TRANSFUSION MEDICINE
Safety and efficacy of cryopreserved platelets in
bleeding patients with thrombocytopenia
Sherrill J. Slichter,1,2 Larry J. Dumont,3,4 Jose A. Cancelas,5 MeLinh Jones,1 Terry B. Gernsheimer,2
Zbigniew M. Szczepiorkowski,3 Nancy M. Dunbar,3 Gautham Prakash,3 Stephen Medlin,6 Neeta Rugg,5
Bridget Kinne,6 Victor W. Macdonald,7 Greggory Housler,7 Manoj Valiyaveettil,7 Peter Hmel,8 and
Janet H. Ransom8
BACKGROUND: The short dating period of room
temperature–stored platelets (PLTs; 5-7 days) limits their
availability at far-forward combat facilities and at remote
civilian sites in the United States. PLT cryopreservation
in 6% DMSO and storage for up to 2 years may improve
timely availability for bleeding patients.
STUDY DESIGN AND METHODS: A dose escalation
trial of DMSO-cryopreserved PLTs (CPPs) compared to
standard liquid-stored PLTs (LSPs) was performed in
bleeding patients with thrombocytopenia. Within each of
four cohorts, six patients received escalating doses of
CPP (0.5 unit, 1 unit, and sequential transfusions of
2 and 3 units) and one received a LSP transfusion.
Patients were monitored for adverse events (AEs),
coagulation markers, PLT responses, and hemostatic
efficacy.
RESULTS: Patients with a World Health Organization
bleeding score of 2 or more received from 0.5 to 3 units
of CPP (n = 24) or 1 unit of LSP (n = 4). There were no
related thrombotic or other serious AEs experienced.
Mild transfusion-related AEs of chills and fever (n = 1),
transient increased respiratory rate (n = 1), DMSO-
related skin odor (n = 2), and headache (n = 1) were
observed after CPP transfusion. Among CPP recipients
14 of 24 (58%) had improved bleeding scores, including
three of seven (43%) patients who had intracerebral
bleeding. CPP posttransfusion PLT increments were
significantly less than those of LSPs; however, days to
next transfusion were the same. After transfusion, the
CPP recipients had improvements in some variables of
thrombin generation tests and thromboelastography.
CONCLUSION: Cryopreserved PLT transfusions
appear to be safe and effective when given to bleeding
patients with thrombocytopenia.
T
he short shelf life of room temperature–stored
platelets (PLTs; 5–7 days) severely limits the avail-
ability of PLT transfusion for wounded soldiers at
forward medical facilities. In response to the
ABBREVIATIONS: AE(s) = adverse event(s); AP = apheresis; aPTT =
activated partial thromboplastin time; AT = antithrombin; CNS =
central nervous system; CPP(s) = cryopreserved platelet(s); DIC =
disseminated intravascular coagulation; ECG = electrocardiogram;
F1+2 = prothrombin fragments 1+2; LSP(s) = liquid-stored platelet(s);
MA = maximum amplitude; PT = prothrombin time; TAT = thrombin
antithrombin; TEAE(s) = treatment-emergent adverse event(s); TEG =
thromboelastography; TGT(s) = thrombin generation test(s).
From the 1Research Institute, Bloodworks Northwest; and the
2
University of Washington School of Medicine, Seattle, Washington;
3
Geisel School of Medicine at Dartmouth and Dartmouth-Hitchcock
Medical Center, Lebanon, New Hampshire; 4Blood Systems
Research Institute, Denver, Colorado; 5Hoxworth Blood Center,
University of Cincinnati; and the 6University of Cincinnati Health
Hospital, Cincinnati, Ohio; 7U.S. Army Medical Research and
Materiel Command, Fort Detrick, Maryland; and 8Fast-Track
Drugs & Biologics, LLC, North Potomac, Maryland.
Address reprint requests to: Larry J. Dumont, Blood Systems
Research Institute, 717 Yosemite Street, Denver, CO 80230; e-mail:
LDumont@bloodsystems.org.
Funding for this research study was provided by contracts
W81XWH-15-C-0047 and W81XWH-13-C-016 from the U.S. Army
Medical Research and Materiel Command. The views, opinions,
and/or findings contained in this report are those of the author(s) and
should not be construed as an official Department of Defense position,
policy, or decision unless so designated by other documentation.
ClinicalTrials.gov Identifier: NCT02078284.
Received for publication March 11, 2018; and accepted March
30, 2018.
doi:10.1111/trf.14780
© 2018 AABB
TRANSFUSION 2018;58;2129–2138
Volume 58, September 2018 TRANSFUSION 2129
SLICHTER ET AL.
military’s need for extended stored PLTs, Dr Robert Valeri,
working at the Naval Blood Research laboratory in Boston,
developed methods of cryopreserving PLTs utilizing DMSO as a
cryoprotectant.1 The best evidence of the hemostatic effective-
ness of cryopreserved PLTs (CPPs) was a randomized trial of
standard versus CPPs during cardiopulmonary bypass surgery.2
Patients transfused with CPPs required fewer blood products
and had reduced postoperative blood loss. Dr Charles Schiffer
also used a similar method for cryopreservation of autologous
PLTs to support the transfusion needs of hematology-oncology
patients who had developed an alloimmunization response to
transfused donor PLTs.3–5 However, CPPs have never been
licensed in the United States.
Despite the presence of the DMSO cryoprotectant, freez-
