Human Haemopoietic Progenitor Cell Mobilization

Michael John Watts

A thesis submitted to the University of London for the degree of Doctor of Philosophy

1999

The Department of Haematology University College London Medical School University College London 98, Chenies Mews LONDON WC1E6HX ProQuest Number: U642271

All rights reserved

INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.

In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest.

ProQuest U642271

Published by ProQuest LLC(2015). Copyright of the Dissertation is held by the Author.

All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code. Microform Edition © ProQuest LLC.

ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 Abstract

The advent of recombinant growth factors in the late 1980's has ushered in a new era of haematopoietic progenitor cell (HPC) therapy by facilitating the mobilization of bone marrow progenitors into the circulation where they can be collected in large numbers by apheresis. The work of this thesis has defined the minimal and optimal CD34+ cell threshold requirements for engraftment. The frequency of poor mobilization was noted and risk factors determined. It was demonstrated that poor mobilization was usually a feature of bone marrow damage rather than a specific mobilization defect. In addition, studies in normal volunteers indicated that there was wide inter-individual variation in G-CSF induced progenitor cell mobilization which was not due to G-CSF pharmacodynamic variability.

The clinical feasibility of CD34+ cell purification was determined and its clinical limitations documented. These purified cells were used clinically to demonstrate that the increased numbers of T-cells and monocytes in a peripheral blood stem cell (PBSC) transplant compared to bone marrow did not account for the more rapid haematological recovery that occurs with PBSC.

These studies answer specific clinically relevant questions and form a platform for future studies in "graft engineering". Acknowledgements

The work contained in this thesis represents 5 years of sometimes intensive but always satisfying effort. I am very grateful to Professor David Linch and Dr Tony Goldstone for their confidence in me (which exceeded my own) to play a major part in the successful development of the peripheral blood stem cell program at University College London Hospitals. This work began with my predecessor, Mark Jones and his wife Sally with our first PBSC transplant at UCLH in 1992.

Above all David's intellectual input and clarity of vision has taught me that the most likely route to successful study is a focussed approach which is contrary to my natural inclination of aimless enthusiast.

I would also like to pay tribute to the conscientious work of Angela Sullivan who joined me in August 1994 as the clinical demands of the blood stem cell program was rising exponentially. When she left to conduct two successful in vivo expansion experiments of her own, resulting in two baby boys I was fortunate to find Stuart Ings. Stu has been a mainstay of the stem cell laboratory, particularly as my thesis came closer to completion. His reward of course is more work and his own PhD registration last September.

I would also like to thank all of the daycare specialist nurses without whom most of the clinical audits would have been impossible and my colleagues in haematology for their unstinting support.

Finally, my wife and three children who have complained about my absence at home may hopefully get a chance to complain about my presence at home.

Mike Watts, December 1998 Table of Contents

Page Title page 1 Abstract 2 Acknowledgements 3 Table of contents 4 List of figures 8 List of tables 10 List of abbreviations IZ List of work published during the course of this thesis 13

Chapter 1 Introduction 1.0.1 Bone marrow the “seed bed” of the blood 15 1.0.2 The bone marrow and the Bomb 18 1.0.3 Early marrow transplantation 18 1.0.4 Improvements in survival 19 1.0.5 The growth of autologous transplantation 20 1.0.6 Stem cell sources 20 1.0.7 "Graft engineering" 21 1.0.8 Transplantation and this thesis 22 1.0.9 Circulating stem cells 23 1.1.0 Possible mechanisms of HPC mobilization 24 1.1.1 Monitoring of PBSC Collection and Transplantation 31 1.1.2 Clonogenic functional assays 32 1.1.3 Flow cytometric CD34+ cell measurement 33 1.1.4 Progenitor threshold requirements 35 1.1.5 Apheresis timing 35 1.1.6 Optimisation of Progenitor Cell Mobilization 36 1.1.7 Post PBSCT Growth Factors 41 1.1.8 Multiple High Dose Therapies 42 1.1.9 Use of Venesection Instead of Apheresis 42 1.2.0 Use of cytoklne-prlmed bone marrow 43 1.2.1 Future Developments 44 1.2.2 CONCLUSION 45 2 General Methods and Materials Page 2.0.1 "General purpose" tissue culture medium 46 2.0.2 Sample handling of apheresis products 46 2.0.3 Cytocentrifuge preparations 46 2.0.4 Viability counts 47 2.0.5 Fiow cytometry for CD34+ cell counts 47 2.0.6 Alkaline Phosphatase-anti-Aikaline Phosphatase (APAAP staining) 49 2.0.7 Manual counting of APAAP stained CD34+ cells 49 2.0.8 Esterase staining 51 2.0.9 Ciinicai CD34+ ceil purification 52 2.1.0 Methyicellulose colony forming cell (CFC) assays 52 2.1.1 In-house methyicellulose media 52 2.1.2 Recombinant growth factors added 54 2.1.3 Commercial methocellulose media 54 2.1.4 Plating concentrations 54 2.1.5 Piiot vial thaw method for CFC 55 2.1.6 Colony counting 55 2.1.7 Mononuclear cell preparation (MNC) and CFC/ml 56 2.1.8 Apheresis CFC dose caiculations 57 2.1.9 Cytokine Assays 58 2.2.0 Data handling and statistical analysis 59

Chapter 3 Progenitor Threshold Requirements for Haematological Recovery 3.0 INTRODUCTION 60 3.1 PATIENTS STUDIED 62 3.2 RESULTS 63 3.2.1 Overali incidence of engraftment delays 63 3.2.2 MNC dose and engraftment 64 3.2.3 CD34-h cell dose and engraftment 64 3.2.4 GM-CFC dose and engraftment 65 3.2.5 Correlation between CD34+ cells and GM-CFC collected at 70 apheresis 3.2.6 Effect of post infusion G-CSF on haematoiogical recovery 72 3.3 DISCUSSION 74

Chapter 4 Factors Predicting for Mobilization Efficiency 4.0 INTRODUCTION 77 4.1.0 PATIENTS AND METHODS 78 4.1.2 PBSC mobiiization and coilection 79 4.1.2 Statisticai analysis of data 80 4.2.0 RESULTS 80 4.2.1 Toxicity of Mobiiization regimen 80 4.2.2 Effect of G-CSF type and leukapheresis machine on PBSC yieids. 80 4.2.3 Factors Predicting for PBSC Yieid 83 4.2.4 Proportion of patients achieving different coilection thresholds 84 4.3 DISCUSSION 87 Chapter 55 Variability of Progenitor Numbers Mobilized by G-CSF in Normal Subjects Page 5.0 INTRODUCTION 90 5.1 METHODS AND MATERIALS 92 5.1.1 Volunteers 92 5.1.2 Study design 92 5.1.3 Progenitor Ceii Measurements 93 5.1.4 Serum G-CSF Assay 94 5.2 RESULTS : 94 5.2.1 Baseline Values 94 5.2.2 White Blood Ceii count 94 5.2.3 Progenitor Ceii Counts 94 5.2.4 Pharmacokinetics 95 5.2.5 individual Variability 96 5.2.6 Adverse events 96 5.3 DISCUSSION 102

Chapter 66 The Role of Back up Bone Marrow for “Poor Mobilizers” 6.0 INTRODUCTION 106 6.1 PATIENTS AND METHODS 108 6.1.1 Patients studied 108 6.1.2 Mobiiization regimens 108 6.1.3 PBSC Harvesting 109 6.1.4 Bone Marrow Han/esting 110 6.1.5 Cryopreservation 110 6.1.6 Progenitor ceii assays 110 6.1.6 High dose therapy regimens 111 6.1.7 Haemopoietic Recovery 112 6.2 RESULTS ; 112 6.2.1 Overaii haematoiogical recovery 112 6.2.2 Haematological recovery of the poorest mobiiizers 112 6.2.3 Back-up bone marrow use 114 6.3 DISCUSSION 119 Chapter 7 Purification Page 7.0 INTRODUCTION 123 7.1 METHODS AND MATERIALS 124 7.1.1 Patients studied 124 7.1.2 Conditioning 124 7.1.3 PBSC mobiiization and coilection 125 7.1.4 Purified CD34+ cell cryopreservation 125 7.1.5 CD34+ ceii purification protocols 126 7.1.6 Processing modifications 126 7.1.7 Progenitor evaluation 127 7.1.8 Caiculations used to assess ceil yieids 128 7.2 RESULTS : 128 7.2.1 Progenitor ceil purity and recovery (yieids) 128 7.2.2 Assessment of Yield variability 132 7.2.3 Effects of the pre-column washing steps 132 7.2.4 Cionogenicity of CD34+ purified ceils 134 7.2.5 Haematoiogical recovery after infusion of purified CD34+ ceils 135 7.2.6 Single versus multiple harvests 136 7.3 DISCUSSION 148

Chapter 88 Accessory Cells do not contribute to Serum G-CSF or IL-6 Cytokine Levels nor to Rapid Haematological Recovery post PBSCT

8.0 INTRODUCTION 152 8.1 METHODS AND MATERIALS 152 8.2.1 Patient groups 152 8.2.2 PBSC mobilization and coilection 153 8.2.3 CD34+ ceii selection 154 8.2.4 Progenitor evaluation 154 8.2.5 Accessory ceii depletion 155 8.2.6 Cytokine assays 155 8.2.7 Statisticai analysis of data 156 8.3 RESULTS : 156 8.3.1 CD34+ ceil purification and accessory cell depletion 156 8.3.2 Cytokine profiles following stem ceil infusions 159 8.3.3 Haematological recovery following infusion of purified cells 162 8.4 DISCUSSION 165

Chapter 99 Discussion and Conclusions 167

References 182

Appendix I 205 List of Figures

Chapter 1 Figure Page 1.1 Clonogenic assays and antigens associated with differentiation of 17 Haemopoietic progenitor cells 1.2 Normal bone marrow structure showing sinusoidal blood supply and 26 egress into the circulation of mature blood cells.

Chapter 2

2.1 CD34+ cell measurement by flow and APAAP. MGG appearances 48 of progenitors and of colonies in methyicellulose 2.2 Correlation of visual and flow cytometric CD34+ cell counts 50

Chapter 3\

3.1.1 CD34+ cell dose infused and engraftment of A) neutrophils and B) 67 platelets 3.1.2 GM-CFC dose infused and engraftment of A) neutrophils and B) 68 platelets 3.1.3 MNC dose infused and engraftment of A) neutrophils and B) 69 platelets 3.2.0 Correlation between CD34+ cells and GM-CFC collected at 71 apheresis 3.3.0 CD34+ cell dose re-infused and post infusion G-CSF effect on 73 A) neutrophil and B) platelet recovery

Chapter 4

4.1 A) Kinetics of progenitor mobilization after low dose 82 cyclophosphamide and filgrastim or lenograstim G-CSF formulations B) Cell doses collected at first apheresis 4.2 Flow chart showing relationship of poorest mobiiizers to prior 83 therapy 4.3 Distribution of progenitor thresholds collected in a single or 86 combined with a second apheresis collection A) CD34+ cell B) GM-CFC yields

8 Chapter 5 Figure Page 5.1 Flow chart of study design 92 5.2 Serial white cell (A), GM-CFC (B) and CD34+ cell (C) counts 97 following G-CSF mobilization in 20 normal subjects. 5.3 Mean serum G-CSF levels after first dose of G-CSF 98 5.4 Individual serial GM-CFC count curves in 20 subjects mobilized 99 alternately with A) glycosylated and B) non-glycosylated G-CSF 5.5 Intra-individual progenitor mobilization between mobilization 100 events using the two formuiations of G-CSF 5.6 Lack of relationship between serum G-CSF levels and peak 101 progenitor levels

Chapter 6

6.1 Flow chart showing patients who failed PBSC mobiiization and 111 went on to receive high dose therapy, 6.2 Distribution plot of days to engraftment against various progenitor 118 thresholds re-infused

Chapter 7

7.1 CD34+ ceii purity post selection is related to start CD34% in 130 apheresis harvest 7.2 Correlation between harvest and purified CD34+ ceii dose 131 7.3 Wide CD34+ cell yield recovery from apheresis harvests 131 7.4.1 Purified CD34+ cells re-infused and haematological recovery 140 7.4.2 Purified CD34+ cell GM-CFC dose re-infused and haematological 141 recovery 7.4.3 Whole PBSC CD34+ cells re-infused and haematological recovery 142 7.4.4 Whole PBSC GM-CFC re-infused and haematological recovery 143 7.4.5 Single PBSC harvest CD34+ cells re-infused and haematological 144 recovery 7.4.6 Single PBSC harvest GM-CFC re-infused and haematological 145 recovery 7.4.7 Multiple PBSC harvest CD34+ cells re-infused and haematological 146 recovery 7.4.8 Multiple PBSC harvest GM-CFC re-infused and haematological 147 recovery

Chapter 8

8.1 Accessory cell depletion following CD34+ cell purification 158 8.2 Serial cytokine changes and haematological recovery 160 8.3 Endogenous cytokine levels post PBSCT: A) G-CSF, B) IL-6 161 8.4 Flaematological recovery of A) neutrophils and B) platelets in 164 patients receiving whole or CD34+ cell purified PBSC List of Tables Chapter 1 Table Page 1.1 Current Haemopoietic Cell (HPC) transplantation 21 1.2 Cross-transplantation studies in the G-CSFR "knock-out" mouse 29 1.3 Mobilization response of mutant mice to blocking antibodies against 31 VLA-4 (expressed on HPC) or to its stromal cell ligand VCAM-1 1.4 Mobilization of breast cancer patients with G-CSF at 10pg/kg/day 39 only or given SCF at 10|ig for three days prior to G-CSF then concomitantly for the remaining 7 days

Chapter 3

3.1 Patients who received high dose therapy with PBSC support 63 3.2 Progenitor dose re-infused and days to haematological recovery 64 3.3 Progenitor dose re-infused and risks of delayed haematological 66 recovery

Chapter 4

4.1 Patients studied 79 4.2 Univariate and multivariate analysis of clinical factors for the 84 achievement of PBSC progenitor thresholds

Chapter 5

5.1 Mean peak white cell and progenitor counts on days 4, 5 and 6 after 95 glycosylated or non-glycosylated G-CSF mobilization 5.2 Pharmocokinetic parameters for glycosylated or non-glycosylated 96 G-CSF after a first dose at 5pg/kg

Chapter 6

6.1 PBSC mobilization regimens and patients studied 109 6.2 Proportion of patients with a poor total progenitor yield of less than 113 1 x10®/kg CD34+ cells 6.3 Haematological recovery in patients receiving high dose therapy 114 and various numbers of haemopoietic progenitor cells 6.4 Patients with less than 1 x10®/kg CD34+ cells and less than 1 117 xIOVkg GM-CFC in whom bone marrow was subsequently harvested

10 Chapter 7 Table Page 7.1 Subjects studied and number of CD34+ cell purification procedures 124 performed 7.2 Effect of harvest CD34+ cell concentration on purity and yield of 129 CD34+ enriched fraction 7.3 Results of identical CD34+ cell processing on consecutive 132 apheresis collection from 19 individuals 7.4 Outcome of CD34+ cell purification with two modifications to 133 standard processing 7.5 Cionogenicity of CD34+ cell selected products 135 7.6.1 Progenitor dose re-infused and risk of delayed haematological 138 recovery for patients receiving one or more whole PBSC collections compared to those receiving CD34-selected from a single harvest 7.6.2 Progenitor dose re-infused and risk of delayed haematological 139 recovery for patients receiving a single whole PBSC harvest compared to those receiving 2 or more whole collections

Chapter 8

8.1 Patients receiving whole or CD34+ cell selected PBSC 153 8.2 Processing results of CD34+ cell purification 157 8.3 Haematological recovery of patients receiving whole or CD34+ cell 163 purified PBSC

11 List of Abbreviations used:

ABMT Autologous Bone Marrow Iransplantation APAAP Alkaline Phosphatase-Anti- Alkaline Phosphatase BFU-E "Bursf'-Forming Unit for Erythroid cells CFC Colony-Forming Cell CFU-GM (GM-CFC) Colony-Forming Unit for Granulo-Monocytic cells DC Dendritic Cell EPO Erythropoietin Flt-3L Fms-like tyrosine kinase receptor ligand G-CSF Granulocyte-Colony Stimulating Factor GM-CFC (CFU-GM) Granulocyte-Monocyte Forming Cell (as for CFU-GM) GM-CSF Granulocyte-Monocyte-Colony Stimulating Factor HD Hodgkin's Disease HGNHL High Grade Non-Hodgkin's Lymphoma HIFCS Heat-Inactivated Fetal Calf Se rum HPC Haemopoietic Progenitor Cell IL-3 lnterLeukin-3 (Other cytokines include IL-6, IL-7, IL-8) IMDM Iscove's Modified Dulbecco's Medium LFA-1, LFA-2 Leukocyte-Activating Factor-1,2 LGNHL Low Grade Non-Hodgkin's Lymphoma LTCIC Long-Term Colony Initiating Cell MNC MonoNuclearCell NHL Non-Hodgkin's Lymphoma PBPC Peripheral Blood Progenitor Ceils (also PBSC) PBS Phosphate-Buffered Saline PBSC Peripheral Blood Stem Cells (also PBPC) RPMI Roswell Park Memorial Institute Media SCF Stem Ceil Factor (c-kit ligand) TBI Total Body Irradiation TBS Tris Buffered Saline TPO Thrombopoietin (c-mpi ligand) VCAM-1 Vascular Cell Adhesion Molecule-1 VLA-4 Very-Late Antigen-4 WBC White Ceii Count

12 Work Published During the Course of this Thesis

1. Jones, H.M., Jones, S.A., Watts, MJ., Khwaja, A., Mills, W., Fielding, A., Goldstone, A.H. & Linch, D.C. (1994), Development of a simplified single-apheresis approach for peripheral- blood progenitor-cell transplantation in previously treated patients with lymphoma. J-Clin- Oncol, 12, 1693-702

2. Watts, M.J., Jones, H.M., Sullivan, A.M., Langabeer, S.E., Jamieson, E., Fielding, A., Williams, C., Berenson, R.J., Goldstone, A.H. & Linch, D.C. (1995). Accessory cells do not contribute to G-CSF or IL-6 production nor to rapid haematological recovery following peripheral blood stem cell transplantation. B rJ Haematol, 91, 767-72.

3. Serke, S., Watts, M.J., Knudsen, L.M., Kreissig, C., Schneider, U., Schwella, N., Linch, D. & Johnsen, H.E. (1996). In-vitro cionogenicity of mobilized peripheral blood CD34-expressing cells: inverse correlation to both relative and absolute numbers of CD34-expressing cells. Br J Haematol, 95, 234-40.

4. Linch, D.C. & Watts, M.J. (1996). Standards for PBSC mobilization and collection. Ann Oncol, 7, 9-10.

5. Fielding, A.K., Goldstone, A.M., Sullivan, A.M., Watts, M.J., Jamieson, E. & Linch, D.C. (1996). Peripheral blood stem cell support does not accelerate haemopoietic recovery from miniBEAM chemotherapy. Bone Marrow Transplant, 18, 507-11.

6. Watts, M.J., Sullivan, A.M., Jamieson, E., Pearce, R., Fielding, A., Devereux, S., Goldstone, A.H. & Linch, D.C. (1997). Progenitor-cell mobilization after low-dose cyclophosphamide and granulocyte colony-stimulating factor: an analysis of progenitor-cell quantity and quality and factors predicting for these parameters in 101 pretreated patients with malignant lymphoma. J Clin Oncol, 15, 535-46.

7. Watts, M.J., Sullivan, A.M., Ings, S.J., Leverett, D., Peniket, A.J., Perry, A.R., Williams, C.D., Devereux, S., Goldstone, A.H. & Linch, D.C. (1997). Evaluation of clinical scale CD34+ cell purification: experience of 71 immunoaffinity column procedures. Bone Marrow Transplant, 20, 157-62.

8. Watts, M.J. & Linch, D.C. (1997). Peripheral blood stem cell transplantation. Vox Sang, 7 3 , 135-42 (invited review).

9. Watts, M.J., Addison, I., Long, S.G., Hartley, S., Warrington, S., Boyce, M. & Linch, D.C. (1997). Crossover study of the haematological effects and pharmacokinetics of glycosylated and non-glycosylated G-CSF in healthy volunteers. B rJ Haematol, 98, 474-9.

10. Serke, S., Arseniev, L., Watts, M.J., Fritsch, G., Ingles-Esteve, J., Johnsen, H.E., Linch, D., Cancelas, J.A., Meyer, O., Kadar, J.G., Huhn, D. & Matcham, J. (1997). Imprecision of counting CFU-GM colonies and CD34-expressing cells. Bone Marrow Transplant, 20, 57-61.

13 11. Collins, P., Watts, M.J., Brocklesby, M., Gerrltsen, B. & Veys, P. (1997). Successful engraftment of haploidentical stem cell transplant for familial haemophagocytic lymphohistiocytosis using both bone marrow and peripheral blood stem cells. Br J Haematol, 96, 644-6.

12. Williams, C.D., Linch, D.C., Watts, M.J. & Thomas, N.S. (1997). Characterization of ceii cycie status and E2F compiexes in mobilized CD34+ ceiis before and after cytokine stimuiation. Blood,90, 194-203.

13. Watts, M.J., Addison, i., Ings, S.J., Long, S.G., Hartley, S., Warrington, S., Boyce, M. & Linch, D.C. (1998a). Optimai timing for coiiection of PBPC after glycosylated G-CSF administration. Bone Marrow Transplant, 21, 365-8.

14. Perry, A.R., Watts, M.J., Peniket, A.J., Goidstone, A.H. & Linch, D.C. (1998). Progenitor cell yields are frequently poor in patients with histoiogicaiiy indolent lymphomas especially when mobilized within 6 months of previous chemotherapy. Bone Marrow Transplant, 21,1201 -5.

15. Peniket, A.J., Perry, A.R., Wiiliams, C.D., MacMillan, A., Watts, M.J., Isaacson, P.G., Goldstone, A.H. & Linch, D.C. (1998). A case of EBV-associated lymphoproiiferative disease foiiowing high- dose therapy and CD34-purified autoiogous peripherai blood progenitor cell transplantation. Bone Marrow Transplant, 22, 307-9.

16. Yong, K.L., Watts, M.J., Shaun Thomas, N., Suiiivan, A., ings, S. & Linch, D.C. (1998). Transmigration of CD34+ celis across speciaiized and nonspecialized endothelium requires prior activation by growth factors and is mediated by PECAM-1 (CD31). Blood, 9 1 ,1196-205

17. Watts, M.J., Sullivan, A.M., Leverett, D., Peniket, A.J., Perry, A.R., Wiliiams, C.D., Devereux, S., Goidstone, A.H. & Linch, D.C. (1998). Back-up bone marrow is frequentiy ineffective in patients with poor peripherai-blood stem-cell mobilization. J Clin Oncol, 1 6 ,1554-60.

18. Watts, M.J., Suiiivan, A.M., Ings, S.J., Barlow, M., Devereux, S., Goldstone, A.H. & Linch, D.C. (1998). Storage of PBSC at -80 degrees C. Bone Marrow Transplant, 21,111-2.

19. Devereux, S., Corney, C., Macdonald, C., Watts, M.J., Sullivan, A., Goidstone, A.H., Ward, M., Bank, A. & Linch, D.C. (1998). Feasibility of multidrug resistance (MDR-1) gene transfer in patients undergoing high-dose therapy and peripherai biood stem ceii transpiantation for lymphoma. Gene Ther, 5, 403-8.

14 Chapter 1

Introduction

1.0.1 "Bone marrow: the seedbed of the blood" The fascinating history of the bone marrow has been eruditely summarised by Tavassoli. It includes ancient references to the marrow as a rich and nutritional "dainty" and also the memorable quotation of Sir Thomas Elliot in 1539 who thought "marrowe is more delectable than the bray ne." (Tavassoli, 1980). The account goes on to describe the feeding of bone marrow as therapy for patients with blood dyscrasias in the 1890's and its revival in the 1920's following Whipple's studies on the haematinic effects of various foods and suggests this practice as a naïve forerunner to modern marrow transplantation. The recognition that the bone marrow is the seat of blood cell formation is accredited to Neumann who in 1869 noted from his observations that "the marrow operates continually in the de novo formation of red blood cells". He reasoned that the proliferation he saw in the marrow must result in cells moving into the general circulation and would have no doubt felt fully vindicated by the currently understood capacity for the cellular and proliferative content of the human bone marrow. The total cellular content is estimated at around 2 xIO^^ cells with a daily turnover of 2 xIO^^ red blood cells, 1 x10^° white cells and 4 xIO^^ platelets (Tavassoli, 1980). With a "cell factory" of this magnitude it is unsurprising that the marrow is highly vulnerable to the toxic side-effects of chemotherapy and radiotherapy and is usually the limiting factor to high dose therapy of cancer treatment. The vast majority of haemopoietic cells are morphologically recognisable and mature end-cells but within the marrow there are much smaller sub-populations of multipotent and lineage restricted cells of around 5 xIO® each, and a minority of repopulating stem cells thought to number around 1-10 xIO® (Moore, 1996). In the model shown in Figure 1.1 haemopoiesis is conceived as an overlapping continuum of cell types with

15 permissive elements such as growth factors and cell-cell/stromal mediated interaction preventing apoptosis and with a strong stochastic component at the single cell level during early development (Ogawa, 1993). Whether the repopulating stem cells self-renew or have very high proliferative capacity is debatable although there is some evidence that stem cells may age through successive divisions. This is suggested by the haemopoietic decline following serial bone marrow transplantation (Mauch & Heilman, 1989) and following chemo/radiotherapy insult (Moore, 1992; van Os et al., 1998).

16 Increasing cell maturity

A. Total haemopoietic cells in adult system (Moore 1996)

106-1 5x106 5x109

Stem cells Multipotent Lineage Morphologically Mature marrow, restricte recognisable blood and tissue differentiating cells LTRC cells

CAFC/LTCIC/PAÔ I------1 HPP-CFC / BL-CFC/ CFU-S

CFU-MIx

BFU-E Key to clonogenic assay abbreviations: LTRC = Long Term Repopulating Cell {in vivo B. Clonogenic assay) ; CAFC = Cobblestone Area Forming GM-CFC Cell; LTCIC = Long Term Culture Initiating Cell; Assays PAÔ ("delta assay") = Plastic Adherent cells giving rise to CFC; HPP-CFC = High GM-Clusters Proliferation Potential-Colony Forming Cell; BL- I 1 CFC = Blast cell - Colony Forming Cell; CFU-S = Spleen-Colony Forming Unit {in vivo assay); CFU-mix = Colony Forming Unit - mixed lineage; BFU-E = Burst Forming Unit-Erythroid; 0. Antigen staining of GM-CFC = Granulocyte / Monocyte-Colony Forming Cells; GM-clusters = Granulocyte / CD34+ cells Monocyte clusters; CFU-E = Colony Forming Units- Erythroid.

-I CD34(MY10) I I CDw90(Thy-1)

] CD38(T10)

HLA-DR

CD33 (MY9)

CD45RA

Figure 1.1 Stem cell/progenitor cell hierarchy, clonogenic assays and differentiation antigens associated with haemopoietic cell populations

17 1.0.2 The bone marrow and the bomb The single greatest stimulus for research in bone marrow transplantation followed the detonation of the first atomic bomb in 1945 in New Mexico. The subsequent effects of this weapon on the populations of Hiroshima and Nagasaki provided an urgent need to understand and treat radiation injuries in man. A very early study by Fabricious-Moeller, in 1922 had shown that radiation induced thrombocytopaenic bleeding could be prevented in guinea pigs by shielding their legs (Fabricious-Moeller, 1922). This work was developed by Jakobson in 1949 who showed that mice could be protected from Sfyk&n fatal radiation induced marrow aplasia by shielding theJJakobson et al., 1949), an important haemopoietic organ in these animals. He then went on in 1951 to show that lethally irradiated mice could be protected by shielding the femur or rescued by the intra-peritoneal infusion of spleen cells (Jakobson eta!., 1951). In the same year, Lorenz showed that lethally irradiated mice and guinea pigs could be rescued with syngeneic bone marrow (Lorenz et a!., 1951). Subsequent investigators showed that the donor cells repopulated the marrow and the term "radiation chimera" was introduced (Ford et a!., 1956). In 1958 six physicists developed radiation induced marrow aplasia following a nuclear accident in former Yugoslavia. One of the physicists died but Mathé described the successful infusion of allogeneic bone marrow and transient engraftment determined by red cell antigen studies in the remaining five. This procedure may have provided temporary support until the subsequent autologous recovery in this group (Mathé et a!., 1959).

1.0.3 Early marrow transplantation One of the most important pioneers in the field was E. Donnod! Thomas working first in Goopertown, New York and then Seattle, Washington. He had carefully set out experiments using dogs to define the radiation doses, marrow harvesting procedures and methods of support required to transplant humans. His report in 1960 on the outcome of clinical transplantation of 19 patients with

18 allogeneic marrow after irradiation proved to be sadly typical for this era (Thomas & Ferrebee, 1960). Thirteen died without engraftment and two had autogenous recovery. The remaining four received transplants from identical twins and engrafted although all relapsed within three months. The specific problems were ultimately identified as rejection of donor cells by the host and conversely by rejection symptoms in the recipient mediated by immuno­ competent donor cells. (Graft versus host disease, GvHD). These problems, combined with infection led to the depressing statistic of only three successful prolonged allografts out of 417 procedures reported by 1966 (Pegg, 1966)

1.0.4 Improvements in survival It is doubtful whether allogeneic transplantation could have continued without the discovery of the HLA histocompatibility system, largely attributable to the work of a Nobel Prizewinner, Jean Daussett (Daussett et ai., 1965) and Terasaki's microassay which made HLA typing practical (Terasaki & McLelland, 1964). This led to the use of transplants using marrow from HLA- matched siblings in the 1970's and from a growing pool of non-sibling HLA- matched donors from donor panels such as the Anthony Nolan panel in the UK. Treatments to suppress GvHD were also evolved including the use of steroids, methotrexate, anti-thymocyte globulin and in 1980 cyclosporin (Powles et ai., 1980). Improvements in supportive care such as new antibiotics, antifungals, antivirals, blood and platelet support and improved venous access with the introduction of the Hickman line (Hickman et ai., 1979) combined to reduce the toxicity of the procedure (Bortin et ai., 1992) and by the 1980's led to a worldwide expansion in transplant activity (Barrett, 1991). At this time registries such as the International Bone Marrow Transplant Registry and the European to Bone Marrow Transplant Registry were set up^provide overviews of activity in the field with statistical analysis of results.

19 1.0.5 The growth of autologous transplantation Early animal experiments had shown sporadic cures in solid tumours following high dose chemo/radiotherapy and autologous bone marrow rescue (McFarlane et al., 1959). With the improvements in allogeneic transplant toxicity, refined conditioning regimens and cryopreservation techniques came an exponential increase in the application of high dose therapy and autografts for lymphoma, leukaemia and solid tumours (Linch & Goldstone, 1984). Recently published European data shows this to be a growing trend and from the period 1983 to 1994 autologous grafts increased from 35% to 65% (Gratwohl & Schmitz, 1996). A further impetus to the more recent growth of autografts has been the introduction of harvesting haemopoietic progenitor cells (HPC) from peripheral blood as an alternative to marrow. This was made of feasible by the use^ growth factors introduced in the late 1980's to improve haematological recovery and which could also mobilize sufficient haemopoietic progenitors into the blood for transplantation (Duhrsen et al., 1988; Socinski et al., 1988). The convenience and reduced toxicity of this procedure has resulted in it gradually replacing autologous marrow autografting (Gratwohl & Schmitz, 1996).

1.0.6 Stem cell sources The direct collection of stem cells from the bone marrow is not the only method for transplantation and the term "haemopoietic progenitor cell" (HPC) is commonly used to embrace haemopoietic support from various sources which can be used in different clinical situations (Table 1.1). In addition to bone marrow and peripheral blood stem cells (PBSC) for autologous or allogeneic use, placental and umbilical cord blood is another source of allogeneic HPC which is being HLA-typed and banked for transplantation (Gluckman, 1994). It has been used successfully in children (Broxmeyer et al., 1990; Gluckman, 1996) but has yet to overcome concerns about graft size for adult recipients (Gluckman et al., 1998). PBSC are in use in the allogeneic setting and raise

20 specific considerations of the mobilization of normal donors (Anderlini et al., 1998) and the impact on graft versus leukaemia and graft versus host disease. These issues have been recently reviewed by Russell, Gratwohl and Schmitz including a summary of the EBMT consensus statement on the use of allogeneic PBSCT (Gratwohl & Schmitz, 1996).

Table 1.1 Current Haemopoietic Progenitor Cell (HPC) transplantation Aiiogeneic Autoiogous indication Haematological malignancy, Haematological and non- Aplastic anaemia, congenital haematologlcal malignancy bone marrow disorders, ?Gene therapy immune deficiency ?Auto-immune diseases

HPC source Bone marrow Bone marrow Peripheral blood Peripheral blood Cord blood Family or unrelated, HLA matched or mis-matched

Toxicity Infective risk high for months or Infective risk mainly in early post­ years, GvHD. Treatment- transplant. Treatment-related related mortality 5-35% mortality <5% according to case-mix

1.0.7 "Graft engineering" The last decade has seen the development of extensive cell manipulation techniques or "graft engineering" in the transplantation field. An early example in the 1980's was the depletion of T-cells from donor marrow to reduce GvHD. This effect was achieved but at the cost of increased problems with rejection and leukaemic relapse (Poynton, 1988). This implied a beneficial graft-versus- leukaemia effect associated with GvHD. Controlled stimulation of this activity using cytokines such as interleukin-2 following autologous and allogeneic transplants (Soiffer et ai., 1992) and the infusion of donor lymphocytes are active areas of interest (Mackinnon et ai., 1995). The recognition that the CD34 antigen identifies primitive haemopoietic cells including cells capable of long

21 term engraftment (Dunbar et al., 1995) and the availability of clinical selection devices (Berenson eta!., 1991 ; Hohaus et a!., 1997; Richel et a!., 1996) has facilitated a new range of areas for transplantation. These include T-cell depletion in the allogeneic setting (Collins et a!., 1997) and possibly CD3» s

1.0.8 Transpiantation and this thesis The work in this thesis began with the development of the peripheral blood stem cell program, including CD34+ cell purification at University College London Hospitals and relates to the issues surrounding the mobilization and collection of sufficient autologous haemopoietic progenitor cells in pre-treated patients with lymphoma, myeloma and solid tumours to support high dose therapy as whole or CD34+ cell purified products. The mobilization of normal subjects and the problem of "poor mobiiizers" are also addressed.

As previously outlined, the use of high dose chemo/radiotherapy requiring autologous stem cell support is being increasingly explored in a range of malignant disorders (Attal et ai., 1996; Bezwoda et ai., 1995; Haioun et ai., 1997; Linch et ai., 1993; Philip et ai., 1995). In 1995 2,702 such procedures were reported to the European Blood and Marrow Transplant Registry and the use of autologous bone marrow stem cells as a source of HPC has largely been replaced by the use of peripheral blood stem cells (PBSC). This has the

22 advantage that harvesting does not require a general anaesthetic and that haematological regeneration, especially of the platelet lineage, is more rapid (Beyer et al., 1995; Gianni et a!., 1989; Schmitz et al., 1996). This results in less time in hospital and a significant saving of resources (Schmitz et al., 1996; Smith et al., 1997). At University College London Hospitals the transition from autologous bone marrow transplantation (ABMT) to PBSCT after BEAM chemotherapy has been associated with a reduction in the number of days on which platelets were transfused from approximately 16 to 3 days and red cell units from 8 to 4 units (Jones et al., 1994). A further potential advantage of PBSC collections is that they may contain less contaminating malignant cells than the bone marrow (Ross etal., 1993).

1.0.9 Circulating stem cells Although in the steady state the majority of haemopoietic stem cells reside in the bone marrow, it has long been known that haemopoietic stem cells also circulate in the blood. In 1951, Brecher and Cronkite (Brecher & Cronkite, 1951) provided evidence of circulating stem cells in rats. In 1962, Goodman and Hodgson (Goodman & Hodgson, 1962) showed that pooled blood isolated from donor mice could restore haemopoiesis in lethally irradiated parent mice and postulated the presence of circulating stem cells. In subsequent years, similar data were obtained with larger animal species such as dogs (Epstein et al., 1966). The difficulties of obtaining a sufficient blood cell dose was highlighted in a study by Storb who showed that 2 xIO® MNC collected by leukapheresis over 7 days were required to engraft lethally irradiated dogs (Storb etal., 1967) which is around 10-fold higher than the equivalent marrow cell dose in this model. An early human attempt to use normal steady state peripheral blood as a source of stem cells for transplantation in identical twins failed (Hershko et al., 1979) and although sporadic successes with this approach were eventually reported in the autologous setting (Kessinger et al., 1988), a subsequent randomised trial showed that steady-state peripheral

23 blood cells did not accelerate marrow recovery following high dose therapy (Lobo etal., 1991). In 1976, Richman (Richman et a!., 1976) showed that there was an increase in the number of circulating progenitor (colony forming) cells during the recovery phase from chemotherapy. Several groups successfully used PBSC collected after standard dose chemotherapy to facilitate subsequent high dose therapy (Gianni et a!., 1989; Juttner et a!., 1985; Korbling et a!., 1986; Reiffers et ai., 1986). It was apparent that this was a labour intensive process and multiple leukapherescs had to be performed to ensure engraftment, particularly of the platelet lineage. Engraftment was, however often rapid compared to that from bone marrow. In 1988, it was reported that administration of granulocyte colony stimulating factor (G-CSF) and granulocyte-macrophage colony- stimulating factor (GM-CSF) resulted in mobilization of progenitor cells into the circulation (Duhrsen et a!., 1988; Socinski et al., 1988). This was shortly followed by a study in which G-CSF mobilized blood was shown to accelerate platelet recovery when used to augment autologous marrow after high dose therapy (Sheridan et al., 1992). Mobilization strategies employing growth factors were rapidly developed and in many circumstances, only one leukapheresis procedure is necessary to ensure engraftment. In some centres, mobilized PBSC were initially given with bone marrow, but PBSC are now nearly always given alone.

1.1.0 Possible mechanisms of HPC mobilization Haemopoiesis within the marrow cavity takes place in highly organised structures (Figure 1.2) in which developing clones are supported and compartmentalised by specialized sinus macrophages (reticular cells). The maturing cells are separated from blood circulating in the vascular sinuses by an endothelial monolayer and discontinuous basement membrane (marrow sinus wall). Cell adhesion molecules (CAMs) of the CD44 family, integrins, selectins and immunoglobulin superfamilies are expressed on haemopoietic

24 progenitors and many of their ligands are expressed constitu^s^tively on the marrow stromal cell mediated environment (Long, 1992; Simmons et al, 1994; Tavassoli & Hardy, 1990). Progenitor cells are retained in the marrow by a combination of cell-cell and extracellular matrix interactions with adhesive proteins such as fibrinectin, vitronectin, collagen, thrombospondin, fibrinogen and von Willebrand factor (Ruoslahti & Pierschbacher, 1987). These interactions are reduced as the cells mature with a concomitant increase in flexibility and/or reduction in size for the final controlled egress of mature blood cells across the sinus wall to join the circulation (Figure 1.2) (Jandl, 1991).

25 Capillaries Vascular sinusoid Bone trabeculae

Circulating blood

Artery

Sinusoidal membrane

BONE Fat cells (Adipocytes)

Egress of mature blood cells across sinus endothelium Neutrophil into circulating blood diapedis Circulating blood *o o ^ ^ ^

Red cell extruding nucleus

Endothelial cells Adipocyte / Platelet release from megakaryocyte

Adventitial macrophage

Figure 1.2 Diagramatic cross-section of mammalian long bone

26 The precise mechanisms involved for many of these interactions in homing, lodgment, retention and proliferation of progenitors are unclear. However, the observation that a number of haemopoietic growth factors, cytotoxic agents and a few cytokines can mobilize MPC from the marrow and that the mobilized CD34+ cells have a lower expression of CAMs such as the fibrinectin receptors VLA-1 and VLA-4 (very-late antigens-1 and 4) and the leukocyte function- associated antigens LFA-1 and LFA-2 (Mielcarek & Torok-Storb, 1997) suggests a common mechanism involving the perturbation of progenitor binding to the marrow stroma. The importance of the p-1 integrin VLA-4 (CD49) in the process was shown directly by Papayannopolou et al who showed that the administration of a function-blocking anti-VLA-4 antibody to mice or non­ human primates induced MPC mobilization (Papayannopoulou & Nakamoto, 1993). The same group could not however demonstrate mobilization in this study with a blocking antibody to the p-2 integrin LFA-1 (GDI8). Fibbe's group in the Netherlands adopted a similar approach to the study of mobilization using blocking antibodies but with a different model based on the cytokine interleukin-8 (IL-8), which they had shown induced the rapid mobilization of MPC in mice and rhesus monkeys. In a series of experiments summarised in their most recent report (Pruijt et a/., 1998) they showed that blocking antibodies to LFA-1 or its ligand ICAM-1 inhibited IL-8 induced mobilization and significantly reduced MPC mobilization with a second cytokine, interleukin-1 (IL-1). They were also able to show that anti-LFA-1 was capable of blocking the mobilization of very early MPC as transplantation with MNCs from such animals were poorly radioprotective (25% survival) compared to 86% survival using MNG from control IL-8 mobilized mice. The central involvement of LFA-1 in this mobilization system implies a role for mature accessory cells which are activated via this ligand such as neutrophils and monocytes. The observation in this study that IL-8 induced mobilization was also inhibited by a blocking antibody to gelatinase-B supports this view. This metalloproteinase is usually associated with neutrophil granules and collagen

27 breakdown and has been shown to have a role in neutrophil migration across basement membrane (Delclaux et al., 1996). The involvement of metalloproteinases in mobilization may also involve more immature cells as has also been suggested in a recent study which showed that both gelatinase- A and B is produced by CD34+ cells from mobilized blood but is absent in CD34+ cells from bone marrow (Janowska-Wieczorek et a!., 1997). These findings may hold an important clue towards identifying a common mobilization mechanism and suggests that if gelatinase release is required myeloid cells may provide an important contribution to the process. An important study in this regard was reported by Liu who examined the role of the G-CSF receptor (G-CSFR) as a common mechanism for HPC mobilization using a G-CSFR deficient ("knock-out") mouse model. The deficient mouse lacks mature myeloid cells but not HPC in the marrow. In contrast to the normal or "wild-type" mouse it did not mobilize blood HPC with G-CSF, cyclophosphamide or IL-8 but did possess the ability to regenerate marrow HPC as shown by increased marrow HPC activity after cyclophosphamide even compared to the wild-type. Interestingly, HPC mobilization in the deficient mouse was almost normal with flt-3 ligand suggesting an alternative mechanism for this growth factor (Liu et a!., 1997b). The deficient mouse had less than half the number of mature granulocytes in blood and marrow than the wild type and the G-CSF receptor may also be expressed on marrow stromal cells. These observations could not therefore distinguish whether the mobilization lesion was associated with these cells or the abnormal HPC of the deficient mouse. This question was addressed in a series of cross­ transplantation experiments (Table 1.2) , (Liu eta!., 1997a). No HPC were mobilized with cyclophosphamide in the G-CSFR knockout mouse or in the wild-type mouse transplanted with deficient HPC (normal stroma, impaired neutrophil maturation). Mobilization was evident however in the deficient mouse transplanted with wild-type HPC (deficient stroma, restored neutrophil maturation). Of particular note was the finding that deficient HPC were

28 capable of mobilization if both types of HPC were transplanted into a wild type mouse. It is tempting to surmise from this data that differentiated myeloid cells are required since they are present in each instance where mobilization was successful. However, Liu has observed that an occasional G-CSFR deficient mouse has normal blood and marrow mature neutrophil counts and still fails to mobilize HPC. These neutrophils have normal morphology, myeloperoxidase and CD11b staining and are capable of migration but crucially their responsiveness to IL-8 has not been reported (Liu et al., 1997b). The role of normal neutrophils deserves further scrutiny however in view of mobilization inhibition by LFA-1 and gelatinase blocking antibodies and clinical evidence such as the study by Sekhsaria which showed very poor mobilization with G- CSF in 18 patients with a hereditary disorder of neutrophils. Chronic Granulomatous Disease (Sekhsaria et a!., 1996). It is unclear at present how the alternative mechanism proposed by Lîu that G-CSFR bearing HPC are required directly for the induction of a general mobilization process might operate.

