Decision making in the face of uncertainty: the challenge of emerging infectious diseases

“What's called a difficult decision is a difficult decision because either way you go there are penalties.”—Elia Kazan Infectious diseases potentially transmissible through blood transfusion continue to emerge or reemerge globally. Ten articles in this issue of TRANSFUSION focus on the most recent of these, Zika virus, the third major arbovirus within the past two decades to have been introduced to the Western Hemisphere. Zika virus follows the introduction of West Nile virus to New York City in 1999 and chikungunya virus to Saint Martin Island in 2013.1, 2 Factors contributing to their introduction and subsequent spread throughout the Western Hemisphere include causes well identified for other emerging diseases, such as increased travel and trade, urbanization, and population growth. These viruses have become transfusion safety threats despite their relatively short durations of viremia because of their high incidence of infection of the human population during outbreaks and because a proportion of those infected remain asymptomatic and donate blood or alternatively may donate blood before developing symptoms.3, 4 Despite certain similarities of these arbovirus infections, important differences in their biology, epidemiology, and clinical impact have distinct implications for transfusion medicine. Although the widely dispersed Culex species mosquitoes that transmit West Nile virus have permitted viral spread throughout much of the Americas, including all states of the contiguous United States, for unclear reasons large human outbreaks have occurred with variable seasonal intensity only in the United States and Canada.5 Humans develop insufficient viremia to efficiently infect mosquitoes and do not contribute to viral transmission, while many bird species produce high-level viremias and serve as reservoirs for West Nile virus.6 The relatively short life span and high turnover of avian reservoir species do not permit the development of long-standing herd immunity; thus, repeated outbreaks will continue indefinitely.5 Given this epidemiology, along with the propensity of West Nile virus to cause severe neuroinvasive disease, particularly among the elderly, blood donor screening has been proven to be of benefit to public health surveillance over a number of years.7 However, the future epidemiology and impact of chikungunya and Zika viruses are far less certain. Both viruses produce sufficient viremia in humans to efficiently infect Aedes aegypti mosquitoes, thus permitting a human-mosquito-human transmission cycle. The A. aegypti mosquito is ubiquitous in urban environments throughout the tropical and subtropical world, enabling both viruses to become established throughout the Western Hemisphere as a result of spread by human travel. However, A. aegypti is endemic only in the southernmost United States, with occasional seasonal introductions further north, thus limiting the geographic potential for autochthonous viral transmission.8, 9 This trend is similar to the observations with dengue, another arbovirus spread from human to human via A. aegypti mosquitoes, which has only caused transient focal outbreaks in southern Florida and Texas in recent decades, and that autochthonous chikungunya and Zika virus transmission has only been documented in these areas to date.10-13 While Aedes albopictus is a competent vector for dengue, chikungunya, and Zika viruses with wider distribution in the United States than A. aegypti, to date, only one case of autochthonous transmission of dengue in New York and none of chikungunya and Zika viruses has been convincingly shown to result from A. albopictus vectored transmission in the United States.14 Nevertheless, thousands of people with travel-associated chikungunya and Zika virus infection return to nearly all areas of the United States after travel to areas of ongoing transmission, thus presenting a potential risk of transfusion transmission in areas without autochthonous transmission.13 Concerns about transfusion transmission of chikungunya virus have been tempered by its usually short-lived illness without permanent sequelae and the likelihood that development of human herd immunity would greatly reduce transmission.3 The latter may have in fact occurred, with the halving of chikungunya cases reported to the Pan American Health Association each year since its introduction to the Western Hemisphere (http://www.paho.org/hq/index.php?option=com_topics&view=readall&cid=5927&Itemid=40931&lang=en).15 At first, Zika virus’ introduction into the Western Hemisphere seemed to raise a level of concern similar to that for chikungunya virus. However, in the fall of 2015, several months after the recognition of mosquito-borne transmission of Zika virus in Brazil, investigators noted a sharp increase in the number of infants born with microcephaly, which prompted the World Health Organization to declare a Public Health Emergency of International Concern on February 1, 2016.16 Insufficient evidence existed at that time to make