The mantra of blood safety: time for a new tune?

Abstract

Over the past years, a paradigm of blood safety in established blood delivery systems has evolved which proposes the concept of the 'Safety Tripod'. In this paradigm, blood safety is approached through a combination of appropriate donor selection, screening tests for pathogens and pathogen reduction. All three 'legs' of the tripod are considered as equally essential contributors to the final safety of transfused blood, blood components and plasma derivatives. Regulatory measures and applicable standards have also emphasized all three components. With the increasing burden of 'zero risk' in public and political perception, it is difficult to question the immutability of this paradigm. However, its relevance and appropriateness is increasingly open to critical assessment. As the number of putative risks from blood-borne infections continues to mount, selecting donors with minimal geographical and behavioural risks is increasingly problematic and may lead to the introduction of known risks into the donor pool in efforts to minimize unknown ones. While nucleic acid technology has resulted in the ability to develop screening tests in a more rapid fashion than standard serology, the complexity and expense of screening platforms affects the logistics of blood delivery and raises the spectre of an unending series of tests to screen out each potential blood-borne pathogen. The potential for pathogen inactivation as the main route of assuring blood safety has been demonstrated for plasma derivatives, where the current excellent safety level has been achieved through inactivation techniques irrespective of the other legs of the tripod. The advent of such technology for the fresh components of mainstream blood banking therefore offers a promising approach to assuring safety in the absence of increasingly difficult and impractical selection and screening procedures. However, it is difficult to envisage these technologies in the current blood centre setting, and their incorporation can only be effected if the blood banking sector progresses further down the road to the manufacturing, rather than medical laboratory, environment. Furthermore, blood transfusion therapy needs to reflect other advances in therapeutic practice, and continuing evidence that much of blood component usage has no evidence base and is inappropriate must start feeding into blood safety policy if the benefits of technology are to be harvested. Measures for appropriate usage need to incorporate the significant advances in bloodless medicine which have been reported over the past decade. The rapid development of blood substitutes, particularly recombinant proteins, needs to be embraced and furthered by the transfusion community. All these factors can contribute to assuring the safety of blood products by channelling resources into measures for products which cannot be substituted in the near future. The transfusion community needs to embrace these developments and contribute a leadership role to their progression, rather than the suspicion with which it currently greets most of them. Thus, rather than a tripod, the future basis for blood safety should rather rest on the sophisticated application of a range of measures, an acceptance that some traditional measures may have to assume a secondary role, and the ever present need for communicating in a transparent and accountable way the relevant risks and preventive measures to a (justifiably) sceptical public and political framework. Assuring safety in the blood system in the developed world has settled down to a repetitive mantra of the absolute adherence of the so-called 'Safety Tripod' which demands observance to the following principles: Selection of appropriate and 'low risk' donors. Usage of screening tests for the relevant markers of infection. Elimination of any pathogens residual to measures (1) and (2) from the final products. The increasing complexity and costs imposed on the blood delivery system through the application of these principles irrespective of the potential effect on global safety and supply mandates their review. Surveys of viral markers in donor populations from well-developed blood systems demonstrate that donor selection procedures result in marker prevalence and incidence which are significantly lower than in the general population [1] (Table 1); furthermore, lower rates in repeat donors attest to the efficiency of the selection and screening process. In the USA, the donor qualification process has, like in other countries, developed into a complex procedure (described in [2]) underpinned by the requirements of the Code of Federal Regulations. Even in relation to behaviours associated with established risks, the sensitivity and specificity of the current donor selection processes are relatively poor. In studies conducted by the Retrovirus Epidemiology Donor Study (REDS) exercise in the USA, a proportion of donors were found to have answered behavioural questions inaccurately, potentially increasing the infectious risks of the donated units [3]. In a classic example of archaism in donor questioning, the Food and Drug Administration's (FDA's) Blood Products Advisory Committee (BPAC) shrank from changing the question excluding any male-to-male sexual activity since 1977, despite compelling agency data suggesting the risk was minimal and the change was desirable in terms of improving donor numbers and achieving harmonization with other regulations [4]. A more recent study has demonstrated the minimal risk associated with this change [5]. The introduction of nucleic acid testing (NAT) which has minimized the serological window for HIV appears to render such a change logical, but strict adherence to the concept of the 'Safety Tripod' would urge against the relaxation of one measure just because another one minimizes its value. Questioning donors for risk factors associated with geographical residence such as vCJD-related measures, is of very low specificity for known pathogens. Exclusion criteria designed to minimize the risk of transfusion-transmitted malaria [6] are known to result in the unnecessary deferral of many safe donors [7]. In Australia, these criteria lose the blood supply about 100 000 units of red cells each year from a total collection of 900 000 whole blood donations. Again, the development of a specific test will, it is assumed, allow access to a considerable proportion of these units [8]. However, the reaction of regulators to this assumption remains to be tested. With the limitations outlined above in mind, the FDA initiated a process of review of the current donor selection questionnaire in the