Multiplex Screening for Blood-Borne Viral, Bacterial, and Protozoan Parasites using an OpenArray Platform
2013; Elsevier BV; Volume: 16; Issue: 1 Linguagem: Inglês
10.1016/j.jmoldx.2013.08.002
ISSN1943-7811
AutoresElena L. Grigorenko, Carolyn Fisher, Sunali Patel, Caren Chancey, María Rios, Hira L. Nakhasi, Robert Duncan,
Tópico(s)Bacterial Identification and Susceptibility Testing
ResumoThe use of nucleic acid tests for detection of pathogens has improved the safety of blood products. However, ongoing pathogen emergence demonstrates a need for development of devices testing for multiple pathogens simultaneously. One approach combines two proven technologies: Taqman chemistry for target identification and quantification and the OpenArray nanofluidic real-time PCR platform for spatial multiplexing of assays. A panel of Taqman assays was developed to detect nine blood-borne pathogens (BBPs): four viral, two bacterial, and three protozoan parasites. The custom BBP OpenArray plate with 18 assays was tested for specificity and analytical sensitivity for nucleic acid from each purified pathogen and with pathogen-spiked human blood and plasma samples. For most targets, the limits of detection (10 to 10,000 copies/mL) were comparable with existing real-time platforms. The testing of the BBP OpenArray with pathogen-spiked coded human plasma or blood samples and negative control specimens demonstrated no false-positive results among the samples tested and correctly identified pathogens with the lowest concentration detected ranging from 10 cells/mL (Trypanosoma cruzi) to 10,000 cells/mL (Escherichia coli). These results represent a proof of concept that indicated the BBP OpenArray platform in combination with Taqman chemistry may provide a multiplex real-time PCR pathogen detection method that points the way for a next-generation platform for infectious disease testing in blood. The use of nucleic acid tests for detection of pathogens has improved the safety of blood products. However, ongoing pathogen emergence demonstrates a need for development of devices testing for multiple pathogens simultaneously. One approach combines two proven technologies: Taqman chemistry for target identification and quantification and the OpenArray nanofluidic real-time PCR platform for spatial multiplexing of assays. A panel of Taqman assays was developed to detect nine blood-borne pathogens (BBPs): four viral, two bacterial, and three protozoan parasites. The custom BBP OpenArray plate with 18 assays was tested for specificity and analytical sensitivity for nucleic acid from each purified pathogen and with pathogen-spiked human blood and plasma samples. For most targets, the limits of detection (10 to 10,000 copies/mL) were comparable with existing real-time platforms. The testing of the BBP OpenArray with pathogen-spiked coded human plasma or blood samples and negative control specimens demonstrated no false-positive results among the samples tested and correctly identified pathogens with the lowest concentration detected ranging from 10 cells/mL (Trypanosoma cruzi) to 10,000 cells/mL (Escherichia coli). These results represent a proof of concept that indicated the BBP OpenArray platform in combination with Taqman chemistry may provide a multiplex real-time PCR pathogen detection method that points the way for a next-generation platform for infectious disease testing in blood. Testing of donations for blood-borne pathogens (BPPs) has reduced significantly the risk of transfusion transmission of such agents. However, the current number of pathogen tests and the increasing number of emerging pathogens that can affect blood safety results in an increased number of screening tests required before release of the blood and blood products. The use of a single pathogen assay for each agent is burdensome and costly and requires volumes of specimens that risk exceeding the maximum collection volume allowed for testing. Devices that allow simultaneous testing for multiple pathogens (multiplex testing) can potentially streamline blood donation testing. To date, the highest number of pathogens that can be assayed in one multiplex test licensed for blood screening is five [HIV-1 group M and group O, HIV-2, hepatitis C virus (HCV), and hepatitis B virus (HBV)].1Margaritis A.R. Brown S.M. Seed C.R. Kiely P. D'Agostino B. Keller A.J. Comparison of two automated nucleic acid testing systems for simultaneous detection of human immunodeficiency virus and hepatitis C virus RNA and hepatitis B virus DNA.Transfusion. 