Dengue in the context of “safe blood” and global epidemiology: to screen or not to screen?

The aim of a safe blood supply is constantly challenged by emerging and reemerging infectious agents. Emergence is favored by modern economy and lifestyles including international travel, circulation of goods and livestock, animal and human migration, expanding arthropod distributions, and other environmental and ecologic factors. A “worst-case” scenario is an emerging blood-borne pathogen that is responsible for mostly asymptomatic infections in healthy individuals (i.e., blood donors) and is only detected when clinically important morbidities and/or mortality are observed in immunosuppressed transfusion recipients revealing the significance of the new transfusion-transmitted pathogen. Among potential emerging infectious pathogens, arboviruses, and particularly Flaviviruses, are probably the best examples of agents recently threatening the US and global blood supplies. The introduction of West Nile virus (WNV) into the United States in 1999 and the subsequent reports of increasing numbers of WNV-induced neuroinvasive disease cases reported in the New York area and later identified in the broader New England area raised early concerns around potential transfusion transmission of this emerging virus.1 After the first cases of WNV transfusion transmission were reported in 2002,2,3 a series of interventions were implemented to mitigate the risk to recipients.4 Most important and effective, WNV nucleic acid testing (NAT) was implemented under an investigational new drug (IND) exemption in 2003,5 less than a year after the first WNV transfusion transmission case was documented, and revealed that 0.02% of blood donations during the first summer of testing had WNV viremia detected by minipool NAT.5-7 It became evident to the blood community and Food and Drug Administration (FDA) that WNV NAT screening had to be systematically implemented and enhanced using targeted individual-donation NAT in high-incidence settings to prevent the transfusion transmission of this serious blood-borne pathogen. Recent studies, based in part on data from blood donor NAT screening and serosurveys, have estimated that WNV has infected more than 3 million people through 2010 and caused approximately 780,000 illnesses.8 The addition of WNV blood screening in 2003 to routine donor screening prevented thousands of WNV transfusion transmissions over the past decade. Dengue, another Flavivirus recently identified as a potentially serious threat to the blood supply,9-11 is widely spread around the world with four dengue viruses (DENV-1 to -4) circulating in more than 100 countries within tropical and subtropical climates.12 This systemic viral infection is transmitted among humans by Aedes aegypti and, less commonly, Aedes albopictus mosquitoes (there is no epidemiologically important animal reservoir).13,14 Although dengue fever has emerged within the past several hundred years, there has been a marked global reemergence of dengue infections and clinical disease in the past 25 years, with more frequent and larger epidemics associated with increasingly severe disease.15-19 Asymptomatic in most of the cases (53%-87%), the infection can lead to serious disease with symptoms ranging from debilitating but self-limited dengue fever to severe dengue with hemorrhagic fever, shock syndrome, and other potentially lethal complications.20 Although there is no effective antiviral therapy, early diagnosis followed by supportive care can reduce mortality rates for dengue hemorrhagic fever and shock syndromes to approximately 2%.21 As mosquito vectors are spreading to wider areas,22,23 there is a risk for dengue to cause repeated outbreaks in nonendemic areas. Indeed, dengue cases were recently reported in the mainland United States in Texas24,25 and Florida26,27 in Australia,28 and in several countries in Europe and Asia such as Greece, Romania, the Russian Federation, and Turkey.29-34 The cocirculation of different dengue virus types and the emergence of new dengue virus types in endemic areas35 are associated with more severe disease because immunity to one of the four dengue virus types can increase disease severity in humans upon subsequent infection with another dengue virus type.36 Dengue has been the focus of many recent studies trying to evaluate the risk for transfusion transmission in endemic areas during outbreak periods as well as in nonendemic areas due to individuals returning from endemic areas. AABB's Transfusion-Transmitted Diseases Committee recently categorized dengue, along with Babesia and variant Creutzfeldt-Jakob disease, as high-priority agents in terms of threats to US transfusion recipient safety.11 Previous studies have documented small numbers of transfusion-transmitted cases in Hong Kong and Singapore,37,38 as well as moderate rates of dengue viremic blood donors in Brazil, Honduras, and Puerto Rico.39,40 Considerable effort is now being deployed to evaluate the risks and consequences of transfusion-transmitted dengue and the need to develop sensitive assays and implement blood donor screening in the United States and elsewhere. In the current and recent issues of TRANSFUSION, four studies report new data on the risk of transfusion-transmitted dengue in two endemic regions: Puerto Rico41-43 and Brazil.44 Petersen and colleagues41 estimate the prevalence of dengue viremia in Puerto Rican blood donations over a 16-year period from 1995 through 2010. The authors applied the modeling procedure that was developed to estimate the risk of transfusion-transmitted WNV infection2,6 to project the prevalence of dengue viremia in donations in Puerto Rico. The annual prevalence of dengue-viremic blood donations was estimated based on dengue case reports from islandwide surveillance, estimates of case underreporting and data on the duration of viremia, and the unapparent-to-apparent dengue infection ratio. Based on this approach, the mean daily prevalence of dengue viremia in Puerto Rico was estimated at 7 per 10,000, with fluctuations depending on the season and the year. A maximum of 45 of 10,000 were estimated during the 2010 dengue outbreak. These results suggest that a considerable number of dengue transfusion transmission cases may have occurred in Puerto Rico over the past 16 years but gone unrecognized. Mohammed and colleagues42 present data demonstrating a very high dengue seroprevalence within the blood donor population in Puerto Rico with 92% of the blood donors in 2006 identified as having been previously exposed to dengue. The study identified immunoglobulin G reactivity by anti-dengue enzyme-linked immunosorbent assay and then used a microneutralization test to characterize the reactivity to specific dengue virus types. The findings indicate that the blood donor population on the island had been exposed to all dengue viruses, with predominant reactivity to DENV-3 and DENV-2 types, which were responsible for secondary infections in most of the cases; in contrast, DENV-1 seems to have been more recently introduced. Stramer and colleagues43 report on the presence of the dengue virus itself in the blood supply collected by the ARC during the 2007 dengue outbreak in Puerto Rico. The authors retrospectively probed donations from donors in Puerto Rico for the presence of dengue RNA by a sensitive transcription-mediated amplification (TMA) assay with confirmation and quantitation of the amount of virus by real-time reverse transcription–polymerase chain reaction (RT-PCR). They also evaluated TMA-reactive donation plasma samples for the presence of anti-dengue immunoglobulin (Ig)M and for infectious virus by cell culture. The authors found high rates of viremia in the blood collected in Puerto Rico in 2007, with 0.19% of the donations testing positive for dengue RNA. Blood donors identified in the acute phase of the infection (IgM negative) had the highest quantities of virus with viral titers of 105 to 109 copies/mL and positive mosquito cell cultures. These very concerning data demonstrate that viremic donors who are in the acute phase of infection can remain asymptomatic and present as healthy donors. Some of these viremic donors subsequently develop symptoms a few days later which, if reported to the blood center, trigger quarantine and lookback procedures. Importantly, this study also identified a case of severe transfusion-transmitted dengue, with a donor reporting severe dengue symptoms 3 days after donation and the corresponding recipient developing febrile illness 3 days after transfusion before progressing to dengue hemorrhagic fever. This third report that dengue can be transmitted through transfusion adds to earlier studies from Hong Kong37 and Singapore.38 Finally, in this issue Dias and colleagues44 report data demonstrating the presence of DENV RNA in samples from healthy donors collected at the Hemotherapy Center of Ribeirao Preto, Brazil, during the 2010 dengue outbreak. The authors used a one-step real-time RT-PCR assay to quantify DENV RNA in samples from 500 blood donors randomly selected between February and May 2010 during an epidemic of dengue in the city. They also further characterized the dengue virus type by heminested RT-PCR and sequencing. The authors established that 0.4% of the tested blood donors harbored DENV-3 viremia. These new studies add to prior work from these and other groups that demonstrate the threat posed by dengue to the blood supply.37,38,45 The rates of dengue viremic donors reported in this issue of TRANSFUSION, ranging from 0.19% in Puerto Rico during the 2007 dengue outbreak in the Caribbean to 0.4% in Ribeirao Preto during the 2010 dengue outbreak in Brazil are consistent with those in previous reports37,38 and higher than the rates of WNV-viremic donors identified in the United States during the active transmission season over the past decade and much higher than the rates of window period infections detected by NAT screening for HIV, HCV, and HBV over the past two decades. From a broader context, the World Health Organization estimates that dengue is responsible for 50 to 100 million cases of dengue fever, 500,000 cases of severe dengue, and 22,000 deaths annually throughout the world. With increasing evidence of high rates of dengue viremia in donors and cases of transfusion transmission, one wonders why dengue virus blood screening has not been implemented. The answer to this question is complicated. Given the documented rates of dengue-viremic donors in focused studies and the magnitude of outbreaks in the general population, it is surprising not to have seen many more reported cases of transfusion-transmitted dengue. Therefore, a wait-and-see approach has been adopted until dengue transfusion transmission and consequent severe illness in recipients are better quantified. It has been hypothesized that transfusion-transmitted dengue may not occur for several reasons including 1) the need for mosquito saliva to enhance dengue replication and virulence; 2) the presence of protective immunity in recipients in endemic areas; 3) frequent cotransfusion of antibody-positive units, which neutralize infectivity of viremic donations; and 4) clinical outcomes may not be as severe in immunosuppressed patients infected with dengue.45 Furthermore, prevention of transfusion transmission seems to be less of a concern in endemic areas where recipients are at greater risk to contract the infection outside or even inside hospitals from mosquito bites than through blood transfusion. Additionally, endemic countries are often developing countries and the need for dengue blood screening there is debatable and should be considered in the larger context of a population whose communities' more basic needs, such as food, housing, public infrastructures, and access to reliable medical care, are often still not fully addressed. In such contexts, prevention of transfusion-transmitted dengue is less of a priority compared to more urgent tasks such as HIV prevention and screening. Nonetheless, increasing data documenting high prevalence rates of dengue viremic donors worldwide, and recent dengue outbreaks reported in Texas24,25 and Florida,26,27 have precipitated increased attention from the US blood safety research community. The National Heart, Lung, and Blood Institute (NHLBI), in collaboration with blood bank organizations and the US FDA and the Centers for Disease Control and Prevention (CDC), are supporting studies to establish the need to screen blood in regions prone to dengue epidemics. Solutions are being evaluated in partnership with companies developing serologic (NS1Ag) and molecular diagnostic (TMA and PCR) screening assays. A prospective study was launched by the ARC in 2009 to evaluate the rate of dengue-infected blood donors in Puerto Rico and interdict infected donations. The Bio-Rad dengue NS1Ag screening assay was implemented under IND on the island in March 2010. This assay was the only direct virus screening test available on the market at that time. The IND allowed testing of all ARC blood collections on the island during the 2010 dengue season. However, a discrepancy was quickly noted between the low yield of NS1Ag-reactive units that could be confirmed by RNA tests and donor follow-up relative to the large outbreak of dengue in Puerto Rico that year, a concern that was confirmed by later testing of stored plasma aliquots from NS1Ag-negative donations for RNA, which demonstrated high rates of missed viremia. Therefore, a new screening strategy was developed for the 2011 dengue season using the highly sensitive Gen-Probe DENV TMA assay.39,40,46 On August 1, 2011, the ARC in collaboration with Gen-Probe, Inc., and with funding from NHLBI, began a “rapid retrospective testing” study that involved retesting of donations identified as NS1Ag-nonreactive using the DENV TMA assay within 1 week of blood collection. This study further confirmed that the NS1Ag screening assay was insensitive as well as nonspecific and hence not useful as a donor screening test.47 Therefore, the current plan is to change the prospective IND-approved NS1Ag screening to prospective IND-approved DENV TMA screening for the 2012 dengue season. This IND screening would be extended to all blood collections in Puerto Rico, with continued careful follow-up of RNA-positive donors, particularly those detected by the ARC. These follow-up studies of viremic donors should contribute unique insights into early viral dynamics and host determinants of dengue disease, as well as establish a repository of serial samples from viremic donors for future use by the larger scientific community, further expanding the contributions of NHLBI's BioLINCC repositories (https://biolincc.nhlbi.nih.gov/home/) to infectious disease research.48 Collaborative efforts have also been launched with partners in developing countries and particularly in Brazil where dengue is responsible for increasingly large seasonal epidemics. The REDS-III study in Brazil, sponsored by NHLBI, is now evaluating dengue virus incidence and prevalence in Brazilian blood donors and recipients, and the rates and correlates of transfusion transmission, as well as clinical manifestations associated with transfusion transmission of dengue. Such data are needed for Brazilian, US, and other international health authorities to evaluate the need for screening of donors for dengue in epidemic regions. This study is also building an “arbovirus repository” for future use to detect additional possible mosquito-transmitted viruses in Brazilian donors and recipients such as yellow fever virus, chikungunya, or new viruses discovered by expanding application of metagenomics to virus discovery.49 Our experience responding to the threat posed by dengue to the blood supply illustrates the challenge the blood community is facing when confronted with an emerging blood-borne pathogen as well as our enhanced capacity to respond to such a challenge. The lessons from the past and the network of collaborative partners and resources deployed over the years have enhanced our “EID Tool Kit,” enabling proactive surveillance and aggressive reaction to new pathogens threatening blood safety.50-52 With the strong support of NHLBI, FDA, and CDC, scientists in the blood community are currently gathering key data to determine, based on prospective studies and sound evidence, whether or not dengue donor screening needs to be implemented to protect blood recipients in the United States and throughout the world. Public–private partnerships have also been forged over the past three decades to effectively react to the emergence of deadly pathogens threatening the safety of blood transfusions by development and rigorous evaluation of potential donor screening assays. In the future, pathogen reduction techniques could represent an alternative to blood screening for dengue and other EIDs, but these technologies are still under development and not yet available in the United States. All strategies to prevent dengue transfusion transmission will come at a significant cost, and hence while trying to address the threat posed by dengue, blood bankers must also consider the consequent increase in the price of the blood components and the capacity of hospitals and national health care systems to fund those increases.53 The question “to screen or not to screen” is simple, but the answer is not. For dengue, the studies in progress will hopefully lead to a sound decision over the next several years. MPB is on the Scientific Advisory Board of Gen-Probe, Inc., which has developed a DENV RNA assay that is being employed in research studies and IND screening, and BSRI receives research support from Novartis Diagnostics and Terumo Corp. (previously CaridianBCT).