Table 1.2 Cross-transplantation studies and cyclophosphamide induced HPC mobilization in the G-CSFR knockout mouse (Liu et al 1997)

Mouse type : DONOR RECIPIENT Mobilization Neutrophil (HPC) (STROMA) maturation WT +++ Yes KO - No KO>WT KO WT - No WT>KO WT KO ++ Yes WT and KO>WT KO and WT WT ++ Yes (WT and KO HPC) WT =wild-type: KO =G-CSFR knockout

An alternative mechanism for mobilization may involve the haemopoietic progenitor c-kit receptor which is downmodulated on marrow HPC before egress into the blood and the c-kit ligand, stem cell factor (SCF) which occurs both as a soluble form and membrane bound on marrow stromal cells. In a study by Cynshi, mice with Steel or WW mutations (that lack the ability to produce soluble SCF or have a defective c-kit receptor respectively) were

29 shown to mobilize very poorly to G-CSF (Cynshi etal., 1991). In clinical studies higher CD34+ cells yields have been shown to be associated with low c-kit expression on these cells and was shown by To's group for example, to be highly correlated (p<0.005) despite the use of six different mobilization protocols (To etal., 1997). Recently an important role for c-kit signalling involving VLA-4A/CAM-1 binding of HPC to marrow stroma has been proposed by Papayannopoloulou's group at the University of Washington in Seattle (Papayannopoulou et a!., 1998). They performed mobilization experiments with blocking antibodies to VLA-4 (expressed by HPC) and VCAM-1 (stromal cell ligand for VLA-4) in mutant mice. They showed that HPC release from the marrow with both antibodies did not rely on the presence of a functional G-CSF, IL-7 or IL-3a receptor. In addition the "motheaten mouse" mutant, meVme'' which has a defect in c-kit tyrosine kinase signalling (SHP-1 knockout) and high steady state circulating progenitors also mobilized significantly above baseline with both antibodies. In contrast, WAAT mutations (impaired c-kit signalling but normal protein expression and c-kit receptor/kit ligand binding) mobilized very poorly with both antibodies. The defective response in the \N/W mice was restored however, 8 weeks after transplantation with HPC from littermate controls with normal c-kit receptors. Steel-Dickie mice (soluble c-kit ligand, but not stromal-bound) mobilized if VLA-4 was blocked but not with the VCAM-1 antibody (Table 1.3). The group proposed that a functional c-kit receptor is required for mobilization via VLA-4A/CAM-1 Interactions which override c-kit receptor dependent signals required for downmodulation of c-kit and VLA-4 on HPC and release of binding through other CAM interactions. The reason for mobilization with the VLA-4 but not the stromal VCAM-1 blocking antibody in the Steele-Dickie mouse is not clear. Also of note in this study was the observation that WA/\T mice were capable of late but eventually normal HPC mobilization levels induced by the cytokine flt-3 ligand. The flt-3 receptor, like c-kit is expressed on HPC and also signals through tyrosine kinase when bound by its ligand and on the basis of

30 both this study and mobilization of the G-CSFR deficient mouse discussed previously, may operate through a different mechanism.

Table 1.3 Moblllzation response of mutant mice to blocking antibodies against VLA-4 (expressed on HPC) or to its stromal cell ligand VCAM-1

Mouse mutant VLA-4 blocked VCAM-1 blocked (on HPC) (on stromal cells)

G-CSFR deficient +++ + 4 - 4 - IL-7 deficient +++ +++ IL-3a deficient +++ +++ Steel-Dickie SI/SI'^ +++ (lacks stomal cell-bound c-kit ligand) W/W" (lacks functional c-kit on HPC) meVme'' "Moth-eaten mouse" ++++ ++++ (constitutively activated c-kit)

Taken together all of the studies described suggest a two-phase general mechanism of mobilization involving an initial triggering event and second multi-step amplification and controlled release phase. Triggering events might act alone and lead to rapid (within minutes) transient mobilization such as the gelatinase-B dependent IL-8 mechanism discussed earlier. Longer-term (within days) amplification of the response may be regulated by signalling pathways involving cross talk between intergrins and cytokines or intergrins and other CAMs.

1.1.1 Monitoring of PBSC Coiiection and Transpiantation A variety of assays have been developed for measuring different but often overlapping haemopoietic cell populations at various stages of the differentiation pathway (Figure 1.2). The in vitro assays are all surrogate markers for engraftment potential but several have proved useful in clinical practice.

31 1.1.2 Clonogenic functional assays The ultimate assay for stem cell function are in vivo assays for long-term repopulating cells (LTRC) which can assess the potential for durable, multi­ lineage engraftment. Such assays for human haemopoiesis have been described in chimaeric immunodeficient animals such as mice and sheep (Civin et ai., 1996; Srour et ai., 1993; Wang et ai., 1998) but are not suitable for clinical use. Clonogenic techniques, which are used in vitro for measurement of the most primitive progenitors, include long term culture initiating cells (LTCIC), cobblestone-forming cells (GAFC), plastic adherence (P-delta) and high performance proliferation cells (HPP). These assays provide the closest in vitro surrogate measurement of LTRC and durable engraftment (Sutherland et ai., 1990), but are difficult to standardise, time-consuming and impractical for routine use. In contrast, simple assays are available for relatively late progenitor cells such as granulocyte-monocyte-colony forming-cells (GM-CFC) which appear to be closely related to those cells giving rise of rapid engraftment (Linch & Watts, 1996). Jones and colleagues clearly demonstrated in the mouse that cells giving rise to short term haemopoietic recovery were distinct from those giving rise to long term repopulation (Jones et ai., 1990). This view has been recently challenged by more recent studies in which purified HPC with the most primitive phenotype and little short-term CFC growth in v/fm (lineage negative, Sca-1+ Thyl .1'°^"®^) were shown to mediate both early and late engraftment in mice (Nibley & Spangrude, 1998; Uchida et ai., 1994). In humans, autografts with highly purified lineage negative, CD34+, Thy-1+ progenitors have been shown to mediate engraftment after high dose therapy although this is delayed compared to when the total CD34+ cell population are infused. (Tricot et ai., 1998). Also in the human setting Kaizer and colleagues found that removal of GM-CFC, but not earlier cells, from infused marrow by incubation by 4-hydroxy-cyclophosphamide resulted in delayed haematological recovery (Kaizer et ai., 1985) and Robertson et ai found that purging with an anti-CD33 antibody which removed GM-CFC but not

32 LTCIC resulted in a similar delay (Robertson et al., 1992). In the majority of clinical situations in which PBSC rescue is used, the high dose therapy employed causes only temporary myelosuppression and the autograft is only required to provide relatively short-term support. GM-CFC assays thus appear to be a fully appropriate means of graft assessment and standardisation. Even when truly myeloablative conditioning is used, such assays also seem to be appropriate as late graft failure after prompt early recovery is exceedingly rare in the autologous setting. GM-CFC assays have their problems however. There is often marked variability between centres, the coefficient of variation of the assay is often wide, and regular quality assurance exercises are difficult. Fresh stem cell preparations are ideally required for quality control purposes because if frozen aliquots are used, this introduces the large potential variability of freeze-thaw recoveries.

1.1.3 Flow cytometric CD34+ cell measurement A simpler measurement of primitive cells is the flow cytometric assessment of the number of cells expressing the CD34 antigen. Measurement of such low proportions of cells by immuno-fluorescence is technically demanding but providing well-standardised staining and analysis protocols are used, it is possible to obtain low coefficients of variation and good concordance between laboratories (Siena etal., 1991; Sovalat etal., 1993). The average CD34+ cell is considerably more mature than the average GM-CFC and might not be expected to be as predictive of engraftment, but the advantages of precision and rapidity counterbalance this possibility and in practice there is little to choose between GM-CFC and CD34+ cell assays for predicting graft adequacy. It is our practice to go ahead with high dose therapy on the basis of CD34+ cell measurements unless the results are marginal in which case the transplant is delayed until the GM-CFC data is available. It has been suggested the measurement of CD34+ subsets may be of value in predicting engraftment potential. The most primitive cells within the CD34+ cell

33 compartment which have been shown to give rise to multi-lineage haemopoiesis in vitro and in immunocompromised animal hosts are CD34+CD38-, CD34+HLA-DR- or CD34+Thy-14- (Craig et al., 1993b; Huang & Terstappen, 1994; Rusten et al., 1994; Terstappen & Huang, 1994). This fraction of CD34+ cells are also thought to be small and quiescent (G0/G1) cells and responsive to growth factors such as SCF, Mip-1-a and flt3 (Miller & Eaves, 1997; Wagner et al., 1995). The proportions of CD34-H cells which are CD38- or HLA-DR- are consistently below 5% which compares to around 25% of CD34+ cells which express Thy-1. The highest proportion of CD34+Thy-In- cells in PBSC were observed during early apheresis collections by Haas and Murray who suggested that these harvests would be most likely to provide long term engraftment potential (Haas et al., 1994b; Murray etal., 1995). However, the co-expression of Thy-1 by mobilized CD34n- is highly variable between individuals and Humeau has demonstrated the heterogeneity of this subset by its lack of correlation with other important phenotypes such as CD34+CD38- (Humeau et al., 1996). The use of CD34n- cell subset measurement to predict engraftment has been evaluated by Dercksen who demonstrated early neutrophil engraftment more significantly with the numbers of mobilized CD34+CD33- re-infused and of platelet engraftment with the number of CD34+ cells co-expressing the platelet glycoprotein marker CD41 than with total CD34-H cell numbers. Dual CD344- and adhesion protein L-selectin+ cells were also shown to predict early platelet engraftment and this was subsequently confirmed in a report by Wanatabe (Dercksen et al., 1995a; Dercksen et al., 1995b; Watanabe et al., 1998). The practical problem with the quantitation of CD34-H subsets in addition to the very low numbers involved is that they are lost or acquired in a gradual and overlapping sequence (Figure 1.1) so that an applied discrimination gate is somewhat subjective. It is difficult to see how such sub-set measurements can be standardised between centres until the problems of standardising total GD34-I- cell counts are resolved.

34 1.1.4 Progenitor threshold requirements With autologous bone marrow transplants, there is considerable controversy concerning progenitor cell requirements (Gorin, 1986; Spitzer etal., 1980) but in the context of PBSCT, there is widespread agreement that there are readily definable progenitor cell thresholds (Demirer & Bensinger, 1995). In our experience, delayed neutrophil recovery is very rare but delayed platelet recovery is not infrequent and closely related to the dose of progenitor cells infused. If more than 3.5 x 10®/kg CD34+ cells or 3.5 xIOVkg GM-CFC are infused the risk of delayed recovery is negligible. Values between 1-3.5 xIOVkg and 1-3.5 xIOVkg CD34+ cells and GM-CFC respectively give a 5-15% risk of delayed platelet recovery whereas if less than 1 x10®/kg CD34+ cells are re-infused the risk of delayed platelet recovery is in the order of 40% and some patients remain severely thrombocytopaenic indefinitely (Watts et al., 1998b). The use of bone marrow in these patients still results in protracted engraftment in the majority indicating that in most patients the poor mobilization is a reflection of poor quality marrow rather than a specific defect in mobilization (Dreger et al., 1995). It must also be borne in mind that in the heavily pre­ treated patient enumeration of progenitor cells alone may not give a complete reflection of the quality of the harvested product. There is some evidence that the threshold requirements are higher when multiple aphereses rather than a single collection are required (Watts etal., 1997^ Weaver et ms).

1.1.5 Apheresis timing CD34+ cell measurement may also be useful in determining when to commence leukapheresis. Using a standardised regimen of cyclophosphamide 1.5g/m^ and G-CSF (Jones et al., 1994) there is a leukocyte nadir on day 8 followed by a rapid rise in the WBC after this time. The optimal time to start harvesting has been identified as the first day on which the whole blood cell count exceeds 5.0 x107l (Watts et al., 1995b) which accords closely with the data of Haas (Haas et al., 1995) who with a different chemotherapy

35 regimen found peak circulating progenitor cell levels commencing when the post nadir WBC reached 4.75 x10Vl. In these well-standardised situations, progenitor cell measurements are not required for harvest timing but with other regimens or in very heavily pre-treated patients, peripheral blood CD34+ cell measurements are advisable. Schwella et al (Schwella et a/., 1995) for example, showed a harvest day peripheral blood GD34+ cell count of 40 x1 OVL to be highly predictive (r=0.94, p<0.0001) for a single apheresis CD34+ cell dose greater than 2.5 xIOVkg. Elliot efa/have described a highly significant correlation between both harvest day (r=0.95, p<0.001) and importantly the preceding day (r=0.85, p <0.001) peripheral blood GD34+ cell counts and subsequent harvest GD34+ cell dose/kg collected. This allowed efficient apheresis planning with the construction of a simple table predictive of the GD34+ cell dose/kg likely at a single apheresis. A collection exceeding 1 xIOVkg GD34+ cells, for example, was obtained in 93% of patients with a preceding day peripheral blood GD34+ cell count of > 10 x10®/L (Elliott et a!., 1996). This study has been recently updated and the utility of this approach confimed by Armitage e ta l (Armitage etal., 1997).

1.1.6 Optimisation of Progenitor Cell Mobilization Selection of a mobilization regimen is a compromise between obtaining the maximum progenitor cell yields possible, and minimising the morbidity and costs of the procedure. In the allogeneic setting, the normal donor can obviously not be given cytotoxic drugs, and cytokines alone must be used. In the autologous setting where the patient will have received prior chemotherapy adequate yields can often be obtained with cytokine alone but it is more usual to give the cytokines after chemotherapy as a number of studies have shown that the combination of chemotherapy and cytokines gives better mobilization than either alone (Kobayashi et al., 1995; Mohle et al., 1994). The chemotherapy can either be specific for the mobilization process or can be a cycle of therapy usually used in the treatment of the disease. A wide variety of

36 cytotoxic drugs and their combinations cause progenitor cell mobilization although it is probably best to avoid the most stem cell toxic agents such as the nitrosoureas and alkylating agents. An exception to the alkylating agents is cyclophosphamide, which is markedly 'stem cell sparing'. One of the first mobilizing regimens was high dose cyclophosphamide, 7g/m^ (Gianni et al., 1989). This has the advantage of considerable anti-tumour activity but is very myelosuppressive, has marked non-haematological toxicities, and can cause considerable morbidity and frequent requirements for hospitalisation. In our institution we have considerable experience with cyclophosphamide 1.5g/m^ which is effective, well tolerated and has an associated hospitalisation rate of only 5% (Watts at a!., 1997c). There have been few comparative studies of different mobilization chemotherapy regimens and no randomised trials. Several studies have suggested that cyclophosphamide 7g/m^ is more effective than 4g/m^ albeit with greater toxicity (Rosenfeld at a!., 1995; Rowlings at al., 1992). In a case-match study we have found the lymphoma salvage regimen ESHAP (Velasquez at al., 1994) to be at least as effective as cyclophosphamide 1.5g/m^, and to have the benefit of greater anti-lymphoma activity (Watts at al., 1996b). In addition, ESHAP gives rise to a significantly higher CD34+ cell percentage in the apheresis product which is an advantage if CD34+ cell purification is planned.

Both G-CSF and GM-GSF are effective progenitor cell mobilizers either when used alone or after chemotherapy (Peters at al., 1993). G-CSF is more widely used however because of its superior toxicity profile (Demuynck at al., 1995). For the majority of patients, adequate mobilization can be achieved with relatively low dose G-CSF e.g. 5|ig/kg/day. There is however, a marked dose response effect and higher doses undoubtedly give better average yields (Hoglund etal., 1996; Roberts etal., 1995; Sekhsaria at al., 1996; Waller at al., 1996).

37 In the study by Hoglund et al (Hoglund of a/., 1996) it was decided that 10p,g/kg/day sc was the maximum tolerated dose for volunteer donors, but some centres use more than this (Waller et a/., 1996). In the autologous setting higher doses are generally tolerated but the cost issues are considerable. Other cytokines when used in combination with G-CSF can also augment mobilization. In a study described by Geissler for example, low progenitor numbers were collected in 6 patients with pretreated lymphoma who were mobilized with G-CSF at 5^ig/kg/day for 5 days. After a treatment-free interval of at least a month, these patients were "primed" with IL-3 at 5fxg/kg/day for 7 days followed by an identical G-CSF mobilization protocol. Circulating GM-CFC were typically double that seen at first mobilization with G-CSF alone and leukapheresis following the IL-3/G-CSF combination achieved progenitor yields above the minimum target of 2 x10®/kg in 3 of 6 patients (Geissler et a!., 1996). Thrombopoietin (IRQ) is the ligand for c-mpi and is believed to be the most important regulator of megakaryocytopoiesis. It has also been shown to mobilize multilineage HPC in clinical studies (Rasko etal., 1997). It has also been given post chemotherapy in doses of 0.03, 0.1, 0.3, 1, 3 and 5|Lig/kg concurrently with G-CSF at 5^ig/kg in a large randomised trial of solid tumour patients (Basser et al., 1997). In this study the fold increase in blood progenitors from baseline to day 15 after chemotherapy was shown to be significantly enhanced in the 25 patients receiving G-CSF and doses of TRO between 0.3 to 5^ig/kg compared to the 10 patients receiving placebo and G- CSF alone (median GM-CFC fold increase 191 vs 3, R=0.002, CD34+ cell fold increase 20 vs 1, R=0.03). It is notable that IL-3 and TRO were used with low doses of G-CSF in these studies however and it is likely that similar progenitor increments could have been achieved by increasing the G-CSF dose. In contrast, in a clinical trial of breast cancer patients reported by Basser G-CSF was given at lO^ig/kg/day sc for 7 days alone or with concurrent SCF at doses of 5, 10 or 15|ig/day sc. RBSC were collected on days 5, 6 and 7 of G-CSF administration. The increase in progenitor yields in the SCF cohorts was

38 disappointing and the only significant trend was in the group receiving 10|Lig/kg SCF (p=0.13 and p=0.07 for GD34+ cells and GM-CFC respectively) (Basser et a/., 1998). During the course of the study however peripheral blood colony assays had been performed for several days after the cessation of growth factors and it was noted that despite a falling white cell count CFC numbers remained high for a further 4 days or so in the SCF cohorts compared to those receiving G-CSF alone (Begley etal., 1997). An additional group of patients was therefore incorporated into the study that were given ten days of SCF at 10M-g/kg/day, the first three days prior to G-CSF administration. This SCF pre­ treatment group had significantly higher peak blood progenitor counts and higher progenitor yields at apheresis (Table 1.4). Furthermore the authors suggesting qualitative differences based on an increased recloning ability in vitro of primary GM colonies from the SCF pre-treated patients (90% secondary colony growth compared to 30% in the patients receiving G-CSF only) (Begley et ai., 1997).

Table 1.4 Mobilization of breast cancer patients with G-CSF at 10|ig/kg/day only or given SCF at 10|xg for three days prior to G-CSF then concomitantly for the remaining 7 days (Basser et ai 1998)

Mobilization Peak blood progenitor Apheresis yields protocol counts (3 harvests days 5,6,7) CD34+ ceils GM-CFC CD34+ cells GM-CFC x10^/L x1(f/L X10P/kg XKF/kg G-CSF only 110 (64-167) 30(14-18) 11.4 (1.9-23.7) 40.2 (8.3-91.6) Pre-treatment 341 (255-417) 155 (89-202) 24.5 (4.5-52.1) 97.5 (28.7-205.1) SCF and G-CSF

P value < 0.001 < 0.001 < 0.001 < 0.001

Major issues raised by these studies are how optimal dosing might be achieved, increased cost, whether the addition of another growth factor can achieve more than merely increasing the dose of G-CSF, and if so whether any increased toxicity is acceptable. A second approach to increase progenitor

39 yields might be to "boost" circulating progenitors immediately prior to apheresis. Single injections of the chemokines IL-8 and MIP-1 a have been shown to transiently boost G-CSF induced progenitor cell levels in mice (Lord etal., 1995; Pruijt et a!., 1998) but initial experience of MIP-1 a in patients has not been encouraging. (Broxmeyer et a!., 1998; Watts et a!., 1996a) As more potent ( and more expensive ) cytokine doses and combinations are defined, it is important to identify those patients who would most benefit from them. In a series of pre-treated patients with high grade lymphoma and Hodgkin's disease at our institution approximately 25% of patients failed to achieve the minimum harvest yield of 1 x 10Vkg CD34+ cells with a single apheresis and about 8% even after combining several apheresis (Watts et a!., 1997c; Watts et a!., 1998b). The failure rate is in general lower in solid tumours but higher in low-grade lymphomas and myeloma. A number of factors have been identified as being associated with poor mobilization including extensive prior chemotherapy (Watts et a!., 1998b) especially alkylating agents (Ketterer et a!., 1998; Morton et a!., 1997) use of the Dexa BEAM / mini BEAM regimen (Dreger etal., 1995; Watts etal., 1997c; Weaver etal., 1998), prior radiotherapy (Haas etal., 1994a; Watts etal., 1997c) and the time from last chemotherapy to mobilization (Perry etal., 1998). Besinger has reported advanced age to be associated with poor mobilization (Bensinger et al., 1995) and Koumakis and Anderlini identified significantly worse results in patients over 50 years old (Anderlini etal., 1997b; Koumakis et al., 1996). In a series of 101 patients from our institution however, 8 out of the poorest 29 mobilizers defined as <1x10®/kg CD34+ cells with the first apheresis, had no adverse risk factors (Watts et al., 1997c). It may be therefore, that many poor mobilizers can only be identified after failing a standard mobilization protocol. This is perhaps not too surprising as even with normal donors there may inter-individual variation in mobilization as wide as in pre-treated cancer patients, with a subset of poor mobilizers (Bishop etal., 1997; Brown etal., 1997; Roberts etal., 1995). This individual response coupled with pre-treatment factors which affect mobilization and the

40 variety of protocols in use make direct comparisons difficult. Most studies have suggested that progenitors collected increase progressively from the recovery phase of standard chemotherapy alone < higher dose chemotherapy alone < cytokines alone < chemotherapy plus cytokine mobilization (Mohle et al., 1994; Van Hoef, 1996). A major issue will be the extent to which novel mobilization regimens can improve the yields in the poor mobilizers rather than just in those patients in whom mobilization was already adequate.

Progress has also been made in the method of stem cell collection. Continuous flow cell separations will allow the processing of larger cell volumes than intermittent flow machines and recent high volume apheresis programs have been developed with improved progenitor cell yields (Murea et a!., 1996). Attention should also be paid to the timing of G-CSF injections relating to the harvest procedure. An injection of G-CSF paradoxically causes a transient fall in circulating progenitor cells (Watts et a!., 1998a) and G-CSF should not therefore be administered for several hours prior to an apheresis.

1.1.7 Post PBSCT Growth Factors With the rapid haematological recovery that occurs following PBSCT it was uncertain whether post infusion cytokines would further accelerate neutrophil recovery, but a number of randomised trials have now shown this to be the case (Klumpp etal., 1995; Linch et al., 1997). Following BEAM and PBSCT there is a 3 day shortening of the time to achieve a neutrophil count of 0.5 x 107l (Linch et al., 1997) and whether this translates into a cost-effective benefit will depend on the local practice and health care system. A retrospective study suggested that administration of G-CSF resulted in delayed platelet recovery (Bensinger et al., 1995) but this has not been confirmed in the randomised trials.

41 1.1.8 Multiple High Dose Therapies A single course of high dose therapy may not be the best way to intensify dose and multiple cycles of high dose treatment with PBSC support is being explored (Brugger et ai., 1993; Shea et ai., 1992; Tepler et ai., 1993). The dose of drugs used is less than when a single high dose procedure is performed because of the cumulative non-haematological toxicity. This usually means that haematological recovery would occur without PBSCT within a couple of weeks so the benefit of the PBSC infusion is often marginal. In a study of PBSCT after the miniBEAM regimen, which results in neutropenia and thrombocytopenia lasting until a median of 15 and 10 days after finishing chemotherapy, no benefit could be seen for PBSCT compared to historical controls (Fielding et ai., 1996). If PBSC are to be used for multiple cycles of chemotherapy they can either be collected 'up-front' and aliquoted out, or collected off the back of successive cycles of chemotherapy. The latter has the advantage of 'in vivo purging' but the problem that yields tend to decline with repeated cycles. There is also evidence from mouse models that the use of cytokines to aid recovery in between treatment cycles may further exacerbate poor yields with this approach (van Os etal., 1998).

1.1.9 Use of Venesection instead of Apheresis In some patients the mobilized progenitor cell levels are sufficiently high (e.g. CD34+ cells > 50 x 10®/L ) that 1 litre of blood can exceed the minimum requirement for platelet engraftment, and can be used instead of apheresis collections. Ossenkoepelle et ai reported on the use of venesection collections after high dose melphalan in patients with myeloma and noted a shortening of the median regeneration times (Ossenkoppele et ai., 1994) although smaller volumes of 450ml of unprocessed mobilized blood were inadequate for supporting early courses of multicycle chemotherapy for breast cancer patients (Vermeulen et ai., 1998). In our experience the majority of patients do not achieve the minimum threshold levels with a 1 litre collection to support high

42 dose therapy and this means that this strategy can only be used with regimens of moderate dose intensity where the period of pancytopenia without stem cells can be well tolerated.

1.2.0 Use of cytokine-primed bone marrow Some investigators have reasoned that the engraftment benefits of mobilized PBSC could be gained with marrow after "cytokine-priming" and provide advantages such as enhanced stromal cell support, HPC support where apheresis is not available, reduced volume of marrow required and donor choice of collection. In a study by Hansen et al 37 patients with haematological malignancies were primed with G-CSF (5|Lig/kg/d for 5 days) or GM-CSF (10|xg/kg/d for 5 days)or IL-3 (10|Lig/kg/d for 10 days) prior to marrow collection. Despite expanded marrow progenitor activity no improvement in engraftment kinetics was seen compared to historical marrow autografts (Hansen et al., 1995). This lack of benefit was confirmed in another study using IL-3 in a further 19 patients (Sosman et ai., 1995). Enhanced neutrophil engraftment comparable to PBSCT has however been reported in a small study of 6 patients using G-CSF-primed marrow (Lowenthal et a!., 1996). Taken together these studies suggested that HPC mobilized into the blood possess some property that facilitates rapid engraftment that is difficult to reproduce with HPC collected from the marrow compartment. Interest has been renewed recently however by reports of the successful use of cytokine-primed marrow. In a randomized trial using high doses of G-CSF at IB^ig/kg for three days to prime autologous blood or marrow HPC in 55 lymphoma patients (in a ratio of 2 blood to 1 marrow), Damiani was able to show equivalent rapid engraftment in both groups (Damiani etal., 1997). Rapid platelet recovery compared to historical controls has also been shown by Durrant in 10 patients following allogeneic transplants using G-CSF-primed marrow from normal donors collected after 5 days of G-CSF at 10iu.g/kg/day (Durrant et ai., 1997). Whether the rapid engraftment in these studies represents a marrow or blood/marrow effect is

43 unclear. Of the 1000 mL or so of marrow oolleoted from each donor around SOOmL is peripheral blood which is likely to provide sufficient progenitors alone particularly in the case of mobilization with high G-CSF doses and in normal donors. Assessing graft adequacy with this method of HPC support may therefore prove difficult due to the variable contribution of concomitant mobilized blood progenitor cells collected.

1.2.1 Future Developments There is widespread interest in the use of purified CD34+ cells (Cagnoni & Shpall, 1996) and a number of clinical scale purification devices are now available (Berenson et a!., 1991; Hohaus et a!., 1997; Richel et a!., 1996). Direct comparison of the different devices has not yet been made. The major rationale for GD34+ cell purification is to effect purging of tumour cells, although CD34+ cell purification should probably be considered as only the first step in this process if high efficiency purging is to be achieved. The big question is whether such purging will improve long term outcome but there are as yet no adequate sized trials addressing this. A problem with GD34+ cell purification is that significant cell losses occur. In our experience of 116 immunoaffinity column (GEPRATE® SG) procedures the median GD34+ cell loss was 52% (Ghapter 7), which is sufficient to restrict the number of patients in whom such procedures can be performed (Watts et a!., 1997b). It should be noted however, that when allowance is made for the number of GD34+ cells infused, the use of purified GD34+ cells is not associated with any delay in engraftment. (Brugger etal., 1994; Hohaus et al., 1997; Watts etal., 1997b).

Gonsiderable attention is being directed to the possibility of expanding PBSG in vitro (Brugger et al., 1993; Haylock et al., 1992). This could take two forms. First, the aim could be to expand the most primitive cells responsible for long term engraftment. This might be particularly useful for cord blood transplants or when performing autologous PBSGT in heavily pre-treated patients although it

44 is likely that the latter category of patients will 'expand' poorly. Furthermore, although in vitro stem cell expansion can probably be achieved in mice (Muench etal., 1993), particularly with very high levels of cytokines (Miller & Eaves, 1997) it may be that in man any cytokine induced proliferation is associated with differentiation and stem cell loss. Indeed Holyoake and colleagues have reported early regeneration and subsequent graft failure when expanded marrow was used (Holyoake et al., 1997). The second aim is to expand the later progenitor / precursor cells in an attempt to further accelerate haemopoietic recovery. Whether these more mature cells will home appropriately to the bone marrow and develop rapidly into functional mature end cells is not yet known. There is also interest in the in-vitro production of dendritic cells from PBSC for use in tumour immunisation programmes and this is reviewed elsewhere (Fisch etal., 1996; Reid, 1997; Steinman, 1996). Finally PBSC are attractive targets for gene therapy although there are many problems to be resolved before this becomes a clinical reality (Devereux et al., 1998).

1.2.2 CONCLUSION The development of PBSCT has been a significant advance in autologous and increasingly, allogeneic stem cell transplantation and further optimisation of the technology is in progress. In the autologous setting it must be noted however that this represents an improvement in supportive care and the most important issue is whether high dose therapy cures more patients than conventional dose therapy. Large randomised trials are required to address this issue.

45 Chapter 2

General Methods and Materials

The reagents used and suppliers are listed in Appendix I (page 205 ).

2.0.1 "General purpose" tissue culture medium A general purpose tissue culture solution of 10% fetal calf serum in RPMI 1640 with L-Glutamine (Gibco) and preservative free heparin at 20U/mL was sterile filtered and used for general dilution, cell washing and preparation for culture assays. The fetal calf serum used (Gibco, BRL) was heat inactivated at 56°G for 30 minutes (heat-inactivated fetal calf serum, HIFCS) dispensed into 50ml polypropylene tubes and stored at -20°C until required.

2.0.2 Sample handling of apheresis products A sterile harvest sample was taken from the apheresis bag pilot line after repeated “stripping back” of the product into the bag which was continually mixed to ensure even cell suspension. This sample was then diluted 1/10 (for cell counts) and 1/100 (for colony assays) in sterile RPMI containing 10% fetal calf serum and 20U/ml heparin. Direct harvest smears and these dilution’s were then used for a full blood count (STKR Coulter inc. Hialeah, FL). Standard May-Grunwald-Giemsa stained smears was used for manual white cell differential (Dacie 1950). Progenitor measurement by CD34+ cell and CFG assays were performed as detailed below.

2.0.3 Cytocentrifuge preparations Gytocentrifuge preparations were made by adding 2 xIO® to 0.5 xIO® of the cell suspension to 2 drops of 20% bovine serum albumin in a disposable cytocentrifuge cup (Shandon ) The cells were spun onto the slides at 500rpm for 10 minutes and air dried.

46 2.0.4 Viability counts The cell viability of purified CD34+ cell fractions and diluted thawed bone marrow or PBSC was assessed visually in a haemocytometer following 1/2 dilution in 0.4% Trypan Blue solution (Sigma T-8154). The proportion of blue staining cells (non-viable) was subtracted from the total nucleated cell count to derive the viable cell count.

2.0.5 Flow cytometry for CD34+ cell counts Peripheral blood and harvest CD34 positive cell numbers were determined by flow cytometry using a modified Milan protocol (Siena et al., 1991). Apheresis samples were diluted 1/20 in AB serum in a lOOjil volume (approximately 0.5 x10® cells) or a 100|il aliquot of peripheral blood was incubated for at least 30 minutes with a directly conjugated phycoerythrin HPCA-2 monoclonal antibody or appropriate matched control (Becton Dickinson). This was followed by a commercial red cell lysis method (Coulter "Q-Prep"); a phosphate buffered saline wash to reduce background staining and flow cytometry (Coulter Epics XL-MCL). A minimum of 50,000 total white cells were counted for each sample and the percentage of cells which were CD34 positive were identified by the dual characteristics of low light scatter and PE staining and gated as a "cluster" (Figure 2.1). The percentage positive result of the irrelevant control antibody was then subtracted. The CD34+ cell percentages were used to derive absolute counts from the peripheral blood and harvest total white cell counts.

47 A. PBSC harvest: CD34+ cells by flow cytometry (HPCA-2-PE antibody) and slide-based APAAP staining. Far right: MGG morphology of cell smear.

• t

PE intensity • ' • , # _ •• # à A •

Log side scatter —i APAAP MGG

B. CD34+ cell purified cells: CD34+ cell purity assessed by flow cytometry, APAAP staining and MGG morphology of cytospin preparations.

: 85.5% S ’ - c ' . /

* PE ; intensity

"T—TTT, r,7l--T—1 1 rrfyy •• 1' , TITT ■ T TTTi T-n Log side scatter —i

0. Low power view of GM-CFC and BFU-E colonies from purified CD34+ cells incubated for 14 days in methylcellulose media

BFU-E GM-CFC (CFU-GM)

Figure 2.1 CD34+ cell measurement and methylcellulose colony assays

48 2.0.6 Alkaline Phosphatase-anti-Alkaline Phosphatase (APAAP staining) Direct visualisation and counting of CD34+ and CD3+ cells was by APAAP staining (Figure 2.1) (Erber etal., 1984). Tris Buffered Saline (TBS, 0.05M) was prepared by diluting a Tris-HCI buffer 1/10 in saline and used throughout for antibody dilution and washing steps. All antibody labelling and staining was performed at room temperature. The cell smears or cytocentrifuge preparations were air dried and stained immediately or tightly wrapped in parafiim and stored at -20°C for future batch staining. Slides for staining were fixed for 10 minutes with methanol (BOH, Analar) at room temperature (RT) and air-dried. The appropriate mouse monoclonal antibody was diluted 1/10 in TBS and lOOul of this flooded over the cells in a damp chamber and left for 1 hour at RT. Unreacting antibody was then removed from the slide by rinsing TBS over the cells from a wash bottle. Rabbit anti mouse antibody (Dako Z259) diluted 1/20 in TBS was then flooded over the cells and incubated for a second 1 hour incubation in the damp chamber. This layer acts as a "bridge" between the first mouse monoclonal antibody and the final layer of mouse alkaline phosphatase-anti-alkaline phosphatase complexes. The final APAAP layer (Dako, D0651) was applied after a second TBS wash for the last 1 hour incubation. A commercial stain for Alkaline Phosphatase was then used for colour development (Sigma, F-4523). This simply involved dissolving a tablet of buffer (TBS) in lOmL of distilled water, adding a tablet of Fast Red dye and filtering the solution through a 20 micron filter directly onto the slides at RT for 30 minutes for full colour development. The slides were then counterstained with Mayer's Haematoxylin for 5 minutes and mounted (Apathy's aqueous mountant, BDH ).

2.0.7 Manual counting of APAAP stained CD34+ cells Prior to the development of flow cytometry for CD34+ cell measurement visual counting of CD34+ cells was used. The inaccuracies inherent in counting low

49 percentages of positive cells was reduced by the use of a "Miller's squares" eyepiece (Graticules Ltd, Tonbridge, Kent, UK). This is a microscope eyepiece graticule marked with a square containing a smaller square within one corner which is one-tenth the area of the larger square. All cells, negative and positive are counted in the small square but the positive cells are scored separately in the large square. This is designed to ensure that 10 times as many cells are screened for the minority positive cell population than for the predominantly negative total. A 100-cell total was counted in the small square and the CD34+ cell cells seen in the large square was recorded resulting in a percentage of positive cells seen in approximately 10,000 cells examined. The visual counting method was compared to flow cytometric CD34 counting for 42 harvests and showed a good correlation (r=0.870. Figure 2.2) as has been reported by ourselves and others (Dyson et al., 1994; Jones et al., 1994; Watts et al., 1997c).

3 -,

0 E %. ü I

û ü

CD34% by APAAP

Figure 2.2 Visual versus flow cytometric CD34+ cell counting of apfieresis harvest samples

50 2.0.8 Esterase staining Monocyte numbers in apheresis collections and in purified CD34+ cell fractions (Chapter 8) were assessed by esterase cytochemical staining based on the method of Li and Yam (Li et ai., 1973). This was applied to cell smears or cytospin preparations. Colour development in this method is based on an azo- dye coupling reaction to a napthol ring which is released by esterase activity within the cell from the napthol-esterase substrate used.

The stock working stain was made up by the addition the following [-reagent^ reagents to 44.5ml M/15 of Phosphate buffer, pH 7.6 (all from Sigma, UK):

1. 50mg alpha-napthyl acetate substrate dissolved in 2.5 ml ethylene glycol monomethyl ester 2. 3ml of hexazotized pararosanalin azo dye (prepared by dissolving 1g pararosanalin hydrochloride in 5ml of concentrated HCL and adding 20ml H^O).

The pH of the working solution was adjusted to 6.1 with IN NaOH and filtered into 1ml aliquots, which were stored at -20°C until required.

For staining, the smears were fixed for 10 minutes in formalin vapour, the thawed working esterase solution filtered onto the slides and allowed to react for 45 minutes. This was then gently rinsed off with distilled water and 1 % methyl green applied for 2 minutes as a nuclear counter-stain. Monocyte esterase activity was evident as a strong diffuse red-brown reaction. The percentage of these cells were scored out of 100 cells examined in apheresis smears where positive cells exceeded 20% and out of 400 cells for purified CD34+ cytospin preparations where the proportion of positive cells was lower. These percentages were then used to derive the absolute monocyte count from the total white cell counts (Chapter 8).

51 2.0.9 Clinical CD34+ cell purification This is detailed in Chapters 7 and 8

2.1.0 Methylceliulose colony forming cell (CFG) assays During the first pilot studies methylceliulose semi-solid assay media for CFG colony growth was prepared "in-house". As the PBSC program expanded however it became necessary and desirable to use commercially available and quality batch controlled media for the purpose (Stem cell technologies, Vancover, Canada) supplemented with recombinant cytokines.

2.1.1 In-house methylceliulose media: This methylceliulose assay system was modified from that described by Gregory, (Gregory, 1976). The culture mix was made up in a 50ml tube from thawed stock materials previously prepared (described below) as follows: 20ml of 2.7% methylceliulose, 18ml of heat inactivated Fetal calf serum (HIFCS), 6ml 10% buffered BSA solution, 4ml of p-mercaptoethanol stock and SmL of conditioned medium with 100U of added erythropoietin, 250ng GM-CSF and SOOng of IL-3. The mixture was warmed to 37°C in an air incubator, well mixed and dispensed into 'ready to use' 2.5 ml aliquots in sterile, disposable Bijou bottles. When required a Bijou of methylceliulose mixture was thawed in a 37°C air incubator, the required volume of test cell suspension added and the bottle thoroughly mixed prior to plating.

The methocellulose matrix was prepared by adding 10.8g of methylceliulose (A4 premium grade Dow Chemical Co., Hounslow, Middlesex) into a sterile, glass bottle containing a magnetic stirrer. This bottle was autoclaved for 15 minutes at 121°C and 400ml of Iscove's modified Dulbecco's medium (IMDM) containing 500U penicillin and 500^g of streptomycin added to the contents which were stirred at 4°C until fully dissolved (typically, 48 - 96 hours).

52 Conditioned medium (5637 CM) was produced as a source of growth factors using the method of Fraser et al. (Fraser at a!., 1988). This was prepared by seeding 5 x 10® of the 5637 cells (obtained as the HTB9 Bladder carcinoma cell line from the European Cell Culture Collection, Porton Down, Wiltshire) in 25 ml Iscove's modified Dulbecco's medium supplemented with antibiotics and 10% HIFCS in 75 cm^ tissue culture flasks. The flasks were incubated at 37°C in 5% COg in air. Having allowed the cells to expand for 8 to 10 days the supernatant tissue culture medium was removed centrifuged and sterile filtered. Aliquots were then stored at -20°C before use.

Bovine serum albumin (BSA, RIA grade. Sigma) was detoxified before use using a modification of the method of Horton etal. (Horton 1969). 91ml of sterile tissue culture grade water was added to 50 g of BSA and the powdered BSA was allowed to sublime without stirring at 4°C for 48 hours. Then lOg mixed bed resin (Analytical grade mixed bed resin AG-501-X8D-20 -50 Mesh, Biorad, Hemel Hemstead, Herts.) was added to the BSA, stirred and left at 4°C for three hours. The solution was stirred every 20 minutes for the two hours, left at room temperature for 30 minutes and decanted through nylon gauze. The solution (approximately 40% w/v at this point) was made isotonic by the addition of 1.1ml of 10 X strength Dulbecco's phosphate buffered saline to each 15ml of BSA solution. The resulting solution (approximately 37% w/v) was further diluted with single strength PBS to 10% and centrifuged at 1400g for thirty minutes to remove particulates. Before use, the 10% BSA solution was buffered with 7.5% NaHCOg solution (Gibco). For the methylceliulose cultures 0.6ml 7.5% NaHCOg solution was added to every 9.4 ml 10% BSA solution. These solutions were then aliquoted and stored at -20°C.

The p-mercaptoethanol stock was prepared by adding 5^il of p-mercaptoethanol (Sigma) to 30ml of IMDM, and storing at -20°C until required.

53 2.1.2 Recombinant growth factors added Three recombinant growth factors were available to supplement the in-house methocellulose method. The stock solutions of these were diluted in sterile water as indicated and the volumes required for 50ml of culture mix added to the conditioned media that was sterile filtered into the methocellulose mixture.

