Journal of Clinical Apheresis 7:132-134 (1992)
Hematopoietic Stem Cell Processing and Cryopreservation Scott
D. Rowley
Clinical Cryobiology Laboratory, Fred Hutchinson Cancer Research Center, Seattle, Washington Either bone marrow or peripheral blood may be harvested to provide hematopoictic stem cells (HSC) for autologous transplantation. Both, however, comprise heterogeneous cell populations. The HSC necessary for successful engraftment constitute a very small fraction of the cells harvested. After collection. the harvested cells usually undergo several processing steps to reducc the product volume, remove cells (such as maturc blood cclls or tumor cells). or t o cryopreservc the cells for later reinfusion. Granulocytes and red blood cells, for example, survive cryopreserviltion poorly using freezing techniques designed for HSC. Therefore, bone marrow5 being cryopreserved must be depleted of mature blood cells to avoid toxicity from infusion of damaged mature blood cells. Mature blood cclls may also impede the variety of tumor cell purging techniques currently being studied. These processings are designed to minimize the loss of HSC while achieving an appropriate HSC product for the individual patient. A number of apheresis devices and cell washers simplify the enrichment of HSC in the harvested cell products. In contrast, tumor cell purging techniques are not standardized between the various transplant centers. 0 1992 Wiley-Liss, Inc.
Key words: autologous bone marrow transplantation, bone marrow, peripheral blood stem cell
RATIONALE FOR PROCESSING
Bone marrow and peripheral blood stem cell (PBSC) products comprise heterogeneous cell populations, only a very small proportion of which are hematopoietic stem cells required for engraftment after transplantation. Harvested bone marrow not only contains large quantities of mature peripheral blood cells [ 11, but also noncellular material, such as plasma, bone spicules, clot, and fat. PBSC collections are cleaner in that cell and marrow debris are not present. The frequency of hematopoietic stem cells in the peripheral blood is lower than that found in bone marrow, so that PBSC transplantation is characterized by the need to collect, process, and infuse large quantities of cells. Marrow and PBSC are manipulated in the laboratory to enrich them with hematopoietic progenietors while decreasing the total volume and quantity of peripheral blood cells. Any manipulation intended to purify or preserve hematopoietic stem cells will also be applied to the large fraction of mature blood cells contained in these products. Unfortunately, ex vivo manipulations that are optimal for hematopoietic cells may be extremely damaging to these other cells. Cryopreservation is an obvious example of this problem. Freeze techniques differ for erythrocytes and hematopoietic stem cells. Granulocytes are poorly cryopreserved with any current technique. Reinfusion of cryopreserved PBSC or marrow products containing a large quantity of mature blood cells may precipitate considerable toxicity because of lysis of granulocytes and 0 1992 Wiley-Liss, Inc.
erythrocytes 12-41, The infusion of cryopreserved whole marrow was associated with renal failure in a number of patients in one study 151. For this reason, PBSC and marrow products are virtually always depleted of mature blood cells before cryopreservation. Another prominent reason for processing of these products is that mature blood cells may interfere with purging of tumor cells. Purging of tumor cells from the cell inoculum is a theoretical necessity for the autologous transplantation for malignancies that arise from, or frequently metastasize to, the marrow (e.g., acute leukemia, multiple myeloma, breast cancer, non-Hodgkin's lymphoma). Tumor cells contained in the harvested product could serve as a source of relapse. Both immunologic- and pharmacologic-based techniques have been developed to kill or separate malignant cells from the hematopoietic cells necessary for engraftment. One of the most common purge techniques involves incubating the cells with a derivative of cyclophosphamide, 4-hydroperoxycyclophosphamide (4HC) that is active ex vivo. 4HC is inactivated by an enzyme contained in blood cells, including erythrocytes. Thus, higher red cell concentrations in the incubation mixture will increase 4HC inactivation and will decrease the intensity of the purge [6,7]. Depletion of red cells from the marrow be-
Address reprint requests to Scott D. Rowley, M.D., Clinical Cryobiology Laboratory, Fred Hutchinson Cancer Research Center, 1124 Columbia St., Seattle, WA 98104.
