Original Paper Vox Sang 1992;62:146-151

Antonio Angelini Alfredo Dragani Anna Berardi Antonio Iacone Guiseppe Fioritoni Glauco Torlontano

Evaluation of Four Different Methods for Platelet Freezing In vitro and in vivo Studies

Chair of Hematology, University of Chieti, Division of Hematology and Blood Bank, Civil Hospital, Pescara, Italy

................................................................................................. Abstract

This report describes our experience with various techniques for the freezing of platelet-rich plasma, removed from the final product after leukapheresis procedures performed on 14 hematological patients. A total of 194 platelet units were frozen for subsequent autologous transfusion, by the following four methods: (1) 6% dimethyl sulfoxide (DMSO); (2) a combination of 5 % DMS0/6% hydroxyethyl starch; (3) 3% glycerol; (4) 5% glycerol/4% glucose. Each technique was evaluated by measuring the percentage of platelet recovery, malondialdehyde (MDA) production, and lactate dehydrogenase release. To investigate the safety and therapeutic effectiveness of the previously frozen platelets, in vivo comparison of four platelet freezing methods was made in 8 thrombocytopenic patients, using corrected platelet increment (CCI), determined at 24 h. Our in vitro results indicate that the cryopreservation with 6% DMSO, without controlled cooling rate, provides significantly ( ~ 4 1 . 0 5great) er platelet recovery (75%) as compared to other systems. The decrease of MDA production and the increase in plasma lactate measured after the thawing process was less in the DMSO-frozen units than in the other platelet units. When platelets, cryopreserved by this method, were subsequently transfused into patients, a significantly better CCI (>5,OOO/pl) was obtained. In our series, 6 patients were entirely supported with frozen autologous platelets. It appears from this study that a better understanding of the physical and biochemical events occurring during the freezing process will improve platelet cryopreservation, allowing a more systematic use of frozen platelets in the support of thrombocytopenic patients.

...........

Treatment with previously cryopreserved autologous platelets during the thromobocytopenic period is an essential part in the supportive care program after bone marrow transplantation. The use of autologous frozen platelets, collected during remission, prevents the major complications of platelet support such as HLA alloimmu-

Received: January 4. I Y Y I Revised manuscript received: August 26, I W I Accepted: Sept 16, l Y Y l

nization and infection with transmissible disease agents [l]. In recent years, peripheral blood has been used as an alternative source of stem cells for autotransplantation in lieu of bone marrow [2,3]. Platelets obtained from leukapheresis procedures can be removed from mononuclear cell suspensions and frozen by different methods and

Dr. Antonio Angclini Division of Hematology Civil Hospital Via Renato Paolini. I-651(K)Pescara (Italy)

1992 S. Karger AG. Barel o(U?-Ylx)7/9?/0h?34I I46 !$?.7S/Il

'Ll

cryoprotective agents [4-61. In this study we have evaluated the efficacy of four different platelet freezing methods. We employed 6% dimethyl sulfoxide (DMSO), a combination of hydroxyethyl starch (HES) and DMSO, 3% glycerol (3% Gly) and 5% glycerol-4% glucose (G/G) with reconstitution in a non plasma diluent. 194 autologous platelet concentrates (PC) were frozen to support 14 patients undergoing autologous peripheral blood stem cell transplantation. These techniques were compared by evaluating the effects of freezing, thawing and washing on platelet function. The effectiveness of transfusions was assayed by posttransfusion corrected platelet count increment (CCI) determinded at 24 h.

