Preventing disease transmission by deceased tissue donors by testing blood for viral nucleic acid

Transcription

1 Cell and Tissue Banking (2005) 6: Ó Springer 2005 DOI /s Preventing disease transmission by deceased tissue donors by testing blood for viral nucleic acid D. Michael Strong 1, *, Karen Nelson 1, Marge Pierce 2 and Susan L. Stramer 3 1 Puget Sound Blood Center/Northwest Tissue Center, 921 Terry Avenue Seattle, WA 98104; 2 American Red Cross, St. Paul, MN; 3 American Red Cross, Gaithersburg, MD; *Author for correspondence ( dmstrong@psbc.org; phone: ; fax: ) Received 1 June 2005; accepted in revised form 15 August 2005 Key words: Deceased tissue donors, Infectious disease, NAT, Nucleic acid testing, Tissue transplantation Abstract Nucleic acid testing (NAT) has reduced the risk of transmitting infectious disease through blood transfusion. Currently NAT for HIV-1 and HCV are FDA licensed and performed by nearly all blood collection facilities, but HBV NAT is performed under an investigational study protocol. Residual risk estimates indicate that NAT could potentially reduce disease transmission through transplanted tissue. However, tissue donor samples obtained post-mortem have the potential to produce an invalid NAT result due to inhibition of amplification reactions by hemolysis and other factors. The studies reported here summarize the development of protocols to allow NAT of deceased donor samples with reduced rates of invalid results. Using these protocols, inventories from two tissue centers were tested with greater than 99% of samples producing a valid test result. Abbreviations: HBV hepatitis B virus; HCV hepatitis C virus; HIV human immunodeficiency virus; IC internal control; IND investigational new drug study protocol approved by the Food and Drug Administration; NAT Nucleic acid testing; PCR polymerase chain reaction; TMA transcriptionmediated amplification Introduction Testing of blood samples from deceased donors who may donate tissue has traditionally been carried out with immunoassays to detect viral antibodies or antigens. The assays that have been routinely employed have been those originally developed for the screening of potential blood donors. Viruses such as hepatitis B virus (HBV), hepatitis C virus (HCV) and human immunodeficiency virus type 1 (HIV) have all been transmitted by tissue transplantation (Conrad et al. 1995; Eastlund 1995; Homan et al. 2003; Eastlund and Strong 2004). These viruses, which have also been transmitted by blood transfusion, have been difficult to detect in asymptomatic donors during the so-called viremic window period, before antibodies are detectable (Busch et al. 1995; Dodd 2000). Although window period estimates have been extensively studied for blood donors (Schreiber et al. 1996; Dodd et al. 2000; Glynn et al. 2000), similar estimates have only recently been made for tissue donors (Stanworth et al. 2000; Zou et al. 2004). As a result, it has been determined that the risk of a viremic tissue donor being in the window period with a negative test for HIV or HCV

2 256 antibodies is substantially higher than that of blood donors. In 1999, new screening methods involving nucleic acid amplification to detect HIV and HCV RNA were implemented in the United States under an investigational new drug (IND) protocol approved by the Food and Drug Administration (FDA) (Stramer et al. 2000; Busch et al. 2000; Busch 2001). This new nucleic acid testing (NAT) protocol is used to test multiple samples in small pools, referred to as minipools. Recently the results of these trials have been published and have demonstrated an improved sensitivity resulting in the prevention of the transmission of approximately five HIV infections and 56 HCV infections through blood transfusion in the US annually, and has reduced the residual risk of transfusion-transmitted HIV and HCV to approximately one in two million blood units screened (Stramer et al. 2004). More recently, HBV NAT has also been shown to reduce the window period and thus potentially provide an increased margin of safety (Kleinman et al. 2005). As a result, the HIV, HCV, and HBV NAT assays have been licensed for screening of blood donations. With the demonstration of the sensitivity of these new assays in closing the window period for the well-known infectious viruses, and from the residual risk estimates for cadaveric donors, the implementation of NAT for screening of samples from deceased tissue donors should also contribute significantly to safety. Previous reports have documented problems with testing post-mortem samples using licensed blood screening assays (Novick et al. 1993; Burtonboy and Delloye 1996; Heim et al. 1999; Belcher et al. 2003). Therefore, studies were performed to determine the feasibility of performing NAT on samples from deceased donors including determining optimal methods for ensuring valid and accurate post-mortem sample test results. had been obtained as serum from tissue donors obtained post-mortem and stored frozen. In addition, samples were obtained from normal donors who