2007; 47: 1783-1793Crossref PubMed Scopus (59) Google Scholar Current multiplex tests require subsequent discriminatory assays to identify which pathogen produced the reactive test result. An improved technology using real-time nucleic acid tests (NATs) that promise to deliver discrimination simultaneously with detection is available in Europe but not the United States (TaqScreen MPX Test, version 2.0; Roche, Roche Molecular Diagnostics, Pleasanton, CA) and other devices that combine amplification with microarray hybridization have also been proposed.2Tomioka K. Peredelchuk M. Zhu X. Arena R. Volokhov D. Selvapandiyan A. Stabler K. Mellquist-Riemenschneider J. Chizhikov V. Kaplan G. Nakhasi H. Duncan R. A multiplex polymerase chain reaction microarray assay to detect bioterror pathogens in blood.J Mol Diagn. 2005; 7: 486-494Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar New technology needs to be developed that can achieve in a single assay both higher levels of multiplicity and simultaneous discrimination of the reactive agent. Application of novel and emerging technologies offers the prospect to develop the next generation of superior performance tests that will detect an increasing number of transfusion-transmittable agents in multiplex format with high sensitivity and specificity, robustness, and adaptability to accomplish detection of new agents. To date, NATs have only been applied to viral pathogens, namely, HIV, HCV, HBV, and West Nile virus (WNV) using plasma specimens. Testing of plasma specimens would be substantially improved if all of these viral NAT assays could be performed on a single device with simultaneous detection and discrimination. Further, no Food and Drug Administration–approved NAT is being performed for blood donor screening of whole blood specimens. Nucleic acid detection of bacteria and protozoan parasites, such as Plasmodium, Trypanosoma cruzi, and Leishmania, that would be associated with whole blood specimens would make a significant contribution to blood safety if available on a multiplex assay of sufficient sensitivity. A multiplex approach can be achieved by using assays with different reporters combined in the same reaction well or by using high densities of physically independent PCR wells. However, both approaches have challenges. The presence of multiple primers may lead to cross-hybridization with each other and the possibility of mispriming with other templates, so optimization is required for any new primer sets in multiplex PCR. As for the other approach, stringent fluidic isolation between adjacent PCR reaction wells is necessary to prevent cross-contamination during loading and temperature cycling on any PCR platform with a high density of independent wells. Recent advancements in high-throughput real-time PCR technologies, such as the OpenArray platform (Life Technologies Corp., Carlsbad, CA), address these disadvantages by assembling selected primer and probe sets in spatially isolated reaction wells and combine the sensitivity and specificity of quantitative PCR with the high-throughput performance of microarrays to achieve large-scale screening of pathogens.3Gonzales T.K. Kulow M. Park D.J. Kaspar C.W. Anklam K.S. Pertzborn K.M. Kerrish K.D. Ivanek R. Dopfer D. A high-throughput open-array qPCR gene panel to identify, virulotype, and subtype O157 and non-O157 enterohemorrhagic Escherichia coli.Mol Cell Probes. 2011; 25: 222-230Crossref PubMed Scopus (23) Google Scholar, 4Morrison T. Hurley J. Garcia J. Yoder K. Katz A. Roberts D. Cho J. Kanigan T. Ilyin S.E. Horowitz D. Dixon J.M. Brenan C.J. Nanoliter high throughput quantitative PCR.Nucleic Acids Res. 2006; 34: e123Crossref PubMed Scopus (176) Google Scholar, 5Slinger R. Moldovan I. Barrowman N. Chan F. Successful nanolitre real-time PCR detection of respiratory pathogens in clinical specimens.Clin Microbiol Infect. 