Materials and Methods Preporatioti of PCs A total of 194 leukapheresis procedures were performedon 14 hematological patients with a continuous flow blood cell seperator (CS 3000, Baxter Healthcare Co., Deerfield, Ill.) according to standard operating procedures. Each collection product was centrifuged at 255 g for 15 min at 22°C in a Sorvall RC-3 centrifuge. The supernatant, platelet-rich plasma (PRP). wasexpressedbyaplasmaextractor into a polyvinyl chloride (PVC) transfer bag (Terumo Corporation, Tokyo. Japan) while samples were collected and tested for platelet count and in vitro assays. The PRP was further concentrated at 22°C by centrifugation at 4,500 g for 10 rnin to sediment platelets. Platelet-poor plasma (PPP) was expressed in a 300-ml plastic transfer bag, to be stored at -20°C for reconstitution of thawed-washed platelets. After 60 niin of undisturbed storage at room temperature (RT), PCs were resuspended with 60ml of 4% human serum albumin (HSA) in 0.9% sodium chloride. At this stage a sample was obtained for platelet count and in vitro studies. Platelet suspensions were then withdrawn in a 60-ml plastic syringe, injected into a Teflon-Kapton freezing bag (DF 700, Gambro, Hechingen, FRG) through the needle port by the corner and cryopreserved as described below. All procedures were performed within a laminar flow cabinet. Plutelet Cryopreservatioti Methods 6% DMSO. Sixty-one platelet units were cryopreserved using DMSO as a cryoprotectant according to a modification of the procedure described by Valeri et al. [7]. After resuspension in 60 ml of 4% serum albumin, platelets were left in the freezing bag at RT under gentle shaking for 30 min. Twelve percent sterile DMSO (Cryoserv, Tera Pharmaceuticals, Buena Park, Calif.) in a 0.9% sterile sodium chloride solution (previously stored at 4°C) was then added (v/v), over a 30-min period, to the platelet suspension through a normal giving set on an oscillating shaker, with a flow rate controlled by the caliber of the plastic tube. The final concentration of DMSO was 6% with 1 unit of platelets per 120 ml. The freezing bag port was heatsealed and cut after removal of air bubbles. The bag was then placed between two metallic holders to ensure homogeneous distribution of the suspension during freezing. The entire assembly was placed horizontally in a -80°C mechanical freezer. This gave roughly a 2-3"C/min rate, though the freezing rate was not controlled. The frozen platelets were thawed by immersion in a 37°C wather bath