ing and thawing of PLTs results in a number of changes
related to damage and activation of at least a portion of the
PLTs including production of PLT microparticles, greater
thrombin generation in an in vitro assay system, 25% to 30%
loss of PLT content, increased P-selectin expression, and
increased expression of phosphatidylserine on the PLT outer
membrane.6,7 The resulting in vitro procoagulant phenotype
is a concern when assessing the in vivo safety of CPPs. In a
recent literature review, we did not find even a single
reported case of thromboembolic complications after the
transfusion of more than 3000 CPP units to 1334 patients.8 In
recent studies of radiolabeled autologous CPPs in 32 healthy
subjects studied in three different laboratories, PLT recover-
ies were reduced (33 ± 10%) and the in vivo survivals were
shorter (7.5 ± 1.2 days) compared with fresh PLT-rich
plasma-derived PLTs (63 ± 9% and 8.6 ± 1.1 days) or histori-
cally observed for standard-of-care liquid-stored PLTs.8,9
The primary objective of this study was to evaluate the
safety of CPPs in hematology-oncology patients with throm-
bocytopenia with active bleeding. Secondary objectives were
to gather additional in vitro descriptive data of the PLT
products and to obtain preliminary efficacy data.
MATERIALS AND METHODS
Study design
This trial was a multicenter, open-label, randomized Phase
1 dose escalation study to evaluate the safety and efficacy of
the Department of Defense’s formulation of CPP after trans-
fusion to actively bleeding patients with thrombocytopenia.
Four sequential cohorts of patients received escalating
doses of CPP transfusions starting with 0.5, 1, 2, and 3 units
each given to six patients with an additional patient in each
cohort receiving 1 unit of liquid-stored PLTs (LSPs) for a
total of 28 patients. Patient assignment to test or control
PLTs was centrally randomized. Safety and efficacy were
evaluated for up to 6 days after study PLT transfusions;
however, mortality was followed for 30 days. The study was
performed at three sites: Bloodworks Northwest and Univer-
sity of Washington Hospital (Seattle, WA); Dartmouth-
2130 TRANSFUSION Volume 58, September 2018
Hitchcock Medical Center (Lebanon, NH); and Hoxworth
Blood Center and University Hospital University of Cincin-
nati (Cincinnati, OH).
Study population
Potential study subjects were hospitalized hematology-
oncology patients more than 17 years of age with thrombo-
cytopenia because of their underlying disease or its
treatment with active World Health Organization (WHO)
Grade 2 or greater bleeding score. Generally, WHO Grade
2 bleeding is any gross organ system bleeding, WHO Grade
3 bleeding is severe enough to require red blood cell (RBC)
transfusion(s), and WHO Grade 4 is life-threatening bleed-
ing. Underlying diagnoses eligible for enrollment were as
follows: any type of leukemia, myeloma, myelofibrosis, mye-
lodysplasia, aplastic anemia, hematopoietic or nonhemato-
poietic solid tumor, and chemotherapy or radiation
therapy–induced marrow aplasia or hypoplasia or any type
of hematopoietic stem cell transplant. Subjects were
excluded if they had evidence of acute or chronic dissemi-
nated intravascular coagulation (DIC) by clinical laboratory
values (D-dimer greater than 8 μg/mL and fibrinogen < 100
mg/dL), receiving treatment with anti-PLT drugs and/or
full anticoagulation therapy, a history or diagnosis of unpro-
voked thrombotic events, immune thrombocytopenia,
thrombotic thrombocytopenic purpura, hemolytic uremic
syndrome, or venoocclusive disease. All patients provided
written informed consent.
PLT product sources and manufacturing
Cryopreserved PLTs were prepared by the Dartmouth-
Hitchcock Medical Center Transfusion Medicine Research
Laboratory as described previously.9 Briefly, 27% DMSO
(Lonza) was aseptically added to apheresis (AP; ≥ 3.0 × 1011
PLTs in 165-375 mL of plasma) within 57 hours of collection
to a final DMSO concentration of 6% (5.6%-6.7%). AP in 6%
DMSO were concentrated to 20 to 35 mL by centrifugation
in 500 mL ethyl vinyl acetate freezing bags (OriGen Biomed-
ical), placed in a cardboard plasma freezing container, fro-
zen in a chest-type mechanical freezer set at –80 C, and
held frozen at not more than –65 C for up to 2 years. CPPs
were shipped on dry ice in a validated container to Blood-
works Northwest and Hoxworth Blood Center as needed.
Product condition and temperature history for each unit
were verified before release for transfusion. Each site
thawed ABO-matched CPP unit(s) at 37 C for approximately
8 minutes, resuspended the PLTs in 25 mL of sterile saline,
and transfused within 4 hours of thawing on Study Day
1. Standard-of-care in-process leukoreduced LSPs were AP
in plasma collected using an apheresis system (Trima, Ter-
umo BCT, Inc.) with ACD-A after the manufacturer’s
instructions, contained at least 3 × 1011 PLTs, and were irra-
diated with 25 or 30 Gy and stored at 20 to 24 C for no
more than 5 days.