USA [9, 10]. It should be noted that the FDA, while accepting the need for simplifying and making more comprehensible the current questionnaire, continues to issue complex guidances for donor selection. A rapidly emerging issue in donor selection involves the validation of selection procedures, particularly the donor questionnaire [11], a process which would bear emulation in relation to regulatory guidances. An assessment of whether the increasingly complex questions introduced over, say, the past decade have enhanced transfusion safety relative to the selection process used in the previous years would be an interesting exercise. Molecular biology now dominates the field of predonation testing for transfusion-transmitted viruses through: The use of techniques which accelerate greatly the viral characterization and epidemiology of putative transfusion-transmitted agents and provide a more rapid screening route than classical serological approaches [12]. The estimation of viral loads during the relevant parts of the viral replication cycle, which can be employed to address questions such as the relationship between viral load and infectivity, the potential for pathogen reduction technology (see below) and other related issues [13]. The closure of the infectious window prior to the development of serological markers allowing the residual risk from the viruses of traditional concern to approach very low levels [14]. NAT in the blood screening environment has been amply reviewed [15] and its cost effectiveness relative to other healthcare interventions has been the cause of substantial comment [16, 17]. The incursion of NAT into the blood banking environment will doubtless progress to mandatory requirements for single donor rather than pooled NAT once the technology is sufficiently developed. Initial regulatory thinking limiting NAT to the role of an in-process control for plasma pools during fractionation appears particularly pertinent for resistant viruses currently unscreened at the donation level [18]. Recent opinions from the FDA viewing the screening of nonenveloped viruses in plasma pools as diagnostic tests subject to the need for investigational new drug (IND) approval routes in the USA will force the industry to identify and follow up donors whose donations result in positive plasma pools [19]. Coupled with FDA requirements to quarantine blood components from such positive donations 'when possible'[20], the industry can expect to have viruses such as parvovirus B19 (B19) and hepatitis A (HAV) as part of the mainstream screening regimen. The introduction of NAT has unleashed a juggernaut which has yet to result in a significant enhancement in blood safety in developed blood delivery systems. Allain [19] has shown how NAT can have a significant effect on the safety of blood collected in situations of high prevalence for the known transfusion-transmitted viruses, but the economic circumstances of the countries involved will not permit the use of NAT in these environments. NAT appears to be the methodology of choice in screening viruses where the infectious phase precedes the development of an antibody response, such as the currently worrisome West Nile Virus (WNV) [21]. However, it is difficult to visualize the constant addition of screening tests, of whatever methodology, to blood supply systems as more agents are recognized. This issue will be discussed further below. In the meantime, the use of technologies other than NAT, such as the ELISA for HCV antigen [22], HIV antigen [23, 24] and haemagglutination [25] for B19 should be considered by countries and blood systems not committed to this methodology. Studies coordinated by the FDA show that using HbsAg assays which are more sensitive than those currently licensed in the USA allow window period closures for HBV which are similar to those acquired through NAT on pools [26]. The haemophilia community's experience is the best indicator of the status of blood safety in relation to manufactured plasma derivatives. Viral inactivation has been the predominant contributor to safety in these patients. Patients receiving Belgian cryoprecipitate compared to French and US patients receiving large pool concentrates indicate that donor selection, usage of domestic-sourced product and plasma pool size also play a role [27]. It is worth noting that FDA-recommended donor selection procedures, which led to the risk of AIDS from blood transfusion decreasing markedly before the introduction of testing, did not prevent the continued brisk penetration of the HIV epidemic in the haemophilia community in the USA over 1983–84 [28]. Studies also show that the plasma pool for such products was still infectious after the introduction of HCV antibody testing [29], while infectivity was not detectable in intramuscular immunoglobulin products with a historical record of safety. The gold standard for viral safety in plasma derivatives is albumin solution heated for 10 h at 60 °C in the final container, a process now known to inactivate the transfusion-transmitted viruses, including B19 [30]. The lack of transmission by this product, but not by other, non-virally-inactivated plasma derivatives, during periods when the manufacturing plasma pool is known to have been heavily contaminated with HIV and HCV confirms the dominant contribution of viral inactivation to the safety of plasma derivatives. The standard inactivation procedures used for plasma derivatives have been inapplicable to labile blood components because of the fragility of the products concerned. Solvent–detergent-treated plasma (SD plasma) established a presence in the North American and European environments but has been associated with some problems. The pooling required for pharmaceutical scale manufacture has increased the risk from agents unscreened at the donation levels such as B19; the introduction of NAT for plasma pools prior to manufacture has decreased this risk [31]. Although the use of methylene blue (MB) to inactivate viruses in red cell and platelet suspensions has been described [32], the pathogen reduction techniques used to sterilize plasma have not been used for other components. The experience of treating plasma has allowed some insight into the possible pitfalls and limitations of pathogen reduction: The transmission of B19 infection by SD plasma demonstrates the risk of introducing a new risk as a result of a putative safety measure. It is noteworthy that this increased risk of B19 infection was predicted before the actual transmissions were noted [33]. All pathogen reduction methods for fresh plasma have an effect on product quality. The decision of the United Kingdom authorities to