Stem Cell Processing and Cryopreservation
fore 4HC incubation results in a more uniform purge IS]. Mature blood cells can similarly decrease the efficacy of immunologic purging. PROCESSING TECHNIQUES
The goal of marrow and PBSC processing is the removal of mature blood cells, plasma, and noncellular debris while conserving the hematopoietic stem cells necessary for engraftment. Hematopoietic stem cells possess the general size and density characteristics of mature lymphocytes, and can therefore be separated with varying degrees of ease from red cells, granulocytes, and platelets. Most simply, cells may be centrifuged in blood transfer packs using a large-capacity centrifuge. Plasma, buffy-coat cells, and red blood cells can then be expressed into separate containers. This technique can be used to isolate buffy-coat cells for further processing and cryopreservation, or to deplete plasma or red cells for allogeneic transplantation into ABO-incompatible recipients. Hydroxyethyl starch may be added to enhance the separation. Marrow and PBSC processing may be performed using apheresis devices or blood cell washers 19- 111. Some of these devices isolate a relatively enriched mononuclear cell preparation. For example, we found that we recovered 85.7% of nucleated cells in the buffy-coat fraction using a COBE 2991 cell separator, compared to 49.7% using a Haemonetics 30 processor [8]. However, both devices recovered 80% of the mononuclear cells. Mononuclear cells also can be isolated from the buffycoat cells by density-gradient centrifugation over FicollHypaque or Percoll. This further purifies the hematopoietic stem cells by depleting residual red cells and granulocytes from the buffy-coat cell collection. The advantages of using an automated cell separator are speed, uniformity of product, efficiency, and decreased risk of microbial contamination. The drawbacks are the cost of software, the complex nature of the devices requiring skilled operators, and the minimum volumes necessary to prime the system. The choice of techniques is dictated by the types of products being processed, the requirements of purging procedures for specific cell populations, and such administrative concerns as operator experience, availability of devices, and the need to use devices for nonmarrow processing procedures. For some centers a device that can be used for apheresis as well as marrow and PBSC processing may be more cost effective than a device devoted to the laboratory alone. Any processing may be complicated by loss of cells. Catastrophic failure of centrifuge bags or apheresis software is a rare occurrence. Cell loss may prolong the period of post-transplant aplasia. Catastrophic loss could
133
preclude transplantation because of the difficulty in harvesting additional marrow cells, especially from patients previously treated with chemotherapy. Microbial contamination may occur [ 12,131 but most likely because of processing in open containers. CRY0PRESERVATlO N 0F HEMAT0 POI ETIC STEM CELLS
Hematopoietic stem cells cannot be conserved in the laboratory with currently available cell culture techniques. Once harvested, a progressive loss of stem cells occurs. One study found about a 50% loss of myeloid precursors after 2 days for whole bone marrow stored at 4°C [ 141. After 4 days of storage, progenitor cell survival fell to about 12%. In contrast, Delforge et al. stored unfractioned bone marrow for 4 days at 4°C and found 97% recovery of myeloid hematopoietic progenitors 151. The clinical feasibility of nonfrozen storage was subsequently reported from a clinical study of successful engraftment after storage of marrow at 4°C for 54 hours [161. Although marrows stored without freezing for periods up to at least 54 hours will still engraft, the usual pretransplant conditioning regimen of high-dose chemotherapy/radiation often requires 6-8 days to administer. To conserve hematopoietic stem cell function during the conditioning regimen, the cells are cryopreserved. The currently most commonly used cryobiology technique is the use of dimethylsulfoxide (DMSO) as a cryoprotectant, with slow rates of cooling, rapid rates of thawing, and storage below - 100°C. Addition of hydroxyethyl starch may allow a decrease in DMSO concentration ~71. Damage to cells by freezing results from ice formation. If the temperature decrease is slow, the formation of ice will be primarily extracellular. With rapid cooling, ice crystallization will occur within the cell. The effects of ice crystal formation differ according to the location of the ice crystal. Intracellular ice crystals may rupture the cell membrane. Formation of extracellular ice absorbs free water, causing concentration of salts, hyperosmolality, and severe dehydration of the cell. The addition of concentrated penetrating cryoprotectants, such as 10% (1.4 M) DMSO, decreases the amount of water taken into ice crystals at any temperature by colligative effects, and thus, the degree of cellular dehydration. (Colligafive refers to the effects dependent on the number of particles present in the solution and not the nature of the particles.) Solutes that decrease the “freezing temperature” of water also decrease the proportion of water contained in the ice crystals at any subfreezing temperature. Colligative cryoprotectants must penetrate the cell (and must be nontoxic at these high concentrations) to
134
Rowley
avoid aggravating the hyperosmolality caused by extracellular ice formation. REFERENCES 1 . Holdrinet RSG. Egmond JV, Wessels JMC. Haanen C . A method
2.