1. Interleukin-3 (IL-3, Sandoz) stock 300|ig/ml (16ul of 1/10 stock in H^O) 2. GM-CSF (Hoechst )Stock 250p,g/ml (lOul of 1/10 stock in HgO) 3. Erythropoietin (Boehringer-Mannheim) Stock 1000U/ml (ISOjil added)

2.1.3 Commercial methocelluiose media This was considerably more simple. Commercial methylceliulose without added growth factors ("methocult" HCC-4230) was purchased in 80ml aliquots and stored at -20°C. To make the working mixture, growth factors were added to 20ml of Iscove's minimal essential media and sterile filtered into the thawed methocult to give final concentrations of IL-3 30ng/ml, GM-CSF 25ng/ml, SCF lOng/ml, G-CSF 25ng/ml and 3U/ml of erythropoietin. This was well mixed and dispensed in 2.5ml volumes into sterile Bijou tubes while warm. These tubes were stored "ready to go" at -20°C until required.

2.1.4 Plating concentrations Samples were plated in three or four wells in 0.5ml 24-well plates (Costar) Optimal plating densities were established which typically gave around 10-40 total colonies per well:

1 2.5 xlO^^/ml for mobilized blood MNCs, diluted apheresis products and fresh bone marrow MNCs. 2 5 xIOVml for steady-state blood MNCs, thawed apheresis products

54 3 1 xIOVml for thawed bone marrow harvests and CD34+ cell depleted fractions following CD34+ cell purification. 4 1 x10^/ml and 2.5 x107ml for CD34+ cell purified cells

2.1.5 Pilot vial thaw method for CFC For some studies pilot vials of cryopreserved marrow or PBSC were evaluated for CFC content. These samples were thawed quickly in a 37°C waterbath in a ziplock bag and 100^1 of the thawed cell suspension transferred to a sterile conical universal. This was resuspended to 2ml dropwise with 1.9ml of 10% HIFCS in RPMI containing 20U/ml of heparin which was pre-warmed to 37°C. The reconstituted cells had WBC counts and viabilities performed immediately and the required sample volume was added to pre-warmed methylceliulose media. The 0.5% DMSO cryopreservative remaining on the diluted thawed cells was not washed off the cells as cells losses can be considerable (Gorin ) and was further diluted when the required volume of thawed cells (typically <100ul) was added to the 2.5mL of methylceliulose.

2.1.6 Colony counting Colonies were counted after 14 days incubation at 37°C in a humidified 4% COg atmosphere using an inverted stereo microscope (Olympus Optical, London. ). The appearance of colonies formed are shown in Figure 2.1. BFU-E were scored for haemoglobinised colonies exceeding 64 cells (colonies usually contained several thousand cells) or considered as CFU-E below this. Similarly, white cell colonies were scored as GM-CFC if cell numbers exceeded 64 or classified as GM-clusters below this. No attempt was made to quantify mixed colonies at the plating densities used as the chances of random association of CFC types was too high. Colony growth with the commercial and in-house methocellulose media was comparable. No significant differences were noted in the same samples plated in GM-CFC colony numbers but BFU-E growth was more consistent in the

55 commercial media. For this reason where the results from both assays were combined for analysis only GM-CFC numbers were considered.

2.1.7 Mononuclear cell preparation (MNC) and CFC/ml blood Light density mononuclear cells were prepared from blood and fresh marrow samples by ficoll sedimentation. A white cell count was performed on the start sample and a measured volume (typically 5ml of blood or 2 ml of marrow) was diluted with an equal volume of RPMI containing 10% HIFCS and 20U/ml heparin. The diluted sample was then carefully layered onto its own volume of ficoll in a 20ml sterile universal tube. This was centrifuged at 400g for 20 minutes and the light density layer pipetted off, diluted in 20ml of 10% HIFCS/RPMI medium and centrifuged at 800g for 10 minutes to pellet the cells. Contaminating ficoll was removed with the supernatant and the pellet resuspended in the 10% HIFCS/RPMI medium and made up to a measured volume of MNCs (typically 0.5 ml for steady-state blood or 5ml for mobilized blood). A white cell count was performed on this and the MNC/m! of blood/marrow calculated as below:

MNC x10®/ml sample = MNC prep, count x10^/ml x MNC prep, volume (ml) Volume of blood/marrow used for MNC prep.

The volume of MNC preparation required for a given plating concentration in 2.5 ml of methocellulose media was calculated as shown below, using steady- state (resting) blood as an example:

56 Resting blood example: MNC plating required at 5 xIOVmL:

Volume of MNC prep, required (p.!) to add to 2,500 ul of methocell for 5 x10* cells/ml :

= 5 x10* X 2500 = 0.05 x10^ x 2500 = volume to add (|il) MNC prep WBC x10®/ml MNC prep WBC x10®/ml

The required MNC volume was added to the warmed methocell, well mixed and plated into three or four 0.5 ml well in a 24-well "Costar" plate. Sterile PBS was added in between and in surrounding wells to prevent methocell-drying artifacts over the culture period and after 14 days colonies counted as below.

The CFC/ml in this example were then derived as follows:

Average number of CFC per well = CFC /2.5 x10* cells plated (0.5 ml well volume used)

Average number of CFC per well x 4 = CFC/10® cells plated and x10 =CFC/10® cells plated.

CFC/ml blood = MNC x1 OVml blood x CFC/10® cells plated.

2.1.8 Apheresis CFC dose calculations Apheresis harvest samples were processed as "MNC preparations" and simply diluted 1/100 in 10% HIFCS in RPMI with 20U/ml heparin. The volume required to add to the prepared 2.5ml of methocell for a plating concentration of 2.5 xIOVml cell count was then calculated as follows:

Volume of harvest prep, required (jil) to add to 2,500 ul of methocell for 2.5 x10* cells/ml :

= 2.5 x10*______X 2500 = 0.025 x10^ x 2500 = volume to add (jil) 1/100 harvest WBC x10®/ml 1/100 harvest WBC xIOVml

57 After culture, the dose/body wieght (kg) of CFC were derived as follows:

Average number of CFC per well = CFC /1.25 xIO* cells plated (0.5 ml well volume used)

Average number of CFC per well x 8 = CFC/10^ cells plated and x10 =CFC/10® cells plated.

CFC/ml harvest = harvest WBC x10^/ml x CFC/10^ cells plated x harvest volume fmh Body weight (kg)

This is expressed as CFC xIOVkg.

2.1.9 Cytokine Assays Plasma concentrations of all growth factors were evaluated by specific enzyme- linked immuno-adsorption assay (ELISA) kits (“Quantikine”, R+D systems- British Biotechnology, Cowly, Oxford, UK). The detection thresholds of the cytokines measured were; G-CSF 5.0 pg/ml, GM-CSF 20 pg/ml, IL-6 10 pg/ml, stem cell factor 5.0 pg/ml and MIP-1-alpha 3.0 pg/ml. The detection method was similar for each assay. A mouse monoclonal antibody specific to the cytokine to be measured was provided adsorbed to 96 well polystyrene plates. Geometric dilution’s of plasma control recombinant human cytokine were added to the wells with test samples in duplicate, and incubated for 2 hours at room temperature. After washing the plate, a peroxidase conjugated goat -anti -mouse antibody was added for a further 2 hour incubation, followed by a second wash to remove un-reacting antibody. Tetramethylbenzidine was then added which is oxidised by peroxidase activity to a blue reaction product. The reaction was stopped after 20 minutes with 2N sulphuric acid and absorbance read at 450nm vs substrate blank on a microtitre plate reader (Dynatech MR700, Dynatech labs. Ltd., Sussex, UK) . Absorbance of the test wells were compared with the standard curve and expressed as pg/ml. Means and SEMs were calculated for various time points.

58 2.2.0 Data handling and statistical analysis As the number of patients enrolled into the PBSCT program increased it became evident that a computerised database was required. This was constructed for recording clinical, mobilization and harvest details using FilemakerPro software (Claris Corporation) and an Apple Macintosh computer. The cross-platform capabilities of this package allowed transfer into a number of statistical analysis packages such as Microsoft Excel, or for use on an IBM PC e.g. Number Cruncher Statistical System 5.x, Univ. Utah.

59 Chapter 3

Progenitor Threshold Requirements for Haematoiogicai Recovery

3.0 INTRODUCTION Progenitor mobilization, harvesting, processing and cryopreservation impose considerable logistic and financial pressures on the resources of a transplantation unit. Aggressive progenitor mobilization and apheresis collection is expensive and not without morbidity. These factors must be wieghed against the consequences of delayed engraftment which may result from inadequate progenitor collection. Without a knowlege of minimal and optimal progenitor requirement targets, plans to minimise unduly aggressive mobilization protocols, apheresis collection, storage, conventional processing and the use of cell manipulation techniques such as CD34+ cell selection which involve progenitor losses are less feasible. This chapter will consider the broad overall relationship of progenitor dose requirements and haematoiogicai recovery. Subsequent chapters involve the detailed examination of subsets of these patients and address issues of “poor mobilizers”, progenitor quality and the feasibility of purified CD34+ cell transplantation.

There is a considerable consensus that the re-infusion of higher PBSC progenitor numbers results in more prompt haematoiogicai engraftment. The higher the level of recovery counts, guarantee of recovery (99% rather than 90% of patients for example) and the shorter the speed of engraftment expected the higher the target progenitor requirement is likely to be. In this chapter a pragmatic hazard assessment approach to engraftment delay is adopted in which the incidence of pre-defined slow or delayed haematoiogicai recovery times are related to the progenitor dose re-infused and thresholds

60 determined. Most other groups have defined progenitor requirements on the basis of statistical differences between mean or median recovery times of neutrophil and platelet counts for a given progenitor dose re-infused against the progenitor dose re-infused (Bensinger et al., 1995; Schwartzberg et al., 1993; Weaver et al., 1997). which is sensitive but can suggest unrealistically high progenitor collection for some patient groups. Even if the logistics and cost of progenitor collection were not limiting it has been shown in many studies that some patient groups mobilize poorly (Haas et ai, 1994a; Watts et ai, 1998b; Weaver et ai, 1997). In particular^ those patients with bone marrow damage associated with primary disease, prior chemotherapy and/or radiotherapy are likely to mobilize low progenitor numbers even if multiple apheresis procedures are undertaken (Haas et ai, 1994a; Weaver et ai, 1995; Weaver et ai, 1997). Sub-optimal yields may also result from the choice of mobilization protocol, apheresis and progenitor assays used. Progenitor threshold requirements assessed by haematoiogicai recovery are further compounded by variables which may be independent of progenitor dose (Morse et ai, 1993). These include clinical factors such as prior therapy and poor progenitor quality, return to a damaged bone marrow micro-enviroment, intervening disease, infection, drug reactions, immunological cytopaenias, use of growth factors or to technical factors such as measurement errors or freeze- thaw losses. It is unlikely therefore that stringent engraftment predictions for any given progenitor threshold will be applicable to all patients in any large clinical series.

Most studies have used CD34 antigen positive cell numbers as a progenitor measurement of harvest adequacy. The flow cytometric method has the advantages of precision and potential for interlaboratory standardisation and most importantly can be used for immediate apheresis guidance. GM-CFC assays require 2 weeks before the results can be assessed and cannot be. used therefore^to guide apheresis in real time but are a measure of progenitor cell

61 clonogenic function which may be necessary to confirm harvest adequacy from patients with low CD34+ cell yields in particular. The MNC yield has been shown in some studies involving well standardised mobilisation protocols to be a reliable measurement of harvest adequacy (Linch et al., 1997; Scott et al., 1995a; Watts etal., 1997c).

3.1 PATIENTS STUDIED Between June 1992 and March 1996, peripheral blood stem cells mobilized with G-CSF ± chemotherapy were collected on 324 consecutive patients with non-leukaemic malignancies. Of these 256 went on to high dose therapy with PBSC support. The patient diagnoses, median age and sex of the transplanted patients studied in this chapter are detailed in Table 3.1. These patients were primarily involved in a study to assess the overall incidence of poor mobilizers and the efficacy of backup bone marrow to ensure engraftment in such patients. This study is described in Chapter 6 where mobilization protocols, pre­ treatment, apheresis and progenitor yields for each patient group are detailed. The purpose of this chapter is to establish the overall relationship between progenitors re-infused and haematoiogicai recovery in the 256 of these patients who went on to transplantation following high dose therapy. MNC, CD34 and GM-CFC counts and dose were performed as described in the general methods and materials section. Chapter 2. Haematoiogicai recovery was assessed from the time of PBSC infusion (24 hours after completing high dose chemo/radiotherapy) to the day on which a sustained neutrophil count of >0.5x1 Q9/L was achieved. Platelet recovery was defined as the number of days before platelet independence was achieved with a platelet count above 15 X1Q9/L without evidence of bleeding. Recovery was abitrarily defined as “slow” if recovery to these parameters occured between 21 and 28 days and “delayed” if recovery was beyond 28 days. Nine patients with some of the poorest apheresis yields ( below 1 x10®/kg CD34 cells) also had autologous bone marrow re-infused although only the PBSC progenitors re-infused are

62 included in this analysis. Twenty four patients received immunoaffinity purified CD34+ cells and were excluded from MNC dose analysis.

Table 3.1 Patients who received high dose therapy with PBSC Support (n=256)

Diagnosis n Sex Age (M:F) Hodgkin’s disease 99 61:38 31 (15-57)

NHL; Low grade 27 19: 8 48 (25-64) High grade 66 46:20 44 (21-66) Myeloma 31 19:12 55 (19:12) Solid tumour 33 13:20 38 (14-60)

3.2 RESULTS 3.2.1 Overall incidence of engraftment delays The overall incidence of neutrophil engraftment beyond 28 days was 3/256 (1%) and of slow neutrophil recovery over 21 days but less than 28 days was 8/256 (3%). Delayed platelet recovery occured in 29/256 (11%) of patients and slow recovery in 9/256 (4%). Table 3.2 shows the median/range of days to neutrophil and platelet recovery and Table 3.3 the incidence of slow and delayed haematoiogicai recovery, both tables related to our previously defined minimal and optimal thresholds of 1 and >3.5 xlO^/kg CD34+ cells, or 1 and >3.5 xIOF/kg GM-CFC or 1 and >3.5 x10^/kg MNC. Distribution plots of the cell dose returned against recovery time are shown in Figures 3.1.1 to 3.1.3. These thresholds and a recovery line at 28 days are indicated.

63 Table 3.2 Progenitor cell dose re-infused and days to haematoiogicai recovery in 256 patients

Days to Neutrophils Days to Platelet of >0.5 x10®/L Independence Ceil Dose Returned < 1 1 -3.5 > 3.5 < 1 1 -3.5 > 3.5

CD34+ cells x10®/kg median days 15 12 11 >28* 13 10 range 9-50 6-27 8-23 8->500 6-62 5-68

GM-CFC xIOVkg median days 16 12 11 >22* 13 10 range 10-50 6-27 8-19 8->500 7-120 5-68

MNC X 10®/kg** median days 13 12 11 1 4 > 12* 1 1 range 10-18 6-50 8-39 8-20 7->500 5-80

The 24 patients who received purified CD34+ cells excluded from MNC analysis * Exact median not possible as some patients died platelet dependent or did not achieve independence

3.2.2 MNC dose and engraftment No clear discrimination thresholds were evident based on the MNC dose returned with a spread of slow and delayed engraftment over a wide range of doses from approximately 1-6 x10®/kg (Figure 3.1.3).

3.2.3 CD34+ cell dose and engraftment In contrast a clear threshold is evident at a dose of 1 x10®/kg CD34-h cells re­ infused (Figure 3.1.1). The 41 patients who received <1 xlO^/kg included all 3 patients with delayed neutrophil recovery and 20 of the 29 patients with delayed platelet recovery including all those who failed to achieve platelet independence within 70 days. If the incidence of delayed platelet recovery is considered against the CD34-H cell dose re-infused in increments of 1 x10®/kg, this was 4/52 at 1-2, 2/36 at 2-3, 3/25 at 3-4, 0/22 at 4-5 and 0/80 above 5 xIO® CD34+ cells/kg. The increased frequency of delayed platelet recovery in those

64 patients who received <1 x10®CD34+ cells/kg compared to those who received 1-2 x10®CD34+ cells/kg was highly significant (P<0.0001) but did not achieve significance between the stepwise increases above 1 x10®CD34+ cells/kg. However the incidence of delayed platelet recovery was significant between those patients who received between 1-3.5 x10®CD34+ cells/kg (8/102) and 1/113 in those who received a CD34+ cell dose above this level (P=0.0157). The one patient who received a CD34+ cell dose above 3.5 x10®CD34+ ce I Is/kg and did not achieve platelet independence until day +60 was a 15 year old boy with a diagnosis of Ewing’s sarcoma in whom slow platelet recovery had been noted following busulphan during his standard chemotherapy. This was the principal drug used in his high dose therapy.

3.2.4 GM-CFC dose and engraftment A re-infused GM-CFC dose of below 1 x10^/kg was also predictive for poor engraftment and included the 3 patients with delayed neutrophil recovery and a high incidence of delayed platelet recovery (12/27, 44%) compared to 15/97 (15%) in those who received between 1-3.5 x10^/kg GM-CFC (P=0.0156). There was also a cluster of patients with delayed platelet recovery (12/43,28%) who received between 1-2 x10^/kg GM-CFC (Figure 3.1.2). The difference between these two incidences was not statistically significant and it is noteworthy that the patients who received <2 x10®/kg GM-CFC accounted for 24 of the 29 patients with delayed platelet recovery including all those who failed to achieve platelet independence within 70 days. No significant differences between the frequency of delayed platelet recovery was evident in stepwise doses above 2 xlO^/kg GM-CFC however. The incidence of delayed platelet recovery was 3/33, 1/34, 1/25 and 0/93 for patients re-infused with 2-3, 3-4, 4-5 and more than 5 x10^/kg GM-CFC respectively. It is particularly of note that all of the patients with delayed platelet recovery who received up to 2 x10^/kg GM-CFC were also among the 41 who received <1 x10®CD34+

65 cells/kg. Amongst these 41 patients 19 had a GM-CFC dose below 1 x10®/kg with 12 cases of delayed platelet engraftment, 18 received 1-2 xlO^/kg GM- CFC with 9 cases of delayed platelet engraftment and the remaining 4 received a GM-CFC dose above 2 xlO^/kg at 2.5, 3.2, 3.7 and 3.7 xIOVkg with no cases of delayed or slow recovery.

Table 3.3 Progenitor cell dose re-infused and risks of delayed haematoiogicai recovery in 256 patients

Neutrophil Platelet Recovery Recovery Cell Dose Returned < 1 1-3.5 > 3.5 < 1 1-3.5 > 3.5

CD34+ cells x1C^/kg n = 41 102 113 41 102 113 Slow ( >21 days) 3 4 1 3 3 3 ( 7%) ( 4%) ( 1%) ( 7%) ( 3%) ( 3%) Delayed ( >28 days) 3 0 0 2 0 8 1 ( 7%) ( 0%) ( 0%) (47%) ( 8%) ( 1%)

GM-CFC x1(f/kg n = 27 97 132 27 97 132 Slow ( >21 days) 3 5 0 1 3 5 (11%) ( 5%) ( 0%) ( 4%) ( 3%) ( 4%) Delayed ( >28 days) 3 0 0 1 2 1 5 2 (10%) ( 0%) ( 0%) (44%) (15%) ( 1%)

MNCx 1QF/kg* n = 14 105 113 14 105 113 Slow ( >21 days) 0 7 1 1 3 4 ( 0%) (7%) ( 1%) ( 0%) (4%) ( 3%) Delayed ( >28 days) 1 0 2 1 1 8 7 ( 7%) ( 0%) (2%) ( 7%) (17%) ( 6%)

‘The 24 patients who received purified CD34+ cells are excluded from MNC analysis

66 100-1

jn 75- IE Q. 2 3 G) (Do X O in (/) o S' Q

(n = 4) 100^ B

Ü0 c "O0 c 0 Q. -D0 c

0 0 0 0_ o ..... o 0 0 Û

CD34+ cells x 106/kg Fig 3.1.1 PBSC CD34+ cell dose reinfused and haematoiogicai recovery of (A) Neutrophils and (B) Platelets in 256 patients. 0 = 9 patients who also received autologous bone marrow.

67 100-,

(fi 75- le 2 cd 30) 1- O Z X 5 0 ^ o ^ o s. co Q 25-

(n = 1 4 >100

B

0 ü 75 H C ■D0 C 0 Q. "O0 C 50-

0 0

CL 25- % ...... 2 (fi S' û

5 10

GM-CFC xlO^/kg

Fig 3.1.2 PBSC GM-CFC dose re-infused and haematoiogicai recovery of (A) Neutrophils and (B) Platelets in 256 patients. 0 = 9 patients who also received autologous bone marrow.

6 8 100n

(/) 75-

i=i2 05 3 O ^ X in

C/D S' Û 25- O ' <$>

> 100n oc# o B 0 o o 75- c 0 ■a c 0 Q. TD0 _C 50-

0 0 5 o CL 254 O ^ ^ o 0 S' o

T" 5 10 15

MNC x108/kg

Fig 3.1.3 PBSC MNC dose re-infused and haematoiogicai recovery of (A) Neutrophils and (B) Platelets in 256 patients. 0 = 9 patients who also received autologous bone marrow.

69 ■

5 » ^ C

3.2.5 Correlation between CD34+ cells and GM-CFC collected at apheresis The total CD34+ cells and GM-CFC collected at apheresis for the 256 patients in this study was not linear but well correlated (r=0.832, Fig 3.2). It is particularly worthy of note that of the 215 patients in whom a CD34+ cell dose of at least 1 xIOVkg was collected only 2 patients had a corresponding GM- CFC dose below 1 xlO^/kg. The first case was an 32 year old female with a diagnosis of Hodgkin’s disease who had received 10 cycles of chemotherapy and mantle zone radiotherapy. She was mobilized with a single dose of cyclophosphamide (1.5gW) followed by G-CSF the following day at a dose of 10|ig/kg/day until apheresis was complete. She had two apheresis collections performed with yields of 0.6 and 0.4 x10®CD34+ cells/kg and 0.2 xlO^/kg GM- CFC in each case. Both collections were returned following BEAM chemotherapy and post infusion G-CSF administered from this day. Neutrophil recovery to 0.5 xIO^L and platelet independence were achieved by day 11 and 17 respectively. The second patient was a 44 year old man mobilized with the same protocol with a diagnosis of low grade to high grade transformation of a low grade Non-Hodgkin’s lymphoma. This involved the bone marrow at diagnosis but was judged morphologically clear at the time of mobilization. He had received 3 cycles of fludarabine and 3 cycles of CHOP chemotherapy prior to mobilization. He had 3 apheresis procedures with yields of 0.6, 0.5 and 0.5 x10®CD34h- cells/kg and 0.3, 0.2 and 0.2 xIOVkg GM-CFC respectively. These cells were returned following cyclophosphamide and total body irradiation conditioning and he was given post infusion G-CSF from day +15. Neutrophil recovery to 0.5 xIO^/L was achieved by day 19 and platelet independence by day 17.

A total apheresis collection of at least 1 xIOVkg CD34+ cells/kg assured a GM- CFC dose of at least 1 xIOVkg in all but the two patients mentioned and in the majority (194/215, 90%) assured 2 xIOVkg GM-CFC. For the 183 patients in whom at least 2 xIoVkg CD34+ cells were collected 179 patients had a

70 corresponding GM-CFC dose of at least 2 x10^/kg (Figure 3.2). The four exceptions had GM-CFC doses of 1.3, 1.4, 1.5 and 1.6 x1 OVkg. All four of these patients experienced rapid neutrophil and platelet recovery within 17 days of PBSC infusion.

9) O)

o <%> o o,

GM-CFC x105/kg

Figure 3.2 Progenitors closes collected in 258 patients at apheresis. A CD34+ cell yield of 2 x1 OVkg asssured a GM-CFC yield of 2 xIOVkg in 98% of patients

71 3.2.6 Effect of post infusion G-CSF on haematoiogicai recovery Planned post infusion G-CSF was administered on the day of PBSC infusion or from day +6 for a total of 81 patients at a dose of Spg/kg/day sc. It was

administered symptomatically for slow neutrophil recovery in a further 10 patients. Neutrophil and platelet recovery of the 256 patients related to the GD34+ cell dose infused is shown in Figure 3.3 with those patients who received early post infusion G-CSF indicated. Of the 3 patients with a neutrophil recovery beyond 28 days, all of whom received <1 x10®CD34+ cells/kg, the two patients with longest recovery times of 39 and 50 days received G-CSF from the day of PBSC infusion and the third with a recovery time of 30 days from day +19. It is notable that none of the 9 patients who took 21-28 days to recover a neutrophil count of 0.5 x107L received post infusion G- CSF. Three of these patients received CD34+ cell doses below 1 x10®/kg and the remaining six received 1.0, 1.7, 2.1, 2.1, 2.9, and 3.6 x10®/kg CD34+ cells. Among the 41 patients who received <1 x10®CD34+ cells/kg the median day (range) to neutrophil recovery of the 13 patients who did not receive post infusion G-CSF was 12 (10-50) compared to 15 (9-30) for the remaining 28 patients. For the 215 patients who received >1 x10®CD34+ cells/kg the median day (range) to neutrophil recovery was 10 (8-17) for the 68 who received early G-CSF and 12 (6-27) for the remaining 147 patients (P=0.05). Slow or delayed platelet recovery was related to CD34+ cell dose re-infused independently of G-CSF administration.

72 50 ^ Early G-CSF

40- O None or late G-CSF

ü)

30- O ! O <> i o ; ^ <

O

> 100130

O

5 10 CD34 dose x10^/kg

Figure 3.3. PBSC CD34+ cell dose returned to 256 patients and days to A) Neutrophil and B) Platelet recovery when given post infusion G-CSF.

73 3.3 DISCUSSION Various figures have been quoted for progenitor cell requirements, reviewed by Demirer (Demirer & Bensinger, 1995), but most values quoted are in the range of 1 to 5 xIOVkg GM-CFC or 1 to 5 x10®/kg CD34+ cells. Part of the variation is dependant on what proportion of patients with slow engraftment is acceptable to a given centre and how this is defined, in a series of 52 patients with lymphoma and solid tumours reported by Schwartzberg et a/., (1993) all 4 patients with slow platelet recovery had received <2 x lOVkg GM-CFC. Zimmerman studied 30 patients receiving high dose therapy for advanced breast cancer. A high level of slow engraftment was seen with a CD34+ cell dose below 0.75 x 10Vkg and a threshold of 2.0 x 10®/kg was suggested for rapid engraftment (Zimmerman et al., 1995). Of 42 lymphoma patients autografted with PBSC reported by Haas a dose of 2.5 xIOVkg CD34+ cells was identified as a threshold to achieve rapid engraftment (Haas et a!., 1994a). Of the 7 patients with delayed recovery below this threshshold in the Haas series, 6 received less than 1.0 x107kg CD34+ cell and required 4 to 11 apheresis collections to achieve this. Excessive apheresis of poor mobilizers to ensure engraftment may not be effective however as reported by Weaver who attempted to increase progenitor numbers by remobilizing and reapheresing patients with poor yields but did not reduce the risk of slow engraftment suggesting poor progenitor quality (Weaver et a!., 1995; Weaver et a!., 1997). Overall these threshold values are similar to those described in our study and suggest that they may be broadly applicable to a range of diseases and conditioning regimens. The highest incidence (around 50%) of delayed engraftment in the present study were seen in the patients with a PBSC yield below 1 x10®/kg CD34+ cells or below 1 xIO^ GM-CFC. However a relatively high incidence (12/43, 28%) was also seen in those patients who received between 1-2 xIOVkg GM-CFC. Although these 12 patients also had a concomitant CD34+ cell dose below 1 x10®/kg minimum progenitor levels of 1 xIOVkg CD34+ cells and 2 xIOVkg GM-

74 CFC appear to be prudent at these borderline levels. A GM-CFC yield of at least 2 xIOVkg was achieved in 179/183 (98%) of patients in whom 2 xIOVkg CD34+ cells were collected at apheresis and these patients could proceed directly to transplantation if required without awaiting colony assays. In those patients with a CD34+ cell yield below 1 xIOVkg it would be preferable to wait and collect additional progenitors with an optimal mobilization protocol (Perry et al., 1998) if permitted by the underlying clinical condition. Some patients with a GM-CFC dose below 2 xIOVkg may be still considered eligible for transplantation depending on the perceived need but this is a clinical decision to be made by the physician in charge in each individual case.

It is noteworthy that in this study the number of MNC was not predictive of recovery compared to the progenitor cell measurements in contrast to our previous experience and that of others (Linch at a!., 1997; Scott at a!., 1995a; Watts at a!., 1997c). In the study reported by Scott, 28 pre-treated lymphoma and myeloma patients were mobilized with cyclophosphamide at 3g/m^ or 7g/m^ and G-CSF, apheresis commencing on a recovery WBC above 3.0 x107L. Those patients where an MNC dose over 4 x10®/kg was collected also had a GM-CFC dose over 1 xIOVkg and experienced rapid engraftment following high dose therapy. Our studies used a mobilizing regimen of cyclophosphamide at 1.5g/m^ and G-CSF with the first harvest carried out on the first day the WBC exceeded 5.0 x 10^/L. This specific mobilization regimen can be used safely without the more complex progenitor cell measurements providing the WBC recovery counts are prompt and MNC yields are high. Where the WBC kinetics are not validated for all of the mobilisation regimens as in this study CD34 or GM-CFC measurements are still mandatory. In addition the MNC count is inappropriate when the harvest has been manipulated including progenitor cell purification.

75 Although not a randomised study those patients who received post-PBSC infusion G-CSF had significantly faster neutrophil recovery without any delay in platelet recovery as had been suggested in one study (Bensinger etal., 1995). The beneficial effect of G-CSF post-PBSC transplantation has however been confirmed by ourselves and others in randomized trials (Klumpp etal., 1995; Linch etal., 1997)

In conclusion, these data suggest minimum progenitor requirements for direct progress to high dose therapy of 1 x10®/kg CD34+ cells and 2 xIOVkg GM-CFC, both thresholds ensured by collecting 2 x1 OVkg CD34+ cells at apheresis. Below these levels additional clinical assessment of the individual case should be made before proceeding to transplantation.

76 Chapter 4 Factors Predicting for Mobiiization Efficiency

4.0 INTRODUCTION A number of factors have been shown to predict for a poor PBSC yield including diagnosis, prior radiotherapy, marrow infiltration or fibrosis and extensive prior chemotherapy (Bensinger et al., 1995; Haas et al., 1993; Passes Coelho et al., 1995; Watts et al., 1997c). The type of prior chemotherapy may also be relevant. In a study by Morton et al for example continuous therapy with the alkylating agent chlorambucil expressed as a period of administration or total grams received was shown to be highly detrimental to CD34+ cell yields (Morton et al., 1997). Significantly poorer yields were also shown recently by Ketterer to be strongly associated with the use of tAe larvtL-meèotoUte f Iu4^f25bine, especially in patients with indolent lymphomas (Ketterer et al., 1998). In study by Dreger et a l , more than one cycle of the stem cell toxic Dexa-BEAM regimen was found to be associated with very poor subsequent PBSC harvests (Dreger et al., 1995). This poor mobilization effect has recently been confirmed by Weaver et al using mini-BEAM, a similar chemotherapy salvage regimen (Weaver etal., 1998). Many different mobilization schemes have been described but there is general agreement that the highest progenitor yields are achieved with a combination of chemotherapy and growth factors (Demirer & Bensinger, 1995; Mohle et al., 1994; Van Hoef, 1996). The chemotherapy can be part of the standard therapy (Brugger et al., 1993; Dreger etal., 1993; Pettengell et al., 1993; Watts et al., 1996b), or can be specific to the mobilization process (Gianni et al., 1989). In our institute we have extensively used relatively low dose cyclophosphamide (1.5g/m^), followed by G-CSF in pre-treated patients with malignant lymphoma (Jones etal., 1994). Such a standardized protocol can be applied to patients at all stages of their disease, and is associated with much less morbidity than

77 higher dose regimens (Gianni etal., 1989; Rosenfeld etal., 1995; Rowlings et al., 1992). This regimen has also allowed successful engraftment from a single apheresis harvest in a proportion of patients (Jones et al., 1994). The purpose of this study was to evaluate the progenitor yields collected from a relatively homogeneous series of 101 pre-treated lymphoma patients all mobilized with this protocol. The relationship between the progenitor yields collected from an optimal timepoint during mobilization and the patient age, sex, diagnosis, prior radiotherapy, the number of cycles and timing of chemotherapy was determined by multivariate analysis. Using this information it has been possible to determine the proportion of patients achieving the different progenitor threshold requirements discussed in Chapters 3 and 7 when considering either a single or a double apheresis collection. A proportion of patients were deemed to be “poor mobilizers” and the factors predicting for this have been evaluated.

4.1 PATIENTS AND METHODS This study refers to 101 patients with relapsed or resistant lymphoma, 61 with Hodgkin’s Disease and 40 with Non Hodgkin's Lymphoma, who were entered into the high dose therapy program at University College London Hospital. The demographics of these patients are shown in Table 4.1. No patient had morphological bone marrow involvement with disease at the time of mobilization.

78 Table 4,1 Patients studied

Hodgkin’s Non-Hodgkin’s Variable Disease Lymphoma (n = 61) (n = 40) Median age 30 44 ( range ) 15-54 20-61 Sex ( male:female ) 36:25 25:15 No. given Prior radiotherapy 30/61 5/40 % 49 13 Cycles of prior chemotherapy: 8 6 median (range) 4-20 3-18 Number of patients recieving 0,1 or 2 courses of mini-BEAM therapy prior to PBSC collection 0 46 31 1 11 9 2 4 0

4.1.1 PBSC mobilization and collection. All patients were mobilized with low dose cyclophosphamide (1.5g/m^) given on day 1 followed by G-CSF given sc at lOjxg/kg (filgrastim) or a single vial of lenograstim (263 pg/vial) 24hrs afterwards and daily thereafter until harvesting was complete. One to three apheresis harvests were collected on days 8 to 12, on a Haemonetics V50 (n=18), COBE Spectra (n=27) or Baxter CS3000 apheresis machine (n=56), commencing when the white count was beginning to rise after the cyclophosphamide induced nadir. Each apheresis procedure was carried out for approximately 4 hours, with the Haemonetics V50 this resulted in the processing of 3.5-5.5 liters of blood , with the COBE Spectra 8- 16 liters and 10 liters (pre-set) with the Baxter CS3000. In the early phases of this study all patients had 3 apheresis collections, and 97 of the 101 patients had at least 2 harvests. As the PBSC program evolved, the timing of the harvest was refined and it became apparent that a single apheresis was adequate in many patients. In the harvest data presented in this study the first harvest is that taken on the day when the recovery WBC first exceeded 5.0 x

79 107L which we have previously determined to be the optimal time for a harvest with this mobilization protocol (Watts et al., 1995b). Differential white cell counts and progenitor assays for CD34+ cells and GM-CFC were measured as described in the general methods and materials section, Chapter 2.

4.1.2 Statistical analysis of data. The relationship between the various clinical parameters and harvest yield were examined by univariate and multivariate analysis using the hazards regression model of Cox (Cox, 1972).

4.2 RESULTS 4.2.1 Toxicity of Mobilization regimen Peripheral blood stem cells were collected after cyclophosphamide 1.5g/m^ and G-CSF in 101 lymphoma patients. During the mobilization period 5/101 (5%) of patients required admission to hospital, 3 with neutropaenic sepsis, one with bone pain and a fifth with disease related symptoms.

4.2.2 Effect of G-CSF regimen and leukapheresis machine on PBSC yields. Two different G-CSF regimens were used: either filgrastim lO^ig/kg/day (n = 43) or lenograstim 1 vial (263|xg) / day (n = 58) from the day after the cyclophosphamide until harvesting was complete. In initial studies on the mobilization achieved with these two different G-CSF regimens, the kinetics and magnitude of the mobilization were found to be similar (Figure 4.1 A). The total MNC, GM-CFC and CD34+ cells collected on the first day that the WBC exceeded 5.0 x 10^/L after cyclophosphamide induced nadir were also similar (Fig 4.1 B). The median dose of MNC, CD34+ cells and GM-CFC collected were 1.6 x 10®/kg, 2.1 x 10®/kg and 2.1 x lOVkg respectively in the filgrastim group and 2.2 x 10®/kg, 2.2 x 10Vkg and 2.7 x 10^/kg in the lenograstim group. In subsequent analyses the two G-CSF regimens have therefore been analyzed together. In the first 18 patients, PBSC were collected using a

80 Haemonetics V50 cell separator. Subsequent PBSC were harvested using continuous flow cell separators, either a COBE Spectra (n = 27) or a Baxter CS3000 (n = 56). The Haemonetics V50 machine collects a buffy coat and the median number of MNC harvested was 1.3 x 10®/kg (range 0.4 - 3.6) on the first day the WBC exceeded 5.0 x 10^/L, which was less than with the continuous flow machines at the same time point ( median 2.1 x 10®/kg - range 0.5 - 17.8). The progenitor yields were similar however; the median number of CD34+ cells and GM-CFC collected with the Haemonetics machine being 2.2 xIOVkg and 2.1 x lOVkg respectively compared to 2.1 x 10®/kg and 2.7 x lOVkg for the continuous flow machines. The discrepancy between lower MNC yield and equivalent progenitor cell yield with the Haemonetics V50 may in part relate to the difficulty in performing accurate MNC counts on a buffy coat.

81 lenograstim G-CSF filgrastim G-CSF 1 via! (263|j.g)/clay 10fxg /kg /day 100000

10000 -

o

1000- E O LL O 100 - È o

10 -

dayl day 10 day 11 day 1 dayl 0 day 11 Day of mobilization

B lenograstim G-CSF (n=58) filgrastim G-CSF (n=43) 1 vial (263|ig)/day 100.00

10.00 - O) O) O) CD 1.0 0 -

0.10 -

0.01 -

0.00 MNC CD34 GM-CFC MNC CD34 GM-CFC Cells collected

Figure 4.1. Kinetics of progenitor mobilization after 1.5g/m2 cyclophosphamide followed by lenograstim or filgrastim G-CSF (A), and cells collected at first apheresis showing medians lines and distribution (B). There was no significant difference for MNC (P=0.7), CD34+ cells (P=0.7) and GM-CFC (P=0.11) collected.

82 4.2.3 Factors Predicting for PBSC Yieid Multivariate analysis was carried out to determine which factors predicted for obtaining the various CD34+ cell and GM-CFC thresholds on a single apheresis collected on the first day the mobilization recovery WBC exceeded 5.0 x10®/L. The factors considered were age, sex, diagnosis, cycles of previous therapy, previous treatment with radiotherapy, intermittent versus continuous flow apheresis, and the G-CSF mobilization regimen employed (Table 4.2). The only factor predicting for a poor yield on a single apheresis (collected on the first day the post nadir WBC exceeded 5 x10®/L) was the receipt of previous radiotherapy. (P=0.04 for GM-CFC yield and P=0.03 for CD34+ cell yield). This only applied if GM-CFC or CD34+ cells were considered as continuous variables but did not hold if the achievement of a single target thresholds of 1 or 2 x10® CD34+ cells or 1 or 2 x10® GM-CFC was used. Paradoxically, the administration of radiotherapy to any site appeared more predictive than when only wide field irradiation was considered. No predictive effect of prior chemotherapy was apparent in this series and 8 of the 29 patients with poor mobilization defined as a yield of 1 x10®/kg CD34+ cells collected at first apheresis when the WBC exceeded 5.0 x10®/L had received neither prior radiotherapy nor more than 6 cycles of chemotherapy (Figure 4.2). A multivariate analysis of factors predicting for yield in terms of GM-CFC collected gave the same result.

101 patients r - - 29 patients <1 x10®/kg CD34+ cells collected at first apheresis when WBC >5.0 x10®/L

7 N 16 patients 13 patients NoRTz RT 8 patients <6 courses 8 patients >6 courses of chemotherapy of chemotherapy

Figure 4.2 Flow chart showing the relationship of the poorest mobilizers to prior therapy

83 Table 4.2 Univariate and multivariate analysis of clinical factors on the achievement of target PBSC progenitor yields in 101 lymphoma patients

Clinical Factor Prediction of Minimum Progenitor Yield at First Apheresis (when recovery WBC first exceeds 5 xlG^L) CD34+ cells GM-CFC 1 xtCf/kg 2 x1(f/kg 1 x1(f/kg 2x10P/kg

Age 0.53 0.33 0.69 0.29 (<33,>33) Sex 0.96 0.57 0.88 0.89 G-CSF type 0.42 0.60 0.96 0.76 (filgrastim/lenograstim) HD or NHL diagnosis 0.87 0.14 0.21 0.22 Chemo cycles (<6, >6) 0.16 0.75 0.69 0.83

Previous radiotherapy 0.25 0.29 0.52 0.74 (none versus any) Prior miniBEAM/BEAM 0.43 0.24 0.63 0.31 (Yor N) Diagnosis to 0.34 0.36 0.87 0.34 mobilization(mo) (<12. >12) End of chemotherapy to 0.22 0.19 0.47 0.09 mobilization (mo) (<3, >3) G-CSF dose (lenograstim) 0.61 0.52 0.49 0.39 Previous radiotherapy none versus any with 0.030 0.038 progenitor yield as continuous variable

4.2.4 Proportion of patients achieving different collection thresholds The proportion of patients achieving the minimum and optimal progenitor cell yields with either one or two leucaphereses has been determined. In addition the attainment of a minimum yield for CD34 purification has been assessed. Six of the patients in this series had CD34+ cells purified on an immunoaffinity stem cell concentration system. The median yields of CD34+ cells and GM- CFC in the purified fraction were 40% and 60% respectively. These losses of around 50% are in accord with our wider experience in a total of 116 CD34+ cell purification procedures discussed in Chapter 7. This implies that if a

84 CD34+ cell dose of 2 x10®/kg CD34+ oelis (and by implication a GM-CFC dose of 2 x10®/kg) was required as discussed in Chapter 3 for direct progress to high dose therapy then approximately 4 xIOVkg CD34+ cells would be required for stem cell purification. A pre-CD34+ cell selection harvest of 2 x10®/kg CD34+ cells could be considered but it would be important to ensure the minimal CD34+ cells dose of 1 x10®/kg resulted from purification and to await colony assays before making the decision to proceed to high dose therapy.

In total 72/101 of patients ( 71%) harvested when the WBC first exceeded 5.0 x IOVL (before day 11) had a single apheresis CD34 dose of at least 1 x10®/kg. All of these patients had GM-CFC >1 xIOVkg and in 65 (90%) of cases they also had above 2 x10®/kg, this latter group satisfying both minimal progenitor thresholds for progress to high dose therapy as discussed in Chapter 3. Around half of the patients (53/101) had a CD34+ cell yield of at least 2 xIOVkg at first apheresis and 49 of these also exceeded 2 xIOVkg GM-CFC. Only 28/101 patients exceeded the optimal threshold CD34-H cell threshold of 3.5 xIOVkg (Figure 4.3). Ninety-five patients in this study had one or more additional harvests after the first day on which the WBC was > 5.0 x107L and two patients with a single collection at this time exceeded the optimum CD34 and GM-CFC dose thresholds. The impact of single and additional harvests on the attainment of progenitor cell yields can therefore be assessed for 97 patients (Figure 4.3). Overall the number of individuals who attained a CD344- cell threshold of 2 xIOVkg increased to 74/97(76%) with the addition of a second apheresis, and the proportion attaining an optimal yield increased to 51/97(53%). It is notable that if two consecutive harvests are pooled, the proportion of patients with lymphoma with adequate cells to consider CD34+ cell purification feasible increases from 53 to 75%.