2012; 18: E286-E288Crossref PubMed Scopus (3) Google Scholar In this study, we evaluated a custom BBP OpenArray panel composed of primer and probe sets developed by scientists from both our institutions and published sources.6Drexler J.F. Kupfer B. Petersen N. Grotto R.M. Rodrigues S.M. Grywna K. Panning M. Annan A. Silva G.F. Douglas J. Koay E.S. Smuts H. Netto E.M. Simmonds P. Pardini M.I. Roth W.K. Drosten C. A novel diagnostic target in the hepatitis C virus genome.PLoS Med. 2009; 6: e31Crossref PubMed Scopus (47) Google Scholar, 7Nicolas L. Prina E. Lang T. Milon G. Real-time PCR for detection and quantitation of leishmania in mouse tissues.J Clin Microbiol. 2002; 40: 1666-1669Crossref PubMed Scopus (172) Google Scholar, 8Piron M. Fisa R. Casamitjana N. Lopez-Chejade P. Puig L. Verges M. Gascon J. Gomez i Prat J. Portus M. Sauleda S. Development of a real-time PCR assay for Trypanosoma cruzi detection in blood samples.Acta Trop. 2007; 103: 195-200Crossref PubMed Scopus (281) Google Scholar We determined its limit of detection for four viral pathogens in human plasma specimens simultaneously and five bacterial and protozoan pathogens in human blood specimens simultaneously. We also evaluated the custom BBP OpenArray's ability to correctly identify coded specimens at the limit of detection. HIV, HCV, and HBV were prepared from Center for Biologics Evaluation and Research (CBER) lot release panel members that were composed of human plasma specimens from infected individuals or infected cell culture, diluted with pooled human plasma defibrinated and delipidated, and tested negative for common BBPs (base matrix) to 500 or 100 copies/mL, inactivated, and stored in 1-mL vials at −80°C. The viral load of the frozen vials was extensively validated, as previously described for HIV9Davis C. Heath A. Best S. Hewlett I. Lelie N. Schuurman R. Holmes H. Calibration of HIV-1 working reagents for nucleic acid amplification techniques against the 1st international standard for HIV-1 RNA.J Virol Methods. 2003; 107: 37-44Crossref PubMed Scopus (29) Google Scholar, 10Lee S. Wood O. Taffs R.E. Hu J. Machuca A. Vallejo A. Hewlett I. Development and evaluation of HIV-1 subtype RNA panels for the standardization of HIV-1 NAT assays.J Virol Methods. 2006; 137: 287-291Crossref PubMed Scopus (12) Google Scholar and HCV.11Saldanha J. Heath A. Lelie N. Pisani G. Nubling M. Yu M. Calibration of HCV working reagents for NAT assays against the HCV international standard: The Collaborative Study Group.Vox Sang. 2000; 78: 217-224Crossref PubMed Scopus (67) Google Scholar The HBV panel members were derived from source material that was a specimen obtained from a blood donor in the United States [HBV DNA, genotype A, serotype adw2; Food and Drug Administration (FDA) identification No. 723534]. The panel members were diluted in base matrix and tested to contain 100 and 500 DNA copies/mL (Steve Kerby, FDA, unpublished data). The lot release panel members used were obtained directly from the Product Testing Section, Division of Emerging and Transfusion Transmitted Diseases, CBER, FDA. WNV was cultured on Vero cells,12Grinev A. Daniel S. Laassri M. Chumakov K. Chizhikov V. Rios M. Microarray-based assay for the detection of genetic variations of structural genes of West Nile virus.J Virol Methods. 2008; 154: 27-40Crossref PubMed Scopus (10) Google Scholar and the supernatant was collected and viral RNA quantified by both plaque-forming units per milliliter determined by plaque assay13Rios M. Zhang M.J. Grinev A. Srinivasan K. Daniel S. Wood O. Hewlett I.K. Dayton A.I. Monocytes-macrophages are a potential target in human infection with West Nile virus through blood transfusion.Transfusion. 2006; 46: 659-667Crossref PubMed Scopus (57) Google Scholar and copies per milliliter in limiting dilution and a TaqMan assay.14Rios M. Daniel S. Chancey C. Hewlett I.K. Stramer S.L. West Nile virus adheres to human red blood cells in whole blood.Clin Infect Dis. 2007; 45: 181-186Crossref PubMed Scopus (59) Google Scholar, 15Rios M. Daniel S. Dayton A.I. Wood O. Hewlett I.K. Epstein J.S. Caglioti S. Stramer S.L. In vitro evaluation of the protective role of human antibodies to West Nile virus (WNV) produced during natural WNV infection.J Infect Dis. 2008; 198: 1300-1308Crossref PubMed Scopus (14) Google Scholar Two bacterial species, Escherichia coli and Yersinia enterocolitica, were cultured in Super Broth (Quality Biological Inc., Gaithersburg, MD) and quantified by measuring optical density at 600 nm and converting to cells per milliliter by the formula 5.0 × 108 cells/mL = 1 absorbance unit with the SmartSpec 3000 spectrophotometer (BioRad Laboratories Inc., Hercules, CA). Leishmania donovani 1S2D promastigote form was cultured as previously described16Debrabant A. Joshi M.B. Pimenta P.F.P. Dwyer D.M. Generation of Leishmania donovani axenic amastigotes: their growth and biological characteristics.Intl J Parasitol. 