had previously tested nonreactive in these assays. The study was designed to model the testing of frozen samples. Serum samples were thawed, and a portion spiked with a Multiprep positive control supplied by Roche to concentrations of 300 IU/ml of HIV, HCV and HBV. Aliquots of the spiked and unspiked samples were refrozen prior to testing. Some hemolyzed samples were selected for testing. The concentration of free hemoglobin in the test samples was estimated by comparison to free hemoglobin standards; standards and unknown samples were tested using the Sysmex XE 2100 analyzer (Sysmex America, Inc., Mundelein, IL). Testing was performed in triplicate in accordance with NAT manufacturer s instructions. Personnel performing the testing were blinded to the identity of the donors whose samples were being tested. As part of the qualification protocol, some samples were diluted immediately prior to testing with normal saline, plasma or with the diluent provided with the assays. Samples studied with the Procleix Ò HIV-1/HCV assay, discriminatory HBV (dhbv) assay (Gen- Probe Inc, San Diego, CA/Chiron Corporation, Emeryville, CA) included serum and plasma samples obtained from blood donors or frozen samples from tissue donors provided by the American Red Cross. Samples were spiked with tissue culture media or plasma containing 200 IU/ ml HIV-1 or HCV prior to testing. Some blood donation samples were hemolyzed mechanically and the concentration of hemoglobin determined by colorimetric assay (Sigma). As part of the qualification protocol, some samples were diluted immediately prior to testing with normal saline or plasma. Materials and methods Samples Samples studied with the COBAS Ampliscreen TM HIV-1, HCV and HBV (Roche Molecular Systems, Pleasanton, CA) assays were provided by the Northwest Tissue Center (NTC) or the Musculoskeletal Transplant Foundation (MTF). Samples Assays The COBAS Ampliscreen TM HIV-1 and HCV assays are based on reverse transcription of viral RNA followed by polymerase chain reaction (PCR) and hybridization. Each assay contains an internal control (IC) specific for the virus tested in order to detect inhibition of amplification and thus avoid a false-negative interpretation. In order for a

3 257 sample test result to be interpreted as valid, the IC must be reactive; samples with nonreactive IC signals are considered invalid and must be repeated. The HIV-1 and HCV assays are licensed for blood donation screening; the HBV assay has been used for blood donation screening under IND. The Procleix Ò HIV-1/HCV and dhbv assays are based on transcription-mediated amplification (TMA) of viral RNA and DNA. TMA also uses an IC to determine the validity of the amplification reaction (as stated above). Samples testing reactive by the FDA licensed, multiplex HIV-1/HCV assay are then tested by discriminatory HIV-1 and HCV assays to determine the specificity of the original test result. The HBV assay has been used for blood donation screening under IND. Results Inhibition of NAT assays by unknown factors in post-mortem samples When thawed samples from deceased donors were tested in the COBAS Ampliscreen TM or Procleix Ò HIV-1/HCV assays, IC failures were noted indicating inhibition of amplification by factors in the samples. When undiluted samples were tested, the extent of inhibition was greater in the PCRbased assays (Figure 1) vs. the TMA assays (Figure 2). Hemolyzed samples created by mechanical disruption of red cells and tested without dilution showed complete inhibition in the TMA assays (Figure 2). Effects of hemolysis, or time from asystole to sample collection, on assay performance In our analysis of the source of inhibitors in deceased donor samples, we evaluated whether hemolysis alone flagged samples at risk of invalid tests, or if there were additional inhibitors other than free hemoglobin that appeared with extended time between asystole and sample collection. We stratified samples collected at times post-asystole (less than 13 h) by the amount of hemolysis, and stratified grossly hemolyzed samples (greater than 5.6 mg/ml) by time of collection. Samples were spiked with virus as described in Materials and methods and tested in the COBAS Ampliscreen TM HIV-1, HCV assays. As shown in Tables 1 and 2, hemolysis is a marker of sample inhibition. Increased time post-asystole did not result in an increase in failed tests in samples with apparent gross hemolysis. Although this does not rule out the presence of other inhibitors, it indicates that hemolysis identifies, or is responsible for, potentially invalid samples. The effects of free hemoglobin were also studied in the Procleix Ò HIV-1/ HCV Assay and discriminatory assays. Figure 2 % valid tests Undiluted Diluted 1:5 HIV-1 HCV HBV Figure 1. Testing of hemolyzed samples with the Ampliscreen TM HIV-1, HCV and HBV assays. Hemolyzed serum samples: 43 undiluted 55 diluted 1:5. Samples were selected by assessment of hemolysis levels and collection of at least 8 h post-asystole. Data are presented as the percentage of tests with a valid IC result.