85 First harvest Combined first and second harvest 1007on A CD34 cell dose w 75%- x106/kg 0 E III □ > 3.5 ^ 50% H 2 - 3.5 .0 l i l CB □ 1 - 2 ÜL III < 1 25%-

0% 1 Apheresis 2 Aphereses

100%n

B GM-CFC dose

CO 75%- x105/kg 0 E 3 □ > 3.5 ^ 50%- 2 - 3.5 0 m 0 CL s 1 - 2 25%- ill ■ < 1

0% 1 Apheresis 2 Aphereses

Fig 4.3. Distribution of progenitor threshoids collected in a singie apheresis or combined with a second collection in pre-treated lymphoma patients ( n=97 ) for A) CD34-H cell, and B) GM-CFC yields

8 6 4.3 DISCUSSION This series of 101 consecutive lymphoma patients mobilized with cyclophosphamide at a dose of 1.5g/m^ followed by G-CSF confirms the convenience and safety of this regimen. Only 4 patients required admission to hospital for complications related to mobilization which compares very favorably with higher dose cyclophosphamide regimens. Hospital admission rates of 100% and 44% have been reported by Rowlings using cyclophosphamide alone for mobilization at doses of 7g/m^ and 4g/m^ respectively, with a single mortality in each group (Rowlings et al., 1992). Nearly 20% of patients in each group required platelet transfusions. Gianni noted that the addition of GM-CSF to a 7g/m^ cyclophosphamide mobilization regimen reduced the toxicity but 13% of patients still required platelet transfusions (Gianni et a!., 1989). Rosenfeld and colleagues found that the addition of G-CSF or GM-CSF to cyclophosphamide 4g/m^ reduced hospitalization rates from 73% to 32% although a procedure related death still occurred (Rosenfeld et a!., 1995). It is particularly notable that severe thrombocytopaenia is most uncommon with the lower dose of cyclophosphamide used here which is a considerable advantage for the insertion of lines and for the apheresis procedures which may further lower the platelet counts.

After the initial 56 patients had been entered into the study, a stepwise analysis of the progenitor cell yields obtained at different post - nadir WBC was carried out using WBC intervals of 1.0 x 107L. This revealed that the optimal time to harvest as determined by progenitor yields was when the post cyclophosphamide induced WBC nadir first exceeded 5.0 x 10®/L (Watts et a!., 1995b). This is higher than the frequently quoted values of around 1.0 x 10®/L (Craig et a!., 1993a; Pettengell et a!., 1993), but is in close accord with data from other groups (Dreger etal., 1993; Haas et al., 1994a; Scott etal., 1995a).

87 During the process of progenitor cell mobilization and harvesting it is necessary to use surrogate markers of engraftment potential. In Chapter 3 these surrogate values were validated by assessment of subsequent outcome in a large series of patients recieving high dose therapy including TBI. Neutrophil engraftment within 28 days is almost invariable above 1 x10®/kg CD34+ cells or 2 xIOVkg GM-CFC but the risks of delayed platelet recovery increase progressively as the number of cells infused decreases. Above threshold values of 3.5 x10®/kg CD34+ cells or 3.5 x10^/kg GM-CFC platelet engraftment delayed beyond 28 days is very rare. Below a level of 1.0 x10®/kg CD34+ cells and 2.0 xIOVkg GM-CFC the risks of engraftment delay are very high (approximately 50%).

Although the results of mobilization with a single dose of cyclophosphamide at 1.5g/m^ plus G-CSF are generally satisfactory there is failure to achieve adequate mobilization in approximately 7% of patients with pre-treated lymphomas. This raises the issue of using alternative mobilization regimens. There is some evidence that higher doses of cyclophosphamide give rise to higher progenitor cell yields (Boiron etal., 1993) and higher doses of G-CSF have a similar effect (Zeller et a!., 1996). However, increasing the dose of cyclophosphamide is associated with significant morbidity and occasional mortality (Gianni etal., 1989; Rosenfeld et al., 1995; Rowlings etal., 1992) and using a higher dose of G-CSF or additional cytokines increases the expense. It would therefore be desirable to restrict the use of alternative regimens to those patients at high risk of mobilization failure. Multivariate analysis has shown that prior radiotherapy is a significant predictor of poor mobilization when the progenitor yield is considered as a continuous variable which is in accord with previous reports (Haas et al., 1994a; Schwartzberg et al., 1993). Previous radiotherapy is of little value in predicting who will achieve a minimum threshold however. It is noteworthy that administration of any radiotherapy was of greater predictive value than when just wide field irradiation was considered.

88 This may however be spurious due to the small patient numbers in each subgroup. We have not observed a significant influence of the amount of previous chemotherapy in this series which is in accord with some others (Kotasek et al., 1992) but which differs from several other reports (Haas et a!., 1994a; Passos Coelho etal., 1995). This may be due to the fact that ours is a relatively homogenous series. All patients had malignant lymphoma and had been pre­ treated, with only 6 patients having had more than 12 cycles of chemotherapy.

With the cyclophosphamide 1.5g/m^ and G-CSF regimen 65% of lymphoma patients have an adequate harvest from a single apheresis defined as 1 xIOVkg CD34+ cells and 2 xIOVkg GM-CFC although only 28% achieve an optimal CD34+ cell yield of at least 3.5 x10®/kg. If a second consecutive harvest is combined with this first collection, the proportion of patients with adequate and optimal yields are 76% and 50% respectively, if CD34+ cell purification is to be carried out then it must be anticipated that approximately 50% of the progenitor cells will be lost during processing. The minimum threshold required to consider selection feasible is therefore 2 x10®/kg CD34+ cells. Fifty three percent of patients obtain this from one harvest and 76% if 2 harvests are pooled, a process that allows a single clinical scale immunoaffinity column to be used. The fact that one quarter of patients do not have adequate cells for processing from two harvests emphasizes the need to improve upon mobilization protocols or mobilize before extensive therapy and to improve upon progenitor cell recoveries from CD34 purification procedures.

89 Chapter 5 Variability in Progenitor Numbers Mobilized by G-CSF in Normal Subjects

5.0. INTRODUCTION Granulocyte colony-stimulating factor (G-CSF) is a growth factor that stimulates the latter stages of neutrophil development and also augments some of the functional properties of these cells (Metcalf, 1990; Rapoport etal., 1992) and is widely used in clinical practice in the treatment of neutropenic conditions. G- CSF administration also results in the mobilization of peripheral blood stem cells and it is the most commonly used growth factor for this purpose. It is usually given following chemotherapy to maximise progenitor yields but is also given alone to some patients and for nearly all normal donors. An important consideration in assessing the efficacy of mobilizing regimens which use this growth factor is the extent of inter-individual variation in response to G-CSF. In a study by Roberts for example a broad range with at least a 10-fold variability was noted in G-CSF mobilization of GM-CFC in pre-treated cancer patients, untreated cancer patients but also in healthy subjects (Roberts at a!., 1995). Similar variability has been noted in the clinical experiences of G-CSF mobilization of normal donors and the incidence of “poor mobilizers” in the normal population is estimated to be around 5-10% (Bishop at a!., 1997; Wiesneth at a!., 1998).

The type of recombinant G-CSF product used may also affect mobilizing efficiency. Native G-CSF is a glycoprotein with a molecular mass of 18-21 Kd, consisting of 174 amino acids (Souza at a!., 1986) and is 0-glycosylated at residue threonine 133 (Kubota at al., 1990). The sugar moiety accounting for approximately 4% of the total weight Two recombinant forms of G-CSF are in wide clinical use. A glycosylated form (lenograstim, Chugai Pharma) which is

90 made in Chinese Hamster Ovary (OHO) ceiis and is giycosyiated on the same site as the naturaiiy occurring moiecule and a non-glycosyiated form (filgrastim, Amgen) which is produced in E. Cell. Glycosyiation can have significant effects on cytokine function. The ncn-giyccsyiated form of erythropoietin is equipctent to the glycosylated form in vitro but has minimal in vivo activity owing to rapid clearance by the liver (Fukuda et al., 1989). Glycosyiation of GM-CSF is reported to decrease its affinity of binding to its cognate receptor (Cebon et ai., 1990) and would thus be expected to reduce its biological potency. In vitro studies with G-CSF suggest that the glycosylated form is more stable than the non-glycosylated form in serum (Kishita et al., 1992) and has greater potency in in vitro colony cultures (Nissen et ai., 1994; Pedrazzoli et ai., 1996). This could be due to differential stability during the culture period but short term studies of the effects of G-CSF on neutrophil priming have suggested greater potency of the giycosyiated product in vitro (Decleva et al., 1995). Studies in our own laboratory have however shown identical potency of the two forms of G-CSF with regard to neutrophil stimulation and priming (Pizzey, 1998). A formal intra­ individual, randomised comparison of giycosyiated G-CSF with non- glycosylated G-CSF, both at a dose of 10pg/kg/day, was made in normal volunteers (Hoglund et ai., 1997) and demonstrated superior stem cell mobilization with glycosylated G-CSF.

In the current study the progenitor mobilizing effects of glycosylated and non- glycosylated G-CSF at a dose of Spg/kg/day was compared with concomitant pharmacokinetic studies to ascertain whether any possible differences in biological function observed were due to drug availability and to determine whether previously described inter-individuai variation in progenitor cell mobilization is related to G-CSF concentrations.

91 5.1. METHODS AND MATERIALS 5.1.1. Volunteers. 20 healthy male volunteers age 20-35 years were entered into this study. All were normal on physical examination and routine haematological and biochemical screening. Individuals were hospitalised for the two periods of G-CSF administration. The study was approved by the local research ethics committee and each volunteer gave written informed consent.

5.1.2 Study design. The study design is outlined in Figure 5.1.

20 Normal Male Volunteers

Randomise ( double blind )

10 receive 5)ig/kg/day lenograstim 6 days 10 receive 5pg/kg/(^y filgrastim 6 days 4 week “wash-out”I period “Cross-over” Repeat mobilisation with alternative G-CSF

Figure 5.1 Flow chart of comparative G-CSF study

The 20 individuals were randomised in a double blind fashion to receive either glycosylated G-CSF (lenograstim) or non-glycosylated G-CSF (filgrastim) at a

dose of 5pg/kg/day subcutaneously for 6 days with G-CSF given at 0900hr each day. Following a minimum 4 week washout period each volunteer was given an identical schedule of the alternative G-CSF preparation. Baseline full

92 blood counts, GM-CFC and CD34+ measurements were made immediately before the first G-CSF administration (day 1) and on days 4, 5 and 6 immediately before the next dose of G-CSF. Samples were also taken at the same time of day on day 7. Serum samples for G-CSF assay and pharmacokinetic analysis were taken on day 1 at time points 0, 1,2, 3, 4, 6 , 8 , 10, 12, 16 and 24 hours and on day 6 at 0, 4, 8 and 24 hours in relation to dosing. Adverse events were recorded.

5.1.3 Progenitor Cell Measurements.

Low density peripheral blood mononuclear cells were plated in triplicate at 5 x lOVmL or 2.5 x lOVmL in a commercially available methylcellulose-based clonogenic assay medium (Terry Fox laboratories, Metachem, UK) containing added recombinant growth factors as described in the general materials and methods chapter. To ensure consistency a single batch of reagents was used throughout the study. GM-CFC and burst-forming units-erythroid (BFU-E) were simultaneously evaluated after 14 days incubation at 37°C in a humidified 4% COg atmosphere. All cultures were scored blindly to eliminate observer bias.

Cytocentrifuge preparations were made from approximately 0.5 x 10® cells for each peripheral blood mononuclear cell sample and stored at -20°C for batch staining with a CD34 monoclonal antibody (HPCA-2, Becton Dickinson) and alkaline phosphatase-anti-alkaline phosphatase (APAAP) colour development as described in the general materials and methods section. A "Miller's squares" eyepiece (Graticules Ltd, Tonbridge, Kent, UK) was used to count the

number of stained cells in a total of 10,000 cells examined.

93 5.1.4 Serum G-CSF Assay. Serum G-CSF concentrations were measured by chemiluminescent immunoassay as described by Kiriyama et al (Kiriyama et a!., 1993). These were performed for this study by Mr K Chichibu, Chugai Pharmaceuticals, Tokyo, Japan. In this assay lenograstim and filgrastim standards give identical dose response curves.

5.2. RESULTS 5.2.1 Baseline Values. There was no significant difference in the baseline blood counts and progenitor cell levels between those randomised to receive lenograstim first and those randomised to receive filgrastim first.

5.2.2 White Blood Cell count. There was as expected a marked rise in the white blood cell (WBC) count by day 4 with similar values achieved on days 5 and 6 . The mean white cell count was higher on day 7 (24 hours after administration of the last dose of G-CSF) than on days 4-6. On all days the total WBC during glycosylated G-CSF was slightly higher than during non-glycosylated G-CSF and the difference in the peak value was significant (p=0.02) (Figure 5.2.1). The difference in WBC counts was largely attributable to differences in neutrophil counts, for which the peak values were significantly different (p=0.03) (data not shown).

5.2.3 Progenitor Cell Counts.

The peak counts of GM-CFC occurred on day 5, 6 or 7 in different individuals (Figure 5.2.2). The peak counts were higher during lenograstim treatment

(p=0.03). The average count of GM-CFC in each individual on days 5, 6 and 7 was also calculated, since progenitor cells are usually collected over a 2 or 3 day period. The mean count achieved with glycosylated G-CSF was 28% higher than with non-glycosylated G-CSF (p=0.003) (Table 5.1).

94 The peak counts of CD34+ cells were also achieved on days 5, 6 or 7. The mean peak level of CD34+ cells was 12% higher with glycosylated G-CSF; but the difference from non-glycosylated G-CSF was not significant (Figure 5.2.3).

The mean CD34+ counts on days 5, 6 and 7 were 17% higher with glycosylated G-CSF than with non-glycosylated G-CSF but the difference did not achieve significance (Table 5.1).

Table 5.1. Mean peak white cell and progenitor counts on days 5, 6 and 7 foliowing glycosylated or non-glycosylated G-CSF

G-CSF type WBC GM-CFC CD34+ cells (5p,g/kg/day) x10^/L per mL blood per mL blood Glycosylated (lenograstim n=20) 40.5 6,037 53,637 Non-glycosylated 37.6 4,705 45,964 (filgrastim n=20) p value 0.02 0.003 A/S

5.2.4 Pharmacokinetics. The concentration-time curves for day 1 are shown in Figure 5.3 for lenograstim

and filgrastim. The area under the curve (AUCo. 24h) was 18% higher with filgrastim, a difference that was significant at the 5% level. C^^^ was 20% greater with filgrastim which was also significant at the 5 % level. There was no significant difference between the mean elimination half lives (Table 5.2).

On day 6 serum concentrations of both drugs were approximately one third of those on day 1 (p<0.001) as judged by the AUC calculated using data from

time points 0, 4, 8 and 24 hours. The mean values were again higher for filgrastim than lenograstim (49,871 vs 39,363 pg/mL.hr) but at the lower levels

on day 6 the difference was not significant.

95 Table 5.2. Pharmocokinetic parameters for non-glycosylated and glycosylated G-CSF after a single dose of Spg/kg

Non-Glycosylated Glycosylated (filgrastim) (lenograstim) Ratio (95%Ci)

AUCo.24hr (pg/mL.hr) 120,451 102,334 1.18 (1.08-1.29)

Cmax(Pg/mL) 14,228 11,849 1.20 (1.07-1.36)

ti/2 (h) 2.75 2.95 1.07 (0.87-1.06)

5.2.5 Individual Variability. It is noteworthy that there was marked inter-individual variation in progenitor cell mobilization with both G-CSF types (Figures 5.4.1 and 5.4.2) and there was a greater than ten-fold difference between the best and the worst mobilizers. In contrast, intra-individual variation in mobilization with the two compounds was generally small (Figure 5.5). It is notable that the individual with the poorest mobilization in the group appears with the lowest blood GM- CFC count /mL on day 7 after first mobilization with glycosylated G-CSF (Figure 5.4.1) and again with the lowest day 7 blood GM-CFC count after “washout” and mobilization with non-glycosylated G-CSF (Figure 5.4.2).

There was no significant correlation between the extent of progenitor cell

mobilization and serum concentrations of G-CSF on day 1 or day 6 as assessed by AUCo. 24h and C^^^ values (Figure56).

5.2.6 Adverse events. Bone pain and arthralgia were the most common adverse events and were equally common during the two treatments. They were controlled with paracetamol.

96 50-1

40-

O) G-CSF type o 30- X Glycosylated U 2 0 - (lenograstim) n =20 CÛ Non-Glycosylated

10 - (filgrastim) n =20

012345678

10000-1 X3O B O 7500- _l E S 5000- O à 2500- (3

0 1 2345678 TD 80000 -I O 0

1 60000 - (f) "Ô) U 40000 - + s Û 20000 - o

012345678

Figure 5.2. Serial white cell (A), GM-CFC (B) and CD34+ cell counts in 20 normal subjects following mobilization with the two types of G-CSF

97 16000

14000

12000 Non-glycosylated G-CSF ^ (filgrastim) c

^ 10000 Glycosylated G-CSF o (lenograstim) 6 8000 E ZJ 6000 (D cn 4000

2000

0 4 8 12 16 20 24

Time (hours)

Figure 5.3. Mean serum G-CSF levels after the first dose of G-CSF in 20 normal subjects alternately mobilized with two types of G-CSF at 5ug/kg s.c.

98 1 0 0 0 0 0 n

10000- ■o o 0 -û _J 1000 4 E O LL O1 100 - 0

10^

1 - h f - Day 100000-1 7

B 10000 - TD O 0

1000 - E O LL O1 100 - O

//

10 -

■ i - t ■ 1 4 7

Figure 5.4. Individual GM-CFC counts in 20 normal subject mobilized alternately with A) glycosylated and B) non-glycosylated G-CSF types.

99 15000n

■D c 05 CO lO CO

■O 10000 - "O o 0 8 E O LL 5000 ^ O1 o c 05 CD E

Non-Glycosylated Glycosylated

G-CSF type given

Figure 5.5. Intra-individual progenitor mobilization in 20 normal subjects mobilized alternately with two formulations of G-CSF

100 G-CSF type: Glycosylated (lenograstim) Non-Glycosylated (filgrastim)

200000

S' 150000- •D

C\J

100000-

50000 -

T------r 5000 10000 15000

200000

CD >. CO 150000-

100000-

5 2 50000 -

0 5000 10000 15000

Average GM-CFC /ml blood (days 5, 6 and 7)

Figure 5.6. No relationship was seen between serum G-CSF concentration achieved and subsequent peak progenitor levels in 20 normal subjects mobilized alternately with the two types of G-CSF

101 5.3. DISCUSSION In this study lenograstim was more potent than filgrastim on a weight for weight basis in terms of the rise in white cell count and the mobilization of GM-CFC. Our data are in accord with data previously reported (Hoglund et a/., 1997) comparing the two forms of G-CSF at a dose of 1 0 pg/kg/day. In both studies the absolute difference in white cell counts was small indicating that this is a poor discriminator of effective G-CSF dose delivered. In the Hoglund study lenograstim was significantly more potent than filgrastim at CD34+ cell mobilization, whereas in our study the trend to greater CD34+ cell mobilization with lenograstim was not significant. This may be because we used a slide- based immunoassay which is less precise than flow cytometry because fewer cells are counted; however the slide-based method has the advantage that the samples can be batch-stained to avoid assay drift.

The difference in potency observed by the Hoglund study and ourselves is small in terms of progenitor mobilization and would only be of clinical relevance in those patients with relatively poor mobilization whose collections are borderline for the progenitor threshold requirements. In the autologous PBSCT clinical setting a peripheral blood CD34+ count of < 20 x 10®/L is often taken to indicate a poor mobilizer (Armitage et al., 1997; Elliott et al., 1996). Using this criterion in our study, 2 patients mobilized inadequately with lenograstim and 4 with filgrastim. In general, intra-individual variation in mobilization with the two formulations was around 30% or less and the 2 patients that failed to mobilize with lenograstim also failed with filgrastim.

Glycosylated G-CSF is more resistant to degradation in serum than non­ glycosylated G-CSF (Kishita et ai, 1992) and it might be expected that the greater potency of lenograstim might simply be related to higher serum levels. That is clearly not the case however since concentrations of filgrastim on day 1 were significantly higher than those of lenograstim. It is possible that the

102 greater potency of glycosylated G-CSF is due to greater affinity of this form of G-CSF for its receptor. Such a difference in receptor affinity might also explain the paradoxical finding of lower serum concentrations of lenograstim than filgrastim. At peak concentrations of G-CSF of about 600pM it would be expected that only about two-thirds of G-CSF receptor sites would be occupied (Khwaja et al., 1993a) and a change in receptor affinity could alter receptor occupancy considerably and thus modulate serum levels. It is noteworthy that as previously described the serum concentrations after 6 days of G-CSF administration were markedly lower than on day 1, probably owing to the increased granulocytic cell mass and consequent removal of a greater proportion of ligand from the circulation (Kuwabara et a!., 1996; Stute et a!., 1992) However, the binding of the two forms of G-CSF to neutrophils has been formally examined and there is no difference (Pizzey, 1998). This raises the possibility then that the difference in efficiency is related to G-CSF compartmentalisation. If glycosylation of G-CSF increased binding to the marrow micro enviroment, then its efficiency on bone marrow progenitors might be greater and serum levels lower. It is well known that many cytokines bind to glycosoaminoglycans found in the stromal compartment of bone marrow resulting in enhanced haemopoietic function (Gupta et a!., 1996; Roberts et a!., 1988; Siczkowski et a!., 1992) but a formal study of this mechanism for G-CSF activity specifically has not been published to date.

Our study clearly illustrates that there is marked inter-individual variation in progenitor cell mobilization, with more than a 10-fold difference between the best and the worst mobilizers that is in accord with previous reports (Hoglund et a!., 1996; Jones eta!., 1996; Roberts et a!., 1995). The difference is unlikely to be due to the occurrence of minor illnesses as volunteers in general mobilized to a similar degree at intervals of at least 4 weeks apart. In contrast a smaller degree of intra-individual variability was seen and notably the poorest

103 mobilizer on first exposure to G-CSF was also the worst on the second occasion. Low intra-individual variability has also been described by (Anderlini et al., 1997a) who reported the same progenitor yields in 13 normal donors who were remobilized identically at least 1 month after the original G-CSF protocol. The reason for wide inter-individual variability is not known but this study shows that it is unlikely to be due to variations in metabolism that could modulate serum G-CSF concentration. A recent study of poor G-CSF mobilization in 53 normal donors mobilized with 5|ig G-CSF/kg/day showed low steady-state lymphocyte or monocyte counts to be a significant predisposing factor and raised the possibility of a genetic component shown by the absence of HLA-A1 in some poor mobilizers although this was not achieve significance (p=0.085) (Bishop et al., 1997). We have mobilized 40 male volunteers with this dose of G-CSF for at least 5 days, twenty of whom were subjects in our cross-over study and remobilized, giving data for a total of 60 such procedures in healthy subjects. We have not found a correlation between baseline lymphocyte, monocyte or neutrophil counts and day 5 blood CD34+ cell or GM-CFC counts in the whole group or in subgroups such as G-CSF type or first mobilization data only. However the Bishop study defined a “poor mobilizer” by the inability to collect 3 xIOVkg CD34+ cells and 6.5 x10®/kg MNC in two apheresis collections and did not use circulating progenitor levels. Based on the assumption that the steady-state number of circulating CD34+ cells is a reflection of the available mobilizable pool. Brown et ai used pre­ cytokine blood CD34+ cell counts to predict CD34+ cell apheresis yields in 47 normal donors. The minimum collection target set was 2.0 x10®/kg recipient wieght within three aphereses. (Brown et a!., 1997). The donors were mobilized with G-CSF at 10jig/kg/day sc and collected from day 5 using a large volume leucapheresis protocol in which 18-20 litres of blood was processed. The target CD34+ cell yield was achieved with one apheresis harvest in 42 donors, 3 required two collections, one donor three collections and in one donor in whom less than 1 xIOVkg CD34+ cells was collected in two

104 collections, a bone marrow harvest was collected. There was a strong correlation between the donor baseline CD34+ cell count in the blood which varied from 0.8-11.4 x10®/L (mean 4.0 x10®/L) and the pre-apheresis count which showed a mean of 114 (range 9.5-320 x10®/L), P=0.0008. The total dose of G-CSF received also correlated with the pre-apheresis CD34+ cell count (P=0.0002). However these two factors together accounted for less than half of the variation in GD34+ cells mobilized and imply that other, as yet undefined factors play an important role.

This variability in mobilization has obvious clinical relevance and a major issue will be whether this innate variation will be reduced by higher doses of G-CSF or the introduction of new approaches such as cytokine combinations.

105 Chapter 6 The Role of Back up Bone Marrow for "Poor Mobilizers"

6.0 INTRODUCTION High dose therapy with autologous haemopoietic support has been shown to be of value in a range of malignant disorders (Attal et al., 1996; Bezwoda et a!., 1995; Haioun et a!., 1997; Linch et a!., 1993; Philip et a!., 1995). The early studies used bone marrow as the source of stem cell support and a minimum number of 1 x10®/kg MNC was typically used. There has been considerable debate concerning the threshold numbers of bone marrow GM-CFC required, with GM-CFC numbers infused being predictive of slow engraftment in some studies but not in others. A correlation between the dose of GM-CFC infused and neutrophil recovery was first reported by Spitzer (Spitzer et a!., 1980) and minimum GM-CFC threshold values of between 0.1 - 1.0 xIOVkg have since been quoted by various authors (Arnold eta!., 1986; Gorin, 1986). In our centre, a minimum of 2 xIOVkg GM-CFC post freeze-thawing of a test aliquot has been considered as adequate to proceed to high dose therapy.

More recently, bone marrow has been largely replaced by mobilized peripheral blood stem cells as the source of autologous haemopoietic stem cells as it leads to faster engraftment particularly of the platelet lineage, with more rapid discharge from hospital and attendant cost savings (Beyer et a!., 1995; Jones et a!., 1994; Schmitz eta!., 1996; Smith eta!., 1997). There is a considerable consensus that there is a threshold requirement for peripheral blood progenitor cells assessed either by CD34+ enumeration or by GM-CFC (Bensinger et a!., 1995; Demirer & Bensinger, 1995; Watts & Linch, 1997; Weaver et a!., 1995). In a previous study of engraftment of 81 patients receiving BEAM chemotherapy (Watts & Linch, 1997) the optimal number of

106 CD34+ cells and GM-CFC was shown to be 3.5 x107kg and 3.5 x107kg respectively, above which values delayed haematological recovery was very rare. A small proportion of patients receiving between 1-3.5 xIOVkg CD34+ cells or 1-3.5 xIOVkg GM-CFC had delayed platelet recovery but below the minimum threshold levels of 1 xIOVkg CD34+ cells or 1 xIOVkg GM-CFC, the risks of delayed platelet recovery was very high (Watts & Linch, 1997). Although GM-CFC and CD34+ cell values are highly correlated, (Serke et al., 1996) in our institution, multivariate analysis has shown that GM-CFC levels are more predictive of haematological recovery. Some patients will not achieve the minimum CD34+ cell threshold but will exceed the GM-CFC threshold and it has in general been our policy to proceed to high dose therapy using the PBSC support in these patients. In patients where neither the CD34+ cell or the GM-CFC minimum thresholds have been obtained, a bone marrow harvest has usually been performed and used with or without PBSC as haematological support.

A number of factors have been shown to predict for a poor PBSC yield including prior radiotherapy, marrow infiltration or fibrosis and extensive prior chemotherapy (Bensinger eta!., 1995; Haas at a!., 1994a; Passes Coelho at a!., 1995; Watts & Linch, 1997). The type of prior chemotherapy may also be relevant and in a study by Dreger the use of the stem cell toxic Dexa-BEAM regimen was found to be associated with poor subsequent PBSC harvests (Dreger at al., 1995). This finding was confirmed subsequently using the similar, miniBEAM regimen (Weaver ef a/., 1998). Prior Dexa-BEAM therapy in the Dreger study did not negatively influence bone marrow harvest yields, however, suggesting that some pre-mobilization chemotherapy regimens may cause a specific defect in progenitor cell mobilization.

107 In this study the engraftment data in patients receiving less than 1 xIOVkg CD34+cells and in particular those patients who had bone marrow collected and reinfused after a “failed” PBSC collection is the primary focus. Delayed haematological recovery was frequent in this latter group suggesting that the poor PBSC harvest reflected an impaired bone marrow rather than a specific defect in progenitor cell mobilization.

6.1 PATIENTS AND METHODS 6.1.1 Patients studied. Between June 1992 and March 1996, G-CSF mobilized peripheral blood stem cells were harvested on 324 consecutive patients with non-leukaemic malignancies. The patient diagnoses, median age and sex, median number of courses of prior chemotherapy and the proportion having received prior radiotherapy are shown in Table 1. No patient had bone marrow disease as assessed by trephine biopsy within the 4 weeks proceeding PBSC mobilization, except for the myeloma patients who had <15% infiltration with plasma cells at the time of mobilization.

6.1.2 Mobilization regimens. In the majority of patients (310 out of 324), priming chemotherapy was used in addition to G-CSF with most receiving cyclophosphamide 1.5g/m^ with G-CSF (273 patients) starting the following day and continuing until harvesting was completed (Table 6.1).

108 Table 6.1 PBSC mobilization regimens and patients studied (n=324)

Median(range) of Proportion Diagnosis Patient prior chemotherapy receiving prior number courses radiotherapy* Hodgkin’s disease 114 8(1-25) 51% NHL : Low grade** 36 7 (4-20) 19% High grade 96 6(1-18) 16% Myeloma 37 6 (2-36) 53% Solid tumour 41 4 (0-17) 22% Age: median (range) 40(13- 69) Sex : maleifemale 204:120

Mobilization regimens : Patient No. G-CSF (filgrastim) alone 14 (14%) 10ug/kg/day Cyclophosphamide 1.5g/m^ and G- 164 (48%) CSF (filgrastim) at 10ug/kg/day Cyclophosphamide 1.5g/m^ and G- 109 (33%) CSF (lenograstim) 263ug/day ESHAP + G-CSF (lenograstim) 13 ( 4%) 263ug/day Other 24 (12%)

*Locai or wide fieid * *One month continuous chiorambucil counted as one course of chemotherapy

6.1.3 PBSC Harvesting. In the 273 patients collected after cyclophosphamide and G-CSF, PBSC were harvested between days 8 - 1 2 after the cyclophosphamide usually starting on day 10 when the white cell count was rising from the cyclophosphamide induced nadir. 13 patients were mobilized with ESHAP and G-CSF and harvested on a rising white cell count of >3.0 x10®/L, usually on or after day 15. Fourteen collections were performed on an intermittent cell separator (Haemonetics V50, Haemonetics Ltd, Leeds, UK) and the remainder were collected on a continuous flow cell separator (COBE Spectra, COBE laboratories Ltd., UK or Baxter CS3000, Baxter Healthcare Ltd., UK). The number of aphereses performed varied between 1 - 4 depending on progenitor

109 cell numbers harvested, although it was unusual to perform more than 3 aphereses even if the progenitor cell yield was poor.

6.1.4 Bone Marrow Harvesting. Bone marrow was harvested from the pelvis and sternum under general anaesthetic between 12 and 79 days after a failed PBSC collection, at a time when the full blood count was in the normal range. Between 800-1000ml of marrow was obtained in each case.

6.1.5 Cryopreservation. Bone marrows were cryopreserved in homologous ABO-compatible, fresh frozen plasma with the addition of 10% dimethyl sulfoxide (final concentration) using controlled rate freezing and stored in a liquid nitrogen freezer (Jones et a/., 1994) Peripheral blood stem cell harvests were cryopreserved either as described above or, in 5% pentastarch with 4% human serum albumin and 5% dimethyl sulfoxide (final concentration) as previously detailed (Makino et al., 1991).

6.1.6 Progenitor cell assays. Full blood count (STKS® Coulter Inc., Hialeah, FL), manual white cell differential and progenitor measurement by CD34+ cell and CFC assays were performed on all apheresis products. CD34+ cell numbers were determined with a CD34 monoclonal antibody (HPCA-2, Becton Dickenson) and a slide- based alkaline phosphatase anti-alkaline phosphatase (APAAP) staining method or flow cytometry. These measurements and granulocyte / monocyte- colony forming cells (GM-CFC) were performed as detailed in Chapter 2 . Samples for colony assays were plated in triplicate, at concentrations of 2.5 xIOVml for PBSC products and 5 xIOVml for thawed bone marrow samples. Colonies were counted after 14 days incubation at 37°C in a humidified 4% COg atmosphere.

1 1 0 6.1.7 High Dose therapy regimens. A total of 40 patients with a PBSC collection of <1 x10® CD34+ cells/kg went on to high dose therapy (Figure 6.1). The high dose regimens used were; 25 patients received the BEAM protocol (BCNU 300mg/m^ iv. on day 1, etoposide 200mg/m2/ day iv days 2 - 5, (total dose 800mg/m^ ) cytosine arabinoside 200mg/m2 b.i.d on days 2 - 5 (total dose 1600mg/m^) and melphalan

140mg/m^/V. on day 6); seven patients received cyclophosphamide and total body irradiation (TBI) ; four melphalan and TBI ; one patient received melphalan and VP16; one patient received melphalan and thiotepa and two patients received high dose melphalan only.

PBSC collected from 324 patients

< 1 xio^kg CD34+ cells

51 patients

28 patients 23 patients < 1 x10®/kg" >1 xlO^kg GM-CFC GM-CFC

18 high dose 22 high dose therapy therapy (12 with BM) (2 with BM)

Includes 1 pa tient with failed culture assay

Figure 6.1 Patients who failed PBSC mobilization and went on to receive high dose therapy

111 6.1.8 Haemopoietic Recovery. Recovery was assessed from the time of PBSC infusion (24 hours after completing high dose chemo/radiotherapy) to the day on which a sustained neutrophil count of > 0.5 x 10^/1 was achieved. Platelet recovery was defined as the number of days before platelet independence was achieved with a platelet count in excess of 15 x107l without evidence of bleeding. Slow recovery was arbitrarily defined as recovery to these parameters between 21 and 28 days and delayed recovery as recovery beyond 28 days as described in Chapter 3.

6.2 RESULTS 6.2.1 Overaii haematoiogicai recovery Peripheral blood stem cells were collected on 324 consecutive patients and in 273 (84%) the total progenitor collection exceeded 1 x10®/kg CD34+ cells. Two hundred and fifteen of these patients went on to high dose therapy and had > 1 x10®/kg CD34+ cells returned. The median time to neutrophil recovery was 11 days (range 6-27) and the median time to platelet independence was 11 days (range 5-68). Six patients (3%) had slow neutrophil recovery but in no case was this delayed beyond 28 days. Sixteen patients also had slow (n=7) or delayed (n=9) platelet engraftment. In those individuals receiving between 1 - 3.5 x10®/kg CD34+ cells there were 8/102 with delayed recovery compared to 1/112 patients who received more than 3.5 xIOVkg CD34+ cells (p= 0.0074).

6.2.2 Haematoiogicai recovery of the poorest mobilizers In 51 patients (16%) a total collection of less than 1 x10®/kg CD34+ cells was collected and the diagnoses of those patients failing to achieve this threshold are shown in Table 6.2.

112 Table 6.2 Proportion of patients with a poor total PBSC collection of less than 1 x10®/kg CD34+ cells

All patients with GM-CFC dose In patients with Malignancy <1 x1(f/kg <1 x1(f/kg CD34+ cells: CD34+ cells <1 x1CP/kg >1x1CF/kg

Hodgkin’s disease 17/114(15%) 9/114 ( 8%) 8/114 ( 7%) NHL: Low grade 10/ 36(28%) 8 / 36 (22%) 2 / 36 ( 6%) High grade 12/ 96(13%) 6 / 96( 6%) 6 / 96 ( 6%) Myeloma 10/ 37 (27%) 4 / 37(11%) 6 / 37(16%) Solid tumours 2 / 41 ( 8%) 1 / 41 ( 4%) 1 / 41 ( 4%)

Totals 51 /324 (16%) 28 /324 ( 9%) 23/324( 7%)

Of these 51 patients, 23 had >1 x10®/kg GM-CFC which we have previously reported as adequate to ensure engraftment. One of these 23 patients did not proceed to high dose therapy however, due to poor clinical condition. Bone marrow was collected and used in two patients who went on to high dose therapy but not in the other 20 high dose therapy recipients. Bone marrow was collected in these 2 patients within two weeks of the PBSC failure as the

CD34+ cell count was <0.5 x 10®/kg and poor GM-CFC yields had been anticipated. In this group of 22 autograft recipients there were two procedure related deaths, one from pulmonary haemorrhage on day 2 and one on day 6 from sepsis. In the 20 patients assessable for engraftment the median time to neutrophil recovery to 0.5 x 10^/1 and platelet independence were 15 days and

19 days respectively (Table 6.3). If the revised minimum GM-CFC dose of 2 xIOVkg outlined in Chapter 3 were applied to this group, four of these patients would have exceeded this. It is notable that only one of these patients experienced slow platelet recovery at 22 days. None of the 20 patients had slow neutrophil recovery, the longest time being 19 days. However acquisition of platelet independence was delayed beyond 21 days in 10 patients (50%) and beyond 28 days in 8 (40%). Platelet independence was however achieved by 6 weeks in all but 4 patients (57, 74, 80 and >120 days). Bone

113 marrow had been given with PBSC in the patient not regenerating to platelet independence until day +80 and in the patient with the longest delay, back-up marrow was re-infused at 120 days because of poor platelet engraftment. This patient is alive at day 500 but maintains a fragile platelet count of around 20

X 10^/1.

Table 6.3 Haematological recovery in patients receiving high dose therapy and re-infused with varying numbers of haemopoietic progenitor cells

PBSC yields Source of Days of Days to Days to Transfusion progenitor ceiis Neufs. Neufs. platelet Requirements < 0.1X1CF/I >0.5x10^/i Indep. red cells pits (units) (days)

>10F CD34+cei!s PBSC 4 11 11 4 4 n=221

<1(f CD34+cells PBSC 5 15 19 6 6 >10^ GM-CFC (n=18) n=22* PBSC + BM (6, 6) (12, 17) (80, >500) (5, 13) (15, 20) (n=2)

PBSC 8.5 18.5 42 13 24 <1(f CD34+ cells (n=6) <10^ GM-CFC n=18** BM ±PBSC 7 15 >47*** 20 20 (n=11*)

Median days to recovery shown where possble. * Two patients died at day+2 and day+ 6, ""One patient died at day+ 8, Unassessable median, 5/11 patients died while platelet dependent

6.2.3 Back-up bone marrow use

In twenty seven of the 51 patients with a total harvest <1 xIOVkg CD34+ cells, the GM-CFC yield was below our previously accepted minimum requirement of 1 x10^/kg and a further patient with low CD34+ cells was placed in this category because of a culture assay failure. A bone marrow harvest was subsequently performed in 18 of these patients who were still considered eligible for high dose therapy. In the 10 in whom marrow was not collected, high dose therapy

114 with autologous stem cell rescue was abandoned in 6. In the four remaining patients however high dose therapy was given with just the “poor quality PBSC” returned. This was early in the series at a time when the the minimum threshold had not been fully clarified. It should be noted that GD34+ cell numbers were only marginally low in 2 of these patients at 0.9 x 10®/kg but were well below our threshold at 0.6 and 0.3 x 10®/kg in the other two. One patient with a CD34+ infusion of 0.9 x10®/kg recovered promptly after a melphalan / TBI graft (14 days to ANC >0.5 x10®/l and 17 days to platelet independence) but two of the remaining three had slow or delayed neutrophil recovery (23 and 39 days) and all three had delayed platelet recovery (39, 45 and 85 days).

In seventeen of the 18 patients who had a marrow harvest, an adequate MNC count of >1 x10®/kg was collected (median 1.8 x 10®/kg, range 0.9 - 2.9 x10®/kg). Four patients did not proceed to high dose therapy on clinical grounds including the one patient with a bone marrow MNC count of < 1 x 10®/kg. Of the

14 patients who went on to high dose therapy, 10 received PBSC and BM, 2 received BM alone and 2 received PBSC with the bone marrow held in reserve. There were thus a total of 6 patients with CD34+ cells and GM-CFC below the minimum thresholds who received PBSC alone and 5 of these had delayed platelet engraftment. In the 12 patients who received bone marrow, the haematological recovery was generally poor (Tablets and Table#). One patient died whilst pancytopaenic on day 8 from a perforated gastric ulcer and sepsis and in the remaining evaluable patients the median time to a neutrophil count >0.5 x107l was 15 days and to platelet independence was unassessable due to protracted delays in engraftment (Table63 and Table#). Five patients have died between days 47 - 200, one of disease progression at day 200, one of a fungal infection on day 47, one of thrombopaenic associated pulmonary haemorrhage on day 78 and the other two of sepsis on day 58 and day 107. All were platelet dependent at the time of death and a further patient survives

115 beyond 400 days but is still platelet dependent. As a consequence of the poor engraftment the blood transfusion requirements in this group were very high (Table 6.3 ).

Of the 18 patients with less than 1 x1 OVkg GD34+ cells and <1 x10®GM-CFC/kg in their PBSC harvests who went on to have a bone marrow harvest, 12 had test vials available for assessment of the number of GM-CFC in the post thaw cryopreserved bone marrow (Table44). Nine of these patients were those who went on to receive high dose therapy with bone marrow ( ± PBSC) support.

Five received >2 xIOVkg GM-CFC and 4 less than this value. Slow or delayed haematological recovery was not related to the number of bone marrow GM- CFC.

116 Table 6.4 Patients with < 1x10®/kg CD34+ ceiis and < 1x10Vkg GM-CFC in whom Bone Marrow was subsequently harvested

Stem cell UPN PBSC BM Haematological Blood source collections harvests recovery Support No CD34 GM- MNC GM- Days Days Day to RBC Days ofaph 10®/kg CFC 10®/kg CFC neuts to pit indep. units of pits 10®/kg 10®ykg < 0.1 neuts (thaw) x10®/i >0.5 x10®/i 0837 4 0.1 1.0 1.9 5.6 8 14 15 5 6 BM + 0961 3 0.3 ND 1.7 0.1 10 30 > 400 84 81 PBSC (n=10) 0971 3 0.1 0.9 1.8 0.2 6 15 >107t 3 20 1047 3 0.7 7.0 2.9 24.8 8 12 >78t 29 78 1074* 2 0.5 6.0 2.2 ND -- -- - 1110 2 0.3 6.4 1.6 ND 11 15 34 12 13 1116 2 0.5 9.5 2.3 3.8 3 11 13 5 5 1121 3 0.6 7.9 1.7 0.3 2 11 8 5 5 1128 2 <0.1 <0.1 1.8 1.0 25 50 >58t 82 32 1181 2 0.3 2.7 1.3 ND 3 26 70 39 69

BM only 1041 2 <0.1 <0.1 1.4 3.8 7 23 >47t 67 22 (n=2) 1049 2 <0.1 <0.1 1.9 5.6 3 15 >200t 20 20

PBSC 0910 3 0.1 3.5 1.1 ND 10 24 59 22 58 only (BM in 0982 3 0.8 3.6 2.5 1.3 7 12 30 9 20 reserve)

Patients in SHE 2 0.0 0.8 0.9 ND whom high KLA 3 0.6 1.9 2.3 1.4 -- - - - dose ACU 3 0.1 3.0 2.3 ND -- ---

was not MOB 2 0.0 2.6 1.6 2.6 - - - - given

UPN = unique patient number. ‘ Recovery data unasessabie on UPN 1074, patient died on day + 8. t = patients who died while platelet dependent (see text for details).