Methods
Liquid-stored PLTs and CPPs were administered by a single
intravenous (IV) infusion through a central line on Study
Day 1 followed by approximately 100 mL of saline rinse.
Multiple units of CPP were infused immediately one after
the other. Volume differences between the products did not
allow blinding of randomization assignment to patients or
their medical staff. After a study transfusion, patients were
followed for evidence of transfusion reactions, thrombotic
events, other adverse events (AEs), coagulation-related vari-
ables, and PLT efficacy endpoints.
Clinical assessments were performed daily at approxi-
mately the same time by a member of the site’s trial staff
and included vital signs; a general physical examination for
evidence of bleeding or thrombotic event; patient or family
member reports of bleeding signs or symptoms; and chart
review for bleeding, thrombotic events, or other treatment-
emergent AEs (TEAEs). All TEAEs were evaluated by the site
investigator for severity (mild, moderate, or severe) and
relationship to the study transfusion (related, possibly
related, or not related). All safety data were reviewed by an
independent medical monitor and a data monitoring com-
mittee before escalation to the next higher dose cohort.
Before and after the transfusion on Study Day 1 and on
Study Day 2, patients were tested for coagulation markers
including fibrinogen, D-dimer, prothrombin fragments 1+2
(F1+2), thrombin antithrombin (TAT), antithrombin (AT),
prothrombin time (PT), activated partial thromboplastin
time (aPTT), thrombin generation test (TGT; calibrated
automated thrombogram, Diagnostica Stago, Inc.), and
thromboelastography (TEG; Thrombelastograph 5000, Hae-
monetics Corp.). Serum troponin was determined on Study
Day 1 before the infusion and on Day 3. A 12-lead electro-
cardiogram (ECG) was performed on Study Day 1 before
the infusion and on Days 2 and 3. Transfusion efficacy was
evaluated by assessing spontaneous bleeding and grading
based on WHO criteria,10 PLT counts to determine count
increments (CIs) and corrected count increments (CCIs),
and the number of days until the next PLT transfusion (days
to next transfusion). On Study Days 1 through 6, patients
were assessed for TEAEs, spontaneous bleeding, vital signs,
PLT counts, and thrombotic events. A patient was consid-
ered an evaluable subject for safety assessments if he or she
received a study infusion and completed study-related Day
3 procedures although the full follow-up period was 6 days.
Criteria for evaluating transfusion outcomes
Primary safety endpoints included all TEAEs, signs or symp-
toms of thrombotic events including ECG changes, and
changes in the following posttransfusion variables: vital
signs, coagulation (fibrinogen, D-dimer, F1+2, TAT, AT, PT,
and aPTT), chemistry (serum creatinine, lactate dehydroge-
nase [LDH], and troponin), and hematology (hematocrit
[Hct], hemoglobin [Hb], and PLT counts). Efficacy endpoints
SAFETY OF CRYOPRESERVED PLTs
included spontaneous bleeding rates and severity scores,
PLT CI and CCI, days to next transfusion, TGT, and TEG.
The study PLTs were characterized in vitro on Study Day
1 by pH, PLT concentration, volume, P-selectin, microparti-
cles, phosphatidylserine expression, mean PLT volume, and
swirling.
Statistical analyses
Descriptive statistics are used to present the study data. Dis-
crete variables are presented as number of observations and
percentages. Continuous variables are given as means, stan-
dard deviations (SDs), median, and range. Coefficients of
variation are also presented for all quantitative PLT vari-
ables assessed by in vitro assays of a pretransfusion aliquot
of the CPP and LSP products. For summary descriptive sta-
tistics, missing data are represented by counts and were
treated as missing at random, and no adjustments were
made. All statistical analyses were performed with computer
software (SAS, Version 9.2, SAS Institute). A post hoc data
analysis was done using a mixed-effect analysis of variance
(ANOVA) with repeated subject measures and a random
intercept stratified by product type (Proc Mixed, SAS 9.4).
RESULTS
Patient recruitment and baseline characteristics
Twenty-eight eligible patients were randomized, received a
study PLT transfusion, and completed the study over a
period of 21 months (Fig. 1): 39.3% were female and 60.7%
were male, 75% were Caucasian, and the mean ± SD age was
51.9 ± 15.8 years (Table 1). One exception to the randomiza-
tion scheme was in Cohort 1 where one patient who was ran-
domly assigned to receive 0.5 units of CPPs was inadvertently
transfused with one entire CPP unit. All patients had WHO
Grade 2 or greater bleeding at the time of the study PLT
transfusion, and half had severe or life-threatening WHO
Grade 3 (n = 7) or Grade 4 (n = 7) bleeding. Of note is that
16 of 28 patients (57%) were either alloimmunized (human
leukocyte antigen [HLA] panel-reactive antibodies of > 20%;
n = 7) and/or were otherwise clinically refractory to standard
PLT transfusions (CI < 11 × 109/L after two consecutive
transfusions; n = 9).11
Characteristics of transfused PLTs
Forty-two units of CPP were thawed and transfused with a
mean (±SD) PLT content per unit of 2.4 × 1011 ± 0.4 × 1011
PLTs (range, 1.8 × 1011-3.4 × 1011) reflecting an expected
20% to 25% decrease in PLT counts after the freeze-thaw pro-
cess. The mean storage time of the transfused CPPs at no
warmer than –65 C was 312 days (range, 89-621 days). The
four LSP units had PLT counts of 3.8 × 1011 ± 0.5 × 1011
PLTs/transfusion (range, 3.3 × 1011-4.3 × 1011) with a storage
time of 2 to 5 days at 20 to 24 C. All study PLTs had pH 22 C
of more than 6.5 before transfusion. CPPs were more