3.
4.
5.
6.
7.
8.
9.
for quantification of peripheral blood admixture in bone marrow aspirates. Exp Hematol 8: 103-107, 1980. Davis JM, Rowley SD, Braine HG, et al. Clinical toxicity of cryopreserved bone marrow graft infusion. Blood 75:781-786, 1990. Kessinger A, Schmit-Pokorny K, Smith D, Arniitage J. Cryopreservation and infusion of autologous peripheral blood stem cells. Bone Marrow Transpl 5(Suppl 1):25-27. 1990. Stroncek DF, Fautsch SK, Lasky LC, Hurd DD, Ramsay NKC, McCullough J . Adverse reactions in patients transfused with cryopreserved marrow. Transfusion 3 1521-526, 1991. Smith DM, Weisenberger DD, Bierman P, et al. Acute renal failure associated with autologous bone marrow transplantation. Bone Marrow Transpl 2:195-201, 1987. Jones RJ, Zuehlsdorf M. Rowley SD, et al. Variability in 4hydroperoxycyclophosphamide activity during clinical purging for autologous bone marrow transplantation. Blood 70: 14901494. 1987. Rowley SD. Jones RJ, Piantadosi S , et al. Efficacy of ex vivo purging for autologous bone marrow transplantation in treatment of acute nonlymphoblastic leukemia. Blood 74:50 1-506. 1989. Rowley SD. Davis J M , Piantadosi S. et al. Density-gradient separation of autologous bone marrow grafts before ex vivo purging with 4-hydroperoxycyclophosphamide. Bone Marrow Transpl 6:321-327. 1990. Cullis HM, Areman E, Carter CS. Nucleated cell separation using
the Fenwal CS3000'". In Gee A (ed): "Bone Marrow Processing and Purging. A Practical Guide." Boca Raton, FL: CRC Press. 1991. pp 53-71. 10. Hart1 ML: Bone marrow processing with the Haemonetics V50 Plus'". In Gee A (ed): "Bone Marrow Processing and Purging. A Practical Guide." Boca Raton, FL: CRC Press. 1991. pp 87-106. 1 1 . McMannis JD: Use of the COBE 2991'" cell processor for bone marrow processing. In Gee A (ed): "Bone Marrow Processing and Purging. A Practical Guide." Boca Raton, FL: CRC Press. 1991. pp 73-85. 12. Rowley SD. Davis J . Dick J , et al. Bacterial contamination of bone marrow grafts intended for autologous and allogeneic bone marrow transplantation: Incidence and clinical significance. Transfusion 28:109- 112, 1988. 13. Henslee J , Kenyon P. Ferrieri P , et al. Prevention of early gram positive (gm + ) septicemia in autologous bone marrow transplant (ABMT) patients (PTS) (abstr). Proc Am Soc Clin Oncol 3: 100. 1984. 14. Kohsaki M. Yanes B, Ungerleider JS, Murphy MJ. Non-frozen preservation of committed hematopoietic stem cells from normal human bone marrow. Stem Cells I : I 11-123. 1981. 15. Delforge A , Ronge-Collard E, Stryckmans P. et al. Granulocytemacrophage progenitor cells preservation at 4°C. Br J Haematol 53:49-54. 1983. 16. Burnett AK, Tansey P. Hills C , et al. Haematological reconstitution following high dose and supralethal chemo-radiotherapy using stored, non-cryopreserved autologous bonc marrow. Br J Haematol 54:309-316, 1983. 17. Stiff PJ, Koester AR. Weidner MK. et al. Autologous bone marrow transplantation using unfractionated cells cryopreserved in dimethylsulfoxide and hydroxyethyl starch Nithout controlledrate freezing. Blood 70:974-978, 1987.