117 60 n

O) 50-

40-

Q. 30-

2 0 -

10 -

100-1

90-

8c 80- TD0) C 70- 0) CL ■a(1) 60- Ç o 50- 0 % CL 40- 2 30- 1 Q 2 0 -

10 -

GM-CFC < 1 > 1 1-3.5 >3.5 x105/kg CD34 less than 1 1-3.5 >3.5 X 106/kg

Progenitor dose returned

Figure 6.2 Distribution plots of haematological recovery post PBSCT at various progenitor thresholds and days to (A) neutrophil recovery and (B) platelet independence (n=252)

118 6.3 DISCUSSION There is general agreement that a threshold number of progenitor cells are required to achieve rapid engraftment after PBSC transplantation, although there is some variation between centres in the definition of these thresholds (Demirer & Bensinger, 1995). In our institution we have previously defined a minimum threshold of 1 x10®/kg CD34+ cells or 1 xIOVkg GM-CFC as the minimum threshold to allow engraftment and 3.5 x1 OVkg CD34+ cells or 3.5 xIOVkg GM-CFC as the optimal threshold to ensure rapid engraftment of neutrophils and platelets within 21 days (Watts et al., 1997c). In our report of 97 patients with histologically aggressive non Hodgkin's lymphoma or

Hodgkin's disease mobilized with cyclophosphamide and G-CSF, 88 % of patients achieved the minimum CD34 threshold with 1 or 2 aphereses (Watts et a/., 1997c). In this larger series of 324 patients the failure rate with up to 3 aphereses was slightly higher at 16%, although a significant proportion of these patients exceeded the minimum GM-CFC threshold. This higher failure rate is due largely to the inclusion of patients with histologically indolent lymphoma and myeloma. The reasons for the poorer mobilization in these categories of disease is not clear but might be due to the higher incidence of bone marrow involvement at diagnosis, the greater prior use of stem cell damaging alkylating agents and possibly the use of fludarabine in the histologically indolent lymphomas.

Of the 20 patients with a PBSC yield of < 1 xIOVkg CD34+ cells, but > 1 x 1 OVkg GM-CFC who were evaluable for engraftment, 18 went on to high dose therapy supported by the marginal PBSC harvests alone and two also received bone marrow cells. Neutrophil engraftment occurred in all patients within 19 days but platelet recovery was delayed in half the patients within the attainment of platelet independence taking more that 4 weeks in about one third of the patients. Four of the patients who recieved PBSC only in this group exceeded the 2 xIOVkg GM-CFC threshold described in chapter 3 and engraftment was

119 rapid in ail but one patient in whom platelet independence took 22 days to achieve. Although this higher level of GM-CFC would give greater assurance all but one patient in the whole group did became platelet independent and engraftment with this number of progenitors thus seems justifiable in the context of the underlying disease.

Twenty eight patients had a CD34 yield below 1 x10®/kg and a GM-CFC yield below 1 xIOVkg. In the earlier phase of our studies six patients in this group were rescued from high dose therapy with those 'inadequate' PBSC alone, two having had a prior bone marrow harvest held in reserve. Three had slow or delayed neutrophil recovery (50%) and 5 had delayed platelet recovery (83%).

Twelve patients with <1 x10®/kg CD34+cells and <1 xIOVkg GM-CFC were rescued with their bone marrow cells, the inadequate PBSC being given at the same time in 10 cases. The median time to neutrophil recovery in the 11 assessable patients was 15 days and 4 patients had recovery delayed beyond 21 days. These 4 patients also had very poor platelet recovery, one surviving beyond 400 days and still platelet dependent. In addition platelet recovery was delayed in 4 other patients who had had prompt neutrophil recovery. As expected from the recovery data the transfusion requirements in this group of patients was very high. It should be noted that 3 patients did have prompt engraftment despite the poor PBSC harvest but this could not be predicted from the bone marrow GM-CFC numbers. The number of post thaw bone marrow GM-CFC was <1 xIOVkg with the exception of one patient (UPN 1047) who despite a high value of bone marrow GM-CFC still had delayed platelet recovery. The poor engraftment in patients rescued with bone marrow after PBSC mobilization failure is indicative of the poor bone marrow quality in these patients. In the majority of poor mobilizers there was no suggestion that there is a specific mobilization failure occuring in the presence of adequate marrow.

120 This differs from the report by Dreger and colleagues in which treatment with Dexa BEAM appeared to compromise the quantity/quality of PBSC more than bone marrow (Dreger et al., 1995). Only one of 12 PBSC failures who received bone marrow in the present study had received previous mini BEAM therapy which is very similar to the Dexa BEAM regimen so the series are not directly comparable although PBSC progenitor yields have also been shown to be impaired with mini BEAM (Weaver eta!., 1998). Scott of al reported on the use of PBSC in 5 patients in whom an inadequate PBSC harvest was obtained. Three of these patients had failed to reach a platelet count of 50 x107l at the time of reporting but they did all achieve platelet independence (Scott et al., 1995b). An alternative to BM harvesting in patients with poor mobilization is to repeat the PBSC mobilization and collection, but Weaver et al reported that such patients still had slow platelet recovery (Weaver et al., 1995; Weaver et al., 1997).

It should be noted that prior to the advent of PBSC transplantation these patients who failed to mobilize would have received autologous bone marrow transplants as the criteria used for bone marrow adequacy was purely the number of MNC collected. Indeed approximately 15% of lymphoma patients receiving BEAM and ABMT had delayed neutrophil and platelet engraftment (Khwaja et ai, 1993b). The use of PBSC reduces the incidence of such delays and the assessment of progenitor cells allows accurate prediction of those at risk for delayed engraftment.

It is possible that a damaged bone marrow stromal cell compartment contributes to the mobilization of PBSC in some patients and that this also leads to the delayed engraftment seen with a subsequent marrow harvest of good cellularity. The clinical implications are however similar.

121 In conclusion, this study confirms that after high dose therapy the infusion of a PBSC harvest of <1 x10® CD34+ cells/kg is frequently associated with delayed haematological engraftment, and in these circumstances the additional enumeration of GM-CFC is highly valuable. In those patients with >1 x10®/kg GM-CFC rapid neutrophil engraftment can be expected. There is a high risk of delayed platelet recovery but nearly all will become platelet independent within 3 months. Those patients with <1 xIOVkg GM-CFC by contrast have a high incidence of delayed neutrophil recovery and many patients are platelet dependent beyond six weeks. One patient remains platelet dependent at 400 days. This results in increased procedure related morbidity and possibly mortality and a marked escalation in the cost of the procedure. The use of additional bone marrow cells in this situation does not overcome the risks of delayed engraftment and very serious consideration should be given as to whether it is appropriate to proceed to high dose therapy in such patients.

122 Chapter 7 Feasibilty of Clinical Scale CD34+ Cell Purification

7.0 INTRODUCTION In Chapter 3 it was shown that the optimal number of autologous CD34+ cells required to support high dose therapy is > 3.5 x10®/kg to ensure rapid neutrophil and platelet recovery. The minimum requirement was defined as 1 x10®/kg CD34+ cells, below which there is a very high risk of delayed or failed platelet engraftment. Either whole peripheral blood stem cells may be infused or the progenitors can first be purified. Such purification procedures serve to “purge” the collection of contaminating tumour cells (Berenson et al., 1991) and are a prerequisite for in vitro stem cell expansion and gene therapy (Brenner, 1993; Briddell at ai., 1997; Williams at ai., 1996). The use of monoclonal antibodies to purify progenitors was first described over 15 years ago (Beverley at ai., 1980) with the initial studies using “negative selection”. Subsequently, positive selection with antibodies to the CD34 antigen was developed (Berenson at ai., 1991; Civin at ai., 1996; Marolleau at ai., 1996; McNiece at ai., 1997; Richel at ai., 1996) and has proved amenable to the “scaling up” required for clinical practice. Several clinical scale devices are now commercially available. Important issues are the purity of the enriched fraction, the cell losses that occur and whether such losses frequently prevent the attainment of the desired progenitor thresholds. The time required for the processing procedure is an important logistical consideration. This chapter describes the experience at UCLH with the CEPRATE system on 116 apheresis collections from 91 individuals mobilised for PBSC (74 patients with lymphoma or myeloma and 17 normal donors) and compares the engraftment of patients receiving high dose therapy supported by GD34+ cell purified or whole PBSC.

123 7.1 METHODS AND MATERIALS 7.1.1 Patients studied Thirty-two patients with relapsed or resistant lymphoma, 42 myeloma patients and 17 normal donors were studied. All CD34 purification procedures were performed on a single apheresis harvest processed within 1 hour of collection. In 25 subjects two consecutive days’ apheresis collections were processed giving a total of 116 CEPRATE column procedures. The patient details and CD34 purification procedures performed are shown in Table 7.1.

Table 7.1 Subjects Studied and Number of CD34+ cell Purification Procedures Performed

Age Sex Paired Total no. Diagnosis No. (median M:F CD34 CD34 range) Procedures Procedures Myeloma 42 55 31:11 6 48 37-65 HD 9 30 4:5 1 10 17-48 HGNHL 23 37 14:9 3 26 20-63 Normal donors 17 40 11:6 15 32 18-63 No.CEPRATE 91 25 116 Procedures

( HD, Hodgkin’s disease; HGNHL, High grade non-Hodgkin’s lymphoma )

7.1.2 Conditioning The ablative therapy used for the lymphoma patients was the BEAM protocol (Linch et al., 1993; Schmitz et al., 1996) ( BCNU 300mg/m^/V. on day 1, etoposide 200mg/mVday iv days 2 - 5, (total dose 800mg/m^ ) cytosine arabinoside 200mg/m" b.i.d on days 2 - 5 (total dose 1600mg/m') and melphalan 140mg/m" iv. on day 6. The ablative therapy for the myeloma patients was 100 to 140mg/m^ melphalan and 8 to 12 Gy of fractionated total body irradiation (Jagannath et al., 1992).

124 7.1.3 PBSC mobilisation and coliection The 17 normal donors were mobilised with G-CSF (filgrastim) at 10p.g/kg/day sc and apheresis collections were performed on days 5 and 6. Ten of the lymphoma patients were mobilised with G-CSF beginning six days after ESHAP chemotherapy (etoposide 40mg/m^ for 4 days, methylprednisolone 500mg daily, cytarabine 2g/m^ and c/s-platin 25mg/m^ over four days). A single vial of G-CSF (lenograstim, 263 pg/vial) was given daily from completion of ESHAP until harvesting was complete. Apheresis commenced from day 15 when the recovery WBC first exceeded 3.0 x107L. The remaining patients were mobilised with a protocol described previously (Jones et al., 1994). Low dose cyclophosphamide (1.5g/m") was given on the first day (“day 1”) followed by G-CSF given scat lOpg/kg (filgrastim) or a single vial of lenograstim (263 pg/vial) 24hrs aften/vards and daily thereafter until harvesting was complete. Two to three apheresis collections were performed on days 8 to 12 commencing when the recovery WBC first exceeded 5.0 xIOVL. PBSC for CD34 purification were collected on a continuous flow cell separator (COBE Spectra, COBE laboratories Ltd., UK or Baxter CS3000, Baxter Healthcare Ltd., UK). At least one unmanipulated apheresis collection was cryopreserved and held in reserve for each patient.

7.1.4 Purified CD34+ celi cryopreservation In all cases the CD34+ selection process resulted in a cell fraction of approximately 90ml which was concentrated by centrifugation to 4.5 ml and an equal volume of cold tissue culture grade PBS (Dulbecco's phosphate buffered saline without calcium or magnesium. Sigma, Dorset, UK), containing 15% DMSG and 8% human serum albumin added as cryopreservative. The cells were then transferred to two 5ml cryopreservative vials (Nalgene, Rochester, NY) and left at -80°C overnight in a controlled rate freezing device (Nalgene, “Cryo 1°C freezing container”). The following day the vials were quickly

125 transferred to sterile 15ml polypropylene tubes (Becton Dickenson, Franklin lakes, NY ), secured in a cardboard folder and stored in liquid nitrogen.

7.1.5 CD34+ cell purification protocols Using the standard CEPRATE protocol the apheresis product was washed twice with 1 litre of tissue culture grade phosphate buffered saline (PBS, Sigma) using a COBE 2991 cell processor (30min). Three mg (1 mg/ml) of the supplied biotinylated GD34 monoclonal antibody was added to the harvest for a 25 minute period at room temperature in a final incubation volume of 75 ml (The standard protocol has since been revised from 3mg antibody in 75-150ml to 3mg in a 150ml volume). The two wash steps with 1 litre of PBS were repeated after the incubation period to remove unbound antibody (30min). The washed CD34 antibody labelled cells were then made up to a final volume of 300ml in PBS and connected to the CEPRATE device. The column run stage of the procedure is automated and takes 35 minutes. As the cells flow through the device, avidin coated polyacrylamide beads within the immunoaffinity column capture the biotin moiety of the CD34 labelled cells allowing the unadsorbed CD34 negative cells to run through to a “waste” bag. The final part of the program is mechanical agitation and rinse of the capture beads with PBS, releasing the bound CD34+ cells (“adsorbed cell fraction”) into a collection bag in an approximate 90ml volume. Overall the standard procedure took two hours.

7.1.6 Processing modifications In the first protocol modification (“no pre-moAb wash”) 3mg of the CD34 antibody was added directly to the harvest without prior PBS washing in a final incubation volume of between 60 ml and 110 ml and the procedure thereafter was as in the standard protocol. Omission of this pre-wash step reduced the processing time from around 2 hours to 1 Vg hours. In the second modification

126 (“no wash” method) both the pre wash and the post antibody PBS wash steps were omitted. Only 1mg of the CD34 antibody was used in this protocol to prevent saturation of the avidin bead column. Following the antibody incubation step the harvest was made up to a final 300ml volume in PBS containing 10U/ml of heparin and the product loaded on the column. CD34+ selected cells were obtained from the apheresis product within 1 hour with this protocol.

7.1.7 Progenitor evaluation In all cases samples were removed for white cell and progenitor counts from the harvest product, the antibody labelled harvest as loaded on the column and from unadsorbed and adsorbed (CD34+ cell purified) fractions. CD34+ cells were measured by flow cytometry a using directly conjugated CD34- phycoerythrin monoclonal antibody (HPCA-2PE, Becton Dickinson). Cytocentrifuge preparations were also prepared from the CD34+ cell purified fraction for each procedure and stained with May-Grunwald/Giemsa for blast cell morphology as a rapid guide to column performance. A second cytospin was stained for CD34+ cells using alkaline phosphatase-anti-alkaline phosphatase (APAAP) immunoenzymatic staining as detailed in the methods and materials section, Chapter 2.

Granulocyte/monocyte-colony forming cells (GM-CFC) and burst-forming units- erythroid (BFU-E) were simultaneously evaluated in a methylcellulose clonogenic assay medium as detailed in the materials and methods Chapter. Harvest PBSC cells were plated in triplicate at concentrations of 2.5 xIOVml, the CD34 unadsorbed cell fraction at 1 xIOVml and purified CD34+ cells at 1 x107ml and 2.5 x107ml. Colonies were counted after 14 days incubation at 37°C in a humidified COg atmosphere.

127 7.1.8 Calculations used to assess cell yields The following calculations were made for progenitor recovery (yields) and overall WBC losses;

CD34 recovery % = Harvest CD34 x10^ xlOO Purified CD34+ cells x10®

GM-CFC recovery % = Harvest GM-CFC x10^ x100 Purified CD34+ cells x10®

Overall WBC losses % = 100 - ______Harvest WBC x10^______x100 WBC recovered from column x10®

7.2 RESULTS 7.2.1 Progenitor cell purity and recovery (yieids) One-hundred and sixteen GD34-I- cell selection procedures were performed on 91 subjects. Overall the median purity of the CD34-H cell enriched fraction was 70% with 52% recovery of the starting harvest CD34-H cells (Table 2). The losses of CD344- cells were comparable to the overall white cell losses which were approximately 50% if the standard processing protocol procedure was employed. The recovery of GM-CFC overall was slightly less than for CD34+ cells at 36%. The CD34-f- cells obtained were correlated with the input CD34-H cell number (r = 0.796) and the median yield from a single column procedure was 2.1 xIOVkg. In 101/116 instances (87%) >1 x10®/kg were obtained which we have previously defined as the minimum target to ensure prompt haematological engraftment (Watts etal., 1997c).

128 Table 7.2 Effect of Harvest Concentration of CD34+ cells on Purity and Yield of CD34+ cell Enriched Fraction (medians/range)

Harvest values CD34 purified fraction CD34 CD34 CD34 CD34 CD34 GM-CFC level CD34% xlO^/kg purity yield >1x10^/kg yield

All 1.4 4.4 70% 52% 87% 36% n = 116 0.2-13.4 0.4-49.8 6-95 8-107 101/116 3-118

>1% 1.7 5.7 77% 46% 98% 39% n = 86 1.0-13.4 1.6-49.8 25-95 10-107 84/86 5-118

<1% 0.7 2.0 50% 64% 57% 32% n = 30 0.2-0.9 0.4-5.4 6-89 8-101 17/30 3-114

Most notably there was a wide range of purity and yields obtained. This in part reflects the variable cellular input from different harvests. A logarithmic relationship was observed between harvest CD34 concentration and resultant purity in the CD34 enriched fraction (Figure. 1). In those patients in whom the percentage of CD34+ cells in the PBSC collection was < 1% (n = 30), the median CD34+ cell dose prior to processing was 2.0 x1 OVkg and the purity of the CD34+ cell fraction was 64% with a median dose in the final fraction of 1.2 xIOVkg. Only 17/30 (57%) of the final fractions contained the minimum requirement of 1 x10®/kg (Table 2). In the remaining 86 individuals who had >1% CD34+ cells in their PBSC collections the median input dose was 5.7 x10®/kg and the final CD34+ cell purity was 77% with a median dose in the final fraction of 2.8 x1 OVkg. Eighty-four out of eighty-six (98%) of procedures resulted in a final yield of >1 x10®/kg CD34+ cells.

Analysis of the CD34+ cells input in absolute cell numbers gives similar data. Of those 17 individuals in whom the starting material contained < 2 xIOVkg only 7/17 (41%) of procedures resulted in an adequate number of CD34+ cells in the final fraction. In those patients with > 2 x10®/kg CD34+ cells in the starting material this was 95/99 (95%).

129 100n r = 0.577

(0 0Ü 80- ® 6 o ° o < ^ "O Î © 03 CO. 0 0 0 60- CO gp Û GO Ü GO 'o "C 40- o 9 3 o o ÛL

2 0 -

0 1 2 3 4 5

Apheresis harvest CD34%

Figure 7.1 CD34+cell purity post selection is related to start CD34% in the apheresis harvest ( n = 116 )

130 g 20-1 r = 0.783 o X (/) Q) 15- ü + co û ü 10 - 000) co X 5 -

0 5 10 Purified CD34+ cells x106/|^g Figure 7.2 Correlation between harvest and purified CD34+ cell dose

20 -

0) _Q E 15- 3 c

■I 10 CL

5

0

% CD34+ cell recovery from harvest

Figure 7.3 CD34+ cell recovery from PBSC harvests after CD34+ cell purification (n = 116 )

131 7.2.2 Assessment of Yield variability Although cell purity was closely related to CD34+ cell input, and there was a reasonable correlation between input and purified CD34+ cell numbers (Figure 7.2, r=0.78), the CD34+ cell yield was highly variable (Figure 7.3) and unpredictable. To assess whether this was due to technical variation or was patient specific, 19 patients had PBSC purified and processed identically on consecutive days. The first day CD34+ cell dose was slightly higher than the second day (p=0.02) for the paired harvests with a median starting dose and percentage of CD34+ cells of 5.7 x107kg and 1.3% compared to 4.0 x1 OVkg and 1.0% in the second harvest. However, although there was again marked inter-individual variability the losses on the two days in individual patients were similar (Table 7.3).

Table 7.3 Results of Identical CD34 Purification Processing on Consecutive Apheresis Collections from 19 individuals (medians/range)

Harvest values CD34 purified fraction Apheresis CD34 CD34 CD34 CD34 GM-CFC harvest ievei x1CF/kg purity yield yield First coiiection 1.3% 5.7 65% 48% 32% 0.4-3.4 1.3-18.1 32-89 15-91 11-91 Second 1.0% 4.0 62% 56% 40% coiiection 0.4-4.0 1.0-9.8 23-90 33-99 4-114 Paired t-test A/S 0.02 A/S A/S NS

7.2.3 Effects of the pre-column washing steps There were considerable non-specific cell losses associated with the recommended full procedure and in an effort to reduce these, and thus hopefully improve the final CD34+ cell yield, modifications were made to remove one or both of the preparative washing steps. The standard protocol was performed in 26 patients and included washing the harvest cells twice in 1L PBS before addition of the CD34 antibody followed by a second PBS wash after the incubation period. To reduce non-specific cell loss during

132 centrifugation and on the column, the antibody was added direct to the apheresis product in 47 cases. This resulted in a reduction in total cell losses from 46% to 34%. To further minimise non-specific WBC losses a final cohort of 43 apheresis products were processed without any pre column wash steps (“no-wash” method). Non-specific white cells losses were reduced further but the purity and yield of GD34+ cells were similar with all three processing methods (Table 7.4). The protocol modifications although not improving GD34+ cell yield did result in a worthwhile time saving from two hours to one hour.

Table 7.4 Outcome of CD34 Purification with Two Modifications to Standard CEPRATE Processing (medians/range)

Harvest values CD34 purified fraction

Processing CD34% CD34 CD34 CD34 GM-CFC Non-specific protocol x1(f/kg purity yield yield WBC losses Standard 1.3 3.3 64% 48% 33% 46% (n = 26) 0.2-4.0 0.6-14.0 6-89 8-102 3-118 13-75

No pre-moAB 1.4 4.4 71% 55% 40% 34% wash (n= 47) 0.2-5.0 0.4-18.3 25-95 14-107 9-114 0-84

“No-wash” 1.4 4.5 70% 48% 36% 1% (n = 43) 0.5-13.4 1.3-49.8 31-92 10-106 5-106 0-30

To further ensure that these modifications were not deleterious to the purified GD34+ cell yield two consecutive apheresis collections were processed comparing PBS wash and “no-wash” methods in 6 of the 25 patients where paired apheresis products were processed. The starting dose of GD34+ cells was similar on both days as was the percentage of GD34+ cells in the starting material, the median values being 3.3 x10®/kg and 1.3% in the collections subjected to the PBS wash procedure and 4.0 x10®/kg and 1.6% in the collections with the “no-wash” method. The total cell losses were significantly reduced from 48% with the PBS wash procedure and 11% with the “no-wash” method (P = 0.0003). As with the larger cohort however the final GD34+ cell

133 yield and purity were similar, being 1.8 x10®/kg and 66% respectively with the PBS wash products and 1.9 xIOVkg and 67% respectively with PBS wash steps omitted. There was again considerable inter-individual variation but no statistically significant differences were obtained in CD34+ cell yields and purities on the two separate days. It should be noted however that the correlation between input CD34+ cells and purified yields shown in FigureTô is not linear and shows some fall off at high input numbers of CD34+ cells. Since this study CellPro Europe have released data showing poorer yields above a target capture number of 600 xIO® CD34+ cells and have suggested doubling the CD34 antibody concentration. The median target number of harvest CD34+ cells in the 6 patients in the paired “wash” versus “no-wash” comparison was only 273 (125-345 xIO®). The upper threshold of 600 xIO® CD34+ cells was exceeded in 27 of the 116 harvests undergoing CD34+ cell selection and showed a median recovery of 41 %. The “No wash” method was applied to 9 of these harvests and showed poorer yields with a median value of 29%. This may be due the lower capture antibody concentration used in this method to prevent fouling of the avidin selection column with free antibody.

7.2.4 Clonogenicity of CD34+ purified ceils Progenitors with GM-CFC clonogenic function reside within the CD34+ cell population. As a measure of clonogenic function of the CD34+ cells in the various fractions the ratio of GM-CFC per CD34+ cell plated in methylcellulose was calculated for the start harvest, CD34+ cell purified and CD34+ cell depleted fractions. The ratios of GM-CFC to CD34+ cells plated for the harvest sample and CD34 depleted fractions were the same and showed a median value of 0.1 (around 10% of CD34+ cells giving rise to GM-CFC). However fewer than expected GM-CFC from purified CD34+ cells plated were evident with a median ratio of 0.06. This finding was evident in the discrepancy between the overall CD34+ cell yield compared to the GM-CFC yield of the 116 procedures which showed medians of 52% and 36% (Table 7.2). A similar

134 discrepancy in GM-CFC yields has been reported by other groups and applies to a greater or lesser degree regardless of the CD34+ cell selection methodology employed. (Table 7.5)

Table 7.5 Discrepant CD34+ cell and GM-CFC yields with various clinical CD34+ cell selection methods

Group and HPC source processed SelectionSelection n Median Median system CD34 yield % GM-CFC yield CEPRATE 71 52 36 (Watts etal., 1997b) (PBSC) Isolex* 11 60 29 ClinlMACS* 11 79 39 (Alcorn etal., 1995) (PBSC) Isolex 7 90 31 (Lemoli etal., 1997) (PBSC) CEPRATE 26 58 45 (Shpall etal., 1994) (BM) CEPRATE 18 52 36 (Firat etal., 1998) (BM) CEPRATE 7 40 35 Isolex 7 44 36 (Rowley etal., 1998) (PBSC) isolexIsolex 60 56 25

* Unpublished observations

7.2.5 Haematological recovery after infusion of purified CD34+ cells Fifty-two of the 74 patients, 21 lymphoma patients and 31 myeloma patients had their purified CD34+ cells returned alone after BEAM chemotherapy or melphalan/TBI respectively. The median time to recovery of neutrophils of 0.5 x10®/L was 12 days (range 9-23 days) with only two patients having slow recovery defined as greater than 21 days. Thirty-four patients received G-CSF from 6 days after the infusion of purified CD34-h cells and the median time to neutrophil recovery in these patients was 12 days (range 9-22 days) compared to 14.5 days (range 11-23 days) in the remainder (P= 0.007). The median time to attainment of platelet independence, defined as a sustained, unsupported platelet count above 15 x107L without bleeding, was 13 days (range 7-100) with G-CSF administration having no effect. Three patients had slow platelet recovery (22, 26 and 27 days) and four patients had delayed recovery at 32, 36, 41 and 100 days. The overall incidence of delayed platelet recovery 4/53 (8%) was similar to that seen in the 262 patients who received whole PBSC

135 (8%) was similar to that seen in the 262 patients who received whole PBSC products (11%) The four patients with delayed platelet recovery after infusion of GD34+ cell purified PBSC received 3.1, 1.6, 1.1, and 0.9 x10®/kg CD34 cells which was in each case below our previously defined optimal threshold of 3.5 x10®/kg (Watts et al., 1997c). The one case that took 100 days to become platelet independent received less than our “minimal” threshold number of IxlOVkg CD34+cells. In a further four patients the purified CD34+ cell dose was considered inadequate to return alone at 0.6, 0.3, 0.3 and 0.01 xIO® /kg and in each case two unmanipulated back up PBSC collections were returned concomittantly. In three patients even the pooled total dose of CD34+ cells collected remained low at 0.8, 0.6 and 0.3 x10®/kg and a bone marrow harvest was collected in the poorest mobiliser. Slow neutrophil recovery was seen in two of these patients at 26, and 24 days and delayed platelet recovery in two at 59 and 80 days respectively. In the fourth patient the two pooled back up PBSC collections were above the minimum threshold level at 2.8 x10®/kg CD34+ cells and recovery was rapid at 12 and 13 days for neutrophil and platelet recovery respectively.

7.2.6 Single versus multiple harvests The patients who received purified cells alone had CD34+ cells selected from a single apheresis harvest. The haematological recovery of these patients was compared with that of 232 patients (detailed in Chapter 1) who received whole PBSC. This group was further divided into the 70 patients who received a single harvest and the remainder who received more than one collection. (Tables 7.5 and 7.6 and distribution plots. Figures 7.5.1 to 7.5.5). Delayed neutrophil recovery to a count of 0.5 x10®/L exceeding 28 days was only seen in 3 non-CD34+ cell selected patients in whom an apheresis yield of 1 x10®/kg CD34+ cell or 1 xIOVkg GM-CFC was not acheived despite multiple collections. It is notable that the slow recovery of the 9 patients who took 21-28

136 days to attain this neutrophil count was not progenitor dose dependent and that none received post transplantation G-CSF. There were more delays in platelet recovery beyond 28 days in the patients who received less than 2 x10®/kg CD34+ cells in multiple whole collections (22/66,36%) compared to those who received this dose in single collections whether from whole or CD34+ cell purified harvests (5/42, 12%) p=0.035. This may have been dose-related however as only 9 of the 42 patients who received <2 x10®/kg CD34+ cells in single harvest products also had <1 x1 OVkg GD34+ cells compared to 33/66 in the multiple apheresis group. Delayed platelet recovery was also significant however between those patients who received multiple collections which contained less than 2 xIOVkg GM-CFC (22/57, 39%) and the patients who received <2 xIOVkg in single collections, (5/42,12%), p=0.024. In this instance the distribution of low progenitors in the two groups was equivalent and was 17 below 1 x107kg GM-CFC among the 45 patients receiving <2 xIOVkg GM-CFC in multiple collections compared these doses in 17 and 42 patients respectively in the single apheresis group (Tables 7.5 and 7.6 and Figures 7.4.1-7.4.8).

137 Table 7.5 Progenitor cell dose re-infused and risk of delayed haematological recovery in 53 patients receiving immunoaffinity purified CD34+ ceils compared to 232 patients receiving one or more whole PBSC collections

Neutrophil Platelet Recovery Recovery Cell Dose Returned < 1 1 -2 2-3.5 >3.5 < 1 1 -2 2-3.5 >3.5 0034+ cell P u rifie d CD34+ cells x1Cf/kg n = 5 18 13 16 5 18 13 16 Slow 0 1 0 1 1 1 1 0 ( >21 days) ( 0%) ( 6%) (0%) ( 5%) (17%) (6%) (8%) ( 0%) Delayed 0 0 0 0 1 2 1 0 ( >28 days) ( 0%) ( 0%) ( 0%) ( 0%) (17%) (11%) (8%) ( 0%)

GM-CFC n = 16 13 11 12 17 13 11 12 Slow 1 0 1 0 1 2 0 0 {>21 days) ( 6%) ( 0%) (9%) ( 0%) ( 6%) (15%) ( 0%) ( 0%) Delayed 0 0 0 0 2 1 1 0 ( >28 days) ( 0%) ( 0%) (0%) ( 0%) (12%) ( 8%) ( 9%) ( 0%) Whole PBSC

CD34+ cells x1Cf/kg n = 37 43 46 106 37 43 46 106 Slow 3 2 2 0 2 2 1 3 ( >21 days) ( 8%) ( 5%) ( 4%) ( 0%) ( 5%) ( 5%) (2%) ( 3%) Delayed 3 0 0 0 1 9 3 3 1 ( >28 days) ( 8%) ( 0%) (0%) (0%) (51%) ( 7%) ( 7%) ( 1%)

GM-CFC n - 19 38 47 128 19 38 47 128 Slow 3 2 2 0 0 2 1 5 ( >21 days) (16%) ( 5%) ( 4%) ( 0%) ( 0%) (15%) ( 0%) ( 4%) Delayed 3 0 0 0 11 1 1 2 2 {>28 days) (16%) ( 0%) ( 0%) ( 0%) (58%) (29%) (4%) ( 2%)

138 Table 7.6. Progenitor cell dose re-infused and risk of delayed haematological recovery in 232 patients who received single (n=70) or multiple (n=162) non-CD34 selected PBSC

Neutrophil Platelet Recovery Recovery Cell Dose Returned < 1 1 -2 2-3.5 >3.5 < 1 1-2 2-3.5 >3.5 Single Collection re-infused

CD34-h cells xlO^/kg n = 4 15 17 34 4 15 17 34 Slow 0 2 1 0 1 1 1 0 ( >21 days) ( 0%) (13%) ( 6%) ( 0%) (25%) ( 7%) ( 6%) ( 0%) Delayed 0 0 0 0 0 2 2 0 ( >28 days) ( 0%) ( 0%) ( 0%) (0%) ( 0%) (13%) (12%) ( 0%)

GM-CFC x1(f/kg n - 2 10 16 42 2 10 16 42 Slow 0 2 0 1 0 1 0 2 {>21 days) ( 0%) (20%) ( 0%) ( 2%) ( 0%) (10%) ( 0%) ( 5%) Delayed 0 0 0 0 0 2 2 0 ( >28 days) ( 0%) ( 0%) ( 0%) ( 0%) ( 0%) (20%) (13%) ( 0%)

Multiple collections re- Infused

CD34+ cells x1Cf/kg n = 33 28 29 72 33 28 29 72 Slow 3 1 0 0 1 1 0 3 ( >21 days) ( 9%) ( 4%) (0%) (0%) ( 9%) ( 4%) ( 0%) (4%) Delayed 3 0 0 0 1 9 1 1 1 ( >28 days) ( 9%) ( 0%) ( 0%) (0%) (58%) ( 4%) ( 3%) ( 1%)

GM-CFC n = 17 28 31 86 17 28 31 86 Slow 3 0 1 0 0 1 1 3 ( >21 days) (18%) ( 0%) (3%) ( 0%) ( 0%) (4%) ( 3%) ( 3%) Delayed 3 0 0 0 11 9 0 2 ( >28 days) (16%) ( 0%) ( 0%) ( 0%) (65%) (32%) ( 4%) ( 2%)

139 100-1

75- _i g s Z X O ID 50- (/) o

25-

O OO

0 5 10 15

lOOi

B

0_ 25 - 0 1 Q

0 5 10 15

CD34+ cells x10®/kg

Figure 7.4.1 Immunoaffinity purified CD34+ cells re-infused and A) Neutrophil, B) Platelet recovery (n=52)

140 100-1

75-

0)3 O % o S 50- (/) d

25-

0 5 10 15 100-1

B 0) U c 75- 0 *D 0C 0Q. -a _c 50- 0 0

CL 0 25- OO 1 a

0 5 10 15

GM-CFC x105/kg

Figure 7.4.2 Immunoaffinity purified CD344- cell: GM-CFC re-infused and A) Neutrophil, B) Platelet recovery (n=52)

141 100-1

U) 75- Q. 2 ^ 03 OG) z ^ O in 50 O) d a

25

« < r .

100

B

75- 0) ü c 0 •U c 0 Q. 50- U0 C

0 0 25- CL O (f) 0 Q r 10 15

CD34+ cells x10®/kg

Figure 7.4.3 Whole PBSC CD34+ cells re-Infused and A) Neutrophil, B) Platelet recovery (n=232)

142 Days to Platelet independence Days to Neutrophils 0 .5 X1Q9/L

cn o o cn o _l_

CQC CD "si

> ^ Z o O CD o =r “ 0 O 00 •O c/5 o z o “n CD O DO GO X :56 D3 T l o m O cn CD CD3

CD i CD

DO 100-1

75-

_ i ô>3 G)O ÎS 50- W o

25-

0 5 10 15

100-1

B

0 ü 75- C ■O0 c 0 Q. ■D0 _C 50-

0 0

CL O 25- 0 0 Q

0 5 10 15

CD34+ cells x106/|^g

Figure 7.4.5 Single whole PBSC harvest: CD34-I- cells re-infused and A) Neutrophil, B) Platelet recovery (n=70)

144 100-1

75-

Q. 2 3 05 0) O 50 - ^ X o LO (/) O S' û 25-

0 I 10 15 100-1

B

0) 75- ü c 0) ■D C 0 CL "O0 50- C

0 0 % Û. 25- 0 1 Q

05 10 15

GM-CFC xlO^/kg

Figure 7.4.6 Single whole PBSC harvest: GM-CFC re-infused and A) Neutrophil, B) Platelet recovery (n=70)

145 100n

_cn 75- Z CL

3 05 0 O Z ^ o in 0 6

Q

100

B

(D ü C “OQ) C Q) Û. ■a(D ç

Q) 0

CL O

05 û r 10 15

CD34+ cells xlO^/kg

Figure 7.4.7 Multiple whole PBSC harvest: CD34+ cells re-infused and A) Neutrophil, B) Platelet recovery (n=162)

146 Days to Neutrophils Days to Platelet independence CD 0 .5 X1Q9/L

C71 ■vl o O cn o I Tl cqc’ Q -vl i f Z H CD " O Q r it O %CD Tm l n TJ CD O m O o ÎI cn u s i

Q.i 3 Q. 7.3 DISCUSSION The fact that haematological regeneration after purified CD34+ cell transplantation was not delayed relative to whole PBSC is in accord with our previous data in a smaller number of cases (Watts et al., 1995a). This implies that accessory cells are not required for rapid haematological recovery even when no post infusion growth factors are given. Identical haematological recovery with CD34 purified or unseparated PBSC has also been reported by others (Brugger etal., 1994; Hohaus et al., 1997). It should be noted however that CD34+ cell losses do occur with purification and if threshold limits are not applied before and after purification, an increase in delayed platelet recovery would be anticipated. The effectiveness of progenitor doses from whole PBSC versus single harvest (whole or CD34+ cell purified) on haematological recovery was also considered in this study. Fewer than expected delays in platelet recovery occurred in the single harvest group patients with low GM-CFC implying poorer progenitor quality where multiple collections were required to achieve the same dose. This could also be due however to a discrepancy between GM-CFC numbers and purified CD34+ cell number plated compared to non­ selected CD34+ cells in the same patient. This effect has been reported in other CD34 purification studies (Alcorn et al., 1995; Firat et al., 1998; Lemoli et al., 1997; Shpall etal., 1994; Watts etal., 1997b). There is some evidence that the lack of accessory cells may be detrimental to colony growth from purified CD34+ cells in semi-solid culture and that colony growth can be improved with marrow stroma "feeder layers" (Rowley et al., 1998). This is currently under investigation in our unit.

This large experience with the CEPRATE purification device confirms that it is reliable and easy to use in the clinical setting. The overall CD34+ cell purity and yield in all 116 CEPRATE procedures of 70% and 52% compares with values of 42% and 52% reported by Sphall ef a/in 18 patients (Shpall et al..

148 1994) 61% and 73% as reported by Brugger ef a/in 21 patients (Brugger et al., 1994) and 47% and 77% as reported by Schiller ef a/in 37 patients (Schiller ef a/., 1995). The median losses of CD34+ cells in these series during purification have ranged from 27% to 53% but in individual cases the losses can be considerably greater. Such losses have an impact on the clinical use of the purified products. We have previously defined the minimum CD34+ cell dose for PBSC transplantation as 1 x10®/kg, and to allow for the losses associated with purification at least 2 x10®/kg GD34+ cells are required prior to purification. In a series of 97 p re-treated lymphoma patients mobilised with cyclophosphamide and G-CSF (Watts & Linch, 1997) only 51 patients (53%) achieved this target with a single apheresis. This means that if CD34+ cell purification is to have wider use two or more purification procedures must be performed on consecutive apheresis collections or the collections pooled and processed on the same column. In the 97 patients referred to above, pooling of two collections allowed achievement of the 2 x1 OVkg CD34+ cell threshold in an additional 20 patients (21%). Twenty-six percent of patients still remain below this threshold, although it is likely that this number would be less if patient progenitor cells were mobilised and collected at an earlier stage of the disease.

A number of difficult decisions now arise concerning CD34+ cell purification for the patient with borderline mobilization. If the primary objective is as a purging step for tumour cells the infusion of fewer cells, providing engraftment was assured could be seen as advantageous. Secondly, there is some concern that T-depleted autografts may be harmful in terms of impaired cell-mediated immunity post transplantation (Anderson etal., 1990; Nachbaur etal., 1997) and we have seen a fatal case of an EBV-associated lymphoma in a myeloma patient following PBSC support of high dose therapy with CD34+ purified cells. The collection of a reserve non-CD34 selected PBSC harvest would seem prudent therefore to cover both slow haematological and immunological

149 recovery. Finally the use of 2 or more CD34 selections per patient raises considerable financial contraints, encouraging most centres to pool collections. Some users have reported poor cell recoveries associated with the processing of PBSC products stored at 4°C however, which may relate to the expression of adhesion proteins, activated platelets and cell clumping (Jaremo et al., 1992; Mentzer eta!., 1986).

A noteworthy finding was that the purity of the final fraction achieved was closely related to the input CD34+ cell percentage. A starting percentage of >1% was found to be associated with a high probability that the final purity would be >60% (72/86 cases). The percentage of CD34+ cells in the starting collection is of course related to the absolute number of CD34+ cells in the harvest and in general a 1% threshold relates to an absolute threshold of about 2 x10®/kg which is required from the viewpoint of 50% losses. In practice therefore it is possible to abort a CD34+ cell purification before cell losses and unnecessary expense have been incurred. Appreciation of the relationship between “input” and “output” will also be important when comparing the performance of different purification devices. From a clinical standpoint “input “ cells must refer to harvest collection values. A statement of final purity and yield without input CD34+ percentages is not meaningful. In this study we have evaluated the impact of reducing the centrifugation/ wash steps to try and minimise the cell losses. The non-specific losses can be significantly ameliorated by such modifications but unfortunately the specific CD34+ cell losses are not reduced.

In conclusion, the CEPRATE device is a reliable system for purification of CD34+ cells on a clinical scale. The results in terms of purity of the final fraction are closely related to the input CD34+ cell count. It is a relatively rapid process and can be further speeded up by omission of recommended wash steps, although this does not improve cell yield. The CD34+ cell losses are

150 approximately 50% and are a significant problem in many cases, preventing the attainment of the CD34+ cell thresholds required to ensure rapid haematological engraftment.

151 Chapter 8 Accessory Cells do not Contribute to G-CSF or IL-6 Production nor to Rapid Haematological Recovery following Peripheral Blood Stem Cell Transplantation

8.1 INTRODUCTION Following very high dose chemo-radiotherapy there is a rise in circulating G- CSF and IL-6 levels which precedes the recovery of neutrophils and platelets (Sallerfors & Olofsson, 1991; Watari et al., 1989). Relatively high leveis of cytokines have been documented in recipients of peripheral blood stem cells (PBSC) and several groups have suggested that the many accessory cells (lymphocytes and monocytes) contained in a leukapheresis collection may contribute to the more rapid haematological recovery that occurs compared to bone marrow stem cell (BMSG) recipients (Ho et a!., 1994; Kawano et a!., 1993; Mielcarek eta!., 1996; Testa eta!., 1994). To address this issue we have assessed haematological recovery in 10 patients with lymphoma or myeloma treated with high dose chemo/radiotherapy supported by purified progenitors (CD34+ cells) and compared this to recovery in 59 patients with lymphoma or myeloma who received unfractionated PBSC. In neither group were growth factors given during the first 11 days following stem cell infusion. In addition, we have evaluated the plasma cytokine profile in 12 patients treated with high dose BEAM chemotherapy or melphalan/total body irradiation, 6 of who received unfractionated PBSC and six purified CD34+ cells.

