From www.bloodjournal.org by guest on March 12, 2019. For personal use only.
Effect of Cell Concentration on Bone Marrow and Peripheral Blood Stem
Cell Cryopreservation
By Scott D. Rowley, William I. Bensinger, Ted A. Gooley, and C. Dean Buckner
The effects of cell concentration during cryopreservationon
bone marrow (BM) orperipheral blood (PB)-derivedhemato-
poietic progenitor cells have not been described. The much
greater numbers of cells harvestedfor autologous PB stem
cell (PBSC) transplantation requires that thecells be frozen
at higher cell concentrations, or in much greater volumes,
compared with BM. We cryopreserved 108 PBSCcollections
from 30 patients at an average (&SDI cell concentration of
3.7 -I- 1.9 x 10’ nucleated cells per mL in 127 & 45 mL. The
proportion of mononuclear cells was 52.9% k 27.2%. The
products also contained 2.9 & 2.1 x los platelets/mL and an
average red cell proportion of 12.9% ? 7.2%. The nucleated
cell recovery after thawing was 75.4y0 & 13.096. The nucle-
ated cell concentration during freezing was not predictive
for the postthawrecoveries of nucleated cells ( P = .38), gran-
ulocyte-macrophagecolony-forming unit ( P = .06) or CD34’
cells (P = .54), or for the viability of mononuclear cells( P =
.W). The platelet and red cell concentrations similarly were
not predictive for these endpoints. Samples (3BM, 7 PBSCI
from 10 patients were simultaneously cryopreserveda t two-
C RYOPRESERVATION OF hematopoietic stem cells
(HSC) is a routine aspect of autologous bone marrow
(BM) transplantation. Progressive loss of HSC viability over
time, beyond that associated with the freezing and thawing
procedures, does not appear to occur if storage conditions
are adequate. Thus, cryopreservation allows the administra-
tion of conditioning regimens requiring multiple days, as
well as storage of HSC for future use. Although no standard
technique is accepted by all centers, variations in techniques
are generally minor.’ virtually all centers performing autolo-
gous BM transplantation cryopreserve cells in dimethylsulf-
oxide (DMSO). Furthermore, engraftment failure or delay
has not been attributed to variations in technique, although
optimal conditions such as concentrations of cells or protein,
and storage temperatures have not been defined for human
HSC.
Recently, several centers have reported the transplantation
of peripheral blood stem cells (PBSC) collected after mobili-
zation by granulocyte-colony stimulating factor (G-CSF) or
granulocyte-macrophage colony-stimulating factor (GM-
CSF).*-’ Compared with collection during rebound after che-
motherapy, large quantities of cells (frequently exceeding 5
X 10” cells) are collected during each apheresis. These cell
quantities exceed by several-fold the quantity of cells usually
harvested for BM transplantation. Cryopreservation of these
cells at the cell concentrations generally used for BM (com-
monly, 2 to 4 X lo7 nucleated cells/mL’) generates large
product volumes containing large quantities of DMSO. Rein-
fusion of these cells may be associated with considerable
toxicity during infusion, unless cryopreservation and reinfu-
sion techniques are modified in response to the quantity of
cells harve~ted.~.’ Options include washing and concentrating
the cells after thawing, or freezing at higher cell concentra-
tions.
We concentrated PBSC collected after G-CSF or GM-
CSF mobilization in minimal volumes resulting in high cell
concentrations during cryopreservation, and prospectively
studied HSC recovery after thawing. No consistent detrimen-
Blood, Vol 83, No 9 (May l ) , 1994 pp 2731-2736
fold, and from 5 additional patients (PBSC) at 6- t o 24-fold
differing cell concentrations. A lower recovery of erythroid
burst forming unit was found for samples frozen at higher
cell concentrations ( P = -04). but no significant differences
were found in theother endpoints listed above. The average
Celt concentration during freezing for each patient’s PBSC
collections (n = 34 patients) did notpredict time t o achieve
a PB count of >500 granulocyteslpl ( P = 51) or platelet
transfusion independence ( P = .39). Patients achieved these
endpoints of engraftment at medians of 12 and13days,
respectively. The infusion of these products was generally
well tolerated. Similarly, the cell concentration at which BM
cells were frozen did notpredict for theduration of granulo-
cyte ( P = .63) or platelet ( P = .36)aplasias for 54 patients
undergoing autologous BM transplantation. These data sug-
gest that PBSC or BM cells collected for transplantation may
be cryopreserved at very high cell concentrations without
loss of engraftment potential or undue infusion-relatedtox-
icity.
0 1994 by The American Society of Hematology.
tal effect of nucleated cell, platelet, or red cell concentrations
during cryopreservation could be shown. Furthermore, no
effect on engraftment kinetics could be determined.
