Dr Jerome Connor, 3606 Research Forest Drive, The Woodlands, TX 77381, U.S.A.
Platelet transfusion represents an important component of the therapy for thrombocytopenic patients. Prolonged storage capabilities for platelets would alleviate many problems associated with blood banking. Unfortunately, current cryopreservation methods are complex to implement and result in loss of cell number and functional activity. Previous in vitro studies have shown that the use of ThromboSolTM, a platelet-stabilizing formulation, in the cryopreservation of platelets results in significant retention of cell number and in vitro functional activities in addition to reducing the DMSO requirement to only 2%. We evaluated the in vivo circulatory parameters of platelets cryopreserved with ThromboSol. Single donor platelet units were obtained from healthy volunteers (n = 16); the units were then split and cryopreserved with either ThromboSol and 2% DMSO or 6% DMSO alone. Following storage at −80°C for 7–10 d the samples were thawed, washed and radiolabelled with either 51Cr or 111In. The paired samples were then mixed and reinfused into the autologous volunteer. At various time intervals following transfusion a blood sample was drawn and the quantity of circulating labelled platelets was determined. The percent recovery and survival time was determined by multiple-hit analysis. The ThromboSol-treated platelets, as compared to the 6% DMSO-treated platelets, displayed statistically higher percent recovery (40.2% v 28.8%) and survival time (166.3 h v 152.1 h). These results demonstrated that platelets cryopreserved with ThromboSol displayed superior in vitro and in vivo characteristics as compared to the standard 6% DMSO method. The use of ThromboSol allowed for a 3-fold reduction in the DMSO concentration in conjunction with a 40% increase in circulating cell number and normal survival times.
Platelets represent an important transfusable blood component for the control of bleeding in thrombocytopenic patients. Current guidelines allow platelets to be stored for a maximum of 5 d at 20–24°C, creating an inventory control problem for hospitals and blood banks (Lazarus et al, 1982; Murphy, 1985). The storage restriction was the result of concerns over the potential for bacterial contamination (Alvarez et al, 1995; Yomotovian et al, 1993). The ability to cryopreserve platelets for extended storage would aid in the management of these storage-associated problems. Unfortunately the current system for cryopreserving platelets is neither a simple nor an effective method. In fact, routine platelet cryopreservation is not considered practical.
The American Association of Blood Banks (AABB) recommends two methods of platelet cryopreservation (Branch et al, 1993). Both systems include procedures that are labour-intensive, require a controlled-rate addition of a plasma/DMSO mixture and require a post-thaw wash step prior to transfusion. Moreover, the current methods of cryopreservation result in significant damage to the platelets following the thaw process (Balduini et al, 1993; Bock et al, 1995; Dullemond-Westland et al, 1987).
ThromboSolTM is a platelet-stabilizing solution consisting of selected second-messenger effectors that inhibit specific activation pathways endogenous to platelets, resulting in platelets that are biochemically stabilized against the detrimental effects of cold storage (Connor et al, 1996; Currie et al, 1998). Recent studies have demonstrated that the use of ThromboSol in the cryopreservation of platelets allowed simple processing, a reduction of the DMSO to 2% and excellent post-thaw retention of cell number and in vitro functional activity (Currie et al, 1998). Specifically, when compared to platelets cryopreserved using 6% DMSO, single donor unit platelets cryopreserved with ThromboSol yielded statistically significant higher retention of cell number, percentage of cells displaying a discoid morphology, ESC and HSR (Currie et al, 1998). In addition, the ThromboSol-treated platelets displayed a statistically relevant reduction in the expression of the activation marker P-selectin.
Overall, using ThromboSol in the cryopreservation of platelets yields a significant improvement in the retention of both platelet morphological indices and in vitro functional activities. Unfortunately, no in vitro assay has been identified as a strong indicator of the in vivo characteristics of transfused platelets. Thus, no strong conclusion can be drawn with regard to whether the improved in vitro indices will translate into improved in vivo circulatory parameters. This report describes a study that directly compared the in vivo circulatory characteristics of platelets cryopreserved with ThromboSol with those cryopreserved using 6% DMSO.
