ORIGINAL ARTICLE In vitro comparison of cryopreserved and liquid platelets: potential clinical implications Lacey Johnson,1 Michael C. Reade,2 Ryan A. Hyland,1 Shereen Tan,1 and Denese C. Marks1
BACKGROUND: Platelet (PLT) concentrates can be cryopreserved in dimethyl sulfoxide (DMSO) and stored at −80°C for 2 years. These storage conditions improve availability in both rural and military environments. Previous phenotypic and in vitro studies of cryopreserved PLTs are limited by comparison to fresh liquid-stored PLTs, rather than PLTs stored over their clinically relevant shelf life. Further, nothing is known of the effect of reconstituting cryopreserved PLTs in plasma stored at a variety of clinically relevant temperatures. STUDY DESIGN AND METHODS: Apheresis PLTs were either stored at room temperature for 5 days or cryopreserved at −80°C with 5% DMSO. Cryopreserved PLTs were thawed at 37°C and reconstituted in plasma (stored at different temperatures) and compared to fresh and expired liquid-stored PLTs. In vitro assays were performed to assess glycoprotein expression, PLT activity, microparticle content, and function. RESULTS: Compared to liquid PLTs over storage, cryopreserved PLTs had reduced expression of the key glycoprotein receptors GPIbα and GPIIb. However, the proportion of PLTs expressing activation markers CD62P and CD63 was similar between cryopreserved and liquid-stored PLTs at expiry. Cryopreserved PLT components contained significantly higher numbers of phosphatidylserine- and tissue factor–positive microparticles than liquid-stored PLTs, and these microparticles reduced the time to clot formation and increased thrombin generation. CONCLUSION: There are distinct differences between cryopreserved and liquid-stored PLTs. Cryopreserved PLTs also have an enhanced hemostatic activity. Knowledge of these in vitro differences will be essential to understanding the outcomes of a clinical trial comparing cryopreserved PLTs and liquid PLTs stored for various durations.
I
n Australia, and many other countries including the United States, platelets (PLTs) stored under standard blood banking conditions, 20 to 24°C with constant agitation, currently have a shelf life of 5 days. This short shelf life can result in logistic and supply problems, so alternatives permitting longer storage, such as cryopreservation, are being investigated. Cryopreserved PLTs have been used by the military to treat bleeding since 2001.1 To date, only a single clinical trial underpins this practice, which demonstrated that cryopreserved PLTs are hemostatically effective when transfused in patients bleeding after cardiopulmonary bypass.2 Despite this encouraging result, a trial involving 73 patients, only 24 of whom received cryopreserved PLTs, is insufficient to support regulatory approval. In an effort to provide the required clinical evidence to satisfy clinicians and regulatory bodies, a pilot, randomized controlled clinical trial of cryopreserved PLTs, the cryopreserved
ABBREVIATIONS: APC = allophycocyanin; DFP = deep-frozen plasma; FFP24 = fresh-frozen plasma stored at 4°C for up to 24 hours; MA = maximum amplitude; R-time = time to clot initiation; TEG = thromboelastogram. From 1Research and Development, Australian Red Cross Blood Service, Sydney, NSW, Australia; 2Burns, Trauma and Critical Care Research Centre, University of Queensland, Brisbane, Queensland, Australia; and Joint Health Command, Australian Defence Force, Canberra, ACT, Australia. Address correspondence to: Lacey Johnson, Research and Development, Australian Red Cross Blood Service, 17 O’Riordan Street, Alexandria, NSW 2015, Australia; e-mail: [email protected]. We acknowledge that the Australian Governments fully fund the Australian Red Cross Blood Service for the provision of blood products and services to the Australian Community. Parts of this study were funded by the Defence Health Foundation. Received for publication June 30, 2014; revision received September 2, 2014, and accepted September 8, 2014. doi: 10.1111/trf.12915 © 2014 AABB TRANSFUSION **;**:**-**. Volume **, ** **
TRANSFUSION
1
JOHNSON ET AL.
versus liquid PLTs for surgical bleeding (CLIP) trial (ACTRN12612001261808), is planned.3 It is well established that PLTs develop a storage lesion during liquid storage, characterized by a progressive decline in in vitro quality (reviewed by Shrivastava4). While it is known that cryopreserved PLTs are more activated than fresh PLTs (1000 patients transfused) of clinical experience with this product.10 However, the logistic advantages of using DFP in austere environments do not apply to civilian hospital settings. Further, DFP is not a standard product in Australia. As such, for logistic and regulatory reasons, the use of fresh-frozen plasma (FFP; plasma frozen and stored at −30°C) for PLT reconstitution in a civilian hospital setting would be preferable. Some hospital blood banks routinely keep a unit of thawed FFP stored at 4°C for up to 24 hours (FFP24), for use in emergencies. Using prethawed plasma would allow cryopreserved PLTs to be supplied in roughly the same time frame as liquid-stored PLTs. However, there are currently no data to support the use of FFP or FFP24 for reconstitution of cryopreserved PLTs. The aim of this study was therefore to comprehensively characterize and compare the in vitro quality of liquid-stored PLTs over the entire shelf life, such as will be used in the CLIP study, with cryopreserved PLTs. In addition, we aimed to compare the effects of different plasma types for PLT reconstitution.
MATERIALS AND METHODS This study had approval from the Australian Red Cross Blood Service Human Research Ethics Committee. All donations were from eligible, voluntary donors. Leukoreduced (200 6.4-7.4
Thawed PLTs FFP24‡
DFP†
FFP‡
299.6 ± 10.4§ 299.9 ± 36.4§|| 7.20 ± 0.02§
300.7 ± 24.1§ 273 ± 39.1|| 7.22 ± 0.01§
Specifications
296.5 ± 16.7§ 252.86 ± 26.2¶ 7.24 ± 0.05§¶
250-350 >200 6.4-7.4
* Values shown as mean ± SD. † n = 12 in each group. ‡ n = 8 in each group. § p < 0.05 between Day 2 and thawed PLTs, determined using ANOVA and Dunnett’s multiple comparison test. || p < 0.05 between Day 5 and thawed PLTs, determined using ANOVA and Dunnett’s multiple comparison test. ¶ p < 0.05 between DFP and indicated plasma, determined using ANOVA and Bonferroni’s multiple comparison test. NA = not applicable.
Volume **, ** **
TRANSFUSION
3
JOHNSON ET AL.
expression of these markers on the cryopreserved PLTs was equivalent to liquid PLTs stored for 5 days. Fresh liquid-stored and cryopreserved PLTs expressed similar basal levels of activated GPIIIa receptor (PAC-1; Fig. 1G). However, cryopreserved PLTs were unable to respond to ADP to induce receptor activation, whereas the liquidstored PLTs maintained a response even after 5 days of storage (Fig. 1H). The type of plasma used for reconstitution did not alter PLT receptor expression. The ability of PLTs to aggregate in response to agonists is an indicator of functional quality. PLT aggregation in response to both collagen and ADP decreased during the 5-day storage period (Figs. 2A-2D). Although the aggregation responses in thawed cryopreserved PLTs were significantly reduced compared to fresh PLTs (Day 2), the average collagen aggregation response was equivalent to Day 5 PLTs (Figs. 2A and 2B). The type of plasma used for reconstitution did not influence the aggregation response. TEG assesses PLT contribution to global clot formation. The R-time was not affected by storage; however, cryopreserved PLTs had a more rapid clotting response than liquid-stored PLTs (Figs. 2E and 2F). Cryopreservation also resulted in a slight, but significant, reduction in clot strength compared to liquid-stored PLTs, as exhibited by the decreased MA (Figs. 2E and 2G). The use of DFP, FFP, or FFP24 resulted in similar TEG responses. PLT microparticles contribute to hemostasis and their concentration is Fig. 1. The surface expression of PLT receptors vary between liquid-stored and reported to increase with storage duracryopreserved PLTs. PLTs were sampled on Days 2 and 5 of liquid storage or immedition and after cryopreservation.13,16 The ately after reconstitution in the indicated plasma. PLTs were stained with the indiabsolute number of phosphatidylserine cated antibodies—(A) CD61-FITC, (B) CD42a-PE, (C) CD41a-FITC, (D) CD42b-PE, (E) -expressing PLT microparticles did not CD62P-PE, (F) CD63-PE, (G) PAC-1-FITC, (H) PAC-1-FITC after stimulation with increase during PLT storage (Fig. 3A), ADP—and analyzed by flow cytometry. The data represent mean ± SD (error bars); but was significantly higher after *p < 0.05 compared to Day 2 liquid storage; †p < 0.05 compared to Day 5 liquid cryopreservation. The absolute number storage. of tissue factor–positive microparticles was much lower than phosphatidylserine-expressing microparticles and simiwas significantly reduced in the cryopreserved PLT comlarly did not increase during storage. Interestingly, ponents (Figs. 1C and 1D), compared to liquid-stored the DFP-reconstituted cryopreserved PLTs contained sigunits. As expected, the proportion of PLTs expressing the nificantly more tissue factor–positive microparticles activation markers P-selectin (CD62P) and CD63 (Fig. 3B). increased during storage (Figs. 1E and 1F), whereas 4
TRANSFUSION Volume **, ** **
CRYOPRESERVED VS. LIQUID-STORED PLTs
Fig. 2. PLT aggregation and TEG responses vary between liquid-stored and cryopreserved PLTs. PLTs were sampled on Days 2 and 5 of liquid storage or immediately after reconstitution in the indicated plasma. PLT aggregation was analyzed using an aggregometer and the following agonists: (A, B) 10 μg/mL collagen or (C, D) 20 μmol/L ADP, with a representative trace shown and the mean ± SD (error bars). TEG variables were assessed using PLTs activated with kaolin, with (E) a representative trace shown and the mean ± SD (error bars) for (F) R-time and (G) MA. *p < 0.05 compared to Day 2 liquid storage; †p < 0.05 compared to Day 5 liquid storage. Volume **, ** **
TRANSFUSION
5
JOHNSON ET AL.