Volume 58, September 2018 TRANSFUSION 2131
SLICHTER ET AL.
Fig. 1. Patient disposition. Fifty-nine patients were assessed for eligibility, and 28 were randomly assigned and received a PLT
transfusion. *One patient was randomly assigned to receive 0.5 units of CPPs but inadvertently received 1 unit. This patient’s data were
analyzed with the 1-unit CPP group. There were no early withdrawals.
activated than LSPs showing higher expression of PLT-
associated P-selectin and phosphatidylserine with more PLT
microparticles (Fig. 2).
Transfusion results
CI showed a dose response with CPP transfusions from
1.20 × 109 ± 3.10 × 109 PLTs/L for 0.5-unit
13.2 × 109 ± 6.38 × 109 PLTs/L for 3-unit transfusions.
When normalized for the number of PLTs transfused, there
was no difference in CCI for different CPP doses with an
average of 3.0 m2/μL per 1011 PLTs (Fig. 3). Both CI and
CCI were much greater for LSP transfusions when com-
pared to even a 3-unit transfusion of CPP.
Posttransfusion PLT survival is approximated by days
to next transfusion. Consistent with the previously reported
long life span of radiolabeled autologous CPPs ranging from
6.7 to 8.0 days,8,9 there is no difference in days to next trans-
fusion for CPP and LSP transfusions in spite of the lower CI
with CPPs (Table 2).8,9
AEs
There were no thrombotic events considered related to
study PLTs after any study transfusion. Eleven serious AEs
were reported in five patients who received CPPs, and all
were considered related to worsening of their underlying
clinical condition and/or its treatment. Four patients devel-
oped sepsis; one case was classified as life-threatening and
three cases were related to the patients’ deaths. Of these
three fatal cases, one patient who received 0.5 units of CPPs
died 30 days later of septic shock; one patient who received
1 unit of CPPs developed a worsening lung infection, adult
respiratory distress syndrome, and worsening sepsis that
resulted in death 10 days after study transfusion; and one
patient who received 3 units of CPPs developed sepsis, with
2132 TRANSFUSION Volume 58, September 2018
worsening thrombocytopenia and stroke, and died of car-
diac arrest 6 days after the study transfusion. One patient
developed an abnormal ECG with prolonged QT interval
due to a medication known to prolong the QT interval
(Dasatinib) that resolved after drug discontinuation.
Treatment-emergent AEs considered related to CPP
transfusion occurred in three patients. One patient had mild
to
chills and fever after receiving 1 unit of CPPs, two patients
had a DMSO skin odor after 0.5 or 3 units of CPPs, and one
patient had a moderate headache the day after a 3-unit CPP
transfusion. One patient had a severe reaction with chills,
rigors, wheezing, and tachypnea after a nonstudy PLT trans-
fusion given 5 days after the patient received 3 units of CPPs.
This reaction was not considered related to his prior CPP
transfusion. There were no TEAEs related to a LSP transfu-
sion (see Table S1 [available as supporting information in the
online version of this paper] for a full summary of AEs).
Clinical bleeding
Bleeding assessed clinically before and after each patient’s
study transfusion is summarized in Table 3. By the end of
Study Day 2, 14 of 24 patients (58%) had improved bleeding
status after a CPP transfusion, and two of four patients
(50%) improved after a LSP transfusion.
Among the CPP recipients, three patients with WHO
Grade 3 or 4 bleeding had remarkable clinical improve-
ments after a CPP transfusion. One patient with Grade
4 central nervous system (CNS) bleeding had altered mental
status, occasional garbled or slurred speech, photophobia,
and headache and was not ambulatory. After receipt of
1 unit of CPPs, all neurologic symptoms resolved within
24 hours, repeat cranial computerized tomography scan at
24 hours showed no progression of the hematoma, and the
 TABLE 1. Patient demographics and medical diagnosis*
All patients
Characteristic 0.5-unit CPP (n = 5) 1-unit CPP (n = 7) 2-unit CPP (n = 6) 3-unit CPP (n = 6) LSP (n = 4) (n = 28)
Sex
Female 4 (80.0) 2 (28.6) 3 (50.0) 1 (16.7) 1 (25.0) 11 (39.3)
Male 1 (20.0) 5 (71.4) 3 (50.0) 5 (83.3) 3 (75.0) 17 (60.7)
Race
American Indian or Alaskan Native 1 (25.0) 1 (3.6)
Black or African American 1 (14.3) 1 (16.7) 2 (7.1)
Native Hawaiian or other Pacific Islander 1 (16.7) 1 (3.6)
Not available 1 (16.7) 1 (3.6)
Other 1 (20) 1 (25.0) 2 (7.1)
White 4 (80) 6 (85.7) 4 (66.7) 5 (83.3) 2 (50) 21 (75.0)
Ethnicity
Not Hispanic or Latino 5 (100) 7 (100) 5 (83.3) 6 (100.0) 4 (100) 27 (96.4)
Not available 1 (16.7) 1 (3.6)
Age (year) 46 ± 17.3δ 57.7 ± 18.6 52.3 ± 15.8 53.3 ± 13 46.3 ± 16.2 51.9 ± 15.8
(28-65) (20-71) (33-75) (30-64) (27-62) (20-75)
Height (cm) 171 ± 2.77 172 ± 8.55 176 ± 11.7 180 ± 11.6 177 ± 13.4 175.2 ± 10
(168-175) (161-185) (161-193) (160-193) (160-190) (160-193)
Weight (kg) 91.4 ± 15.3 87.2 ± 23.9 87.7 ± 28.3 97.8 ± 16.4 81.1 ± 11.4 89.5 ± 20.1
(72.6-111.9) (61.6-119.2) (53.5-127) (76.6-113) (70.8-97.5) (53.5-127)
Diagnosis
AML 3 (1 PBSC) 2 1
ALL 2 (1 cord) 2 (1 PBSC) 1 CAR-T
Myelofibrosis 2 (1 cord, 1 PBSC)
Myeloma 1 (autoPBSC)
CML 1
NHL 1 PBSC
ANLL 3 (1 PBSC, 1 BMT) 3 (1 PBSC) 2 (1 PBSC, 1 autoPBSC) 2 (1 PBSC)
Aplastic anemia 1
* Data are reported as number (%), mean ± SD (range), or number (number transplanted).
ALL = acute lymphocytic leukemia; AML = acute myelogenous leukemia; ANLL = acute nonlymphocytic leukemia; autoPBSC = autologous peripheral blood stem cell transplant; BMT = allogeneic
bone marrow transplant; CAR-T = CAR-T cell infusion; CML = chronic myelogenous leukemia; cord = cord blood allogeneic hematopoietic progenitor cell transplant; NHL = non-Hodgkin’s
lymphoma; PBSC = allogeneic peripheral blood stem cell transplant.