2004; 34: 205-217Crossref PubMed Scopus (185) Google Scholar and quantified with a Beckman/Coulter particle counter. T. cruzi Columbiana strain parasites were cultured as epimastigotes as previously described17McDaniel J.P. Dvorak J.A. Identification, isolation, and characterization of naturally-occurring Trypanosoma cruzi variants.Mol Biochem Parasitol. 1993; 57: 213-222Crossref PubMed Scopus (50) Google Scholar and quantified with a Beckman/Coulter particle counter. Plasmodium falciparum was cultured as previously described18Mahajan B. Zheng H. Pham P.T. Sedegah M.Y. Majam V.F. Akolkar N. Rios M. Ankrah I. Madjitey P. Amoah G. Addison E. Quakyi I.A. Kumar S. Polymerase chain reaction-based tests for pan-species and species-specific detection of human Plasmodium parasites.Transfusion. 2012; 52: 1949-1956Crossref PubMed Scopus (30) Google Scholar and quantified by microscopic determination of the percentage of red blood cells infected and determination of the number of red blood cells per milliliter with a Beckman/Coulter particle counter, then the Plasmodium cells were calculated by the product of these two numbers. For HIV, HCV, and HBV, the 1-mL aliquoted lot release panel members were used directly as test specimens without further dilution. WNV-spiked specimens were prepared (under biosafety level 3 containment) from live WNV cell culture stock at a concentration of 1010 copies/mL as determined by various methods in collaborative studies. An intermediate dilution of 1:1000 containing 107 copies/mL of WNV was prepared in PBS, followed by a second intermediate dilution of 1:100 in human plasma to have 105 copies/mL, which was further subjected to a 10-fold serial dilution in 1-mL aliquots of human plasma to prepare the testing specimens that contained 104, 103, and 102 copies/mL. E. coli, Y. enterocolitica, L. donovani, T. cruzi, and P. falciparum were similarly diluted and spiked into whole heparinized human blood. The blood and plasma were obtained from deidentified donors to the research program at the Department of Transfusion Medicine, National Institutes of Health, under an approved protocol (principal investigator R.C.D.). Nucleic acid from 140 μL of WNV-spiked plasma was extracted using the QIAamp Viral Mini Kit (Qiagen, Valencia, CA) according to the manufacturer protocol and RNA eluted twice with 100 μL of elution buffer each elution into the same tube. Nucleic acid from 1 mL of all other plasma specimens was extracted with the QIAamp MinElute Virus Spin Kit (Qiagen) according to the manufacturer protocol and eluted in 180 μL of elution buffer. DNA from 1-mL whole blood specimens was extracted with the QIAamp DNA Blood Mini Kit (Qiagen) according to the manufacturer protocol and eluted in 180 μL of elution buffer. The panel for detection of BBPs consisted of primers and Taqman probes for viral, bacterial, and parasitic targets (Tables 1 and 2) designed for this study, published in the literature or commercially available from Life Technologies Corp. A proprietary algorithm was applied for a subset of Taqman assays designed by Life Technologies Corp. This algorithm evaluates a set of optimal assays, considering criteria of melting temperature and nucleotide composition of primer-pair combinations. The algorithm selects the assays with highest specificity, based on nucleic acid sequence comparison of assay primers and probes with genomic sequences from other closely related species. Based on sequence comparison, the assay with the highest mismatch score with other organisms was chosen, which minimizes the possibility of generation of false-positive results during the testing. Primers and probes designed by the FDA laboratory as part of this study were selected from genomic sequence using MacVector software version 11.1.2 (MacVector, Inc., Cary, NC), avoiding hairpin loops, primer duplexes, and optimizing melting temperatures.Table 1Assays for Plasma and Whole BBPs Found in Public Sources or Available on RequestSpeciesTarget geneAssay nameAmpliconReferenceAccession No.