Fig. 3. PLT microparticle number and functionality vary between liquid-stored and cryopreserved PLTs. PLT microparticles were stained with (A) CD61-APC and annexin V-FITC or (B) CD61-APC and CD142-PE and enumerated using flow cytometry. Phosphatidylserine (PS)- and tissue factor (TF)-dependent microparticle function was measured from supernatant. PS-dependent (C) clotting time was measured with the Procoag-PPL assay and (D, E) thrombin generation was initiated using 1 pmol/L tissue factor (PRP reagent). TF-dependent (F, G) thrombin generation was initiated with 4 μmol/L phospholipids (MP reagent). The data represent mean ± SD (error bars); *p < 0.05 compared to Day 2 liquid storage; †p < 0.05 compared to Day 5 liquid storage.
6
TRANSFUSION Volume **, ** **
The contribution of microparticles to hemostatic function was assessed from the supernatant fraction, using the PPL clotting assay, which is dependent on the presence of phosphatidylserine. The clotting time was not affected by storage, but was significantly reduced after cryopreservation (Fig. 3C). Again, the type of plasma used for PLT reconstitution had no effect on clotting time. The functional effect of these microparticles was also assessed by measuring their contribution to thrombin generation using the PRP reagent, which relies on the presence of phosphatidylserine in the sample. The lag time to thrombin generation was not affected by storage or the type of plasma used for reconstitution; however, cryopreserved PLTs had a significantly shorter lag time than liquid-stored PLTs at any time during storage (Fig. 3D). The peak amount of thrombin capable of being generated was significantly higher in the cryopreserved PLTs, compared to the liquid-stored PLTs (Fig. 3E). Thrombin generation can also be triggered by the presence of tissue factor. The contribution of tissue factor to the thrombin-generating capacity was assessed using the MP reagent (Figs. 3F and 3G). The data demonstrate that the peak of thrombin generation was increased during PLT storage. However, the supernatants of cryopreserved PLTs were still capable of faster and greater tissue factor–induced thrombin generation than liquid-stored PLTs. In summary, it is likely that the greater number of microparticles present in the supernatant of the cryopreserved units (regardless of type of plasma used for reconstitution) may be responsible for mediating the phosphatidylserine- and tissue factor– dependent procoagulant activity. The presence of clinically relevant soluble factors was also measured in the PLT supernatant. The absolute amount of P-selectin (soluble CD62P) was higher in the cryopreserved units (Fig. 4A), compared to liquid-stored PLTs throughout the shelf life. Similarly, thromboxane levels were 50-fold higher in the cryopreserved PLT units,
CRYOPRESERVED VS. LIQUID-STORED PLTs
DISCUSSION In this in vitro study, a comprehensive analysis of the characteristics of thawed cryopreserved PLTs was compared to liquid-stored apheresis PLTs over their shelf life (2-5 days postcollection). Cryopreserved PLT components have a greater volume, but contain a similar number of PLTs to liquid-stored units. Cryopreserved PLT units also have significantly more microparticles than liquid-stored PLTs and are more hemostatically active, despite having a lower expression of important PLT glycoproteins and compromised aggregation responses. This study shows, for the first time, that these differences exist regardless of the duration of storage for liquid PLTs up to the current shelf life (5 days). This study differs from previous Fig. 4. The amount of soluble mediators varies between liquid-stored and comparisons of cryopreserved PLTs as it cryopreserved PLTs. PLT supernatant was collected from the PLT components and assessed the functional characteristics the concentration of soluble (A) P-selectin, (B) thromboxane, (C) RANTES, and (D) of cryopreserved PLTs compared to IL-27 was measured by ELISA. The concentration was multiplied by the unit volume those of liquid-stored PLTs over the to report the total dose/unit. The data represent mean ± SD (error bars); *p < 0.05 potential transfusable shelf life, rather compared to Day 2 liquid storage; †p < 0.05 compared to Day 5 liquid storage. than immediately after collection. Further, in previous prefreeze and postthaw comparisons,11,13 the prefreeze PLTs were not irradiated. In this study, the liquid-stored PLT components were irradiated before compared to the liquid-stored PLTs, but were not affected release, as this is standard practice in Australia for all by storage time or the plasma used for reconstituliquid PLT components, while the cryopreserved PLTs tion (Fig. 4B). RANTES appears to be released from were not. Gamma irradiation is known to damage PLTs;17,18 the PLT granules after cryopreservation, but was at a similar level as PLTs stored for their shelf life (5 days; however, the changes induced by irradiation are likely Fig. 4C), while the amount of IL-27 present in the PLT minimal compared to those induced by cryopreservation.5 components was comparable between liquid-stored It is well established that PLTs undergo changes and cryopreserved PLTs, indicating that this cytokine was during ex vivo storage, known collectively as the PLT not released from the PLTs in substantial amounts storage lesion, evidenced by altered metabolism, (Fig. 4D). increased expression of activation markers (CD62P, CD63, Both liquid-stored and cryopreserved apheresis PLT and phosphatidylserine), and decreased aggregation components are suspended in plasma, so the contriburesponse.4 Data presented in this article confirm these tion of coagulation factors to the overall hemostatic effect results for liquid-stored PLTs, but also demonstrate that must be considered. As expected, the FVIII levels in the several indicators of PLT quality do not change after liquid-stored PLTs were comparable to cryopreserved PLTs cryopreservation as much as might have been expected. reconstituted in FFP or FFP24, but significantly lower than For example, liquid-stored components at expiry and the cryopreserved PLTs reconstituted in DFP (Table 2). cryopreserved components have a comparable collagen The activity of protein S in the PLT concentrate was aggregation response and contain a similar proportion of reduced by Day 5 of PLT storage, compared to fresh (Day activated PLTs (as identified by CD62P and CD63 expres2) and cryopreserved PLTs. FV activity decreased during sion). Further, clinical data indicate that cryopreserved the storage of PLTs and was also significantly lower in the PLTs may have a longer in vivo survival rate than whole cryopreserved PLTs. The reduction in these factors also blood–derived liquid-stored PLTs depending on the resulted in prolongation of the prothrombin time in the method of preparation,5,19 which supports the notion that cryopreserved PLTs, which was similar to liquid-stored traditional in vitro indicators do not truly represent the in PLTs at expiry. vivo potential.20 Volume **, ** **
TRANSFUSION
7
JOHNSON ET AL.