4 258 % valid tests Undiluted Diluted 1:5 Hemolysed sera HIV-1 Hemolysed sera HCV Cadaveric sera HIV-1 Cadaveric sera HCV Figure 2. Testing of hemolyzed samples with the ProcleixÒHIV-1/HCV Assay. Hemolyzed serum samples: six blood donation samples frozen as whole blood, thawed and assayed; results were obtained from discriminatory tests. Deceased donor serum samples: 50 pre- or post-mortem tested by the discriminatory assays. Approximately 50% of the samples were post-mortem and had some degree of hemolysis. Data are presented as the percentage of tests with a valid IC result. shows that mechanically hemolyzed samples spiked as described in Materials and methods gave uniformly invalid tests. Dilution of samples to avoid inhibition; effect on assay sensitivity Diluting the post-mortem samples with saline at a ratio of 1:5 reduced inhibition as noted by fewer IC failures (Figures 1 and 2). Diluted samples retained sensitivity in both assays (data not shown). Testing frozen samples from deceased donors In cases where an initial test was invalid, dilution of samples was used to obtain valid tests for frozen serum samples from deceased donors who had tissue in inventory at the Northwest Tissue Center or at the American Red Cross. The protocol derived for the COBAS assays involves dilution at 1:5; initial invalid test results are resolved by retesting diluted samples in triplicate. The protocol for the Procleix assays also includes initial testing that is of the undiluted sample; initial invalid Table 1. Effect of increased time post-asystole on samples with apparent hemolysis. Time post asystole (h) HIV Analyte pos (%) HIV IC pass (%) HCV Analyte pos (%) HCV IC pass (%) Samples were selected for hemolysis (free hemoglobin >5.6 mg/ml) and stratified on the basis of time from asystole until sample collection; samples were spiked with HIV and HCV at 300 IU/ml and tested undiluted in the COBAS Ampliscreen TM HIV-1, HCV assays. Table 2. Effect of hemoglobin concentration (Hb) on samples collected 9 13 h post asystole. Hb (mg/ml) HIV Analyte pos (%) HIV IC pass (%) HCV Analyte pos (%) HCV IC pass (%) > Samples were selected for time of collection post-asystole and stratified on the basis of hemoglobin concentration; samples were spiked with HIV and HCV at 300 IU/ml and tested undiluted in the COBAS Ampliscreen TM HIV-1, HCV assays.

5 259 results are resolved by performing the 1:5 dilution and retesting. Table 3 presents a summary of the data achieved with the two assays. Valid repeat tests were obtained with over 99% of samples tested. The HBV assays showed the highest rate of invalid tests. These data suggest that the presence of inhibitory substances in most samples can be reduced by dilution. Discussion Testing of donor blood for disease markers plays an important role in reducing the risk of disease transmission. As assays improve in sensitivity, the risk is further reduced by limiting the so-called window period, from the point of infection until detectability. The probability of collecting a sample during this window period has been extensively evaluated in blood donors (Schreiber et al. 1996; Dodd et al. 2000, 2002; Glynn et al. 2000). As new and more sensitive assays are implemented, donors, who may have early infections, will be detected. An example of such a case in the tissue donor population was published by Conrad et al. (1995). Banked sera from tissue donors that had previously been found to be nonreactive in the first generation anti-hcv enzyme immunoassay (EIA) and nonreactive for antibodies to hepatitis B core antigen were retested upon implementation of a new generation assay (second generation anti- HCV EIA). In that case, previously test-nonreactive serum samples from two donors were reactive in the new assay; follow-up testing of recipients from the tissue of those donors were determined to be infected with HCV. The tissue recipients along with the donor tested positive for HCV RNA by PCR. More recently, an organ and tissue donor whose serum samples tested nonreactive by the second generation anti-hcv EIA were determined to have transmitted HCV (Cieslak et al. 2003; Homan et al. 2003). Upon further analysis, stored, frozen serum obtained pre-mortem from the donor tested nonreactive by a more sensitive test for anti- HCV (third generation EIA), but the sample was positive for HCV RNA. These cases demonstrate that had nucleic acid amplification been available, and had these donors been tested, these cases might have been prevented. The quality of samples obtained from organ and tissue donors has been of some concern. Massive blood loss and intravascular volume replacement by transfusion of crystalloid solutions and blood can cause hemodilution and result in unreliable donor test results for infectious diseases (LeFor et al. 1995; Eastlund 2000). In 1987, a case of HIV transmission to multiple recipients of organs derived from an infected donor was reported (Centers for Disease Control 1987). The donor tested anti-hiv nonreactive when the sample was obtained immediately after the donor had received blood transfusions and large volumes of crystalloid solution. When the donor was tested 48 h later, the samples tested positive due to replenishment of immunoglobulin from extravascular sites. As a result, US federal regulations were published which set hemodilution limits and required calculations of dilution effects to ensure that testing is accurate and samples that are tested truly represent