∗Accession numbers are for GenBank (http://www.ncbi.nlm.nih.gov/genbank).CoordinatesSize (bp)HIVGAGHIV-1AJN417241.1684-797113I. Hewlett, FDA, unpublished dataHIVGAGHIV-1BJN417241.1677-75578I. Hewlett, FDA, unpublished dataHIVGAGHIV-1CJX973372724-79470R.C.D., FDA, unpublished dataHCV3'UTR X-tailHCV-AEU835523.13-56536Drexler J.F. Kupfer B. Petersen N. Grotto R.M. Rodrigues S.M. Grywna K. Panning M. Annan A. Silva G.F. Douglas J. Koay E.S. Smuts H. Netto E.M. Simmonds P. Pardini M.I. Roth W.K. Drosten C. A novel diagnostic target in the hepatitis C virus genome.PLoS Med. 2009; 6: e31Crossref PubMed Scopus (47) Google ScholarLeishmaniaMinicircleLEI-1AB678348.110-1251157Nicolas L. Prina E. Lang T. Milon G. Real-time PCR for detection and quantitation of leishmania in mouse tissues.J Clin Microbiol. 2002; 40: 1666-1669Crossref PubMed Scopus (172) Google ScholarLeishmania18S rRNALEI-2FR799614.11020493- 1020707214R.C.D., FDA, unpublished dataT. cruziMinisatellitesTCF-1AY520036†Two assays only differ by probe sequence.26-1901648Piron M. Fisa R. Casamitjana N. Lopez-Chejade P. Puig L. Verges M. Gascon J. Gomez i Prat J. Portus M. Sauleda S. Development of a real-time PCR assay for Trypanosoma cruzi detection in blood samples.Acta Trop. 2007; 103: 195-200Crossref PubMed Scopus (281) Google ScholarT. cruziMinisatellitesTCF-2AY520036†Two assays only differ by probe sequence.26-187161R.C.D., FDA, unpublished databp, base pair.∗ Accession numbers are for GenBank (http://www.ncbi.nlm.nih.gov/genbank).† Two assays only differ by probe sequence. Open table in a new tab Table 2Commercial Taqman Assays Available from Life Technologies Corp.Assay nameLT assay No.OrganismGeneTarget accession No.HCV-BPa03453408_s1HCV5′UTRAF009606HBV-APa03453405_s1HBVP, SX04615HBV-BPa03453406_s1HBVP, XX04615WNV-APa04329496_s1WNV3′UTRAF202541.1WNV-BPa04329497_s1WNV3′UTRAF533540.1PLA-1Pa04329498_s1Plasmodium speciesMitochondrial rRNAGQ355486.1PLA-2Pa04329499_s1Plasmodium speciesMitochondrial rRNAAY791633.1GNEG-1Pa04329500_s1Gram-negative bacteriagapAX02662.1GNEG-2Pa04329501_s1Gram-negative bacteriagapAX02662.1 Open table in a new tab bp, base pair. Nucleic acid solution (10 μL) extracted from plasma samples was reverse transcribed using the High Capacity Reverse Transcription Kit (Life Technologies Corp., P/N 4368814) with random hexamer primers or gene-specific primers (pool of all assay primers diluted 1:5 relative to the concentration in the PCR reaction), in a total volume of 20 μL, incubated at 37°C for 2 hours per manufacturer protocol. To increase sensitivity for targets present at low concentration, target-specific preamplification was performed. A 10-microliter sample of cDNA was mixed with 10 μL of a mixed pool of PCR assay oligonucleotides at 180 nmol/L-50 nmol/L primer pair-probe concentration and 20 μL of Taqman preamplification master mix (Life Technologies Corp., P/N 4384267). The preamplification reaction was performed on an ABI9700 at the following cycling conditions: single cycle (95°C for 10 minutes), 18 cycles (95°C for 15 seconds and 60°C for 1 minute), and one cycle (95°C for 5 minutes). Preamplification products were diluted 1:10 in 0.1× TE buffer and were used on the BBP OpenArray platform. The OpenArray technology is based on a metal plate the size of a microscope slide that has been photolithographically patterned and etched to form a rectilinear array of 3072 through-holes, organized in 48 subarrays with 64 through-holes each. Each through-hole is loaded with individual Taqman assays and contains 33 nL of PCR mixture. Previous work4Morrison T. Hurley J. Garcia J. Yoder K. Katz A. Roberts D. Cho J. Kanigan T. Ilyin S.E. Horowitz D. Dixon J.M. Brenan C.J. Nanoliter high throughput quantitative PCR.Nucleic Acids Res. 2006; 34: e123Crossref PubMed Scopus (176) Google Scholar has shown that the PCR assay performance in the nanoplates is equivalent to the same assay in microplates but with a >150-fold lower reaction volume (33-nL versus 5-μL PCR reaction