TABLE 2. Coagulation variables of liquid-stored and cryopreserved PLTs* Coagulation parameter APTT (sec) PT (sec) Fibrinogen (g/L) FII (%) FV (%) FVII (%) FVIII (%) FXI (%) FXII (%) Antithrombin (%) Protein C (%) Protein S (%)
Liquid-stored Day 2† Day 5† 36.9 ± 3.2 39.4 ± 2.8 13.7 ± 0.6 15.9 ± 0.7 2.6 ± 0.4 2.7 ± 0.3 88.9 ± 9.5 87.7 ± 7.0 92.1 ± 22.7 65.1 ± 14.0 91.4 ± 17.3 83.2 ± 17.3 58.2 ± 29.4 52.1 ± 30.9 98.7 ± 17.9 99.8 ± 16.0 83.3 ± 22.0 88.3 ± 24.4 102.5 ± 8.1 106.8 ± 11.0 85.4 ± 31.0 87.1 ± 35.5 70.6 ± 23.6 29.1 ± 10.9
DFP† 40.3 ± 15.1 15.1 ± 0.9§|| 2.9 ± 0.6 95.8 ± 7.8 52.3 ± 12.1§|| 100.8 ± 17.0|| 92.0 ± 39.5§|| 103.5 ± 14.1 75.4 ± 18.0 103.5 ± 8.8 82.2 ± 35.6 74.3 ± 25.7||
Freeze-thawed FFP‡ 37.6 ± 3.3 15.2 ± 0.5§ 2.9 ± 0.9 90.5 ± 9.1 47.6 ± 9.6§|| 97.8 ± 9.6 80.6 ± 37.1 108.1 ± 23.1 81.0 ± 18.3 99.1 ± 11.8 85.3 ± 21.1 59.9 ± 9.1||
FFP24‡ 37.9 ± 5.0 15.4 ± 0.9§ 3.2 ± 0.4 87.3 ± 11.9 49.8 ± 13.2§|| 98.8 ± 19.8 50.5 ± 9.5¶ 106.4 ± 19.2 97.9 ± 31.7 95.9 ± 9.6 102.6 ± 27.3 63.5 ± 14.6||
* Values shown as mean ± SD. † n = 12 in each group. ‡ n = 8 in each group. § p < 0.05 between Day 2 and thawed PLTs, determined using ANOVA and Dunnett’s multiple comparison test. || p < 0.05 between Day 5 and thawed PLTs, determined using ANOVA and Dunnett’s multiple comparison test. ¶ p < 0.05 between DFP and indicated plasma, determined using ANOVA and Bonferroni’s multiple comparison test. APTT = activated partial thromboplastin time; PT = prothrombin time.
It is apparent that cryopreserved PLT components have a higher hemostatic potential than liquid-stored PLTs, regardless of the duration of liquid storage within the current shelf life. The cryopreserved PLTs promote faster in vitro clotting, as measured with the TEG and PPL assay, and also contribute to greater and more rapid thrombin generation. The importance of both phosphatidylserine and tissue factor in these processes is also demonstrated. It has been suggested that microparticles support hemostasis in thrombocytopenia and might partially substitute for PLTs.16 Our data support this hypothesis, as the supernatant alone, containing microparticles, is sufficient to reduce the clotting time and induce thrombin generation. However, it remains to be determined whether these in vitro results translate to a similar clinical effect. In an effort to understand the procoagulant capacity of cryopreserved PLTs, endpoint measurements such as TEG or ROTEM should be assessed after PLT transfusion in any clinical trial, and this has been included in the design of the CLIP trial. One of the most striking differences between liquidstored and cryopreserved PLTs is the absolute number of microparticles. While it has previously been suggested that the number of microparticles in cryopreserved units may be similar to, or even lower than, that found in fresh liquid-stored PLTs,5,6 our data clearly demonstrate that regardless of the PLT shelf life, cryopreserved PLTs contain significantly more microparticles than liquid-stored PLTs. PLT microparticles formed as a result of storage are known to express phosphatidylserine and tissue factor.16 This study confirms these findings and extends the context to cryopreserved PLTs. Due to the important role that microparticles appear to play in potentiating the hemostatic activity of cryopreserved PLTs, future detailed char8
TRANSFUSION Volume **, ** **
acterization of their phenotype and functional potential is warranted. Such studies should also examine potential adverse effects of transfusing a greater number of microparticles, such as acute lung injury.21 In addition to microparticles, other soluble mediators that reportedly mediate transfusion reactions were assessed. Thromboxane is released by activated PLTs and is an important participant in hemostasis, acting as a vasoconstrictor and facilitating PLT aggregation at the site of vessel injury.22 The role of thromboxane in mediating the hemostatic effect of cryopreserved PLTs has been demonstrated in previous studies,2,23 but this study demonstrates that the total amount of thromboxane transfused with the cryopreserved PLT component would be 50-fold greater than that for a liquid-stored component. Such a difference could plausibly influence the hemostatic activity of PLTs, both endogenous and transfused. Both IL-27 and RANTES have been shown to be released from activated PLTs and have been associated with acute transfusion reactions.24,25 Our data demonstrate that the transfusion of either a liquid-stored or cryopreserved PLT unit would expose the recipient to similar amounts of these soluble mediators. It is known that the granules within PLTs are repositories for many coagulation factors that are released upon PLT activation.26 As cryopreserved PLTs are reconstituted in plasma before transfusion, and given the coordination between plasma coagulation proteins and PLTs for maintaining hemostasis, the presence and activity of coagulation factors in the transfused units must be considered. Over the shelf-life, the activity of FV, FVII, FVIII, and protein S were reduced in the liquid-stored PLTs, which is to be expected based on previous studies.27 The activity of FV was significantly lower in the cryopreserved PLT group
CRYOPRESERVED VS. LIQUID-STORED PLTs
than stored PLTs, which may be due to freeze-thaw effects, although we hypothesize that FV may be binding to the phospholipid present on the PLTs or microparticles.28,29 Despite some predictable differences in the coagulation factor activity of plasma stored under different conditions, the type of plasma used for reconstitution did not significantly influence PLT quality, as very few differences were seen between the DFP, FFP, or FFP24 groups. This allows flexibility with the choice of reconstitution solution and the potential to eliminate any substantial delay in supplying cryopreserved PLTs from a hospital blood bank. The transfusion of any PLT component is not without risk.30 However, it is important to determine whether cryopreserved PLTs are associated with a greater risk of adverse events than standard liquid-stored PLTs. This study, among others,8,13,23 presents evidence suggesting that cryopreserved PLTs have a greater hemostatic capacity than liquid-stored PLTs. This may raise concerns of prothrombotic complications. Cryopreserved PLTs contain higher amounts of soluble P-selectin (CD62P), elevated microparticle and thrombin generation potential, all of which have been associated with increased risk of venous thromboembolism.31-33 However, the number of PLT transfusions should ideally be titrated to effect, and the greater hemostatic activity of cryopreserved PLTs may allow fewer units to be given, potentially reducing other risks while conferring no greater potential for venous thromboembolism. This uncertain balance of risk and benefit cannot be answered by in vitro laboratory tests. Only a clinical trial actively monitoring possible adverse events, such as the CLIP trial, can answer this question. PLT transfusions are either prophylactic, to prevent spontaneous bleeding in markedly thrombocytopenic patients, such as those treated for hematologic malignancy, or therapeutic, in patients with or at risk of bleeding due to trauma or invasive procedures. Cryopreserved PLTs have been used for both indications.10,34,35 However, the results of this study, together with those previously published demonstrating cryopreservation-associated alterations in PLT surface marker expression, microparticle content, and immediate hemostatic effect, suggest that cryopreserved PLTs may be more suitable for therapeutic rather than prophylactic use. This is further evidenced by clinical studies in both humans and baboons, which demonstrate that cryopreserved PLTs are very capable of stopping bleeding, despite reduced in vivo recovery compared to fresh PLTs.2,5,19,36 This study provides valuable information to influence the design and interpretation of clinical trials of cryopreserved PLTs, such as the CLIP study. The CLIP study will use liquid PLTs stored for varying durations, as it aims to compare cryopreserved PLTs to PLTs in conventional clinical use. If a substantial effect of storage duration on liquidPLT structure and function had been observed, it would have been necessary to understand this potential con-
founding factor in the clinical trial. Further, we have also shown that cryopreserved PLT structure and function are not markedly affected by reconstitution in FFP or FFP24, compared to DFP, which comprises the bulk of cryopreserved PLT experience. Using FFP or FFP24 will be logistically easier to supply should cryopreserved PLTs become part of routine clinical practice. Based on the results of this study, the CLIP trial will be conducted in Australia using cryopreserved PLTs as described in this study, that is, apheresis, leukoreduced, nonirradiated, group O PLTs, resuspended in whole blood–derived group AB FFP or FFP24, to assess the feasibility and safety of cryopreserved PLTs in reducing bleeding in cardiac surgical patients. ACKNOWLEDGMENTS The technical assistance of Rachel Webb and Mirande Gorrion is acknowledged. CONFLICT OF INTEREST The authors have disclosed no conflicts of interest.