8.2 MATERIALS AND METHODS 8.2.1 Patient groups Fifty-eight patients with relapsed or resistant lymphoma and 11 myeloma patients were studied. The patient demographics are summarised in Table 8.1.

152 The ablative therapy used for the lymphoma patients was the BEAM protocol (Chopra et al., 1993; Linch et al., 1993) ( BCNU 300mg/m^/V. on day 1, etoposide 200mg/mVday iv days 2 - 5, (total dose 800mg/m^ ) cytosine arabinoside 200mg/m" b.i.d on days 2 - 5 (total dose 1600mg/m") and melphalan 140mg/m" iv. on day 6. The ablative therapy for the myeloma patients was 100 to 140mg /m^ melphalan and 8 to 9 Gy of fractionated total body irradiation (Jagannath et al., 1992).

Table 8.1 Patients receiving unfractionated or CD34+ ceii selected peripheral blood stem cells.

Clinical details Whole PBSC Purified CD34+ cells (n= 59) (n=10) Hodgkin’s disease 42 2 NHL 10 3 Myeioma 7 5 Maies : Females 41 : 18 4 :6 Median Age (range) 31 (15-67) 44 (17-59) Previous wide field irradiation 5/59 0/10 Previous cycles of chemotherapy 7(3-17) 6(3-11) median (range) Bone marrow involvement at time of PBSC harvest 0/59 0/10

8.2.2 PBSC mobilization and collection. All patients were mobilized with low dose cyclophosphamide (1.5g/m^) given on the first day (“day 1”) followed by G-CSF given scat lOpg/kg (filgrastim) or a single vial of lenograstim (263 pg/vial) 24hrs afterwards and daily thereafter until harvesting was complete. One to three apheresis collections were performed on days 8 to 12, on a COBE Spectra or Baxter CS3000, commencing when the recovery WBC first exceeded 5.0 xIO^L. The apheresis collection was then cryopreserved in an equal volume of autologous plasma containing 20% dimethyl sulphoxide (DMSG) followed by controlled rate freezing and storage in liquid nitrogen. When required the PBSC were thawed at the bedside at 40°C and returned via a fast running central catheter. In 10

153 patients, a single apheresis was subjected to CD34+ cell selection as described below. The CD34+ selection process resulted in a 90ml volume cell fraction which was concentrated and cryopreserved in two 4.5 mL aliquots as detailed in Chapter 7. When stem cell re-infusion was required, each individual vial was removed from its outer 15ml tube, thawed in a sealed bag in a water bath at 37°G and slowly reconstituted with warmed 8% human serum albumin in PBS up to a 30ml volume. The re-suspended cells were then drawn into a 50ml syringe and returned to the patient via a 3-way tap into a fast running intra-venous saline drip.

8.2.3 CD34+ cell selection Single apheresis collections were processed for clinical scale CD34+ cell purification using a CEPRATE SC immunoaffinity column and associated microprocessor device using the manufacture’s protocol as detailed in Chapter 7. A 0.5ml aliquot of the purified CD34+ cell fraction was taken for cytocentrifuge preparations for blast cell morphology, CD34 numbers by alkaline phosphatase-anti-alkaline phosphatase (APAAP) immunoenzymatic staining and for flow cytometry and colony forming cell (CFC) cultures.

8.2.3 Progenitor evaluation A sterile harvest sample was taken from the apheresis bag pilot line after repeated “stripping back” of the product into the bag which was continually mixed to ensure even cell suspension. This sample was then diluted 1/10 and 1/100 in RPMI containing 10% fetal calf serum and 20U/ml heparin. Direct harvest smears and these dilution’s were then used for a full blood count (STKR Coulter inc. Hialeah, FL.), manual white cell differential and progenitor measurement by CD34+ cell and CFC assays as detailed in the general methods materials Chapter 2 and for CD34+ cell selection specifically, in Chapter 7.

154 8.2.4 Accessory cell depletion. Smears from unfractionated harvest products, pre CD34+ cell processing harvest smears and corresponding cytospins from the purified CD34+ fractions were tightly wrapped in parafilm and stored at -20°C for batch staining. Monocytes were identified by esterase staining (Li et al., 1973) and T-cells by APAAP staining (Erber et ai, 1984) with a CD3 mouse monoclonal (UCH-T1, Dako Ltd., Bucks, England). Positively reacting cells were counted for each cell marker by examining an approximated 2000 cells with a “Miller’s squares” eyepiece counting graticule (Graticules Ltd., Tonbridge, Kent UK).

8.2.4 Cytokine assays. Tripotassium-EDTA specimens (Vacutainer, Becton Dickenson) were collected from patients before high dose therapy, (“pre”), on the day of stem cell infusion, (day 0) and on alternate days up to day 23. The samples were centrifuged at 3500g for 15 minutes within 4hrs of collection and plasma separated and stored at -20°C until required.

Plasma concentrations of all growth factors were evaluated by specific enzyme- linked immuno-adsorption assay (ELISA) kits (“Quantikine”, R+D systems- British Biotechnology, Cowly, Oxford, UK). The detection thresholds of the cytokines measured were; G-CSF 5.0 pg/ml, GM-CSF 20 pg/ml, IL-6 10 pg/ml, stem cell factor 5.0 pg/ml and MIP-1-alpha 3.0 pg/ml. The detection method was similar for each assay. A mouse monoclonal antibody specific to the cytokine to be measured was provided adsorbed to 96 well polystyrene plates. Geometric dilution’s of plasma control recombinant human cytokine were added to the wells with test samples in duplicate, and incubated for 2 hours at room temperature. After washing the plate, a peroxidase conjugated goat -anti -mouse antibody was added for a further 2 hour incubation, followed by a second wash to remove un-reacting antibody. Tetramethylbenzidine was then added which is oxidised by peroxidase activity to a blue reaction product. The reaction was stopped after 20 minutes with 2N

155 sulphuric acid and absorbance read at 450nm vs. substrate blank on a microtitre plate reader (Dynatech MR700, Dynatech labs, ltd., Sussex, UK) . Absorbance of the test wells were compared with the standard curve and expressed as pg/ml. Means and SEMs were calculated for various time points for each of the two groups of patients (whole PBSC infused or CD34+ stem cells only) at pre, day 0, then grouped at intervals of days 1-3, 4-6, 7-9, 10-12, 13-15, 16-18 and finally days 19-21.

8.2.3 Statistical analysis of data. Recovery times of patients given unfractionated PBSC were compared with those for patients who received purified CD34 cells by the Mann-Whitney U test for non-parametric, unpaired data on an IBM PC ( Number Cruncher Statistical System 5.x, Univ. Utah ).

8.3 RESULTS 8.3.1 CD34+ cell purification and accessory cell depletion Peripheral blood progenitor cells from a single apheresis collection were purified on a CEPRATE SC immunoaffinity column and subsequently used as haematological support following high dose therapy in 10 patients. In this group, the median dose (range) of CD34+ cells infused was 3.0 xIOVkg (0.5- 6.9 x10®/kg) with a median purity of 81% (44-93%) (Table 8.2). The CD34 purification process resulted in a greater than 2 log depletion of monocytes and greater than 3 log depletion of T cells (Figure 8.1). Six patients were also studied who received whole PBSC from 2 or 3 apheresis collections. The comparative numbers of accessory cells infused in this group are also shown in Figure 8.1 with the accessory cells present in individual harvests compared with that of the pooled harvests re-infused in this group.

156 Table 8.2 Processing results of CD34+ cell selection:

Harvest values Post processing results

Go o L ' Patient Diag MNC CD34 CD34 CD34 CFC number x1(f purity xKf/kg yield yield CD34% 1 HD 23 1.4% 81% 3.5 60% 43% ^ (o 2 HD 42 1.8% 88% 6.1 48% 80% \2_ 3* NHL 23 2.8% 54% 1.9 21% 34% 4* NHL 13 2.4% 87% 1.6 30% 12% < 3 5* NHL 19 1.5% 80% 2.8 102% 84% 2 6* MM 22 1.1% 77% 3.1 78% 53% - U T MM 19 1.6% 86% 2.8 75% 33% 8* MM 50 2.4% 93% 6.9 50% 53% I V 9 MM 19 1.3% 44% 0.5 11% 10% 1 0 MM 31 1.6% 61% 3.1 47% 45% 6.C

M edians 23 1.6% 81% 3 .0 49% 44% (range) (13-50) (1.1-28) (44-93) (0.5-6.9) (11-102) (10-84)

'"Serial plasma cytokines were measured post PBSC infusion in patients 3 to 8

157 Monocyte numbers T-cells numbers Whole PBSC CD34 selected whole PBSC CD34 selected

c O=3 Ü

0 Ü

Individual Total Pre-Post CD34 Individual Total Pre-Post CD34 harvests returned selection harvests returned selection

Figure 8.1 Accessory cell depletion following CD34+ cell purification. Results are shown for a single and total harvest infused for the whole PBSC group. This is compared with the single collection used for the CD34 purified patients.

158 8.3.2 Cytokine profiles following stem cell Infusions. Serial plasma G-CSF levels were measured in 12 patients following haemopoietic stem cell infusions and IL-6 in 11. The G-CSF and IL-6 profiles are almost a mirror image of the neutrophil and platelet levels, both cytokines rising as the blood counts fall and falling to baseline levels as the blood counts recover (Figure 8.2). With both cytokines there was an abrupt rise in plasma levels between days 5 and 14 post stem cell infusion (Figures 8.3A and 8.3B). Six patients had received purified CD34+ cells and 6 received whole PBSC and there was no significant difference between these two groups. The cytokine levels were not different between those patients with lymphoma receiving BEAM chemotherapy and those with myeloma receiving melphalan/TBI conditioning (data not shown). The median number of cycles of chemotherapy previously given in the 6 patients receiving purified CD34+ cells was 5 compared to 8 in the recipients of whole PBSC and one of the latter group had received wide field irradiation compared to none in the former group.

Serial plasma stem cell factor and GM-CSF levels were also measured in 11 patients. There was no detectable change in these levels at all time points (data not shown). Similarly, plasma MIP1-alpha levels were measured in five patients and showed no significant changes over the recovery period (data not shown).

159 Neutrophil count X109/L------r 1000

4-

Platelets xIO /L 100

G-CSF (pg/ml)

#v

IL-6 (pg/ml)

preO 2 4 6 8 10 12 14 16 18 20 22

Days post stem cell infusion

Fig 8.2 Serial cytokine changes associated with haematological recovery post high dose therapy and stem cell infusion. Median values shown, n=12

1 6 0 - o - ’ " purified CD34+ cells

10000-1 whole PBSC

E D) Q.

1000 - u_ CO Ü 100 - Ô CO E cn jO CL

Pre 3 6 9120 15 18 Days post stem cell infusion

100-1

E Q. CO CD _0)> 1 / C£)

CO E CO _C0 CL

Pre 0 3 6 9 12 15 18 21 24

Days post stem cell infusion

Figure 8.2. Endogenous plasma cytokine levels post PBSCT. A) G-CSF and B) IL-6

161 8.3.4 Haematological recovery following infusion of purified cells. Ten patients were rescued from high dose chemo/radiotherapy by purified CD34+ cells from a single apheresis. This group contained the 6 patients on whom the cytokine profile was measured and a further 4 patients. Five patients had lymphoma (2 Hodgkin's disease and 3 non-Hodgkin's lymphoma) and 5 had multiple myeloma

Although none of these patients was given G-CSF immediately following stem cell re-infusion, two patients, patient numbers 1 and 6 were given a single injection of G-CSF on days 18 and 22 respectively because of delayed neutrophil recovery. Although neutrophils >0.1 x10^/L were detectable by day 11 the subsequent rise was slow. In both patients the neutrophil count was >0.5 xIO^/L the following day after the G-CSF injection and no further G-CSF was given. The neutrophil count subsequently remained above 0.5 xIO^/L.

The recovery data is shown in Table 8.3 and a more detailed comparison with the recovery in 59 patients (39 with Hodgkin’s disease, 14 with non-Hodgkin’s lymphoma and 6 with myeloma) following whole PBSCT is shown in Figures 8.4A and 8.4B in which recovery is plotted against the total GM-CFC dose infused. There was no discernible difference in recovery between those patients with lymphoma or multiple myeloma. None of the recipients of whole PBSCT received G-CSF immediately after stem cell infusion, although it was given to 5 patients beyond day 11 because of slow neutrophil recovery. Overall the neutrophil recovery to 0.5 x10®/L was marginally slower in the recipients of purified CD34+ cells than in recipients of whole PBSC ( p<0.05 ). Recovery is however related to dose of GM-CFC infused, with a higher proportion of slow recoveries in recipients of < 3.5 xIOVkg. This lower dose was more frequent in the recipients of purified CD34+ cells as they were derived from a single apheresis whereas the whole PBSC recipients received an average of 2 collections. In addition, CD34 purification results in some

162 losses (approximately 50%). If only those patients who receive a GM-CFC dose of < 3.5 x1 OVkg are considered, there is no significant delay associated with the use of purified CD34 cells. Platelet recovery was not significantly different between the recipients of whole PBSC and purified CD34 cells.

Table 8.3 Haematological recovery of patients receiving CD34+ purified cells compared to patients receiving whole PBSC

Patient diag Days of Days to Days to Platelet Days to number ANC ANC Platelet units Platelets <0.1x10^/L 0.5x10^/L indep. given >50x10®/L 1 * HD 6 23 15 10 19 2 HD 4 12 13 10 17 3 NHL 2 13 9 10 11 4 NHL 5 17 21 55 22

5 NHL 9 18 14 30 21 6 * MM 11 19 41 105 44 7 MM 5 18 13 20 20 8 MM 6 12 11 10 12 9 MM 7 16 18 25 41 1 0 MM 7 16 8 35 13

CD34 6 16.5 13.5 22.5 20 med/range (4-9) (11 -23) (9-41) (11 -41) (10-105) Whole 6 1 2 1 1 1 5 22 PBPC: (1-11) (6-27) (6 - >85) (5 - 230) (9->100) (n =59)

*A single injection of G-CSF was administered to patient 1 on day 22 and a single dose to patient 6 on day 18

163 100-1 purified CD34+ cells: n=10

O) 8 0 - whole PBSC: n=59 o X m d A 6 0 “ O z < o 4 0 - (/) S' Q

0 5 10 15 20 25

100-1 B 80- OQ) 0)C "O c 0) 0)Q. "D C 40- CL O (/) 2 0 - Q o o

0 5 10 15 20 25

GM-CFC dose x 105/kg

Figure 8.4 Haematological recovery of A) neutrophils and B) platelets in patients receiving whole or CD34+ cell purified PBSCT

164 8.4 DISCUSSION This study confirms that following high dose chemotherapy or radiotherapy followed by PBSCT there is a marked rise in plasma G-CSF and IL-6 levels. The levels of G-CSF peaked around day 5 at approximately 1 ng/ml and IL-6 of approximately 50pg/ml. The peak G-CSF levels are in accord with data published by Baiocchi et aland Ho eta! (Baiocchi eta!., 1993; Ho at a!., 1994), but are somewhat lower than that reported by other groups (Haas et a!., 1993; Kawano eta!., 1993; Testa et a!., 1994) This may represent methodological differences but is also likely to represent the different patient groups that have been studied. Testa et al, found that the levels achieved were related to the underlying disease (and thus the pre-transplant treatment given) and that patients with ovarian cancer gave higher levels than those with haematological malignancies (Testa et al., 1994). In accord with this, we have studied one patient receiving high dose therapy for breast cancer and found peak G-CSF levels on days 4 to 8 of approximately 5.0 ng/ml.

The levels of stem cell factor did not change significantly in the post-transplant period and this is in accord with the report from Testa (Testa et al., 1994). The levels of GM-CSF were below the levels of detection following transplantation, and similar results have been reported by others (Ho et al., 1994; Kawano et al., 1993). Testa, in contrast, did find a post transplant rise in GM-CSF in a proportion of patients (Testa et al., 1994), The reason for this discrepancy is not clear.

In this study, we observed no significant difference in the level of G-CSF of IL-6 achieved between patients with lymphoma or myeloma although the conditioning regimen used for the two diseases was different. There was also no difference in the cytokine profiles of the recipients of purified CD34+ cells compared to those of whole PBSC even though the former received approximately 3 logs less monocytes and even fewer T-cells. This strongly

165 suggests that such re-infused accessory cells do not contribute to the production of cytokines during the neutropaenic and thrombocytopaenic period.

The recipients of accessory depleted progenitor cells had marginally slower neutrophil recovery than the whole PBSC recipients, but this could be largely attributed to the smaller numbers of progenitors (GM-CFC) that were re-infused to these patients illustrating that there is a threshold number of progenitors required below which the risks of slow engraftment increase (Bensinger et al., 1995; Schwartzberg et a!., 1993; Tricot et a!., 1995; Weaver et a!., 1995; Weaver et a!., 1997). One of the recipients of the purified CD34+ cells had delayed platelet recovery, which is in accord with the experience obtained with whole PBSC re-infusions. These data indicate that re-infused accessory cells do not make a major contribution to the rapid haematological recovery seen with PBSCT compared to ABMT although as this was not a randomized study a minor contribution of accessory cells cannot be fully excluded. It must also be noted that of the 10 patients who received purified PBSC, 5 were conditioned with BEAM chemotherapy which is not fully myeloablative. Some caution must therefore be exercised in determining any accessory cell requirements for the engraftment/proliferation of the most primitive stem cells.

As the more rapid regeneration is not apparently due to accessory cells in the PBSC, this implies it is due to the infusion of a functionally different subset of progenitor cells. None of the patients in this study received growth factors immediately following stem cell re-infusion, and it seems highly likely from the recovery data presented in the previous chapters and recent randomized trials (Klumpp et a!., 1995; Linch et a/., 1997) that the recovery of neutrophils may have been accelerated by the administration of G-CSF.

166 Discussion

In Chapter 3 progenitor thresholds were defined for predicting rapid engraftment after high dose therapy with peripheral blood stem cell support. The general consensus is that a PBSC progenitor dose between 1-5 x10®/kg CD34+ cells or 1-5 xIOVkg GM-CFC is required to ensure rapid engraftment (Demirer & Bensinger, 1995) and that higher progenitor numbers result in more rapid engraftment (Bensinger et al., 1995). Many reports have shown a reduction in mean engraftment times with progressively higher doses but there is still an obligate period of cytopaenia and a steep dose relationship so that gains rapidly diminish above around 1 x10®/kg CD34+ cells (Haas et a!., 1995; Weaver et a!., 1995). Against the need to balance high progenitor yields against over-agressive mobilization and apheresis protocols, particularly in heavily pre-treated patients and to maximise available resources, a pragmatic hazard/risk assessment was adopted in Chapter 3. In this approach the incidence of pre-defined slow (21-28 days) or delayed (>28 days) recovery times to achieve platelet independence or a neutrophil count of 0.5 x1 OVL, were compared with respect to MNC, CD34+ cell and GM-CFC numbers re-infused for 256 patients after high dose therapy. MNC numbers were not predictive in this diverse group of patients although we and others have reported this can be the case using well-standardized mobilization regimens (Scott et a!., 1995a; Watts et al., 1997c). A clear threshold was evident in this series at 1 xIOVkg CD34+ cells re-infused. The 41 patients who received CD34+ cells below this dose included all of the 3 patients with delayed neutrophil recovery and 20 of the 29 patients with delayed platelet recovery including all of those who failed to achieve platelet independence within 70 days. No significant increase in the incidence of delayed platelet recovery was noted among the patients receiving CD34+ cell doses between 1-3.5 xIOVkg (which was 8/102 overall). There was however a significantly reduced risk of

167 delayed platelet recovery between this group and recovery in those who received a CD34+ cell dose above 3.5 x10®/kg (1/113, P=0.0157) in line with our previous report (Watts et al., 1997c). A reinfused GM-CFC dose below 1 xIOVkg was also highly predictive of poor engraftment and included the three patients with delayed neutrophil recovery and 12 of the 29 patients with delayed platelet recovery. However, it was found in this larger series that 12 patients who experienced delayed platelet recovery were among the 43 patients who received between 1-2 xIOVkg GM-CFC which was not significantly different from those who received less than 1 x1 OVkg GM-CFC. If a GM-CFC threshold were used alone therefore a minimum criterion of 2 xIOVkg would be required. Notably, all of the 12 patients referred to above who had delayed platelet recovery and received 1-2 xIOVkg GM-CFC were also among those patients who received a CD34 dose below 1 x10®/kg. Above 3.5 xIOVkg GM-CFC delayed platelet recovery was rare. We have previously found low CD34+ cell and GM-CFC counts to be closely correlated (Watts et a!., 1997c), and this revised data has increased our previous minimum progenitor thresholds for direct progress to transplant from 1 x10®/kg CD34+ cells and 1 xIOVkg GM-CFC to 1 x10®/kg and 2 xIOVkg for CD34+ cells and GM-CFC respectively. Fortunately when >2 x107kg CD34+ cells are collected 98% of patients will have >2 xIOVkg GM-CFC. We therefore believe that if >2 xIOVkg CD34+ cell are collected it is appropriate to proceed to transplantation without waiting for culture data. It should be noted that haematological engraftment in this thesis refers to short-term engraftment sufficient to ensure independence from platelet support. Haas has shown in a retrospective study of 145 consecutive PBSCT patients that recovery to a platelet count of at least 150 x10®/L within 10 months was more likely in those patients who received threshold levels of 2.5 x10®/kg CD34+ cells or 1.2 xIOVkg GM-CFC. In this study there were 9 patients who received a median of only 0.18 (range 0.02- 0.49 ) xIOVkg GM-CFC who took 1-6 years to achieve normal counts. However none of the 93 patients assessable for long-term engraftment, with a maximum

168 follow-up of 15 years, developed graft failure (Haas et al., 1995). The clinical implications of slower long-term engraftment are not clear and an audit is in progress for our own patients. It was noted in Chapter 3 that of the 9 patients with slow neutrophil recovery between 21-28 days that there were 3 in each group receiving CD34+ cell thresholds of <1, 1-3.5 and >3.5 x10®/kg and that none of these patients were among the 81 patients who received early post transplant G-CSF. This is consistent with the beneficial effect of post infusion G-CSF which has been shown in randomised trials (Linch, Klumpp).

In Chapter 4, parameters predicting for progenitor yields were assessed for 101 pre-treated lymphoma patients all mobilized with cyclophosphamide 1.5g/m^ and G-CSF, with the first apheresis performed when the recovery WBC first exceeded 5.0 x 107L. The relationship between the number of progenitor cells collected and patient age, sex, diagnosis, prior radiotherapy and time since last chemotherapy was determined by multivariate analysis. Prior radiotherapy was found to be predictive for poor mobilization when the progenitor yield was considered as a continous variable in line with other reports (Haas eta!., 1994a; Schwartzberg eta!., 1993). Paradoxically wide field irradiation was not more detremental than local treatment and this may / have been a statistical "quirk" of small numbers. Interesting, however a recent study by Sharp et al showed that cytokine-induced HPC mobilization in mice was inhibited or abolished whether applied to a single limb, upper or lower half of the body. Furthermore, humoural inhibition of mobilization was seen when plasma from an irradiated mouse was injected into a normal mouse prior to cytokine adminstration (Sharp et al., 1998). This implies that impaired HPC mobilization cannot be assumed to only affect the area irradiated. The prediction of poor mobilization to threshold levels of 1 or 2 x107kg CD34+ cells at first apheresis against a dichotomous variable of <6 or >6 cycles of chemotherapy in Chapter 3 did not achieve significance and only became

169 suggestive where the number of chemotherapy cycles were considered as a continuous variable (P=0.06 for both 1 and 2 x107kg CD34+ cell thresholds). This lack of association is very similar to the findings in some studies (Ketterer e ta i, 1998; Kotasek etal., 1992), but not others (Drake et ai, 1997; Haas et ai, 1994a; Schwartzberg e ta i, 1993). In the Drake study a haematological toxicity scoring system for prior chemotherapy agents was devised which showed significantly poorer mobilization between those patients with heavy pre­ treatment and high scores compared to those with low scores. This was not completely predictive however and could not accurately determine which individuals would achieve a given progenitor threshold. This is perhaps not surprising in view of the wide baseline of progenitor mobilization among normal individuals (Roberts et ai, 1995) and the identification of "poor mobilizers" at present therefore relies on failure to achieve a minimal progenitor yield following a well-standardized protocol.

In Chapter 5 the extent of inter-individual progenitor cell mobilization was assessed in 20 healthy male volunteers who were part of a cross-over study to compare the mobilization efficacy of two types of G-CSF, filgrastim (non­ glycosylated) and lenograstim (glycosylated). The average peak levels of GM- CFC mobilized with the glycosylated G-CSF was 28% higher than the non­ glycosylated form but the intra-individual variation between each mobilization was much less than between individuals. There was more than a 10-fold variation between the best and worst mobilizers in agreement with other studies (Hoglund et ai, 1996; Jones et ai, 1996; Roberts et al., 1995; Watts et ai, 1997a). In contrast, even accounting for the difference in G-CSF type administered, the intra-individual progenitor numbers mobilized were relatively similar, in accord with a remobilization study reported by Anderlini (Anderlini et ai, 1997). Notably the worst mobilizer at first G-CSF exposure in Chapter 5 was also the worst on the second occasion. The reason for such wide inter­ individual variabilty in mobilization is unknown, but this study showed it was

170 unlikely to be due to variations in metabolism that could modulate serum G- CSF concentrations. Some studies have attempted to predict individuals who will mobilise HPC poorly on the assumption that the incremental rise in circulating progenitors will be fairly constant and can be predicted by measuring baseline or "steady state" circulating GD34+ cell numbers. In a study reported by Fruehauf et al for example, pre-mobilization circulating CD34+ cells were measured in 15 pre­ treated cancer patients and an individual estimate of subsequent yield calculated with a 95% probability of success. A recent study by Brown et ai, also used baseline pre-cytokine CD344- cell counts was used in a similar way to predict the subsequent progenitor yield in 47 normal donors and is detailed in Chapter 5, p104 (Brown etai., 1997). A major disadvantage of this approach of poor mobilizer identification is the requirement for exceptionally precise flow cytometry. The steady-state blood CD34 numbers for those patients who would require more than one apheresis to achieve the minimum progenitor yield of 2.5 x10®/kg in the Fruehauf report ranged from 1.0-0.2 x10®/L (Fruehauf et a/.,

1995). Assuming a WBC of 10 x 1 0 7 L for example, this equates to a CD34+ cell concentration of 0.01-0.002%. In the Brown study the baseline counts of healthy donors ranged from 0.01-0.15% (mean 0.05%) which gave absolute counts of 0.8-11.4 x10®/L (mean 4.3 x107L) (Brown et a/., 1997). Many clinical laboratories would be likely to find this level of precision challenging judging by the wide variability in CD34-h cell measurement at much higher levels seen in inter-laboratory quality assurance exercises (Barnett et a/., 1998; Chin-Yee et a/., 1997; Gratama etai., 1997; Gratama etai., 1998). Despite the demonstration of a high degree of correlation between the immediate pre-apheresis blood CD34+ cell count of G-CSF mobilized normal donors and baseline CD34+ cell counts or total G-CSF received in the Brown study, these two factors accounted for less than half of the variability in CD34+ cells mobilized (Brown etai., 1997), suggesting other important reasons for mobilization variability.

171 It is possible that since bone marrow stores of stem cells would be expected to exceed a normal lifespan there may be little evolutionary pressure to restrict random genetic variation in stem cell frequency above the minimum level required. Some evidence for this notion has been demonstrated in mice where the chromosomal location of "stem cell frequency genes" has been identified. In separate studies reported by Muller-Sieberg and Roberts, a narrow range of LTCIC numbers was seen within individual mouse strains but at least a 10-fold variation between strains (Muller-Sieburg & Riblet, 1996; Roberts et al., 1997). In a subsequent report, Muller-Sieberg described two mice strains with high marrow LTCIC frequency in one instance and low in the other. Both types were mobilized with G-CSF and showed the mobilization of blood LTCIC in proportionally high or low numbers.

In Chapter 6 the value of HPC support provided by the collection of autologous bone marrow was assessed for those patients who failed to collect a minimum yield of peripheral blood stem cells. The lowest level of PBSC CD34+ cells to ensure engraftment is thought to be 1 xIOVkg (Bensinger et al., 1994; Watts at al., 1997c) and there was a failure to harvest this dose in 51/324 of consecutively mobilised patients. Twenty three of these had >1x10Vkg GM- CFC, 22 proceeding to HOT with concomitant "back-up" bone marrow in two of these patients with very poor CD34+ yields of <0.5 x10®/kg. Neutrophil recovery was within 21 days but platelet independence was delayed (>28 days) in 8. Of the 28 patients with <1x107kg GM-CFC, 6 received HOT with PBSC alone and 5 had delayed engraftment. Twelve patients with <1x10Vkg GM-CFC received HOT supported by bone marrow collected after PBSC collection failure. Eleven patients were evaluable for engraftment and 4 had slow (>21 days) or delayed (>28 days) neutrophil recovery and 8 delayed platelet recovery. In the group of patients receiving <1x107kg GM-CFC there were 5 procedure related deaths. This study illustrates that delayed haematological recovery is frequent if <1x10Vkg CD34+ cells are infused after

172 HDT particulary with GM-CFC <1x10Vkg. The procedure related mortality in this latter group is high. In most patients whose PBSC harvest contains <1x10Vkg GM-CFC the use of bone marrow cells does not improve engraftment suggesting that poor PBSC mobilization usually indicates poor marrow function. It should be noted that poor PBSC yields are not always the result of impaired marrow stores. With some stem cell toxic mobilizing regimens for example, poor mobilization may not adversely affect marrow progenitor yields ((Dreger et al., 1995). In some instances poor yields may result from inappropriate mobilizing chemotherapy (Demirer & Bensinger, 1995), insufficient recovery from prior treatment (Perry et a/., 1998), or technical failures in harvest timing or apheresis procedures. If clinical circumstances permit it is probably preferable to allow at least 3 months recovery from chemotherapy and remobilize and harvest with an optimal protocol before assuming a mobilization failure.

In Chapter 7, one hundred and sixteen PBSC collections were subject to CD34+ cell purification using the CEPRATE® SC stem cell concentration system. The overall median purity of CD34+ cells was 70% (6-95%). CD34+ cell, and GM-CFC recoveries were 52% (8-107%) and 36% (3-118%). Purity was logarithmically related to the input percentage of CD34+ cells and starting requirements were established of 1% CD34 cell content for optimal purity and a minimum of 2 xIOVkg CD34+ cells to ensure recovery of our minimum engraftment threshold of 1 xIOVkg CD34+ cells. Reduction of the washing steps reduced non-specific cell losses and shortened the procedure but did not affect progenitor cell recovery. Purified CD34+ cells were reinfused following high dose therapy in 52 patients. The median time to neutrophil recovery of 0.5 x10®/L was 12 (9-23) days and to the attainment of platelet independence was 13 (7-100) days. The risks of delayed platelet recovery were related to the CD34+ cell dose infused and were identical to the risks when non-purified PBSC collections were used. Purification of CD34+ cells using the CEPRATE

173 device was shown to be reliable and the purified product resulted in prompt engraftment. The cell losses that occur do however restrict its use in many patients.

Haemopoietic recovery is more rapid after peripheral blood stem cell (PBSC) transplantation than after autologous bone marrow transplantation, and the aim of the study in Chapter 8 was to assess the role of the large number of lymphocytes and monocytes (accessory cells) in a PBSC leucapheresis product in this rapid regeneration. It has been noted by Mielcarek for example that PBSC harvests contain many more cells associated with cytokine production including at least 1-log more T-cells and 50-fold more monocytic cells (CD 14+) compared to a typical marrow graft (Mielcarek & Torok-Storb, 1997). Mielcarek provided in vitro evidence that sorted CD14+ cells from PBSC were associated with at least 10-fold higher levels of supernatent G-CSF and IL-6 cytokine production when incubated on long-term bone marrow stroma than equivalent marrow MNC and proposed that this may facilitate faster haematological engraftment with PBSC in vivo (Mielcarek et al., 1996). Higher plasma G-CSF and IL-6 levels post PBSCT compared to marrow autografts were noted by a number of other authors (Baiocchi at al., 1993; Ho at al., 1994; Kawano at al., 1993; Testa at al., 1994). Haematological recovery was therefore assessed in 10 PBSC recipients with lymphoma or myeloma in whom monocytes and T-cells were depleted by a median of 2.3 and 3.3 logs by CD34+ cell selection using the CEPRATE® SC stem cell concentration system and compared with recovery in 59 recipients who received whole PBSC. After allowing for the number of progenitor cells reinfused, there was no significant delay in engraftment induced by accessory cell depletion. Plasma levels of granulocyte-colony stimulating factor (G-CSF), granulocyte/monocyte-colony stimulating factor (GM-CSF), interleukin-6 (IL-6), stem cell factor (SCF) and macrophage-inhibitory pnitèin-alpha (MIP-1-alpha) during the transplant procedure were similar whether or not accessory cells were given. The G-CSF

174 and IL-6 levels rose between days 5 to 14 post transplantation to approximately 1 ng/ml and 50 pg/ml respectively. This study indicates that accessory cells reinfused with PBSC collections are not responsible for the subsequent cytokine profile or rapid haematological recovery. Intriguingly, in a recent in vitro study by Gupta et al, purified GD34+ bone marrow cells were co-cultured with irradiated stromal cells either in direct contact or separated by a transwell membrane and supernatant G-CSF and IL-6 concentrations measured. A significant 5-fold increase over baseline stromal supernatant levels to around lOOOpg/mL for IL-6 and 250pg/mL for G-CSF was observed in both culture conditions with CD34+ cells over a seven day period and suggests humoural stimulation of stromal cell cytokine production (Gupta et al., 1998). When more mature myelo-monocytic cells (CD15+CD14+) were incubated with the irradiated stromal cells supernatent IL-6 and G-CSF levels were also increased but only by 1.5-2-fold and transiently after 48hrs incubation. This study raises the interesting possibility that infused CD34+ cells in vivo may be directly responsible for a significant contribution to elevated cytokine levels.

The requirements of PBSC autografting are now well established and the work of these thesis contributes to the understanding of progenitor requirements, the feasibility and limitations of progenitor purification and addresses some of the issues of poor mobilization. This experience provides a platform which is essential to facilitate the major advances in graft engineering and cellular therapy anticipated in the next few years. A notable feature of PBSCT has been its rapid clinical application based on the previous long experience with marrow transplantation. Despite some biological differences between these progenitor sources this has proved highly successful and durable for autografting (Haas et al., 1995). A more cautious approach is justifiable for the steadily rising replacement of marrow allografting with blood MPC however as these cells will provide the sole source of engraftment for the lifetime of the recipient.

175 Although there are at least 10-fold more T-cells infused in allogeneic PBSCT there has been no significant increase in acute GvHD reported and, as in the autologous setting, haematological recovery is significantly more rapid (Hagglund et al., 1998; Ringden eta!., 1998) with resultant savings of 30% in the cost of the first 100 days estimated in one study (Faucher et al., 1998). The greater number of stem cells possible with PBSC has also facilitated the use of so called low-intensity or "mini-transplants" in which chimerism and hopefully a graft versus tumour effect is established without fully myeloablative therapy (Slavin et al., 1998). This appears to be less toxic than conventional transplants and can therefore be used for patients previously excluded because of higher age or organ dysfunction for example (Giralt et al., 1997).

Despite these apparent advantages, the safety of the donor is of paramount concern and the adverse consequences of the mobilization and collection of PBSC must be weighed against previous experience with bone marrow donation. There have been two major studies addressing the risks of marrow donation in normal donors. Firstly, Buckner and colleagues reported their experience of 1,270 such collections. Six donors had life-threatening complications including two cardiopulmonary arrests and a cerebrovascular event although there were no deaths. Significant operative site morbidity occurred in 10 donors with haematomas causing pressure neuropathies in 6 (Buckner et ai, 1984). Secondly, Stroncek et al have reported on the first 493 unrelated donors in the US National Bone Marrow Donor Program. Only one life-threatening event occurred but there was considerable morbidity that persisted for longer than might be generally appreciated. Twenty nine donors (6%) suffered from acute complications including hypotension, infection and haemorrhage at the marrow collection site. Between one to two weeks after the procedure 24% of donors were feeling fatigued and 27% were still suffering from pain at the collection site. 13% of donors had not fully recovered by 30 days (Stroncek et ai, 1993). In summary, marrow donation requires a general

176 anaesthetic and operating theatre procedure and although there were no fatalities these series they show that it cannot be regarded as a trivial procedure and it against this background that the benefits and risks of G-CSF administration and apheresis collection of PBSC must be assessed.

Comprehensive, long-term donor safety following G-CSF treatment and PBSC collection is rapidly accumulating but still requires continual monitoring. A concensus view has emerged that the short-term safety profile for normal donors receiving up to 10^ig/kg/day sc is acceptable for clinical use (Anderlini et al., 1998). A positive dose-response relationship between higher G-CSF levels and progenitor yield has been shown in several studies but they may produce more severe side-effects [Stroncek, 1996 #361]. There are several consistent adverse consequences of G-CSF administration in normal donors. Leukocytosis seldom exceeds 60-70 x10^/L but should prompt intervention significantly above this level as this is a risk factor for thrombosis (Anderlini et a!., 1998). A study of 102 normal donors receiving G-CSF for 5 or 10 days at doses of 2, 5, 7.5 or 10|ig/kg was reported by Stroncek (Stroncek et a!., 1996). The overall incidence of side-effects in donors was as follows; bone pain 83%, headache 39%, body ache 23% and of fatigue was 14%. Nausea and vomiting affected 12% and abnormal liver function tests were also common findings but were not symptomatic. These adverse events were shown to be increased in those receiving higher doses of G-CSF. Of at least equal importance to adverse donor events is the apheresis protocol used to collect PBSC. In a recent international survey for example the risks of central line insertion was considered completely unacceptable for normal donors by 10/28 countries who would only allow peripheral venous access (Cleaver & Goldman, 1998). The immediate side-effects of leukapheresis have been documented by Wiesneth ef a/who reported a study of 97 normal donors following a total of 197 apheresis collections. These included citrate toxicity in around 5% of donors and a drop in platelet count of around 30%. Resulting

177 thrombocytopaenia (<80 x10®/L) in 16% of patients was treated by re-infusion of their platelet rich plasma (Wiesneth etal., 1998). A recent report by Novotny et al of 48 normal donors who underwent a median of 2 (1-4) large volume leukapheresis collections following G-CSF administration twice daily at 5p.g/kg sc. was of particular concern as long-term side-effects were observed. Lymphocyte counts were significantly below the median pre-mobilization count of 2.3 x107L at 1 month post collection (1.3 x107L, P<0.0001) and even at 11 months (1.5 x10®/L, P=0.0005). In four cases nadir lymphocyte counts below 1 x10®/L were noted (0.4, 0.7, 0.9 and 0.9 x107L), although this was not associated with an apparent susceptibility to infection (Novotny et a!., 1998).

In order to minimise these risks to healthy donors therefore it is necessary to establish progenitor requirements for different types of allogeneic transplantation protocols. The minimum CD34+ cell dose used by the Seattle group in an allogeneic PBSC study was 7.1 xIOVkg recipient body weight (Bensinger et a!., 1996) although Schmitz et ai have used as little as 2.1 xIOVkg (Schmitz etal., 1995) and a Spanish collaborative study reported by Urbane showed no advantage to using more than 2.5 xIOVkg (Urbano-lspizua et al., 1996). A dose of 3 x1 OVkg CD34+ cells appears acceptable to date. The Spanish group recently reported their experience of 62 allogeneic transplantation procedures for haematological malignancies using purified CD34+ cells from HLA-matched sibling donors. Haematological recovery was rapid in those patients who received over 1 x1 OVkg CD34+ purified cells and acute graft versus host disease was reduced compared to their previous experience with marrow or unmanipulated PBSC transplantation. Grade 0, I, II and III acute GvHD was seen in 43, 8, 4 and 1 patient respectively. No patients experienced grade IV GvHD but six patients developed chronic GvHD which was limited in two cases but extensive in four. T-cells (CD3+) in the CD34+ cell purified fraction were present at a median dose (range) of 0.4 (0.01-2 xIOVkg) for the 38 patients who received immunoaffinity purified CD34+ cells

178 (CEPRATE system) and 0.14 (0.03-2.5) xIOVkg for the remaining 24 who had their donor CD34+ cells selected by immunomagnetic beads (Baxter Isolex system) (Urbano-lspizua et al., 1998). This data is in accord with our own experience of these systems where in a recent audit of three clinical CD34 cell selection systems, the CEPRATE, Isolex and CliniMAGS devices, we have observed median T-cell log depletions from PBSC of 3, 3.6 and 5.7 respectively (unpublished observations). The effectiveness of the CliniMACS device for T-cell depletion is particularly interesting as it may facilitate HLA mismatched transplantation without severe GvHD.

Looking further ahead, gene therapy is a natural progression within the field of haemopoietic and graft engineering. Since the first human gene therapy trial to treat adenine deaminase deficiency in 1989 there have been more than 170 approved gene therapy trials in the US alone and more than 1500 patients have been enrolled in gene therapy trials worldwide (Knoell & Yiu, 1998). It is not yet known which vector system will prove most effective. Proof of principle has been shown but transduction rates remain disappointingly low and definitive clinical benefit has yet to be shown (Devereux et a!., 1998; Prince, 1998).

Dendritic cells (DCs) are the most potent of antigen presenting cells and are uniquely able to stimulate naive T-cells. Hsu et ai demonstrated that this ability could be activated in a pilot trial of four patients with B-cell lymphoma who received tumour antigen-pulsed autologous dendritic cells. There was a measurable cellular response in all patients and clinical regression of disease in three. This has led a number of clinical trials where DCs are exposed to clinically relevant antigens in patients with viral infections or malignancy (Lotze etal., 1997). The discovery that DCs can be generated from CD34+ cells under the influence of GM-CSF and TNFa or differentiated from blood monocytes with GM-CSF and IL-4 (Steinman, 1996), has further intensified

179 research activity. In the first instance because CD34+ cell division and expansion may facilitate retro-viral gene transfer and in the second instance, because large numbers of blood monocytes are readily collectable by apheresis. The large number of biological variables in the DC immunotherapy protocol and patient group chosen is likely to mean that judging clinical efficacy will be time-consuming but any clear clinical benefits should be evident in the next 3 years or so.

An interesting cytokine in the field of cellular therapy which is likely to make some impact in the next few years is Flt3 ligand, mentioned previously in the introduction section on mechanisms of HPC mobilization. It appears to mobilize HPC independently from G-CSF, cyclophosphamide, or integrin- mediated mechanisms. Serum Flt3 ligand levels have also been reported as a marker of marrow activity and to show an inverse relationship to circulating CD34+ cell counts (Haug et al., 1998). Moreover, it has been given to normal volunteers and uniquely, mobilizes circulating dendritic cells (Maraskovsky et a!., 1997). It has also been shown to mediate tumour specific responses in animal models (Lyman, 1998; Lynch, 1998) and may be an important adjuvant for DC immunotherapy. It also synergises with a wide range of other colony stimulating factors and interleukins and has been shown to support LTCIC expansion in mice in combination with stem cell factor (Miller & Eaves, 1997). It may therefore be an important component in attempts to expand progenitors from cord blood and autografts from heavily pre-treated patients as discussed in the introduction.