Volume 58, September 2018 TRANSFUSION 2133

SAFETY OF CRYOPRESERVED PLTs
SLICHTER ET AL.

Fig. 2. Transfused PLT characteristics. Displayed are the percentages of PLTs in the unit (□, CPP units; ▪, LSP units) expressing P-selectin,
phosphatidylserine, and the percentage of microparticles. Also given are the number of PLTs/unit × 1011. The data are given as the mean ± 1 SD.

Fig. 3. Posttransfusion PLT responses. PLT CI (□) is the difference in posttransfusion minus pretransfusion patient PLT counts. CCI ( ) ▪
is the CI normalized for patient body surface area and the number of PLTs transfused; that is, CCI = (PLT increment [103/μL] × body
surface area/m2)/number of PLTs transfused × 1011. Data are given as the mean ± 1 SD.

patient was discharged from the hospital 5 days after
transfusion.
Another patient with Grade 4 CNS bleeding based on
computerized tomography scan and confirmed by magnetic
resonance image had intermittent headaches that worsened
with coughing. The headaches were controlled with four
doses of acetaminophen the day before the study transfu-
sion. There were mild intermittent episodes of headache
after study transfusion and reduction of acetaminophen to
only one dose daily up to Study Day 3 and no complaints of
headaches at the outpatient visit on Study Day 4. It was
concluded based on improvements in the severity and fre-
quency of his headaches and decreased need for high-dose
acetaminophen that CNS bleeding had probably resolved.
Furthermore, he was released from the hospital 2 days after
the CPP transfusion.
Another patient with WHO Grade 3 bleeding based on
a hemothorax that required RBC transfusions had no chest

2134 TRANSFUSION Volume 58, September 2018

 SAFETY OF CRYOPRESERVED PLTs

TABLE 2. Days to next PLT transfusion

CPP units LSP
Parameter 0.5 1 2 3 1
Number of observations 5 6* 5* 5* 3*
Median (days) 1 1.5 1.0 2.0 1.3
Range 0-1 1-2 0-2 1-3 1-2
* One patient in each of these groups did not have any additional PLT transfusions.