REFERENCES 1. Lelkens CC, Koning JG, De Kort B, et al. Experiences with frozen blood products in the Netherlands military. Transfus Apher Sci 2006;34:289-98. 2. Khuri SF, Healey N, Macgregor H, et al. Comparison of the effects of transfusions of cryopreserved and liquidpreserved platelets on hemostasis and blood loss after cardiopulmonary bypass. J Thorac Cardiovasc Surg 1999;117: 172-83. 3. Reade MC, Marks DC, Johnson L, et al. Frozen platelets for rural Australia: the CLIP trial. Anaesth Intensive Care 2013; 41:804-5. 4. Shrivastava M. The platelet storage lesion. Transfus Apher Sci 2009;41:105-13. 5. Dumont LJ, Cancelas JA, Dumont DF, et al. A randomized controlled trial evaluating recovery and survival of 6% dimethyl sulfoxide–frozen autologous platelets in healthy volunteers. Transfusion 2013;53:128-37. 6. Johnson LN, Winter KM, Reid S, et al. Cryopreservation of buffy-coat-derived platelet concentrates in dimethyl sulfoxide and platelet additive solution. Cryobiology 2011;62: 100-6. 7. Noorman F, Strelitski R, Badloe J. −80°C frozen platelets are activated compared to 24 hour liquid stored platelets and quality of frozen platelets is unaffected by a quick preparation method (15 min) which can be used to prepare platelets for the early treatment of trauma patients in military theatre. Transfusion 2012;52:62A. 8. Barnard MR, Macgregor H, Ragno G, et al. Fresh, liquidpreserved, and cryopreserved platelets: adhesive surface receptors and membrane procoagulant activity. Transfusion 1999;39:880-8. Volume **, ** **
TRANSFUSION
9
JOHNSON ET AL.
9. Hornsey VS, McMillan L, Morrison A, et al. Freezing of buffy coat-derived, leukoreduced platelet concentrates in 6 percent dimethyl sulfoxide. Transfusion 2008;48: 2508-14. 10. Noorman F, Badloe J. −80°C frozen platelets, efficient logistics: available, compatible, safe and effective in the treatment of trauma patients with or without massive blood loss in military theatre. Transfusion 2012;52(Suppl): 33A. 11. Johnson L, Reid S, Tan S, et al. Pas-G supports platelet reconstitution after cryopreservation in the absence of plasma. Transfusion 2013;53:2268-77. 12. Council of Europe. Guide to the preparation, use and
22. Fitzgerald G. Mechanisms of platelet activation: thromboxane A2 as an amplifying signal for other agonists. Am J Cardiol 1991;68:11B-15B. 23. Valeri CR, Macgregor H, Ragno G. Correlation between in vitro aggregation and thromboxane A2 production in fresh, liquid-preserved, and cryopreserved human platelets: effect of agonists, pH, and plasma and saline resuspension. Transfusion 2005;45:596-603. 24. Savage WJ, Savage JH, Tobian AA, et al. Allergic agonists in apheresis platelet products are associated with allergic transfusion reactions. Transfusion 2012;52:575-81. 25. Hamzeh-Cognasse H, Damien P, Nguyen KA, et al. Immune-reactive soluble OX40 ligand, soluble CD40
quality assurance of blood components. 14th ed.
ligand, and interleukin-27 are simultaneously oversecreted
Strasbourg, France: Council of Europe Publishing; 2008.
in platelet components associated with acute transfusion
13. Johnson L, Coorey CP, Marks DC. The hemostatic activity
reactions. Transfusion 2014;54:613-25.
of cryopreserved platelets is mediated by phosphatidylserine-expressing platelets and platelet
26. Blair P, Flaumenhaft R. Platelet alpha-granules: basic biology and clinical correlates. Blood Rev 2009;23:177-89.
microparticles. Transfusion 2014;54:1917-26. 14. Johnson L, Winter KM, Kwok M, et al. Evaluation of the quality of blood components prepared using the Reveos
27. Weiss DR, Franke D, Strasser EF, et al. Von Willebrand factor, clotting factors, and clotting inhibitors in apheresis platelet concentrates. Transfusion 2014;54:633-9.
automated blood processing system. Vox Sang 2013;105: 225-35.
28. Chen YW, Chen JK, Wang JS. Strenuous exercise promotes shear-induced thrombin generation by increasing the
15. National Blood Serivce. Guidelines for the blood transfusion services in the UK. 7th ed. London: The Stationery Office; 2005. 16. Keuren JF, Magdeleyns EJ, Govers-Riemslag JW, et al. Effects of storage-induced platelet microparticles on the initiation and propagation phase of blood coagulation. Br J Haematol 2006;134:307-13.
shedding of procoagulant microparticles from platelets. Thromb Haemost 2010;104:293-301. 29. Gilbert GE, Sims PJ, Wiedmer T, et al. Platelet-derived microparticles express high affinity receptors for factor VIII. J Biol Chem 1991;266:17261-8. 30. Katus MC, Szczepiorkowski ZM, Dumont LJ, et al. Safety of platelet transfusion: past, present and future. Vox Sang
17. Marrocco C, D’Alessandro A, Girelli G, et al. Proteomic analysis of platelets treated with gamma irradiation versus
2014;107:103-13. 31. Blann AD, Noteboom WM, Rosendaal FR. Increased
a commercial photochemical pathogen reduction technology. Transfusion 2013;53:1808-20. 18. Picker SM, Steisel A, Gathof BS. Cell integrity and mito-
soluble p-selectin levels following deep venous thrombosis: cause or effect? Br J Haematol 2000;108:191-3. 32. Pabinger I, Thaler J, Ay C. Biomarkers for prediction of
chondrial function after Mirasol-PRT treatment for pathogen reduction of apheresis-derived platelets: results of a three-arm in vitro study. Transfus Apher Sci 2009;40: 79-85. 19. Valeri CR, Macgregor H, Giorgio A, et al. Circulation and hemostatic function of autologous fresh, liquid-preserved, and cryopreserved baboon platelets transfused to correct an aspirin-induced thrombocytopathy. Transfusion 2002; 42:1206-16. 20. Van Der Meer PF, Kerkhoffs JL, Curvers J, et al. In vitro comparison of platelet storage in plasma and in four platelet additive solutions, and the effect of pathogen reduction: a proposal for an in vitro rating system. Vox Sang 2010;98: 517-24. 21. McVey M, Tabuchi A, Kuebler WM. Microparticles and acute lung injury. Am J Physiol Lung Cell Mol Physiol 2012; 303:L364-81.
10
TRANSFUSION Volume **, ** **
venous thromboembolism in cancer. Blood 2013;122: 2011-8. 33. Hernandez C, Orbe J, Roncal C, et al. Tissue factor expressed by microparticles is associated with mortality but not with thrombosis in cancer patients. Thromb Haemost 2013;110:598-608. 34. Pedrazzoli P, Noris P, Perotti C, et al. Transfusion of platelet concentrates cryopreserved with thrombosol plus lowdose dimethylsulphoxide in patients with severe thrombocytopenia: a pilot study. Br J Haematol 2000;108: 653-9. 35. Schiffer CA, Aisner J, Wiernik PH. Clinical experience with transfusion of cryopreserved platelets. Br J Haematol 1976; 34:377-85. 36. Valeri CR, Giorgio A, Macgregor H, et al. Circulation and distribution of autotransfused fresh, liquid-preserved and cryopreserved baboon platelets. Vox Sang 2002;83:347-51.
