BLOOD COMPONENTS Inactivation of viruses in platelet and plasma products using a riboflavin-and-UV–based photochemical treatment Shawn D. Keil,1 Abderrahmane Bengrine,2 Richard Bowen,3 Susanne Marschner,1 Nick Hovenga,1 Lindsay Rouse,1 Denise Gilmour,1 Gilles Duverlie,2 and Raymond P. Goodrich1
BACKGROUND: Multilayered blood safety programs reduce the risk of transfusion-transmitted diseases; however, there remains a risk of window period transmission of screened viruses and transmission of unscreened and emerging viruses from asymptomatic donors. To reduce this risk, a riboflavin-and-UV-light– based pathogen reduction process was evaluated against eight viral agents. STUDY DESIGN AND METHODS: Riboflavin and UV light was evaluated against the following eight viral agents: encephalomyocarditis virus (EMC), hepatitis A virus (HAV), hepatitis C virus (HCV), influenza A (FLUAV), La Crosse virus (LACV), pseudorabies virus (PRV), sindbis virus (SINV), and vesicular stomatitis virus (VSV). Before treatment, a sample was removed to determine the product’s initial viral load. After treatment the product’s viral load was reevaluated and the log reduction was calculated. RESULTS: Virus reduction after treatment with riboflavin and UV light is equivalent in platelet (PLT) and plasma units, as demonstrated by a 3.2-log reduction of EMC in plasma, PLTs, and PLT additive solution containing 35% plasma. Additionally, the following viral reductions values were observed: HAV 1.8 log, HCV at least 4.1 log, FLUAV at least 5.0 log, LACV at least 3.5 log, PRV 2.5 log, SINV 3.2 log, and VSV at least 6.3 log. CONCLUSIONS: The results observed in this study suggest that treating PLT and plasma products with a riboflavin-and-UV-light–based pathogen reduction process could potentially eliminate window period transmission of screened viruses and greatly reduce the risk of transfusion transmission of unscreened viruses.
espite a multilayered approach to blood safety, residual transfusion risks associated with viral agents still exist. Physical examination eliminates symptomatic donors, the donor questionnaire excludes high-risk donors, and screening removes blood units found to contain known agents (e.g., human immunodeficiency virus [HIV], hepatitis C virus [HCV], hepatitis B virus [HBV]) from the transfusion pool. Although the use of serology and/or nucleic acid testing (NAT) is quite sensitive at detecting known agents, it still relies on the specific target to be at or above a minimum concentration to be detected.1 The length of time that the specific target remains below the detection threshold for a given screening assay defines the residual risk for that pathogen. This risk is typically associated either with the initial asymptomatic phase or when the donor has a silent chronic infection where the virus does not reach detectable levels nor does it trigger an antibody response. A
ABBREVIATIONS: ATCC 5 American Type Culture Collection; CHIKV 5 chikungunya virus; CPE 5 cytopathic effect; EMC 5 encephalomyocarditis virus; FFU 5 focusforming units; FLUAV 5 influenza A; GLP 5 good laboratory practice; LACV 5 La Crosse virus; PFU 5 plaque-forming units; PRT 5 pathogen reduction technology; PRV 5 pseudorabies virus; SINV 5 sindbis virus; TCID50 5 50% tissue culture infectious dose; VSV 5 vesicular stomatitis virus; WNV 5 West Nile virus. From 1Terumo BCT, Lakewood, Colorado; 2Biobanque de Picardie, EA4294, UPJV, CHU-Amiens, Amiens, France; and 3
Colorado State University, Fort Collins, Colorado.
Address reprint requests to: Raymond P. Goodrich, PhD, Terumo BCT, 10810 W. Collins Avenue, Lakewood, CO 80215; e-mail: [email protected]
All studies were funded by Terumo BCT. Received for publication July 18, 2014; revision received December 17, 2014; and accepted December 26, 2014. doi:10.1111/trf.13030 C 2015 AABB V
TRANSFUSION 2015;55;1736–1744 1736 TRANSFUSION Volume 55, July 2015
INACTIVATION OF VIRUSES USING RIBOFLAVIN AND UV
recent publication from the Canadian Blood Services estimated the residual risk of transfusion transmission of HIV, HCV, and HBV using the incidence rates of donor seroconversions from all donations from 2006 to 2009. They reported a residual risk of 1 per 8 million donations (HIV), 1 per 6.7 million donations (HCV), and 1 per 1.7 million donations (HBV).2 Using the same modeling they estimated the following window period days: 9.5 days (HIV), 8.0 days (HCV), and 38.3 days (HBV).2 Another analysis of window period risk days of HCV, HIV, and HBV by Weusten and colleagues3 evaluated minipool sizes of 16 and 8, along with individual NAT: HCV 3.1, 2.7, and 1.5 days; HIV 6.3, 5.5, and 3.3 days; and HBV 24.4, 22.2, and 15.6 days, respectively. Although these two studies both estimate a relatively small window period, there have been documented transfusion transmission of these viruses.4 In 2008, a repeat donor donated a unit of whole blood in Missouri.4 His donation was screened using both an enzyme immunoassay (Genetic Systems HIV-1/HHIV-2 Plus O EIA, Bio-Rad Laboratories, Redmond, WA) and by 16-donation minipool NAT (Procleix HIV-1 NAT, Gen Probe, San Diego, CA). Both the tests were negative; however, the