Relating type-structures partial variations on a theme of Friedman and Statman - Springer
Academic edition
Find out how to access preview-only content
Relating type-structures partial variations on a theme of Friedman and StatmanAndrea AspertiAffiliated withDipartimento di Informatica, Giuseppe LongoAffiliated withDipartimento di Informatica
* Final gross prices may vary according to local VAT.Proof nets Construction and Automated Deduction in Non-Commutative Linear Logic: extended abstract
PasswordRemember meSign in via your institutionSign in via your institution
Loading...
You have selected 1 citation for export.
RIS (for EndNote, ReferenceManager, ProCite)
Citation Only
Citation and Abstract
JavaScript is disabled on your browser. Please enable JavaScript to use all the features on this page.
, 1998, Pages 1-20
Open Access
Proof nets Construction and Automated Deduction in Non-Commutative Linear Logic: extended abstract
Opens overlay
Opens overlay
aLORIA UMR 7503 – Univ. Henri Poincaré Campus Scientifique, BP 239, 54506 Vandoeuvre-lès-Nancy Cedex, FrancebEcole Normale Supérieure de Lyon 46, Allée d'Italie, 69364 Lyon Cedex 07, France
Under a Creative Commons
Proof nets can be seen as a multiple conclusion natural deduction system for Linear Logic (LL) and form a good formalism to analyze some computation mechanisms, for instance in type-theoretic interpretations. This paper presents an algorithm for automated proof nets construction in the non-commutative multiplicative linear logic that is useful for applications including planning, concurrency or sequentiality. The properties of this algorithm can be proved from a recently defined graph-theoretic characterization of non-commutative proof nets. Involving simple construction principles improved in the commutative case, it leads also to a new proof search method for the non-commutative fragment. Moreover because of the relationships between the non-commutative linear logic and the Lambek calculus we can derive from it an alternate method for automatic construction of proof nets in this calculus.
V.M. Abrusci
Phase semantics and sequent calculus for pure noncommutative classical linear propositional logic
Journal of Symbolic Logic, 56 (4) (1991), pp.
V.M. Abrusci
Noncommutative proof nets
J.Y. Girard, Y. Lafont, L. Regnier (Eds.), Advances in Linear Logic, Cambridge University Press (1995), pp. 271–296
V.M. Abrusci. Lambek calculus, cyclic multiplicative-additive linear logic, noncommutative multiplicative-additive linear logic: language and sequent calculus. In V.M. Abrusci and C. Casadio, editors,
Proofs and Linguistic Categories – Proceedings 1996 Roma Workshop, pages 21–48, 1996.
A. Asperti and G. Dore. Yet another correctness criterion for multiplicative linear logic with mix. In
Logic at St Petersburg '94, Symposium on Logical Foundations of Computer Science, LNCS 813, pages 34–46, St Petersburg, Russia, July 1994.
G. Bellin, J. van de Wiele
Subnets of proof-nets in MLL-
J.Y. Girard, Y. Lafont, L. Regnier (Eds.), Advances in Linear Logic, Cambridge University Press (1995), pp. 249–260
V. Danos, L. Régnier
The structures of multiplicatives
Arch. Math. Logic, 28 (1989), pp. 181–203
D. Galmiche. Connection methods in Linear Logic and Proof nets Construction.
Theoretical Computer Science, 1998. Accepted for publication.
D. Galmiche and B. Martin. Automated proof nets construction in multiplicative and additive linear logic. Technical report, CRIN-CNRS, 1997.
D. Galmiche and B. Martin. Proof search and proof nets construction in linear logic. In
4th Workshop on Logic, Language, Information and Computation, Wollic'97, Fortaleza, Brasil, August 1997. Logic Journal of IGPL, vol. 5–6, pp 883–885.
D. Galmiche and G. Perrier. A procedure for automatic proof nets construction. In
LPAR'92, International Conference on Logic Programming and Automated Reasoning, LNAI 624, pages 42–53, St. Petersburg, Russia, July 1992.
D. Galmiche, G. Perrier
On proof normalization in linear logic
Theoretical Computer Science, 135 (1) (1994), pp. 67–110
J.Y. Girard
Linear logic
Theoretical Computer Science, 50 (1) (1987), pp. 1–102
J.Y. Girard. Towards a geometry of interaction. In J.W. Gray and A. Scedrov, editors,
Conference on Categories in Computer Science and Logic, pages 69–108, Boulder, Colorado, June 1987.
J.Y. Girard
Linear logic: its syntax and semantics
J.Y. Girard, Y. Lafont, L. Regnier (Eds.), Advances in Linear Logic, Cambridge University Press (1995), pp. 1–42
J.Y. Girard
Proof nets: the parallel syntax for proof theory
Ursini, Agliano (Eds.), Logic and Algebra, M. Dekker, New York (1995)
G. Gonthier, M. Abadi, and J.J. Levy. Linear logic without boxes. In
7th IEEE Symposium on Logic in Computer Science, pages 223–234, Santa-Cruz, California, June 1992.
P. Lincoln, J. Mitchell, A. Scedrov, N. Shankar
Decision problems for propositional linear logic
Annals of Pure and Applied Logic, 56 (1992), pp. 239–311
G. Morrill
Clausal proofs and discontinuity
L. J. of the IGPL, 3 (2,3) (1995), pp. 403–427
M. Nagayama, M. Okada
A graph-theoretic characterization theorem for multiplicative fragment of non-commutative linear logic – extended abstract
Electronic Notes in Theoretical Computer Science, 3 (1996), p. 11
M. Pentus. Equivalence of multiplicative fragments of cyclic linear logic and noncommutative linear logic. In 4th
Int. Symposium LFCS'97, LNCS 1234, pages 306–311, Yaroslavl, Russia, 1997.
