+ count
~ 10 cells/Âµl
Enumeration of CD34+ cell in stem cell product
PB CD34+ count
~ 10 cells /Âµl
Enumeration of CD34+ cell in stem cell product
Continuous G-CSF support until â‰¥ 2 x 106/kg CD34+ cells were collected
Daily WBC monitoring
Daily WBC monitoring
Monitoring days to neutrophil engraftment after autologous stem cell transplantation
CFU-GM clonogenic assay
3g/m2 Cyclophosphamide (Day 1)
2 x 5Âµg/kg/day G-CSF
The apheresis product containing PBSC was transferred to stem cell transplant laboratory for cryopreservation on the same day. Fresh freezing medium was prepared daily as stock solution for each apheresis product containing 20% dimethyl sulfoxide (DMSO) (Cat. WAK-DMSO-10, WAK-Chemie Medical GmbH, Germany), 70% Hanks' balanced salt solution (HBSS) (Cat. 14170-112, Gibco, USA) and 10% autologous plasma. Freezing medium was kept in 4oC until use. Volume of the harvested stem cell product was measured by aspiration syringe and diluted by HBSS to desired volume. Equal volume of stock freezing medium was slowly added to the apheresis product which was pre-cooled at 4oC for 30 minutes with gentle agitation. The nucleated cell concentration of each bag was adjusted to 1 - 1.5 x 108/ml. Final concentration of DMSO in the cryopreserved stem cell product would be 10%, containing 5% autologous plasma. Mixture of stem cell product was then divided into cryocyte-freezing bags (Nexell, Irvine, CA, USA) or CryoMACS freezing bags (Miltenyi Biotec, GmbH, Germany) together with three cryogenic vials each containing 0.5ml of stem cells component as quality control. The cell products were placed into a controlled-rate freezer (Kryo-10, Planer, England) and cooled under a stepwise cooling program (Ramp 1: 10oC to 4oC at 5oC/min., Ramp 2: hold at 4oC for 10 min., Ramp 3: 4oC to -5oC at 2oC/min., Ramp 4: -5oC to -40oC at 1oC/min., Ramp 5: -40oC to -160oC at 5oC/min.). The cryopreserved bags together with the vials were stored in a liquid nitrogen storage tank until ready for re-infusion.
2.4 CFU-GM clonogenic assay
An aliquot of apheresis product was diluted and adjusted to 0.5 x 106/ml WBC with Hanks' balanced salt solution (Cat. 14170-112, Gibco, USA). 0.4ml at 0.5 x 106/ml cells suspension was dispensed into 4ml methylcellulose based media containing colony-stimulating factors (MethoCult H4534, Stem cell technologies, Vancouver, British Columbia, Canada). 1 ml aliquot of cell-methylcellulose media was plated in triplicate in 35-mm culture dishes. The final concentration of each culture dish contained 0.5 x 105 nucleated cells. After 14-days of humidified (>95%) incubation at 37oC with 5% CO2, CFU-GM colonies were counted using an inverted microscope. A cluster of 30 cells formed together would be classified as one colony (fig 3a, 3b). The frequency of CFU-GM colonies was averaged and total CFU-GM was calculated based on total nucleated cells in the apheresis product.
Total number of CFU-GM was calculated by the following equation:
Total WBC in the PBSC product
0.5 x 105
Average CFU-GM colonies
Total CFU-GM =
Total CFU-GM/kg was determined by dividing the total CFU-GM by body weight (kg) of recipient.