MATERIALS AND METHODS
Patient selection and transplantation procedures. Patients eligi-
ble for PBSC transplantation underwent stem cell mobilization using
either G-CSF (Amgen Inc, Thousand Oaks, CA) or GM-CSF (Im-
munex Corp, Seattle, WA).5 G-CSF (16 pg/kg/d subcutaneously for
5 or 6 days) was administered during steady-state hematopoiesis
without chemotherapy rebound for most patients (patients with
unique patient numbers [UPNs] 7262, 7364, 7902, and 7923 listed
in Table 3 received G-CSF after cyclophosphamide or cyclophospha-
mide plus etoposide administration, with PBSC collection during
recovery from neutropenia). PBSC collections were performed for
either 3 or 4 days starting on the fourth day of G-CSF administration.
GM-CSF (250 pg/m2) was administered daily to a limited number
of patients treated for multiple myeloma during the recovery phase
from cyclophosphamide-induced marrow hypoplasia, with collection
of PBSC after PB white cell count exceeded 1 X 103/pL.All patients
underwent 12-L blood volume leukapheresis daily using a COBE
Spectra (COBE BCT, Lakewood, CO) as previously de~cribed.~ Dur-
ing apheresis, patients were anticoagulated with acid-citrate dextrose
formula A (ACD-A; Fenwal, Deerfield, IL) and heparin (5,000 U/
500 mL ACD-A). In addition, 20 to 40 mL of ACD-A was added
Fromthe Clinical Research Division, Fred Hutchinson Cancer
Research Center, Seattle, WA.
Submitted July 9, 1993: accepted December 23, 1993.
Supported in part by Grants No. CA47748 and CA15704 from
the National Cancer Institute, Bethesda, MD.
Address reprint requests to Scott D. Rowley, MD, FACP, Clinical
Cryobiology Laboratory, M227, Fred Hutchinson Cancer Research
Center, 1124 Columbia St, Seattle, WA 98104.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1994 by The American Society of Hematology.
0006-4971/94/8309-0014$3.00/0
2731
 From www.bloodjournal.org by guest on March 12, 2019. For personal use only.
2732
to the collection container before the start of the apheresis run to
decrease the risk of cell clumping. BM was collected under general
or regional anesthesia from the iliac crests using techniques pre-
viously described and without stimulation with G-CSF or other cyto-
kines.* Informed consent under Institutional Review Board-approved
protocols for collection and transplantation of BM and/or PBSC was
obtained for all patients.
Cryopreservation technique. Excess plasma was removed from
PBSC products by centrifugation in 600-mL blood-transfer packs in
a Sorvall RC-3 centrifuge (Du Pont CO, Wilmington, DE) at 3,000
rpm for 10 minutes. The volume of the residual cell pellet was
adjusted with autologous plasma as necessary, and 15- to 30-mL
aliquots were added to the number of freezing bags necessary. A
volume of cryoprotectant solution (see below) equal to the volume
of the cells was added and the cells cooled at 1"Clmin to -4O"C,
and then 10"C/min to -80°C using a rate-controlled freezer (Cryo-
med, New Baltimore, MD) before transfer into the vapor phase of
nitrogenat - 180°C or below.'BM cells were similarly cryopre-
served in a minimum of two bags (50 to 60 mL each). BM cells
were processed before cryopreservation by either collection of buffy-
coat cells or separation of light-density cells (specific gravity <
1.078 g/&) on Ficoll-Hypaque gradients (LSM, Organon Teknika,
Durham, NC) using a COBE 2991 BloodCellWasher (COBE
BCT).'"
Fresh cryopreservation solution consisting of 20% DMSO (Cryo-
serv, Research Industries, Salt Lake City, UT) and 40% autologous
plasma in TC199 (Gibco, Grand Island, NY) was prepared for each
PBSC or BM product. This was added to the cell products at equal
volume to achieve a final concentration of 10% DMSO and 20%
plasma. The cryoprotectant solution was generally chilled to4°C
before use, but the cells were not chilled before addition.
Before reinfusion, the cells were rapidly thawed in a 37°C water
bath at the patient's bedside. A volume of ACD-A equal to 20% of
the bag volume was added to prevent cell clumping. Samples for
analysis after thawing were obtained after the addition of ACD.
Each bag of cells was infused over 5 to IO minutes through a large
bore, intravenous catheter. All patients were hydrated and medicated
with diphenhydramine, mannitol, and hydrocortisone immediately
before cell infusion."
Cell counts. Nucleated cell and platelet counts, and hematocrits
were obtained for fresh PBSC products using a Sysmex E2500 (Toa.
Inc, Chicago, IL). Nucleated cell counts for BM products before
cryopreservation and for both BM and PBSC products after thawing
were obtained using a Coulter ZM (Coulter, Inc, E).The proportion
of mononuclear cells (defined as lymphocytes and monocytes) was
determined from 200-cell differential counts of Wright-stained speci-
mens.
Hematopoietic cell assays. Cells were cultured at 5 X IO4 cells/
mL in methylcellulose supplemented with Iscove's Modified Dul-