donor blood. A second complication is the quality of premortem or post-mortem blood samples. Standard organ and tissue donor collection practices range from a collection of pre-mortem plasma samples to post-mortem collection of serum samples up to 24 h following death. These post-mortem practices increase the chance of obtaining highly hemolyzed specimens that may compromise the sensitivity and specificity of assays. Novick and et al. (1993) demonstrated that there is indeed a significant difference Table 3. Testing of frozen samples from deceased tissue donors with tissue allografts in inventory. Assay Initial invalid rate (%) Repeat invalid rate (%) COBAS Ampliscreen TM HIV-1 (N = 218) COBAS Ampliscreen TM HCV (N = 220) COBAS Ampliscreen TM HBV (N = 229) Procleix Ò HIV-1/HCV (N = 1089) Procleix Ò dhbv (N = 472) If initial testing invalid, repeat testing performed after 1:5 dilution. The number of samples tested is also indicated.

6 260 in the performance of different manufacturers test kits when using grossly hemolyzed cadaveric serum samples vs. testing living donor blood samples. They found an increase in false-positive results (specifically for some test kits). In the case of detection of viral nucleic acids in blood samples from cadavers, the issue is a potential decrease in sensitivity due to the inhibition of reverse transcription or polymerase chain reaction steps by substances present in the samples (Padley et al. 2003). A simple 1:5 dilution step appears to overcome this inhibition without significantly compromising sensitivity. Therefore, the recommended protocols include the use of a dilution step if the initial undiluted sample yields an inhibited result. In the case of PCR, a 1:5 dilution is made from the start on all samples as suggested by Tobler et al. (2002). Although there was no significant differences in samples procured at different times to asystole, others have noted an increase in false-positive immunoassay results and a decrease in sensitivity with PCR in samples collected >24 h post-mortem (Burtonboy and Delloye 1996). As reported here, it was apparent that hemolysis can also interfere with the TMA assay, in addition to being an indicator of other interferences. However, the data summarized here indicate that dilution compensates for the presence of inhibitors and results in valid results in over 99% of samples from deceased donors. Sensitivity using a 1:5 dilution has been noted to be similar to that obtained with testing pooled samples of blood donors (Tobler et al. 2002). Studies comparing serum and plasma samples at various storage conditions have shown that room temperature storage results in a marked reduction in the concentrations of HCV RNA; in addition, repeated freeze-thaw cycles can cause a moderate reduction in viral levels (Busch et al. 1992). Standards of the American Association of Tissue Banks require that programs freeze a serum sample for archive purposes so that testing may be conducted on samples from donors who may have tissue in inventory should a new more sensitive test become available (American Association of Tissue Banks 2002). Therefore, studies were carried out to validate NAT protocols that could be used with stored cadaveric donor samples. Samples collected post-mortem will be de facto serum samples whether drawn into an anticoagulant or not. Both manufacturers recommend collection into a variety of anticoagulants; however, EDTA consistently gives the most reliable results. The serum samples from the previously reported transmission cases were frozen and thawed but still tested NAT-reactive. It is clear that serum samples can be used, even if plasma is preferred (Anderson et al. 2003). Viral nucleic acid deteriorates over time depending on storage conditions, which is why the manufacturers provide specific directions in their package inserts limiting storage at 4 C in whole blood to less than 72 h with additional storage allowed as plasma. Although there is a loss of sensitivity due to freezing and thawing or storage in serum compared to plasma, high viral copy numbers in most identified recently infected donors in the window period still allow nucleic acid detection (Galel et al. 2002; Stramer et al. 2004). Studies of the prevalence of viral confirmedpositive tissue donors have recently been published (Stanworth et al. 2000; Archibald et al. 2003; Zou et al. 2004). In these studies, the prevalence rates of HBV, HCV, HIV, and HTLV infections were lower among tissue donors than in the general population but considerably higher than among first-time blood donors. Despite the improvement in detection of window period donors using NAT, it is expected that infections may still be missed due to early infections and low viral copy numbers resulting in false-negative test results (Najioullah et al. 2004). Nevertheless, the addition of NAT to the screening of tissue donors should significantly reduce the risk of these infections among recipients of donated tissues (Gordon et al. 2002). Acknowledgements The authors thank Wendy Yadock, Carol Taylor, and Sandra Linauts for their technical support, and Laura Schreiber for editorial assistance from Puget Sound Blood Center. We would also like to thank Joyce Davis and her staff at Medical Marketing Consultants for clinical monitoring for the American Red Cross. References American Association of Tissue Banks Standards for Tissue Banking 10th edition. American Association of Tissue Banks, McLean, VA.