Proof Nets for Lambek Calculus
Journal of Logic and Computation, 2 (2) (1992), pp. 211–231
D.N. Yetter
Quantales and (noncommutative) linear logic
Journal of Symbolic Logic, 55 (1) (1990), pp. 41–64
Remember meUpdate on the use of pathogen-reduced human plasma and platelet concentrates - Seltsam - 2013 - British Journal of Haematology -
Wiley Online Library
Advertisement
The use of pathogen reduction technologies (PRTs) for labile blood components is slowly but steadily increasing. While pathogen-reduced plasma is already used routinely, efficacy and safety concerns impede the widespread use of pathogen-reduced platelets. The supportive and often prophylactic nature of blood component therapy in a variety of clinical situations complicates the clinical evaluation of these novel blood products. However, an increasing body of evidence on the clinical efficacy, safety, cost-benefit ratio and development of novel technologies suggests that pathogen reduction has entered a stage of maturity that could further increase the safety margin in haemotherapy. This review summarizes the clinical evidence on PRTs for plasma and platelet products that are currently licensed or under development.Improvements in donor screening and testing have significantly reduced the risk of transfusion-transmitted infections. Also, evidence-based transfusion guidelines as well as patient blood management programmes aiming at the optimal use of blood components contribute to blood safety by limiting the exposure of patients to potentially infectious blood. However, a relevant risk of transmission of viruses, bacteria, protozoa and prions to recipients remains. The relatively high incidence of bacterial contamination of platelet products, pathogens not detected by conventional blood screening methods, viruses missed due to low titres (window period), and newly emerging transfusion-transmitted agents continue to threaten the safety of the blood supply. Pathogen reduction technologies (PRTs) have the potential to close or at least reduce this safety gap. Whereas early PRTs, such as solvent/detergent (S/D) and methylene blue (MB) treatment, are only suitable for plasma, newer methods can treat fragile cellular blood components. These new-generation methods use ultraviolet (UV) light with or without photosensitizers or photoreagents and inactivate pathogens without destroying the membranes of cellular therapeutic products (Webert et&al, ). Most of them have already been evaluated in plasma and platelets, and their potential use in red blood cells (RBCs) is being explored. Novel PRTs allow for a more generalized approach to pathogen inactivation and pave the way for a paradigm shift in transfusion medicine.Pathogen reduction technologies have moved the current blood safety paradigm from reactive to proactive (McCullough, ). A consensus conference organized by the Canadian Blood Service and H&ma-Qu&bec in 2007 recommended the implementation of pathogen reduction as soon as a feasible and safe method to inactivate a broad spectrum of transfusion-relevant pathogens became available (Klein et&al, ). It was also concluded that the decision whether to introduce pathogen reduction should not be delayed until PRTs suitable for all cellular blood components became available.Concerns about reduced component quality and toxicity have, however, limited or prevented the routine use of PRTs in many countries. Increasing evidence of the safety and efficacy of pathogen-reduced blood products suggests the need for careful assessment of the risks and benefits of PRTs. The aim of this review is to provide a brief update on advances in PRT treatment of blood components and to summarize the strengths and weaknesses of the available PRTs. Given that experience with PRTs for RBCs is still limited, this paper focuses on pathogen-reduced plasma and platelet products.Pathogen reduction technologiesCurrent PRTs for labile components either target cell membranes or nucleic acids (Table&). Most target nucleic acids and are thus ineffective for prion diseases. The mechanism of action determines the efficacy and applicability of PRTs. The S/D and MB plus light techniques are only applicable to plasma due to their detrimental effect on cell membranes. Ultra violet (UV) light-based technologies have been developed to also treat cellular blood products, such as platelet concentrates (PCs).Table&1.&Current pathogen reduction technologies for plasma and platelets in use or under clinical developmentMechanism of actionDisruption of lipid membranesMB intercalates into nucleic acid and mediates the formation of singlet oxygen upon illuminationAmotosalen (S-59) intercalates into nucleic acid and induces covalent cross-linking upon UVA exposureRiboflavin associates with nucleic acids and mediates an oxygen-independent electron transfer upon UV exposureUVC directly interacts with nucleic acids, causing the formation of nucleotide dimersBlood componentsPlasmaPlasmaPlasma and plateletsPlasma and plateletsPlasma and plateletsProducts Octaplas (Octapharma) OctaplasLG (Octapharma) Uniplas (Octapharma) Plasmasafe (Kedrion) Plasma viro-att&nu& solvent/detergent (French Blood Service, EFS) Bioplasma FDP (National Bioproducts Institute of Pinetown, South Africa) &Mini-pool& systems (V.I.P.S. SA, Switzerland)
THERAFEX MB (Macopharma) Springe method (Grifols) INTERCEPT Blood system for plasma and platelets (Cerus)MIRASOL PRT system for plasma and platelets (Terumo)THERAFLEX UV-Platelets (Macopharma)StatusIn clinical useIn clinical useIn clinical useIn clinical useUnder clinical developmentSolvent/detergent treatmentIn S/D treatment, batches of 60&380&l of plasma are pooled to yield a standardized biopharmaceutical product with uniform plasma protein concentrations. Treatment inactivates lipid-enveloped viruses, cells and most protozoa (Horowitz et&al, ; Hellstern & Solheim, ). Although SD treatment has no effect on non-enveloped viruses, it prevents transmission of hepatitis A virus (HAV) and parvovirus B19 (PB19) by lowering the virus load in the starting plasma through pooling and by neutralizing immune antibodies commonly present in the initial plasma pools (Solheim et&al, ). SD treatment is preceded by nanofiltration to remove cells, cell fragments and membrane-associated viruses, and is followed by sterile filtration and aseptic filling (Pelletier et&al, ). The production process for OctaplasLG, Octapharma's second-generation S/D plasma product, includes an affinity ligand chromatography step designed to bind prion agents at the level of prion infectivity, thus potentially reducing the risk of prion disease transmission (Heger et&al, ).Methylene blue plus visible light (MB/light)Methylene blue plus visible light was developed as a photodynamic viral inactivation technique for pathogen reduction in single units of fresh-frozen plasma (FFP) (Lambrecht et&al, ). MB is a phenothiazine dye that intercalates into viral nucleic acid. It produces singlet oxygen-mediated destruction of viral nucleic acid when illuminated with visible light, thus preventing pathogen replication. MB/light inactivates parasites and a broad range of different DNA and RNA viruses in plasma (Seghatchian et&al, ). However, it has detrimental effects on cell membranes, which limits its suitability to plasma. There are two MB-based systems available for treating plasma for transfusion: THERAFLEX MB-Plasma (Macopharma, Mouvaux, France) and the original &Springe method& (Grifols, Barcelona, Spain). The former removes lymphocytes and residual MB by filtration, whereas the latter uses a freeze-thaw step to destroy contaminating lymphocytes without an MB removal step.Amotosalen plus ultraviolet A light (amotosalen/UVA)The INTERCEPT Blood System (Cerus Corporation, Concord, CA, USA) is a photochemical pathogen reduction method used to treat plasma and PCs. It uses amotosalen as the photoactive compound. This alkylating agent penetrates cellular and nuclear membranes and binds to the double-stranded regions of DNA and RNA. When activated by low-energy UVA light (320&400&nm), amotosalen cross-links nucleic acids, thus blocking DNA/RNA replication. Amotosalen/UVA efficiently inactivates a broad spectrum of enveloped viruses, bacteria, protozoa and residual leucocytes, but has inconsistent effects on non-enveloped viruses (Wollowitz, ; Irsch & Lin, ). The system includes a compound adsorption device to remove residual amotosalen and photoproducts.Riboflavin plus ultraviolet light (riboflavin/UV)MIRASOL (TerumoBCT, Lakewood, CO, USA) is a riboflavin/UV-based PRT system used for pathogen reduction of plasma and platelets. Riboflavin (vitamin B2) acts as a photosensitizer. Upon exposure to UVA and UVB light (285&365&nm), it mediates selective damage to nucleic acids without binding to cells or proteins. Riboflavin associates with nucleic acids and mediates oxygen-independent electron transfer, causing irreversible damage to nucleic acids. Riboflavin/UV treatment has been proven effective against a range of pathogens, including bacteria, enveloped viruses, protozoa, leucocytes and some non-enveloped viruses, as well as for inactivation of white blood cells (WBCs) (Goodrich et&al, ; Marschner & Goodrich, ). Because naturally occurring vitamin B2 and its photodegradation products are considered as non-toxic, no removal step is intended prior to transfusion.Ultraviolet C (UVC)THERAFLEX UV-Platelets (Macopharma), a novel system for UVC-based pathogen reduction without photoactive substances, is currently undergoing clinical efficacy and safety testing (Mohr et&al, ; Seltsam & M&ller, ). Shortwave UVC light (254&nm) directly interacts with nucleic acids, resulting in the formation of pyrimidine dimers that block the elongation of nucleic acid transcripts (Douki et&al, ). Because UVC irradiation mainly affects the nucleic acids of pathogens and WBCs while maintaining plasma and platelet quality, this technology is suitable for the treatment of both plasma and platelet products. UVC treatment significantly reduces the infectivity of plasma and PCs contaminated by disease-causing viruses, bacteria and protozoa. Because this technology does not require the addition and removal of photoactive substances, UVC treatment is just as simple but less time-consuming than gamma irradiation.Potential of pathogen reduction technologyIn the past, procedures designed to increase blood safety often had the unintended effect of decreasing the availability of blood components. Precautionary exclusion of donor groups with a potential risk of disease transmission and the implementation of new donor screening tests, which always produce a certain rate of false-positive results, have resulted in increased donor deferral (Linden et&al, ; Custer et&al, ). However, the availability of blood itself is a safety issue. Pathogen reduction is an attractive and pragmatic approach to further reducing the risk of disease transmission while maintaining an adequate blood supply. PRTs are effective against a wide range of pathogens and therefore target most of the infectious risks in transfusion medicine (Table&).Table&2.&Potential of pathogen reduction of blood componentsElimination of the residual risk of window period infection from transfusion-relevant pathogens not detected by donor screening (e.g., HIV, HBV, HCV, syphilis)Reduction of the risk of recognizable infectious agents that currently cannot be prevented completely (e.g. bacteria, CMV, HAV)Proactive protection against risks from emerging infectious agents (i.e. known or as-yet-unknown pathogens) entering the blood supply of a certain communityGlobal protection against transfusion-associated graft-versus-host diseaseAlthough sensitive screening tests for the most common transfusion-transmitted pathogens, such as human immunodeficiency virus (HIV), hepatitis B virus (HBV) and hepatitis C virus (HCV), have decreased the risk of infection through blood transfusion to extremely low levels, infections due to undetectably low titres of pathogens still occur (Busch et&al, ). PRTs may raise the blood safety margin by inactivating pathogens that remain undetectable due to window periods, chronic viraemia or test failure.In addition, PRTs could significantly reduce the risk of unscreened viruses (e.g. HAV or PB19) and could eliminate pathogens that are not completely preventable (e.g. bacteria and cytomegalovirus). Bacterial contamination of platelets still poses the highest risk of transfusion-transmitted disease despite the availability of bacterial detection methods (Blajchman, ; Brecher et&al, ). Bacteria can be missed in early testing due to low bacterial concentrations and/or late log-phase growth. During storage, such undetected bacteria in PCs may grow to levels that can induce acute sepsis and even fatal reactions in recipients (Stormer et&al, ). A recent study estimated the risk per-patient of receiving a bacterially contaminated unit of apheresis platelets in the USA at 1:250, and that of developing a septic transfusion reaction at 1:1000 (Kleinman et&al, ). Blood products usually have very low levels of bacterial contamination at the time of donation (Brecher et&al, ). Therefore, it should be possible to significantly reduce the risk of receiving bacterially contaminated blood