becco's Medium (IMDM; GIBCO, Grand Island, NY), 30% fetal
bovine serum (FBS;Hyclone, Logan, UT), 1% bovine serum albu-
min (Boehringer Mannheim Corp, Indianapolis, IN), molL 2-
mercaptoethanol (Sigma Chemical CO, St Louis, MO), molL
methylprednisolone sodium succinate (Upjohn CO,Kalamazoo, MI),
30 U/mL GM-CSF (Amgen), 100 U/mL interleukin-3 (Amgen), and
1 U/mL erythropoietin (Amgen). Erythroid burst-forming unit (BFU-
E) and myeloid (Cm-GM) colonies were identified after 14 days
of culture in a fully humidified 5% to 6% CO, in air atmosphere.
Samples (1 mL) of thawed cells were obtained after the adhtion of
ACD-A, serially diluted in 5 steps to 36 mL by addition of equal
volumes of phosphate-buffered saline containing 1% FBS, and
washed twice in IMDM before culture as described above. The total
quantity of progenitor cells before freezing was calculated from the
number of colonies enumerated, the seeding density of the culture
dishes, and the total number of cells frozen. The quantity of progeni-
tors cells after thawing was similarly calculated, but with adjustment
ROWLEY ET AL
for the recovery of the nucleated cefls after serial dilution and wash-
ing.
Quant@cation oj viable CD34+ cells. Redblood cells (RBCs)
from samples of cells obtained before cryopreservation or after thaw-
ing and serial dilution were removed by hypotonic lysis using ammo-
nium chloride. The cells were thenstained with a fluorescein isothio-
cyanate or phycoerythrin-conjugated CD34 antibody (8G 12; Becton
Dickinson, San Jose, CA) or an irrelevant isotype control, washed.
and counterstained with propidium iodide (PI; Becton Dickinson)
or 7-aminoactinomycin D (7AAD; Sigma Chemical CO). After an
additional wash and within 2 hours of staining, the proportions of
viable (PI or 7AAD excluding) cells, mononuclear cells, and CD34+
cells in the fresh and thawed specimens were determined byflow
cytometric analysis (FACScan, Becton Dickinson). CD34' cells
showed fluorescence greater than 99.8% of isotype control-stained
cells. The quantities of these cell populations in each product before
and after freezing were determined as described above.
Statistical analysis. The relation of the cell concentration during
cryopreservation to therecovery of nucleated cell recovery, mononu-
clear cell viability, and the recoveries of myeloid (Cm-GM), ery-
throid (BFU-E), and CD34' cells was evaluated by linear regression
analysis and calculation of Pearson's correlation coefficient. The
significance of the correlation parameters was tested by Student's 1-
test. The relationship of samples frozen simultaneously at two cell
concentrations was evaluated by the Wilcoxon signed-rank test. The
prognostic importance of cell concentration in predicting en-
graftment (censored for death) was assessed using the proportional
hazards regression model of COX.'^ For all evaluations, time refers
to the interval between cell infusion (day 0) and day of event (en-
graftment or death). No adjustments for multiple comparisons were
made in calculating the reported P values. For this reason, P values
between .01 and .05 should be viewed as suggestive and not conclu-
sive evidence of a difference.
RESULTS
Effect of cell concentrations during freezing on HSC re-
covery. To determine the effect of cell concentration during
cryopreservation on the recovery of hematopoietic progeni-
tor cells, we studied 108 PBSC products harvested from 30
patients (Table 1). We subsequently collated data on 57 of
these products from 22 patients after thawing. The average
PBSC collection contained 4.8 ? 3.4 X 10" cells (mean t-
SD; range, 0.6 to 14.9 X 10'") cryopreserved at an average
cell concentration of 3.7 ? 1.9 X IO' nucleated cells/mL
(range, 0.4 to 8.0 X 10'). Large quantities of platelets and
RBCs were also cryopreserved (Table 1). The nucleated cell
recovery after thawing was 75.4% t 13.0%. Nucleated cell
concentration during cryopreservation did not predict nucle-
ated cell recovery or mononuclear cell viability as deter-
mined by PI dye exclusion after thawing (Fig 1, A and B).
Although the cell concentration during freezing was border-
line (P = .06) and poorly( r = .29) predictive for the recovery
of CFU-GM progenitors after thawing (Fig ID), it did not
predict the recovery of viable CD34+ cells (Fig IC) or ery-
throid progenitors ( r = .17, P = .27, Fig not shown).
The proportion of mononuclear cells contained in these
products was determined by light microscopyandranged
from 10.5% to 100% of the nucleated cells. To determine if
the presence of granulocytes that predominately composed
the remainder of the nucleated cells of these products af-
fected the recovery of hematopoietic progenitors after cryo-
preservation, we similarly attempted to correlate, specifi-
cally, the concentration of mononuclear cells, and separately,
 From www.bloodjournal.org by guest on March 12, 2019. For personal use only.