BACKGROUND: Platelet (PLT) concentrates can be cryopreserved in dimethyl sulfoxide (DMSO) and stored at −80°C for 2 years. These storage conditions improve availability in both rural and military environments. Previous phenotypic and in vitro studies of cryopreserved PLTs are limited by comparison to fresh liquid-stored PLTs, rather than PLTs stored over their clinically relevant shelf life. Further, nothing is known of the effect of reconstituting cryopreserved PLTs in plasma stored at a variety of clinically relevant temperatures. STUDY DESIGN AND METHODS: Apheresis PLTs were either stored at room temperature for 5 days or cryopreserved at −80°C with 5% DMSO. Cryopreserved PLTs were thawed at 37°C and reconstituted in plasma (stored at different temperatures) and compared to fresh and expired liquid-stored PLTs. In vitro assays were performed to assess glycoprotein expression, PLT activity, microparticle content, and function. RESULTS: Compared to liquid PLTs over storage, cryopreserved PLTs had reduced expression of the key glycoprotein receptors GPIbα and GPIIb. However, the proportion of PLTs expressing activation markers CD62P and CD63 was similar between cryopreserved and liquid-stored PLTs at expiry. Cryopreserved PLT components contained significantly higher numbers of phosphatidylserine- and tissue factor–positive microparticles than liquid-stored PLTs, and these microparticles reduced the time to clot formation and increased thrombin generation. CONCLUSION: There are distinct differences between cryopreserved and liquid-stored PLTs. Cryopreserved PLTs also have an enhanced hemostatic activity. Knowledge of these in vitro differences will be essential to understanding the outcomes of a clinical trial comparing cryopreserved PLTs and liquid PLTs stored for various durations.
I
n Australia, and many other countries including the United States, platelets (PLTs) stored under standard blood banking conditions, 20 to 24°C with constant agitation, currently have a shelf life of 5 days. This short shelf life can result in logistic and supply problems, so alternatives permitting longer storage, such as cryopreservation, are being investigated. Cryopreserved PLTs have been used by the military to treat bleeding since 2001.1 To date, only a single clinical trial underpins this practice, which demonstrated that cryopreserved PLTs are hemostatically effective when transfused in patients bleeding after cardiopulmonary bypass.2 Despite this encouraging result, a trial involving 73 patients, only 24 of whom received cryopreserved PLTs, is insufficient to support regulatory approval. In an effort to provide the required clinical evidence to satisfy clinicians and regulatory bodies, a pilot, randomized controlled clinical trial of cryopreserved PLTs, the cryopreserved
ABBREVIATIONS: APC = allophycocyanin; DFP = deep-frozen plasma; FFP24 = fresh-frozen plasma stored at 4°C for up to 24 hours; MA = maximum amplitude; R-time = time to clot initiation; TEG = thromboelastogram. From 1Research and Development, Australian Red Cross Blood Service, Sydney, NSW, Australia; 2Burns, Trauma and Critical Care Research Centre, University of Queensland, Brisbane, Queensland, Australia; and Joint Health Command, Australian Defence Force, Canberra, ACT, Australia. Address correspondence to: Lacey Johnson, Research and Development, Australian Red Cross Blood Service, 17 O’Riordan Street, Alexandria, NSW 2015, Australia; e-mail: [email protected]. We acknowledge that the Australian Governments fully fund the Australian Red Cross Blood Service for the provision of blood products and services to the Australian Community. Parts of this study were funded by the Defence Health Foundation. Received for publication June 30, 2014; revision received September 2, 2014, and accepted September 8, 2014. doi: 10.1111/trf.12915 © 2014 AABB TRANSFUSION **;**:**-**. Volume **, ** **
TRANSFUSION
1
JOHNSON ET AL.
versus liquid PLTs for surgical bleeding (CLIP) trial (ACTRN12612001261808), is planned.3 It is well established that PLTs develop a storage lesion during liquid storage, characterized by a progressive decline in in vitro quality (reviewed by Shrivastava4). While it is known that cryopreserved PLTs are more activated than fresh PLTs (1000 patients transfused) of clinical experience with this product.10 However, the logistic advantages of using DFP in austere environments do not apply to civilian hospital settings. Further, DFP is not a standard product in Australia. As such, for logistic and regulatory reasons, the use of fresh-frozen plasma (FFP; plasma frozen and stored at −30°C) for PLT reconstitution in a civilian hospital setting would be preferable. Some hospital blood banks routinely keep a unit of thawed FFP stored at 4°C for up to 24 hours (FFP24), for use in emergencies. Using prethawed plasma would allow cryopreserved PLTs to be supplied in roughly the same time frame as liquid-stored PLTs. However, there are currently no data to support the use of FFP or FFP24 for reconstitution of cryopreserved PLTs. The aim of this study was therefore to comprehensively characterize and compare the in vitro quality of liquid-stored PLTs over the entire shelf life, such as will be used in the CLIP study, with cryopreserved PLTs. In addition, we aimed to compare the effects of different plasma types for PLT reconstitution.
MATERIALS AND METHODS This study had approval from the Australian Red Cross Blood Service Human Research Ethics Committee. All donations were from eligible, voluntary donors. Leukoreduced (200 6.4-7.4
Thawed PLTs FFP24‡
DFP†
FFP‡
299.6 ± 10.4§ 299.9 ± 36.4§|| 7.20 ± 0.02§
300.7 ± 24.1§ 273 ± 39.1|| 7.22 ± 0.01§
Specifications
296.5 ± 16.7§ 252.86 ± 26.2¶ 7.24 ± 0.05§¶
250-350 >200 6.4-7.4
* Values shown as mean ± SD. † n = 12 in each group. ‡ n = 8 in each group. § p < 0.05 between Day 2 and thawed PLTs, determined using ANOVA and Dunnett’s multiple comparison test. || p < 0.05 between Day 5 and thawed PLTs, determined using ANOVA and Dunnett’s multiple comparison test. ¶ p < 0.05 between DFP and indicated plasma, determined using ANOVA and Bonferroni’s multiple comparison test. NA = not applicable.
Volume **, ** **
TRANSFUSION
3
JOHNSON ET AL.
expression of these markers on the cryopreserved PLTs was equivalent to liquid PLTs stored for 5 days. Fresh liquid-stored and cryopreserved PLTs expressed similar basal levels of activated GPIIIa receptor (PAC-1; Fig. 1G). However, cryopreserved PLTs were unable to respond to ADP to induce receptor activation, whereas the liquidstored PLTs maintained a response even after 5 days of storage (Fig. 1H). The type of plasma used for reconstitution did not alter PLT receptor expression. The ability of PLTs to aggregate in response to agonists is an indicator of functional quality. PLT aggregation in response to both collagen and ADP decreased during the 5-day storage period (Figs. 2A-2D). Although the aggregation responses in thawed cryopreserved PLTs were significantly reduced compared to fresh PLTs (Day 2), the average collagen aggregation response was equivalent to Day 5 PLTs (Figs. 2A and 2B). The type of plasma used for reconstitution did not influence the aggregation response. TEG assesses PLT contribution to global clot formation. The R-time was not affected by storage; however, cryopreserved PLTs had a more rapid clotting response than liquid-stored PLTs (Figs. 2E and 2F). Cryopreservation also resulted in a slight, but significant, reduction in clot strength compared to liquid-stored PLTs, as exhibited by the decreased MA (Figs. 2E and 2G). The use of DFP, FFP, or FFP24 resulted in similar TEG responses. PLT microparticles contribute to hemostasis and their concentration is Fig. 1. The surface expression of PLT receptors vary between liquid-stored and reported to increase with storage duracryopreserved PLTs. PLTs were sampled on Days 2 and 5 of liquid storage or immedition and after cryopreservation.13,16 The ately after reconstitution in the indicated plasma. PLTs were stained with the indiabsolute number of phosphatidylserine cated antibodies—(A) CD61-FITC, (B) CD42a-PE, (C) CD41a-FITC, (D) CD42b-PE, (E) -expressing PLT microparticles did not CD62P-PE, (F) CD63-PE, (G) PAC-1-FITC, (H) PAC-1-FITC after stimulation with increase during PLT storage (Fig. 3A), ADP—and analyzed by flow cytometry. The data represent mean ± SD (error bars); but was significantly higher after *p < 0.05 compared to Day 2 liquid storage; †p < 0.05 compared to Day 5 liquid cryopreservation. The absolute number storage. of tissue factor–positive microparticles was much lower than phosphatidylserine-expressing microparticles and simiwas significantly reduced in the cryopreserved PLT comlarly did not increase during storage. Interestingly, ponents (Figs. 1C and 1D), compared to liquid-stored the DFP-reconstituted cryopreserved PLTs contained sigunits. As expected, the proportion of PLTs expressing the nificantly more tissue factor–positive microparticles activation markers P-selectin (CD62P) and CD63 (Fig. 3B). increased during storage (Figs. 1E and 1F), whereas 4
TRANSFUSION Volume **, ** **
CRYOPRESERVED VS. LIQUID-STORED PLTs
Fig. 2. PLT aggregation and TEG responses vary between liquid-stored and cryopreserved PLTs. PLTs were sampled on Days 2 and 5 of liquid storage or immediately after reconstitution in the indicated plasma. PLT aggregation was analyzed using an aggregometer and the following agonists: (A, B) 10 μg/mL collagen or (C, D) 20 μmol/L ADP, with a representative trace shown and the mean ± SD (error bars). TEG variables were assessed using PLTs activated with kaolin, with (E) a representative trace shown and the mean ± SD (error bars) for (F) R-time and (G) MA. *p < 0.05 compared to Day 2 liquid storage; †p < 0.05 compared to Day 5 liquid storage. Volume **, ** **
TRANSFUSION
5
JOHNSON ET AL.