Hematopoietic Stem Cell Processing and Cryopreservation Scott
D. Rowley
Clinical Cryobiology Laboratory, Fred Hutchinson Cancer Research Center, Seattle, Washington Either bone marrow or peripheral blood may be harvested to provide hematopoictic stem cells (HSC) for autologous transplantation. Both, however, comprise heterogeneous cell populations. The HSC necessary for successful engraftment constitute a very small fraction of the cells harvested. After collection. the harvested cells usually undergo several processing steps to reducc the product volume, remove cells (such as maturc blood cclls or tumor cells). or t o cryopreservc the cells for later reinfusion. Granulocytes and red blood cells, for example, survive cryopreserviltion poorly using freezing techniques designed for HSC. Therefore, bone marrow5 being cryopreserved must be depleted of mature blood cells to avoid toxicity from infusion of damaged mature blood cells. Mature blood cclls may also impede the variety of tumor cell purging techniques currently being studied. These processings are designed to minimize the loss of HSC while achieving an appropriate HSC product for the individual patient. A number of apheresis devices and cell washers simplify the enrichment of HSC in the harvested cell products. In contrast, tumor cell purging techniques are not standardized between the various transplant centers. 0 1992 Wiley-Liss, Inc.
Key words: autologous bone marrow transplantation, bone marrow, peripheral blood stem cell
RATIONALE FOR PROCESSING
Bone marrow and peripheral blood stem cell (PBSC) products comprise heterogeneous cell populations, only a very small proportion of which are hematopoietic stem cells required for engraftment after transplantation. Harvested bone marrow not only contains large quantities of mature peripheral blood cells [ 11, but also noncellular material, such as plasma, bone spicules, clot, and fat. PBSC collections are cleaner in that cell and marrow debris are not present. The frequency of hematopoietic stem cells in the peripheral blood is lower than that found in bone marrow, so that PBSC transplantation is characterized by the need to collect, process, and infuse large quantities of cells. Marrow and PBSC are manipulated in the laboratory to enrich them with hematopoietic progenietors while decreasing the total volume and quantity of peripheral blood cells. Any manipulation intended to purify or preserve hematopoietic stem cells will also be applied to the large fraction of mature blood cells contained in these products. Unfortunately, ex vivo manipulations that are optimal for hematopoietic cells may be extremely damaging to these other cells. Cryopreservation is an obvious example of this problem. Freeze techniques differ for erythrocytes and hematopoietic stem cells. Granulocytes are poorly cryopreserved with any current technique. Reinfusion of cryopreserved PBSC or marrow products containing a large quantity of mature blood cells may precipitate considerable toxicity because of lysis of granulocytes and 0 1992 Wiley-Liss, Inc.
erythrocytes 12-41, The infusion of cryopreserved whole marrow was associated with renal failure in a number of patients in one study 151. For this reason, PBSC and marrow products are virtually always depleted of mature blood cells before cryopreservation. Another prominent reason for processing of these products is that mature blood cells may interfere with purging of tumor cells. Purging of tumor cells from the cell inoculum is a theoretical necessity for the autologous transplantation for malignancies that arise from, or frequently metastasize to, the marrow (e.g., acute leukemia, multiple myeloma, breast cancer, non-Hodgkin's lymphoma). Tumor cells contained in the harvested product could serve as a source of relapse. Both immunologic- and pharmacologic-based techniques have been developed to kill or separate malignant cells from the hematopoietic cells necessary for engraftment. One of the most common purge techniques involves incubating the cells with a derivative of cyclophosphamide, 4-hydroperoxycyclophosphamide (4HC) that is active ex vivo. 4HC is inactivated by an enzyme contained in blood cells, including erythrocytes. Thus, higher red cell concentrations in the incubation mixture will increase 4HC inactivation and will decrease the intensity of the purge [6,7]. Depletion of red cells from the marrow be-
Address reprint requests to Scott D. Rowley, M.D., Clinical Cryobiology Laboratory, Fred Hutchinson Cancer Research Center, 1124 Columbia St., Seattle, WA 98104.