PERIPHERAL BLOOD STEM CELL CRYOPRESERVATION 2733

Table 1. Cryopreservation Cell Concentrations volume contained in each freeze bag were the same. We
and Cell Recoveries After Thawing adjusted the cell concentration of PBSC products with autol-
Mean ? SD Range ogous plasma collected during apheresis to again maintain
identical concentrations of DMSO and plasma in each bag.
Cryopreserved"
Nucleated cell count ( x 10") 4.8 2 3.4 0.6-14.9 After thawing and anticoagulation with ACD-A, samples
Volume frozen (mL) 127 -C 45 34-300 were obtained from each bag for serial dilution and analysis.
Nucleated cell concentration ( x 108/mL) 3.7 i: 1.9 0.4-8.0 We found no consistent differences based on the cell concen-
Platelet concentration (X i09/mL) 2.9 -C 2.1 0.4-10.9 tration during freezing on the recovery of nucleated cells,
Hematocrit (%) 12.9 -C 7.2 2.8-44.7 mononuclear cell viability, or the recoveries of viable CD34+
Proportion mononuclear cells (%) 52.9 ir 27.2 10.5-100.0 cells or CFU-GM (Table 3). The recovery of BFU-E was
Proportion viable cells (%) 97.9 -C 1.4 91.2-99.2 lower for samples frozen at higher cell concentrations. This
No. viable CD34' cells ( x lo7) 9.3 -c 8.1 0.2-28.6 difference and a similar trend for the recovery of CFU-GM
NO. CFU-GM ( x lo7) 1.2 2 1.2 0.0-7.2 andCD34' cells may be artifacts of the dilution process
NO. BFU-E ( x lo7) 2.6 2 2.3 0.2-28.6
because samples frozen at the higher cell concentrations
Thawed*
were more likely to clump after the wash step, and the recov-
Recovery of nucleated cells (%) 15.4 2 13.0 43.2-102.8
Proportion viable cells (56)
ery of nucleated cells after washing was entered into the
All cells 63.6 t 15.2 31.2-85.0 calculation of progenitor cell recovery. CFU-GM-derived
Mononuclear cells 84.4 2 5.5 73.8-95.2 colonies per 5 X lo4 cells plated averaged 45.4 2 58.7
Recovery of viable CD34' cells (%l 71.4 I 51.9 8.9-226.5 (+-SD) for samples frozen at the higher concentrations and
Recovery of CFU-GM (%l 45.3 I 45.9 0.0-217.8 46.0 2 61.2 for samples frozen at the lower concentrations
Recovery of BFU-E (%) 48.8 ? 33.9 0.0-170.7 ( P = .88). Although the number of samples analyzed was
Shown are data from the cryopreservation of 108 products from limited, analysis of the five PBSC samples frozen at 6- to 24-
30 patients. Flow cytometric analysis for viability and CD34 phenotype fold differences in cell concentration showed no significant
analysis were available for 60 products, and progenitor cell assays difference in any of these parameters of HSC survival ( P >
were available for 67 products, before cryopreservation. After thaw- .l9 for all analyses).
ing, data were available for 57 products from 22 patients. The number Effect of cell concentration during cryopreservation on
of samples available after thawing for analysis of each parameter are engraftment kinetics. A total of 34 patients were trans-
shown in Fig 1. planted with PBSC alone. These patients were treated for
breast cancer (n = 14), non-Hodgkin's lymphoma (n = lo),
multiple myeloma (n = 5), and a varietyof other solid tumors
the concentration of granulocytic cells, with the recovery of
(n = 5). Although 1 patient underwent only two collections
hematopoietic progenitor cells after thawing. The concentra-
and 4 patients required six to nine collections in two series
tion of mononuclear cells during cryopreservation did not
of G-CSF mobilization to achieve adequate numbers of mo-
predict nucleated cell recovery ( r = -. 13, P = .33), mononu-
nonuclear cells for infusion, most patients underwent either
clear cell viability ( r = -.19, P = .28), or the recoveries of three (n = 22) or four (n = 7) aphereses while receiving G-
viable CD34+ cells ( r = -. 1l, P = .53), CFU-GM ( r = CSF. Patients were conditioned withbusulfanand cyclo-
-.l$ P = .33), or BFU-E ( r = -.13, P = .40). The concen- phosphamide with (n = 25) or without (n = 4) total body
tration of granulocytes also did not predict cell recovery ( r irradiation (TBI) with lung and liver shielding, or cyclophos-
= -.03, P = .80), mononuclear cell viability (r = .06, P =
phamide, TBI, and etoposide (n = 3). One patient each was
.72), or the recoveries of viable CD34' cells ( r = -.04, P conditioned with etoposide, BCNU, and cyclophosphamide,
= .82), CFU-GM ( r = .22, P = .17), or BFU-E ( r = -.lo, or etoposide, thiotepa, and cyclophosphamide. Six patients
P = S2). received G-CSF, 5 pgkgld, and another 6 patients received
We also explored the predictive value of platelet or RBC GM-CSF, 250 pg/m*/d, starting the day of PBSC infusion.
concentrations on these five parameters of progenitor cell The cell concentration during cryopreservation for the total
survival. Increasing platelet concentration very poorly pre- cells collected was calculated for each patient and averaged
dicted lower nucleated cell recovery ( r = .27, P = .04), but (+SD) 3.8 2 1.9 X lo8 nucleated cells/mL, with a range 0.2
was not predictive for mononuclear cell viability, or the X IO8 to 7.4 X lo8 cells/mL. These 34 patients reached
recoveries of viable CD34+ cells or erythroid or myeloid greater than 500 granulocytes/pL at a median of 12 days
progenitors (Table 2). The red cell content of the cryopreser- (range, 8 to 15 days), and platelet-transfusion independence
vation mixture was not predictive for any of these parameters at a medianof 13 days (range, 7 to 73 days). We assessed the
(Table 2). Despite the presence of large numbers of mature prognostic importance of cell cryopreservation in predicting
blood cells, including granulocytes, platelets, and RBCs, time to these two engraftment endpoints. The average cell
clumping after thawing was not observed in these products. concentration during cryopreservation did not predict either
To further explore the effect of cell concentration on he- time to achieving greater than 500 granulocytes/pL ( P =
matopoietic progenitor cell recovery after thawing, we stud- .51) or time to platelet-transfusion independence ( P = .40)
ieda limited number of BM or PBSC products split for in univariate analysis. Adjusting for diagnosis and growth
freezing at 2-fold or greater different cell concentrations. factor administration did not alter this conclusion. The lim-
The cell concentrations of the light-density cells isolated ited range in granulocyte aplasia duration also suggests that
from three BM collections were adjusted with tissue-culture cryopreservation of PBSC at these cell concentrations did
medium before the addition of the cryoprotectant solution not deleteriously affect the survival of cells responsible for
so that the final protein and DMSO concentrations, and the hematologic recovery after reinfusion.
 From www.bloodjournal.org by guest on March 12, 2019. For personal use only.