body weight of recipient (kg)
Total CFU-GM/kg =
Fig. 3a. CFU-GM, 40x
Fig. 3b. CFU-GM, 100x
2.5 Enumeration of CD34+ cell and gating strategy
The CD34+ flow cytometric analysis in peripheral blood and stem cell product was carried out by Flow Cytometry Laboratory, Department of Haematology, Queen Mary Hospital, Hong Kong. Analysis of absolute count of CD34+ cells was performed by commercial analytical kit Stem-Kitâ„¢. Cells staining was performed according to the manufacturer (ref. IM3630, Beckman Coulter, USA). CD34+ cells determination was according to ISHAGE guidelines in which the Stem-Count flourospheres enabled absolute counting of CD34+ cells. Briefly, two tubes labelled CD45/CD34/7-AAD each containing 20Âµl anti-CD45-FITC/anti-CD34-PE were prepared for CD34+ cell enumeration. Another tube labelled CD45/Ctrl/7-AAD containing 20Âµl anti-CD45-FITC/IsoClonic Control-PE was prepared for negative control. 100Âµl of fresh blood/apheresis sample in WBC concentration â‰¤ 15 x 109/ml (if necessary, sample needed to be diluted in DPBS to reach the desired concentration) was added into each tube. The tubes were immediately vortex vigorously for 5 seconds and incubated at room temperature for 15 minutes in the dark. Two millilitres (2ml) of ammonium chloride lysing solution was added to each of the three tubes after incubation to lyses red cells with gently vortex. Then, all tubes were incubated at room temperature for another 15 minutes in the dark. After ammonium chloride lysis, 100Âµl of Stem-Count flourospheres was added to each tube with immediately vortex. Samples were kept in dark on melting ice and ready for flow cytometric analysis within 1 hour. The sample was well vortexed before data acquisition.
2.5.1 Gating Strategy
Enumeration of viable CD34+ cells in blood/apheresis specimen was performed by single-platform protocol with sequential gating method according to ISHAGE (International society of hematotherapy and graft engineering) guidelines. Beckman Coulter CytoMic FC500 (USA) flow cytometer equipped with CXP system software was employed to perform samples analysis. As shown in fig. 4, histogram 1 displayed side-scatter versus 7-AAD where region J (0.3%) was set to include all the non-viable events. Histogram 2 displayed all viable events on histogram 1. Region A on histogram 2 was positioned to include all leukocytes except cell debris, platelet aggregates and CD45- events. Viable CD34+ cells (reddish cluster with low side scatter and low/immediate CD45+ staining) and lymphocytes region (bright CD45+, low side scatter) are also displayed within region A. A total number of 75,000 CD45+ events were collected on histogram 2. Appropriate setting of region is important to serve as a denominator for the calculation of percentage of CD34+ cells. Histogram 3 was gated on region A. A discrete cluster from dim to bright viable CD34+ events with low side scatter was displayed in region B. Region C on histogram 4 which was set on region A and B to exclude platelet aggregates and cell debris to characterize true CD34+ events. Region D was gated by region A, B and C whether other events with weak CD34+ and CD45+ staining will be eliminated. Same procedure was done in negative control. After gating procedures established, absolute number of viable CD34+ cells per Âµl was automatically calculated by the system software.
Histogram 4 (CD34+ cells)
Histogram 6 (Negative control)
Fig. 4. Gating strategy of CD34+ enumeration by using Stem-Kitâ„¢. Methodology of data acquisition was described in patients and methods. (Section 2.6.1)
2.6 Autologous Stem cell transplant conditioning regimen
Patients received high dose chemotherapy (Melphalan 200 mg/m2 patient surface area) 2 days before stem cell transplant, according to the transplantation protocol in the Bone Marrow Transplantation Unit, Queen Mary Hospital. Cryopreserved PBSC was thawed by immersion in a 37oC to 40oC waterbath with continuous shaking and gently massage until visible ice was completely melted. Thawed products were transfused immediately to the patients.
2.7 Statistical analysis
Results are expressed as mean ï‚± standard error of mean (SEM). Comparison between groups of data was evaluated by 2-sided t-test. Comparison of neutrophil and platelet engraftments between groups was evaluated by Kruskal-Wallis H Tests. Pearson's correlation coefficient was used to examine the correlation between CD34+ and CFU-GM. Statistical calculations were performed with computer software IBM SPSS, Version 13. A p-value of less than 0.05 is considered statistical significant.
3.1 Patient population
From Jan-2009 to Oct-2010, a total number of 28 multiple myeloma patients were enrolled in this study. Eighteen patients received CY-G-CSF mobilization. Ten patients were mobilized by G-CSF alone of whom five of them have required plerixafor at the discretion of treating haematologists. The baseline characteristics of the three groups were tabulated in Table 2. One patient whose PBSC was mobilized have not underwent autologous PBSCT.
