Technical report High-efficiency volume reduction of cord blood using pentastarch

(2001) 27, 457 461 2001 Nature Publishing Group All rights reserved 0268 3369/01 $15.00 www.nature.com/bmt Technical report High-efficiency volume reduction of cord blood using pentastarch H Yang 1,2, JP Acker 3, D Abley 1, LE McGann 1,2,3 and J Akabutu 1,4 1 Alberta Cord Blood Bank, 2 Canadian Blood Services, Departments of 3 Laboratory Medicine and Pathology, 4 Pediatrics, University of Alberta, Edmonton, Alberta, Canada Summary: Human umbilical cord blood (UCB) has been used successfully to treat a variety of genetic, hematological, and oncologic disorders. However, the low number of hematopoietic progenitor cells available in donated cord blood samples limits transplantation of cord blood to children and small adults. Reduction of the volume of umbilical cord blood is widely used in cord blood banking to reduce the storage requirements in large-scale UCB banks. Unfortunately, during the volume reduction process, up to 40% or more of the progenitor cells are lost using current reduction methods. This study describes a highly reproducible, double collection technique using Pentaspan to reduce UCB volume by red cell depletion. This results in the preservation of critical hematopoietic progenitor cells. The final volume of the leukocyte concentrates (LC) was 19.8 ± 0.4 ml with 95% red cell depletion. The recovery of nucleated cells (NC), mononuclear cells (MNC), CD34 + cells and colony-forming units (CFU) following double collection was 89%, 94%, 96%, and 106%, respectively. This is significantly higher than the recovery from single collections, where recovery was 74%, 77%, 84%, and 91% for NC, MNC, CD34 + and CFU, respectively. The double collection technique provides an efficient and highly reproducible method for the preparation of UCB for long-term storage and transplantation. Bone Marrow Transplantation (2001) 27, 457 461. Keywords: umbilical cord blood; flow cytometry; volume reduction; leukocytes; CD34 + Over the last decade, more than 1000 cases of successful human umbilical cord blood transplantation have been reported. This has established the utility of using umbilical cord blood as an alternative source of hematopoietic stem cells for bone marrow reconstitution. Hematopoietic progenitors from umbilical cord blood have many advantages over adult bone marrow and peripheral blood stem cells. 1 4 The main disadvantage of umbilical cord blood is the Correspondence: Dr H Yang, Alberta Cord Blood Bank, 3rd Floor Canadian Blood Services Building, 8249 114 Street, Edmonton, Alberta T6G 2R8, Canada Received 13 July 2000; accepted 30 October 2000 restricted volume of the donated cord blood sample. This limits the absolute number of hematopoietic progenitor cells available for transplantation. A low nucleated cell dose can significantly decrease recipient survival after hematopoietic progenitor cell transplantation. 5 Maximizing the number of hematopoietic progenitor cells available for transplantation is critical, and efforts have focused on improving the number of cells that are recovered during cord blood banking procedures. is widely used in cord blood banking to improve available storage space. 6,7 The current method for volume reduction was developed by Rubinstein et al 6 and is followed by most cord blood banks worldwide. The volume of cord blood is reduced by enhanced sedimentation of red cells by rouleaux formation that is induced by a strong sedimenting agent such as 6% hydroxyethyl starch (Hespan). However, the number of nucleated cells lost during this procedure can be 40% or more. 1,8,9 As a result of this cell loss, mainly children meet the optimal nucleated cell dose of 3.7 10 7 /kg of recipient body weight that is used as a minimum requirement for a successful transplantation outcome. 10 In order to expand the pool of patients able to receive transplants, only donated cord blood samples that have a large initial volume are banked. Unfortunately, valuable cord blood samples that do not meet the minimum volume criteria or minimum recovery of leukocytes are routinely discarded. Without more efficient methods that reduce the volume of umbilical cord blood samples, use of cord blood as an alternative source of hematopoietic progenitor cells will remain limited to children and small adults. This study describes an efficient and reproducible method using Pentaspan to reduce the total volume of cord blood samples without significantly reducing the number of nucleated cells during the process. Materials and methods Cord blood collection Registered mothers, with informed consent, donated cord blood samples from normal full-term deliveries under a protocol reviewed by the Institution s Health Research Ethics Board. Cord blood was collected from the umbilical vein by gravity drainage into a blood collection bag containing 23 ml citrate phosphate dextrose adenine (CPDA-

458 1) anticoagulant (Medsep, Covina, CA, USA). The collected cord blood was stored at room temperature and processed within 36 h of collection. Volume reduction Single collection: Ten percent pentastarch (Pentaspan; DuPont, Mississauga, Ontario, Canada) was added to the cord blood in the collection bag to obtain a final concentration of 2% and was incubated at room temperature for 5 min. Centrifugation time was determined by the volume of cord blood with CPDA-1 using the following, empirically derived formula (unpublished): K CT = (1) L M where CT = centrifugation time, K = 7.7227, L = ln (volume of cord blood with 23 ml CPDA), M = 26.242. The bag was held vertically in a centrifuge bucket using styrofoam inserts and centrifuged at 52 g at 10 C for the time calculated from the above formula. After completion of the centrifugation, the buffy coat was expressed into a 150 ml transfer bag (No. 1 transfer bag; Baxter Healthcare, Deerfield, IL, USA) using a plasma extractor (Baxter Healthcare). The transfer bag was then centrifuged at 400 g for 10 min at 10 C. The plasma was expressed into a second transfer bag (No. 2 transfer bag) resulting in approximately 20 ml of leukocyte concentrate (LC) remaining in the first transfer bag. The leukocyte concentrate was then assessed. Double collection: The first collection of the double collection was the same as for the single collection except that plasma from No. 1 transfer bag was returned to the collection bag (containing depleted red blood cells and leukocytes). The closed processing system was reassembled by reconnecting the plastic tubing of No. 1 transfer bag and the collection bag using a sterile tubing welder (Terumo Medical, MD, USA). The tube on the collection bag was then sealed using a hand sealer (Sebra, Tucson, AZ, USA) at 10 cm from the bag and then cut. The collection bag was held vertically in a centrifuge bucket using styrofoam inserts, and centrifuged at 52 g at 10 C for the time calculated from Equation 1. When multiple samples were processed, the average centrifuge time as calculated from Equation 1 was used. After completion of the centrifugation, the collection bag was re-connected to No. 1 transfer bag (containing the initial leukocyte concentrate) using the sterile tube welder as before. The No. 1 transfer bag was then centrifuged at 400 g for 10 min at 10 C. After completion of the centrifugation, the plasma was expressed into a second transfer bag (No. 2 transfer bag) using a plasma extractor, leaving a final volume of 20 ml of leukocyte concentrate in the No. 1 transfer bag. Enumeration and viability of CD34 + cells CD34 + cells were quantified using the flow cytometric technique described by Keeney et al. 11 Briefly, 2 10 6 nucleated cells from either cord blood, LC from first collection, or LC from double collection were treated with either 20 l CD34-PE (580 clone)/cd45-fitc (J33 clone) or CD45-FITC/Isoclonic PE at room temperature for 20 min in the dark. Two milliliters of 10% ammonium chloride was added to the tubes followed by 20 l 7-amino-actinomycin D (7-AAD, 1 g/ml; Molecular Probes, Eugene, OR, USA). The tubes were incubated for 15 min in the dark, and then 100 l of Coulter Stem-Count beads was added. Within one hour of sample preparation, flow cytometric analysis was performed on a Coulter EPICS XL/MCL cytometer (Coulter, Hialeah, FL, USA). The 7-AAD functions to determine cell viability. 