import plasma from the United States and then treat it with MB to decrease infectious risk [34] may be perceived as a case of trading one, unknown, risk (for vCJD) for the real risk of exposing patients to a product, which has been imperfectly virally inactivated, from a higher risk donor population. The current interest in pathogen reduction technologies (PRT) for cellular components must therefore be moderated by the experience of other areas of blood safety. While risks from pooling are not likely to be an issue in cellular components [35], there is a possibility that patient exposure to components treated with the various PRTs (reviewed in [36]) will have to be increased due to the effect of the PRT on product quality. This can only be confirmed by clinical studies. It should be noted that the loss of red cells resulting from leucocyte reduction did not lead to an increased usage of red cells [37]. While further clinical studies with red cells are awaited, the inactivation processes for platelets have shown an effect on in vivo characteristics. The S-59/UVA technology results in platelets which are haemostatically equivalent to untreated platelets but which are associated with higher transfusion requirements [38], while recovery and survival studies with platelets treated with riboflavin and light infused into macaques showed decreases in these parameters in treated platelets [39]. In the past years, the risk of emerging pathogens has affected markedly the blood safety policy of the developed world, and has strained severely the ability to maintain the blood supply while practising the principles of the 'Safety Tripod'. As the risks, unknown or proven, of pathogens previously unassociated with blood transmission have arisen, they have been addressed through implementing whichever of the three measures described have been possible. It is now established that whenever the threat of the introduction of an exogenous pathogen into a national blood system occurs, the first measure which is considered is the deferral or restriction of donors travelling from, or resident in, the geographical location of origin of the pathogen. The principle was firmly established for countering the risk of vCJD from donors resident or travelling in areas of (relatively) high vCJD incidence. In this instance, irrespective of the issue of the proportionality of the response, it is reasonable to rely on this measure as the mainstay of protecting the safety of whole blood and component transfusion, given the lack of a screening test. Similar measures have been actuated by agencies in response to the threat of WNV [40] and Severe Acute Respiratory Distress Syndrome (SARS) [41]. Studies in Australia have shown that vCJD-directed measures resulted in selectively excluding repeat donors who came from the same socio-economic groups most likely to have lived and travelled in the UK [42]. Whyte has shown [43] that the loss of relatively modest numbers of repeat donors can have a dramatic effect on product supply. It is therefore likely that if the current policy paradigm of deferring on the basic of geographically-associated risk persists, the effect on the blood supply will be substantial. Kreill [44] makes the same argument for mosquito-vectored viruses other than WNV [45, 46]. As this review is being written, international public health concerns around SARS are precipitating donor deferrals based upon travel to SARS-infected areas. The speed with which these deferrals were undertaken by regulatory and public health authorities confirms the dominance of the precautionary approach, in this instance applied to an agent which, on the basis of its biological features, poses little threat to the blood supply. While performing such microbiological testing during a period of undoubted risk is a valid blood safety measure, it is difficult to predict what measures will be taken by agencies once the epidemic rescinds. The tendency for measures, particularly those promulgated by the FDA, to spread from the North American environment to other geographical areas with a different risk profile is cause for concern. In Australia, the national regulator was able to limit WNV-related measures to donors associated with a geographically-associated risk, and to allow the usage of plasma for fractionation on the basis of inactivation of the virus during manufacturing [47]. WNV screening would be used as a substitute for deferral when a test became available, and would selectively target the donors identified as being at risk through the selection questionnaire [48]. In the blood systems of the developed world such an indiscriminate approach to safety measures is already causing a strain in blood supply management and financing, but these effects are amplified in developing countries where the logistical and financial burdens are higher, together with the actual blood-related risks themselves. Of relevance is the rapidly expanding use of NAT in these testing strategies. It appears that serology will be of limited use in the assessment of asymptomatic viraemia in both established and emerging pathogens, and a 'quantum leap' in blood screening methodology has established NAT as the first technological recourse. The intrinsic costs of NAT, together with the possibility of expensive patent restrictions such as those recently imposed for HCV would make the access of such technologies by developing countries difficult [26]. The ultimate costs of current blood safety measures make the prospect of PRTs attractive. However, several prerequisites will be required before pathogen reduction acquires the same role for blood components as viral inactivation currently plays in the safety of plasma derivatives: Robust and safe technologies. Currently described technologies need to be wide in their scope in terms of, e.g. inactivation of the resistant nonenveloped viruses [49]. Furthermore, possible toxic and other potential adverse effects of the technologies require substantial processing manipulations to remove the agents involved in the inactivation process. Integration into the mainstream blood banking environment. The production of components using the proposed pathogen reduction systems will require an acceleration of blood banking into the kind of manufacturing environment more akin to pharmaceutical manufacturing under the aegis of Good Manufacturing Practices (GMPs). Blood components will be highly manipulated products whose manufacture will require precise adherence to complex operating procedures to ensure standardized and safe products. Revisio

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 31 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.