Finally, blood HPC transplantation, progenitor selection and other manipulations of autologous and allogeneic grafts has grown exponentially during the course of this thesis due to rise in the biotechnology industry and the availability of growth factors, early success with the procedures and the accessibility of blood components. It seems likely that the technology

180 developed for current applications of graft engineering will come to include related cell manipulations for therapeutic use which are now emerging. These include the isolation and expansion of circulating endothelial cell progenitors for angiogenesis (Asahara et al., 1997; Gupta etal., 1997) and the selection and expansion of marrow-derived mesenchymal stem cells which can be differentiated into mesenchymal tissues such as bone, cartilage or muscle in the appropriate culture conditions (Fleming et al., 1998; Majumdar et al., 1998).

These are exciting times for the foundation of cellular therapy and there are so many current and developmental studies ongoing that most future predictions are likely to be wrong. The only certainty is that expansion of the field is inevitable.

181 References

Alcorn, M.J., Holyoake, T.L., Richmond, L.J., Pearson, C., Farrell, E., Kyle, B., Dunlop, D.J., Fitzsimmons, E., Stewart, W.P., Pragrïôil, I.B. & Franklin, I.M. (1995). CD34 selection using the Baxter Isolex 300 system. Comparison between fresh PBPC and cryopreserved PBPC. Blood, 86, 230a (abstract 906).

Anderlini, P., Lauppe, J., Przepiorka, □., Seong, D., Champlin, R. & Korbling, M. (1997). Peripheral blood stem cell apheresis in normal donors: feasibility and yield of second collections. B rJ Haematol, 96, 415-7.

Anderlini, P., Przepiorka, □., Korbling, M. & Champlin, R. (1998). Blood stem cell procurement: donor safety issues. Bone Marrow Transplant, 21 SuppI 3, S35-9.

Anderson, K.C., Soiffer, R., DeLage, R., Takvorian, T., Freedman, A.S., Rabinowe, S.L., Nadler, L.M., Dear, K., Heflin, L., Mauch, P. & et al. (1990). T-cell-depleted autologous bone marrow transplantation therapy: analysis of immune deficiency and late complications. Blood, 76, 235- 44.

Armitage, S., Hargreaves, R., Samson, D., Brennan, M., Kanfer, E. & Navarrete, C. (1997). CD34 counts to predict the adequate collection of peripheral blood progenitor cells. Bone Marrow Transplant, 20, 587-91.

Arnold, R., Schmeiser, T., Heit, W., Frickhofen, N., Pabst, G., Heimpel, H. & Kubanek, B. (1986). Hemopoietic reconstitution after bone marrow transplantation. Exp Hematol, 14, 271-7.

Asahara, T., Murohara, T., Sullivan, A., Silver, M., van der Zee, R., Li, T., Witzenbichler, B., Schatteman, G. & Isner, J.M. (1997). Isolation of putative progenitor endothelial cells for angiogenesis. Science, 275, 964-7.

Attal, M., Harousseau, J.L., Stoppa, A.M., Sotto, J.J., Fuzibet, J.G., Rossi, J.F., Casassus, P., Maisonneuve, H., Façon, T., If rah, N., Payen, C. & Bataille, R. (1996). A prospective, randomized trial of autologous bone marrow transplantation and chemotherapy in multiple myeloma. Intergroupe Français du Myelome. N-Engl-J-Med, 335, 91-7

Baiocchi, G., Scambia, G., Benedetti, P., Menichella, G., Testa, U., Pierelli, L., Martucci, R., Foddai, M.L., Bizzi, B., Mancuso, S. & et al. (1993). Autologous stem cell transplantation: sequential production of hematopoietic cytokines underlying granulocyte recovery. Cancer Res, 53, 1297-303.

Barnett, D., Granger, V., Storie, I., Peel, J., Pollitt, R., Smart, T. & Reilly, J.T. (1998). Quality assessment of CD34+ stem cell enumeration: experience of the United Kingdom National External Quality Assessment Scheme (UK NEQAS) using a unique stable whole blood preparation. B rJ Haematol, 102, 553-65.

Barrett, J. (1991). Worldwide bone marrow transplant activity In the last decade. New strategies in bone marrow transplantation. Wiley-Liss Inc.: New York.

Basser, R.L., Rasko, J.E., Clarke, K., Cebon, J., Green, M.D., Grigg, A.P., Zalcberg, J., Cohen, B., O'Byrne, J., Menchaca, D.M., Fox, R.M. & Begley, C.G. (1997). Randomized, blinded,

182 placebo-controlled phase I trial of pegylated recombinant human megakaryocyte growth and development factor with filgrastim after dose-intensive chemotherapy in patients with advanced cancer [published erratum appears in Blood 1997 Sep 15;90(6):2513]. Blood, 89, 3118-28.

Basser, R.L, To, LB., Begley, C.G., Maher, D., Juttner, C., Cebon, J., Mansfield, R., Olver, I., Duggan, G., Szer, J., Collins, J., Schwartz, B., Marty, J., Menchaca, □., Sheridan, W.P., Fox, R.M. & Green, M.D. (1998). Rapid hematopoietic recovery after multicycle high-dose chemotherapy: enhancement of filgrastim-induced progenitor-cell mobilization by recombinant human stem-cell factor. J Clin Oncol, 16, 1899-908.

Begley, C.G., Basser, R., Mansfield, R., Thomson, B., Parker, W.R., Layton, J., To, B., Cebon, J., Sheridan, W.P., Fox, R.M. & Green, M.D. (1997). Enhanced levels and enhanced clonogenic capacity of blood progenitor cells following administration of stem cell factor plus granulocyte colony-stimulating factor to humans. Blood, 90, 3378-89.

Bensinger, W., Appelbaum, F., Rowley, S., Storb, R., Sanders, J., Lilleby, K., Gooley, T., Demirer, T., Schiffman, K., Weaver, C. & et al. (1995). Factors that influence collection and engraftment of autologous peripheral-blood stem cells. J Clin Oncol, 13, 2547-55.

Bensinger, W.I., Clift, R., Martin, P., Appelbaum, F.R., Demirer, T., Gooley, T., Lilleby, K., Rowley, S., Sanders, J., Storb, R. & Buckner, C.D. (1996). Allogeneic peripheral blood stem cell transplantation in patients with advanced hematologic malignancies: a retrospective comparison with marrow transplantation. Blood, 88, 2794-800.

Bensinger, W.I., Longin, K., Appelbaum, F., Rowley, S., Weaver, C., Lilleby, K., Gooley, T., Lynch, M., Higano, T., Klarnet, J. & et al. (1994). Peripheral blood stem cells (PBSCs) collected after recombinant granulocyte colony stimulating factor (itiG-CSF): an analysis of factors correlating with the tempo of engraftment after transplantation [see comments]. Br-J-Haematol, 87, 825-31

Berenson, R.J., Bensinger, W.I., Hill, R.S., Andrews, R.G., Garcia-Lopez, J., Kalamasz, D.F., Still, B.J., Spitzer, G., Buckner, C.D., Bernstein, I.D. & et al. (1991). Engraftment after infusion of CD34+ marrow cells in patients with breast cancer or neuroblastoma. Blood, 77,1717-22.

Beverley, P.C., Linch, D. & Delia, D. (1980). Isolation of human haematopoietic progenitor cells using monoclonal antibodies. Nature, 287, 332-3.

Beyer, J., Schwella, N., Zingsem, J., Strohscheer, I., Schwaner, I., Oettle, H., Serke, S., Huhn, D. & Stieger, W. (1995). Hematopoietic rescue after high-dose chemotherapy using autologous peripheral-blood progenitor cells or bone marrow: a randomized comparison. J-Clin-Oncol, 1 3, 1328-35 .

Bezwoda, W.R., Seymour, L. & Dansey, R.D. (1995). High-dose chemotherapy with hematopoietic rescue as primary treatment for metastatic breast cancer: a randomized trial [see comments]. J-Clin-Oncol, 13, 2483-9

Bishop, M.R., Tarantolo, P.J., Bierman, J.M., Vose, J.O., Armitage, Z.S., Pavletic, J., Lynch, J., Kollath, R.L., Reddy, P.I., Warkentin, A. & Kessinger, A. (1997). Predictive factors for the identification of allogeneic blood stem cell donors as "poor mobilizers" prior to stem cell collection. Blood, 90, 592a (abstract 2632).

183 Boiron, J.M., Marit, G., Faberes, C., Cony-Makhoul, P., Foures, C., Ferrer, A.M., Cristol, G., Sarrat, A., Girault, D. & Reiffers, J. (1993). Collection of peripheral blood stem cells in multiple myeloma following single high-dose cyclophosphamide with and without recombinant human granulocyte- macrophage colony-stimulating factor (rhGM-CSF). Bone Marrow Transplant, 12, 49-55.

Bortin, M.M., Horowitz, M.M., Gale, R.P., Barrett, A.J., Champlin, R.E., Dicke, K.A., Gluckman, E., Kolb, H.J., Marmont, A.M., Mrsic, M. & et al. (1992). Changing trends in allogeneic bone marrow transplantation for leukemia in the 1980s. Jama, 268, 607-12.

Brecher, G. & Cronkite, E.P. (1951). Postirradiation parabiosis and survival in rats. Proa Soc Exp Biol Med, 77, 292-294.

Brenner, M.K. (1993). Gene transfer into human hematopoietic progenitor cells; a review of current clinical protocols. J Hematother, 2, 7-17.

Briddell, R.A., Kern, B.P., Zilm, K.L., Stoney, G.B. & McNiece, I.K. (1997). Purification of CD34+ cells is essential for optimal ex vivo expansion of umbilical cord blood cells. J Hematother, 6, 145- 50.

Brown, R.A., Adkins, D., Goodnough, L.T., Haug, U.S., Todd, G., Wehde, M., Hendricks, D., Ehlenbeck, C., Laub, L. & DiPersio, J. (1997). Factors that influence the collection and engraftment of allogeneic peripheral-blood stem cells in patients with hematologic malignancies. J Clin Oncol, 15, 3067-74.

Broxmeyer, H.E., Gluckman, E., Auerbach, A., Douglas, G.W., Friedman, H., Cooper, S., Hangoc, G., Kurtzberg, J., Bard, J. & Boyse, E.A. (1990). Human umbilical cord blood: a clinically useful source of transplantable hematopoietic stem/progenitor cells. Int J Cell Cloning, 8 SuppI 1, 76-89; discussion 89-91.

Broxmeyer, H.E., Orazi, A., Hague, N.L., Sledge, G.W., Jr., Rasmussen, H. & Gordon, M.S. (1998). Myeloid progenitor cell proliferation and mobilization effects of BB10010, a genetically engineered variant of human macrophage inflammatory protein-1 alpha, in a phase I clinical trial in patients with relapsed/refractory breast cancer. Blood Cells Mol Dis, 24,14-30.

Brugger, W., Birken, R., Bertz, H., Hecht, T., Pressler, K., Frisch, J., Schulz, G., Mertelsmann, R. & Kanz, L. (1993). Peripheral blood progenitor cells mobilized by chemotherapy plus granulocyte-colony stimulating factor accelerate both neutrophil and platelet recovery after high- dose VP16, ifosfamide and cisplatin. Br J Haematol, 84, 402-7.

Brugger, W., Henschler, R., Heimfeld, S., Berenson, R.J., Mertelsmann, R. & Kanz, L. (1994). Positively selected autologous blood CD34+ cells and unseparated peripheral blood progenitor cells mediate identical hematopoietic engraftment after high-dose VP16, ifosfamide, carboplatin, and epirubicin. Blood, 84, 1421-6.

Buckner, C.D., Clift, R.A., Sanders, J.E., Stewart, P., Bensinger, W.I., Doney, K.C., Sullivan, K.M., Witherspoon, R.P., Deeg, H.J., Appelbaum, F.R. & et al. (1984). Marrow harvesting from normal donors. Blood, 64, 630-4.

Cagnoni, P.J. & Shpall, E.J. (1996). Mobilization and selection of CD34-positive hematopoietic progenitors. Blood Rev, 10, 1-7.

184 Cebon, J., Nicola, N., Ward, M., Gardner, I., Dempsey, P., Layton, J., Duhrsen, U., Burgess, A.W., Nice, E. & Morstyn, G. (1990). Granulocyte-macrophage colony stimulating factor from human lymphocytes. The effect of glycosylation on receptor binding and biological activity. J Biol Chem, 265, 4483-91.

Chin-Yee, I., Anderson, L., Keeney, M. & Sutherland, D.R. (1997). Quality assurance of stem cell enumeration by flow cytometry. Canadian CASE Study Group. Cytometry, 30, 296-303.

Chopra, R., McMillan, A.K., Linch, D.C., Yuklea, S., Taghipour, G., Pearce, R., Patterson, K.G. & Goldstone, A.H. (1993). The place of high-dose BEAM therapy and autologous bone marrow transplantation in poor-risk Hodgkin's disease. A single-center eight- year study of 155 patients. Blood, 1137-45.

Civin, C.I., Almeida-Porada, G., Lee, M.J., Olweus, J., Terstappen, L.W. & Zanjani, E.D. (1996). Sustained, retransplantable, multilineage engraftment of highly purified adult human bone marrow stem cells in vivo. Blood, 88, 4102-9.

Cleaver, S.A. & Goldman, J.M. (1998). Use of G-CSF to mobilise PBSC in normal healthy donors- an international survey. Bone Marrow Transplant, 21 SuppI 3, S29-31.

Collins, P., Watts, M., Brocklesby, M., Gerritsen, B. & Veys, P. (1997). Successful engraftment of haploldentical stem cell transplant for familial haemophagocytic lymphohistiocytosis using both bone marrow and peripheral blood stem cells. Br J Haematol, 96, 644-6.

Cox, D.R. (1972). Regression models and life tables. J R Stat Soc, 34.

Craig, J.I., Anthony, R.S., Stewart, A., Thomson, E.B., Giilon, J. & Parker, A.C. (1993a). Peripheral blood stem cell mobilization using high-dose cyclophosphamide and G-CSF in pretreated patients with lymphoma. Br J Haematol, 85, 210-2.

Craig, W., Kay, R., Cutler, R.L. & Lansdorp, P.M. (1993b). Expression of Thy-1 on human hematopoietic progenitor cells. J Exp Med, 177, 1331-42.

Cynshi, O., Satoh, K., Shimonaka, Y., Hattori, K., Nomura, H., I mai, N. & Hirashima, K. (1991). Reduced response to granulocyte colony-stimulating factor in W/Wv and Sl/SId mice. Leukemia, 5, 75-7.

Damiani, D., Fanin, R., Silvestri, F., Grimaz, S., Inf anti, L., Geromin, A., Cerno, M., Michieli, M., Rinaldi, C., Savignano, C., Trani, G., Fiacchini, M. & Baccarani, M. (1997). Randomized trial of autologous filgrastim-primed bone marrow transpiantation versus filgrastim-mobilized peripheral blood stem cell transplantation in lymphoma patients. Blood, 90, 36-42.

Daussett, J., Rapaport, F.T., Ivanyi, P. & Colombani, J. (1965). Tissue alloantlgens and transplantation. Histocompatibility Testing. Munksgaard: Copenhagen.

Decieva, E., Cramer, R. & Zabucchi, G. (1995). Glycosylation improves the priming effect exerted by recombinant human granulocyte colony-stimulating factor (lenograstim) on human neutrophil superoxide production. Int J Tissue React, 17, 191-8.

Delclaux, C., Delacourt, C., D'Ortho, M.P., Boyer, V., Lafuma, C. & Harf, A. (1996). Role of gelatinase B and elastase in human polymorphonuclear neutrophil migration across basement membrane. Am J Respir Cell Mol Biol, 14, 288-95.

185 Demirer, T. & Bensinger, W.l. (1995). Optimization of peripheral blood stem cell collection. Curr Opin Hematol, 2, 219-26.

Demuynck, H., Delforge, M., Verhoef, G., Zachee, P., Vandenberghe, P. & Boogaerts, M. (1995). Comparative study of peripheral blood progenitor cell collection in patients with multiple myeloma after single-dose cyclophosphamide combined with rhGM-CSF or rhG-CSF [see comments]. Br-J-Haematol, 90, 384-92

Dercksen, M.W., Gerritsen, W.R., Rodenhuis, S., Dirkson, M.K., Slaper-Cortenbach, 1.0., Schaasberg, W.P., Pinedo, H.M., von dem Borne, A.E. & van der Schoot, C.E. (1995a). Expression of adhesion molecules on 0034+ cells: CD34+ L-selectin+ cells predict a rapid platelet recovery after peripheral blood stem cell transplantation. Blood, 85, 3313-9.

Dercksen, M.W., Rodenhuis, S., Dirkson, M.K., Schaasberg, W.P., Baars, J.W., van der Wall, E., Slaper-Cortenbach, I.C., Pinedo, H.M., Von dem Borne, A.E., van der Schoot, C.E. & et al. (1995b). Subsets of CD34+ cells and rapid hematopoietic recovery after peripheral-blood stem­ cell transplantation. J Clin Oncol, 13, 1922-32.

Devereux, S., Corney, C., Macdonald, C., Watts, M., Sullivan, A., Goldstone, A.H., Ward, M., Bank, A. & Linch, D.C. (1998). Feasibility of multidrug resistance (MDR-1) gene transfer in patients undergoing high-dose therapy and peripheral blood stem cell transplantation for lymphoma. Gene Ther, 5, 403-8.

Drake, M., Ranaghan, L., Morris, T.C., Nolan, L., Desai, Z.R., Irvine, A.E., Jordan, A., Magill, K. & Price, S. (1997). Analysis of the effect of prior therapy on progenitor ceil yield: use of a chemotherapy scoring system [published erratum appears in Br J Haematol 1997 Nov;99(2):479]. B rJ Haematol, 98, 745-9.

Dreger, P., Kloss, M., Petersen, B., Haferlach, T., Loftier, H., Loeffler, M. & Schmitz, N. (1995). Autologous progenitor cell transplantation: prior exposure to stem celi-toxic drugs determines yield and engraftment of peripheral blood progenitor cell but not of bone marrow grafts. Blood, 86, 3970-8

Dreger, P., Marquardt, P., Haferlach, T., Jacobs, S., Mulverstedt, T., Eckstein, V., Suttorp, M., Loftier, H., Muller-Ruchholtz, W. & Schmitz, N. (1993). Effective mobilisation of peripheral blood progenitor cells with 'Dexa- BEAM' and G-CSF: timing of harvesting and composition of the leukapheresis product. Br J Cancer, 68, 950-7.

Duhrsen, U., Villeval, J.L., Boyd, J., Kannourakis, G., Morstyn, G. & Metcalf, D. (1988). Effects of recombinant human granuiocyte coiony-stimulating factor on hematopoietic progenitor cells in cancer patients. Blood, 72, 2074-81.

Dunbar, C.E., Cottler-Fox, M., O'Shaughnessy, J.A., Doren, S., Carter, C., Berenson, R., Brown, S., Moen, R.C., Greenblatt, J., Stewart, F.M. & et al. (1995). Retrovirally marked CD34-enriched peripheral blood and bone marrow cells contribute to long-term engraftment after autologous transplantation. Blood, 85, 3048-57.

Durrant, S.T.S., Hutchins, C.J., Weatern, R., Choo, P., Morton, J. & Fanning, S. (1997). Rapid platelet engraftment following allogeneic bone marrow transplantation using filgrastim-primed bone marrow. Blood, 90, abstract 4600.

186 Elliott, C., Samson, D.M., Armitage, S., Lyttelton, M.P., McGuigan, D., Hargreaves, R., Giles, C., Abrahamson, G., Abboudi, Z., Brennan, M. & Kanfer, E.J. (1996). When to harvest peripheral- blood stem cells after mobilization therapy: prediction of GD34-positive ceil yield by preceding day CD34-positive concentration in peripheral blood. J-Clin-Oncol, 14, 970-3

Epstein, R.B., Graham, T.C., Buckner, C.D. & al, e. (1966). Allogeneic marrow engraftment by cross-circulation in lethally irradiated dogs. Blood, 28, 692-707.

Erber, W.N., Pinching, A.J. & Mason, D.Y. (1984). Immunocytochemical detection of T and B cell populations in routine blood smears. Lancet, 1, 1042-6.

Fabricious-Moeller, J. (1922). Experimental studies of Hemorrhagic diathesis from X-ray sickness. Levin and Munksgaard: Copenhagen.

Faucher, C., Fortanier, C., Viens, P., Le Corroller, A.G., Chabannon, C., Camerlo, J., Novakovitch, G., Gastaut, J.A., Maraninchi, D., Moatti, J.P. & Blaise, D. (1998). Clinical and economic comparison of ienograstim-primed blood cells (BC) and bone marrow (BM) allogeneic transplantation. Bone Marrow Transplant, 21 SuppI 3, 892-8.

Fielding, A.K., Goldstone, A.M., Sullivan, A.M., Watts, M.J., Jamieson, E. & Linch, D.C. (1996). Peripheral blood stem cell support does not accelerate haemopoietic recovery from miniBEAM chemotherapy. Bone Marrow Transplant, 18, 507-11.

Firat, H., Giarratana, M.C., Kobari, L, Poloni, A., Bouchet, S., Labopin, M., Gorin, N.C. & Douay, L. (1998). Comparison of CD34+ bone marrow cells purified by immunomagnetic and immunoadsorption cell separation techniques. Bone Marrow Transplant, 21, 933-8.

Fleming, J.E., Jr., Haynesworth, S.E., Cassiede, P., Baber, M.A. & Caplan, A.I. (1998). Monoclonal antibody against adult marrow-derived mesenchymal stem cells recognizes developing vasculature in embryonic human skin. Dev Dyn, 212, 119-32.

Ford, C.E., Hamerton, J.L., Barnes, D.W.H. & Loutit, J.F. (1956). Cytologicai identification of radiation chimeras. Nature, 177, 239-47.

Fraser, J.K., Nicholls, J., Coffey, C., Lin, F.K. & Berridge, M.V. (1988). Down-modulation of high- affinity receptors for erythropoietin on murine erythrobiasts by interleukin 3. Exp Hematol, 1 6, 769-73.

Fruehauf, S., Haas, R., Conradt, C., Murea, S., Witt, B., Mohle, R. & Hunstein, W. (1995). Peripheral blood progenitor ceil (PBPC) counts during steady-state hematopoiesis allow to estimate the yield of mobilized PBPC after filgrastim (R-metHuG-CSF)-supported cytotoxic chemotherapy [see comments]. Blood, 85, 2619-26

Fukuda, M., Sasaki, H. & Fukuda, M.N. (1989). Erythropoietin metabolism and the influence of the carbohydrate structure. In Erythropoietin: from molecular structure to clinical application, Baldamus, P., Scigalla, L., Wieczorek, L. & Koch, K.M. (eds) pp. 76, 78-89. Karger: Basel.

Geissler, K., Peschel, C., Niederwieser, D., StrobI, H., Goldschmitt, J., Ohier, L., Bettelheim, P., Kahls, P., Huber, C., Lechner, K., Hocker, P. & Kolbe, K. (1996). Potentiation of granulocyte colony-stimulating factor-induced mobilization of circulating progenitor ceils by seven-day pretreatment with interleukin-3. Blood, 87, 2732-9 .

187 Gianni, A.M., Bregni, M., Siena, S., Villa, S., Sciorelli, G.A., Ravagnani, F., Pellegris, G. & Bonadonna, G. (1989). Rapid and complete hemopoietic reconstitution following combined transplantation of autologous blood and bone marrow cells. A changing role for high dose chemo-radiotherapy? Hematol Oncol, 7, 139-48.

Giralt, S., Estey, E., Albitar, M., van Besien, K., Rondon, G., Anderlini, P., O'Brien, S., Khouri, I., Gajewski, J., Mehra, R., Claxton, D., Andersson, B., Beran, M., Przepiorka, D., Koller, C., Kornblau, S., Korbling, M., Keating, M., Kantarjian, H. & Champlin, R. (1997). Engraftment of allogeneic hematopoietic progenitor cells with purine analog-containing chemotherapy: harnessing graft-versus-leukemia without myeloablative therapy. Blood, 89, 4531-6.

Gluckman, E. (1994). European organization for cord blood banking [see comments]. Blood Cells, 20, 601-8.

Gluckman, E. (1996). Bone marrow transplantation in children with hereditary disorders. Curr Opin Pediatr, 8, 42-4.

Gluckman, E., Rocha, V. & Chastang, C. (1998). European results of unrelated cord blood transplants. Eurocord group. Bone Marrow Transplant, 21 SuppI 3, 887-91.

Goodman, J.W. & Hodgson, G.8. (1962). Evidence of stem cells in the peripheral blood of mice. Blood, 19, 702-714.

Gorin, N.C. (1986). Collection, manipulation and freezing of haemopoietic stem cells. Clin Haematol, 15, 19-48.

Gratama, J.W., Kraan, J., Levering, W., Van Bockstaele, D.R., Rijkers, G.T. & Van der Schoot, C.E. (1997). Analysis of variation in results of CD34+ hematopoietic progenitor cell enumeration in a multicenter study. Cytometry, 30, 109-17.

Gratama, J.W., Orfao, A., Barnett, D., Brando, B., Huber, A., Janossy, G., Johnsen, H.E., Keeney, M., Marti, G.E., Preijers, P., Rothe, G., Serke, S., Sutherland, D.R., Van der Schoot, C.E., Schmitz, G. & Papa, S. (1998). Flow cytometric enumeration of CD34+ hematopoietic stem and progenitor cells. European Working Group on Clinical Cell Analysis. Cytometry, 34,128-42.

Gratwohl, A. & Schmitz, N. (1996). First international symposium on allogeneic peripheral blood precursor cell transplants. Bone Marrow Transplant, 17 SuppI 2, S I-3.

Gupta, K., Ramakrishnan, S., Browne, P.V., Solovey, A. & Hebbel, R.P. (1997). A novel technique for culture of human dermal microvascular endothelial cells under either serum-free or serum-supplemented conditions: isolation by panning and stimulation with vascular endothelial growth factor. Exp Cell Res, 230, 244-51.

Gupta, P., Blazar, B.R., Gupta, K. & Verfaillie, C.M. (1998). Human CD34(+) bone marrow cells regulate stromal production of interleukin-6 and granulocyte colony-stimulating factor and increase the colony-stimulating activity of stroma. Blood, 91, 3724-33.

Gupta, P., McCarthy, J.B. & Verfaillie, C.M. (1996). Stromal fibroblast heparan sulfate is required for cytokine-mediated ex vivo maintenance of human long-term culture-initiating cells. Blood, 8 7, 3229-36.

188 Haas, R., Gericke, G., Witt, B., Cayeux, S. & Hunstein, W. (1993). Increased serum levels of granulocyte colony-stimulating factor after autologous bone marrow or blood stem cell transplantation [see comments]. Exp Hematol, 21, 109-13.

Haas, R., Mohle, R., Fruhauf, S., Goldschmidt, H., Witt, B., Flentje, M., Wannenmacher, M. & Hunstein, W. (1994a). Patient characteristics associated with successful mobilizing and autografting of peripheral blood progenitor cells in malignant lymphoma. Blood, 83, 3787-94 .

Haas, R., Mohle, R., Murea, S., Goldschmidt, H., Pforsich, M., Witt, B. & Hunstein, W. (1994b). Characterization of peripheral blood progenitor cells mobilized by cytotoxic chemotherapy and recombinant human granulocyte colony- stimulating factor. J Hematother, 3, 323-30.

Haas, R., Witt, B., Mohle, R., Goldschmidt, H., Hohaus, S., Freuhauf, S., Wannenmacher, M. & Hunstein, W. (1995). Sustained long-term hematopoiesis after myeloablative therapy with peripheral blood progenitor cell support. Blood, 85, 3754-61 .

Hagglund, H., Ringden, O., Remberger, M., Lonnqvist, B., Sparrelid, E., Tammik, L. & Kumlien, G. (1998). Faster neutrophil and platelet engraftment, but no differences in acute GVHD or survival, using peripheral blood stem cells from related and unrelated donors, compared to bone marrow. Bone Marrow Transplant, 22,131-6.

Haioun, C., Lepage, E., Gisselbrecht, C., Bastion, Y., Coiffier, B., Brice, P., Bosly, A., Dupriez, B., Nouvel, C., Tilly, H., Lederlin, P., Biron, P., Briere, J., Gaulard, P. & Reyes, F. (1997). Benefit of autologous bone marrow transplantation over sequential chemotherapy in poor-risk aggressive non-Hodgkin's lymphoma: updated results of the prospective study LNH87-2. Groupe d'Etude des Lymphomes de l'Adulte. J Clin Oncol, 15, 1131-7.

Hansen, P.B., Knudsen, H., Gaarsdal, E., Jensen, L., Ralfkiaer, E. & Johnsen, H.E. (1995). Short-term in vivo priming of bone marrow haematopoiesis with rhG-CSF, rhGM-CSF or rhlL-3 before marrow harvest expands myelopoiesis but does not improve engraftment capability. Bone Marrow Transplant, 16, 373-9.

Haug, J.S., G. Todd, G., R. Bremer, R., R. Brown, R. & DiPersioHaug, J.F. (1998). Serum flt-3 ligand levels correlate with circulating CD34 levels. Blood, 92, 121a [abstract 490].

Haylock, D.N., To, L.B., Dowse, T.L., Juttner, C.A. & Simmons, P.J. (1992). Ex vivo expansion and maturation of peripheral blood CD34+ cells into the myeloid lineage. Blood, 80,1405-12.

Hershko, C., Gale, R.P., Ho, W.G. & Cline, M.J. (1979). Cure of aplastic anaemia in paroxysmal nocturnal haemoglobinuria by marrow transfusion from identical twin: Failure of peripheral- leucocyte transfusion to correct marrow aplasia. Lancet, 1, 945-7.

Hickman, R.O., Buckner, C.D., Clift, R.A., Sanders, J.E., Stewart, P. & Thomas, E.D. (1979). A modified right atrial catheter for access to the venous system in marrow transplant recipients. Surg Gynecol Obstet, 148, 871-5.

Ho, A.D., Maruyama, M., Maghazachi, A., Mason, J.R., Gluck, S. & Corringham, R.E. (1994). Soluble CD4, soluble CD8, soluble CD25, lymphopoieitic recovery, and endogenous cytokines after high-dose chemotherapy and blood stem cell transplantation. Blood, 84, 3550-7.

189 Hôglund, M., Smedmyr, B., Bengtsson, M., Totterman, T.H., Cour-Chabernaud, V., Yver, A. & Simonsson, B. (1997). Mobilization of CD34+ cells by glycosylated and nonglycosylated G-CSF in healthy volunteers-a comparative study. Eur J Haematol, 59,177-83.

Hôglund, M., Smedmyr, B., Simonsson, B., Totterman, I. & Bengtsson, M. (1996). Dose- dependent mobilisation of haematopoietic progenitor cells in healthy volunteers receiving glycosylated rHuG-CSF. Bone Marrow Transplant, 18,19-27.

Hohaus, S., Pforsich, M., Murea, S., Abdallah, A., Lin, Y.S., Funk, L., Voso, M.T., Kaui, S., Schmid, H., Wallwiener, D. & Haas, R. (1997). Immunomagnetic selection of CD34+ peripheral blood stem cells for autografting in patients with breast cancer. B rJ Haematol, 97, 881-8.

Holyoake, T.L., Alcorn, M.J., Richmond, L., Farrell, E., Pearson, 0., Green, R., Dunlop, D.J., Fitzsimons, E., Pragnell, I.B. & Franklin, I.M. (1997). CD34 positive PBPC expanded ex vivo may not provide durable engraftment following myeloablative chemoradiotherapy regimens. Bone Marrow Transplant, 19, 1095-101.

Huang, S. & Terstappen, L.W. (1994). Lymphoid and myeloid differentiation of single human CD34+, HLA-DR+, CD38- hematopoietic stem cells. Blood, 83, 1515-26.

Humeau, L., Bardin, F., Maroc, C., Alario, T., Galindo, R., Mannoni, P. & Chabannon, C. (1996). Phenotypic, molecular, and functional characterization of human peripheral blood CD34+/THY1 + cells. Blood, 87, 949-55.

Jagannath, S., Vesole, D.H., Glenn, L, Crowley, J. & Barlogie, B. (1992). Low-risk intensive therapy for muitiple myeloma with combined autologous bone marrow and blood stem cell support. Blood, BO, 1666-72.

Jakobson, L.O., Marks, E.K., Gaston, E.O. & Zirkle, R.E. (1949). Effect of spleen protection on mortality following X-irradiation. Journal of Laboratory and Clinical Medicine, 34,1538.

Jakobson, L.O., Simmons, E.L., Marks, E.K. & Eldridge, J.H. (1951). Recovery from radiation injury. Science, 113, 510-11.

Jandl, J.H. (1991). Blood:Pathophysiology. Blackwell Scientific Publications: Boston.

Janowska-Wieczorek, A., Chang, H., Cabuhat, M.L. & Marquez, L.A. (1997). CD34+ progenitor cells mobilized from peripheral blood but not CD34+ cells from normal marrow secrete matrix metalloproteinases (MMP-2 and MMP-9). Blood, 90, 95a.

Jaremo, P., Shaba, A. & Kutti, J. (1992). Some storage characteristics of buffy coats used for preparation of platelet concentrates. Ann Hematol, 65, 269-73.

Jones, H.M., Jones, S.A., Watts, M.J., Khwaja, A., Mills, W., Fielding, A., Goldstone, A.H. & Linch, D.C. (1994). Development of a simplified single-apheresis approach for peripheral-blood progenitor-cell transplantation in previously treated patients with lymphoma. J-Clin-Oncol, 1 2, 1693-702 .

Jones, H.M., Picozzi, V.J. & Price, H.T. (1996). Peripheral blood progenitor harvest following mobilization using G-CSF: experience in collections from 99 normal subjects. Blood, 88, 679a (abstract 2705).

190 Jones, R.J., Wagner, J.E., Celano, P., Zicha, M.S. & Sharkis, S.J. (1990). Separation of pluripotent haematopoietic stem cells from spleen colony- forming cells [see comments]. Nature, 347, 188-9.

Juttner, C.A., To, L.B., Haylock, D.N., Branford, A. & Kimber, R.J. (1985). Circulating autologous stem cells collected in very early remission from acute non-lymphoblastic leukaemia produce prompt but incomplete haemopoietic reconstitution after high dose melphalan or supralethal chemoradiotherapy. Br J Haematol, 61, 739-45.

Kaizer, H., Stuart, R.K., Brookmeyer, R., Beschorner, W.E., Braine, H.G., Burns, W.H., Fuller, D.J., Korbling, M., Mangan, K.F., Saral, R. & et al. (1985). Autologous bone marrow transplantation in acute leukemia: a phase I study of in vitro treatment of marrow with 4- hydroperoxycyclophosphamide to purge tumor cells. Blood, 65, 1504-10.

Kawano, Y., Takaue, Y., Saito, S., Sato, J., Shimizu, T., Suzue, T., Hirao, A., Okamoto, Y., Abe, T., Watanabe, I . & et al. (1993). Granulocyte colony-stimulating factor (CSF), macrophage-CSF, granulocyte-macrophage CSF, interleukin-3, and interleukin-6 levels in sera from children undergoing blood stem cell autografts. Blood, 81, 856-60.

Kessinger, A., Armitage, J.O., Landmark, J.D., Smith, D.M. & Welsenburger, D.D. (1988). Autologous peripheral hematopoietic stem cell transplantation restores hematopoietic function following marrow ablative therapy. Blood, 71, 723-7.

Ketterer, N., Salles, G., Moullet, I., Dumontet, C., ElJaafari-Corbin, A., Tremisi, P., Thieblemont, C., Durand, B., Neidhardt-Berard, E.M., Samaha, H., Rigal, D. & Coiffier, B. (1998). Factors associated with successful mobilization of peripheral blood progenitor cells in 200 patients with lymphoid malignancies. Br J Haematol, 103, 235-42.

Khwaja, A., Carver, J., Jones, H.M., Paterson, D. & Linch, D.C. (1993a). Expression and dynamic modulation of the human granulocyte colony- stimulating factor receptor in immature and differentiated myeloid cells. Br J Haematol, 85, 254-9.

Khwaja, A., Mills, W., Leveridge, K., Goldstone, A.H. & Linch, D.C. (1993b). Efficacy of delayed granulocyte colony-stimulating factor after autologous BMT. Bone Marrow Transplant, 11, 479- 82.

Kiriyama, R., Chichibu, K., Matsuno, T. & Ohsawa, N. (1993). Sensitive chemiluminescent immunoassay for human granulocyte colony- stimulating factor (G-CSF) in clinical applications. Clin Chim Acta, 220, 201-9.

Kishita, M., Motojima, H., Oh-eda, M., Kojima, T., Kamachi, S., Fukushima, M. & Ono, M. (1992). Stability of granulocyte colony stimulating factor (G-CSF) in serum. The clinical report, 26, 221.

Klumpp, T.R., Mangan, K.F., Goldberg, S.L., Pearl man, E.S. & Macdonald, J.S. (1995). Granulocyte colony-stimulating factor accelerates neutrophil engraftment following peripheral- blood stem-cell transplantation: a prospective, randomized trial. J Clin Oncol, 13,1323-7.

Knoell, D.L. & Yiu, I.M. (1998). Human gene therapy for hereditary diseases: a review of trials. Am J Health Syst Pharm, 55, 899-904.

Kobayashi, N., Kasai, M., Masauzi, N., Ogasawara, M., Kiyama, Y., Naohara, T., Higa, T., Hashino, S., Tanaka, J. & Imamura, M. (1995). Comparison of peripheral blood stem cells mobilized with

191 granulocyte colony-stimulating factor with or without prior standard-dose chemotherapy in patients with malignancy. Jpn-J-Clin-Oncol, 25, 250-7 issn: 0368-2811.

Korbling, M., Dorken, B., Ho, A.D., Pezzutto, A., Hunstein, W. & Fliedner, T.M. (1986). Autologous transplantation of blood-derived hemopoietic stem cells after myeloablative therapy in a patient with Burkitt's lymphoma. Blood, 67, 529-32.

Kotasek, D., Shepherd, K.M., Sage, R.E., Dale, B.M., Norman, J.E., Charles, P., Gregg, A., Pillow, A. & Bolton, A. (1992). Factors affecting blood stem cell collections following high-dose cyclophosphamide mobilization in lymphoma, myeloma and solid tumors. Bone Marrow Transplant, 9, 11 -7.

Kubota, N., Orita, T., Hattori, K., Oh-eda, M., Ochi, N. & Yamazaki, T. (1990). Structural characterization of natural and recombinant human granulocyte colony-stimulating factors. J Biochem (Tokyo), 107, 486-92.

Kuwabara, T., Kobayashi, S. & Sugiyama, Y. (1996). Kinetic analysis of receptor-mediated endocytosis of G-CSF derivative, nartograstim, in rat bone marrow cells. Am J Physiol, 271, E73- 84.

Lemoli, R.M., Fortune, A., Raspadori, □., Ventura, M.A., Martlnelll, G., Gozzetti, A., Leopardi, G., Ratta, M., Cavo, M. &Tura, S. (1997). Selection and transplantation of autologous hematopoietic CD34+ cells for patients with multiple myeloma. Leuk Lymphoma, 26 Supp11, 1-11.

Li, C.Y., Lam, K.W. & Yam, L.T. (1973). Esterases in human leukocytes. J HIstochem Cytochem, 21, 1- 12.

Linch, D.C. & Goldstone, A.H. (1984). Autologous bone marrow transplantation. B rJ Haematol, 58, 1-7.

Linch, D.C., Milligan, D.W., Winfield, D.A., Kelsey, S.M., Johnson, S.A., Littlewood, T.J., Smith, G.M., Hutchinson, R.M., Goldstone, A.H., Fielding, A.K. & Vaughan Hudson, G. (1997). G-CSF after peripheral blood stem cell transplantation In lymphoma patients significantly accelerated neutrophil recovery and shortened time in hospital: results of a randomized BNLI trial [see comments]. Br J Haematol, 99, 933-8.

Linch, D.C. & Watts, M.J. (1996). Standards for PBSC mobilisation and collection. Ann Oncol, 7 , 9-10.

Linch, D.C., Winfield, D., Goldstone, A.H., Moir, D., Hancock, B., McMillan, A., Chopra, R., Milligan, D. & Hudson, G.V. (1993). Dose Intensification with autologous bone-marrow transplantation in relapsed and resistant Hodgkin's disease: results of a BNLI randomised trial. Lancet, 341, 1051-4 .

Liu, F., Betsuyaku, T., Poursine-Laurent, J., Senior, R.M. & Link, D.C. (1997a). G-CSF receptor- dependent signals generated by hematopoietic cells are required for the mobilization of murine hematopoietic progenitors by cyclophosphamide. Blood, 90, abstract 4319.

Liu, F., Poursine-Laurent, J. & Link, D.C. (1997b). The granulocyte colony-stimulating factor receptor is required for the mobilization of murine hematopoietic progenitors into peripheral blood by cyclophosphamide or interleukin-8 but not flt-3 ligand. Blood, 90, 2522-8.

192 Lobo, F., Kessinger, A., Landmark, J.D., Smith, D.M., Weisenburger, D.D., Wigton, R.S. & Armitage, J.O. (1991). Addition of peripheral blood stem cells collected without mobilization techniques to transplanted autologous bone marrow did not hasten marrow recovery following myeloablative therapy. Bone Marrow Transplant, 8, 389-92.

Long, M.W. (1992). Blood cell cytoadhesion molecules. Exp Hematol, 20, 288-301.

Lord, B.I., Woolford, L.B., Wood, L.M., CzaplewskI, L.G., McCourt, M., Hunter, M.G. & Edwards, R.M. (1995). Mobilization of early hematopoietic progenitor cells with BB-10010; a genetically engineered variant of human macrophage inflammatory proteln-1 alpha. Blood, 85, 3412-5 .

Lorenz, E., Uphoff, D.E., Reid, T.R. & Shelton, E. (1951). Modification ofacute irradiation injury in mice and guinea pigs by bone marrow injection. Radiology, 58, 863-77.

Lotze, M.T., Shurin, M., Davis, I., Amoscato, A. & Storkus, W.J. (1997). Dendritic cell based therapy of cancer. Adv Exp Med Biol, 417, 551-69.

Lowenthal, R.M., Sullivan, S.A., Parker, N. & Marsden, K.A. (1996). G-CSF-primed bone marrow cells for autologous transplantation [letter; comment]. Lancet, 347, 1125.

Lucarelll, G., GalimbertI, M., Polchi, P., Angelucci, E., Baronciani, D., Giardini, C., PolitI, P., DurazzI, S.M., Muretto, P. & Albertini, F. (1990). Bone marrow transplantation in patients with thalassemia. N Engl J Med, 322, 417-21.

Lyman, S.D. (1998). Biologic effects and potential clinical applications of Flt3 ligand. Curr Opin Hematol, 5, 192-6.