TABLE 3. Improvements in posttransfusion bleeding

Pretransfusion WHO bleeding grade
Transfusion dose Grade 2* Grade 3* Grade 4* Total
0.5-unit CPP 1/1 (100%) 2/4 (50%) 3/5 (60%)
1-unit CPP 2/4 (50%) 1/1 (100%) 1/2 (50%) 4/7 (57%)
2-unit CPP 2/3 (66%) 1/1 (100%) 1/2 (50%) 4/6 (66%)
3-unit CPP 2/3 (66%) 1/3 (33%) 3/6 (50%)
Total 7/11 (64%) 4/6 (66%) 3/7 (43%) 14/24 (58%)

1-unit LSP 1/3 (33%) 1/1 (100%) 2/4 (50%)

* Overall, in 14 of 24 (58.3%) CPP-transfused patients, bleeding improved including three of seven (43%) patients with CNS bleeding.

tube drainage for 6 hours (a fourfold decrease in his total
chest drainage from the previous day) immediately after
receiving 2 units of CPPs. No patient had bleeding worsen
after receiving a study PLT transfusion. Interestingly, there
was no evidence of a dose response based on bleeding
reduction and the number of CPP units transfused.
Clinical laboratory assessments
PT, aPTT, fibrinogen, and D-dimer
All patients entered into this study had evidence of coagulo-
pathy of disease and/or its treatment; PT and aPTT clotting
times were increased, D-dimers were elevated, and fibrino-
gen was reduced in some patients; however, DIC was ruled
out in all patients. Mean levels of PT and aPTT tended to be
above the upper limit of normal but generally stable over
the course of the study (Table S2 and S3, respectively, avail-
able as supporting information in the online version of this
paper). Fibrinogen, an acute-phase reactant, was elevated in
most study subjects, but remained stable throughout the
study period, therefore presenting no evidence of a signifi-
cant consumptive process such as DIC (Table S4, available
as supporting information in the online version of this
paper). D-dimer, a fibrin breakdown product, was also ele-
vated in these patients consistent with an active breakdown
of fibrin. D-dimer was stable over the course of the study
with only small changes observed (Table S5, available as
supporting information in the online version of this paper).
AT, TAT, and F1+2
Median levels of AT tended to be toward or below the low
end of the normal ranges in all patients before the study PLT
transfusion (Table S6, available as supporting information in
the online version of this paper). Mean changes from base-
line on the day after the study transfusion were small and
insignificant (p > 0.05). As thrombin is formed, TAT
increases are expected as part of the in vivo coagulation con-
trol mechanisms. Mean TAT levels were above the laboratory
reference range before the study PLT transfusion. TAT
increased after study PLT transfusion in the 1-unit CPP group
(p < 0.05) and the 3-unit CPP group (p > 0.10) but decreased
in the 2-unit CPP and LSP study groups (Table S7, available
as supporting information in the online version of this
paper). TAT levels were highly variable in the study patients.
As with D-dimer, F1+2 was frequently elevated in the study
subjects before the first study PLT transfusion (Table S8, avail-
able as supporting information in the online version of this
paper). Mean and median F1+2 changes from baseline were
highly variable between patients (SD up to 821 pmol/L for Day
2 data for the 0.5-unit CPP group). In this study, we observed
that the levels of F1+2 and TAT were correlated (r = 0.863,
p < 0.0001). After CPP transfusions, there was a consistent
increase in both F1+2 and TAT, although this only reached sig-
nificance in the 1-unit CPP dose cohort and did not show a
dose–response relationship. AT levels were not consistently cor-
related with F1+2 or TAT concentrations across all patients.
With the small LSP cohort (n = 4), there were no significant
changes in either F1+2 or TAT after transfusion.
Other clinical laboratory tests
Other clinical laboratory outcomes of creatinine, troponin,
LDH, Hct, and Hb are shown in Tables S9 through S13
(available as supporting information in the online version of
this paper). There were no clinically relevant changes in any
of these assays from baseline values.

Volume 58, September 2018 TRANSFUSION 2135

SLICHTER ET AL.

Fig. 4. Differences in TGT test variables before to after transfusion. Height of the bars represents the mean ± 1 SD. Data were analyzed
using a mixed-effect ANOVA with repeated subject measures and a random intercept stratified by product type (Proc Mixed, SAS 9.4).
Overall, the pre- to posttransfusion effects for peak thrombin were significant for 2- and 3-unit CPP transfusions (p = 0.005 and
p = 0.002, respectively), and time to peak was close to significantly shortened (p = 0.058) for 3-unit transfusions. ( ) Day 1, before
transfusion; ( ) Day 1, after transfusion; ( ) Day 2.