MATERIALS AND METHODS
Healthy volunteers (both female and male) gave informed consent to participate in this project, which was reviewed and approved by The University of Texas M. D. Anderson Cancer Center (UTMDACC) Institutional Review Board. All donors conformed to the UTMDACC blood bank donor criteria. The participants donated a standard single-donor apheresis unit (SDU) at the UTMDACC blood bank. The SDU was processed and tested according to standard blood-banking procedures. On the day following donation, two 30 ml samples of the platelets were obtained and placed into T-3101 150 ml transfer packs (CharterMed, Lakewood, N.J.). A sample of the fresh platelets was analysed for in vitro functional activity as described below.
Cryopreservation of platelets
One platelet sample was treated with ThromboSol according to the following protocol. ThromboSol, containing amiloride, adenosine and sodium nitroprusside (SNP), was prepared as a 50-fold concentrate in DMSO and sterile-filtered using a 0.2 μm DMSO compatible filter. Using a sample site coupler, 600 μl of the ThromboSol solution was added to the platelet sample yielding a final DMSO concentration of 2%. The resulting ThromboSol-treated platelet preparation contained the following reagents: amiloride 0.25 mm, adenosine 0.1 mm and SNP 50 μm. The treated platelets were then mixed by gentle shaking and directly inserted into a −80°C freezer in an aluminium cassette. The second platelet sample was treated with 6% DMSO based on the method outlined by the American Association of Blood Banks (Valeri Method), followed by insertion into a −80°C freezer in an aluminium cassette (Branch et al, 1993).
Platelet processing and radiolabelling
Following cryopreservation storage for 7–10 d, the paired platelet samples were removed from the −80°C freezer and thawed by insertion into a 37°C water bath. A sample of thawed platelets from both storage conditions was retained for the evaluation of post-cryopreservation in vitro functional activity as described below. The cell samples were washed and randomly radiolabelled with either 51Cr or 111In according to standard techniques (Holme et al, 1993; Keegan et al, 1992; Snyder, 1986; Wadenvik & Kutti, 1991). Following radiolabelling, a mixed sample of the platelets from each storage condition was prepared which contained approximately 555 kBq of each isotope. An aliquot of the sample was retained for determination of the radioactivity of the infusion dose, which represented the total input value.
Infusion and in vivo circulation
The mixed radiolabelled platelet sample was infused into the autologous donor via a venous injection. At 2 h and 1 d, 2 d, 4 d, 6 d and 10 d, a 10 ml blood sample was obtained by venepuncture. An aliquot of the whole blood was counted using a gamma counter to determine the amount of radioactivity of each isotope remaining in circulation. A correction factor was employed to correct for spillover between the two radioisotopes and for leaching of the radiolabel from the cells during circulation (Holme et al, 1993; Keegan et al, 1992; Snyder, 1986). The radioactivity remaining in the cells at the 10 d time point was used to correct for uptake of the radiolabel marker by circulating red cells as described previously (Keegan et al, 1992). In addition, the input sample, obtained prior to infusion, was corrected for cell sticking of the radiolabel as described previously (Holme et al, 1993). The corrected amount of each radioisotope remaining in circulation, at the indicated time point, was compared to the corrected input dose yielding the percentage of infused platelets from each storage condition remaining in circulation (Holme et al, 1993; Keegan et al, 1992; Snyder, 1986). The percent recovery and survival time were determined from a curve fitting program and the percent recovery and survival time were determined using the multiple-hit analysis (Snyder, 1986).
In vitro analysis
Fresh and cryopreserved platelet samples were analysed for the retention of in vitro morphological indices and functional activity. The platelets were assayed for cell number directly following removal from the storage bag and after a single wash step performed by centrifugation at 950 g for 20 min. The platelet samples were also analysed for mean platelet volume (MPV), percent discoid, ESC, HSR, ADP/collagen-induced aggregation and surface expression of the platelet proteins GPIb and P-selectin (CD62) as described previously (Connor et al, 1996; Currie et al, 1998).
The results are expressed as mean values plus and minus standard deviations. The mean values of the control and treated sample were compared by the two-tailed paired t-test using an α level = 0.05.