7 261 Anderson S.M., Gocke D., Belzer S.W., Gamble A., Campanella C. and Bankowski M.J Detection of Hepatitis C virus RNA in Cadaveric Serum or Plasma Samples using Nucleic acid Amplification. Presented at the 27th Annual Meeting, American Association of Tissue Banks, San Diego CA, August 23 26, 2003, S-1. Archibald L.K., Mills C.R., Mills A., Kainer M.A. and Roberts M.R Bloodborne viral infections among cadaveric tissue donors: implications for serological screening and allograft tissue safety. Presented at the 27th Annual Meeting, American Association of Tissue Banks, San Diego CA, August 23 26, 2003, S-3. Belcher L., Chandler L., Kasprisin D. and Gilcher R.O A case study: HIV positive donor with negative PCR. Presented at the 27th Annual Meeting, American Association of Tissue Banks, San Diego CA August 23 26, 2003, S-5. Burtonboy G. and Delloye C Polymerase chain reaction in cadaveric blood and tissue. Transplant Proc. 28: Busch M.P Closing the windows on viral transmission by blood transfusion. In: Stramer S.L. (ed.), Blood Safety in the New Millennium. AABB Press, Bethesda MDpp Busch M.P., Kleinman S.H., Jackson B., Stramer S.L., Hewlett I. and Preston S Committee report. Nucleic acid testing of blood donors for transfusion-transmitted infectious diseases: report of the Interorganizational Task Force on Nucleic Acid Amplification Testing of Blood Donors. Transfusion 40: Busch M.P., Lee L.L. and Satten G.A Time course of detection of viral and serologic markers preceding human immuno-deficiency virus type 1 seroconversion: implications for screening of blood and tissue donors. Transfusion 35: Busch M.P., Wilber J.C., Johnson P., Tobler L. and Evans C.S Impact of specimen handling and storage on detection of hepatitis C virus RNA. Transfusion 32: Centers for Disease Control Human immunodeficiency virus infection transmitted from an organ donor screened for HIV antibody: North Carolina. MMWR 36: Cieslak P.R., Hedberg K., Thomas A.R., Kohn M.A., Chai F., Nainan O.V., Williams I.T., Bell B.P., Tugwell B.D. and Patel P.R Hepatitis C virus transmission from an antibody-negative organ and tissue donor United States MMWR 52: Conrad E.U., Gretch D.R., Obermeyer K.R., Moogk M.S., Sayers M., Wilson J.J. and Strong D.M Transmission of the hepatitis-c virus by tissue transplantation. J. Bone Joint Surg. 77-A: Dodd R.Y Current viral risks of blood and blood products. Ann. Med. 32: Dodd R.Y., Notari E.P.IV and Stramer S.L Current prevalence and incidence of infectious disease markers and estimated window-period risk in the American Red Cross blood donor population. Transfusion 42: Dodd R.Y., Stramer S.L., Aberle-Grasse J. and Notari E.IV Risk of hepatitis and retroviral infections among blood donors and introduction of nucleic acid testing (NAT). Dev. Biol. Stand. 102: Eastlund T Infectious disease transmission through cell, tissue, and organ transplantation: reducing the risk through donor selection. Cell Transplant 4: Eastlund T Hemodilution due to blood loss and transfusion, and reliability of cadaver infectious disease testing. Cell Tissue Bank 1: Eastlund T. and Strong D.M Infectious Disease Transmission through Tissue Transplantation Advances in Tissue Banking, Vol. 7. Phillips GO, World Scientific, Singapore, pp Esslinger S., Mortensen J.E., Woll J. and Blair K A comparison of Abbott and BioRad Hepatitis B surface antigen (HBsAg) and HIV 1/2 