a causal link between maternal Zika virus infection and birth defects; considerable skepticism remained particularly since mosquito-borne viruses had never been linked definitively to human birth defects. Another confounding factor in February 2016 was reports of sexual transmission of Zika virus, particularly since sexual contact had never been associated with transmission of any other mosquito-borne virus.17 Accumulating case reports and a case series of infants with microcephaly began to strengthen the causal relationship between Zika virus and microcephaly and to define a specific phenotype among affected infants consisting of severe microcephaly, intracranial calcifications, redundant scalp skin, hypertonia/spasticity, clubfoot, and congenital joint contractures.18 These data, along with experimental data showing that Zika virus produced cell death and attenuated future growth of human neural progenitor cells, led the CDC to conclude in April 2016 that Zika virus caused microcephaly and other serious brain defects.19, 20 Subsequent animal models demonstrated that Zika virus is a teratogen, and a case-control study demonstrated a strong relationship between maternal infection and microcephaly.21-25 The full spectrum of adverse fetal outcomes and the risk associated with infection throughout pregnancy is currently unknown and remains an area of active investigation. What is now clear, however, is the potential for serious lifelong adverse impact to a child if Zika virus were to be transmitted by transfusion to a pregnant woman or her sexual partner. The articles focusing on Zika virus in this issue collectively describe a tremendous amount of knowledge gained over the relatively short time span of about a year. For a graphic illustration of the progression of Zika virus and progress in knowledge made regarding the virus in 2016, please see Oussayef and coworkers (https://www.cdc.gov/mmwr/volumes/65/wr/mm6552e1.htm?s_cid=mm6552e1_e).26 For further discussion, the articles in this issue are categorized into five themes. When the first case of Zika virus infection was reported in Puerto Rico by the CDC at the end of December 2015 (https://www.cdc.gov/media/releases/2015/s1231-zika.html), no commercial tests were available in the United States for its detection. However, work conducted about a decade prior on West Nile virus provided a directly relevant paradigm for the development of NAT for screening of the blood supply. In addition, NAT had previously been used for the detection of Zika virus in outbreak settings outside of the Western Hemisphere. Such prior work helped facilitate both test development and the implementation of screening. Several different noncommercial entities initiated the development of laboratory developed tests for diagnostic purposes, and two commercial sponsors experienced in NAT screening tests for blood engaged in the development of commercial assays to screen potential donors. The three articles in this issue relevant to the development of NAT screening of the blood supply are illustrative of the progress made in this field. Bielaire and colleagues27 describe the use of their Zika virus laboratory developed test that was used at the time of the 2013 to 2014 Zika virus outbreak in French Polynesia, which they had determined to have a limit of detection of approximately 100 copies/mL. For context, during the period when the samples screened were collected there was a 30-day deferral in place for symptoms of arbovirus infection at the French Polynesia blood bank in Tahiti. Given the relatively high asymptomatic infection rate with Zika virus, it is not too surprising that using minipools of 3 samples they found that 2.8% of donor samples were reactive for Zika virus RNA. No transfusion-transmitted cases of Zika virus were detected. With the development of more sensitive NAT assays and knowledge of the increased sensitivity provided by individual donor NAT described below, it is interesting to speculate that the actual percentage of reactive donor samples may have been higher than that described in French Polynesia, where it was estimated that approximately 11.5% of the population reported symptomatic infection. Stone and colleagues28 describe the evaluation of a 25-member panel of samples by 11 laboratories using 17 different assays. The comparison of the various assays provided the notable finding that enhanced sensitivity was associated with using a greater sample volume of plasma for the initial RNA extraction. Although not designed to compare the investigational commercial assays developed for use in screening the blood supply, the study found that both of these assays had 100% detection of standards at levels of down to 10 to 40 estimated copies/mL and LOD50 values of less than 5 copies/mL. Given that differences in the sensitivity of various assays are a well-recognized phenomenon with NAT, the development of a reference standard to harmonize results between different assays and laboratories was highly desirable. Once again emanating from global collaboration and using data provided by 21 different