Stem Cell Processing and Cryopreservation
fore 4HC incubation results in a more uniform purge IS]. Mature blood cells can similarly decrease the efficacy of immunologic purging. PROCESSING TECHNIQUES
The goal of marrow and PBSC processing is the removal of mature blood cells, plasma, and noncellular debris while conserving the hematopoietic stem cells necessary for engraftment. Hematopoietic stem cells possess the general size and density characteristics of mature lymphocytes, and can therefore be separated with varying degrees of ease from red cells, granulocytes, and platelets. Most simply, cells may be centrifuged in blood transfer packs using a large-capacity centrifuge. Plasma, buffy-coat cells, and red blood cells can then be expressed into separate containers. This technique can be used to isolate buffy-coat cells for further processing and cryopreservation, or to deplete plasma or red cells for allogeneic transplantation into ABO-incompatible recipients. Hydroxyethyl starch may be added to enhance the separation. Marrow and PBSC processing may be performed using apheresis devices or blood cell washers 19- 111. Some of these devices isolate a relatively enriched mononuclear cell preparation. For example, we found that we recovered 85.7% of nucleated cells in the buffy-coat fraction using a COBE 2991 cell separator, compared to 49.7% using a Haemonetics 30 processor [8]. However, both devices recovered 80% of the mononuclear cells. Mononuclear cells also can be isolated from the buffycoat cells by density-gradient centrifugation over FicollHypaque or Percoll. This further purifies the hematopoietic stem cells by depleting residual red cells and granulocytes from the buffy-coat cell collection. The advantages of using an automated cell separator are speed, uniformity of product, efficiency, and decreased risk of microbial contamination. The drawbacks are the cost of software, the complex nature of the devices requiring skilled operators, and the minimum volumes necessary to prime the system. The choice of techniques is dictated by the types of products being processed, the requirements of purging procedures for specific cell populations, and such administrative concerns as operator experience, availability of devices, and the need to use devices for nonmarrow processing procedures. For some centers a device that can be used for apheresis as well as marrow and PBSC processing may be more cost effective than a device devoted to the laboratory alone. Any processing may be complicated by loss of cells. Catastrophic failure of centrifuge bags or apheresis software is a rare occurrence. Cell loss may prolong the period of post-transplant aplasia. Catastrophic loss could
133
preclude transplantation because of the difficulty in harvesting additional marrow cells, especially from patients previously treated with chemotherapy. Microbial contamination may occur [ 12,131 but most likely because of processing in open containers. CRY0PRESERVATlO N 0F HEMAT0 POI ETIC STEM CELLS
Hematopoietic stem cells cannot be conserved in the laboratory with currently available cell culture techniques. Once harvested, a progressive loss of stem cells occurs. One study found about a 50% loss of myeloid precursors after 2 days for whole bone marrow stored at 4°C [ 141. After 4 days of storage, progenitor cell survival fell to about 12%. In contrast, Delforge et al. stored unfractioned bone marrow for 4 days at 4°C and found 97% recovery of myeloid hematopoietic progenitors 151. The clinical feasibility of nonfrozen storage was subsequently reported from a clinical study of successful engraftment after storage of marrow at 4°C for 54 hours [161. Although marrows stored without freezing for periods up to at least 54 hours will still engraft, the usual pretransplant conditioning regimen of high-dose chemotherapy/radiation often requires 6-8 days to administer. To conserve hematopoietic stem cell function during the conditioning regimen, the cells are cryopreserved. The currently most commonly used cryobiology technique is the use of dimethylsulfoxide (DMSO) as a cryoprotectant, with slow rates of cooling, rapid rates of thawing, and storage below - 100°C. Addition of hydroxyethyl starch may allow a decrease in DMSO concentration ~71. Damage to cells by freezing results from ice formation. If the temperature decrease is slow, the formation of ice will be primarily extracellular. With rapid cooling, ice crystallization will occur within the cell. The effects of ice crystal formation differ according to the location of the ice crystal. Intracellular ice crystals may rupture the cell membrane. Formation of extracellular ice absorbs free water, causing concentration of salts, hyperosmolality, and severe dehydration of the cell. The addition of concentrated penetrating cryoprotectants, such as 10% (1.4 M) DMSO, decreases the amount of water taken into ice crystals at any temperature by colligative effects, and thus, the degree of cellular dehydration. (Colligafive refers to the effects dependent on the number of particles present in the solution and not the nature of the particles.) Solutes that decrease the “freezing temperature” of water also decrease the proportion of water contained in the ice crystals at any subfreezing temperature. Colligative cryoprotectants must penetrate the cell (and must be nontoxic at these high concentrations) to
134
Rowley
avoid aggravating the hyperosmolality caused by extracellular ice formation. REFERENCES 1 . Holdrinet RSG. Egmond JV, Wessels JMC. Haanen C . A method
2.