recipient of the plasma component later tested positive for HIV. The donor did not indicate any high-risk behaviors on the screening questionnaire at the time of donation; however, later the donor admitted to not being truthful on the donor screening questionnaire. In addition to HIV, HBV, and HCV, blood donations can also be screened for other viruses, such as West Nile virus (WNV) in the United States. Again despite the low risk, documented transfusion transmissions have occurred with WNV.5-8 In 2012, a fatal case of WNV transmission in Colorado was reported as a probable platelet (PLT) unit transfusion–associated event from an asymptomatic donor.5 The PLT unit implicated as the source of the transmission was part of a 5-unit minipool NAT that tested positive using the Cobas TaqScreen West Nile test (Roche Molecular Systems, Pleasanton, CA). All units were rescreened as individual NAT using the same assay; however, all units tested negative and were released for transfusion per the Food and Drug Administration guidance at the time.9 Beyond the low-level risk associated with screened viruses lies a potentially greater risk from unscreened and emerging viruses. Some of these agents are unscreened because they are well characterized and pose little threat to the average blood recipient; however, some of these agents can have more serious consequences if they are transfused to an immune-suppressed recipient, for example, human parvovirus B19.10,11 Other viruses may only have recently become a threat due to expanding ranges of the vector (e.g., dengue virus12-14), mutations of the virus (e.g., chikungunya virus [CHIKV]15), or a recent zoonotic event (e.g., swine flu, avian flu, and Middle East respiratory syndrome). When viruses appear suddenly, screening tests are typically unavailable due to the novelty of the
TABLE 1. Genomic, structural, and family composition of viral agents evaluated using riboflavin and UV light Virus
EMC HAV Human HCV FLUAV LACV PRV SINV VSV
(1) ss RNA (1) ss RNA (1) ss RNA (–) ss RNA (–) ss RNA ds DNA (1) ss RNA (–) ss RNA
No No Yes Yes Yes Yes Yes Yes
Picornaviridae Picornaviridae Flaviviridae Orthomyxoviridae Bunyaviridae Herpesviradae alpha Togavirdae Rhabdoviridae
agent. Development and approval time only further delay the availability of these tests. In the case of WNV it took 9 months, after screening had been recommended, before a commercial assay was available.16 One potential way to provide an additional layer of safety to help prevent transfusion transmission of viruses is through the use of a broad-spectrum pathogen reduction process. The Mirasol Pathogen Reduction Technology (PRT) System for Platelets and Plasma, a system that utilizes riboflavin as a photosensitizer in combination with a UV light illumination device (Terumo BCT, Lakewood, CO), has been developed to reduce the infectivity of a broad range of blood-borne pathogens.17-23 The purpose of this study was to expand on the existing published viral reduction data and establish the effectiveness of this PRT system against a broad range of viral families. In this study the reduction of encephalomyocarditis virus (EMC), hepatitis A virus (HAV), HCV, influenza A (FLUAV), La Crosse virus (LACV), pseudorabies virus (PRV), sindbis virus (SINV), and vesicular stomatitis virus (VSV) was evaluated in either plasma or PLT units using in vitro cell culture assay systems (see Table 1 for virus characteristics). The viruses selected for this study are composed of known human pathogens (HAV, HCV, FLUAV, and LACV) and recommended model viruses (EMC, PRV, SINV, and VSV).24 The list was not meant to be inclusive of all possible transfusion-transmitted pathogens, but rather evaluate a broad family of enveloped and nonenveloped viruses.
MATERIALS AND METHODS PLT and plasma products All blood products were collected at an accredited blood bank and shipped to Terumo BCT. PLT units collected and treated in plasma met the following collection specifications: 90 to 360 mL volume and 0.80 to 1.5 3 106 PLTs/lL concentration. All apheresis PLT units were collected in ACD-A and units were rested (no agitation) for a minimum of 2 hours after collection and before treatment. After the rest period and before treatment PLT units were evaluated for PLT swirl and basic blood gas variables to ensure optimal product quality. Cell counts were performed on a Volume 55, July 2015 TRANSFUSION 1737
KEIL ET AL.
hematology analyzer (AcT diff, Beckman-Coulter, Brea, CA) and blood gases were tested using a blood gas analyzer (ABL 725, Radiometer, Copenhagen, Denmark). Plasma units were derived from both apheresis and whole blood collections and had a treatment range of 170 to 360 mL. Plasma units consisted of both fresh-frozen plasma (frozen to –30 C within 8 hr of collection) and recovered plasma (frozen to –20 C within 8-24 hr after collection). Plasma combined with PLT additive solution (PAS) met the following specifications after PAS addition: 250 to 450 mL total volume with a 30% to 45% plasma content. The PAS used was SSP1 (MacoPharma UK Ltd, Twickenham, UK) and the solution was manually added to the plasma units at Terumo BCT.