without agitation for 5 min. The container was removed and thawed platelets tansferred form the freezing bag to a 600-ml PVC transfer bag through the second needle port of the freezing bag. Platelets were gently mixed during the transfer. One more sample was taken at this stage for platelet count and function studies. Frozen-thawed platelets were then diluted rapidly at RT with a 400-ml volume of 4% human albumin previously stored at 4°C and centrifuged at 4,500 g for 5 min in a Sorvall RC-3 centrifuge at 22°C. The supernatant solution, containingmost oftheDMSO, wasexpressed byaplasmaextractorinto a 600-ml plastic transfer bag and discarded and 500 ml 4% HSA was then added to the PC. After resuspension by gentle agitation, platelets were centrifuged again at 4,500g for 5 min, the supernatant removed and the PCs reconstituted in autologous frozen-thawed plasma. The complete procedure required approximately 30 min. Platelets were then incubated with agitation at 6 rpm using an elliptical rotator (Fenwal Laboratories, Deerfield, Ill.) at RTfor about 1 h, prior to collecting another sample for platelet count. 5 % D M S O d % HES. Forty-six platelet units were cryopreserved with a mixture of DMSO and HES according to the method of Stiff et al. [8]. A volume of 500ml of freezing solution was prepared as follows: 60 g of HES powder (Baxter, Labs) and 540 mg of dextrose powder were added to 240 ml Normosol R (Abbot, Labs) in a 1,000cm3 graduated beaker. This viscous solution was stirred until completely dissolved, transferred into a 500-ml sterile evacuated bottle (Baxter, Labs) and then autoclaved. After cooling at RT, 160 ml of 25% HSA were added and the mixture stored at 4°C. At the time of use, 6 ml of sterile DMSO were added to 54 ml of the mixture in a 150-ml PVC transfer bag, to yield a 60-ml volume of cryoprotective solution containing 10% DMSO, 12% HES and 8% albumin. After cooling to 4"C, this freezing solution was rapidly injected into a Teflon-Kapton freezing bag, containing 60 ml of PC diluted in 4% albumin, using an 18-gauge needle and a 60-ml plastic syringe, through a sampling site coupler, over a 5-min period. The final productwas5% inDMSOd% HES. Trappedairwas carefullyremoved by aspiration and the bag was heat-sealed, placed between two metallic plates, and laid flat for storage in a -80°C mechanical freezer with a 2-3"Uminrate. The freezing rate wasnot controlled. For autologous transfusion, frozen platelets were rapidly thawed by immersion in a 37°C water bath for 5 min and transferred to a 600-ml PVC transfer bag. Samples were collected for platelet counts and function assays. The suspensions were then diluted (v/v) with 4% HSA. previously stored at 4°C. and centrifuged twice at 4,500 g for 10 min at 22°C to remove most of cryoprotectant. After washing, platelets were reconstituted with autologous fresh frozen plasma and stored undisturbed at RTfor60 min. Furthersampleswere then takenforplatelet counts. 3%Glycerol. Forty-five platelet units were frozen using glycerol as a cryoprotectant, according to Dayian et al. [9]. The PC, resuspended in 60 ml of 4% albumin in saline, were gently mixed over a 5-min period with a equal volume of a cryoprotective solution containing 6.3 ml of highly concentrated (57%) sterile glycerol (Baxter Lab) in physiological saline. The final glycerol concentration was 3%. Air was removed and the bag port was sealed, using a heat sealing device, and cut. Glycerolized platelets were placed between two metallic holders and cryopreserved using a programmed freezer Nicool ST 20 (Compagnie Franqaise de Produits OxygCnC).The cooling rate was 10"C/min from 15°C until the heat fusion (-3/4"C), followed by 35"C/min down to -140°C. Temperature changes were monitored ina referencebag. containinga60-mlvolumeof4% HSAandanequal volume of cryoprotective solution. After freezing, bags were removed from the freezing chamber and quickly transferred free of

147

holders to the vapor phase (-120°C) of liquid nitrogen. For autotransfusion, bagswere thawedbyimmersionina42"Cwaterbathforabout 3 min with gentle agitation and rapidly transferred to a 600-ml PVC bag. Samples were taken for in vitro function studies; autologous plasma was added to reduce glycerol concentration and samples were collected from diluted material for platelet counts. No washing was done. Diluted platelets were infused after incubation for 1h at RT. 5% Glycerol-4% Glucose. Forty-two platelet units were frozen with a glycerol-glucose cryoprotective solution according to Dayian et al. [lo]with minor modifications. A 60-ml volume of PC was mixed with an equal volume of freezing solution containing 10.5 ml of 57% sterile glycerol and9.6 ml of 50% sterileglucose in normal saline. The final concentrations were 5% glycerol and 4% glucose. Glycerolized plateletswerefrozeninanelectronicfreezerNicoolST20bythesame cooling program as above and stored in the vapor phase of liquid nitrogen. Thawing was performed at 22°C in a water bath with gentle shaking. Samples were then aseptically removed for in vitro testing. while thawed platelets were transferred in a PVC transfer bag and centrifuged at 4,500 g for 10 min in a Sorvall RC 3 centrifuge at 22°C. PCs were then diluted 15 in a reconstitution medium (pH 6 4 , containing tri-sodium citrate (dihydrate). 0.2 g citric acid (monohydrate),3.1 ganhydrousdextrose, 0.4 g KCI, 6.3 g NaCl and placed into a 37°C water bath for 30 min with gentle agitation. At the end of the incubation, diluted platelets were recentrifuged at 4,500gfor 5 min, the supernatant expressed in a PVC bag and the final PC resuspended with autologous fresh frozen plasma. The suspension was allowed to stand undisturbed at RT for 60 min before transfusion.