Ex vivo PLT function tests TEG

TGTs
Patients generally had reduced to absent in vitro throm-
bin generation in the TGT assay pretransfusion; extended
lag time, reduced peak thrombin, and extended time to
peak (Fig. 4 and Table S14, available as supporting infor-
mation in the online version of this paper). Endogenous
thrombin potential was not consistently reported since
the end of reactions did not always converge permitting
an area integration of the data. One set of curves for one
patient who received 1 unit of CPPs is shown to illustrate
the in vitro thrombin generation response before transfu-
sion, after transfusion, and the day after transfusion (Day
2; Fig. S1, available as supporting information in the
online version of this paper). Overall, the pre- to post-
transfusion effect for peak thrombin were significant for
2- and 3-unit CPP transfusions (p < 0.005), and time to
peak tended to be shortened for 3-unit transfusions
(p = 0.06). For 1-unit LSP transfusions, no significant
effects were seen in this small cohort (n = 4) for any TGT
variables before to after transfusion. The thrombin gener-
ation response, while partially corrected after transfusion,
does not approach those observed in healthy controls
using test conditions of a PLT concentration of approxi-
mately 100 × 109 PLTs/L per test in PLT-rich plasma
(data not shown).
Patients generally had reduced response to kaolin activation
before transfusion as indicated by extended R-time, reduced
angle, and reduced maximum amplitude (MA; Fig. 5 and
Table S15, available as supporting information in the online
version of this paper). Very little to no lysis at 30 minutes
was observed before transfusion, after transfusion, or on
Day 2. Improvements were generally observed in the TEG
variables in response to CPP transfusions (reduced R-time,
increased angle, and increased MA), although these effects
did not always persist through the next day.
Overall, the pre- to posttransfusion effect for R-time
was significant for 0.5- and 1-unit CPP transfusions
(p < 0.05). MA increased after transfusion, reaching signifi-
cance for 3-unit transfusions and was similar to the increase
observed with LSP 1-unit transfusions. MA was generally
maintained through Day 2 observations. For 1-unit LSP
transfusions, no significant effects were seen in this small
cohort (n = 4) for any TEG variables except for MA that was
significantly increased after transfusion (p = 0.01).
DISCUSSION
We report the results of a dose escalation safety and efficacy
study comparing 0.5, 1, 2, and 3 units of CPPs given sequen-
tially to 1 unit of LSPs in bleeding hematology-oncology

2136 TRANSFUSION Volume 58, September 2018

 SAFETY OF CRYOPRESERVED PLTs

Fig. 5. Differences in TEG test variables before to after transfusion. Height of the bars represents the mean ± 1 SD. Data were analyzed
using a mixed-effect ANOVA with repeated subject measures and a random intercept stratified by product type (Proc Mixed, SAS 9.4).
Overall, the before to after transfusion effect for R-time was significant for 0.5-unit CPP (p = 0.048) and close to significant for 1-unit
CPP (p = 0.052) transfusions. MA increased after transfusion, reaching significance (p = 0.021) for 3-unit CPP transfusions, and was
similar to the increase observed with LSP 1-unit transfusions that was also significant (p = 0.011) after transfusion. MA is the maximum
strength or stiffness of the developed clot. R-time is the reaction time and reflects initial fibrin formation. ( ) Day 1, before transfusion;
( ) Day 1, after transfusion; ( ) Day 2.

patients with thrombocytopenia. There were four cohorts
with six patients per cohort given escalating doses of CPP
and one patient per cohort given LSP. Patients were fol-
lowed for 6 days after their study transfusions.
The primary endpoint was safety with the biggest con-
cern being the development of a thromboembolic event as
CPPs are highly activated and contain increased numbers of
microparticles compared with LSPs (Fig. 2). There were no
thromboembolic events related to any study PLT transfusion
(24 CPP recipients) consistent with an exhaustive literature
review of prior CPP transfusions without a single reported
thromboembolic event.8 There were four mild transfusion-
related AEs that were considered related to a CPP transfu-
sion; that is, one patient had mild chills and fever, two had
DMSO-related skin odor, and one had a moderate headache
the day after transfusion.
The PLT CI in patients receiving CPPs increased in a
dose–response manner, but remained much less than LSP
CI even with a 3-CPP-unit transfusion. In spite of these poor
CIs in 16 of 28 (57%) patients who were clinically refractory
to standard-of-care PLT transfusions and were actively
bleeding, 14 of 24 (58%) patients receiving CPP had their
bleeding improve after transfusion including three of seven
(43%) patients with WHO Grade 4 CNS bleeding.
Furthermore, bleeding improvements were unrelated to
CPP dose (Table 3).
As expected in this clinically ill patient population,
coagulation tests demonstrated an overall hypercoagulable
state as is known for patients with malignancies,12 most
especially, low AT, elevated TAT, elevated F1+2, and ele-
vated D-dimer. In spite of the status of these patients before
transfusion as indicated by this panel of hemostasis
markers, they were actively bleeding with abnormalities
observable in sensitive global TGT and TEG coagulation
tests. The coagulation markers D-dimer, fibrinogen, PT, and
PTT were generally stable or had only small changes after
CPP PLT transfusions, consistent with the finding that there
was no clinical evidence of thrombosis in study patients
after a study PLT transfusion despite being in a pretransfu-
sion hypercoagulable state.
These preliminary data suggest that TGT and TEG
responses are diminished as a component of an overall coa-
gulopathy of disease and/or its treatment with a reduced
capacity to generate thrombin. Transfusion of CPP at least
partially corrects these outcomes in a dose-dependent man-
ner, and this effect can be observed for 24 hours posttrans-
fusion. TGT and TEG responses are confounded by PLT
count and other clinical factors such as uremia and