Fig. 3. PLT microparticle number and functionality vary between liquid-stored and cryopreserved PLTs. PLT microparticles were stained with (A) CD61-APC and annexin V-FITC or (B) CD61-APC and CD142-PE and enumerated using flow cytometry. Phosphatidylserine (PS)- and tissue factor (TF)-dependent microparticle function was measured from supernatant. PS-dependent (C) clotting time was measured with the Procoag-PPL assay and (D, E) thrombin generation was initiated using 1 pmol/L tissue factor (PRP reagent). TF-dependent (F, G) thrombin generation was initiated with 4 μmol/L phospholipids (MP reagent). The data represent mean ± SD (error bars); *p < 0.05 compared to Day 2 liquid storage; †p < 0.05 compared to Day 5 liquid storage.
6
TRANSFUSION Volume **, ** **
The contribution of microparticles to hemostatic function was assessed from the supernatant fraction, using the PPL clotting assay, which is dependent on the presence of phosphatidylserine. The clotting time was not affected by storage, but was significantly reduced after cryopreservation (Fig. 3C). Again, the type of plasma used for PLT reconstitution had no effect on clotting time. The functional effect of these microparticles was also assessed by measuring their contribution to thrombin generation using the PRP reagent, which relies on the presence of phosphatidylserine in the sample. The lag time to thrombin generation was not affected by storage or the type of plasma used for reconstitution; however, cryopreserved PLTs had a significantly shorter lag time than liquid-stored PLTs at any time during storage (Fig. 3D). The peak amount of thrombin capable of being generated was significantly higher in the cryopreserved PLTs, compared to the liquid-stored PLTs (Fig. 3E). Thrombin generation can also be triggered by the presence of tissue factor. The contribution of tissue factor to the thrombin-generating capacity was assessed using the MP reagent (Figs. 3F and 3G). The data demonstrate that the peak of thrombin generation was increased during PLT storage. However, the supernatants of cryopreserved PLTs were still capable of faster and greater tissue factor–induced thrombin generation than liquid-stored PLTs. In summary, it is likely that the greater number of microparticles present in the supernatant of the cryopreserved units (regardless of type of plasma used for reconstitution) may be responsible for mediating the phosphatidylserine- and tissue factor– dependent procoagulant activity. The presence of clinically relevant soluble factors was also measured in the PLT supernatant. The absolute amount of P-selectin (soluble CD62P) was higher in the cryopreserved units (Fig. 4A), compared to liquid-stored PLTs throughout the shelf life. Similarly, thromboxane levels were 50-fold higher in the cryopreserved PLT units,
CRYOPRESERVED VS. LIQUID-STORED PLTs
DISCUSSION In this in vitro study, a comprehensive analysis of the characteristics of thawed cryopreserved PLTs was compared to liquid-stored apheresis PLTs over their shelf life (2-5 days postcollection). Cryopreserved PLT components have a greater volume, but contain a similar number of PLTs to liquid-stored units. Cryopreserved PLT units also have significantly more microparticles than liquid-stored PLTs and are more hemostatically active, despite having a lower expression of important PLT glycoproteins and compromised aggregation responses. This study shows, for the first time, that these differences exist regardless of the duration of storage for liquid PLTs up to the current shelf life (5 days). This study differs from previous Fig. 4. The amount of soluble mediators varies between liquid-stored and comparisons of cryopreserved PLTs as it cryopreserved PLTs. PLT supernatant was collected from the PLT components and assessed the functional characteristics the concentration of soluble (A) P-selectin, (B) thromboxane, (C) RANTES, and (D) of cryopreserved PLTs compared to IL-27 was measured by ELISA. The concentration was multiplied by the unit volume those of liquid-stored PLTs over the to report the total dose/unit. The data represent mean ± SD (error bars); *p < 0.05 potential transfusable shelf life, rather compared to Day 2 liquid storage; †p < 0.05 compared to Day 5 liquid storage. than immediately after collection. Further, in previous prefreeze and postthaw comparisons,11,13 the prefreeze PLTs were not irradiated. In this study, the liquid-stored PLT components were irradiated before compared to the liquid-stored PLTs, but were not affected release, as this is standard practice in Australia for all by storage time or the plasma used for reconstituliquid PLT components, while the cryopreserved PLTs tion (Fig. 4B). RANTES appears to be released from were not. Gamma irradiation is known to damage PLTs;17,18 the PLT granules after cryopreservation, but was at a similar level as PLTs stored for their shelf life (5 days; however, the changes induced by irradiation are likely Fig. 4C), while the amount of IL-27 present in the PLT minimal compared to those induced by cryopreservation.5 components was comparable between liquid-stored It is well established that PLTs undergo changes and cryopreserved PLTs, indicating that this cytokine was during ex vivo storage, known collectively as the PLT not released from the PLTs in substantial amounts storage lesion, evidenced by altered metabolism, (Fig. 4D). increased expression of activation markers (CD62P, CD63, Both liquid-stored and cryopreserved apheresis PLT and phosphatidylserine), and decreased aggregation components are suspended in plasma, so the contriburesponse.4 Data presented in this article confirm these tion of coagulation factors to the overall hemostatic effect results for liquid-stored PLTs, but also demonstrate that must be considered. As expected, the FVIII levels in the several indicators of PLT quality do not change after liquid-stored PLTs were comparable to cryopreserved PLTs cryopreservation as much as might have been expected. reconstituted in FFP or FFP24, but significantly lower than For example, liquid-stored components at expiry and the cryopreserved PLTs reconstituted in DFP (Table 2). cryopreserved components have a comparable collagen The activity of protein S in the PLT concentrate was aggregation response and contain a similar proportion of reduced by Day 5 of PLT storage, compared to fresh (Day activated PLTs (as identified by CD62P and CD63 expres2) and cryopreserved PLTs. FV activity decreased during sion). Further, clinical data indicate that cryopreserved the storage of PLTs and was also significantly lower in the PLTs may have a longer in vivo survival rate than whole cryopreserved PLTs. The reduction in these factors also blood–derived liquid-stored PLTs depending on the resulted in prolongation of the prothrombin time in the method of preparation,5,19 which supports the notion that cryopreserved PLTs, which was similar to liquid-stored traditional in vitro indicators do not truly represent the in PLTs at expiry. vivo potential.20 Volume **, ** **
TRANSFUSION
7
JOHNSON ET AL.