products by treating blood products soon after production with PRTs shown to achieve robust and effective broad-spectrum inactivation of bacteria.Pathogen reduction may be the only safeguard against unknown and emerging pathogens entering the blood supply. HIV, an impressive example from the early 1980s, shows that transfusion transmission can occur before a new virus is identified and screening tests can be implemented. Contemporary examples of transfusion-related emerging pathogens include West Nile virus, which spread across the United States in the first decade of this century, Chikungunya virus, which caused outbreaks on the French islands and in Italy, and Dengue virus, which caused outbreaks in Puerto Rico and several Southern European countries (Harrington et&al, ; Stramer et&al, ). PRT treatment was shown to be an effective proactive approach to preventing transfusion-transmitted infections during a recent Chikungunya virus epidemic (Rasongles et&al, ).Viable leucocytes in blood components can cause or contribute to adverse reactions, including alloimmunization and transfusion-associated graft-versus-host disease (TA-GVHD), in recipients (Ruhl et&al, ; Vamvakas & Blajchman, ). TA-GVHD occurs when viable donor T-lymphocytes proliferate and engraft after transfusion of cellular blood components and elicit an immune response in the lymphoid tissues of recipients unable to mount an immune response to the allogeneic donor cells due to immunosuppression or human leucocyte antigen (HLA) compatibility. The current standard for preventing TA-GVHD is gamma irradiation of cellular components for patients at risk. The pathogen reduction systems discussed in this review disable leucocytes by DNA modification or cell membrane disruption. In vitro and animal models have shown that these methods can effectively prevent the development of GVHD from transfused cells, making them potential alternatives to gamma irradiation (Grass et&al, ; Fast et&al, ; Seltsam et&al, ). Moreover, inactivation of leucocytes may reduce other leucocyte-induced non-infectious risks, such as febrile non-haemolytic transfusion reactions, transfusion-related immunomodulation and alloimmunization of recipients (Fast et&al, ). Routine PRT treatment of platelet units would obviate the need for a special inventory of products for patients at risk of GVHD (Mintz, ).Concerns and limitationsEfficacyEach PRT system has some weakness or gap in efficacy. Most PRT systems have a limited capacity to inactivate bacterial spores. This is of particular concern with PCs as spores can develop into vegetative forms and grow to clinically relevant numbers during PC storage (Stormer et&al, ).Pathogen reduction strategies for plasma fractionation are multi-step and batch-wise processes that must achieve a reduction of pathogen titres per ml by at least 6-log steps. This performance criterion does not apply to single blood units containing fragile components such as blood cells, as this would impair their therapeutic performance (Committee for proprietary medicinal products (CPMP), ).The efficacy of PRT treatment is therefore determined based on the effective rate of infection risk reduction compared to the risk in untreated blood products as opposed to the complete elimination of disease transmission. Thus, the effectiveness of risk reduction depends not only on the capabilities of the available PRTs, but also on the dynamics and epidemiology of infections and the type of screening tests or other safety measures in place in a given region. When effective screening tests are performed, pathogen reduction can eliminate or reduce the residual risk of infection due to window periods or donors with occult infection. For known or emerging viruses for which no effective screening tests are available, the degree of risk reduction varies with the inactivation efficacy of the PRT used. As recently discussed by Goodrich et&al (), methods achieving levels of inactivation below peak viraemia may still afford protection from disease transmission by completely eliminating the infectious particles present at low levels during the early and chronic phases of infection. Similarly, PRT treatment can be expected to be highly effective in reducing the risk of bacterial infections and sepsis given the low levels of 10&100 bacteria per product in contaminated donations at the beginning of storage (Brecher et&al, ). However, it is still not clear where the evidence for the actual bacterial transmissions to patients is derived from. There is concern about spore-forming bacteria causing infections and on their ability to resist PRT treatment. Although spores are almost insensitive to PRT treatment and high levels of bacterial contamination may occur, experimental data on contaminated platelet units suggest that pathogen reduction could prevent the vast majority of clinically relevant bacterial complications (Goodrich et&al, ; Mohr et&al, ; Schmidt et&al, ).Tolerability and qualityOne major concern regarding the implementation of PRTs is their impact on the integrity of blood components and the toxicity of the chemicals used in these systems. Ideally, the active substances in PRTs must be very selective and toxic to a broad range of pathogens but, at the same time, non-toxic to blood cells/proteins and transfusion recipients. Inactivation methods using chemical compounds (MB, amotosalen, riboflavin) have been subjected to extensive preclinical toxicology testing. As only small quantities of photochemical compounds are used in PRTs, the available data suggest that they provide sufficient safety margins and do not cause toxicity, carcinogenicity, mutagenicity, genotoxicity and reproductive toxicity (Ciaravino et&al, , ; Reddy et&al, ; Seghatchian et&al, ). Residual solvent and detergent levels in S/D plasma are far below the maximum limits for human exposure (Pelletier et&al, ). The UVC-based pathogen inactivation procedure works without photoactive substances, thus eliminating the risk of photoreagent-related adverse events (Seltsam & M&ller, ). It is currently the only single-blood-unit PRT system that does not use photoreagents.In vitro studies have shown that all of the PRTs discussed in this review influence the quality of plasma proteins and platelets (van Rhenen et&al, ; Ruane et&al, ; AuBuchon et&al, ; Apelseth et&al, ; Mohr et&al, ,). Pathogen reduction of plasma invariably affects relevant proteins and causes an up to 40% loss of coagulation factors and inhibitors (Prowse, ). As coagulation factor activity ranges from 50% to 200% in healthy individuals, reduction on this scale may be clinically irrelevant. However, the