Lynch, D.H. (1998). Induction of dendritic cells (DC) by Flt3 Ligand (FL) promotes the generation of tumor-specific immune responses in vivo. Grit Rev Immunol, 18, 99-107.

Mackinnon, S., Papadopoulos, E.B., Carabasi, M.H., Reich, L., Collins, N.H. & O'Reilly, R.J. (1995). Adoptive immunotherapy using donor leukocytes following bone marrow transplantation for chronic myeloid leukemia: is T cell dose Important in determining biological response? Bone Marrow Transplant, 15, 591-4.

Majumdar, M.K., Thiede, M.A., Mosca, J.D., Moorman, M. & Gerson, S.L. (1998). Phenotypic and functional comparison of cultures of marrow-derived mesenchymal stem cells (MSCs) and stromal cells. J Cell Physiol, 176, 57-66.

Makino, S., Harada, M., Akashi, K., TanlguchI, S., Shibuya, T., Inaba, S. & Niho, Y. (1991). A simplified method for cryopreservation of peripheral blood stem cells at -80 degrees C without rate-controlled freezing [see comments]. Bone Marrow Transplant, 8, 239-44.

Maraskovsky, E., Roux, E., Teepe, M., Braddy, S., Hoek, J., Lebsack, M. & McKenna, H.J. (1997). Ftl-3 ligand increases peripheral blood dendritic cells In healthy volunteers. Blood, 9 0 , 581a {abstract 2585}.

Marolleau, J.P., Brice, P., Dal Cortivo, L., Fermand, J.P., Robert, I., Miclea, J.M., Chaibi, P., Gisselbrecht, C., Mills, B., Kunkel, L., Oldham, F. & Benbunan, M. (1996). CD34+ selection by Immunomagnetic selection (lsolex®300) for patients with malignancies. Blood, 88, 110a (abstract 427).

193 Mathé, G., Jammet, H., Pendic, B. & al, e. (1959). Transfusions and grafts of homologous bone marrow In humans accidentally Irradiated to high doses. Revue Française d'Etudes Cliniques et Biologiques, 4, 226-38.

Mauch, P. & Heilman, S. (1989). Loss of hematopoietic stem cell self-renewal after bone marrow transplantation. Blood, 74, 872-5.

McFarlane, W., Granville, N.B. & Dameshek, W. (1959). Autologous bone marrow Infusion as an adjunct In therapy of malignant disease. Blood, 14, 503-521.

McNIece, I., Briddell, R., Stoney, G., Kern, B., Zllm, K., Recktenwald, D. & Mlltenyl, S. (1997). Large-scale Isolation of CD34+ cells using the Amgen cell selection device results In high levels of purity and recovery. J Hematother, 6, 5-11.

Mentzer, S.J., Guy re, P.M., Burakoff, S.J. & Faller, D.V. (1986). Spontaneous aggregation as a mechanism for human monocyte purification. Ceil Immunol, 101, 312-9.

Metcalf, D. (1990). The colony stimulating factors. Discovery, development, and clinical applications. Cancer, 65, 2185-95.

MIelcarek, M., Roecklein, B.A. & Torok-Storb, B. (1996). CD14+ cells In granulocyte colony- stimulating factor (G-CSF)-moblllzed peripheral blood mononuclear cells Induce secretion of lnterleukln-6 and G-CSF by marrow stroma. Blood, 87, 574-80.

MIelcarek, M. & Torok-Storb, B. (1997). Phenotype and engraftment potential of cytoklne- moblllzed peripheral blood mononuclear cells. Curr Opin Hematol, 4, 176-82.

Miller, C.L. & Eaves, C.J. (1997). Expansion In vitro of adult murine hematopoietic stem cells with transplantable lympho-myelold reconstituting ability. Proc Natl Acad Sol U S A, 94,13648-53.

Mohle, R., Pforsich, M., Fruehauf, S., Witt, B., Kramer, A. & Haas, R. (1994). Filgrastim post­ chemotherapy mobilizes more CD34+ cells with a different antigenic profile compared with use during steady-state hematopoiesis [published erratum appears In Bone Marrow Transplant 1995 Apr;15(4):655] [see comments]. Bone-Marrow-Transplant, 14, 827-32 .

Moore, M.A. (1992). Does stem cell exhaustion result from combining hematopoietic growth factors with chemotherapy? If so, how do we prevent It? [editorial; comment]. Blood, 80, 3-7.

Moore, M.A. (1996). Overview of Haemopoiesis and Haemopoietic reconstruction. Cell Therapy. Cambridge University Press: Cambridge.

Morse, E.E., Tuck, D., Ascensao, J., Sorba, S. & Dalnlak, N. (1993). Factors affecting recovery after peripheral blood stem cell transplantation. Ann Clin Lab Sol, 23, 89-96.

Morton, J., Morton, A., Bird, R., Hutchins, C. & Durrant, S. (1997). Predictors for optimal mobilization and subsequent engraftment of peripheral blood progenitor cells following Intermediate dose cyclophosphamide and G-CSF. Leuk Res, 21, 21-7.

Muench, M.O., FIrpo, M.T. & Moore, M.A. (1993). Bone marrow transplantation with lnterleukln-1 plus kit-llgand ex vivo expanded bone marrow accelerates hematopoietic reconstitution In mice without the loss of stem cell lineage and proliferative potential. Blood, 81, 3463-73.

194 Murea, S., Goldschmidt, H., Hahn, U., Pforsich, M., Moos, M. & Haas, R. (1996). Successful collection and transplantation of peripheral blood stem cells in cancer patients using large-volume leukaphereses. J Clin Apheresis, 11, 185-94.

Murray, L, Chen, B., Galy, A., Chen, S., Tushinski, R., Uchida, N., Negrin, R., Tricot, G., Jagannath, S., Vesole, D. & et al. (1995). Enrichment of human hematopoietic stem cell activity in the CD34+Thy- 1+Lin- subpopulation from mobilized peripheral blood. Blood, 85, 368-78.

Nachbaur, D., Kropshofer, G., Feichtinger, H., Allerberger, F. & Niederwieser, D. (1997). Cryptosporidiosis after CD34-selected autologous peripheral blood stem cell transplantation (PBSCT). Treatment with paromomycin, azithromycin and recombinant human interleukin-2. Bone Marrow Transplant, 19, 1261-3.

Nibley, W.E. & Spangrude, G.J. (1998). Primitive stem cells alone mediate rapid marrow recovery and multilineage engraftment after transplantation. Bone Marrow Transplant, 21, 345-54.

Nissen, C., Dalle Cabonare, V. & Moser, Y. (1994). In v/fro comparison of the biological potency of glycosylated versus non-glycosylated rG-CSF. Drug Investigation, 7, 346-352.

Novotny, J., Kadar, J., Hertenstein, B., Schroder, □., Petersen, D., Arseniev, L, Stangel, W. & Ganser, A. (1998). Sustained decrease of peripheral lymphocytes after allogeneic blood stem cell aphereses. Br J Haematol, 100, 695-7.

Ogawa, M. (1993). Differentiation and proliferation of hematopoietic stem cells. Blood, 81, 2844- 53.

Ossenkoppele, G.J., Jonkhoff, A.R., Huijgens, P.C., Nauta, J.J., van der Hem, K.G., Drager, A.M. & Langenhuijsen, M.M. (1994). Peripheral blood progenitors mobilised by G-CSF (filgrastim) and reinfused as unprocessed autologous whole blood shorten the pancytopénie period following high-dose melphalan in multiple myeloma. Bone Marrow Transplant, 13, 37-41.

Papayannopoulou, T. & Nakamoto, B. (1993). Peripheralization of hemopoietic progenitors in primates treated with anti-VL A4 integrin. Proc Natl Acad SclUSA, 90, 9374-8.

Papayannopoulou, T., Priestley, G.V. & Nakamoto, B. (1998). Anti-VLA4/VCAM-1-induced mobilization requires cooperative signaling through the kit/mkit ligand pathway. Blood, 91, 2231- 9.

Passes Coelho, J.L, Braine, H.G., Davis, J.M., Huelskamp, A.M., Schepers, K.G., Ohiy, K., Clarke, B., Wright, S.K., Noga, S.J., Davidson, N.E. & et al. (1995). Predictive factors for peripheral-blood progenitor-cell collections using a single large-volume leukapheresis after cyclophosphamide and granulocyte-macrophage colony-stimulating factor mobilization. J-Clin- Oncol, 13, 705-14 .

Pedrazzoli, P., Gibelli, N., Pavesi, L, Preti, P., Piolini, M., Bertolini, F. & Robustelli della Cuna, G. (1996). Effects of glycosylated and non-glycosylated G-CSFs, alone and in combination with other cytokines, on the growth of human progenitor cells. Anticancer Res, 16, 1781-5.

Pegg, D.E. (1966). Allogeneic bone marrow transplantation In man. Lloyd-Luke Medical Books: London.

195 Perry, A.R., Watts, M.J., Peniket, A.J., Goldstone, A.H. & Linch, D.C. (1998). Progenitor cell yields are frequently poor in patients with histologically indolent lymphomas especially when mobilized within 6 months of previous chemotherapy. Bone Marrow Transplant, 21, 1201-5.

Peters, W.P., Rosner, G., Ross, M., Vredenburgh, J., Meisenberg, B., Gilbert, 0. & Kurtzberg, J. (1993). Comparative effects of granulocyte-macrophage colony-stimulating factor (GM-CSF) and granulocyte colony-stimulating factor (G-CSF) on priming peripheral blood progenitor cells for use with autologous bone marrow after high-dose chemotherapy. Blood, 81, 1709-19 .

Pettengell, R., Morgenstern, G.R., Well, P.J., Chang, J., Rowlands, M., Young, R., Radford, J.A., Scarffe, J.H., Testa, N.G. & Crowther, D. (1993). Peripheral blood progenitor cell transplantation in lymphoma and leukemia using a single apheresis. Blood, 82, 3770-7.

Philip, T., Guglielmi, C., Hagenbeek, A., Somers, R., Van der Lelie, H., Bron, D., Sonneveld, P., Gisselbrecht, C., Cahn, J.Y., Harousseau, J.L. & et al. (1995). Autologous bone marrow transplantation as compared with salvage chemotherapy in relapses of chemotherapy-sensitive non-Hodgkin's lymphoma [see comments]. N-Engl-J-Med, 333, 1540-5 .

Pizzey, A. Unpubll^d observations (1998).

Powles, R.L., Clink, H.M., Spence, D., Morgenstern, G., Watson, J.G., Selby, P.J., Woods, M., Barrett, A., Jameson, B., Sloane, J., Lawler, S.D., Kay, H.E., Lawson, D., McElwain, T.J. & Alexander, P. (1980). Cyclosporin A to prevent graft-versus-host disease in man after allogeneic bone-marrow transplantation. Lancet, 1, 327-9.

Poynton, C.H. (1988). T cell depletion in bone marrow transplantation. Bone Marrow Transplant, 3, 265-79.

Prince, H.M. (1998). Gene transfer: a review of methods and applications. Pathology, 30, 335- 47.

Pruijt, J.F., van Kooyk, Y., Figdor, C.G., Lindley, I.J., Willemze, R. & Fibbe, W.E. (1998). Anti- LFA-1 blocking antibodies prevent mobilization of hematopoietic progenitor cells Induced by interleukin-8. S/ood, 91, 4099-105.

Rapoport, A.P., Abboud, C.N. & DiPersio, J.F. (1992). Granulocyte-macrophage colony- stimulating factor (GM-CSF) and granulocyte colony-stimulating factor (G-CSF): receptor biology, signal transduction, and neutrophil activation. Blood Rev, 6, 43-57.

Rasko, J.E., Basser, R.L., Boyd, J., Mansfield, R., O'Malley, C.J., Hussein, S., Berndt, M.C., Clarke, K., O'Byrne, J., Sheridan, W.P., Grigg, A.P. & Begley, C.G. (1997). Multilineage mobilization of peripheral blood progenitor cells In humans following administration of PEG- rHuMGDF. Br J Haematol, 97, 871-80.

Reiffers, J., Bernard, P., David, B., Vezon, G., Sarrat, A., Marit, G., Moulinier, J. & Broustet, A. (1986). Successful autologous transplantation with peripheral blood hemopoietic cells in a patient with acute leukemia. Exp Hematol, 14, 312-5.

Richel, D., Johnsen, H., Canon, J., Schaafsma, M., Schenkeveld, C., Hansen, S., McNiece, I., Ewen, C., Davies, R., Miltenyi, S., Guillaume, I . & Symann, M. (1996). Highly purified CD34+ cells isolated with the Amgen cell selector provide rapid engraftment following high-dose chemotherapy in breast cancer. Blood, 8 8 ,110a (abstr.429).

196 Richman, C.M., Weiner, R.S. & Yankee, R.A. (1976). Increase in circulating stem cells following chemotherapy in man. Blood, 47, 1031-9.

Ringden, O., Hagglund, H., Runde, V., Basu, O., Kroschinsky, F., Stockschlader, M. & Potter, M.N. (1998). Faster engraftment of peripheral blood progenitor cells compared to bone marrow from unrelated donors. Bone Marrow Transplant, 21 SuppI 3, 881-4.

Roberts, A.W., DeLuca, E., Begley, C.G., Basser, R., Grigg, A.P. & Metcalf, D. (1995). Broad inter-individual variations in circulating progenitor cell numbers induced by granulocyte colony- stimulating factor therapy. Stem-Cells-Dayt, 13, 512-6 issn: 1066-5099.

Roberts, R., Gallagher, J., Spooncer, E., Allen, T.D., Bloomfield, F. & Dexter, T.M. (1988). Heparan sulphate bound growth factors: a mechanism for stromal cell mediated haemopoiesis. Nature, 332, 376-8.

Robertson, M.J., Soiffer, R.J., Freedman, A.S., Rabinowe, S.L., Anderson, K.C., Ervin, T.J., Murray, C., Dear, K., Griffin, J.D., Nadler, L.M. & et al. (1992). Human bone marrow depleted of CD33-positive cells mediates delayed but durable reconstitution of hematopoiesis: clinical trial of MY9 monoclonal antibody-purged autografts for the treatment of acute myeloid leukemia. Blood, 79, 2229-36 .

Rosenfeld, C.S., Shadduck, R.K., Zeigler, Z.R., Andrews, F. & Nemunaitis, J. (1995). Cyclophosphamide-mobilized peripherai blood stem cells in patient with lymphoid malignancies. Bone-Marrow-Transplant, 15, 433-8 .

Ross, A.A., Cooper, B.W., Lazarus, H.M., Mackay, W., Moss, T.J., Ciobanu, N., Tallman, M.S., Kennedy, M.J., Davidson, N.E., Sweet, D. & et al. (1993). Detection and viability of tumor cells in peripheral blood stem cell collections from breast cancer patients using immunocytochemical and clonogenic assay techniques [see comments]. Blood, 82, 2605-10 .

Rowley, S.D., Loken, M., Radich, J., Kunkle, L.A., Mills, B.J., Gooley, T., Holmberg, L., McSweeney, P., Beach, K., MacLeod, B., Appelbaum, F. & Bensinger, W.l. (1998). Isolation of CD34+ cells from blood stem cell components using the Baxter Isolex system. Bone Marrow Transplant, 2^, 1253-62.

Rowlings, P.A., Bayly, J.L., Rawling, C.M., Juttner, C.A. & To, L.B. (1992). A comparison of peripheral blood stem cell mobilisation after chemotherapy with cyclophosphamide as a single agent in doses of 4 g/m2 or 7 g/m2 in patients with advanced cancer. Aust-N-Z-J-Med, 22, 660-4

Ruoslahti, E. & Pierschbacher, M.D. (1987). New perspectives in cell adhesion: RGD and integrins. Science, 238, 491-7.

Rusten, L.S., Jacobsen, S.E., Kaalhus, O., Veiby, O.P., Funderud, S. & Smeland, E.B. (1994). Functional differences between CD38- and DR- subtractions of CD34+ bone marrow cells, e/ooof, 84, 1473-81.

Sallerfors, B. & Olofsson, T. (1991). Granulocyte-macrophage colony-stimulating factor (GM-CSF) and granulocyte colony-stimulating factor (G-CSF) in serum during induction treatment of acute leukaemia. Br J Haematol, 78, 343-51.

197 Schiller, G., Vescio, R., Freytes, G., Spitzer, G., Sahebi, F., Lee, M., Wu, C.H., Cao, J., Lee, J.C., Hong, C.H. & et al. (1995). Transplantation of CD34+ peripheral blood progenitor cells after high- dose chemotherapy for patients with advanced multiple myeloma. Blood, 86, 390-7.

Schmitz, N., Dreger, P., Suttorp, M., Rohwedder, E.B., Haferlach, T., Loffler, H., Hunter, A. & Russell, N.H. (1995). Primary transplantation of allogeneic peripheral blood progenitor cells mobilized by filgrastim (granulocyte colony-stimulating factor) [see comments]. Blood, 85, 1666- 72.

Schmitz, N., Linch, D.C., Dreger, P., Goldstone, A.H., Boogaerts, M.A., Ferrant, A., Demuynck, H.M., Link, H., Zander, A., Barge, A. & et al. (1996). Randomised trial of fllgrastlm-moblllsed peripheral blood progenitor cell transplantation versus autologous bone-marrow transplantation In lymphoma patients [see comments]. Lancet, 347, 353-7 .

Schwartzberg, L., Birch, R., Blanco, R., WIttlln, F., Muscato, J., Tauer, K., Hazelton, B. & West, W. (1993). Rapid and sustained hematopoietic reconstitution by peripheral blood stem cell Infusion alone following high-dose chemotherapy. Bone Marrow Transplant, 11, 369-74.

Schwella, N., Slegert, W., Beyer, J., Rick, O., Zingsem, J., Eckstein, R., Serke, S. & Huhn, D. (1995). Autografting with blood progenitor cells: predictive value of preapheresis blood cell counts on progenitor cell harvest and correlation of the relnfused cell dose with hematopoietic reconstitution. Ann-Hematol, 71, 227-34 Issn: 0939-5555.

Scott, M.A., Ager, S., Apperley, J.F., Jestlce, H.K., Bloxham, D.M., Boraks, P., Mahendra, P. & Marcus, R.E. (1995a). Peripheral blood progenitor cell harvesting In multiple myeloma and malignant lymphoma. Leuk Lymphoma, 19, 479-84.

Scott, M.A., Ager, S., Jestlce, H.K., Mahendra, P. & Marcus, R.E. (1995b). Failure to mobilise and harvest PBPC does not necessarily preclude the use of high-dose therapy and autologous stem cell rescue [letter]. Bone Marrow Transplant, 15, 487-8.

Sekhsarla, S., Flelsher, T.A., Vowells, S., Brown, M., Miller, J., Gordon, I., Blaese, R.M., Dunbar, G.E., Leltman, S. & Malech, H.L. (1996). Granulocyte colony-stimulating factor recruitment of GD34-I- progenitors to peripheral blood: Impaired mobilization In chronic granulomatous disease and adenosine deamlnase-deflclent severe combined Immunodeficiency disease patients. Blood, 88, 1104-12 .

Serke, S., Watts, M., Knudsen, L.M., Krelssig, G., Schneider, U., Schwella, N., Linch, D. & Johnsen, H.E. (1996). In-vltro clonogenlclty of mobilized peripheral blood GD34-expresslng cells: Inverse correlation to both relative and absolute numbers of GD34-expresslng cells. Br J Haematol, 95, 234-40.

Sharp, J.G., Kessinger, A., Clausen, S.R., Mann, S.L. & O'Kane-Murphy, B. (1998). Concurrent partial body radiation prevents cytokine mobilization of blood progenitor cells: an effect mediated by a circulating factor. J Hematother, 7, 343-9.

Shea, I.e ., Mason, J.R., Stornlolo, A.M., Newton, B., Breslln, M., Mullen, M., Ward, D.M., Miller, L., Christian, M. & Taetle, R. (1992). Sequential cycles of high-dose carboplatin administered with recombinant human granulocyte-macrophage colony-stimulating factor and repeated Infusions of autologous peripheral-blood progenitor cells: a novel and effective method for delivering multiple courses of dose- Intensive therapy. J Clin Oncol, 10, 464-73.

198 Sheridan, W.P., Begley, C.G., Juttner, C.A,, Szer, J., To, L.B., Maher, D., McGrath, K.M., Morstyn, G. & Fox, R.M. (1992). Effect of peripheral-blood progenitor cells mobilised by filgrastim (G- CSF) on platelet recovery after high-dose chemotherapy [see comments]. Lancet, 339, 640- 4.

Shpall, E.J., Jones, R.B., Bearman, S.I., Franklin, W.A., Archer, P.G., Curlel, T., Bitter, M., Claman, H.N., Stemmer, S.M., Purdy, M. & et al. (1994). Transplantation of enriched CD34- posltlve autologous marrow Into breast cancer patients following high-dose chemotherapy: Influence of CD34- positive peripheral-blood progenitors and growth factors on engraftment. J Clin Oncol, 12, 28-36.

SIczkowskI, M., Clarke, D. & Gordon, M.Y. (1992). Binding of primitive hematopoietic progenitor cells to marrow stromal cells Involves heparan sulfate. Blood, 80, 912-9.

Siena, S., Bregni, M., Brando, B., Belli, N., Ravagnani, F., Gandola, L, Stern, A.C., Lansdorp, P.M., Bonadonna, G. & Gianni, A.M. (1991). Flow cytometry for clinical estimation of circulating hematopoietic progenitors for autologous transplantation In cancer patients. Blood, 77, 400-9.

Simmons, P.J., Zannettino, A., Gronthos, S. & Leavesley, D. (1994). Potential adhesion mechanisms for localisation of haemopoietic progenitors to bone marrow stroma. Leuk Lymphoma, 12, 353-63.

Slavin, S., Nagler, A., Naparstek, E., Kapelushnik, Y., Aker, M., CIvldalll, G., VaradI, G., KIrschbaum, M., Ackerstein, A., Samuel, S., Amar, A., Brautbar, C., Ben-Tal, O., Eldor, A. & Or, R. (1998). Nonmyeloablatlve stem cell transplantation and cell therapy as an alternative to conventional bone marrow transplantation with lethal cytoreductlon for the treatment of malignant and nonmallgnant hematologic diseases. Blood, 91, 756-63.

Smith, T.J., Hlllner, B.E., Schmitz, N., Linch, D.C., Dreger, P., Goldstone, A.H., Boogaerts, M.A., Ferrant, A., Link, H., Zander, A., Yanovich, S., KItchIn, R. & Erder, M.H. (1997). Economic analysis of a randomized clinical trial to compare filgrastim- mobilized peripheral-blood progenitor­ cell transplantation and autologous bone marrow transplantation In patients with Hodgkin's and non-Hodgkln's lymphoma. J Clin Oncol, 15, 5-10.

SoclnskI, M.A., Cannlstra, S.A., Ellas, A., Antman, K.H., Schnlpper, L. & Griffin, J.D. (1988). Granulocyte-macrophage colony stimulating factor expands the circulating haemopoietic progenitor cell compartment In man. Lancet, 1,1194-8.

Soiffer, R.J., Murray, C., Cochran, K., Cameron, C., Wang, E., Schow, P.W., Daley, J.F. & RItz, J. (1992). Clinical and Immunologic effects of prolonged Infusion of low-dose recombinant lnterleukln-2 after autologous and T-cell-depleted allogeneic bone marrow transplantation. Blood, 79, 517-26.

Sosman, J.A., Stiff, P.J., Bayer, R.A., Pellska, J., Peace, D.J., Loutfl, S., Stock, W., Oldenburg, D., Unverzagt, K., Bender, J. & et al. (1995). A phase I trial of Interleukin 3 (IL-3) pre-bone marrow harvest with granulocyte-macrophage colony-stimulating factor (GM-CSF) post-stem cell Infusion In patients with solid tumors receiving high-dose combination chemotherapy. Bone Marrow Transplant, 16, 655-61.

Souza, L.M., Boone, T.C., Gabrllove, J., Lai, P.M., Zsebo, K.M., Murdock, D.C., Chazin, V.R., BruszewskI, J., Lu, H., Chen, K.K. & et al. (1986). Recombinant human granulocyte colony- stimulating factor: effects on normal and leukemic myeloid cells. Science, 232, 61-5.

199 Sovalat, H., Wunder, E., Tienhaara, A., Olofsson, T., Fritsch, G., Silvestri, F. & Serke, S. (1993). Commentary: prospects for standardization of stem ceii determination within Europe. J Hematother, 2, 293-6.

Spitzer, G., Verma, D.S., Fisher, R., Zander, A., Vellekoop, L, Litam, J., McCredie, K.B. & Dicke, K.A. (1980). The myeloid progenitor cell-its value in predicting hematopoietic recovery after autologous bone marrow transplantation. Blood, 55, 317-23.

Srour, E.F., Zanjani, E.D., Cornetta, K., Traycoff, C.M., Flake, A.W., Hedrick, M., Brandt, J.E., Leemhuis, I. & Hoffman, R. (1993). Persistence of human multilineage, self-renewing lymphohematopoietic stem cells in chimeric sheep. Blood, 82, 3333-42.

Steinman, R.M. (1996). Dendritic cells and immune-based therapies. Exp Hematol, 24, 859-62.

Storb, r., Epstein, R.B., Ragde, H. & Thomas, E.D. (1967). Marrow engraftment by allogeneic leukocytes in lethaily irradiated dogs. Blood, 30, 805-811.

Stroncek, D.F., Clay, M.E., Petzoldt, M.L, Smith, J., Jaszcz, W., Oldham, F.B. & McCullough, J. (1996). Treatment of normal individuals with granulocyte-colony-stimulating factor: donor experiences and the effects on peripheral blood CD34+ ceii counts and on the collection of peripheral blood stem cells [see comments]. Transfusion, 36, 601 -10.

Stroncek, D.F., Holland, P.V., Bartch, G., Bixby, T., Simmons, R.G., Antin, J.H., Anderson, K.C., Ash, R.C., Bolwell, B.J., Hansen, J.A. & et al. (1993). Experiences of the first 493 unrelated marrow donors in the National Marrow Donor Program. Blood, 81,1940-6.

Stute, N., Santana, V.M., Rodman, J.H., Schell, M.J., Ihie, J.N. & Evans, W.E. (1992). Pharmacokinetics of subcutaneous recombinant human granulocyte colony- stimulating factor in children. Blood, 79, 2849-54.

Sutherland, H.J., Lansdorp, P.M., Henkeiman, D.H., Eaves, A.C. & Eaves, C.J. (1990). Functional characterization of individual human hematopoietic stem cells cultured at limiting dilution on supportive marrow stromal layers. Proa Natl Acad SolUSA, 87, 3584-8.

Tavassoli, M. (1980). Bone marrow: The seedbed of the blood. In Blood pure and eloquent, Wintrobe, M.M. (ed) pp. 57-79. McGraw-Hill: New York.

Tavassoli, M. & Hardy, C.L. (1990). Molecular basis of homing of intravenously transplanted stem cells to the marrow. Blood, 76, 1059-70.

Tepler, I., Cannistra, S.A., Frei, E.d., Gonin, R., Anderson, K.C., Demetri, G., Niloff, J., Goodman, H., Muntz, H., Muto, M. & et al. (1993). Use of peripheral-biood progenitor cells abrogates the myelotoxicity of repetitive outpatient high-dose carbopiatin and cyclophosphamide chemotherapy. J Clin Oncol, 11, 1583-91.

Terasaki, P.I. & McLelland, J.D. (1964). Microdroplet assay of human cytotxins. Nature, 204.

Terstappen, L.W. & Huang, S. (1994). Analysis of bone marrow stem cell. Blood Cells, 20, 45- 61.

200 Testa, U., Martucci, R., Rutella, S., Scambia, G., Sica, S., Benedetti Panici, P., Pierelli, L., Menichella, G., Leone, G., Mancuso, S. & et al. (1994). Autologous stem cell transplantation: release of early and late acting growth factors relates with hematopoietic ablation and recovery. Blood, 84, 3532-9.

Thomas, E.D. & Ferrebee, J.W. (1960). Irradiation and marrow transplantation: Studies in Cooperstown. Lancet, 1289-90.

To, L.B., Haylock, D.N., Simmons, P.J. & Juttner, G.A. (1997). The biology and clinical uses of blood stem cells. Blood, 89, 2233-58.

Tricot, G., Gazitt, Y., Leemhuis, T., Jagannath, S., Desikan, K.R., Siegel, □., Passas, A., Tindle, S., Nelson, J., Juttner, C., Tsukamoto, A., Hallagan, J., Atkinson, K., Reading, C., Hoffman, R. & Barlogie, B. (1998). Collection, tumor contamination, and engraftment kinetics of highly purified hematopoietic progenitor cells to support high dose therapy in multiple myeloma. Blood, 9 1 , 4489-95.

Tricot, G., Jagannath, S., Vesole, D., Nelson, J., Tindle, S., Miller, L., Cheson, B., Crowley, J. & Barlogie, B. (1995). Peripheral blood stem cell transplants for multiple myeloma: identification of favorable variables for rapid engraftment in 225 patients. Blood, 85, 588-96.

Uchida, N., Aguila, H.L., Fleming, W.H., Jerabek, L. & Weissman, I.L. (1994). Rapid and sustained hematopoietic recovery in lethaily irradiated mice transplanted with purified Thy-I.llo Lin-Sca-1+ hematopoietic stem cells. Blood, 83, 3758-79.

Urbano-lspizua, A., Solano, C., Brunet, S., de la Rubia, J., Odriozola, J., Zuazu, J., Figuera, A., Caballero, □., Martinez, C., Garcia, J., Sanz, G., Torrabadella, M., Alegre, A., Perez-Oteiza, J., Jurado, M., Oyonarte, S., Sierra, J., Garcia-Conde, J. & Rozman, C. (1998). Allogeneic transplantation of selected CD34+ cells from peripheral blood: experience of 62 cases using immunoadsorption or immunomagnetic technique. Spanish Group of Allo-PBT. Bone Marrow Transplant, 22, 519-25.

Urbano-lspizua, A., Solano, C., Brunet, S., Hernandez, P., Sanz, G., Alegre, A., Petit, J., Besalduch, J., Vivancos, P., Diaz, M.A., Moraleda, J.M., Carreras, E., Ojeda, E., de la Rubla, J., Benet, I., Domingo-Albos, A., Garcia-Conde, J. & Rozman, C. (1996). Allogeneic peripheral blood progenitor cell transplantation: analysis of short-term engraftment and acute GVHD incidence in 33 cases, alio- PBPCT Spanish Group. Bone Marrow Transplant, 18, 35-40.

Van Hoef, M.E. (1996). Guidelines for efficient peripheral blood progenitor cell collection. Neth J Med, 48, 29-37. van Os, R., Robinson, S., Sheridan, T., Mislow, J.M., Dawes, D. & Mauch, P.M. (1998). Granulocyte colony-stimulating factor enhances bone marrow stem cell damage caused by repeated administration of cytotoxic agents. Blood, 92, 1950-6.

Velasquez, W.S., McLaughlin, P., Tucker, S., Hagemeister, F.B., Swan, P., Rodriguez, M.A., Romaguera, J., Rubenstein, E. & Cabanillas, F. (1994). ESHAP-an effective chemotherapy regimen in refractory and relapsing lymphoma: a 4-year follow-up study [see comments]. J Clin Oncol, 12, 1169-76.

201 Vermeulen, M., Le Pesteur, F., Gagnerault, M.C., Mary, J.Y., Sainteny, R & Lepault, F. (1998). Role of adhesion molecules in the homing and mobilization of murine hematopoietic stem and progenitor cells. Blood, 92, 894-900.

Wagner, J.E., Collins, D., Fuller, S., Schain, L.R., Berson, A.E., Almici, 0., Hall, M.A., Chen, K.E., Okarma, T.B. & LebkowskI, U.S. (1995). Isolation of small, primitive human hematopoietic stem cells: distribution of cell surface cytokine receptors and growth In SCID-Hu mice. Blood, 86, 512- 23.

Waller, C.F., Bertz, H., Wenger, M.K., Fetscher, S., Hardung, M., Engelhardt, M., Behringer, D., Lange, W., Mertelsmann, R. & Finke, J. (1996). Mobilization of peripheral blood progenitor cells for allogeneic transplantation: efficacy and toxicity of a high-dose rhG-CSF regimen. Bone Marrow Transplant, 18, 279-83.

Wang, J.C., Lapidot, T., Cashman, J.D., Doedens, M., Addy, L., Sutherland, D.R., Nayar, R., Laraya, P., Minden, M., Keating, A., Eaves, A.C., Eaves, C.J. & Dick, J.E. (1998). High level engraftment of NOD/SCID mice by primitive normal and leukemic hematopoietic cells from patients with chronic myeloid leukemia in chronic phase. Blood, 91, 2406-14.

Watanabe, T., Dave, B., Heimann, D.G., Jackson, J.D., Kessinger, A. & Talmadge, J.E. (1998). Cell adhesion molecule expression on CD34+ cells in grafts and time to myeloid and platelet recovery after autologous stem cell transplantation. Exp Hematol, 26,10-8.

Watari, K., Asano, S., Shirafuji, N., Kodo, H., Ozawa, K., Takaku, F. & Kamachi, S. (1989). Serum granulocyte colony-stimulating factor levels in healthy volunteers and patients with various disorders as estimated by enzyme immunoassay. Blood, 73, 117-22.

Watts, M.J., Addison, I., Ings, S.J., Long, S.G., Hartley, S., Warrington, S., Boyce, M. & Linch, D.C. (1998a). Optimal timing for collection of PBPC after glycosylated G-CSF administration. Bone Marrow Transplant, 21, 365-8.

Watts, M.J., Addison, I., Long, S.G., Hartley, S., Warrington, S., Boyce, M. & Linch, D.C. (1997a). Crossover study of the haematological effects and pharmacokinetics of glycosylated and non­ glycosylated G-CSF in healthy volunteers. B rJ Haematol, 98, 474-9.

Watts, M.J., Addison, I., Sullivan, A.M., Hunter, M., Czaplewski, L., Wood, M. & Linch, D.C. (1996a). Effects of a single dose of BB-10010 on progenitor mobilization after cyclophosphamide and G-CSF. B rJ Haematol, 96, 277a (abstract 1049).

Watts, M.J., Jones, H.M., Sullivan, A.M., Langabeer, S.E., Jamieson, E., Fielding, A., Williams, C., Berenson, R.J., Goldstone, A.H. & Linch, D.C. (1995a). Accessory cells do not contribute to G-CSF or IL-6 production nor to rapid haematological recovery following peripheral blood stem cell transplantation. B rJ Haematol, 91, 767-72.

Watts, M.J. & Linch, D.C. (1997). Peripheral blood stem cell transplantation. Vox Sang, 73, 135- 42.

Watts, M.J., Lverett, D., Sullivan, A.M., Peniket, A.J., Perry, A.R., MacMillan, A., Goldstone, A.H. & Linch, D.C. (1996b). A comparison of ESHAP+G-CSF versus cyclophosphamide 1.5g/m2+G- CSF for PBSC mobilisation in pre-treated lymphoma patients. Blood, 88, 396a.

202 Watts, M.J., Sullivan, A.M., Ings, S.J., Leverett, D., Peniket, A.J., Perry, A.R., Williams, C.D., Devereux, S., Goldstone, A.H. & Linch, D.C. (1997b). Evaluation of clinical scale CD34+ cell purification: experience of 71 immunoaffinity column procedures. Bone Marrow Transplant, 20, 157-62.

Watts, M.J., Sullivan, A.M., Jaimeson, E., Chavda, N., Fielding, A., Goidstone, A.H. & Linch, D.C. (1995b). A single apheresis algorithm for PBSCT. Br J Haematol, 89, 22a (abstract 79).

Watts, M.J., Suliivan, A.M., Jamieson, E., Pearce, R., Fielding, A., Devereux, S., Goldstone, A.H. & Linch, D.C. (1997c). Progenitor-cell mobilization after low-dose cyclophosphamide and granulocyte colony-stimulating factor: an analysis of progenitor-cell quantity and quality and factors predicting for these parameters in 101 pretreated patients with malignant lymphoma. J Clin Oncol, 15, 535-46.

Watts, M.J., Sullivan, A.M., Leverett, D., Peniket, A.J., Perry, A.R., Williams, C.D., Devereux, S., Goldstone, A.H. & Linch, D.C. (1998b). Back-up bone marrow is frequently ineffective in patients with poor peripheral-blood stem-cell mobilization. J Clin Oncol, 16, 1554-60.

Weaver, C.H., Hazelton, B., Birch, R., Palmer, P., Allen, C., Schwartzberg, L. & West, W. (1995). An analysis of engraftment kinetics as a function of the CD34 content of peripheral blood progenitor cell collections in 692 patients after the administration of myeloablative chemotherapy. Blood, 86, 3961-9.

Weaver, C.H., Potz, J., Redmond, J., Tauer, K., Schwartzberg, L.S., Kaywin, P., Drapkin, R., Grant, B., Linger, P., Allen, C., Longin, K., Zhen, B., Hazelton, B. & Buckner, C.D. (1997). Engraftment and outcomes of patients receiving myeloablative therapy followed by autologous peripheral blood stem cells with alow CD34+ cell content. Bone Marrow Transplant, 19, 1103- 10.

Weaver, C.H., Zhen, B. & Buckner, C.D. (1998). Treatment of patients with malignant lymphoma with Mini-BEAM reduces the yield of CD34+ peripheral blood stem cells [letter]. Bone Marrow Transplant, 21,1169-70.

Wiesneth, M., Schreiner, T., Friedrich, W., Bunjes, D., Duncker, C., Krug, E., Maccari, B., Muller, S., Nowak, S. & Kubanek, B. (1998). Mobilization and collection of allogeneic peripherai blood progenitor cells for transplantation. Bone Marrow Transplant, 21 SuppI 3, S21-4.

Williams, C.D., Linch, D.C., Watts, M.J. & Thomas, M.S. (1997). Characterization of cell cycle status and E2F complexes in mobiiized CD34+ cells before and after cytokine stimulation. Blood, 90, 194-203.

Williams, S.F., Lee, W.J., Bender, J.G., Zimmerman, T., Swinney, P., Blake, M., Carreon, J., Schilling, M., Smith, S., Wiiliams, D.E., Oldham, F. & Van Epps, D. (1996). Selection and expansion of peripheral blood CD34+ cells in autologous stem cell transplantation for breast cancer. Blood, 87, 1687-91.

Zeller, W., Gutensohn, K., Stockschlader, M., Dierlamm, J., Kroger, N., Koehne, G., Hummel, K., Kabisch, H., Weh, H.J., KuhnI, P., Hossfeld, D.K. & Zander, A.R. (1996). Increase of mobilized CD34-positive peripheral blood progenitor cells in patients with Hodgkin's disease, non- Hodgkin's iymphoma, and cancer of the testis. Bone Marrow Transplant, 17, 709-13.

203 Zimmerman, T.M., Lee, W.J., Bender, J.G., Mick, R. & Williams, S.F. (1995). Quantitative CD34 analysis may be used to guide peripheral blood stem cell harvests. Bone Marrow Transplant, 1 5, 439-44.

204 Appendix I

List of Reagents used and Suppliers;

ITEM SIZESUPPLIER CAT NO Filter 0.22pm Sartorlus 165 34 K Kwiii/straw 5" Avon Medicals 93G13JF01 Sterile polypropylene 50ml Greiner 210261 universal Sterile polystyrene universal 30ml Sterllln 128A Sterile polypropylene 15ml Falcon 2097 universal Bijou polypropylene vials 7ml Sterllln 129A Hand sealer clips Baxter FDR4418 Line stripper Baxter - Sterile glass pipette 1 ml Volac W076348 Sterile plastic pipettes 5ml Sterllln 03122

lOmIs 03123

25mls 03124 Pipette tips 200ul, lOOOul Alpha labs Microscope slides Menzel-Glaser 04 11404 24 well tissue culture plate Costar 3524 Sterile needles 0.8 X 40mm Neolus NN-213 8R Capped plastic test tubes for Becton Dickinson Falcon 2054 flow cytometry Division Disposable syringes 5ml, 10ml, 50ml Becton Dickinson Sterile sampling site coupler Baxter FMC1401 Disposable cytofunnels Shandon 5991040 Nescofllm 5cm X 40m BDH 235041401

Romanowski staining : Methanol BDH 10158 Lelshmans stain BDH 35022 Gelmsa BDH 35086SP Sorensons Buffer Mercia Diagnostics 02925

Viability stain 0.4% Trypan Blue Sigma T-8154

205 ITEM SIZE SUPPLIER CAT NO General cell culture reagents RPM11640 with L-glutamlne 500ml Gibcc BRL 21875-034 Iscove's (IMDM) 500ml Gibcc BRL 21980-032 Foetal calf serum Gibcc, Europe Ltd., 10106-078 Paisley Scotland Heparin (monoparin) 1,000u/ml CP Pharmaceuticals Ltd., BN95013559 Wrexham, UK Penicillin / Streptomycin 5000U/ 5mg Sigma P0906

Ficcll-Paque 500ml Pharmacia Biotech 17-0840-03 20% Human Serum Albumin ALPHA PL4447/0008 Dimethyl Sulfoxide 10 mL ampuole Sigma D2650 (Tissue culture (DMSO) grade) Phosphate Buffered Saline without Ca"^ or Sigma □5527 IVIg+

In-house methocell Methylcellulose A4 premium Dow Chemical co. grade Hondslow, Middlesex Conditioned media (5637CM) From HTB9 European Cell Culture bladder collection, Porton Down, carcinoma cell Wiltshire line Bovine Serum Albumin RIA grade Sigma A7888 Mixed bed resin AG-501-X8D- Biorad, Hemel 142-6425 50-20 mesh Hempstead, Herts Dulbecco's phosphate X I0 strength Sigma buffered saline yS-mercaptoethanol Sigma M-3148 ' NaHCOg solution 7.5% Gibcc

206 ITEM SIZE SUPPLIER CAT NO

Commercial methocell 80ml Stem ceii Technologies HCG-4230 Ltd (Dist by Metachem. Diagnostics Ltd in UK) Recombinant growth Factors : G-CSF 250|ig/mi Chugai Pharma GM-CSF 250|ig/ml Hoechst GF-409-61 Stem Ceii Factor 10|ig/mi Sigma S-9790 IL-3 300|ig/ml Sandoz (clinical trial vial) Erythropoietin 1000U/mi Boehringer Mannheim PL15722

Flow cytometry/APAAP CD34 (HPCA-2-PE) 2mi Becton Dickinson 348057 Matched irrelavent lgG1-PE 2ml Becton Dickinson 349043 controi CD3 -FITC (UCHT1) 1ml Dako F0818 matched irreiavent lgG1-FITC 1mi Dako X0927 controi AB serum 10mi L.S.E.Z. Diagnostics

APAAP reagents CD34 (HPCA-2) 1 mi Becton Dickinson 348050 CD3 (UCHT1) 1 ml Dako M0835 Rabbit-anti-Mouse 10 mi Dako Z259 APAAP complexes 1ml Dako D651 Iris 1kg Sigma T-8524 Saline 1 litre for Baxter F7124 irrigation (hospital pharmacy) Water 1 litre Baxter UKF7114

(hospital pharmacy) Napthoi-ASBI substrate MakeslOmL of F-4523 alkaline Fast red tablets phosphatase Sigma staining reagent

207