Quantification of Hepatitis B Virus (HBV) DNA with a TaqMan HBV Analyte-Specific Reagent following Sample Processing with the MagNA Pure LC
http://www.100md.com
微生物临床杂志 2006年第4期
Division of Clinical Microbiology
Section of Biostatistics, Mayo Clinic, Rochester, Minnesota
Department of Pathology and Laboratory Medicine, King Faisal Specialist Hospital, Riyadh, Saudi Arabia
ABSTRACT
TaqMan hepatitis B virus (HBV) analyte-specific reagent (ASR; Roche Molecular Systems, Inc., Branchburg, NJ) is designed for the quantification of HBV DNA in serum or plasma. The performance characteristics of TaqMan HBV ASR following automated sample processing with the MagNA Pure LC instrument (MP; Roche Applied Science, Indianapolis, IN) were evaluated in this study. Analytical sensitivity and precision were assessed with commercially available HBV standards, while clinical serum specimens from HBsAg-seropositive patients and healthy blood donors were used to determine clinical sensitivity, specificity, and correlation with other commercially available assays. Analytical studies yielded a limit of detection of 2.4 IU/ml, with good linearity and correlation (R2 = 0.9958) with expected HBV DNA titers over a wide range (6.0 x 100 to 2.1 x 108 IU/ml). Clinical sensitivity and specificity of the assay combined with automated sample processing were both 100%. Comparison of TaqMan HBV ASR and VERSANT HBV DNA 3.0 assay (bDNA; Bayer HealthCare LLC, Tarrytown, NY) results among clinical specimens yielded good correlation (R2 = 0.9237), with a mean difference in titer of –0.213 log10 IU/ml (95% confidence interval, –0.678 to 1.10 log10 IU/ml). The overall test failure rate was 2.0% among 204 clinical serum specimens tested. Total time required for MP sample processing and automated postelution handling of 24 samples was 224 min, with 57 min of actual hands-on time. MP is a reliable, labor-saving platform suitable for use with TaqMan HBV ASR, providing sensitive and accurate quantification of HBV DNA levels over a range of 8 log10 IU/ml.
INTRODUCTION
Hepatitis B virus (HBV) infection continues to be a leading cause of chronic liver disease, with more than 350 million people chronically infected worldwide (7). Chronic HBV carriers are at increased risk for the development of cirrhosis and hepatocellular carcinoma (2). The direct detection of HBV DNA in serum or plasma has become an important tool in the diagnosis of chronic HBV infection (13). Furthermore, serum HBV DNA level may be an important prognostic indicator (8, 9, 16) as well as an important marker for measuring therapeutic response (10) and the development of resistance to antiviral agents (12). A variety of commercially available HBV DNA assays reporting in standardized units with improved sensitivity and wide dynamic ranges are currently available (11, 14, 15).
The COBAS TaqMan HBV test for research use only (TaqMan HBV RUO; Roche Molecular Systems, Inc., Branchburg, NJ) is a commercially available real-time PCR assay designed for the quantitative detection of HBV DNA in human serum or plasma. Like the TaqMan HBV analyte-specific reagent (ASR; Roche Molecular Systems, Inc.), this assay has been developed for use with the COBAS TaqMan 48 analyzer (CTM 48; Roche Molecular Systems, Inc.). Use of the TaqMan HBV ASR requires testing laboratories to establish and validate their own test methods and provide appropriate test controls, whereas TaqMan HBV RUO is a complete test kit currently designated for research use only in the United States.
Currently, TaqMan HBV RUO comprises two products, the High Pure system viral nucleic acid kit (High Pure; Roche Molecular Systems, Inc.) and the COBAS TaqMan HBV test kit, which are packaged and sold separately. High Pure is a manual sample preparation method configured specifically for processing multiple samples in batches of 12 or 24. Specimen preparation is followed by automated PCR amplification and real-time detection of cleaved dual-fluorescent-dye-labeled oligonucleotide (TaqMan) probes that allow the specific and simultaneous detection and quantification of the HBV precore/core target sequence as well as an internal HBV quantitation standard (QS) on the CTM 48 (14).
The benefits of nucleic acid amplification and detection using real-time PCR include substantial reductions in labor, decreased test turnaround time, and reduced potential for contamination with exogenous DNA. The disadvantages of technically demanding manual sample preparation methods are also numerous. Most importantly, these complex procedures may be a potential source of run-to-run variability as well as sample-to-sample contamination. Use of automated sample processing in clinical diagnostic laboratories provides a labor-saving approach to reducing the number of failed sample preparations, while potentially limiting the occurrence of specimen-to-specimen contamination during processing. This approach may also reduce laboratory space requirements and decreased dependence on laboratory technologists skilled in molecular techniques.
Recent evaluations of the MagNA Pure (MP) LC instrument (Roche Applied Science, Indianapolis, IN) used in conjunction with the MP total nucleic acid isolation kit (large volume; Roche Applied Science) have shown it to be a flexible and reliable platform suitable for the extraction and purification of hepatitis C virus RNA from clinical specimens (1, 6). However, similar approaches utilizing the large-volume MP total nucleic acid isolation kit have not been as successful when used in conjunction with TaqMan HBV RUO (M. Qutub, J. Germer, W. Harmsen, J. Mandrekar, P. Mitchell, and J. Yao, Abstr. 21st Annu. Clin. Virol. Symp., abstr. TP22, 2005; J. Smith and V. Cintron, Abstr. 21st Annu. Clin. Virol. Symp., abstr. TP21, 2005). A possible explanation for these findings obtained with HBV DNA is the presence of a covalently bound polypeptide that can reduce the efficiency of the MP extraction process. Boom et al. previously described the effects of this polypeptide and proposed the use of proteinase K to minimize the adverse effects of this polypeptide on HBV DNA recovery (4).
In this study, we evaluated the performance characteristics of TaqMan HBV ASR on HBV standards and clinical serum specimens processed by the MP using the MP DNA isolation kit (large volume; Roche Applied Science). The sample processing method was further modified to incorporate the use of the HBV QS and an additional proteinase K digestion step as previously described (Smith and Cintron, Abstr. 21st Annu. Clin. Virol. Symp.). Analytical sensitivity and precision of the assay were determined with HBV standards, while clinical sensitivity, specificity, correlation, and agreement were assessed by using previously characterized clinical serum specimens.
MATERIALS AND METHODS
HBV standards. Commercially available panels of HBV standards at concentrations of 20,000,000, 2,000,000, 200,000, 20,000, 2,000, 200, and 0 IU/ml (OptiQuant HBV DNA; AcroMetrix Corp., Benicia, CA) were used in this evaluation. Three additional dilutions at 20, 6, and 2 IU/ml were also prepared from these standards with NAT dilution matrix (AcroMetrix Corp.). In addition, a high-titer clinical serum specimen was also diluted to achieve a target concentration of 2.2 x 108 IU/ml based on a previous result obtained from the VERSANT HBV DNA 3.0 assay (bDNA; Bayer HealthCare LLC, Tarrytown, NY) and reanalyzed by bDNA (10 replicates of a 1:100 dilution), yielding a mean titer of 98,019,400 IU/ml. The 6- and 2-IU/ml dilutions were prepared and analyzed in a single test run on the day of preparation, while aliquots of the remaining standards and high-titer patient specimen were stored at –70°C prior to analysis by TaqMan HBV ASR, performed daily (in duplicate) for a total of 5 days.