plasma quality and production method can also impact coagulation factor levels. Thromboembolic adverse events occurred in U.S. liver transplant patients transfused with S/D plasma that had not been observed in Europe. These events, attributed to a high loss of protein S, antiplasmin and antitrypsin activity during the pathogen reduction process, resulted in the termination of S/D plasma production in the USA in 2002&03 (Hellstern & Solheim, ). Site-specific variations in production may also play a role in the reduced fibrinogen levels in MB plasma that were recently reported in France [Agence francaise de s&curit& sanitaire des produits de sant& (AFSSAPS), 2011].There is reasonable concern that PRT treatment could have a negative effect on the haemostatic potential of transfused PCs. In vitro studies have consistently shown increased metabolic activity and moderate activation of platelets after treatment with all available PRTs (Apelseth et&al, , ; Mohr et&al, ; Ruane et&al, ; van Rhenen et&al, ). The 20&30% reduction of recovery and survival of PRT-treated platelets after reinfusion to healthy subjects suggests an impact on platelet viability, although the results are close to the limits (66% recovery and more than 58% survival) proposed by the U.S. Food and Drug Administration (Snyder et&al, ; AuBuchon et&al, ; Bashir et&al, ). The mechanism by which PRT treatment causes early removal of transfused platelets from the circulation remains unclear as the available in vitro findings do not adequately predict the changes in recovery and survival (Holme et&al, ). It also remains unclear how the impaired viability of PRT-treated platelets or their increased activation might affect the haemostatic balance in transfused patients.Another adverse effect that can cause intolerability is the elicitation of an immune response to plasma and platelet products. PRT treatment may induce modifications in plasma and platelet proteins, thereby creating new antigens (neoantigens) that are recognized as foreign by the recipient's immune system. Such an immune response could result in the formation of antibodies that bind to these neoantigens and cause transfusion reactions and refractoriness. PRTs using photoactive compounds may be more prone to induce immune reactions as the compounds or their photoproducts may be immunogenic or constitute neoantigens by interacting with plasma and platelet proteins. Some authors suspect that sensitization to MB could be responsible for allergic reactions after transfusion of MB-treated plasma in some cases (Nubret et&al, ; Mertes et&al, ). No such immune response has been reported for pathogen-reduced PCs, particularly MIRASOL and INTERCEPT-treated PCs containing (residual) photosensitizers. However, cases of immunization to chemical-dependent antigens following repeated transfusions have limited the clinical evaluation of pathogen reduction methods for RBCs (Benjamin et&al, ).Clinical evaluationPlasmaMost clinical efficacy studies designed to compare PRT-treated and untreated plasma involve only small numbers of patients and thus are not adequately powered to detect statistically significant minor differences. In addition, they assess increments in coagulation factor levels and changes in in vitro coagulation blood tests rather than clinically meaningful outcomes (Pehta, ; Williamson et&al, ; Beck et&al, ; Lerner et&al, ; de la Rubia et&al, ; Hambleton et&al, ; Alvarez-Larran et&al, ; de Alarcon et&al, ; Mintz et&al, ,; del Rio-Garma et&al, ; Bartelmaos et&al, ; Bindi et&al, ). Only two trials are sufficiently powered to provide valid data on the equivalence of PRT-treated and untreated plasma (Table&). The first compared amotosalen/UVA-treated plasma with untreated plasma in patients with acquired coagulopathy of liver disease (Mintz et&al, ). Using changes in prothrombin time (PT) and partial thromboplastin time (PTT) as primary outcome measures, the non-inferiority/equivalence of treatment was shown for the former but not for the latter. The second study compared the efficacy of MB/light-treated, S/D-treated plasma and untreated fresh-frozen plasma in French liver transplant patients using the volume of plasma transfused during transplantation as the clinical endpoint (Bartelmaos et&al, ). The transfused plasma volume was higher with MB/light-treated plasma, but this finding can at least partly be explained by differences in unit volumes and imbalances in bleeding risk factors. Neither of the two studies had sample sizes allowing for the detection of differences in clinical outcomes, such as bleeding and adverse events.Table&3.&Major clinical studies of pathogen-reduced plasmaDesignDouble-blind, prospective, multicentric RCTDouble-blind, prospective, multicentric RCTPlasma types testedQ-FFP vs. A-FFPQ-FFP vs. MB-FFP vs. S/D-FFPNumber of patients58 vs. 5897 vs. 100 vs. 96Clinical settingAcquired coagulopathy of liver diseaseLiver transplantationPrimary endpointsChanges in PT and PTT in response to first transfusionMean plasma volume and number of plasma units transfused per transplantationSecondary endpointsChanges in PT and PTT, factor VII levels, clinical haemostasis, blood component usage and safety up to 7&d following FFP transfusionBlood loss during surgical intervention, correction of laboratory coagulation variables, adverse eventsResults No difference after adjustment of PT and PTT for FFP dose and patient weighEquivalent changes in PT among groups No equivalence of changes in PTT No significant differences in secondary endpoints
After adjustment for bleeding risk factors: Transfused volume of MB-FFP was 14% higher than that of S/D-FFP and Q-FFP MB-FFP and S/D-FFP increased the number of plasma units transfused by 11% and 12%, respectively, compared to Q-FFP No significant differences in secondary endpoints An active haemovigilance programme was launched at a regional blood centre in France to establish a safety database on routine use of amotosalen/UVA-treated plasma with the aim of collecting data on low-frequency unexpected adverse events and therapeutic indications not studied in clinical trials (Cazenave et&al, ). Adverse events were reported for eight of the first 7483 transfusions of amotosalen-UVA-treated plasma, which was similar to the rate observed with conventional plasma.The haemovigilance data collected in France are of particular value for safety assessment of pathogen-reduced plasma, as three different types of PRTs (MB/light, S/D and amotosalen/UVA) are routinely used and surveyed in the same system. A haemovigilance survey of the period from 2005 to 2009 raised concerns over high rates