In vitro analysis
The retention of in vitro morphological indices and functional activities of the cryopreserved platelets from both storage systems are displayed in 1Table I. Both cryopreserved platelet populations yielded >94% recovery of cell number when thawed. Following a single centrifugation wash step, the platelets cryopreserved using 6% DMSO lost 20% of the starting cell number, whereas the ThromboSol-treated population retained 100% of the thaw cell number, a statistically significant difference (P < 0.01). As observed previously (Currie et al, 1998), the ThromboSol reagents protected the platelets from centrifugation-induced storage lesion by blocking the damaging activation events that occur during this procedure. Both cryopreserved samples displayed a significant increase in the cell volume as compared to fresh platelets (P < 0.01). The ThromboSol-treated platelets retained a statistically significant higher percentage of cells displaying a discoid morphology as compared to platelets cryopreserved with 6% DMSO (P < 0.01), yielding a value which was >80% retention of the value for fresh platelets. Both cryopreserved samples displayed retention of the expression of the surface marker GPIb that was not statistically different from the value for the fresh platelets.
Table 1. Table I. Retention of in vitro functional activity. Values are expressed as mean ±standard deviation.* P < 0.01 for cryopreserved platelet samples as compared to fresh platelet samples.† P < 0.01 for platelets cryopreserved with ThromboSol as compared to platelets cryopreserved with 6% DMSO.
The ThromboSol-treated platelets yielded statistically higher retention of ESC and HSR as compared to the 6% DMSO-cryopreserved platelets (P < 0.01). The retention of ESC and HSR for the ThromboSol-treated cells was 51% and 66% of the values for fresh platelets respectively. In contrast, the 6% DMSO-cryopreserved platelets' functional activity values for ESC and HSR fell to 33% and 45% of the fresh values, respectively. The platelets cryopreserved with 6% DMSO displayed less expression of the activation marker P-selectin than the ThromboSol-treated cells, but this difference did not reach statistical significance. The ADP/collagen-induced aggregation assay yielded no significant difference between the platelet samples stored using the two different cryopreservation systems. Both cryopreserved samples yielded retention of agonist-induced aggregation of ~40% of the value for fresh cells.
In vivo analysis
The circulatory parameters, following infusion into the autologous participants, of the platelet samples cryopreserved with ThromboSol or 6% DMSO are displayed in 2Table II. The percent recovery and the survival time were determined using the multiple-hit curve fit analysis. The platelets cryopreserved with ThromboSol yielded a statistically significant higher recovery of the infused sample (40.2% v 28.8%; P < 0.005). The ThromboSol-treated platelets also displayed longer circulation times than the platelets cryopreserved with 6% DMSO, though the difference was not statistically significant (P = 0.09). The survival curves for the two cryopreserved populations are shown in Fig 1. At the 2 h, 1 d, 2 d and 4 d circulation time points the platelets cryopreserved with ThromboSol displayed statistically higher number of circulating cells as compared to the platelets cryopreserved with 6% DMSO (P < 0.01 for 2 h and 4 d; P < 0.001 for 1 d and 2 d).
Table 2. Table II.In vivo circulatory parameters. * P < 0.005 for platelets cryopreserved with ThromboSol versus platelets cryopreserved with 6% DMSO.† P = 0.0901 for platelets cryopreserved with ThromboSol versus platelets cryopreserved with 6% DMSO.
Over the last two decades platelet transfusions have been used with increasing frequency to control thrombocytopenia. The higher demand for platelet transfusions has emphasized the need for long-term storage of platelets and has motivated the pursuit of improved methods for cryopreservation of platelets. Recent studies have shown that ThromboSol, a newly developed platelet storage formulation, allows for the cryopreservation of platelets that are superior to platelets cryopreserved using 6% DMSO, with regards to in vitro functional activity (Currie et al, 1998). Since there is no specific in vitro functional assay that can absolutely predict the in vivo status of stored platelets following transfusion, it is necessary to provide in vivo clinical evidence of the effectiveness of this storage system.