volumes) and with the ability to profile multiple targets using the same sample. cDNA and genomic DNA samples were tested on custom BBP OpenArray plates containing 18 Taqman assays spotted in triplicate. Genomic DNA or cDNA (1.2 μL of each) was mixed with 1.3 μL of PCR-grade water and 2.5 μL of Taqman OpenArray real-time PCR master mix (Life Technologies Corp., P/N 4462159). Then, the reaction mixture was dispensed on a BBP OpenArray plate using the automated sample loading system Accufill (Life Technologies, Corp.). Real-time PCR occurred in an OpenArray custom computer-controlled imaging thermal cycler under software control, where imaging data for up to 9216 individual reaction wells (three plates) were collected during 40 cycles of PCR. The postacquisition step includes calculation of quantification cycle (Cq) and the CT confidence parameter, which were used for further data analysis and in the decision tree for pathogen identification. The CT confidence parameter is calculated using Life Technologies Corp.'s proprietary algorithm, and its value reflects the quality of the amplification curve generated during PCR. The CT confidence values >300 indicate good quality of amplification. The CT confidence values of 200 through 300 are considered marginal, and each amplification curve with such CT confidence values has to be manually checked before the result can be considered valid. Reactions with CT confidence values <200 are not considered valid. The blind testing of coded specimens was ensured by nucleic acid extraction of the spiked specimens in the FDA laboratory, aliquoting the clear liquid into identical tubes that were labeled with a code. Coded specimens were shipped to the Life Technology laboratory, where they were tested with the BBP OpenArray and the result interpreted. The code was broken only after the final interpretation had been recorded. The Cq data in Table 3 represent repeated measures; thus, after testing for normality of distribution by the Rankits Correlation Method,19Sokal R. Rohlf F. Biometry: The Principles and Practice of Statistics in Biological Research.ed 2. W. H. Freeman and Company, New York1981Google Scholar the means and 95% CIs were calculated by the statistical functions in Microsoft Office Excel 2003 (Microsoft Inc., Redmond, WA). The overall correct and incorrect percentages given in Table 4 are best represented by the binomial distribution, so the 95% CIs were calculated by the Wilson score interval.20Wilson E.B. Probable inference, the law of succession, and statistical inference.J Am Stat Assoc. 1927; 22: 209-212Crossref Scopus (2496) Google Scholar In Table 5, which gives the Cq value data, the same statistical approach was used as in Table 3, and Table 6 data were analyzed as in Table 4.Table 3BBP OpenArray Results With Virus-Spiked Plasma SpecimensPathogen (copies/mL)Assay 1Assay 2Assay 3NameMean Cq (95% CI)No. of positive/total through-holesNameMean Cq (95% CI)No. of positive/total through-holesNameMean Cq (95% CI)No. of positive/total through-holesHIV (500)HIV-1A14.3 (±0.1)24/24HIV-1B17.4 (±0.1)12/24HIV-1C14.2 (±0.2)5/24HCV (500)HCV-ANA0/24HCV-B19.2 (±0.04)24/24HBV (500)HBV-A18.5 (±0.05)24/24HBV-B19.4 (±0.04)24/24WNV (100)WNV-A21.3 (±0.06)24/24WNV-BNA0/24WNV (1000)WNV-A19.2 (±0.05)24/24WNV-B25.8 (±0.27)24/24WNV (10,000)WNV-A14.7 (±0.05)24/24WNV-B20.0 (±0.07)24/24NA, not applicable. Open table in a new tab Table 4Detection of Unknown Specimens Plasma Spiked with Virus or Control PlasmaCoded specimen type (copies/mL)No. of specimensLOD (copies/mL)TotalTPTNFPFNNegative plasma control44NAHBV (100)211HBV (500)33500HCV (100)211HCV (500)33500HIV-1 (100)22HIV-1 (500)431500WNV (10)1110WNV (100)33WNV (1000)11Overall total2516405No. (%) overall correct/incorrect∗The sum of the true positives and true negatives is the number overall correct, the sum of the false positives and false negatives is the number overall incorrect. [95% CI]2520 (80) [60.9–91.1]5 (20) [8.9–39.1]FN, false negative; FP, false positive; LOD, limit of detection; NA, not applicable; TN, true negative; TP, true positive.