PRT process for PLTs and plasma The Mirasol PRT process for PLTs and plasma has been described previously in detail.25 Briefly, PLT and plasma units were aseptically transferred to a 1-L citrate plasticized polyvinyl chloride extended-life PLT (ELP) bag and were dosed with 35 6 5 mL of 500 mmol/L riboflavin solution in 0.9% sodium chloride, pH 4.0 to 5.0 (Terumo BCT). After inoculation of the plasma and PLT units with virus, the units were placed into a Mirasol illuminator (Terumo BCT) for UV treatment. PLT units collected and treated in plasma, along with plasma units, were individually illuminated with 6.2 J/mL energy. Plasma units treated in the presence of PAS were illuminated with a comparable energy dose based on the percentage of plasma remaining in the unit. Plasma was used instead of PLTs due to greater availability and its longer shelf life. Data presented in this article show no difference between products that contain PLTs versus products that do not.
Viral reduction studies The viral reduction studies were performed at a variety of locations. The locations include in-house studies (Terumo BCT), collaborations with public entities (e.g., universities), and work with good laboratory practice (GLP) qualified for-profit viral validation laboratories. Due to the for-profit nature of viral validation laboratories, not all assay information is available. All agents tested except HCV and LACV were conducted in compliance with GLP regulations. The laboratories used to test HCV and LACV primarily do research and are not equipped to adhere to GLP. Viruses tested in house (EMC and HAV) were harvested when the cytopathic effect (CPE) reached more than 75%. Viral supernatant was centrifuged at 3200 3 g for 10 minutes to remove cellular debris. Further clarification of the viral stock was accomplished by filtering the virus through a 0.22 mm filter. The clarified viral stock was stored at 280 C until needed. 1738 TRANSFUSION Volume 55, July 2015
Apheresis PLT products collected for SINV, VSV, and PRV were sent overnight from Colorado in a blood bank– approved shipping container to the appropriate sites. Upon arrival at the viral validation laboratories the PLTs were reevaluated for swirl to ensure product quality after shipment.
EMC EMC virus (American Type Culture Collection [ATCC] #VR-129B) obtained from ATCC (Manassas, VA) was propagated in Vero cells (ATCC #CCL-81) grown at 37 C and 5% CO2 for 2 to 3 days in Dulbecco’s modified Eagle’s medium (DMEM; Corning Cellgro, Manassas, VA) with 10% heat-inactivated fetal bovine serum (FBS). EMC reduction studies were performed at the Terumo BCT facility. A total of 55 apheresis PLT products and 29 plasma products were collected and transported to Terumo BCT. EMC virus was inoculated into each of the units at 1% to 5% of the total volume. Units were treated with the standard PRT process as described above. Pretreatment samples, after riboflavin addition, and posttreatment samples were serially diluted in DMEM with 10% FBS and titered over CV-1 cells (ATCC #CCL-70). Titers were performed by plating 0.1 mL of sample onto 10 replicate wells per dilution using a 96-well plate format. Plates were incubated at 37 C and 5% CO2 for 7 to 8 days at which point they were scored for CPE. The calculated titers are expressed as 50% tissue culture infectious dose (TCID50)/mL.
HAV HAV (ATCC #VR-1402) was propagated in FRhK-4 cells (ATCC #CRL-1688) grown at 35 C and 5% CO2 for 8 to 10 days in DMEM with 10% FBS. HAV reduction studies were also performed at Terumo BCT. A total of 11 human apheresis PLT products and six plasma products were collected and transported to Terumo BCT. Only products from antiHAV–free donors were used (data not shown). This ensured the virus was not neutralized by native antibodies. HAV virus was inoculated into each of the units at 3% to 5% of the total volume. Units were treated with the standard PRT process as described above. Pretreatment and posttreatment samples were serially diluted in DMEM with 10% FBS and titered over FRhK-4 cells. Titers were performed by plating 0.1 mL of sample onto 10 replicate wells per dilution using a 96-well plate format. Plates were incubated at 35 C and 5% CO2 for 21 to 22 days at which point they were scored for CPE. The calculated titers are expressed as TCID50/mL.
