Prions and precautions: be careful for what you ask

Blood safety has become fertile soil for the precautionary principle, or at least for a particular embodiment of it. Many in our field take as a matter of dogma that a measure offering promise of improvement in blood safety should be implemented. This philosophy is at variance both with the original formulation and with the context of the precautionary principle and ignores authoritative commentary around the issue. In fact, the principle originally established that actions that might harm the environment should not be undertaken unless the proponents of those actions could establish that they were harmless. Even when the principle is reformulated to favor action to prevent harm, the concept that the action itself should be without appreciable harm is a critical component. Formal implementation of the principle also incorporates concepts that the action taken should be proportional to the chosen level of protection, that it should be consistent with similar measures already taken in comparable circumstances, and that it should be economically rational (as based on, for example, cost/benefit analyses).1 In other words, it is not the case that the precautionary principle requires a safety action to be implemented automatically if it is available. Rather, it requires that all of the consequences of a proposed action should be carefully considered to make a fully reasoned decision. In North America, implementation of measures to defer blood donors potentially exposed to bovine spongiform encephalopathy through residence in, or prolonged travel to the United Kingdom and parts of Europe has been hailed as a triumph of precaution, once it became clear that transmission of variant Creutzfeldt-Jakob disease (vCJD) via transfusion was a fact, rather than a theoretical possibility.2 But it is, in fact, impossible to determine whether this measure has actually prevented any such transmissions in the United States or Canada. We do know that it has eliminated many willing blood donors from the pool and that, at a minimum, has increased the cost of blood by requiring the recruitment of new donors to replace those that were eliminated. In truth, I do not think that we can tell whether or not this measure has validated or violated the precautionary principle. This issue will probably never be determined one way or the other; this uncertainty is not a problem unless this approach is used as a standard against which to measure other possible interventions. In this edition of TRANSFUSION, differing aspects of two such possible interventions are presented; prion removal technology for red blood cells (RBCs), and potential tests for pathologic prions in the blood, before the appearance of disease. These are, indeed, important examples of science directed toward a worthy outcome, but it will be important to be sure that they do not stimulate an errant and automatic implementation of thoughtless precautionism. Until recently, infectivity assays for prions have largely been based on animal inoculation studies, which are extremely time-consuming and costly, particularly for low levels of infectivity. They are further complicated by species barriers that mean that it is essentially impossible to perform direct determinations of the infectivity of human transmissible spongiform encephalopathies (TSEs). Sowemino-Coker and colleagues3 have used a novel, cell culture–based infectivity assay to show that meaningful levels of TSE infectivity can be removed by some, but not all, prototype prion removal filters. The test system described uses a mouse-adapted scrapie TSE and a brain homogenate spiking system, so it is far removed from human endogenous vCJD infection, the true target of the filtration process. Nevertheless, the technique is rapid, taking only 2 weeks or so and is inexpensive and offers the potential for an automatable readout. Thus, it represents an ideal screening approach, allowing for infectivity assays during the early stages of development of prion filters. The study also showed that the data obtained through the cell culture method were congruent with those obtained using a hamster model and endogenous infectivity. It is to be hoped that a successful cell culture infectivity assay approach will eventually be extended to human TSE systems, thus increasing confidence that prion filters will actually do exactly what is expected of them. One of the perceived benefits of prion removal is that it offers safety improvement but without adverse consequences. The study by Wiltshire and colleagues4 in the British National Health Service is somewhat encouraging in this regard, but it is not without some cautionary messages. The prion removal filter that was tested did not appear to have any serious adverse effects on the RBCs themselves, although there were minor increases in hemolysis, potassium leakage, and 2,3-diphosphoglycerate decline. Inasmuch as could be determined from the report, there was no impact on the expression of RBC antigens after passing through the filter, although the methods used would probably not detect the generation of neoantigens. The response of treated cells to gamma irradiation and freezing was also essentially unaffected. What was striking was the loss of RBCs, as expressed by a 7- to 8-g decline in hemoglobin (Hb) in the treated products, attributable to volume loss on the filter. Although this loss did not impact the overall acceptability of units prepared by the “top-and-top” method, it did have such an impact on “bottom-and-top” units that they no longer met current QC requirements (>75% of units must contain >40 g of Hb per unit). It was estimated that implementation of this method would result in a need to collect 10,000 to 