7-AAD is an analog of actinomycin D that binds between cytosine and guanine bases of DNA. 12 The 7-AAD/DNA complex has a maximum absorption at 546 nm with a red fluorescent emission at 647 nm. These spectral characteristics result in 7-AAD being well suited for argon laser equipped cytometry that has an excitation wavelength of 488 nm. There is little spectral overlap between 7-AAD and FITC or PE, further enhancing its suitability for assessing CD34 + cells with single platform flow cytometry. Viability of nucleated cells The viability of nucleated cells was assessed in the 7-AAD stained samples, and analyzed using a flow cytometer. Hematological cell counts The number of total NC and MNC was determined using a Coulter A c.t diff analyzer (Beckman-Coulter, Miami, FL, USA). Any abnormal flags on the Coulter report were verified by manual count of 200 nucleated cells using Sure Stain, a Wright-Giemsa based stain (Fisher Scientific, Pittsburgh, PA, USA) following the manufacturer s instructions. Colony-forming assay (CFU) CFU in pre-processed (whole cord blood), LC from single collection and LC from double collection were assessed using a commercially prepared complete methylcellulose medium, Methocult GF H4434 (StemCell Technologies, Vancouver, BC, Canada). Cells were plated without further separation in triplicate at a concentration of 1.25 10 4 cells per well. After incubation at 37 C for 14 days in humidified air containing 5% CO 2, granulocyte macrophage (CFU- GM), erythroid (BFU-E) and granulocyte erythroid macrophage megakaryocyte (CFU-GEMM) colonies were scored by microscopic examination. Bacteriology To check for contamination, 0.5 ml samples of pre-processed cord blood and LC collected after the double collection technique were tested for sterility. Bacterial and fungal contamination were evaluated using the Bactec method (Becton Dickinson, Mountain View, CA, USA). Statistics A statistical comparison between the recovery obtained using the single and double collection methods was perfor-

Volume reduction, red cell depletion, viability and contami- n Mean ± s.e.m. (range) Pre-processed Double collection Table 1 nation NC, MNC and CD34 + cells. The results also indicate that the range for all cell types is narrower when the double collection technique is used compared to the single collection technique. The recovery from the double collection was much more consistent. 459 Volume (ml) 22 79.2 ± 5.1 19.8 ± 0.4 (45.0 131) (18.0 20.4) RBC depletion (%) 22 NA 94.5 ± 0.7 (88.3 98.6) Viability (%) Nucleated cells 19 92.8 ± 2.2 93.4 ± 0.29 (87.4 99.1) (89.8 99.3) CD34 + cells 22 98.1 ± 0.8 97.1 ± 0.5 (91 100) (91 100) Sterility test 22 Negative Negative NA = not applicable. med. A one-way ANOVA was used with a level of significance set at 0.05. Results Volume reduction Table 1 shows that using the double collection technique reduces the volume of cord blood by three quarters from an original volume of 79 ± 5.1 ml to 19.8 ± 0.4 ml without a significant loss of viability of nucleated cells or CD34 + cells. The achieved red cell depletion was 94.5 ± 0.7%. Cell recovery Table 2 shows total cell counts of pre-processed samples, and samples processed by single collection or double collection method. More cells are recovered from the double collection than from the single collection method. In Table 2, percent recovery of NC, MNC, CD34 + cells and CFU following processing is shown. Double collection technique improved the average recovery of NC, MNC, CD34 + cells, and CFU by 15%, 18%, 12% and 15%, respectively. The increased recovery obtained using the double collection technique is statistically significant for Bacteriology As showed in Table 1, there was no bacterial or fungal contamination of either the pre-processed samples or the samples from the double collection process following a 7-day culture. Discussion This study describes a double collection method that can be used to significantly enhance the recovery of the LC from umbilical cord blood samples while providing excellent volume reduction. The single collection permits depletion of red cells, reduction of plasma volume and collection of LC. This single collection method is widely used in clinical cord blood banks. However, the recovery of NC and MNC is highly variable. The cord blood banking standard for recovery of NC and MNC has been proposed to be 60% and 80%, respectively. 8 In Table 2, recovery of NC after a single collection was 73.5% with a range of 46.6 to 100%. According to the proposed standard, two out of the 22 cord blood specimens in this study, or 9%, would have had to be discarded. However, when the double collection procedure was employed recovery was increased to 89.2% with a range of 81 99.7%. In this case, none of the samples would have had to have been discarded. It is shown in Table 1 that the recovery of NC, MNC, CD34 + cells and CFU from a double collection was significantly higher than from a single collection. More important is the fact that the cell recovery from the double collection method was highly consistent within a very narrow range. These data suggest that the proposed