Human HCV The HCV reduction study was performed as collaboration between Terumo BCT and the Biobanque de Picardie
INACTIVATION OF VIRUSES USING RIBOFLAVIN AND UV
(Amiens, France). This study utilized plasma collected from five different whole blood donors and diluted in PAS. The HuH-7 cell line (RCB1366)26 was used to propagate the virus and as the reporter cell line. The cells were grown at 37 C and 5% CO2 in DMEM (Jacques Boy, Reims, France) supplemented with 10% FBS. Virus production was achieved by using a plasmid encoding the JFH1-CS-A4 genome, a modified version of the full-length JFH1 genome (Genotype 2a; GenBank Accession Number AB237837; provided by T. Wakita). HCV RNA was produced by in vitro transcription of the plasmid encoding JFH1-CS-A4 genome, which was electroporated into HuH-7 cells as previously described.27 Supernatant from the electroporated cells was recovered 10 days postelectroporation. The supernatant was centrifuged at 300 3 g for 10 minutes to remove cellular debris and filtered through a 0.45-mm-pore-sized membrane. This stock was used to perform two successive infections of HuH-7 cells by collecting supernatant 72 hours after infection, filtering, and then transferring virus-containing supernatants onto na€ıve cells. The clarified viral supernatant was aliquoted (5 mL) and stored at 280 C until needed. Due to the difficulty of obtaining large quantities of high-titer stock virus, the HCV reduction study was performed in a 48-well plate using a 250-mL treatment volume. Each 48-well plate was treated with a scaled dose that was verified to be equivalent to the standard PRT process (see Supplement S1, available as supporting information in the online version of this paper). For each unit tested a total of 862 mL of plasma was combined with 1271 mL of SSP1, 275 mL of riboflavin, and 342 mL of virus. Each of the five donor plasma samples was split into three treated replicates and one untreated product. The untreated products were stored in the dark while the treated replicates were undergoing the illumination process. Each treated replicate consisted of three 250mL volumes that were combined after treatment. Infectivity of the treated and nontreated samples was measured by immunofluorescence 48 hours after inoculation of the reporter HuH-7 cells using a monoclonal antibody directed against the glycoprotein E2 of HCV.28 Pretreatment samples were serially diluted and three replicate wells were plated at each dilution using 100 mL/well. To avoid cytotoxicity of the test sample on the reporter cells (data not shown), treated replicates were diluted 1:10 in DMEM with 10% FBS (approx. 7.5 mL final volume) and the entire volume was plated out over 24 wells, with each well containing 300 mL of sample. Titers before and after treatment were determined by counting the number of foci. The calculated titers are expressed as focus-forming units (FFU)/mL.
FLUAV The FLUAV reduction studies were performed at both MDS Pharma Services (Bothell, WA) and BioReliance
(Rockville, MD). The strain of FLUAV used at BioReliance was A/PR/8/34 (ATCC #VR-1469); the source of FLUAV strain used at MDS Pharma Services was not provided. Between the two study sites, three human apheresis PLT products and one plasma product were used to evaluate FLUAV reduction. FLUAV was inoculated into each of the units at 3.5% to 5% of the total volume. The units were treated with the standard PRT process. Pretreatment and posttreatment samples were serially diluted and titered over MDCK cells (ATCC #CCL-34). The samples collected from the plasma unit treated at MDS Pharma Services were titered by plating 0.5 mL of sample onto individual wells using a plaque assay format on six-well plates. Samples tested at BioReliance were also titered over MDCK cells; however, a TCID50 assay was used instead. Plates were monitored daily for CPE. The calculated titers are expressed as plaque-forming units (PFU)/mL (MDS Pharma Services) and TCID50/mL (BioReliance).
LACV The LACV reduction study was performed as a collaboration between Terumo BCT and Colorado State University (Fort Collins, CO). The stock of LACV was prepared by inoculating two 150-cm2 flasks of Vero cells with approximately 106 PFU of the virus (H78 strain). Three days after inoculation the virus-rich medium was collected and centrifuged at 5000 3 g for 10 minutes. The clarified supernatant was frozen at 280 C until needed. LACV was inoculated into each of the six recovered plasma units at 6% of the total volume. Units were treated with the standard PRT process. Pretreatment samples were serially diluted 10-fold in BA-1 medium, whereas the posttreatment samples were only plated at the last noncytotoxic, noninterfering dilution. Confluent monolayers of Vero cells, grown on six-well plates and drained of medium, were inoculated with 300 mL of sample. The plates were rocked every 10 to 15 minutes for 45 minutes, after which a 2-mL overlay (MEM, 5% FBS, 0.5% agarose) was added to each well. The plates were placed into a 5% CO2 incubator at 37 C. Two days later a second overlay, identical to the first, but containing 0.03% neutral red, was added to each well. Plaques were counted the following day. Virus titers were calculated based on plaque count and dilution and are expressed as PFU/mL.
PRV The PRV reduction study was performed at MDS Pharma Services. PRV was inoculated into each of the 6 PLT units at 5% of the total volume. The units were treated with the standard PRT process. Pretreatment and posttreatment samples were serially diluted and titered over Vero cells. Titers were performed by plating 0.5 mL onto individual wells using a six-well plate format. Plates were monitored Volume 55, July 2015 TRANSFUSION 1739
KEIL ET AL.
TABLE 2. Reduction of different viral species when treated with riboflavin and UV light Mean titer (log) Virus
Reporter cell line
3.5 6 0.6 TCID50/mL 3.3 6 0.8 TCID50/mL 3.4 6 0.2 TCID50/mL
5.7 6 0.3 TCID50/mL 5.9 6 0.2 TCID50/mL
4.1 6 0.3 TCID50/mL 3.7 6 0.3 TCID50/mL
Human HCV FLUAV
4.2 6 0.1 FFU/mL 5.2 TCID50/mL 5.8 PFU/mL
0.1 FFU/mL 0.8 TCID50/mL 0.3 PFU/mL
LACV PRV SINV VSV
Vero Vero Vero Vero
55 23 6 84 11 6 17 5 1 1 2 6 6 6 6
6.7 6 0.4 TCID50/mL 6.5 6 0.6 TCID50/mL 6.6 6 0.1 TCID50/mL
PLTs in plasma Plasma Plasma in PAS Overall reduction PLTs in plasma Plasma in PAS Overall reduction Plasma in PAS PLTs in plasma Plasma Overall reduction Plasma PLTs in plasma PLTs in plasma PLTs in plasma
4.4 6 0.04 PFU/mL 6.2 6 0.1 PFU/mL 4.4 6 0.2 PFU/mL 7.4 6 0.6 PFU/mL
0.92 PFU/mL 3.7 6 0.1 PFU/mL 1.2 6 0.3 PFU/mL 1.1 PFU/mL
3.2 6 0.4 3.2 6 0.4 3.2 6 0.1 3.2 6 0.4 1.6 6 0.4 2.2 6 0.4 1.8 6 0.5 4.1* 4.4 5.5 5.0 3.5† 2.5 6 0.1 3.2 6 0.2 6.3‡
Plaque Plaque Plaque Plaque
* Five of five replicates at the limit of detection. † Six of six replicates at the limit of detection. ‡ Three of six replicates at the limit of detection.