2734 ROWLEY ET AL

A. B

20 I 40
0 1 2 3 4 5 8 7 8 0 1 2 3 4 5 6 7 8
CELL CONCENTRATION CELL CONCENTRATION

C D

.
250 y=84.164.3&]

r=-0.11
250 -r y=SB.91-7.05x
,=.a 29
n=33 n=42
p=o54 1 p=0.00
200
Fig 1. Mononuclear cell via-

150 . . bility (B) and recoveries after

thawing of (A) nucleated cells,
= IC1 viable CD34'cells, and (D)
V
+l00
, .. ?l00 , CFU-GM. Correlation coeffi-

V
8
z : . . ... cients and the linear regression
equation are shownfor each
.. .
50 .mg*
.. 8

..m
.
curve. None of the slopes differ

0 '
I
, , I ;
.m
0, ,
,
,
; ..m
-, I., = -5;-
significantly from0, as shown by
the P values for each equation.
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8 Cell concentration
is x W per
CELL CONCENTRATION CELL CONCENTRATION mL.

Similarly, a total of 54 patients received cryopreserved cryopreserved for 30 patients; the other patients received
BM (without supplementation with PBSC). Light-density unpurged buffy-coat cells separated by centrifugation using
cells collected after density-gradient separation and immuno- a COBE 2991 Cell Washer. The patients were transplanted
logic purging using a panel of B-cell or T-cell directed mu- for the treatment of a variety of malignancies including non-
rine monoclonal antibodies and rabbit complement were Hodgkin's lymphoma (n = 24), Hodgkin's disease (n = 13),
acute lymphoblastic leukemia (n = 6), multiple myeloma (n
= 4), breast cancer (n = 4), and other solid tumors (n = 3).
Table 2. Effect of Platelets and Erythrocytes Most patients received hematopoietic cytokines (GM-CSF,
on Cryopreservation of PBSC G-CSF, or IL-3) after marrow infusion. The cell concentra-
r P n
tion at which the light-density cells were frozen ranged
from 1.03 X lo7to 3.46 X 10' nucleated cells/mL (median,
Predictive value of platelet concentration for
7.14 X IO'). Buffy-coat cells were frozen over a range of
cell survivals*
2.79 X IO7 to 1.96 X 10' nucleated cells/mL (median, 9.50
Mononuclear cell viability (%) - .09 .61 35
X lo7). These patients achieved greater than 500
Recovery of (%):
Nucleated cells -.27 .04 57 granulocytes/pL at medians of 12 days (range, 9 to 100
Viable CD34' cells -.l8 .32 33 days) and 15 days(range, 10 days to 36 days) for recipients
CFU-GM 25 .l1 42 of buffy-coat and light-density cells, respectively. Median
BFU-E .l8 .27 42 time to platelet-transfusion independence was 27 days
Predictive value of RBC concentration (range, 10 to 100 days) and 23.5 days (range, 5 to 278
(hematocrit) for cell survivals* days), respectively. The cell concentration at which these
Mononuclear cell viability (%) -.l6 .36 35 cells were frozen did not predict time to achieving greater
Recovery of (%l:
than 500 granulocytes/pL ( P = .63) or last platelet transfu-
Nucleated cells -20 .l3 57
sion ( P = .36) in univariate analysis stratified by initial
Viable CD34' cells .l0 .57 33
CFU-GM .06 .68 42
marrow processing. When a number of possibly clinically
BFU-E -.002 .99 42 relevant variable^,'^ including age, diagnosis, use of growth
factors, cryopreservation cell concentration, and PB counts
* Shown are the correlation coefficient (r)and significance of corre-
lation ( p ) for the effects of platelets and RBCs on various measures
on the day of marrow harvesting were entered into multivar-
of hematopoietic cell recovery after cryopreservation. Samples were iate analysis, only platelet count on day of harvesting (me-
obtained before and after cryopreservation from 57 products from 22 dian, 257 X 103/pL; range 26 to 800 X 103/pL) remained
patients. The numbers available for analysis of thevarious parameters prognostic for duration of granulocyte (P = .001) and plate-
are shown (nl. let aplasia ( P = .03).
 From www.bloodjournal.org by guest on March 12, 2019. For personal use only.