∗ The sum of the true positives and true negatives is the number overall correct, the sum of the false positives and false negatives is the number overall incorrect. Open table in a new tab Table 5BBP OpenArray Results with DNA Pathogen–Spiked Whole Blood SpecimensPathogen (cells/mL)Assay nameMean Cq (95% CI)No. of positive/total through-holes∗Total number of through-holes observed among three experiments.Assay NameMean Cq (95% CI)No. of positive/total through-holes∗Total number of through-holes observed among three experiments.E. coli (10,000)†Lowest concentration detected.GNEG-127.5 (±0.3)22/30GNEG-227.6 (±0.4)22/30Y. enterocolitica (10,000)†Lowest concentration detected.GNEG-128.0 (±0.6)24/30GNEG-227.2 (±0.3)28/30Leishmania (10,000)†Lowest concentration detected.LEI-124.0 (±0.5)30/30LEI-226.6 (±0.5)30/30Leishmania (1000)LEI-10 (NA)LEI-20 (NA)Leishmania (100)LEI-128.8 (±1.3)4/24LEI-230.6 (NA)2/24Plasmodium (10,000)†Lowest concentration detected.PLA-132.4 (±0.7)29/30PLA-233.0 (±0.6)28/30Plasmodium (1000)PLA-133.5 (±3.3)3/24PLA-234.5 (±0.8)6/24Plasmodium (100)PLA-1NR (NA)PLA-2NR (NA)T. cruzi (10,000)TCF-123.0 (±0.2)30/30TCF-222.4 (±0.2)30/30T. cruzi (1000)†Lowest concentration detected.TCF-126.6 (±0.7)23/24TCF-225.7 (±0.5)23/24T. cruzi (100)TCF-129.8 (±1.5)9/24TCF-229.2 (±2.0)7/24NA, not applicable; NR, not replicated.∗ Total number of through-holes observed among three experiments.† Lowest concentration detected. Open table in a new tab Table 6Summary of Testing-Coded Specimens of Pathogen-Spiked Whole BloodSample type (cells/mL)No. of specimensLeast detected (cells/mL)TotalTPTNFPFNNegative blood control44E. coli (10,000)2210,000Y. enterocolitica (10,000)11Y. enterocolitica (1000)111000L. donovani (10,000)22L. donovani (1000)111000L. donovani (100)211P. falciparum (10,000)22P. falciparum (1000)22P. falciparum (100)11100T. cruzi (10,000)11T. cruzi (1000)11T. cruzi (100)22T. cruzi (10)1110Overall total231841No. (%) overall correct/incorrect [95% CI]2322 (96) [79.5–99.3]1 (4) [0.7–20.5]FN, false negative; FP, false positive; TN, true negative; TP, true positive. Open table in a new tab NA, not applicable. FN, false negative; FP, false positive; LOD, limit of detection; NA, not applicable; TN, true negative; TP, true positive. NA, not applicable; NR, not replicated. FN, false negative; FP, false positive; TN, true negative; TP, true positive. A custom BBP OpenArray was assembled with 18 primer and probe sets (designated assays for simplicity) with two assays for each target pathogen except HIV, which had three assays. Each subarray of 64 through-holes contained the 18 assays in triplicate. In each experiment, specimens were loaded in multiple subarrays to achieve a high level of replicates for each assay. Recognizing that most viruses can be detected in plasma and many blood-borne viruses have RNA genomes, we developed a protocol in which each blood specimen collected would be divided into a plasma sample and a whole blood sample. The plasma samples were extracted with the QIAamp MinElute Virus Spin Kit (Qiagen) such that 1 mL of plasma was eluted as approximately 180 μL of nucleic acid solution. Using virus-spiked human plasma and CBER reference preparations, a procedure was optimized in which each nucleic acid sample was reverse transcribed with random primers or a pool of gene-specific primers, then preamplified for 18 cycles with a pool of all 18 primer pairs. A 1:10 dilution of this preamplification mix was used as template in the BBP OpenArray. Two criteria were used for assessment of a standard experiment with BBP OpenArray plates: counting the number of through-holes with amplification detected for each reactive assay of the total number of through-holes used for that assay and the mean Cq values for the reactive through-holes. A series of experiments that led to the determination of the limits of detection for each virus from spiked samples are given in Table 3, where the results are reported for the three assays for HIV and two assays for the other viruses. This set of experiments suggested that on a BBP OpenArray plate, HIV and HBV could be detected below 500 copies/mL, HCV was not consistently detected at 500 copies/mL, and WNV could be detected consistently at 1000 copies/mL. This is an unexpectedly high concentration for the limit of detection for WNV and will be discussed further below. Unspiked control specimens (pathogen-free plasma) were also tested on the same OpenArray plates and s
Referência(s)