TABLE 2. Coagulation variables of liquid-stored and cryopreserved PLTs* Coagulation parameter APTT (sec) PT (sec) Fibrinogen (g/L) FII (%) FV (%) FVII (%) FVIII (%) FXI (%) FXII (%) Antithrombin (%) Protein C (%) Protein S (%)
Liquid-stored Day 2† Day 5† 36.9 ± 3.2 39.4 ± 2.8 13.7 ± 0.6 15.9 ± 0.7 2.6 ± 0.4 2.7 ± 0.3 88.9 ± 9.5 87.7 ± 7.0 92.1 ± 22.7 65.1 ± 14.0 91.4 ± 17.3 83.2 ± 17.3 58.2 ± 29.4 52.1 ± 30.9 98.7 ± 17.9 99.8 ± 16.0 83.3 ± 22.0 88.3 ± 24.4 102.5 ± 8.1 106.8 ± 11.0 85.4 ± 31.0 87.1 ± 35.5 70.6 ± 23.6 29.1 ± 10.9
DFP† 40.3 ± 15.1 15.1 ± 0.9§|| 2.9 ± 0.6 95.8 ± 7.8 52.3 ± 12.1§|| 100.8 ± 17.0|| 92.0 ± 39.5§|| 103.5 ± 14.1 75.4 ± 18.0 103.5 ± 8.8 82.2 ± 35.6 74.3 ± 25.7||
Freeze-thawed FFP‡ 37.6 ± 3.3 15.2 ± 0.5§ 2.9 ± 0.9 90.5 ± 9.1 47.6 ± 9.6§|| 97.8 ± 9.6 80.6 ± 37.1 108.1 ± 23.1 81.0 ± 18.3 99.1 ± 11.8 85.3 ± 21.1 59.9 ± 9.1||
FFP24‡ 37.9 ± 5.0 15.4 ± 0.9§ 3.2 ± 0.4 87.3 ± 11.9 49.8 ± 13.2§|| 98.8 ± 19.8 50.5 ± 9.5¶ 106.4 ± 19.2 97.9 ± 31.7 95.9 ± 9.6 102.6 ± 27.3 63.5 ± 14.6||
* Values shown as mean ± SD. † n = 12 in each group. ‡ n = 8 in each group. § p < 0.05 between Day 2 and thawed PLTs, determined using ANOVA and Dunnett’s multiple comparison test. || p < 0.05 between Day 5 and thawed PLTs, determined using ANOVA and Dunnett’s multiple comparison test. ¶ p < 0.05 between DFP and indicated plasma, determined using ANOVA and Bonferroni’s multiple comparison test. APTT = activated partial thromboplastin time; PT = prothrombin time.
It is apparent that cryopreserved PLT components have a higher hemostatic potential than liquid-stored PLTs, regardless of the duration of liquid storage within the current shelf life. The cryopreserved PLTs promote faster in vitro clotting, as measured with the TEG and PPL assay, and also contribute to greater and more rapid thrombin generation. The importance of both phosphatidylserine and tissue factor in these processes is also demonstrated. It has been suggested that microparticles support hemostasis in thrombocytopenia and might partially substitute for PLTs.16 Our data support this hypothesis, as the supernatant alone, containing microparticles, is sufficient to reduce the clotting time and induce thrombin generation. However, it remains to be determined whether these in vitro results translate to a similar clinical effect. In an effort to understand the procoagulant capacity of cryopreserved PLTs, endpoint measurements such as TEG or ROTEM should be assessed after PLT transfusion in any clinical trial, and this has been included in the design of the CLIP trial. One of the most striking differences between liquidstored and cryopreserved PLTs is the absolute number of microparticles. While it has previously been suggested that the number of microparticles in cryopreserved units may be similar to, or even lower than, that found in fresh liquid-stored PLTs,5,6 our data clearly demonstrate that regardless of the PLT shelf life, cryopreserved PLTs contain significantly more microparticles than liquid-stored PLTs. PLT microparticles formed as a result of storage are known to express phosphatidylserine and tissue factor.16 This study confirms these findings and extends the context to cryopreserved PLTs. Due to the important role that microparticles appear to play in potentiating the hemostatic activity of cryopreserved PLTs, future detailed char8
TRANSFUSION Volume **, ** **
acterization of their phenotype and functional potential is warranted. Such studies should also examine potential adverse effects of transfusing a greater number of microparticles, such as acute lung injury.21 In addition to microparticles, other soluble mediators that reportedly mediate transfusion reactions were assessed. Thromboxane is released by activated PLTs and is an important participant in hemostasis, acting as a vasoconstrictor and facilitating PLT aggregation at the site of vessel injury.22 The role of thromboxane in mediating the hemostatic effect of cryopreserved PLTs has been demonstrated in previous studies,2,23 but this study demonstrates that the total amount of thromboxane transfused with the cryopreserved PLT component would be 50-fold greater than that for a liquid-stored component. Such a difference could plausibly influence the hemostatic activity of PLTs, both endogenous and transfused. Both IL-27 and RANTES have been shown to be released from activated PLTs and have been associated with acute transfusion reactions.24,25 Our data demonstrate that the transfusion of either a liquid-stored or cryopreserved PLT unit would expose the recipient to similar amounts of these soluble mediators. It is known that the granules within PLTs are repositories for many coagulation factors that are released upon PLT activation.26 As cryopreserved PLTs are reconstituted in plasma before transfusion, and given the coordination between plasma coagulation proteins and PLTs for maintaining hemostasis, the presence and activity of coagulation factors in the transfused units must be considered. Over the shelf-life, the activity of FV, FVII, FVIII, and protein S were reduced in the liquid-stored PLTs, which is to be expected based on previous studies.27 The activity of FV was significantly lower in the cryopreserved PLT group
CRYOPRESERVED VS. LIQUID-STORED PLTs
than stored PLTs, which may be due to freeze-thaw effects, although we hypothesize that FV may be binding to the phospholipid present on the PLTs or microparticles.28,29 Despite some predictable differences in the coagulation factor activity of plasma stored under different conditions, the type of plasma used for reconstitution did not significantly influence PLT quality, as very few differences were seen between the DFP, FFP, or FFP24 groups. This allows flexibility with the choice of reconstitution solution and the potential to eliminate any substantial delay in supplying cryopreserved PLTs from a hospital blood bank. The transfusion of any PLT component is not without risk.30 However, it is important to determine whether cryopreserved PLTs are associated with a greater risk of adverse events than standard liquid-stored PLTs. This study, among others,8,13,23 presents evidence suggesting that cryopreserved PLTs have a greater hemostatic capacity than liquid-stored PLTs. This may raise concerns of prothrombotic complications. Cryopreserved PLTs contain higher amounts of soluble P-selectin (CD62P), elevated microparticle and thrombin generation potential, all of which have been associated with increased risk of venous thromboembolism.31-33 However, the number of PLT transfusions should ideally be titrated to effect, and the greater hemostatic activity of cryopreserved PLTs may allow fewer units to be given, potentially reducing other risks while conferring no greater potential for venous thromboembolism. This uncertain balance of risk and benefit cannot be answered by in vitro laboratory tests. Only a clinical trial actively monitoring possible adverse events, such as the CLIP trial, can answer this question. PLT transfusions are either prophylactic, to prevent spontaneous bleeding in markedly thrombocytopenic patients, such as those treated for hematologic malignancy, or therapeutic, in patients with or at risk of bleeding due to trauma or invasive procedures. Cryopreserved PLTs have been used for both indications.10,34,35 However, the results of this study, together with those previously published demonstrating cryopreservation-associated alterations in PLT surface marker expression, microparticle content, and immediate hemostatic effect, suggest that cryopreserved PLTs may be more suitable for therapeutic rather than prophylactic use. This is further evidenced by clinical studies in both humans and baboons, which demonstrate that cryopreserved PLTs are very capable of stopping bleeding, despite reduced in vivo recovery compared to fresh PLTs.2,5,19,36 This study provides valuable information to influence the design and interpretation of clinical trials of cryopreserved PLTs, such as the CLIP study. The CLIP study will use liquid PLTs stored for varying durations, as it aims to compare cryopreserved PLTs to PLTs in conventional clinical use. If a substantial effect of storage duration on liquidPLT structure and function had been observed, it would have been necessary to understand this potential con-
founding factor in the clinical trial. Further, we have also shown that cryopreserved PLT structure and function are not markedly affected by reconstitution in FFP or FFP24, compared to DFP, which comprises the bulk of cryopreserved PLT experience. Using FFP or FFP24 will be logistically easier to supply should cryopreserved PLTs become part of routine clinical practice. Based on the results of this study, the CLIP trial will be conducted in Australia using cryopreserved PLTs as described in this study, that is, apheresis, leukoreduced, nonirradiated, group O PLTs, resuspended in whole blood–derived group AB FFP or FFP24, to assess the feasibility and safety of cryopreserved PLTs in reducing bleeding in cardiac surgical patients. ACKNOWLEDGMENTS The technical assistance of Rachel Webb and Mirande Gorrion is acknowledged. CONFLICT OF INTEREST The authors have disclosed no conflicts of interest.