OOD PERIPHERAL STEM CELL CRYOPRESERVATION 2735

Table 3. Comparison of Simultaneous Freezingat Dtfaring Cell Concentrations on Cryopraaervation of BM or PBSC

Postthaw Recovery of 1%)
Cell Concentration Mononuclear Cell
(x 10' mL) Viability 1%) Nucleated Cells CFU-GM BFU-E CD34' Cells

UPN* Bag 1 Bag 2 Bag 1 Bag 2 Bag 1 Bag 2 Bag 1 Bag 2 Bag 1 Bag 2 Bag 1 Bag 2

704390.8 90.6
6.28 3.14 34.9 37.1 5.8 5.7 9.7 9.2 15.1 14.7
87.8
7055 91.1
0.74 0.37 92.4 86.5 27.7 20.9 21.0 19.6 179.9 113.5
711791.7 0.88
91.9 0.44 71.B 74.5 50.9 67.6 62.3 76.9 ND ND
7206 5.54 2.77 ND ND 73.0 106.7 4.8 12.2 13.3 24.5 ND ND
7281 4.68 2.52 ND ND 68.2 62.6 95.2 31.0 38.4 26.1 ND ND
729680.5 94.2
6.67 3.33 75.2 65.3 34.8 58.7 27.7 38.0 68.4 81.1
731693.0 13.60
61.4 6.80 36.9 42.8 180.3 196.4 43.4 58.9 34.1 75.4
733683.9 4.79
86.8 2.39 81.O 90.0 59.1 59.5 54.4 59.5 21.3 26.6
734192.8 7.06
91.6 3.53 68.7 76.4 17.7 40.9 35.7 52.9 85.5 137.8
737693.2 91.8
4.65 2.33 97.6 91.3 37.6 36.6 36.7 33.3 11.3 19.4
726299.4 6.68
99.4 0.89 91.9 91.1 43.2 46.0 60.5 60.6 82.4 73.8
736498.8 6.07
98.3 0.26 89.0 97.9 48.0 78.3 37.7 73.5 34.6 59.4
788397.8 7.70
98.0 0.39 90.1 89.7 5.1 14.5 14.6 14.4 63.2 85.1
790298.9 6.05
99.1 0.67 66.0 88.7 32.9 46.6 36.5 49.6 34.9 52.8
792398.3 2.15
98.6 0.37 104.3 106.8 77.5 69.7 54.7 65.0 53.5 64.7
Mean - - 91.B 90.5 76.1 80.0 48.0 52.3 36.4 44.1 57.0 67.0
SD - 9.8- 9.6 19.4 20.0 43.3 44.1 16.3 21.2 44.1 35.0
P - - .92 .25 1
.l .04 .09
___ ~~~

Abbreviations: ND, no data; SD, standard deviation.

* BM (UPNs 7043,7055, and 7117) or PBSCs (all other patients) were cryopreserved at twofold or up to 24-fold different cell concentrations.
After thawing, mononuclear cell viability and the recoveries of nucleated cells, viable CD34' cells, and hematopoietic progenitors (quantify after
thawindquantity before freezing X 100) were determined. Statistical evaluation of differences between bags 1 and 2 was performed using
Wilcoxon signed-rank test.