cord blood banking standard requiring cord blood samples to be discarded after unsuccessful volume reduction 8 is not necessary if a double collection process is employed especially for large-volume cord blood samples. The process is appropriately named double collection as Table 2 Total cell count and yield of cord blood from pre-processed, single and double collection Cell type n Cell count/unit ( 10 6 ), mean ± s.e.m. (range) Yield (%), mean ± s.e.m. (range) Pre-processed Single collection Double collection Single collection Double collection P Nucleated cells 22 1399 ± 145 1012 ± 123 1241 ± 129 73.5 ± 3.4 89.2 ± 1.2 0.001 (683 3415) (488 2067) (594 3338) (46.6 100) (81 99.7) Mononuclear cells 22 627 ± 76 456 ± 53 589 ± 72 76.6 ± 3.0 94.4 ± 1.4 0.001 (262 1837) (208 887) (254 998) (48.9 92.9) (84.4 110.9) CD34 + 9 4.9 ± 1.4 4.2 ± 1.2 4.8 ± 1.4 84.2 ± 4.4 96.3 ± 2.2 0.015 (1.6 14.5) (0.3 12.1) (0.33 14.6) (63.1 108.1) (85.3 105.0) Total CFU 9 2.1 ± 0.4 1.5 ± 0.4 2.2 ± 0.4 91.1 ± 7.8 106.4 ± 9.1 0.26 (1.6 4.2) (0.12 3.4) (0.13 4.7) (71.5 116.8) (85.3 115.0)

460 the residual cord blood and expressed plasma is mixed and centrifuged to collect the leukocytes lost in the first collection. The end result is increased leukocyte recovery. The velocity ratio of the leukocytes to red blood cells is an important parameter in this separation. It is practically impossible to separate leukocytes from red blood cells without the addition of a strong sedimenting agent, such as HES, or pentastrach. These agents induce rouleaux, or cell packing, which causes the red blood cells to settle more quickly. In the absence of rouleaux, leukocytes would sediment at the same rate due to a similar cell density as red cells. 13 As the packing of red blood cells becomes more and more compact during sedimentation, the spaces between the packed red cells become smaller and smaller. The leukocytes, therefore, concentrate in the channels between packed red cells, and collide. As a result of this collision of leukocytes, recovery of the cells is reduced. 13 This is supported by the work of Akabutu et al 14 who observed that the recovery of leukocytes is inversely proportional to the volume of the cord blood sample. Therefore, to enhance the recovery of trapped leukocytes, the plasma from the first collection can be returned to the residual cord blood and re-centrifuged. There were three concerns raised when the double collection technique was applied. The first was regarding the quality of the recovered cells following two separate collection procedures. It was therefore important to assess the viability of the NC, CD34 + cells, and CFU throughout the procedure. As seen in Table 1, there was no increase in the amount of damage to CD34 + cells and NC following the double collection. This is further exemplified by the fact that the recovery of CFUs was not affected by adding a second collection process (Table 2). The second concern was with the increased length of processing time. A double collection method takes significantly more time than a single collection method, so understanding what effect this will have in the clinical environment is important. For samples of an identical volume, the double collection process will take twice as long as the single collection, but this disadvantage should be compensated for by consistently higher recovery of cells with a lower proportion of samples being discarded. In fact, when there are multiple samples being processed, and each sample has a different volume, the double collection method can be more effective. As each sample has a different volume, a different centrifugation time will be required according to Equation 1 and as a result the samples would need to be processed separately. However, the double collection time is not as sensitive to sample volume (unpublished) and the centrifugation time can be averaged when multiple samples are processed. In practical terms, the double collection technique can be faster than the single collection method as it allows multiple cord blood samples to be processed together using an average centrifugation time (unpublished). The third concern was bacterial and fungal contamination due to the increased number of transfers required during the double collection process. To prevent this, a sterile tubing welder was used to connect transfer bags. This allowed processing of the umbilical cord blood in a closed system. No contamination was detected in any of the processed samples (Table 1). Pentaspan was used in this study to reduce the volume of cord blood instead of the widely used Hespan. In addition to the fact that Hespan is not clinically approved or licensed in a number of countries, like Canada, there are a number of advantages to the use of Pentaspan over HES that makes it more suitable for use in cord blood banking. As HES can be detected years after its administration, 15 the safety of this sedimenting agent has recently been questioned. 