3.
4.
5.
6.
7.
8.
9.
for quantification of peripheral blood admixture in bone marrow aspirates. Exp Hematol 8: 103-107, 1980. Davis JM, Rowley SD, Braine HG, et al. Clinical toxicity of cryopreserved bone marrow graft infusion. Blood 75:781-786, 1990. Kessinger A, Schmit-Pokorny K, Smith D, Arniitage J. Cryopreservation and infusion of autologous peripheral blood stem cells. Bone Marrow Transpl 5(Suppl 1):25-27. 1990. Stroncek DF, Fautsch SK, Lasky LC, Hurd DD, Ramsay NKC, McCullough J . Adverse reactions in patients transfused with cryopreserved marrow. Transfusion 3 1521-526, 1991. Smith DM, Weisenberger DD, Bierman P, et al. Acute renal failure associated with autologous bone marrow transplantation. Bone Marrow Transpl 2:195-201, 1987. Jones RJ, Zuehlsdorf M. Rowley SD, et al. Variability in 4hydroperoxycyclophosphamide activity during clinical purging for autologous bone marrow transplantation. Blood 70: 14901494. 1987. Rowley SD. Jones RJ, Piantadosi S , et al. Efficacy of ex vivo purging for autologous bone marrow transplantation in treatment of acute nonlymphoblastic leukemia. Blood 74:50 1-506. 1989. Rowley SD. Davis J M , Piantadosi S. et al. Density-gradient separation of autologous bone marrow grafts before ex vivo purging with 4-hydroperoxycyclophosphamide. Bone Marrow Transpl 6:321-327. 1990. Cullis HM, Areman E, Carter CS. Nucleated cell separation using
the Fenwal CS3000'". In Gee A (ed): "Bone Marrow Processing and Purging. A Practical Guide." Boca Raton, FL: CRC Press. 1991. pp 53-71. 10. Hart1 ML: Bone marrow processing with the Haemonetics V50 Plus'". In Gee A (ed): "Bone Marrow Processing and Purging. A Practical Guide." Boca Raton, FL: CRC Press. 1991. pp 87-106. 1 1 . McMannis JD: Use of the COBE 2991'" cell processor for bone marrow processing. In Gee A (ed): "Bone Marrow Processing and Purging. A Practical Guide." Boca Raton, FL: CRC Press. 1991. pp 73-85. 12. Rowley SD. Davis J . Dick J , et al. Bacterial contamination of bone marrow grafts intended for autologous and allogeneic bone marrow transplantation: Incidence and clinical significance. Transfusion 28:109- 112, 1988. 13. Henslee J , Kenyon P. Ferrieri P , et al. Prevention of early gram positive (gm + ) septicemia in autologous bone marrow transplant (ABMT) patients (PTS) (abstr). Proc Am Soc Clin Oncol 3: 100. 1984. 14. Kohsaki M. Yanes B, Ungerleider JS, Murphy MJ. Non-frozen preservation of committed hematopoietic stem cells from normal human bone marrow. Stem Cells I : I 11-123. 1981. 15. Delforge A , Ronge-Collard E, Stryckmans P. et al. Granulocytemacrophage progenitor cells preservation at 4°C. Br J Haematol 53:49-54. 1983. 16. Burnett AK, Tansey P. Hills C , et al. Haematological reconstitution following high dose and supralethal chemo-radiotherapy using stored, non-cryopreserved autologous bonc marrow. Br J Haematol 54:309-316, 1983. 17. Stiff PJ, Koester AR. Weidner MK. et al. Autologous bone marrow transplantation using unfractionated cells cryopreserved in dimethylsulfoxide and hydroxyethyl starch Nithout controlledrate freezing. Blood 70:974-978, 1987.