Table 2 Patients characteristics
G-CSF + Plerixafor
Body Weight (kg)
Myeloma type according to type of paraprotein
VCMD + PAD
VAD + VTD + TD
VTD + Thal/Dex
#Thal/Dex: Thalidomide/Dexamethasone; PAD: Bortezomib, Adriamycin, Dexamethasone; VTD: Bortezomib, Thalidomide, Dexamethasone; VCMD: Vincristine, Cyclophosphamide, Melphalan, Dexamethasone; VAD: Vincristine, Adriamycin, Dexamethasone
3.2 Clinical characteristics during stem cell mobilization
The median hospital stay in the CY+G-CSF group was 13 days (range: 11-15 days). The median nadir neutrophil and platelet count were 0.095 x 106/ml (range: 0 - 0.9 x 106/ml) and 68.5 x 106/ml (range: 13 - 175 x 106/ml) despite continuous G-CSF administration. Patients in the G-CSF or G-CSF-P groups did not need hospitalization and there was no decrease in neutrophil or platelet counts during mobilization (Table 3).
Table 3. Clinical characteristics during stem cell mobilization
Nadir neutrophil counts (x 106 /ml)
Nadir platelet counts (x 106 /ml)
3.3 Impact of CD34+ cell dose in different mobilization regimen
The total number of CD34+ cells/kg collected was shown in fig. 2. The median number of total CD34+ cells collected in CY+G-CSF, G-CSF and G-CSF+P groups were 8.265 x 106/kg (range: 3.1 - 24.5 x 106/kg), 2.8 x 106/kg (range: 2.10 - 3.56 x 106/kg) and 2.38 x 106/kg (range: 2.06 - 3.33 x 106/kg) respectively .
Fifteen out of eighteen (15/18) patients in the CY+G-CSF group achieved the minimum number of 2 x 106/kg CD34+ stem cell in one apheresis session. Three patients achieved sufficient PBSC in two sessions. All patients in the G-CSF and 4 patients in G-CSF+P group have required at least two apheresis sessions. Plerixafor significantly increased PBSC yield compared with patients who received G-CSF only (Figure 4 and Table 5).
Table 5. Apheresis yield of CD34+ cells/kg (x 106/kg) in each group
No. of patients
Median values and (range)
8.265 (1.64 - 24.4)
1.81 (1.34 - 2.01)
0.84 (0.31 - 2.89)
1.46 (0.91 - 1.72)
0.92 (0.31 - 1.55)
1.43 (0.64 - 2.07)
2.8 (2.01 - 3.56)
2.38 (2.06 - 3.33)
Fig. 4. Total CD34+ cells /kg collected in each group
3.4 Impact of CFU-GM clonogenic assay
The median doses of infused CFU-GM in the CY-G-CSF, G-CSF and G-CSF-P groups were 119 x 104/kg (range: 44.7 - 311 x 104/kg), 67.76 x 104/kg (range: 20.6 - 107.2 x 104/kg) and 102.84 x 104/kg (range: 41 - 251.6 x 104/kg) respectively.
Table 5. Clonogenicity of CFU-GM (x 104/kg) infused in PBSC product
No. of patients
Median values and (range)
119 (25.1 - 311)
35 (5.3 - 66.5)
25.2 (10.3 - 73.6)
28.1 (19.6 - 35.9)
30 (15.3 - 52.4)
42.05 (30 - 57.9)
119 (44.7 - 311)
65 (20.6 - 107.2)
73.6 (41 - 251.6)
Fig. 6. Total CFU-GM/kg (x 104) infused under different mobilization regimen
3.5 Correlation between CD34+ cell and CFU-GM
The correlation coefficient between CD34+ and CFU-GM in CY-G-CSF and G-CSF groups were shown in Figure 6. There was statistical correlation in CD34+ cells collected and transplanted CFU-GM in CY group (r = 0.822, p < 0.001). However, we found that there was a weak correlation and no significant correlation between collected CD34+ cells and transplanted CFU-GM in G-CSF group (r = 0.377, p = 0.282).
Fig. 7a. CY group
Fig. 7b. G-CSF and G-CSF+P group
Fig. 7 Correlation graph relating total CD34+/kg collected and total CFU-GM/kg infused. a). correlation between collected CD34+ cells (x 106/kg) and CFU-GM infused (x 104/kg) in CY group; r = 0.822, p < 0.0001. b). correlation between collected CD34+ cells (x 106/kg) and CFU-GM infused (x 104/kg) in G-CSF and G-CSF+P group; r = 0.377, p = 0.282.