16 In comparison, Pentaspan is a lower molecular weight, hetastarch-analog that is cleared from the circulation relatively rapidly. For this reason, and the fact that Pentaspan is equally as effective as HES in inducing red blood cell sedimentation, 17 Pentaspan was used as the sedimenting agent in this study. In conclusion, the double collection technique is an efficient and reproducible method of reducing cord blood volume without depleting the sample of NC, MNC, CD34 + cells and CFU. This procedure has the immediate advantage of decreasing, and perhaps eliminating, the number of clinical samples that are discarded due to low recovery of transplantable cells. Furthermore, by increasing the recovery of stem cells from donated cord blood, more patients, of larger weights, will meet the cell dose criteria required for transplantation. Acknowledgements This study was supported by the Dr Charles Allard Foundation and a grant in aid from the Western Economic Partnership Agreement (Alberta Health and Wellness, Government of Alberta and Western Diversification, Government of Canada), 1999. References 1 Broxmeyer HE, Gordon GW, Hangoc G et al. Human umbilical cord blood as a potential source of transplantable hematopoietic stem/progenitor cells. Proc Natl Acad Sci USA 1989; 86: 3828 3832. 2 Broxmeyer HE, Hangoc G, Cooper S et al. Growth characteristics and expansion of human umbilical cord blood and estimation of its potential for transplantation in adults. Proc Natl Acad Sci USA 1992; 89: 4109 4113. 3 Vaziri H, Dragowska W, Allsop RC et al. Evidence for a mitotic clock in human hematopoietic stem cells: loss of telomeric DNA with age. Proc Natl Acad Sci USA 1994; 91: 9857 9860. 4 Rocha V, Wagner JE, Sobocinski KA et al. Graft-versus-host disease in children who have received a cord-blood or bone marrow transplant from an HLA-identical sibling. New Engl J Med 2000; 342: 1846 1854. 5 Donaldson C, Armitage WJ, Laundy V et al. Impact of obstetric factors on cord blood donation for transplantation. Br J Haematol 1999; 106: 128 132. 6 Rubinstein P, Dobrila L, Richard E et al. Processing and cryopreservation of placental/umbilical cord blood for unrelated bone marrow reconstitution. Proc Natl Acad Sci USA 1995; 92: 10119 10122. 7 Perutelli P, Gatellani S, Scarso L et al. Processing of human cord blood by three different procedures for red blood cell

depletion and mononuclear cell recovery. Vox Sang 1999; 76: 237 240. 8 Fraser J, Cairo MS, Wagner E et al. Cord blood transplantation study (COBLT): cord blood bank standard operating procedures. J Hematother 1998; 7: 521 561. 9 Sousa T, De Sousa ME, Godinho MI et al. Umbilical cord blood processing: volume reduction and recovery of CD34 + cells. Bone Marrow Transplant 1997; 19: 311 313. 10 Gluckman E, Rocha V, Boyer-Chammard A et al. Outcome of cord blood transplantation from related and unrelated donors. New Engl J Med 1997; 337: 373 381. 11 Keeney M, Chin-Yee I, Weir K et al. Single platform flow cytometric absolute CD34 + cell counts based on the ISHAGE guidelines. Cytometry 1998; 34: 61 70. 12 Haugland RP. Handbook of Fluorescent Probes and Research Chemicals. Molecular Probes: Eugene, OR, 1996. 13 Brown RI. The physics of continuous flow centrifugal cell separation. Artificial Organs 1989; 13: 4 20. 14 Akabutu J, Yang H, McGann LE. Centralized public banking of umbilical cord blood stem cells: a Canadian perspective. Blood 1999; 94 (Suppl. 1): 10 (Abstr. 2538). 15 Trivedi SM, Humphrey RL, Braine HD et al. Hydroxyethyl starch serum levels in leukapheresis donors measured by modified periodic acid-schiff staining technique. Transfusion 1984; 24: 260 263. 16 Lee JH, Leitman SF, Klein HG. A controlled comparison of the efficacy of hetastarch and pentastarch in granulocyte collections by centrifugal leukapheresis. Blood 1995; 86: 4664 4666. 17 Strauss RG. In vitro comparison of the erythrocytes sedimenting properties of dextran, hydroxyethyl starch and a new low-molecular weight hydroxyethyl starch. Vox Sang 1979; 37: 268 271. 461