Volume 58, September 2018 TRANSFUSION 2137

SLICHTER ET AL.
pharmacologic effects of treatment drugs. There are no sug-
gestions from TGT, TEG, or other coagulation test data indi-
cating a thrombotic risk from CPP transfusions. In contrast,
the global tests of TGT and TEG suggest that CPP transfu-
sion partially corrects the pretransfusion coagulopathy. In
conclusion, CPP transfusions appear to be safe and effec-
tive, and a Phase 2 efficacy and safety trial in cardiopulmo-
nary bypass surgery patients with postsurgical bleeding is
planned.
ACKNOWLEDGMENTS
The authors acknowledge the patients that participated in this study.
The authors also thank the technical assistance of Jill S. Corson, Shawn
Bailey, Esther Pellham, Irena Gettinger, and Todd Christoffel from
BloodWorks NW; Shawnagay Nestheide, Sarah Hill, and Fatima Moh-
moud from Hoxworth Blood Center; Kathleen Grindle, Renee Geissler,
Deborah Dumont, and Melissa Barber, Dartmouth-Hitchcock; and the
nurses and clinical support staff of the Hematology/Oncology Wards of
the University of Washington Hospital, Dartmouth-Hitchcock Medical
Center, and the University Hospital of Cincinnati. The authors are
deeply appreciative of the extraordinary administrative support pro-
vided by Ginny Knight from BloodWorks NW in the preparation of this
manuscript and Stephen Ransom, of Fast-Track Drugs & Biologics, LLC
for statistical analyses of the study data.
CONFLICT OF INTEREST
The authors have disclosed no conflicts of interest.
REFERENCES
1. Valeri CR, Feingold H, Marchionni LD. A simple method for
freezing human platelets using 6 percent dimethylsulfoxide
and storage at -80 degrees C. Blood 1974;43:131-6.
2. Khuri SF, Healey N, MacGregor H, et al. Comparison of the
effects of transfusions of cryopreserved and liquid-preserved
2138 TRANSFUSION Volume 58, September 2018
platelets on hemostasis and blood loss after cardiopulmonary
bypass. J Thorac Cardiovasc Surg 1999;117:172-83.
3. Schiffer CA, Aisner J, Wiernik PH. Clinical experience with trans-
fusion of cryopreserved platelets. Br J Haematol 1976;34:377-85.
4. Schiffer CA, Buchholz DH, Aisner J, et al. Frozen autologous
platelets in the supportive care of patients with leukemia.
Transfusion 1976;16:321-9.
5. Schiffer CA, Aisner J, Wiernik PH. Frozen autologous platelet
transfusion for patients with leukemia. N Engl J Med 1978;299:
7-12.
6. Freyssinet JM. Cellular microparticles: what are they bad or
good for? J Thromb Haemost 2003;1:1655-62.
7. Böck M, Schleuning M, Heim MU, et al. Cryopreservation of
human platelets with dimethyl sulfoxide: changes in biochem-
istry and cell function. Transfusion 1995;35:921-4.
8. Slichter SJ, Jones M, Ransom J, et al. Review of in vivo studies
of dimethyl sulfoxide cryopreserved platelets. Transfus Med
Rev 2014;28:212-25.
9. Dumont LJ, Cancelas JA, Dumont DF, et al. A randomized con-
trolled trial evaluating recovery and survival of 6% dimethyl
sulfoxide-frozen autologous platelets in healthy volunteers.
Transfusion 2013;53:128-37.
10. Miller AB, Hoogstraten B, Staquet M, et al. Reporting results of
cancer treatment. Cancer 1981;47:207-14.
11. Davis KB, Slichter SJ, Corash L. Corrected count increment and
percent platelet recovery as measures of post transfusion plate-
let response: problems and a solution. Transfusion 1999;39:
586-92.
12. Sheth RA, Niekamp A, Quencer KB, et al. Thrombosis in cancer
patients: etiology, incidence, and management. Cardiovasc
Diagn Ther 2017;7:S178-85.
SUPPORTING INFORMATION
Additional Supporting Information may be found in the
online version of this article.
Tables S1–S15: Supporting Information.
Fig S1: Supporting Information.