DISCUSSION in this study (total for three collections), resulting in infu-

This study found no consistent detrimental effects from the
cryopreservation of PBSC products at varying, but relatively
high cell concentrations. Similarly, the freeze concentration
of BM cells at varying, but lower cell concentrations was not
predictive for the kinetics of engraftment after transplantation. postthaw washing or infusion over several days would be
These findings are in agreement withthe previously described
uniform survival of murine spleen colony-forming unit( C m -
S ) cryopreserved over the range of 5 to 210 X IO6 cells/
m L . I 4 However, these conclusions are based primarily on the
recoveries of nucleated cells and hematopoietic progenitors
such as CD34+ cells or myeloidorerythroidprogenitors.
Likewise, the murine studies were limited to the detection of
CFU-S, not engraftment success after transplantation. In this
study, there was nodiscernible effect upon engraftment kinet-
ics or durability, however, suggesting a major effect on the
survival of primitive and committed hematopoieticstem cells
is unlikely. Also, only the effects of relatively high cell con-
centrations during cryopreservationwereinvestigated. The
previously described murinestudies found a significant deteri- taining less than 1 gof DMSO per kilogram recipient weight.
oration of CFU-S survival when marrow cells were cryopre-
served at concentrations less than 5 X IO6c e l l d d , a situation during infusion, infusion of these large quantities of cells
that may occur when CD34+ cells are highly enriched for
human transplantation, for e~amp1e.I~
A wide range of cryopreservation cell concentrations are
used by the various autologous transplant programs.' It had
been previously recommended that BM cells not be cryopre-
served at high cell concentrations, with 2 X lo7 nucleated
cells/mL suggested as a reasonable concentration.I6Thus, the
large quantities of PBSC after cytokine mobilization would
require cryopreservation in volumes of about 7 L for patients
sions of over 10 g of DMSO per kilogram of patient weight.
Although the lethal dose of DMSO for humans has not been
determined, the lethal dose for 50% of animals (LD50) is 3.1
to 9.2 &g for mice, and 2.5 g/kg for dogs.I7 Therefore,
required if such concentrations are used. Another practical
consideration that affects laboratory decisions is the desire
to split a product into two bags instead of one for freezing.
This would lower the cell concentration for some products,
but would primarily beof concern if small cell quantities
are being stored.
The infusions were generally well tolerated. Infusion-re-
lated toxicities have been reported by a number of cent er^.^.^
These complications appear volume related and could result
from the quantity of DMSO, the quantity of cells infused,
or both. Cryopreservation of PBSC at the cell concentrations
used in this study resulted in total volumes of products that
were usually less than 10 mLkg of recipient weight con-
Although we did not specifically quantitate patient symptoms
in relatively small volumes were generally well tolerated,
suggesting that the amount of DMSO, not the quantity of
cells, may be the primary cause of the previously reported
infusion-related toxicities. One patient, who concomitantly
received both BM (purged with murine antibodies and rabbit
complement) and PBSC, developed pulmonary decompensa-
tion requiring ventilatory support 4 hours after the infusion.
Bronchospasm is occasionally observed in patients receiving
immunologically purged BM cells at this center and others,"
 From www.bloodjournal.org by guest on March 12, 2019. For personal use only.
2736
although the severity of the reaction was probably greatly
increased by the large quantity of cells infused. A second
patient with preinfusion vascular instability became hypoxic
after each of two infusions, separated by several hours, of
168 and 262 mL (2.0 and 3.2 mL/kg). Subsequent infusions,
each separated by several hours on the following day, of 69,
138, and 141 mL were tolerated without complaint. Similar
events have not been observed in over 105 patients to date
with the infusion of PBSC alone, PBSC in combination with
unpurged BM, or PBSC infused on a different daythan
infusion of immunologically purged marrow cells.
A major concern about freezing large numbers of highly
concentratedcellswasthe risk of cellclumpingoccurring
immediately before freezingor after thawing. These products
required secondary centrifugation after collection to concen-
trate the cells for cryopreservation, and also contained large
and variable quantitiesof mature blood cells. The freeze pro-
cess was initiated within 1 hour of this secondary centrifuga-
tion and obvious cell clumping didnot occur. A single product
appeared gelatinous after concentration and additional ACD
(10% vol/vol) was added immediately to prevent clumping.
None of the products clumped after thawing, perhaps because
of the routine addition of ACD before infusion.
Since the development ofeffective cryoprotectants, many
aspects of BM and PBSC cryopreservation and storage have
not been defined. Current techniques appear to be adequate in
preserving sufficient quantitiesof HSC for successful reconsti-
