Prevention of transfusion‐acquired CMV infection: is there a role for NAT?

CMV infection can be transmitted by blood transfusion. This has resulted in the implementation of strategies such as using CMV seronegative and/or WBC-reduced cellular blood components to prevent its transmission to transfusion recipients at high risk of experiencing CMV-associated morbidity and mortality. In order to evaluate the success of these programs and to determine whether additional interventions, such as NAT of blood donors for CMV DNA, would further reduce transfusion-acquired (TA)-CMV infection, it is important to understand the pathogenesis of TA-CMV infection. Two papers in this issue of TRANSFUSION by Roback et al.1 and Drew et al.2 provide important new information that can be used to address this issue. The prevention of TA-CMV infections is associated with several challenges that are unique when compared to many other transfusion -transmitted infections. First, evidence of previous exposure as measured by the presence of CMV antibody is highly prevalent in blood donors; 42 percent of US blood donors are CMV seropositive.1 Although data on CMV seroconversion rates in blood donors are limited, rates also appear to be in the range of 1 percent per year.3 Most importantly, CMV establishes a latent infection in cells of hemopoietic lineage in all subjects who have experienced infection and are seropositive and infectious virus can be reactivated from these cells.4 These reactivation events could occur in the blood donor before infusion, during storage after collection, or in the recipient after transfusion. Although CMV latency and reactivation is incompletely understood, there are significant new data with respect to CMV persistence in man that are relevant to TA-CMV infection.4 CMV latency appears to be limited to CD14+ monocytes in peripheral blood and CD34+ and CD33+ cells in marrow of healthy seropositive donors. Estimates suggest that the frequency of latently infected cells in the peripheral blood of these donors is low, in the range of 0.01 to 0.12 percent or 0.004 to 0.01 percent of mononuclear cells after granulocyte-colony-stimulated mobilization. A model that has been suggested to explain CMV persistence is as follows. The CMV genome is carried in myeloid lineage progenitor cells in the marrow of seropositive healthy subjects and is maintained in the cells as they reside quiescently or self renew. CMV DNA is also maintained as the cells differentiate down the myeloid lineage into PBMNCs. If latently infected progenitor cells or monocytes differentiate along very specific pathways and are exposed to a particular proinflammatory cytokine milieu, reactivation of lytic virus occurs and CMV replication ensues in fully differentiated tissue dendritic cells or macrophages. An allogeneic reaction, as one might expect after blood transfusion using non-WBC-reduced blood components, appears to be a powerful stimulant for this reactivation process with the elaboration of γ-IFN playing a critical role. Additional mechanisms involved in virus reactivation include systemic inflammation associated with release of TNF-α, use of cAMP-elevating drugs, and stress events associated with high plasma catecholamine levels.5 Although CMV reactivation is known to occur frequently in immunosuppressed seropositive patients, recent data suggest that CMV reactivation may also be common in nonimmunosuppressed populations, including healthy blood donors.5 CMV-seropositive donors have a high frequency of CMV-specific effector T cells in peripheral blood (0.1-3%), an observation that suggests frequent interaction between the virus and the immune system in the setting of low-level endogenous reactivation.6 This immune stimulation also likely impacts the humoral arm of the immune response, providing reassurance that antibody-negative, latently CMV- infected, or “seroreverting” blood donors are unlikely to be a real phenomenon. How often does CMV reactivation occur in blood donors and does it have any effect on the risk of TA-CMV infection? Some preliminary data addressing this question are provided by the two studies published in this issue where WBC DNA fractions or plasma from CMV seropositive blood donors were studied for the presence of CMV DNA. Roback et al.1 could reproducibly detect CMV DNA in the WBCs, but not in the plasma, of only two of 416 seropositive blood donors. From the data available, it is not possible to determine whether these blood donors were experiencing primary infection or reactivation events. Drew et al.2 were unable to detect CMV DNA in plasma of 68 CMV seropositive, nonseroconverting donors followed over time. Data from Roback et al.1 would suggest that plasma is not the optimal sampling site for detection of these events. What can we conclude from these results? Although it would be tempting to conclude that CMV reactivation events are infrequent in healthy blood donors, this conclusion is inconsistent with the immunologic evidence. We must remember that the commercial NAT assay used in both of these studies was designed to detect primary and reactivation CMV infection in immunosuppressed populations, but is not sensitive enough to detect latency and may not be sufficiently sensitive to detect low-grade reactivation events. Dumont et al.7 recently reported the fascinating observation that a significant proportion of CMV-seropositive blood donors experienced transient CMV reactivation, sometimes associated with production of infectious virions, on a seasonal basis during periods associated with high environmental pollen counts. Clearly, more work needs to be done to study CMV reactivation events in seropositive blood donors using appropriately sensitive assays and sample types in order to determine the frequency of these events, their seasonality, and their relationship to CMV transmission by blood components. A central unresolved issue with respect to TA-CMV infection is whether CMV infection in transfusion recipients is largely the result of receipt of blood components from CMV-seroconverting blood donors and donors experiencing CMV reactivation, in which case an infectious subset of blood donors should be identifiable, or whether CMV infection is the result of transfusion of latent virus in blood components and subsequent reactivation of virus in the transfusion recipient. If the latter is the case, screening of blood donors other than for previous exposure to the virus (i.e., antibody screening) would be irrelevant, because recipient factors rather than donor factors would define risk. Attempts to identify an “infectious” subset of blood donors using donor viruria and screening for CMV-specific IgM have yielding equivocal results with respect to transmission risk.3 Whether assays for the detection of CMV DNA represent a more reliable tool for the detection of these “infectious” donors remains to be proven. In contrast, historically very high rates of CMV transmission have been documented in the setting of exchange transfusion in neonates and granulocyte transfusion.3 This suggests that most seropositive blood donors can transmit CMV infection when a sufficient number of cells are transferred, there is a appropriate period of microchimerism to permit CMV reactivation in donor cells before they are destroyed, and an appropriate environment to promote CMV reactivation exists in the host. It is most likely that both donors experiencing primary and reactivation CMV infection and reactivation of infectious CMV after transfusion of latently infected cells into recipients contribute to the overall burden of TA-CMV infection. In order to optimize strategies for prevention, it would be important to determine the relative importance of these two mechanisms of transmission. The use of either CMV seronegative or WBC-reduced blood components has significantly reduced the incidence of TA-CMV disease in high-risk populations. It has been suggested that these strategies are “imperfect” with respect to the prevention of TA-CMV infection. This conclusion is based largely on a single important and large multicenter study of TA-CMV infection in seronegative recipients of seronegative marrow transplants who were randomly assigned to receive either seronegative cellular blood components or blood components WBC-reduced by filtration.8 Although the authors concluded that CMV seronegative blood components were equivalent to WBC-reduced blood components, “breakthrough” CMV infections occurred in 1.3 and 2.4 percent of patients receiving seronegative and WBC-reduced blood components, respectively. In interpreting the “breakthroughs” in this study it is important to remember that in both arms of the study, the technology for the provision of CMV-‘safe’ blood components was not optimized. In the CMV-seronegative arm, an assay known to have poorer sensitivity than other commercial assays and requiring subjective interpretation was used. Roback et al.1 found that 7 percent of blood donor samples screened for CMV antibody using commercial screening assays commonly used in blood centers yielded discordant screening results. Although CMV DNA was not detected in any of these blood donors, it illustrates the need for optimizing the sensitivity and specificity of existing CMV antibody screening assays prior to exploring the need for costly prevention strategies, such as NAT of blood donors. Even approaches using two screening assays for CMV antibody in tandem may be more cost-effective that NAT. In the study by Bowden et al.,8 bedside filtration was also used rather than WBC reduction before storage. The latter approach is known to be associated with better quality control and more reproducible WBC removal. In the marrow transplant setting, hemopoiesis in the patient's stem cells is highly stimulated in an environment that often includes both GVHD and sepsis. Repetitive blood transfusions in the setting of immunosuppression that leads to at least transient microchimerism and a cytokine microenvironment that promotes CMV reactivation from latency may explain the high rate of TA-CMV infection in this population. The same blood components transfused into other populations may not be associated with the same risk. Before concluding that current strategies for TA-CMV infection are inadequate, it would be important to study the incidence of TA-CMV infection in a number of at-risk populations when technology associated with the provision of seronegative or WBC-reduced blood components has been optimized. Additional controversy was engendered, particularly with respect to the efficacy of antibody screening as a preventive strategy, by reports of high rates of CMV DNA detection using nucleic acid detection assays not only in seropositive but also in seronegative blood donors.3 Using a standardized specimen panel, Roback et al.9 observed significant interlaboratory differences in the detection of CMV DNA using commercial and “in-house” assays. Using two of the most sensitive, specific, and reproducible of these assays and a representative panel of 1000 US blood donors, Roback et al.1 observed that CMV DNA could only be reproducibly detected in a very small subset of CMV seropositive donors. CMV could not be reproducibly detected in seronegative blood donors. These investigators conclude that earlier published reports of CMV DNA detection in seronegative blood donors are most likely due to technical artifacts leading to false positive results. This conclusion not only represents a vindication of the sensitivity of antibody screening, it is also consistent with clinical observations in the setting of blood transfusion and organ transplantation. It also significantly decreases the evidence supporting the need for an expensive technology (NAT) to replace relatively inexpensive serologic screening to identify blood donors capable of transmitting CMV infection. Theoretically at least, it is in the setting of the “window period” or acutely seroconverting donor that both antibody screening and WBC-reduction strategies are most likely to fail, and NAT has the greatest potential for utility. How frequently are these donors seen and how “infectious” are they? In this issue of TRANSFUSION, Drew et al.2 report finding CMV DNA in the plasma samples of only two of 192 blood donors who had evidence of recent seroconversion, only one of whom had nondetectable CMV antibody at the time of CMV DNA detection. Although it could be argued that this study used widely spaced sampling and a suboptimal sampling site, these findings are consistent with those observed by others.3 That is, CMV viremia is low grade and short-lived in immunocompetent asymptomatic subjects experiencing primary infection. Caution must be exercised in using CMV DNA detection as a surrogate marker for the presence of infectious virions. Studies in acutely infected patients indicate that CMV DNA can be detected in WBCs for a significantly longer period of time than infectious virus detected using standard tissue culture methods. Although it is possible that molecular methods are more sensitive than tissue culture for detecting infectious virus, it is clear that not all, or even most, CMV DNA detected is associated with virions capable of transmitting infection. Boom et al.10 recently observed that the CMV DNA present in the plasma and serum of renal transplant recipients with primary CMV infection was highly fragmented and therefore unlikely to represent circulating infectious virus. If this finding can be validated by others, it would allay concerns expressed by Drew et al.2 and others that plasma viremia may be source of residual CMV transmission when WBC-reduced or seronegative cellular blood components are used. It would also explain the clinical observation that plasma and plasma products do not appear to transmit CMV.3 Although WBC reduction by filtration or apheresis has been found to significantly reduce CMV DNA levels in blood donors and immunosuppressed patients experiencing CMV reactivation,7 small amounts of residual CMV DNA in the absence of infectious virus detectable by tissue culture were detected in some components after WBC reduction. Determining whether this represents noninfectious fragmented CMV DNA in plasma or is virion-associated DNA is extremely important in evaluating the potential efficacy of NAT of components after WBC reduction as a CMV prevention strategy. Considerably more data are required to evaluate the potential efficacy of NAT for CMV prevention. Specifically, it is important to determine the true residual burden of disease when technology associated with current strategies of WBC reduction and antibody screening are optimized. Further studies are needed to determine the incidence of CMV reactivation and primary infection in blood donors, the duration of infectivity of these donors, and whether they contribute significantly to the residual burden of disease. Finally, it is important to determine what CMV DNAemia means clinically; how its detection relates to infectivity in cellular and cell-free blood components. Given the evidence available, the time to consider implementing NAT of blood donors for CMV DNA for the prevention of TA-CMV infection has not yet come.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.