The ThromboSol-cryopreserved platelets displayed superior recovery following infusion, showing a 40% increase in the number of cells remaining in circulation as compared to platelets cryopreserved with 6% DMSO. This enhanced number of circulating platelets was sustained over the full time course of circulation, resulting in a slightly higher survival time. Moreover, following the wash step, the platelets cryopreserved with 6% DMSO yielded a 75% retention of cell number, whereas the ThromboSol-treated platelets retained 94% of the starting cell number. If this washing-induced cell loss is taken into account when comparing the recovery of circulating cell number following infusion, the ThromboSol cryopresevation system would yield a 75% increase in circulating cell number, over the 6% DMSO system, following transfusion. The low cell number recovery, observed with the 6% DMSO system, is a consequence of the required wash step to remove the cryoprotectant, which is a damaging process for the platelets. In contrast, the low concentration of DMSO used in the ThromboSol cryopreservation protocol allows for the potential for direct transfusion of platelets following thawing. Thus, the exclusion of a wash step, with the ThromboSol system, could yield even higher numbers of circulating cells than those obtained during this study.
In order to determine the relevance of the percent recovery of the treated platelets, values for in vivo recovery of preserved platelets was compiled from published reports. 3Table III shows the percent recovery and survival time of fresh platelets and platelets stored for 5 d at room temperature according to standard methods. These circulatory parameter values were obtained using protocols similar to those described in this report and analysed by the multiple-hit analysis. The platelets cryopreserved with ThromboSol yield in vivo recovery values clearly in the range seen with conventionally stored platelets for 5 d at room temperature. Interestingly, comparison of the in vitro functional activity of these two populations reveal that both display similar retention of many characteristics including ESC, HSR and discoid morphology (Connor et al, 1996; Currie et al, 1998). Therefore the ThromboSol-treated cryopreserved platelets could be expected to be as effective as conventionally stored cells with the advantage of extended storage periods. In contrast, the in vivo recovery of fresh platelets is superior to the cryopreserved treated platelets (Table III). Similarly, the in vitro retention of functional activity of fresh platelets is also higher than the cryopreserved platelet population. In practice, though, the clinical availability of fresh platelets is limited and unattainable with regards to autologous donation.
The ability to predict in vivo circulatory characteristics and recovery of platelets, based on the outcome of a specific in vitro analysis, would be beneficial to both the development of storage systems and the prediction of the efficacy of platelet units at the blood bank prior to transfusion. Regression analysis of the in vitro assays performed in this study, in relation to the in vivo percent recovery, revealed no in vitro assay system that satisfied this criterion. Previous work has shown some correlation of in vivo parameters to ESC and HSR (Holme et al, 1997; Murphy et al, 1994), but these assays were not predictive in our experiments. It is of interest to note that the in vivo percent recovery of the ThromboSol cryopreserved platelets was 68% of the referenced value for fresh platelets. This comparison was similar to that observed with fresh versus ThromboSol-treated platelets in the HSR analysis (66%) and ESC analysis (52%). Thus although not directly correlative, these in vitro functional assays may be considered predictive of the potential in vivo parameters.
In addition to the enhanced recovery of circulating platelets, an important consequence of the use of ThromboSol is the ability to reduce the requirement for DMSO as a cryoprotectant. Previous studies testing the use of reduced concentrations of DMSO during infusion indicated that the recipients showed no adverse effects, even following a 10-year period (Handin & Valeri, 1972; Lazarus et al, 1981; Melaragno et al, 1985; Spector et al, 1977). The development of a cryopreservation system that contains transfusable components would eliminate both the damaging effects of the wash step and the need to break the closed system.
The ability to achieve long-term storage of directly transfusable platelets possessing in vivo circulatory characteristics equivalent to conventional 20–24°C stored platelets would significantly impact clinical transfusion medical practices. ThromboSol represents a unique method for the long-term storage of platelets which retain in vivo circulatory parameters similar to standard 20–24°C stored platelets. Future experiments are being developed to analyse the effectiveness of full unit transfusions of ThromboSol-cryopreserved platelets, both washed and unwashed, in thrombocytopenic patients.
We thank the following individuals for their technical assistance: Kelly Larrabee, BSN, Judianne Jones and Toy Leach. This research was supported in part by the U.S. Navy Contract N00014-96-C-0120.