daily for plaque development. The calculated titers are expressed as PFU/mL.
SINV The SINV (Strain Ar-339) reduction study was performed at WuXi AppTec (Philadelphia, PA). SINV was inoculated into each of the 6 PLT units at 2.5% of the total volume. The units were treated with the standard PRT process. Pretreatment samples were serially diluted, whereas the posttreatment sample was only plated at the last noncytotoxic, noninterfering dilution. The virus was titered over Vero cells. Plates were monitored for plaque assay development. The calculated titers are expressed as PFU/mL.
VSV The VSV reduction study was also performed at MDS Pharma Services. VSV was inoculated into each of the 6 PLT units at 5% of the total volume. The units were treated with the standard PRT process. Pretreatment and posttreatment samples were serially diluted and titered over Vero cells. Titers were performed by plating 0.5 mL onto individual wells using a six-well plate format. Plates were monitored daily for plaque assay development. The calculated titers are expressed as PFU/mL.
Calculation of viral reduction In general, the reduction factor (log reduction) was calculated by applying the following equation29: RF 5 log10 [(Volumebefore illumination) 3 (Concentrationbefore illumination)/(Volumetreated) 3 (Concentrationtreated)]. 1740 TRANSFUSION Volume 55, July 2015
In the above studies the Volumebefore illumination and Volumetreated are identical. When a posttreatment sample was negative for the presence of virus, the limit of detection for the assay had been reached. The following equation was used to calculate the theoretical detection limit. All values at the limit of detection are considered less than or equal to the calculated detection limit: logðPÞ log 12 Vv N LOD ¼ log V N¼
where N is the lowest number of particles in product that can be detected with 1 – P confidence, P is the probability that a virus will be undetected (for 95% confidence of detecting a virus, P 5 0.05), V is the total product volume; and v is the (volume inoculated mL) 3 (number of replicates) 3 (last dilution inoculated).
RESULTS The reduction of eight different viral agents was evaluated using a riboflavin-and-UV-light–based PRT process. These agents represent a wide range of viral families, which included both RNA and DNA viruses, along with enveloped and nonenveloped viruses (see Table 1 for virus characteristics). All work involved using in vitro cell cultures systems that measured viable virus. Table 2 summarizes the log reduction achieved when these agents were treated with riboflavin and UV light. Additionally Table 2 describes the
INACTIVATION OF VIRUSES USING RIBOFLAVIN AND UV
TABLE 3. Viral agents previously evaluated using riboflavin and UV light treatment Virus CHIKV35,36 HEV34 HIV cell associated25 HIV intracellular25 Porcine parvovirus25 WNV25
2.2 6 0.9 and 3.7 3 5.9 6 0.2 4.5 6 0.4 5.0 5.1*
(1) ss RNA (1) ss RNA (1) ss RNA
Yes No Yes
Togavirdae Hepevivirdae Retroviridae
(1) ss DNA (1) ss RNA
* Ruane and coworkers25 reported a 5.2-log reduction for WNV. After publication the authors found a small error in the calculation of the assay detection limit. This correction caused a 0.1-log difference in the WNV reduction reported.