Clinical specimens. A total of 204 clinical serum specimens submitted to the Mayo Clinic Hepatitis/HIV Laboratory for routine testing between February 2004 and September 2005 were selected retrospectively for inclusion in this study. These specimens included 32 hepatitis B virus surface antigen (HBsAg)-positive/hepatitis B virus e antigen (HBeAg)-negative/hepatitis B virus e antibody (anti-HBe)-positive and 28 HBsAg-positive/HBeAg-positive sera, 50 sera from healthy blood donors, and 94 serum specimens previously tested by bDNA. HBsAg screening of sera from patients and blood donors was performed with the Auzyme monoclonal assay (Abbott Laboratories, Abbott Park, IL), while HBeAg and anti-HBe screening was performed by the ETI-EBK PLUS and ETI-ABEBK PLUS assays (DiaSorin Inc., Stillwater, MN), respectively. Among the 94 specimens previously tested by bDNA, HBV DNA titers ranged from 432 to 479,121,000 IU/ml and they consisted of the following HBV genotypes: A (n = 10), B (n = 2), C (n = 9), D (n = 2), and H (n = 1). There were 70 specimens without HBV genotype determination. Twenty-two of these specimens that initially yielded HBV DNA levels above the upper limit of quantification of bDNA (17,857,100 IU/ml) were diluted 1:100 in normal human plasma prior to retesting. The actual HBV DNA titers of these specimens were derived from the results of the diluted specimens.
Seven well-characterized HBV DNA-positive clinical serum specimens representing HBV genotypes A to G were also chosen to compare PCR amplification efficiencies among the various genotypes. HBV DNA concentration in each specimen was adjusted to 1.0 x 106 IU/ml (undiluted) and diluted 1:10 and 1:100 with NAT dilution matrix prior to testing by TaqMan HBV ASR.
Retrospectively collected clinical specimens were stored at –20°C for up to 19 months prior to testing by TaqMan HBV ASR and TaqMan HBV RUO. Their use in this study was reviewed and approved by the Mayo Foundation Institutional Review Board.
HBV genotype determination. HBV genotyping was performed with the INNO-LiPA HBV genotyping kit (Innogenetics N.V., Ghent, Belgium) according to the manufacturer's instructions following automated sample processing and a modified PCR amplification protocol (M. Qutub, J. Germer, J. Mandrekar, and J. Yao, Abstr. 21st Annu. Clin. Virol. Symp., abstr. TP23, 2005).
MP sample processing and TaqMan HBV ASR. HBV DNA was extracted from 500-μl aliquots of HBV standards, clinical specimens, and assay controls utilizing the MP DNA isolation kit (large volume; Roche Applied Science) and MP software, version 3.0.11. The TaqMan HBV ASR QS was added directly to the MP lysis/binding buffer just prior to the start of automated processing. For a batch of 24 samples, 120 μl of HBV QS was added to 14.88 ml of MP lysis/binding buffer and gently mixed prior to dispensing it into the appropriate MP reagent reservoir. In a further modification of the procedure, reconstituted proteinase K from the large-volume MP DNA isolation kit was added to the MP elution buffer in a 1:100 ratio. For a batch of 24 samples, 120 μl of proteinase K was added to 11.88 ml of MP elution buffer and gently mixed prior to dispensing it into the appropriate MP reagent reservoir. In contrast to the manual High Pure sample processing method (75 μl elution volume), the final MP elution volume for each sample was set at 100 μl, the minimum volume allowed by the MP software application used (DNA LV Blood_300_500.blk).
Following completion of MP sample processing, TaqMan HBV ASR working master mixture was prepared (after a minimum 30-min room temperature equilibration period) by combining and gently mixing 50 mM manganese acetate (191 μl) with TaqMan HBV ASR (1.4 ml). TaqMan HBV ASR working master mix was transferred to a 2.0-ml microcentrifuge tube. This 2.0-ml microcentrifuge tube and 24 open COBAS TaqMan kinetic reaction tubes (K-tubes) contained in a K-tube holder (K-carrier) were placed onto a prototype MP postelution handling block (noncooling) designed specifically for use with the K-carrier format. Working master mix (50 μl) and appropriate sample eluates (50 μl) were added and mixed automatically in the K-tubes by the MP. Then, the K-carrier containing the K-tubes was removed from the MP, and individual K-tubes were manually sealed prior to loading onto the CTM 48. Amplification and detection were initiated within 60 min of working master mix preparation on the CTM 48 utilizing AMPLILINK software, v 3.0.1 (Roche Molecular Systems, Inc.), with the following amplification profile: 2 precycles of 5 min at 50°C and 2 min at 95°C and 55 cycles of 20 s at 95°C and 18 s at 59°C, followed by a postcycle hold of 2 min at 40°C.
Calibration of the TaqMan HBV ASR was performed by using commercially available panels of HBV standards at 20,000,000, 2,000,000, 200,000, 20,000, 2,000, and 200 IU/ml (OptiQuant HBV DNA). This six-member panel was run in quadruplicate in a single run, and the calibration coefficients were calculated by using the ASR external calibration software, version 2.1 (Roche Molecular Systems, Inc.). Calibrators were excluded from the final calculation of calibration coefficients if the back-calculated titers deviated by >0.2 log10 IU/ml from the expected value.
TaqMan HBV RUO. HBV DNA was extracted from a subset of clinical specimens utilizing the High Pure manual sample preparation method (500-μl aliquots) followed by TaqMan HBV RUO testing performed according to the manufacturer's instructions. This subset (38 of 94 specimens previously tested by bDNA and TaqMan HBV ASR) included 14 specimens with discordant bDNA and TaqMan HBV ASR results (differing by >0.5 log10 IU/ml) and all 24 specimens with known HBV genotypes.
Statistical analysis. Observed results of HBV standards tested by TaqMan HBV ASR were compared with the expected results by linear regression, while the lower limit of detection (LLOD) was determined by probit analysis (95% detection rate) (5). Precision of TaqMan HBV ASR was estimated by determining the coefficients of variation at nine different HBV DNA concentrations (6.0 x 100 to 2.1 x 108 IU/ml), tested in duplicate for a total of 5 days. The P value for the difference between median HBV DNA titer of HBeAg-positive and HBeAg-negative specimens was determined by the Wilcoxon rank sum test. Correlation and strength of agreement between TaqMan HBV ASR and bDNA results were determined by linear regression and Bland-Altman plot (3), respectively. Potential differences in target amplification efficiency by TaqMan HBV ASR were examined by comparing the slopes and coefficients of determination (R2) of the linear regression lines drawn from the results of seven clinical specimens (HBV genotypes A to G), each tested at 1.0 x 106 IU/ml (undiluted) and at 1:10 and 1:100 dilutions.
RESULTS
The LLOD or analytical sensitivity of TaqMan HBV ASR combined with sample processing by MP was determined to be 2.4 IU/ml based on probit analysis (95% detection rate) of the results of replicate testing of HBV standards at concentrations of 20, 6, 2, and 0 IU/ml (Table 1). Good linearity was observed across a wide range of HBV DNA levels extending from 6.0 x 100 to 2.1 x 108 IU/ml (Fig. 1), with good correlation (R2 = 0.9958) and coefficients of variation ranging from 7% to 48% among the nine HBV DNA concentrations tested in duplicate for 5 days (Table 2).
Clinical specificity of TaqMan HBV ASR was 100% (95% confidence interval [CI], 93% to 100%) with no HBV DNA detected in any of the 50 HBsAg-negative sera from healthy blood donors examined in this study. Among the 60 HBsAg-positive patient serum specimens, clinical sensitivity of TaqMan HBV ASR was 100% (95% CI, 94% to 100%) when compared to HBsAg positivity. The median HBV DNA titers found in the 32 HBsAg-positive/HBeAg-negative and 28 HBsAg-positive/HBeAg-positive sera were 3.49 and 7.88 log10 IU/ml (P < 0.0001), respectively, with a mean difference of 3.45 log10 IU/ml (95% CI, 2.39 to 4.50 log10 IU/ml).
Comparison of TaqMan HBV ASR and bDNA among the 94 clinical serum specimens previously tested by bDNA demonstrated good correlation (R2 = 0.9237) of the quantitative results from these two assays (Fig. 2). A Bland-Altman plot showed a mean difference of –0.213 log10 IU/ml (95% CI, –0.678 to 1.10 log10 IU/ml) between the corresponding titers obtained by TaqMan HBV ASR and bDNA, with titer differences within 2 standard deviations (SD) (0.909 log10 IU/ml) of the mean difference for 93.6% (88 of 94) of the specimens (Fig. 3). Five specimens yielded titer differences of >2 SD from the mean, while the remaining specimen had no detectable HBV DNA by TaqMan HBV ASR and was excluded from these analyses (Table 3).
Fourteen discordant specimens (including the 6 specimens in Table 3 and 8 additional specimens with discordant bDNA and TaqMan HBV ASR results differing by >0.5 log10 IU/ml) and the 24 specimens with known HBV genotypes were also tested by TaqMan HBV RUO. Among these 38 specimens, a single specimen (D2) failed to yield quantifiable HBV DNA by TaqMan HBV RUO (<6 IU/ml), while the specimen (D1) previously found to contain no detectable HBV DNA by TaqMan HBV ASR also failed to yield detectable HBV DNA by TaqMan HBV RUO. TaqMan HBV ASR and TaqMan HBV RUO yielded good correlation (R2 = 0.9885) of results among the remaining 36 specimens with quantifiable HBV DNA by both assays (data not shown). A Bland-Altman plot indicated that the differences between the corresponding titers obtained by TaqMan HBV ASR and TaqMan HBV RUO were within 2 SD (0.444 log10 IU/ml) of the mean difference in titer (0.024 log10 IU/ml) for 91.7% (33 of 36) of the specimens (data not shown). Among the five specimens with discordant bDNA and TaqMan HBV ASR results and quantifiable HBV DNA by all three assays (samples D3 to D6), the mean titer difference between bDNA and TaqMan HBV ASR was 1.23 log10 IU/ml, while the mean titer difference between TaqMan HBV ASR and TaqMan HBV RUO was just 0.02 log10 IU/ml.
The slopes of the linear regression lines derived from seven clinical specimens containing HBV genotypes A, B, C, D, E, F, and G and tested undiluted and at 1:10 and 1:100 dilutions yielded a median value of 0.8194. The slopes ranged from 0.7478 (genotype G) to 0.8501 (genotype E), while the associated R2 values ranged from 0.9991 (genotype D) to 1.000 (genotype C) among the various genotypes, with a median value of 0.9998.
The hands-on set up time for the MP was 20 min. The time required to prepare and manually load specimens into the MP sample cartridge prior to automated processing was also 20 min. When the postelution handling option was used, the total time required for processing 24 samples by MP was 224 min, an increase of 69 min relative to the estimated time required to perform the manual extraction method (High Pure). However, the actual hands-on time was reduced from an estimated 145 min required by the manual method to just 57 min using MP sample processing and postelution handling; the processing included two extended periods (125 min and 42 min) of hands-off time. Total time duration for the entire procedure (21 samples plus 3 controls), including sample preparation, MP sample processing, and amplification/detection, was an estimated 6.5 h, of which 1 h consisted of actual hands-on time. During the course of this study, there were no failures of sample processing associated with the MP, and the overall test failure rate of TaqMan HBV ASR (an "invalid" test result as defined by the AMPLILINK, v 3.0.1, software) was 2.0% among the 204 clinical specimens tested.
DISCUSSION
The results of our analytical evaluation suggest that TaqMan HBV ASR with sample processing by MP is a sensitive, accurate assay for the detection and quantification of HBV DNA, with a broad dynamic range extending over 8 log10 IU/ml. Comparison of our analytical data to recently published data obtained by TaqMan HBV RUO performed following manual sample processing by High Pure (using the recommended conversion factor of 5.82 copies/IU) suggests that the LLODs and levels of precision of the two assays are quite similar (14). The LLOD (95% detection rate) was determined to be 6 IU/ml for TaqMan HBV RUO in the previous study compared to 2.4 IU/ml in our current evaluation of TaqMan HBV ASR. Total imprecision also ranged from 15.0% to 99.8% in the previous evaluation of TaqMan HBV RUO, while the imprecision of the TaqMan HBV ASR with sample processing by MP ranged from 7% to 48% in our study. The broad dynamic range and increased sensitivity of these assays will improve the clinician's ability to reliably detect the presence of HBV DNA in serum and plasma and monitor changes in HBV viral load in patients, particularly those with very low HBV DNA titers.
While our clinical evaluations were limited in scope, they did demonstrate that TaqMan HBV ASR results were in complete agreement with HBsAg test results obtained by enzyme immunoassay among both HBeAg-positive and HBeAg-negative serum specimens. There was also generally good correlation (R2 = 0.9237) and agreement (93.6%) with the HBV DNA titers obtained by bDNA. However, TaqMan HBV ASR demonstrated a broader dynamic range than bDNA (8 versus 5 log10 IU/ml, respectively), without noticeable differences in PCR amplification efficiencies among the various HBV genotypes.
The single specimen (D1) without detectable HBV DNA by TaqMan HBV ASR and TaqMan HBV RUO yielded a result of 479 IU/ml by bDNA, near the lower limit of quantification (357 IU/ml). Possible explanations for this discordance include loss of HBV DNA during 16 months of storage at –20°C prior to testing by TaqMan HBV ASR and the possibility of a false-positive bDNA result. A recent multicenter evaluation of bDNA indicated clinical specificity of 99.2% among 600 unique HBV-seronegative sera (15), whereas clinical specificity of 100% was found among 316 sera tested by TaqMan HBV RUO (14). It is important to note that all five of the false-positive specimens identified in this previous evaluation of bDNA initially yielded extremely low titers of 387, 393, 476, 557, and 6,411 IU/ml (calculated with the manufacturer-recommended conversion factor of 5.6 copies/IU). These findings suggest that a likely explanation for the discordant results obtained with specimen D1 was a false-positive bDNA result. However, the limited volume of specimen prevented repeat testing by bDNA and/or analysis by another assay. While small, the difference in assay specificity further suggests that TaqMan HBV ASR and TaqMan HBV RUO are better suited than bDNA for the detection of HBV DNA in serum or plasma for diagnostic purposes.
The majority of clinical specimens yielding quantitative results by both bDNA and TaqMan HBV ASR were in agreement, while there was a small subset of specimens yielding significantly discordance results (specimens D2 to D6 in Table 3). These findings were consistent with an earlier report of nonagreement between these two assays (E. Konnick, L. Comanor, J. Di Canzio, E. Ashwood, and D. Hillyard, Abstr. 21st Annu. Clin. Virol. Symp., abstr. TP20, 2005). TaqMan HBV RUO results obtained from specimens D2 to D6 strongly supported the results of TaqMan HBV ASR combined with sample processing by MP, with a mean difference of just 0.02 log10 IU/ml. There was also good agreement with TaqMan HBV ASR among the larger subset of specimens yielding quantitative results by TaqMan HBV RUO (n = 36), with a mean titer difference of 0.024 log10 IU/ml. Despite using a common unit of measure, significant discordance between the quantitative results obtained from some specimens tested by TaqMan HBV ASR and bDNA (but not TaqMan HBV RUO) highlights the importance of using the same assay when sequentially monitoring individual patients undergoing anti-HBV therapy.
Consistent with our previous evaluation of MP automated sample processing for the COBAS TaqMan HCV test (6), we experienced no significant problems related to the performance of MP and found no evidence of sample-to-sample contamination during this evaluation, despite the processing of both HBV DNA-negative and high-titer samples within the same batches on the MP. However, we have found that improper pipetting techniques may result in sample cross-contamination during manual addition of samples into the MP sample cartridge prior to automated sample processing. The potential for specimen-to-specimen contamination during manual High Pure sample processing has also been identified and addressed specifically in a manufacturer's bulletin sent to laboratories performing TaqMan HBV RUO (reagent bulletin 04-031, dated 1 March 2004, Roche Diagnostics Corp.).
MP sample processing and postelution handling provided two extended periods of hands-off time, thus allowing laboratory technologists time to perform other tasks. Despite an overall increase in processing time relative to the manufacturer-recommended High Pure manual extraction procedure, the hands-on time required for automated sample processing by MP was reduced by 1 h with or without use of the postelution handling capability of MP for a batch size of 24 samples.
Although the MP performed well in this evaluation utilizing a batch size of 24, differences in MP and CTM 48 configuration (maximum batch sizes of 32 and 48 samples, respectively) limit the overall efficiency of MP sample processing for TaqMan HBV ASR. Use of multiple MP instruments may be necessary in clinical laboratories with high specimen throughput demands (6). Alternatively, omission of postelution handling by MP in favor of manual K-tube loading could decrease the total time required for sample processing by MP and increase specimen throughput.
In summary, this study demonstrated that the MP is a reliable, labor-saving sample processing platform suitable for use with TaqMan HBV ASR. TaqMan HBV ASR with sample processing by MP provided sensitive and accurate quantitative HBV DNA results over a range of 8 log10 IU/ml. Our findings, generated using an analyte-specific reagent (TaqMan HBV ASR), suggest that MP sample processing may also be suitable for use with TaqMan HBV RUO (COBAS TaqMan HBV test).
ACKNOWLEDGMENTS
We thank Jarman V. Smith of the Roche Diagnostics Applications Laboratory, Indianapolis, IN, for providing technical guidance with the automated sample processing portion of this study. Roche Diagnostics Corporation provided the MagNA Pure LC and TaqMan reagents used in this study.
This work was supported in part by a grant from Roche Diagnostics Corporation.
REFERENCES
Barbeau, J. M., J. Goforth, A. M. Caliendo, and F. S. Nolte. 2004. Performance characteristics of a quantitative TaqMan hepatitis C virus RNA analyte-specific reagent. J. Clin. Microbiol. 42:3739-3746.
Beasley, R. P. 1988. Hepatitis B virus. The major etiology of hepatocellular carcinoma. Cancer 61:1942-1956.
Bland, J. M., and D. G. Altman. 1986. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet i:307-310.
Boom, R., C. J. Sol, R. Heijtink, P. M. Wertheim-van Dillen, and J. van der Noordaa. 1991. Rapid purification of hepatitis B virus DNA from serum. J. Clin. Microbiol. 29:1804-1811.
Finney, D. J. 1971. Probit analysis, 3rd ed. Cambridge University Press, Cambridge, United Kingdom.
Germer, J. J., W. S. Harmsen, J. N. Mandrekar, P. S. Mitchell, and J. D. C. Yao. 2005. Evaluation of the COBAS TaqMan HCV test with automated sample processing using the MagNA Pure LC instrument. J. Clin. Microbiol. 43:293-298.
Lee, W. M. 1997. Hepatitis B virus infection. N. Engl. J. Med. 337:1733-1745.
Lok, A. S., E. J. Heathcote, and J. H. Hoofnagle. 2001. Management of hepatitis B: 2000—summary of a workshop. Gastroenterology 120:1828-1853.
Martinot-Peignoux, M., N. Boyer, M. Colombat, R. Akremi, B. N. Pham, S. Ollivier, C. Castelnau, D. Valla, C. Degott, and P. Marcellin. 2002. Serum hepatitis B virus DNA levels and liver histology in inactive HBsAg carriers. J. Hepatol. 36:543-546.
Mommeja-Marin, H., E. Mondou, M. R. Blum, and F. Rousseau. 2003. Serum HBV DNA as a marker of efficacy during therapy for chronic HBV infection: analysis and review of the literature. Hepatology 37:1309-1319.
Pawlotsky, J. M. 2003. Hepatitis B virus (HBV) DNA assays (methods and practical use) and viral kinetics. J. Hepatol. 39(Suppl. 1):S31-S35.
Puchhammer-Stockl, E., C. W. Mandl, J. Kletzmayr, H. Holzmann, A. Hofmann, S. W. Aberle, F. X. Heinz, B. Watschinger, and H. Hofmann. 2000. Monitoring the virus load can predict the emergence of drug-resistant hepatitis B virus strains in renal transplantation patients during lamivudine therapy. J. Infect. Dis. 181:2063-2066.
Saito, T., H. Shinzawa, T. Uchida, O. Kawamata, S. Honma, H. Watanabe, L. Shao, K. Saito, H. Togashi, and T. Takahashi. 1999. Quantitative DNA analysis of low-level hepatitis B viremia in two patients with serologically negative chronic hepatitis B. J. Med. Virol. 58:325-331.
Weiss, J., H. Wu, B. Farrenkopf, T. Schultz, G. Song, S. Shah, and J. Siegel. 2004. Real time TaqMan PCR detection and quantitation of HBV genotypes A-G with the use of an internal quantitation standard. J. Clin. Virol. 30:86-93.
Yao, J. D., M. G. Beld, L. L. Oon, C. H. Sherlock, J. Germer, S. Menting, S. Y. Se Thoe, L. Merrick, R. Ziermann, J. Surtihadi, and H. J. Hnatyszyn. 2004. Multicenter evaluation of the VERSANT hepatitis B virus DNA 3.0 assay. J. Clin. Microbiol. 42:800-806.
Yu, M. W., S. H. Yeh, P. J. Chen, Y. F. Liaw, C. L. Lin, C. J. Liu, W. L. Shih, J. H. Kao, D. S. Chen, and C. J. Chen. 2005. Hepatitis B virus genotype and DNA level and hepatocellular carcinoma: a prospective study in men. J. Natl. Cancer Inst. 97:265-272.(Jeffrey J. Germer, Mohamm)
Section of Biostatistics, Mayo Clinic, Rochester, Minnesota
Department of Pathology and Laboratory Medicine, King Faisal Specialist Hospital, Riyadh, Saudi Arabia
ABSTRACT
TaqMan hepatitis B virus (HBV) analyte-specific reagent (ASR; Roche Molecular Systems, Inc., Branchburg, NJ) is designed for the quantification of HBV DNA in serum or plasma. The performance characteristics of TaqMan HBV ASR following automated sample processing with the MagNA Pure LC instrument (MP; Roche Applied Science, Indianapolis, IN) were evaluated in this study. Analytical sensitivity and precision were assessed with commercially available HBV standards, while clinical serum specimens from HBsAg-seropositive patients and healthy blood donors were used to determine clinical sensitivity, specificity, and correlation with other commercially available assays. Analytical studies yielded a limit of detection of 2.4 IU/ml, with good linearity and correlation (R2 = 0.9958) with expected HBV DNA titers over a wide range (6.0 x 100 to 2.1 x 108 IU/ml). Clinical sensitivity and specificity of the assay combined with automated sample processing were both 100%. Comparison of TaqMan HBV ASR and VERSANT HBV DNA 3.0 assay (bDNA; Bayer HealthCare LLC, Tarrytown, NY) results among clinical specimens yielded good correlation (R2 = 0.9237), with a mean difference in titer of –0.213 log10 IU/ml (95% confidence interval, –0.678 to 1.10 log10 IU/ml). The overall test failure rate was 2.0% among 204 clinical serum specimens tested. Total time required for MP sample processing and automated postelution handling of 24 samples was 224 min, with 57 min of actual hands-on time. MP is a reliable, labor-saving platform suitable for use with TaqMan HBV ASR, providing sensitive and accurate quantification of HBV DNA levels over a range of 8 log10 IU/ml.
INTRODUCTION
Hepatitis B virus (HBV) infection continues to be a leading cause of chronic liver disease, with more than 350 million people chronically infected worldwide (7). Chronic HBV carriers are at increased risk for the development of cirrhosis and hepatocellular carcinoma (2). The direct detection of HBV DNA in serum or plasma has become an important tool in the diagnosis of chronic HBV infection (13). Furthermore, serum HBV DNA level may be an important prognostic indicator (8, 9, 16) as well as an important marker for measuring therapeutic response (10) and the development of resistance to antiviral agents (12). A variety of commercially available HBV DNA assays reporting in standardized units with improved sensitivity and wide dynamic ranges are currently available (11, 14, 15).
The COBAS TaqMan HBV test for research use only (TaqMan HBV RUO; Roche Molecular Systems, Inc., Branchburg, NJ) is a commercially available real-time PCR assay designed for the quantitative detection of HBV DNA in human serum or plasma. Like the TaqMan HBV analyte-specific reagent (ASR; Roche Molecular Systems, Inc.), this assay has been developed for use with the COBAS TaqMan 48 analyzer (CTM 48; Roche Molecular Systems, Inc.). Use of the TaqMan HBV ASR requires testing laboratories to establish and validate their own test methods and provide appropriate test controls, whereas TaqMan HBV RUO is a complete test kit currently designated for research use only in the United States.
Currently, TaqMan HBV RUO comprises two products, the High Pure system viral nucleic acid kit (High Pure; Roche Molecular Systems, Inc.) and the COBAS TaqMan HBV test kit, which are packaged and sold separately. High Pure is a manual sample preparation method configured specifically for processing multiple samples in batches of 12 or 24. Specimen preparation is followed by automated PCR amplification and real-time detection of cleaved dual-fluorescent-dye-labeled oligonucleotide (TaqMan) probes that allow the specific and simultaneous detection and quantification of the HBV precore/core target sequence as well as an internal HBV quantitation standard (QS) on the CTM 48 (14).
The benefits of nucleic acid amplification and detection using real-time PCR include substantial reductions in labor, decreased test turnaround time, and reduced potential for contamination with exogenous DNA. The disadvantages of technically demanding manual sample preparation methods are also numerous. Most importantly, these complex procedures may be a potential source of run-to-run variability as well as sample-to-sample contamination. Use of automated sample processing in clinical diagnostic laboratories provides a labor-saving approach to reducing the number of failed sample preparations, while potentially limiting the occurrence of specimen-to-specimen contamination during processing. This approach may also reduce laboratory space requirements and decreased dependence on laboratory technologists skilled in molecular techniques.
Recent evaluations of the MagNA Pure (MP) LC instrument (Roche Applied Science, Indianapolis, IN) used in conjunction with the MP total nucleic acid isolation kit (large volume; Roche Applied Science) have shown it to be a flexible and reliable platform suitable for the extraction and purification of hepatitis C virus RNA from clinical specimens (1, 6). However, similar approaches utilizing the large-volume MP total nucleic acid isolation kit have not been as successful when used in conjunction with TaqMan HBV RUO (M. Qutub, J. Germer, W. Harmsen, J. Mandrekar, P. Mitchell, and J. Yao, Abstr. 21st Annu. Clin. Virol. Symp., abstr. TP22, 2005; J. Smith and V. Cintron, Abstr. 21st Annu. Clin. Virol. Symp., abstr. TP21, 2005). A possible explanation for these findings obtained with HBV DNA is the presence of a covalently bound polypeptide that can reduce the efficiency of the MP extraction process. Boom et al. previously described the effects of this polypeptide and proposed the use of proteinase K to minimize the adverse effects of this polypeptide on HBV DNA recovery (4).
In this study, we evaluated the performance characteristics of TaqMan HBV ASR on HBV standards and clinical serum specimens processed by the MP using the MP DNA isolation kit (large volume; Roche Applied Science). The sample processing method was further modified to incorporate the use of the HBV QS and an additional proteinase K digestion step as previously described (Smith and Cintron, Abstr. 21st Annu. Clin. Virol. Symp.). Analytical sensitivity and precision of the assay were determined with HBV standards, while clinical sensitivity, specificity, correlation, and agreement were assessed by using previously characterized clinical serum specimens.
MATERIALS AND METHODS
HBV standards. Commercially available panels of HBV standards at concentrations of 20,000,000, 2,000,000, 200,000, 20,000, 2,000, 200, and 0 IU/ml (OptiQuant HBV DNA; AcroMetrix Corp., Benicia, CA) were used in this evaluation. Three additional dilutions at 20, 6, and 2 IU/ml were also prepared from these standards with NAT dilution matrix (AcroMetrix Corp.). In addition, a high-titer clinical serum specimen was also diluted to achieve a target concentration of 2.2 x 108 IU/ml based on a previous result obtained from the VERSANT HBV DNA 3.0 assay (bDNA; Bayer HealthCare LLC, Tarrytown, NY) and reanalyzed by bDNA (10 replicates of a 1:100 dilution), yielding a mean titer of 98,019,400 IU/ml. The 6- and 2-IU/ml dilutions were prepared and analyzed in a single test run on the day of preparation, while aliquots of the remaining standards and high-titer patient specimen were stored at –70°C prior to analysis by TaqMan HBV ASR, performed daily (in duplicate) for a total of 5 days.
Clinical specimens. A total of 204 clinical serum specimens submitted to the Mayo Clinic Hepatitis/HIV Laboratory for routine testing between February 2004 and September 2005 were selected retrospectively for inclusion in this study. These specimens included 32 hepatitis B virus surface antigen (HBsAg)-positive/hepatitis B virus e antigen (HBeAg)-negative/hepatitis B virus e antibody (anti-HBe)-positive and 28 HBsAg-positive/HBeAg-positive sera, 50 sera from healthy blood donors, and 94 serum specimens previously tested by bDNA. HBsAg screening of sera from patients and blood donors was performed with the Auzyme monoclonal assay (Abbott Laboratories, Abbott Park, IL), while HBeAg and anti-HBe screening was performed by the ETI-EBK PLUS and ETI-ABEBK PLUS assays (DiaSorin Inc., Stillwater, MN), respectively. Among the 94 specimens previously tested by bDNA, HBV DNA titers ranged from 432 to 479,121,000 IU/ml and they consisted of the following HBV genotypes: A (n = 10), B (n = 2), C (n = 9), D (n = 2), and H (n = 1). There were 70 specimens without HBV genotype determination. Twenty-two of these specimens that initially yielded HBV DNA levels above the upper limit of quantification of bDNA (17,857,100 IU/ml) were diluted 1:100 in normal human plasma prior to retesting. The actual HBV DNA titers of these specimens were derived from the results of the diluted specimens.
Seven well-characterized HBV DNA-positive clinical serum specimens representing HBV genotypes A to G were also chosen to compare PCR amplification efficiencies among the various genotypes. HBV DNA concentration in each specimen was adjusted to 1.0 x 106 IU/ml (undiluted) and diluted 1:10 and 1:100 with NAT dilution matrix prior to testing by TaqMan HBV ASR.
Retrospectively collected clinical specimens were stored at –20°C for up to 19 months prior to testing by TaqMan HBV ASR and TaqMan HBV RUO. Their use in this study was reviewed and approved by the Mayo Foundation Institutional Review Board.
HBV genotype determination. HBV genotyping was performed with the INNO-LiPA HBV genotyping kit (Innogenetics N.V., Ghent, Belgium) according to the manufacturer's instructions following automated sample processing and a modified PCR amplification protocol (M. Qutub, J. Germer, J. Mandrekar, and J. Yao, Abstr. 21st Annu. Clin. Virol. Symp., abstr. TP23, 2005).
MP sample processing and TaqMan HBV ASR. HBV DNA was extracted from 500-μl aliquots of HBV standards, clinical specimens, and assay controls utilizing the MP DNA isolation kit (large volume; Roche Applied Science) and MP software, version 3.0.11. The TaqMan HBV ASR QS was added directly to the MP lysis/binding buffer just prior to the start of automated processing. For a batch of 24 samples, 120 μl of HBV QS was added to 14.88 ml of MP lysis/binding buffer and gently mixed prior to dispensing it into the appropriate MP reagent reservoir. In a further modification of the procedure, reconstituted proteinase K from the large-volume MP DNA isolation kit was added to the MP elution buffer in a 1:100 ratio. For a batch of 24 samples, 120 μl of proteinase K was added to 11.88 ml of MP elution buffer and gently mixed prior to dispensing it into the appropriate MP reagent reservoir. In contrast to the manual High Pure sample processing method (75 μl elution volume), the final MP elution volume for each sample was set at 100 μl, the minimum volume allowed by the MP software application used (DNA LV Blood_300_500.blk).
Following completion of MP sample processing, TaqMan HBV ASR working master mixture was prepared (after a minimum 30-min room temperature equilibration period) by combining and gently mixing 50 mM manganese acetate (191 μl) with TaqMan HBV ASR (1.4 ml). TaqMan HBV ASR working master mix was transferred to a 2.0-ml microcentrifuge tube. This 2.0-ml microcentrifuge tube and 24 open COBAS TaqMan kinetic reaction tubes (K-tubes) contained in a K-tube holder (K-carrier) were placed onto a prototype MP postelution handling block (noncooling) designed specifically for use with the K-carrier format. Working master mix (50 μl) and appropriate sample eluates (50 μl) were added and mixed automatically in the K-tubes by the MP. Then, the K-carrier containing the K-tubes was removed from the MP, and individual K-tubes were manually sealed prior to loading onto the CTM 48. Amplification and detection were initiated within 60 min of working master mix preparation on the CTM 48 utilizing AMPLILINK software, v 3.0.1 (Roche Molecular Systems, Inc.), with the following amplification profile: 2 precycles of 5 min at 50°C and 2 min at 95°C and 55 cycles of 20 s at 95°C and 18 s at 59°C, followed by a postcycle hold of 2 min at 40°C.
Calibration of the TaqMan HBV ASR was performed by using commercially available panels of HBV standards at 20,000,000, 2,000,000, 200,000, 20,000, 2,000, and 200 IU/ml (OptiQuant HBV DNA). This six-member panel was run in quadruplicate in a single run, and the calibration coefficients were calculated by using the ASR external calibration software, version 2.1 (Roche Molecular Systems, Inc.). Calibrators were excluded from the final calculation of calibration coefficients if the back-calculated titers deviated by >0.2 log10 IU/ml from the expected value.
TaqMan HBV RUO. HBV DNA was extracted from a subset of clinical specimens utilizing the High Pure manual sample preparation method (500-μl aliquots) followed by TaqMan HBV RUO testing performed according to the manufacturer's instructions. This subset (38 of 94 specimens previously tested by bDNA and TaqMan HBV ASR) included 14 specimens with discordant bDNA and TaqMan HBV ASR results (differing by >0.5 log10 IU/ml) and all 24 specimens with known HBV genotypes.
Statistical analysis. Observed results of HBV standards tested by TaqMan HBV ASR were compared with the expected results by linear regression, while the lower limit of detection (LLOD) was determined by probit analysis (95% detection rate) (5). Precision of TaqMan HBV ASR was estimated by determining the coefficients of variation at nine different HBV DNA concentrations (6.0 x 100 to 2.1 x 108 IU/ml), tested in duplicate for a total of 5 days. The P value for the difference between median HBV DNA titer of HBeAg-positive and HBeAg-negative specimens was determined by the Wilcoxon rank sum test. Correlation and strength of agreement between TaqMan HBV ASR and bDNA results were determined by linear regression and Bland-Altman plot (3), respectively. Potential differences in target amplification efficiency by TaqMan HBV ASR were examined by comparing the slopes and coefficients of determination (R2) of the linear regression lines drawn from the results of seven clinical specimens (HBV genotypes A to G), each tested at 1.0 x 106 IU/ml (undiluted) and at 1:10 and 1:100 dilutions.
RESULTS
The LLOD or analytical sensitivity of TaqMan HBV ASR combined with sample processing by MP was determined to be 2.4 IU/ml based on probit analysis (95% detection rate) of the results of replicate testing of HBV standards at concentrations of 20, 6, 2, and 0 IU/ml (Table 1). Good linearity was observed across a wide range of HBV DNA levels extending from 6.0 x 100 to 2.1 x 108 IU/ml (Fig. 1), with good correlation (R2 = 0.9958) and coefficients of variation ranging from 7% to 48% among the nine HBV DNA concentrations tested in duplicate for 5 days (Table 2).