tution of the recipient’s hematopoietic function after marrow-
lethal conditioning regimens. However, what may not be evi-
dent is the possible loss of HSC that may affect engraftment
kinetics. Improved cryopreservation techniques may improve
the acceptability of cell infusion to the patients who frequently
develop low-grade toxicities!,’ The data reported in this study
show that cell concentration is not a limitation when freezing
large quantities ofPBSCandBM cells. Cryopreservation at
high cellconcentrationminimizesthetotalproductvolume
tobe infused, and may decreasethe risk ofDMSO-related
complications.This finding may not extend to cryopreservation
atverylowcell concentrations (<5 X IO6 nucleatedcells/
m L ) , 1 4 , 1 5 and enrichment of HSC before cryopreservation may
require different freezing techniques.
ACKNOWLEDGMENT
We are indebted to Beth MacLeod. Kale Slechta, JoAnn Dalton,
David Yadock, and, especially, Daphne McDonald for their expertise
in processing these cell samples, CD34 analysis, hematopoietic pro-
genitor cell culture, and data collation.
REFERENCES
1. Rowley SD: Hematopoietic stem cell cryopreservation. A re-
view of current techniques. J Hematother 1:233, 1992
2. Korbling M, Holle R, Haas R, Knauf W, Dorken B, Ho AD,
Kuse R, Pralle H, Fleidner TM, Hunstein W: Autologous blood stem-
ROWLEY ET AL
cell transplantation in patients with advanced Hodgkin’s disease and
prior radiation to the pelvic site. J Clin Oncol 8:978, 1990
3. Haas R, HoAD, Bredthauer U, Cayeux S, Egerer G, Knauf
W, Hunstein W: Successful autologous transplantation of blood stem
cells mobilized with recombinant human granulocyte-macrophage
colony-stimulating factor. Exp Hematol I8:94, 1990
4. Sheridan WP, Begley CG, Juttner CA, Szer J, To LB. Maher
D, McGrath KM, Morstyn G, Fox RM: Effect of peripheral-blood
progenitor cells mobilized by Filgrastim (G-CSF) on platelet recov-
ery after high-dose chemotherapy. Lancet 1:640, 1992
5 . Bensinger W, Singer J, Appelbaum F, Lilleby K, Longin K,
Rowley S, Clarke E, Clift R, Hansen J, Shields T, Storb R, Weaver C,
Weiden P, Buckner CD: Autologous transplantation with peripheral
blood mononuclear cells collected after administration of recombi-
nant granulocyte stimulating factor. Blood 8 1:3158, 1993
6 . Davis JM, Rowley SD, Braine HG, Piantadosi S. Santos GW:
Clinical toxicity of cryopreserved bone marrow graft infusion. Blood
75:78 I , 1990
7 . Kessinger A, Schmit-Pokorny K, Smith D, Annitage J: Cryo-
preservation and infusion of autologous peripheral blood stem cells.
Bone Marrow Transplant 5:25, 1990 (suppl I )
8. Thomas ED, Storb R: Technique for human marrow grafting.
Blood 36507, 1970
9. Rowley SD, Byrne DV: Low temperature storage of bone mar-
row in nitrogen vapor-phase refrigerators. Transfusion 32:750, 1992
IO. Rowley SD, Davis JM, Piantadosi S, Jones M,Yeager AM,
Santos GW: Density-gradient separation of autologous bone marrow
grafts before ex vivo purging with 4-hydroperoxycyclophosphamide.
Bone Marrow Transplant 6:321, 1990
I I . Rowley SD: Techniques of bone marrow and stem cell cryo-
preservation and storage, in Sacher RA. AuBuchon JP (eds): Marrow
Transplantation: Practical Aspects of Stem Cell Reconstitution.
Bethesda, MD, Am Assoc of Blood Banks, 1992, p 105
12. Cox DR: Regression models and life-tables (with discussion).
J R Stat SOC[B] 34:187, 1972
13. Rowley SD, Piantadosi S, Marcellus DC, Jones RJ, Davidson
NE, Davis JM, Kennedy J, Wiley JM, Wingard JR, Yeager AM,
Santos GW: Analysis of factors predicting speed of hematologic
recovery after transplantation with 4-hydroperoxycyclophospha-
mide-purged autologous bone marrow grafts. Bone Marrow Trans-
plant 7:183, 1991
14. Dicke KA, Vellekoop L, Spitzer G, Zander AR, Vema DS,
Granati L, Litam J, Lafferty M: The role of autologous bone marrow
transplantation in neoplasia, in Okunewick JP, Merdith RE (eds):
Graft-versus-Leukemia in Manand Animal Models, Boca Raton,
FL, CRC, 1981, p 68
15. Berenson RJ, Bensinger WI, Hill RS, Andrews RG, Garcia-
Lopez J, Kalamasz DF, Still BJ, Spitzer G, Buckner CD, Bernstein
ID, Thomas ED: Engraftment after infusion of CD34’ marrow cells
in patients with breast cancer or neuroblastoma. Blood 77:1717,
1991
16. Gorin NC: Collection, manipulation and freezing of haemo-
poietic stem cells. Clin Haematol 15:19, 1986
17. Willhite CC, Katz PI: Dimethyl sulfoxide. J Appl Toxicol
4: 155, 1984
18. Pamphilon DH: The use of Campath monoclonal antibodies
for ex vivo treatment of bone marrow, in Gee AP (ed): Bone Marrow
Processing and Purging. A Practical Guide. Boca Raton, F L , CRC,
1991, p 231
 From www.bloodjournal.org by guest on March 12, 2019. For personal use only.

1994 83: 2731-2736

Effect of cell concentration on bone marrow and peripheral blood

stem cell cryopreservation
SD Rowley, WI Bensinger, TA Gooley and CD Buckner

Updated information and services can be found at:

http://www.bloodjournal.org/content/83/9/2731.full.html
Articles on similar topics can be found in the following Blood collections

Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests

Information about ordering reprints may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprints

Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml

Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.