assay type, reporter cell line used, and matrix the virus was treated in. EMC virus data presented in Table 2 demonstrate equal reduction between treated plasma and PLT units, indicating that the riboflavin and UV light process is not affected by presence of PLTs and the system yields similar reduction values regardless of the product type. For studies conducted in PAS, plasma units were diluted with PAS to a 35% to 40% plasma carryover to represent PLTs stored in additive solution. A total of 84 blood products were used to evaluate the reduction of EMC virus using PRT. The blood products included 55 units of PLTs treated in plasma, 23 units of plasma, and 6 units of plasma combined with PAS. All treatment conditions achieved equivalent 3.2-log reduction (Table 2). This agent was used to internally evaluate the robustness of the riboflavin and UV light process under different treatment conditions, thus the large sample size. All units met the system specifications for the riboflavin and UV process. The PRT system was also evaluated against HAV, another nonenveloped virus. A 1.6 6 0.4-log reduction was observed in the 11 PLT units treated in plasma and a 2.2 6 0.4-log reduction was observed for the 6 plasma/PAS units (Table 2). The overall combined log reduction for HAV is 1.8 6 0.5 log. FLUAV is a common human virus; thus, there was a potential that previous donor exposure to the wild-type virus could produce units containing neutralizing antibody. Units were not screened for the presence of antibody to FLUAV; however, the initial viral titer sample was evaluated against the stock viral titer. Units with initial titers that significantly deviated from the theoretical starting titer (>1.0 log) were not used as it was suspected they contained neutralizing antibodies (data not shown). One PLT unit and 1 plasma unit were evaluable and demonstrated at least 4.4 and at least 5.5 log (Table 2). The combined reduction of the two units is at least 5.0 log. The recent isolation of a cell culture–adapted strain of HCV has allowed for PRT studies to be carried out using an in vitro cell culture system. However, due to a limited supply of viral stock and the relatively low-stock viral titer, the assay was limited to 250-mL illuminations. A total of 5
different plasma units were used and following treatment using riboflavin and UV light there was no detectable HCV virus. The calculated log reduction was at least 4.1 log. PRV, SINV, and VSV were all treated in standard apheresis-collected PLTs. A total of 6 units were treated for each agent and the log reduction observed was 2.5 6 0.1, 3.2 6 0.2, and at least 6.3 6 0.6, respectively (Table 2). A total of 6 plasma units were used to evaluate LACV. The log reduction observed was at least 3.5 6 0.04 (Table 2).
DISCUSSION In order to provide a broad spectrum of protection against viral agents that could be found in donated blood, a pathogen reduction process must demonstrate a robust level of effectiveness against a diverse profile of agents. The purpose of these studies was to assess the effectiveness of riboflavin and UV light against a broad range of viral pathogens. Some of the viruses chosen are known human pathogens with transfusion risk (e.g., HAV and HCV), whereas other agents were chosen to demonstrate the robustness of the PRT system. One challenge of demonstrating the robustness of a PRT process is in determining how much viral reduction must be achieved for the process to be considered robust in performance.30 A conservative estimate for the sensitivity of screened viruses using NAT testing is fewer than 50 copies/ mL.3,31 Thus when using a minipool NAT strategy of 16 donors, virally infected units with greater than 800 copies/ mL (2.9 log) would be detectable by NAT. A minipool size of six and eight donors reduces this to 300 copies/mL (2.5 log) and 400 copies/mL (2.6 log), respectively. This analysis suggests that a PRT process that demonstrates a modest 3-log reduction of screened viruses, when used in combination with minipool NAT, should eliminate most if not all window period virus transmission of HIV, HBV, and HCV. This coincided with the European Committee on Blood Transfusion recommendation that states a PRT process should reduce screened pathogens by at least 1000-fold (3 log).32 In a previous study we reported the ability of riboflavin and UV light to reduce both intracellular and cellassociated HIV by 4.5 and 5.9 log, respectively25 (Table 3), Volume 55, July 2015 TRANSFUSION 1741
KEIL ET AL.
along with the at least 4.1-log reduction of HCV observed in this study. For both of these screened viruses, implementation of this PRT process could potentially eliminate window period transmissions. Due to the lack of a robust in vitro cell culture system for measuring infectivity, the effectiveness of riboflavin and UV light against HBV has only been assessed with PCR amplification assays and has been shown in this system to result in prevention of amplification of samples containing up to 10,000 gEq/mL.33 Assessing the effectiveness against emerging viruses represents a challenge due to the variable nature of these agents such as their titers in donated blood, the length of the asymptomatic phase, and the number of copies required to provide an infectious dose. Together these variables prohibit a firm statement regarding the potential effectiveness of PRT processes against them. However, a way to assess the likely effectiveness of PRT methods against emerging agents is to determine their performance against a broad spectrum of known viral agents. In addition to the viruses tested in this study, prior evaluations have demonstrated the ability of riboflavin and UV light to inactivate at least 3 logs of hepatitis E virus (HEV) virus,34 at least 5.1 logs of WNV,25 at least 5.0 logs of porcine parvovirus,25 and 2.2 logs to at least 3.7 logs of CHIKV35,36 (Table 3). Altogether, this article summarizes the effectiveness of riboflavin and UV light against 10 different viral families, composed of various combinations of RNA, DNA, single-stranded, double-stranded, enveloped, and