immunoassays in a cadaveric donor population. Presented at the 27th Annual Meeting, American Association of Tissue Banks, San Diego CA. August 23 26, 2003, S-23. Galel S.A., Strong D.M., Tegtmeier G.E., Holland P.V., Kuramoto I.K., Kemper M., Pietrelli L. and Gallarda J Comparative yield of HCV RNA testing in blood donors screened by 2.0 versus 3.0 antibody assays. Transfusion 42: Glynn S.A., Kleinman S.H. and Schreiber G.B Trends in incidence and prevalence of major transfusion-transmissible viral infections in US blood donors, 1991 to 1996: The Retrovirus Epidemiology Donor Study (REDS). JAMA 284: Gordon S.M., Carey W., Yen-Lieberman B., Dennis V., Nurko S., Hoeltge G. and Goldfarb D Letter to the Editor: A fatal case of hepatitis C seroconversion following living related kidney transplantation: another argument for nucleic acid amplification testing of transplant recipients and donors. Transplant Infect Dis 4: 113. Heim A., Wagner D., Rothamel T., Hartmann U., Flik J. and Verhagen W Evaluation of serology screening of cadaveric sera for donor selection for cornea transplantation. J. Med. Virol. 58: Homan H., Gilbert D.N., Corless C., Kainer M., Kuhnert W., Jerningan D., Kiang K. and Lofy K Hepatitis C virus transmission from an antibody-negative organ and tissue donor United States, MMWR 52: Kleinman S., Strong D.M., Gorlin J., Holland P.V., Cousins C., Tegtmeier G., Chiacchierini R.P. and Pietrelli L HBV DNA screening of blood donations in minipools using the COBAS AmpliScreen TM HBV Test. Transfusion 45: LeFor W.M., Shires D.L.Jr, McGonigle A.F. and Shires D.L.III Hemoconcentration prior to serology testing in hemodiluted cadaver bone and tissue donors. Clin. Transplant. 9: Najioullah F., Barlet V., Renaudier P., Guitton C., Crova P., Guerin J.C., Peyramond D., Trabaud M.A., Coudurier N., Tardy J.C. and Andre P Failure and success of HIV tests for the prevention of HIV-1 transmission by blood and tissue donations. J. Med. Virol. 73: Novick S.L., Schrager J.A., Nelson J.A. and Baskin B.L A comparison of two HBsAg and two HIV-1 (p24) antigen EIA test kits with hemolyzed cadaveric blood specimens. Tissue Cell Rep. 1: 2 3. Padley D.J., Lucas S.B. and Saldanha J Elimination of false-negative hepatitis C virus RNA results by removal of inhibitors in cadaver-organ donor blood specimens. Transplantation 76:

8 262 Schreiber G.B., Busch M.P., Kleinman S.H. and Korelitz J.J The risk of transfusion-transmitted viral infections: the retrovirus epidemiology donor study. N Eng. J. Med. 334: Stanworth S.J., Warwick R.M., Ferguson M. and Barbara J.A A UK survey of virology testing of cadaver tissue donors. Vox Sang 79: Stramer S.L., Caglioti S. and Strong D.M NAT of the United States and Canadian blood supply. Transfusion 40: [Erratum, Transfusion (2001) 41: 1079] Stramer S.L., Glynn S.A., Kleinman S.H., Strong D.M., Caglioti S., Wright D.J., Dodd R.Y. and Busch M.P Detection of HIV-1 and HCV infections among antibodynegative blood donors by nucleic acid amplification testing. N Eng. J. Med. 351: Tobler L.H., Dockter J., Stramer S.L., Kleinman S.H., Todd D., Giachetti C. and Busch M.P Use of Procleix HIV-1 and HCV discriminatory assays to detect HIV and HCV RNA in whole blood. Transfusion 42: Zou S., Dodd R.Y., Stramer S.L. and Strong D.M Probability of viremia with HBV, HCV, HIV, and HTLV among tissue donors in the United States. N Eng. J. Med. 351: