Quantitative Real-Time Polymerase Chain Reaction for Evaluating DNAemia due to Cytomegalovirus, Epstein-Barr Virus, and BK Virus in Solid-Organ Transplant Recipients

Thomas F. Smith 0 Mark J. Espy 0 Jayawant Mandrekar 0 Mary F. Jones 0 Franklin R. Cockerill 0 Robin Patel 0 Divisions of 0 Clinical Microbiology 0 Biostatistics 0 Infectious Diseases 0 Mayo Clinic 0 Foundation 0 Rochester 0 Minnesota 0 0 Received 28 March 2007; accepted 29 May 2007; electronically published 12 September 2007. and Foundation , 200 1st St. SW, Rochester, MN 55905 Testing for cytomegalovirus-, Epstein-Barr virus-, and BK virus-specific gene targets in specimens from solid-organ transplant recipients for DNA by quantitative real-time polymerase chain reaction has been implemented in many diagnostic facilities. This technology provides rapid, accurate, and reproducible results for early detection, monitoring, and medical management of patients with these infections. Because these assays are becoming commonly used in clinical practice, the technical variables associated with specimen processing (e.g., nucleic acid extraction, gene target, and result reporting), amplification, and unique patient characteristics (e.g., age, sex, underlying diseases, immune status, and immunosuppressive regimens received) are factors that may influence the understanding and interpretation of test results. We emphasize the need for standardization of existing variables through parallel comparative and proficiency testing, uniform units for expressing results, to provide for clinical correlation with the results of these molecular assays. - jective interpretation of the test results [57]. Importantly, both shell vial cell cultures (qualitative) and antigenemia tests (quantitative) have been shown in comparative studies to be less sensitive than PCR. In addition, DNAemia was detected in an earlier stage of nucleic acid amplification [610]. As an alternative to cell culture and the antigenemia test, implementation of molecular techniques for the rapid and sensitive detection of nucleic acid targets has yielded new levels of capabilities for laboratory diagnosis of many microbial infections [11]. Implementation of these tests has been facilitated by the availability of real-time PCR instrumentation with automation of nucleic acidtarget amplification and amplicondetection steps in a closed system (i.e., reaction tubes never opened during or after amplification). Therefore, for both qualitative and quantitative detection of viruses, automated realtime PCR has generally replaced the time-consuming and laborintensive conventional amplification and detection of products by gel electrophoresis and Southern blot or ELISA methods that frequently cause contamination events and false-positive results because of the inadvertent transfer of high-copy nucleic acid products to other specimens. REAL-TIME PCR The first applications of real-time PCR (7 years ago at The Mayo Clinic; Rochester, MN) were the routine, sensitive, and specific assays for herpes simplex virus and varicella-zoster virus as replacements for cell-culture methods; the menu of assays has now been extensively expanded to include the laboratory detection of many microbial targets [1114]. More recently, quantitative real-time PCR has been developed for assessing copy levels of DNA of CMV, EBV, and BKV in blood specimens obtained from patients receiving solid-organ transplants. For target amplification, oligonucleotide primers and probes for amplification and detection, respectively, of nucleic acid are selected from conserved nucleotide sequences within a viral gene; these products constitute the first level of sensitivity and specificity for quantitative real-time PCR. Together with other components, the assay is subsequently adjusted to permit the polymerase enzyme to function optimally and to produce sensitive and specific signals from labeled probes that are proportional to the amount of target DNA present in the blood sample [11, 15]. Three to 5 commercial quantitative standards are included in the quantitative test. The software for the real-time PCR instrument generates a standard curve with use of these quantitative standards. This plot relates the cycle number in which the amplified nucleic acid target from the standards is detected (by measuring fluorescence) to the amount of target present in the standards. The quantitative level of viral nucleic acid in a test specimen is then determined by comparing the cycle number (crossover point) of the specimen with the standard curve generated with the known levels of the target nucleic acid [11]. NUCLEIC ACID EXTRACTION The preanalytical extraction of nucleic acid from a blood specimen is a critical step for the determination of the viral load [11]. Commercial kits are available for manual and automated extraction of nucleic acids from specimens. Automated extraction platforms vary regarding the number of specimens processed, cost, and time requirements for sample throughput. Specimen extraction must achieve effective recovery of the target nucleic acid from the specimen, but the process should also remove inhibitory substances (i.e., heme present in blood samples). Differences in technical aspects of manual and automated extraction methods may be significant variables that affect downstream generation of results of a PCR assay. For practicality, consistency, and reproducibility of results, most laboratories that perform high-volume molecular testing process specimens with automated systems that reduce manual operational procedures and may reduce ergonomic repetitive-motion injuries among laboratory personnel. Overall, the preanalytical nucleic acidextraction step, with use of manual or automated methods, represents a major challenge for standardization of PCR assays among laboratories. RESULT REPORTING Result reporting of quantitative PCR assays varies widely among laboratories. For quantitative PCR, the lower limit of detection is determined by probit analysis. For example, both the laboratory and the clinician need to know the lowest level for detecting target DNA that is of sufficient copy level to obtain a positive result if the specimen is retested. If a 50-mL extract has only a single copy of target DNA and 10 reactions are performed, each 5-mL aliquot of the sample would yield a result in only 1 test vessel; the other nine 5-mL aliquots would yield negative results (figure 1A). Similarly, if 2 copies of target DNA were present in the original extract, a positive result may be expected in a maximum of 2 of 10 test vessels (figure 1B). Finally, high copy levels of target DNA would be expected to produce positive results in each 5-mL aliquot of an extract (figure 1C). Standardization of result information is critical for each laboratory, so that derived values can be compared and be relatively uniform among institutions that perform quantitative PCR testing. An essential component to this goal is the use of commercially available quantitative standards prepared in units of, for example, CMV DNA copies per microliter (as used in our laboratory). Subsequent amplifi (...truncated)