of allergic reactions following transfusion of MB/light-treated plasma. Consequently, the French regulatory authority decided to discontinue the use of MB/light plasma in France (AFSSAPS, ; Mertes et&al, ). However, when limited to the years in which MB/light-treated plasma was actually used (), statistical analysis of the updated 2010 haemovigilance data did not confirm a significantly higher incidence of severe allergic reactions with MB/light plasma than with other types of plasma (Seltsam & Mueller, ). A recent analysis including the 2011 haemovigilance data revealed a trend with borderline significance towards a higher incidence of severe allergic reactions with MB/light plasma (Mertes et&al, ). The variability of these data demonstrates that imbalances in use and the effects of slight variations in incidence on statistical outcome have to be carefully considered when analysing and interpreting data on very rare events.PlateletsNumerous clinical studies have evaluated the haemostatic efficacy of pathogen-reduced platelets. Most focused on amotosalen/UVA-treated platelets and only one assessed riboflavin/UV-light-treated platelets. Table& characterizes the most relevant clinical studies for pathogen-reduced platelets, including patient numbers and study design. All of these randomized controlled trials (RCTs) were performed in haematology-oncology patients because this population typically receives prophylactic platelet transfusion therapy. However, the studies differ significantly in design and parameters, such as endpoints, transfusion triggers, blinding, test and reference preparations, and age of transfused platelets (Table&). The SPRINT (Therapeutic efficacy and safety of platelets treated with a photochemical process for pathogen inactivation) study, the largest and only RCT that used bleeding as a primary endpoint, did not reveal a significant increase in bleeding complications with pathogen-reduced platelets compared to untreated platelets. Thus, the investigators concluded clinical equivalence between treated and untreated platelets (McCullough et&al, ). However, in almost all trials, transfusion of pathogen-reduced platelets resulted in lower platelet count increments (CIs), lower corrected count increments (CCIs), shorter intervals between platelet transfusions, and an increased number of platelet transfusions per patient (van Rhenen et&al, ; Cazenave et&al, ; Kerkhoffs et&al, ; Lozano et&al, ). This is in accordance with the findings of studies reporting reduced survival and recovery of pathogen-reduced platelets in healthy volunteers and indicates early and/or accelerated removal of treated platelets from the circulation (Snyder et&al, ; AuBuchon et&al, ; Bashir et&al, ).Table&4.&Major clinical studies of pathogen-reduced plateletsTechnologyAmotosalen plus UVAAmotosalen plus UVAAmotosalen plus UVAAmotosalen plus UVAAmotosalen plus UVAAmotosalen plus UVARiboflavin plus UVTest preparationPooled BC platelets in PAS IIIApheresis platelets in PAS IIIApheresis platelets in PAS IIIPooled BC platelets in PAS IIIApheresis or pooled BC platelets in PAS IIIApheresis platelets in PAS IIIApheresis platelets in 100% plasmaReference preparationPooled BC platelets in PAS II or 100% plasmaApheresis platelets in 100% plasmaApheresis platelets in 100% plasmaPooled BC platelets in PAS III or 100% plasmaApheresis or pooled BC platelets in PAS III or PAS III-MApheresis platelets in PAS IIIApheresis platelets in 100% plasmaGamma irradiated (T/R)Only R if indicated99&8% vs. 99&8%Partly vs. 100%If indicated74% vs. 14%0% vs. 100%n.a.Number of patients (T/R)52 vs. 51318 vs. 32722 vs. 2185 vs. 94/99101 vs. 9844 vs. 7256 vs. 54Number of transfusions (T/R)390 vs. 2562678 vs. 2041103 vs. 115391 vs. 381/357101 vs. 98220 vs. 653368 vs. 310Pretransfusion count (T/R)17&&&109/l vs. 15&&&109/l15&&&109/l vs. 15&&&109/l19&&&109/l vs. 17&&&109/l16&&&109/l vs. 17&&&109/l vs. 18&&&109/l9&8&&&109/l vs. 9&6&&&109/l16&&&109/l vs. 15&&&109/l12&&&109/l vs. 17&&&109/lTransfusion trigger&20&&&109/l&10&&&109/l at most study sites&20&&&109/l&10&&&109/l10&20&&&109/ln.a.&10&&&109/l without risk factor &20&&&109/l with risk factorPrimary endpointCCI and CI at 1&hBleeding Grade 2CCI and CI at 1&hCCI at 1&hCCI at 1&hCCI at 1&hCCI at 1&hReduction of CI (T/R)1&h (%)&23&27&24&31&10&4n.a.24&h (%)&34&39&23&33&27&7n.a.Reduction of CCI (T/R)1&h (%)&12&31&23&25&13+4&3124&h (%)&39&34&30&32&30&2&32The Prophylactic PLAtelet DOse (PLADO) study confirmed that a lower CCI after transfusion of conventional platelets must not necessarily translate into reduced haemostatic efficacy, but rather in an increased number of transfusions per patient (Slichter et&al, ). The results of meta-analyses assessing the haemostatic capacity of pathogen-reduced PCs based on the data from all available clinical studies varied with the approaches and eligibility criteria used for calculation. One meta-analysis of combined amotosalen/UVA and riboflavin/UV trial data suggests that the therapeutic efficacy of pathogen-reduced platelets is inferior to that of untreated platelets (Vamvakas, ). However, in a later meta-analysis limited to RCTs with amotosalen/UVA-treated platelets, the same author concluded that pathogen reduction may not increase the risk of severe bleeding complications (Vamvakas, ). To achieve greater homogeneity of the variables studied, the most recent meta-analysis also focused on amotosalen/UVA trials but excluded data from the HOVON (Dutch-Belgian Hemato-Oncology Cooperative Group) study [the only non-double-blinded study, which did not use the World Health Organization (WHO) scale to grade bleeding and reported much lower rates of bleeding than other clinical platelet studies] (Cid et&al, ). This approach revealed no significant differences in bleeding risk between pathogen-reduced and reference PCs.Among the RCTs on pathogen-reduced platelets, the HOVON study attracted special attention as it was prematurely stopped due to a higher bleeding rate in the amotosalen/UVA arm (32%) in comparison to the reference study arms [plasma: 19%, PAS III (InterSol, Fenwal Inc., Lake Zurich, IL, USA) 15%] (Kerkhoffs et&al, ). This finding raised major concerns about the haemostatic efficacy of amotosalen/UVA-treated platelets in the blood transfusion community. It is important to take into account for interpretation that this study had CCI as primary endpoint and was not powered to detect differences in bleeding. Therefore, it is still an open question whether this difference in bleeding is based on a functional deficiency of the PRT-treated platelets or has simply occurred by chance. The HOVON study was criticized for its lack of blinding and for other design attributes (Corash & Sherman, ; Kerkhoffs, ). The relatively high number of off-protocol transfusions (27%) in the study illustrates the challenges in the conduct of complex platelet transfusion trials including more than two arms. It is also important to mention that the pathogen-reduced platelets were, in addition, gamma-irradiated to a high percentage. This treatment may have caused an unneeded damage of the pathogen-reduced platelets and could have resulted in a reduced viability of the platelets. The lessons learned from the discussion on the HOVON trial could be that clinical studies with a clear and simple design and a strict adhesion to the protocol may be superior to complex and multifaceted trials in producing meaningful results to support valid conclusions.A Swiss group conducted a comparison study of pathogen-reduced and control PCs designed to minimize confounding factors by standardizing the concentration of platelets and additive solution and by confining gamma irradiation to control platelets (Sigle et&al, ). The study demonstrated similar CCIs in both groups, but its informative value is limited because of the high number of &off-protocol transfusions& in the test arm. Transfusion across ABO barriers and the age of transfused platelets are other potential factors that influence platelet transfusion efficacy (Slichter et&al, ). Imbalances of these parameters between test and control groups may have contributed to the discrepant results of previous clinical studies with pathogen-reduced platelets.Observational studies and haemovigilance programmes surveying the universal application of pathogen-reduced platelets have presented evidence that these platelets are effective and safe when transfused to a large variety of patients. Surveillance reports for amotosalen/UVA-treated platelet units revealed that PRT-treated platelet components could be implemented into routine practice without substantially changing the use of platelet and red blood cell components (Osselaer et&al, ; Cazenave et&al, ). These findings suggest that either the use of pathogen-reduced platelets does not result in an increase in bleeding or, if bleeding is increased, that it does not translate into serious morbidity and mortality.Data of clinical and haemovigilance studies indicate that the overall safety profile of pathogen-reduced platelets is comparable to that of conventional PCs (Snyder et&al, ; Osselaer et&al, ,). However, there have been concerns over acute respiratory distress associated with UV light and photosensitizer-treated platelets, as reported in the SPRINT and Mirasol Clinical Evaluation (MIRACLE) studies (McCullough et&al, ; Cazenave et&al, ). Results of animal studies suggest that platelets exposed to UV light may mediate acute pulmonary toxicity (Gelderman et&al, ; Zhi et&al, ). Re-analysis of the data of the SPRINT study by an expert panel did not confirm significant differences in the rate of acute lung disorders between treatment and reference arms (Corash et&al, ). It will be difficult to provide definitive evidence for a link between respiratory clinical adverse events and transfusion of pathogen-reduced platelets for two reasons. First, the results from animal studies do not necessarily reflect the situation in severely ill thrombocytopenic patients. Second, a clinical study adequately powered to detect statistically significant differences in the occurrence of such infrequent adverse events between treatment groups requires very large numbers of patients. The discussion on MB plasma-induced severe allergic reactions demonstrates that effective surveillance of appropriate numbers of test and reference products is necessary for valid interpretation of possible rare side-effects (Seltsam & Mueller, ). Haemovigilance has to be further developed to allow for meaningful assessment of rarely occurring adverse effects resulting from the therapeutic use of pathogen-reduced blood products.Bleeding studiesAs platelet transfusions are used to treat and prevent bleeding, there is an obvious need for clinical trials investigating novel platelet transfusion procedures or platelet products (e.g., pathogen-reduced PCs) to assess their efficacy and effectiveness with regard to bleeding. The use of bleeding as a relevant outcome implies that differences in the haemostatic potentials of novel and reference platelet products translates into qualitative or quantitative changes in clinically relevant bleeding events in patients treated with novel products versus reference products. Accurate recording and adjunction into defined bleeding grades still remains a major challenge in platelet transfusion studies. While in previous studies bleedings were assessed on the WHO scale or the on the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTCAE), a new assessment tool is currently under evaluation which may allow for a more valid and standardized categorization of bleeding severities in transfused patients (Webert et&al, ).The PLADO trial is the largest trial of prophylactic platelet transfusion that uses bleeding as the primary endpoint. Evaluating 1272 patients (&400 per study arm), this RCT revealed that doses of 1&1&4&4&&&1011 platelets/m2 body-surface area had no effect on the incidence of WHO grade 2 or higher bleeding despite significantly different platelet increments (Slichter et&al, ). These results suggest that platelet transfusion studies comparing the clinical efficacy of two different platelet preparations are probably unable to show a significant difference in prevention of bleeding unless they differ substantially.A recently published study (Wandt et&al, ) further challenges the use of bleeding as an outcome parameter to evaluate the haemostatic efficacy of a novel platelet transfusion procedure or platelet product. This study is the first large randomized multicentre study (N&=&200 per arm) investigating prophylactic versus therapeutic platelet transfusion therapy for haematological cancers. It detected differences in bleeding events of WHO grade 2 and higher but was unable to show significant differences in the clinically more relevant bleeding of WHO grade 3 and 4. The authors estimated that a total of 2000 patients would be needed to determine differences in severe bleeding. Recruitment of such a large number of patients is hardly feasible. Therefore, it does not seem possible to perform a study adequately powered to determine differences in clinically relevant bleeding events, even in a progressive setting comparing the traditional prophylactic platelet transfusion strategy