Clinical specificity of TaqMan HBV ASR was 100% (95% confidence interval [CI], 93% to 100%) with no HBV DNA detected in any of the 50 HBsAg-negative sera from healthy blood donors examined in this study. Among the 60 HBsAg-positive patient serum specimens, clinical sensitivity of TaqMan HBV ASR was 100% (95% CI, 94% to 100%) when compared to HBsAg positivity. The median HBV DNA titers found in the 32 HBsAg-positive/HBeAg-negative and 28 HBsAg-positive/HBeAg-positive sera were 3.49 and 7.88 log10 IU/ml (P < 0.0001), respectively, with a mean difference of 3.45 log10 IU/ml (95% CI, 2.39 to 4.50 log10 IU/ml).
Comparison of TaqMan HBV ASR and bDNA among the 94 clinical serum specimens previously tested by bDNA demonstrated good correlation (R2 = 0.9237) of the quantitative results from these two assays (Fig. 2). A Bland-Altman plot showed a mean difference of –0.213 log10 IU/ml (95% CI, –0.678 to 1.10 log10 IU/ml) between the corresponding titers obtained by TaqMan HBV ASR and bDNA, with titer differences within 2 standard deviations (SD) (0.909 log10 IU/ml) of the mean difference for 93.6% (88 of 94) of the specimens (Fig. 3). Five specimens yielded titer differences of >2 SD from the mean, while the remaining specimen had no detectable HBV DNA by TaqMan HBV ASR and was excluded from these analyses (Table 3).
Fourteen discordant specimens (including the 6 specimens in Table 3 and 8 additional specimens with discordant bDNA and TaqMan HBV ASR results differing by >0.5 log10 IU/ml) and the 24 specimens with known HBV genotypes were also tested by TaqMan HBV RUO. Among these 38 specimens, a single specimen (D2) failed to yield quantifiable HBV DNA by TaqMan HBV RUO (<6 IU/ml), while the specimen (D1) previously found to contain no detectable HBV DNA by TaqMan HBV ASR also failed to yield detectable HBV DNA by TaqMan HBV RUO. TaqMan HBV ASR and TaqMan HBV RUO yielded good correlation (R2 = 0.9885) of results among the remaining 36 specimens with quantifiable HBV DNA by both assays (data not shown). A Bland-Altman plot indicated that the differences between the corresponding titers obtained by TaqMan HBV ASR and TaqMan HBV RUO were within 2 SD (0.444 log10 IU/ml) of the mean difference in titer (0.024 log10 IU/ml) for 91.7% (33 of 36) of the specimens (data not shown). Among the five specimens with discordant bDNA and TaqMan HBV ASR results and quantifiable HBV DNA by all three assays (samples D3 to D6), the mean titer difference between bDNA and TaqMan HBV ASR was 1.23 log10 IU/ml, while the mean titer difference between TaqMan HBV ASR and TaqMan HBV RUO was just 0.02 log10 IU/ml.
The slopes of the linear regression lines derived from seven clinical specimens containing HBV genotypes A, B, C, D, E, F, and G and tested undiluted and at 1:10 and 1:100 dilutions yielded a median value of 0.8194. The slopes ranged from 0.7478 (genotype G) to 0.8501 (genotype E), while the associated R2 values ranged from 0.9991 (genotype D) to 1.000 (genotype C) among the various genotypes, with a median value of 0.9998.
The hands-on set up time for the MP was 20 min. The time required to prepare and manually load specimens into the MP sample cartridge prior to automated processing was also 20 min. When the postelution handling option was used, the total time required for processing 24 samples by MP was 224 min, an increase of 69 min relative to the estimated time required to perform the manual extraction method (High Pure). However, the actual hands-on time was reduced from an estimated 145 min required by the manual method to just 57 min using MP sample processing and postelution handling; the processing included two extended periods (125 min and 42 min) of hands-off time. Total time duration for the entire procedure (21 samples plus 3 controls), including sample preparation, MP sample processing, and amplification/detection, was an estimated 6.5 h, of which 1 h consisted of actual hands-on time. During the course of this study, there were no failures of sample processing associated with the MP, and the overall test failure rate of TaqMan HBV ASR (an "invalid" test result as defined by the AMPLILINK, v 3.0.1, software) was 2.0% among the 204 clinical specimens tested.
DISCUSSION
The results of our analytical evaluation suggest that TaqMan HBV ASR with sample processing by MP is a sensitive, accurate assay for the detection and quantification of HBV DNA, with a broad dynamic range extending over 8 log10 IU/ml. Comparison of our analytical data to recently published data obtained by TaqMan HBV RUO performed following manual sample processing by High Pure (using the recommended conversion factor of 5.82 copies/IU) suggests that the LLODs and levels of precision of the two assays are quite similar (14). The LLOD (95% detection rate) was determined to be 6 IU/ml for TaqMan HBV RUO in the previous study compared to 2.4 IU/ml in our current evaluation of TaqMan HBV ASR. Total imprecision also ranged from 15.0% to 99.8% in the previous evaluation of TaqMan HBV RUO, while the imprecision of the TaqMan HBV ASR with sample processing by MP ranged from 7% to 48% in our study. The broad dynamic range and increased sensitivity of these assays will improve the clinician's ability to reliably detect the presence of HBV DNA in serum and plasma and monitor changes in HBV viral load in patients, particularly those with very low HBV DNA titers.
While our clinical evaluations were limited in scope, they did demonstrate that TaqMan HBV ASR results were in complete agreement with HBsAg test results obtained by enzyme immunoassay among both HBeAg-positive and HBeAg-negative serum specimens. There was also generally good correlation (R2 = 0.9237) and agreement (93.6%) with the HBV DNA titers obtained by bDNA. However, TaqMan HBV ASR demonstrated a broader dynamic range than bDNA (8 versus 5 log10 IU/ml, respectively), without noticeable differences in PCR amplification efficiencies among the various HBV genotypes.
The single specimen (D1) without detectable HBV DNA by TaqMan HBV ASR and TaqMan HBV RUO yielded a result of 479 IU/ml by bDNA, near the lower limit of quantification (357 IU/ml). Possible explanations for this discordance include loss of HBV DNA during 16 months of storage at –20°C prior to testing by TaqMan HBV ASR and the possibility of a false-positive bDNA result. A recent multicenter evaluation of bDNA indicated clinical specificity of 99.2% among 600 unique HBV-seronegative sera (15), whereas clinical specificity of 100% was found among 316 sera tested by TaqMan HBV RUO (14). It is important to note that all five of the false-positive specimens identified in this previous evaluation of bDNA initially yielded extremely low titers of 387, 393, 476, 557, and 6,411 IU/ml (calculated with the manufacturer-recommended conversion factor of 5.6 copies/IU). These findings suggest that a likely explanation for the discordant results obtained with specimen D1 was a false-positive bDNA result. However, the limited volume of specimen prevented repeat testing by bDNA and/or analysis by another assay. While small, the difference in assay specificity further suggests that TaqMan HBV ASR and TaqMan HBV RUO are better suited than bDNA for the detection of HBV DNA in serum or plasma for diagnostic purposes.