nonenveloped viruses. The treatment resulted in a reduction to the limit of detection for the two flaviviruses tested (HCV and WNV), a reduction of 2.2 to at least 3.7 for the two togaviruses tested (CHIKV and SINV), and a 1.6- and 3.2-log reduction of the two picornaviruses tested (HAV and EMC, respectively). This suggests model viruses within the same viral family may provide an estimate of performance for untested agents; however, it will always be best practice to directly evaluate the agent of concern using an infectivity assay, if such an assay exists. Another difficulty in conducting in vitro pathogen reduction experiments is understanding how the results translate in a clinical setting with a virally contaminated blood product being transfused into a recipient. In general, cell culture–based infectivity assays test the ability of a pathogen to replicate under ideal conditions. Viruses grown in laboratory cell cultures have often been optimized for bench work and have been naturally selected to be robust in the number of infectious particles. Furthermore, the cell culture systems grow apart from and isolated from the effects of host immune cells and complement. All of these factors make it likely that virus infectivity measured using an in vitro assay is higher than what is observed in vivo with native viral strains. While it is true that implementation of a PRT system will add cost, it also offers the potential to remove costs associated with producing safer blood products. Provided 1742 TRANSFUSION Volume 55, July 2015
demonstration of effectiveness of a PRT process for inactivating bacteria, WNV, cytomegalovirus, Chagas, and white blood cells, the potential exists for the PRT process to both replace testing of these agents and stop the use of gamma irradiation. In many centers in Europe where these technologies are employed, these changes have occurred successfully. In addition, the use of a PRT process may also eliminate the need for donor deferrals due to malaria restrictions on travel and implementation of new tests for emerging agents such as Babesia, dengue viruses, Middle East respiratory syndrome virus, and CHIKV. Unlike testing, PRT is a proactive approach and has the potential to remove existing costs, reduce further future expenses and complexity, and maintain blood supply availability, all while affording reduced risks of transfusion-related complications in patients. Custer and coworkers37 evaluated the cost-effectiveness of pathogen reduction compared to current screening methods. A reported cost-effectiveness of $1,423,000/quality-adjusted life-year may be an underestimate since emerging pathogens were not considered in the analysis. Implementing a PRT system also has its limitations that need to be weighed against its benefits. Most PRT processes are not effective against spore-forming bacteria and do not remove prions from the donated product. PRT-treated products may also induce oxygen-mediated damage to pathogens, proteins, and cellular components.38 In addition, the inactivation of a pathogen in the donated product may not reach 100%; however, no detection system is 100% effective at preventing transfusion transmission of a pathogen either. In summary, we evaluated the effectiveness of a PRT process against eight viral agents of varying composition and structure in this study. When combined with the previously published results for six other viruses of varying composition and structure, the data suggest that this system could potentially eliminate window period transmission of screened viruses and greatly reduce the risk of transfusion transmission of nonscreened viruses entering the blood supply. CONFLICT OF INTEREST SK, SM, NH, DG, LR, and RG are employees of Terumo BCT, the manufacturer of the technology described in this article. RB, GD, and AB have disclosed no conflicts of interest.
REFERENCES 1. Roth WK, Busch MP, Schuller A, et al. International survey on NAT testing of blood donations: expanding implementation and yield from 1999 to 2009. Vox Sang 2012;102:82-90. 2. O’Brien SF, Yi QL, Fan W, et al. Current incidence and residual risk of HIV, HBV and HCV at Canadian Blood Services. Vox Sang 2012;103:83-6.
INACTIVATION OF VIRUSES USING RIBOFLAVIN AND UV
3. Weusten J, Vermeulen M, van Drimmelen H, et al. Refinement of a viral transmission risk model for blood donations in seroconversion window phase screened by nucleic acid testing in different pool sizes and repeat test algorithms. Transfusion 2011;51:203-15.
against Babesia microti in apheresis platelets and plasma. Transfusion 2010;50:1019-27. 19. Cardo LJ, Rentas FJ, Ketchum L, et al. Pathogen inactivation of Leishmania donovani infantum in plasma and platelet concentrates using riboflavin and ultraviolet light. Vox Sang
4. Centers for Disease Control and Prevention (CDC). HIV transmission through transfusion—Missouri and Colo-
2006;90:85-91. 20. Keil SD, Kiser P, Sullivan JJ, et al. Inactivation of Plasmo-
rado, 2008. MMWR Morb Mortal Wkly Rep 2010; 59:
dium spp. in plasma and platelet concentrates using
1335-9. 5. Centers for Disease Control and Prevention (CDC). Fatal West Nile virus infection after probable transfusionassociated transmission—Colorado, 2012. MMWR Morb Mortal Wkly Rep 2013;62:622-4. 6. Montgomery SP, Brown JA, Kuehnert M, et al. Transfusionassociated transmission of West Nile virus, United States 2003 through 2005. Transfusion 2006;46:2038-46.
riboflavin and ultraviolet light. Transfusion 2013;53:227886. 21. Goodrich RP, Gilmour D, Hovenga N, et al. A laboratory comparison of pathogen reduction technology treatment and culture of platelet products for addressing bacterial contamination concerns. Transfusion 2009;49: 1205-16. 22. Rentas F, Harman R, Gomez C, et al. Inactivation of Orientia
7. Petersen LR, Busch MP. Transfusion-transmitted arboviruses.
tsutsugamushi in red blood cells, plasma, and platelets with
Vox Sang 2010; 98:495-503. 8. Meny GM, Santos-Zabala L, Szallasi A, et al. West Nile virus
riboflavin and light, as demonstrated in an animal model. Transfusion 2007;47:240-7.