laboratories, Baylis and colleagues29 report the development of such a standard: IS 11468/16 for Zika virus RNA. The understanding of the critical variables for assay sensitivity, along with the availability of a reference standard, should greatly facilitate testing with appropriate sensitivity and allow comparison of results obtained across the globe. Although a publication has appeared describing the incidence of blood donations positive for Zika virus in Puerto Rico, to date there has not been a scientific publication regarding blood donations in the rest of the United States.30 The articles by Galel and colleagues31 and Williamson and colleagues32 represent the initial positive results from the commercial investigational blood screening tests of Roche Molecular Systems, Inc., and Hologic, Inc., respectively. The Roche test was the first to be implemented, and its use was allowed to proceed under an investigational new drug application in March 2016, just 3 months after the report of the first case in Puerto Rico.30, 31 The results of this study were highly informative regarding both the assay itself and the epidemiology of Zika virus in the United States outside of Puerto Rico. There were 23 initially reactive donations out of 358,786 samples tested using individual-donor NAT, and after follow-up testing, 14 of these were determined to represent true-positive donations. All of the positive donations identified were collected in Florida. Of note regarding the assay itself, using simulated minipools of 6 samples, only seven of these donations were identified. This finding is consistent with the Zika virus assay sensitivity characteristics described above. Ten of the 14 positive donations came from individuals who had risk factors for sexual transmission or who had traveled to areas with local transmission of Zika virus within the 90 days prior. However, three individuals had neither sexual nor travel risk factors identified. Roughly 6 months after the first case of Zika was reported in Puerto Rico, testing using the Hologic assay began in June 2016.32 The report by Williamson and colleagues provides complementary data to that above for regions of the United States outside of Florida and Puerto Rico. Of 466,834 donations screened with individual-donor NAT, five were found to be positive by supplemental testing. These donations were collected in Nevada, New York, Arizona, California, and Texas. One donor had donated platelets (PLTs) 1 week before the donation that was identified as being positive, and follow-up testing available did not suggest that Zika virus was transmitted to the recipient. Based on characteristics of the donors, the authors speculate that the individuals providing the donations were likely at the tail end of their RNA-positive period yet acknowledge that there is still uncertainty regarding how long transmission can occur after initial infection. The small number of positive donations identified with either of these assays in the United States outside of Puerto Rico stands in stark contrast to the average of 1% positive donations reported from Puerto Rico during the summer months of 2016. Pathogen reduction technology is currently approved in the United States for use with apheresis PLTs and plasma, and the ability of amotosalen and UV light pathogen inactivation technology to reduce Zika virus in plasma has been previously reported.33, 34 The article in this issue by Laughhunn and colleagues35 reporting on pathogen reduction of Zika virus in red blood cell (RBC) components is significant because of both its findings and its potential implications for safety of the blood supply in the future. In vitro use of an investigational technology for pathogen reduction of RBCs, amustaline (S-303) and glutathione, was associated with complete inactivation of more than 7.75 log genomic equivalents of Zika virus RNA using NAT and 5.99 log of infectivity relative to sham treatment using a cell culture assay. This work is significant in that it provides further evidence that after appropriate additional studies, pathogen reduction technologies may ultimately be effective in mitigating arboviral threats in addition to a variety of other pathogens in RBCs, PLTs, and plasma. Although inactivation of Zika, an enveloped virus, by solvent/detergent (S/D) treatment would be expected based on evidence obtained with other similar pathogens, formal documentation of this inactivation, as well as documentation of the effect of other viral clearance interventions, is welcome. The trio of articles by Blümel, Farcet, and Kühnel and their colleagues36-38 document that Zika is indeed inactivated by S/D treatment and by standard pasteurization technology (58 to 60°C for 2 hr for albumin). Indeed, Zika virus seems to be more sensitive to heat than other closely related viruses. In addition, nanofiltration with a pore size of 40 nm or less was found to remove all Zika virus infectious activity.36 All of these findings represent reassuring news regarding Zika virus and the safety of plasma derivatives. Given that Zika virus infection is frequently