with a therapeutic strategy. The results of the PLADO and Wandt studies suggest that any comparison of two platelet preparations will most probably fail to show differences in mortality regardless of whether it is based on a prophylactic or a therapeutic transfusion regimen.All completed and on-going clinical studies of pathogen-reduced platelets follow the design of the PLADO trial in two ways: (i) prophylactic transfusions are given to thrombocytopenic patients as the standard of care to prevent bleeding complications, and (ii) the number of functional platelets in the pathogen-reduced PCs is lower than that in the reference products, as indicated by lower CI and CCI values. Thus, it is unlikely that clinical studies of pathogen reduction, even &bleeding studies&, will be able to determine the haemostatic potential of a novel platelet product by showing significant differences in clinically relevant bleeding events. Of course, this conclusion is only valid if PRT treatment damages only a subpopulation of platelets that are then removed from circulation and does not profoundly impair all platelets, including those that survive in the circulation. However, the results of a recent pilot study of platelet function in pathogen-reduced platelets isolated from the circulation of patients after transfusion suggest that these circulating cells may elicit haemostatic responses comparable to those in untreated platelets (Johansson et&al, ).Cost-effectivenessThe health economics of PRTs have also to be considered when deciding to implement these technologies, particularly in times of limited or even negative economic growth. A number of dependent and independent reports have been published on the cost-effectiveness of PRTs for different technologies and different countries (AuBuchon & Birkmeyer, ; Jackson et&al, ; Pereira, ; Bell et&al, ; Riedler et&al, ; Staginnus & Corash, ; Postma et&al, ; Moeremans et&al, ; Custer et&al, ). However, meaningful cost-effectiveness analyses in the field of pathogen reduction are limited by the fact that, due to the complexity of transfusion risks and practices, they are based on assumptions and simplifications and often focus on specific aspects only. In addition, the progress in PRTs and other technologies for prevention of transfusion-transmitted infections as well as the change in the infectious risk associated with transfusions (mainly due to emerging pathogens) make transfusion safety a moving target. The most comprehensive cost-effectiveness analysis to date considered whole blood as well as plasma and platelet pathogen reduction approaches as an addition to current interventions (Custer et&al, ). The estimated cost-effectiveness of whole blood PRT and of platelet-and-plasma PRTs was consistent with established thresholds for value in blood safety. While this analysis assessed infections, TA-GVHD, febrile reactions and transfusion-induced immune modulation, it did not include the risk of unknown pathogens. Later, the same authors included in their analysis emerging infections modelled on HIV or West Nile virus. They concluded that the occurrence of a chronic agent and even more the occurrence of an acute agent would significantly improve the cost-effectiveness of PRTs (J. Boyer, unpublished observations). Moreover, the discontinuation or modification of current infection screening methods and interventions may probably have an additional positive impact on the cost-effectiveness ratio of PRTs. Nevertheless, the decision to introduce pathogen reduction may not only be determined by the results of cost-benefit analyses. The thresholds that are commonly regarded as cost-effective in clinical practise do not seem to be relevant for blood safety, at least in developed countries in which a near-to zero-risk for infectious threats is usually intended.Current trends and outlookSeveral PRTs for plasma and platelets have been licensed and are in use in Europe and elsewhere. Some countries (e.g. Switzerland and Belgium) have adopted pathogen-reduction technologies nationwide, whereas other countries (e.g. France, Germany and Spain) only perform pathogen reduction at some regional blood transfusion services. Concerns over possible side effects, particularly component quality and pulmonary toxicity, have impeded the introduction of pathogen reduction in North America. The European outlook for introducing PRTs is multifaceted, and national and regional blood services are showing more and more interest in the evaluation and implementation of pathogen-reduced plasma and platelets. This trend is probably driven by positive risk-benefit assessments of this new blood safety technology by decision-makers in the blood banking business. This view is supported by the results of a Canadian risk model, which show that the availability of PRTs for both platelets and plasma would reduce the expected number of transfusion-transmitted infections by 40% (Kleinman et&al, ). In addition, a recently published study using a combination of clinical data and mathematical modelling showed a favourable risk-benefit profile for the implementation of pathogen-reduced platelets in countries like the United States (Kleinman et&al, ). Currently, the potential benefits appear to outweigh the concerns over excess bleeding and toxicity possibly associated with the use of PRTs, at least in parts of Europe.Stakeholders in the field of transfusion medicine have a strong desire to be able to inactivate pathogens in all blood components in order to increase the safety of the entire blood supply. This implies the need for PRTs for pathogen inactivation of RBCs as well as platelets and plasma. Thus, pathogen reduction cannot achieve its full potential for enhancing blood safety as long as pathogen reduction technologies for RBCs and whole blood are not available. The developments in Europe show that experts and an increasing number of health authorities already consider the implementation of pathogen reduction systems for platelets and plasma an important step towards increasing blood safety. As risk-benefit ratios may vary between countries, health policy-makers will have to shape their own specific regulatory and health-care environments at the national level.Disclosure of conflict of interestsThe authors cooperate with Macopharma in a development project for the THERAFLEX UV-Platelets system. Apart from project grants from the &Forschungsgemeinschaft der DRK-Blutspendedienste e.V.& and from Macopharma, the authors did not receive any financial support relevant to this manuscript.Author contributionsBoth authors designed and wrote the paper.
Advertisement}