15,000 more units annually in England, where 2.1 million collections are undertaken annually. Another disturbing statistic was that 5.9% of units failed to filter properly, although it appears that appropriate adjustments to the filter sets were able to overcome this problem. Another technology that has been anticipated to be of value is the ability to detect infectivity in the blood of asymptomatic or presymptomatic individuals, in a manner analogous to testing routinely used for human immunodeficiency virus, hepatitis C virus, hepatitis B virus, and other well-known transfusion-transmissible agents. There are enormous challenges around this objective, as the circulating levels of pathologic prions are thought to be extremely low, yet accompanied by a vast excess of the normal protein, which is essentially antigenically indistinguishable. To date, attempts to develop an assay for blood have not been very successful, in large part due to these restrictions. Recently, the development of protein misfolding cyclic amplification (PMCA) by Saborio and colleagues5 has offered some hope of a new approach. In essence, this technique is an in vitro adaptation of the in vivo multiplication of pathologic prions. Successive cycles of sonication, followed by incubation, result in amplification of the pathologic conformer of PrP. By performing this technique in the presence of a suitable source of normal PrP, vanishingly low amounts of prion can be amplified to levels that are readily detectable by standard methods, such as the Western blot. Tattum and colleagues6 have used serial PMCA in a novel way to amplify prion proteins from whole blood samples of infected, asymptomatic animals, a few days before the anticipated development of disease. This approach is promising, but one that is far from implementation. Apart from the fact that there are, as yet, no data about the efficacy of this test at earlier phases of infection, it is technically cumbersome, requiring multiple lengthy cycles of amplification, followed by a Western blot detection phase. Although one of the problems with PMCA has been the apparent spontaneous generation of “amplicons,”7 no such artifacts were seen in appropriate controls in this study. This is very important, as a key to the use of prion testing in blood donors will be assurance of a high level of specificity. A further report by the same team discusses an immunoconcentration method and carefully selected monoclonal antibodies to achieve what appear to be even higher levels of analytic sensitivity. In this context, Guntz and colleagues8 at the French EFS have reported on the first phase of an extensive evaluation of the specificity of another candidate prion assay. Their study suggests a gratifyingly high specificity of 99.90%, based on duplicate repeat testing of initially reactive samples. This figure, however, seems to be quite sensitive to sample handling and test kit age. Of more concern is the fact that, even though this specificity is admirable, the test would be expected to have a poor positive predictive value, given that the expected frequency of true positives in the French donor population has been estimated at 1 in 120,000. In other words, for every true positive, there would be 120 false positives. This calculation underlines the need for a confirmatory test before any attempt is made to implement a prion test for blood donors or their donations. It is already quite apparent that the implementation of any such test might have a chilling impact on the willingness to give blood, at least on the basis of studies undertaken in England. These measures will also need to be considered in the context of the future risk of transfusion transmission of vCJD (and perhaps other TSEs), which remains unknown. Current data certainly suggest that the primary epidemic of clinical vCJD is on the wane, at least in the United Kingdom, where careful control of the food chain has minimized continuing exposure. There does continue to be uncertainty about the extent and implications of asymptomatic infection; the finding of vCJD prions among individuals who do not have the MM genotype at the 129 codon of PrP has also led to speculation that there may be a further wave of disease, perhaps with a very extended incubation period. Such issues are the concern of the UK Transfusion Medicine Epidemiology Review Group (TMER), who have provided a report on a very careful and thoughtful evaluation of two cases of vCJD, potentially linked to a single donor.9 The caution used to approach this case is exemplary and the authors conclude that the circumstances do not support the interpretation that one, or perhaps even two, additional transfusion transmissions have occurred. Of course, the alternative interpretation that these were indeed transfusion transmissions would have significant implications, particularly with respect to the potential period of infectivity in an asymptomatic donor. It does not appear that there will be a definitive resolution to this case and, at most, it should be seen as a strong signal to continue the diligent work of the TMER. No matter how encouraging the early advances illustrated by these articles, when the time comes to think seriously about implementation of interventions, it is to be hoped that care will be taken to consider the original formulation of the precautionary principle. The pressure to act in avert the preventable transmission of a dread disease is understandable and would appear to be nonnegotiable, but the proponents of the intervention will also bear the onus of establishing that it will not cause undue harm by restricting the availability of safe blood or adversely affecting healthy blood donors. None.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.