3.6 Kinetics of circulating CD34+ cells after Plerixafor administration
There were 5 patients in the G-CSF-P group. Their PB CD34+ cell counts and yields were listed in Table X. Plerixafor administration resulted in significant increase in PB CD34+ counts 12 hours after administration (before: 14.75 ï‚± 2.14/Âµl vs after: 41.75 ï‚± 4.85/Âµl, p=0.02). The total CD34+ cell yield was also increased (before: 0.88 ï‚± 0.33 x 106/kg vs after: 2.49 ï‚± 0.29 x 106/kg, p=0.021).
Table 6. Fold increase of peripheral blood CD34+ cell count after plerixafor administration
Fold increase (x)
PB CD34+ (cell/Âµl)
PB CD34+ (cell/Âµl)
3.7 Impact on number of apheresis procedures
Only three patients under CY+G-CSF groups needed for second apheresis sessions while all the patients in G-CSF groups and 4 patients in the G+CSF-P were required at least 2 apheresis sessions to collect sufficient stem cells.
Table 7. Number of apheresis procedures
No. of patients
3.8 Impact on engraftment days
In the total 28 patients that underwent stem cell mobilization, one patient in the G-CSF-P group did not proceed to autologous PBSCT. Neutrophil engraftments were similar in the three groups (Kruskal-Wallis H Test, p = 0.062). Furthermore, no significance difference were observed in platelet engraftment among the three groups (Kruskal-Wallis H Test, p = 0.319). In fact, there was no correlation between neutrophil and platelet engraftments and the CD34+ or CFU-GM doses in each transplant (Figure XX)
Table 8. Days of neutrophil and platelet engraftment in different group
No. of patient
Engraftment Days* :
*Neutrophil engraftment: ANC â‰¥ 0.5 x 106/ml for 3 consecutive days
Platelet engraftment: Plt > 50 x 106/ml
r = -0.366, p = 0.135
Fig. Xx Correlation of total CD34+ cells/kg collected and total CFU-GM/kg infused to neutrophil engraftment in CY+G-CSF group
r = -0.4, p = 0.1
Fig. Xx Correlation of total CD34+ cells/kg collected and total CFU-GM/kg infused to platelet engraftment in CY+G-CSF group
r = -0.395, p = 0.105
r = -0.224, p = 0.371
Fig. Xx Correlation of total CD34+ cells/kg collected and total CFU-GM/kg infused to neutrophil engraftment in G-CSF group
r = 0.706, p = 0.182
r = 0.494, p = 0.397
Fig. Xx Correlation of total CD34+ cells/kg collected and total CFU-GM/kg infused to platelet engraftment in G-CSF group
r = -0.423, p = 0.478
r = 0.152, p = 0.807
Fig. Xx Correlation of total CD34+ cells/kg collected and total CFU-GM/kg infused to neutrophil engraftment in G-CSF + P group
Fig. Xx Correlation of total CD34+ cells/kg collected and total CFU-GM/kg infused to platelet engraftment in G-CSF + P group
r = -0.825, p = 0.175
r = -0.466, p = 0.534
High dose chemotherapy followed by autologous PBSCT is currently the standard treatment for patients with multiple myeloma but the optimal mobilization regimens have not been determined (Lie & To, 1997). In this study, we demonstrated that Cy+G-CSF mobilization resulted in higher CD34+ cell yield and fewer number of apheresis sessions compared with those who received GCSF with or without plerixafor. Similar observations have been extensively reported. A number of retrospective studies have demonstrated a 1 to 5 fold increase in CD34+ cell yield in CY+G-CSF compared with G-CSF alone (Gertz et al. 2009; Alegre et al. 1997 and Mark et al. 2008). CY may damage the ability of HSC to adhere to the BM microenvironment, thereby enhancing their mobilization into the circulation Cesana et al. (1998). The difference in CD34+ cell yield between chemomobilization and G-CSF alone might be overcome by a higher dose of G-CSF (24Âµl/kg/day) of the latter as demonstrated by Kroger et al. (1998). The apparent superiority of chemomobilization over standard dose of G-CSF has led to the use of former regimen in Queen Mary Hospital, Hong Kong. However, as new therapeutic agents for myeloma treatment have emerged that are much less myelotoxic than conventional chemotherapy, and novel mobilization agent plerixafor has become available, it is possible that HGF alone may suffice as mobilization agent. This issue was tested in the present study in which we addressed its feasibility in a single institution and there were a number of issues which might have an impact on patient management.
First, we demonstrated that a higher CD34+ cell yield in the CY+G-CSF group had no clinical advantage into terms of engraftment or hospital stay during transplantation. A cell dose of 2 x106 cells/kg was considered minimal for successful engraftment. Our results suggested that a higher cell dose had no advantage and corroborated with those of Stiff et al. (2010) and Narayanasami et al. (2001) who showed that neutrophil and platelet engraftment were not cell dose dependent. However, this was different from those of Ketterer et al. (1998), Shpall et al. (1998) and Statkute et al. (2007) who demonstrated a more rapid hematopoietic recovery with higher cell dose. The results were also in contrast to those of umbilical cord blood transplantation in which engraftment and transplantation outcome were critically dependent on the cell dose in the graft (Wagner et al. 2002, Avery et al. 2010). Therefore, a lower cell yield in the G-CSF groups with or without plerixafor had no negative impact on patient outcome. However, the long-term outcome in these patients would have to be carefully evaluated in prospective study.
Second, we demonstrated that the use of cyclophosphamide during mobilization necessitate hospitalization and resulted in profound neutropenia and thrombocytopenia during mobilization. This was associated with potential risk of potentially fatal infection and bleeding. All 18 patients undergoing CY+G-CSF mobilization recovered uneventfully from cytopenia. However, mortality during chemomobilization has been reported (Fitoussi et al. 2001)( Ref, let me know if you can't find it).
Third, despite the fact that our patients have received minimal myelotoxic agents in their initial treatment, five out of 10 patients failed to achieve the minimal cell dose for transplantation and have required the addition of plerixafor. An average of 2.57-fold increase in circulating CD34+ cells was observed and all patients achieved sufficient cell yield after plerixafor. Plerixafor was normally given on day 4 evening in a 5 days G-CSF mobilization protocol (Calandra et al. 2008; Fowler et al. 2009; Basak et al. 2010). However, in the present study, plerixafor was only given to patients in whom the treating haematologists considered likely to fail G-CSF mobilization based on low PB CD34+ counts or CD34+ cell yield in the initial harvests. This "risk-adapted" strategy might limit the use of this expensive agent to patients who really need it and may reduce the drug cost (HKD $ 50,000 per dose) incurred to the institution.
Finally, we have also examined CFU-GM in the PBSC graft. Before the wide accessibility to flow cytometry, To et al. (1986) suggested that CFU-GM in the graft could be used to predict the haematological recovery after transplantation. Later, a strong correlation between circulating CD34+ cells and CFU-GM were demonstrated (Siena et al. 1991). Similar correlation was observed in the present study. In the present study, we demonstrated a close relationship between CFU-GM/kg and CD34+ cells/kg in CY+G-CSF group with r value of 0.822, (p < 0.0001) but not in G-CSF group (r = 0.377, p = 0.282). The correlation coefficient in CY+G-CSF group of this study was higher than that reported by Jansen et al. (2007). On the other hand, the lack of significant correlation in the G-CSF could be explained by the smaller patient number compared with the CY+G-CSF group. Whether there was a genuine functional difference in HSC or progenitors mobilized by different mobilization regimens would have to be further investigated. At present, CD34+ cell enumeration remained to be the standard in most transplantation centre although CFU-GM may be helpful in assurance of quality of the cryopreserved graft after thawing.
Our study is limited in sample size and its prospective nature. In the future, prospective and carefully controlled study should be performed to compare the PBSC yield and clinical outcome of the three regimens. Furthermore, the indication for the use of plerixafor should be standardized so that the results could be translated into practical guidelines based on the "risk-adapted" approach in this study. The impact of cyclophosphamide as in vivo purging on long term disease control would have to be prospectively evaluated.
The availability of newer mobilization agent plerixafor has enabled us to evaluate the feasibility of G-CSF mobilization in our centre in which CY+G-CSF has been the standard. The CY+G-CSF regimen was superior in PBSC yield and few apheresis session needed but these benefits were offset by the need of hospitalization and profound cytopenia at nadir during mobilization and the lack of advantages in terms of hematopoietic recovery. The long-term outcome of patients mobilized with these protocols would have to be evaluated prospectively.