vitro Evaluatioti Platelet counts were performed on an electronic particle counter (CoulterCountermodel S plus), before and immediatelyafterfreezethaw-wash-resuspension procedures, to determine the percentage of platelet recovery. The pH was monitored in fresh PPP and after resuspension in autologous plasma, at 22°C under anaerobic conditions, using a model PHM84 pH meter (Radiometer. Copenhagen, Denmark). Lactate Delzydrogeriase ( L D H ) Release. The LDH activity was measured spectrophotometrically at 340 nm using an assay based on the reduction of NAD during the conversion of lactate to pyruvate by LDH (Sigma Chemical Co., St. Louis, Mo.). Enzyme levels were measured in each freezing bag after thawing and expressed as a percent of total (per 10' platelets) available activity. To measure total LDH activity, a sample of fresh PRP was removed aseptically from each PVC bag and adjusted to a platelet count of 1OY/ml.After three freeze-thaw cycles to disrupt the platelets, the sample was centrifuged at 4,500g for 10 min and LDH activity measured in the supernatant. This value was defined as total intraplatelet LDH activity per 10" platelets. Immediately after thawing, samples were removed from each platelet freezing bag and centrifuged at 4,500g in a microfuge (Labofuge A, Heraeus-Christ GmbH, Osterode, FRG). The supernatant LDH activity was normalized to LDH/lO' platelets and expressed as a percentage of total intraplatelet activity. Malotidialdehyde ( M D A ) Production. A 2-ml aliquot was removed from each platelet bag and immediately after thawing, adjusted to 10' platelets/ml and centrifuged at 4,500 g for 5 min. The supernatant was removed and the platelet pellet incubatedfor 60 min in a 37°C water bath with a 2-ml solution containing N-ethylmaleimide (1 mM) in phosphate-buffered saline (pH 7.4). At the end of the incubation, the reaction was stopped by the addition of 2 ml of 2.3% perchloric acid containing 0.53% 2-thiobarbituric acid. After agIt1

148

itation, the solution was heated for 15 min in boiling water. After cooling at RT. the solution was centrifuged and the optical density of the pink chromogen in the supernatant read spectrophotometrically at 532 nm [ll]. The amount of MDA measured after thawing was compared to that produced before freezing and expressed as a percentage per 10' platelets. Hypotonic Shock Reactioti. Platelet response to hypotonic shock was assayed after the thawing process as follows: a sample of thawed plateletswas asepticallyremovedfrom each freezing bag and adjusted to a count of 400,000/~1with autologous PPP. For the assay, 1-ml aliquots of the diluted suspension were placed in a photometer (OD 610) and the changes in light absorbance read after the addition of 0.5 ml distilled water to cause an osmotic stress. Readings were performed at 30°C, over a continuous 10-minperiod. Recovery from osmotic stress was expressed as a percent of the control value, obtained assaying an equal volume of fresh platelets added to 0.5 ml of normal saline [12]. Bacteriological Control. Fresh and thawed-washed samples were removed aseptically from platelet bags, inoculated into soy-trypticase broth and streaked onto blood agar media. 111vivo Evaluation

Autologous transfusions were performed when platelet counts fell below 25,OOO/pl. Prior to transfusion 20r3 reconstituted PCs were pooled into a transfer bag and infused in about 30 min, through a blood component recipient set, with an in-line filter. Twenty-fourhour posttransfusion counts were obtained and the results were expressed as corrected count increments according to the following equation: (posttransfusion platelet count minus pretransfusion platelet count) x body surface area (m')divided by the number of platelet transfused X 10". CCIs obtained from afebrile patients. with no evidence of sepsis before or at least 24 h after transfusion, were considered suitable for evaluation. The order of platelet unit administration was randomized.

Statistics For comparison a t test for paired samples or a two-tailed test was used as appropriate. All statistical results were reported as significant for p