infection transmitted by granulocyte transfusion. Blood 2011;117:5778-9. 9. US Food and Drug Administration. Guidance for industry: use of nucleic acid tests to reduce the risk of transmission of West Nile virus from donors of whole blood and blood components intended for transfusion. Rockville (MD): US Food and Drug Administration; 2009 [cited 2014 Dec 16]; Available from: http://www.fda.gov/downloads/biologicsbloodvaccines/guidancecomplianceregulatoryinformation/guidances/blood/ucm189464.pdf. 10. Florea AV, Ionescu DN, Melhem MF. Parvovirus B19 infection in the immunocompromised host. Arch Pathol Lab Med 2007;131:799-804. 11. De Renzo A, Azzi A, Zakrzewska K, et al. Cytopenia caused by parvovirus in an adult ALL patient. Haematologica 1994; 79:259-61. 12. Wilder-Smith A. Dengue infections in travellers. Paediatr Int Child Health 2012;32Suppl1:28-32. 13. Guzman A, Isturiz RE. Update on the global spread of dengue. Int J Antimicrob Agents 2010;36Suppl1:S40-2. 14. Banu S, Hu W, Guo Y, et al. Dynamic spatiotemporal trends of dengue transmission in the Asia-Pacific region, 19552004. PLoS One 2014;9:e89440. 15. Vazeille M, Moutailler S, Coudrier D, et al. Two Chikungunya
23. Keil SD, Saakadze N, Bowen R, et al. Riboflavin and ultraviolet light for pathogen reduction of murine cytomegalovirus in blood products. Transfusion 2005;55: 858-63. 24. ICH Q5A. Viral safety evaluation of biotechnology products derived from cell lines of human or animal origin. Geneva: ICH; 1998. Appendix 2, p. 26. 25. Ruane PH, Edrich R, Gampp D, et al. Photochemical inactivation of selected viruses and bacteria in platelet concentrates using riboflavin and light. Transfusion 2004;44: 877-85. 26. Nakabayashi H, Taketa K, Miyano K, et al. Growth of human hepatoma cells lines with differentiated functions in chemically defined medium. Cancer Res 1982;42: 3858-63. 27. Helle F, Vieyres G, Elkrief L, et al. Role of N-linked glycans in the functions of hepatitis C virus envelope proteins incorporated into infectious virions. J Virol 2010;84: 11905-15. 28. Fournier C, Helle F, Descamps V, et al. Natural selection of adaptive mutations in non-structural genes increases transencapsidation of hepatitis C virus replicons lacking envelope protein genes. J Gen Virol 2013;94:996-1008. 29. Committee for Proprietary Medicinal Products (CPMP).
isolates from the outbreak of La Reunion (Indian Ocean)
Note for guidance on virus validation studies: the design,
exhibit different patterns of infection in the mosquito, Aedes albopictus. PLoS One 2007;2:e1168-
contribution and interpretation of studies validating the inactivation and removal of viruses. CPMP/BWP/268/95.
16. Centers for Disease Control and Prevention (CDC). Detec-
London: The European Agency for the Evaluation of Medici-
tion of West Nile Virus in blood donations—United States, 2003. MMWR Morb Mortal Wkly Rep2003;52:769-72.
nal Products; 1996. 30. Goodrich RP, Custer B, Keil S, et al. Defining adequate patho-
17. Cardo LJ, Salata J, Mendez J, et al. Pathogen inactivation of Trypanosoma cruzi in plasma and platelet concentrates using riboflavin and ultraviolet light. Transfus Apher Sci 2007;37:131-7. 18. Tonnetti L, Proctor MC, Reddy HL, et al. Evaluation of the Mirasol pathogen [corrected] reduction technology system
gen reduction performance for transfused blood components. Transfusion 2010;50:1827-37. 31. Assal A, Barlet V, Deschaseaux M, et al. Comparison of the analytical and operational performance of two viral nucleic acid test blood screening systems: Procleix Tigris and cobas s 201. Transfusion 2009;49:289-300. Volume 55, July 2015 TRANSFUSION 1743
KEIL ET AL.
32. Keitel S. Guide to the preparation, use and quality assurance of blood components. 17th ed. Strasbourg: Council of Europe; 2013. p. 356-61. 33. Keil SD, Miklauz MM, Gilmour DI, et al. Photochemical inactivation of human hepatitis B virus using the Mirasol pathogen reduction technology (PRT) system as measured by PCR [poster abstract]. Vox Sang 2010;91516. 34. Owada T, Kaneko M, Matsumoto C, et al. Establishment of culture systems for Genotypes 3 and 4 hepatitis E virus
36. Tan LK, Lam S, Low SL, et al. Evaluation of pathogen reduction systems to inactivate dengue and chikungunya viruses in apheresis platelets suspended in plasma. Adv Infect Dis 2013;1-9. 37. Custer B, Agapova M, Martinez RH. The cost-effectiveness of pathogen reduction technology as assessed using a multiple risk reduction model. Transfusion 2010;54:2641-73. 38. Mundt JM, Rouse L, Van den Bossche J, et al. Chemical and biological mechanisms of pathogen reduction technologies. Photochem Photobiol 2014;90:957-64.
(HEV) obtained from human blood and application of HEV inactivation using a pathogen reduction technology system. Transfusion 2014;54:2820-7. 35. Vanlandingham DL, Keil SD, Horne KM, et al. Photochemical inactivation of chikungunya virus in plasma and platelets using the Mirasol pathogen reduction technology system. Transfusion 2013;53:284-90.
1744 TRANSFUSION Volume 55, July 2015
SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher’s Web site: Supplement S1. 48-well plate verification.