The majority of clinical specimens yielding quantitative results by both bDNA and TaqMan HBV ASR were in agreement, while there was a small subset of specimens yielding significantly discordance results (specimens D2 to D6 in Table 3). These findings were consistent with an earlier report of nonagreement between these two assays (E. Konnick, L. Comanor, J. Di Canzio, E. Ashwood, and D. Hillyard, Abstr. 21st Annu. Clin. Virol. Symp., abstr. TP20, 2005). TaqMan HBV RUO results obtained from specimens D2 to D6 strongly supported the results of TaqMan HBV ASR combined with sample processing by MP, with a mean difference of just 0.02 log10 IU/ml. There was also good agreement with TaqMan HBV ASR among the larger subset of specimens yielding quantitative results by TaqMan HBV RUO (n = 36), with a mean titer difference of 0.024 log10 IU/ml. Despite using a common unit of measure, significant discordance between the quantitative results obtained from some specimens tested by TaqMan HBV ASR and bDNA (but not TaqMan HBV RUO) highlights the importance of using the same assay when sequentially monitoring individual patients undergoing anti-HBV therapy.
Consistent with our previous evaluation of MP automated sample processing for the COBAS TaqMan HCV test (6), we experienced no significant problems related to the performance of MP and found no evidence of sample-to-sample contamination during this evaluation, despite the processing of both HBV DNA-negative and high-titer samples within the same batches on the MP. However, we have found that improper pipetting techniques may result in sample cross-contamination during manual addition of samples into the MP sample cartridge prior to automated sample processing. The potential for specimen-to-specimen contamination during manual High Pure sample processing has also been identified and addressed specifically in a manufacturer's bulletin sent to laboratories performing TaqMan HBV RUO (reagent bulletin 04-031, dated 1 March 2004, Roche Diagnostics Corp.).
MP sample processing and postelution handling provided two extended periods of hands-off time, thus allowing laboratory technologists time to perform other tasks. Despite an overall increase in processing time relative to the manufacturer-recommended High Pure manual extraction procedure, the hands-on time required for automated sample processing by MP was reduced by 1 h with or without use of the postelution handling capability of MP for a batch size of 24 samples.
Although the MP performed well in this evaluation utilizing a batch size of 24, differences in MP and CTM 48 configuration (maximum batch sizes of 32 and 48 samples, respectively) limit the overall efficiency of MP sample processing for TaqMan HBV ASR. Use of multiple MP instruments may be necessary in clinical laboratories with high specimen throughput demands (6). Alternatively, omission of postelution handling by MP in favor of manual K-tube loading could decrease the total time required for sample processing by MP and increase specimen throughput.
In summary, this study demonstrated that the MP is a reliable, labor-saving sample processing platform suitable for use with TaqMan HBV ASR. TaqMan HBV ASR with sample processing by MP provided sensitive and accurate quantitative HBV DNA results over a range of 8 log10 IU/ml. Our findings, generated using an analyte-specific reagent (TaqMan HBV ASR), suggest that MP sample processing may also be suitable for use with TaqMan HBV RUO (COBAS TaqMan HBV test).
ACKNOWLEDGMENTS
We thank Jarman V. Smith of the Roche Diagnostics Applications Laboratory, Indianapolis, IN, for providing technical guidance with the automated sample processing portion of this study. Roche Diagnostics Corporation provided the MagNA Pure LC and TaqMan reagents used in this study.
This work was supported in part by a grant from Roche Diagnostics Corporation.
REFERENCES
Barbeau, J. M., J. Goforth, A. M. Caliendo, and F. S. Nolte. 2004. Performance characteristics of a quantitative TaqMan hepatitis C virus RNA analyte-specific reagent. J. Clin. Microbiol. 42:3739-3746.
Beasley, R. P. 1988. Hepatitis B virus. The major etiology of hepatocellular carcinoma. Cancer 61:1942-1956.
Bland, J. M., and D. G. Altman. 1986. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet i:307-310.
Boom, R., C. J. Sol, R. Heijtink, P. M. Wertheim-van Dillen, and J. van der Noordaa. 1991. Rapid purification of hepatitis B virus DNA from serum. J. Clin. Microbiol. 29:1804-1811.
Finney, D. J. 1971. Probit analysis, 3rd ed. Cambridge University Press, Cambridge, United Kingdom.
Germer, J. J., W. S. Harmsen, J. N. Mandrekar, P. S. Mitchell, and J. D. C. Yao. 2005. Evaluation of the COBAS TaqMan HCV test with automated sample processing using the MagNA Pure LC instrument. J. Clin. Microbiol. 43:293-298.
Lee, W. M. 1997. Hepatitis B virus infection. N. Engl. J. Med. 337:1733-1745.
Lok, A. S., E. J. Heathcote, and J. H. Hoofnagle. 2001. Management of hepatitis B: 2000—summary of a workshop. Gastroenterology 120:1828-1853.
Martinot-Peignoux, M., N. Boyer, M. Colombat, R. Akremi, B. N. Pham, S. Ollivier, C. Castelnau, D. Valla, C. Degott, and P. Marcellin. 2002. Serum hepatitis B virus DNA levels and liver histology in inactive HBsAg carriers. J. Hepatol. 36:543-546.
Mommeja-Marin, H., E. Mondou, M. R. Blum, and F. Rousseau. 2003. Serum HBV DNA as a marker of efficacy during therapy for chronic HBV infection: analysis and review of the literature. Hepatology 37:1309-1319.
Pawlotsky, J. M. 2003. Hepatitis B virus (HBV) DNA assays (methods and practical use) and viral kinetics. J. Hepatol. 39(Suppl. 1):S31-S35.
Puchhammer-Stockl, E., C. W. Mandl, J. Kletzmayr, H. Holzmann, A. Hofmann, S. W. Aberle, F. X. Heinz, B. Watschinger, and H. Hofmann. 2000. Monitoring the virus load can predict the emergence of drug-resistant hepatitis B virus strains in renal transplantation patients during lamivudine therapy. J. Infect. Dis. 181:2063-2066.
Saito, T., H. Shinzawa, T. Uchida, O. Kawamata, S. Honma, H. Watanabe, L. Shao, K. Saito, H. Togashi, and T. Takahashi. 1999. Quantitative DNA analysis of low-level hepatitis B viremia in two patients with serologically negative chronic hepatitis B. J. Med. Virol. 58:325-331.
Weiss, J., H. Wu, B. Farrenkopf, T. Schultz, G. Song, S. Shah, and J. Siegel. 2004. Real time TaqMan PCR detection and quantitation of HBV genotypes A-G with the use of an internal quantitation standard. J. Clin. Virol. 30:86-93.
Yao, J. D., M. G. Beld, L. L. Oon, C. H. Sherlock, J. Germer, S. Menting, S. Y. Se Thoe, L. Merrick, R. Ziermann, J. Surtihadi, and H. J. Hnatyszyn. 2004. Multicenter evaluation of the VERSANT hepatitis B virus DNA 3.0 assay. J. Clin. Microbiol. 42:800-806.
Yu, M. W., S. H. Yeh, P. J. Chen, Y. F. Liaw, C. L. Lin, C. J. Liu, W. L. Shih, J. H. Kao, D. S. Chen, and C. J. Chen. 2005. Hepatitis B virus genotype and DNA level and hepatocellular carcinoma: a prospective study in men. J. Natl. Cancer Inst. 97:265-272.(Jeffrey J. Germer, Mohamm)