asymptomatic and that serious complications outside of pregnancy are uncommon, defining a population most at risk of complications from transfusion-transmitted Zika virus is a reasonable undertaking. Note that recent publications indicate the Zika virus is most likely to cause congenital malformations with infection during the first or early second trimesters.25, 39 Infection during that gestational period is estimated to be associated with rates of microcephaly of 11% to 13%.40, 41 The article by Murphy and colleagues42 reports on the number of women receiving blood transfusions in a large tertiary care hospital in Ottawa, Canada. They note that in their hospital only 0.04% of expectant mothers receive a transfusion during the first trimester. Although this information is a welcome addition to the literature, it must be interpreted with caution for two reasons. First, the data are not directly applicable to areas with higher rates of sickle cell disease and other hemoglobinopathies. Second, and perhaps of much greater relevance, is that the data must be interpreted in the context of the potential for sexual transmission of Zika virus.43 At this time, the potential for male-to-female sexual transmission is well documented, with presence of Zika virus RNA in semen reported for up to over 90 days after infection, and the rate of clearance has been reported to be variable.44 The mean duration during which semen is infectious is not yet known. Therefore, one can conclude that the issue is larger than just transfusing a pregnant woman—it is transfusing her male partner. Pending further data on the period of infectivity, an analysis is required that takes into consideration men receiving transfusions who might then have sexual contact in the next 90 days with women who are in the first or second trimester of pregnancy. Zika virus represents yet another in a series of emerging or reemerging threats to the blood supply.45, 46 Though a virus of known identity for several decades, the potential implications of the extensive Zika virus outbreak in the Western Hemisphere became more apparent with each month as the year 2016 progressed. Given the uncertainty regarding the spread of the epidemic and the broad range of the potential vector of Aedes mosquitos in the United States, a cautious approach led initially to implementation of donor deferrals for travel to areas with local Zika virus transmission and then to testing throughout the United States and its territories. As the articles in this issue illustrate, building on experience with prior emerging infectious diseases, much has been learned in the relatively brief period of a year both about the nature of the virus and its epidemiology. This knowledge is invaluable as we refine the response to this epidemic. However, in addition to uncertainty regarding whether Zika virus will spread further or become endemic in some areas, there is also much that remains unknown about the complications of infection itself. Clearly, universal screening of the blood supply was a significant undertaking in the United States, and concern regarding resource utilization is understandable.47 However, it is too early to tell whether or not such continued universal screening is necessary. This potential requirement should become clearer during the next year as we observe whether warmer months in the Northern Hemisphere are associated with a resurgence of spread of the virus. In the meantime, as noted by Galel and colleagues, in addition to helping to facilitate an adequate blood supply by removing regional deferrals and allowing collection to continue in places such as Puerto Rico and Florida, screening of the blood supply has had the benefit of leading to the prompt reporting of reactive donors, facilitating a rapid public health response to evaluate and address potential local transmission of Zika virus.31 With the combination of the rapidity of the Zika virus outbreak in the Western Hemisphere, the potential for adverse fetal and other outcomes, and the uncertainty involved, decision-making certainly was not easy. Indeed there are real costs associated with the course that was taken that must be balanced against the potential costs that could have been incurred with a different course of action and different potential outcomes. The articles in this issue fill important gaps in our knowledge as we continue to learn more about this arboviral pathogen. The authors have disclosed no conflicts of interest. The opinions expressed herein are those of the authors and do not represent those of the Centers for Disease Control and Prevention, the U.S. Food and Drug Administration, the Department of Health and Human Services, or the U.S. Government. Peter W. Marks, MD, PhD1 e-mail: [email protected] Lyle R. Petersen, MD, MPH2 1Center for Biologics Evaluation and Research U.S. Food and Drug Administration Silver Spring, MD 2Division of Vector-Borne Diseases National Center for Emerging and Zoonotic Infectious Diseases Centers for Disease Control and Prevention Atlanta, GA

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.