REFERENCES 1. Lelkens CC, Koning JG, De Kort B, et al. Experiences with frozen blood products in the Netherlands military. Transfus Apher Sci 2006;34:289-98. 2. Khuri SF, Healey N, Macgregor H, et al. Comparison of the effects of transfusions of cryopreserved and liquidpreserved platelets on hemostasis and blood loss after cardiopulmonary bypass. J Thorac Cardiovasc Surg 1999;117: 172-83. 3. Reade MC, Marks DC, Johnson L, et al. Frozen platelets for rural Australia: the CLIP trial. Anaesth Intensive Care 2013; 41:804-5. 4. Shrivastava M. The platelet storage lesion. Transfus Apher Sci 2009;41:105-13. 5. Dumont LJ, Cancelas JA, Dumont DF, et al. A randomized controlled trial evaluating recovery and survival of 6% dimethyl sulfoxide–frozen autologous platelets in healthy volunteers. Transfusion 2013;53:128-37. 6. Johnson LN, Winter KM, Reid S, et al. Cryopreservation of buffy-coat-derived platelet concentrates in dimethyl sulfoxide and platelet additive solution. Cryobiology 2011;62: 100-6. 7. Noorman F, Strelitski R, Badloe J. −80°C frozen platelets are activated compared to 24 hour liquid stored platelets and quality of frozen platelets is unaffected by a quick preparation method (15 min) which can be used to prepare platelets for the early treatment of trauma patients in military theatre. Transfusion 2012;52:62A. 8. Barnard MR, Macgregor H, Ragno G, et al. Fresh, liquidpreserved, and cryopreserved platelets: adhesive surface receptors and membrane procoagulant activity. Transfusion 1999;39:880-8. Volume **, ** **
TRANSFUSION
9
JOHNSON ET AL.
9. Hornsey VS, McMillan L, Morrison A, et al. Freezing of buffy coat-derived, leukoreduced platelet concentrates in 6 percent dimethyl sulfoxide. Transfusion 2008;48: 2508-14. 10. Noorman F, Badloe J. −80°C frozen platelets, efficient logistics: available, compatible, safe and effective in the treatment of trauma patients with or without massive blood loss in military theatre. Transfusion 2012;52(Suppl): 33A. 11. Johnson L, Reid S, Tan S, et al. Pas-G supports platelet reconstitution after cryopreservation in the absence of plasma. Transfusion 2013;53:2268-77. 12. Council of Europe. Guide to the preparation, use and
22. Fitzgerald G. Mechanisms of platelet activation: thromboxane A2 as an amplifying signal for other agonists. Am J Cardiol 1991;68:11B-15B. 23. Valeri CR, Macgregor H, Ragno G. Correlation between in vitro aggregation and thromboxane A2 production in fresh, liquid-preserved, and cryopreserved human platelets: effect of agonists, pH, and plasma and saline resuspension. Transfusion 2005;45:596-603. 24. Savage WJ, Savage JH, Tobian AA, et al. Allergic agonists in apheresis platelet products are associated with allergic transfusion reactions. Transfusion 2012;52:575-81. 25. Hamzeh-Cognasse H, Damien P, Nguyen KA, et al. Immune-reactive soluble OX40 ligand, soluble CD40
quality assurance of blood components. 14th ed.
ligand, and interleukin-27 are simultaneously oversecreted
Strasbourg, France: Council of Europe Publishing; 2008.
in platelet components associated with acute transfusion
13. Johnson L, Coorey CP, Marks DC. The hemostatic activity
reactions. Transfusion 2014;54:613-25.
of cryopreserved platelets is mediated by phosphatidylserine-expressing platelets and platelet
26. Blair P, Flaumenhaft R. Platelet alpha-granules: basic biology and clinical correlates. Blood Rev 2009;23:177-89.
microparticles. Transfusion 2014;54:1917-26. 14. Johnson L, Winter KM, Kwok M, et al. Evaluation of the quality of blood components prepared using the Reveos
27. Weiss DR, Franke D, Strasser EF, et al. Von Willebrand factor, clotting factors, and clotting inhibitors in apheresis platelet concentrates. Transfusion 2014;54:633-9.
automated blood processing system. Vox Sang 2013;105: 225-35.
28. Chen YW, Chen JK, Wang JS. Strenuous exercise promotes shear-induced thrombin generation by increasing the
15. National Blood Serivce. Guidelines for the blood transfusion services in the UK. 7th ed. London: The Stationery Office; 2005. 16. Keuren JF, Magdeleyns EJ, Govers-Riemslag JW, et al. Effects of storage-induced platelet microparticles on the initiation and propagation phase of blood coagulation. Br J Haematol 2006;134:307-13.
shedding of procoagulant microparticles from platelets. Thromb Haemost 2010;104:293-301. 29. Gilbert GE, Sims PJ, Wiedmer T, et al. Platelet-derived microparticles express high affinity receptors for factor VIII. J Biol Chem 1991;266:17261-8. 30. Katus MC, Szczepiorkowski ZM, Dumont LJ, et al. Safety of platelet transfusion: past, present and future. Vox Sang
17. Marrocco C, D’Alessandro A, Girelli G, et al. Proteomic analysis of platelets treated with gamma irradiation versus
2014;107:103-13. 31. Blann AD, Noteboom WM, Rosendaal FR. Increased
a commercial photochemical pathogen reduction technology. Transfusion 2013;53:1808-20. 18. Picker SM, Steisel A, Gathof BS. Cell integrity and mito-
soluble p-selectin levels following deep venous thrombosis: cause or effect? Br J Haematol 2000;108:191-3. 32. Pabinger I, Thaler J, Ay C. Biomarkers for prediction of
chondrial function after Mirasol-PRT treatment for pathogen reduction of apheresis-derived platelets: results of a three-arm in vitro study. Transfus Apher Sci 2009;40: 79-85. 19. Valeri CR, Macgregor H, Giorgio A, et al. Circulation and hemostatic function of autologous fresh, liquid-preserved, and cryopreserved baboon platelets transfused to correct an aspirin-induced thrombocytopathy. Transfusion 2002; 42:1206-16. 20. Van Der Meer PF, Kerkhoffs JL, Curvers J, et al. In vitro comparison of platelet storage in plasma and in four platelet additive solutions, and the effect of pathogen reduction: a proposal for an in vitro rating system. Vox Sang 2010;98: 517-24. 21. McVey M, Tabuchi A, Kuebler WM. Microparticles and acute lung injury. Am J Physiol Lung Cell Mol Physiol 2012; 303:L364-81.
10
TRANSFUSION Volume **, ** **
venous thromboembolism in cancer. Blood 2013;122: 2011-8. 33. Hernandez C, Orbe J, Roncal C, et al. Tissue factor expressed by microparticles is associated with mortality but not with thrombosis in cancer patients. Thromb Haemost 2013;110:598-608. 34. Pedrazzoli P, Noris P, Perotti C, et al. Transfusion of platelet concentrates cryopreserved with thrombosol plus lowdose dimethylsulphoxide in patients with severe thrombocytopenia: a pilot study. Br J Haematol 2000;108: 653-9. 35. Schiffer CA, Aisner J, Wiernik PH. Clinical experience with transfusion of cryopreserved platelets. Br J Haematol 1976; 34:377-85. 36. Valeri CR, Giorgio A, Macgregor H, et al. Circulation and distribution of autotransfused fresh, liquid-preserved and cryopreserved baboon platelets. Vox Sang 2002;83:347-51.
