- Email: [email protected]
Preclinical evaluation of the versant HCV RNA 3.0 assay (BDNA)
HEt'ATOLOGYVol. 34, No. 4, Pt. 2, 2001
AASLD ABSTRACTS
563A
1563
1564
PRECLINICAL EVALUATION OF THE VERSANT HCV RNA 3.0 ASSAY (BDNA). Andy Worlock, John Arruda, Tonya Black, Susan Bonapace, Julie
BIOCHEMICAL MARKERS OF LIVER FIBROSIS IN PATIENTS INFECTED BY HEPATITIS C VIRUS: LONGITUDINAL VALIDATION IN A RANDOMIZED TRIAL. Thierry Poynard, Francoise Imbert-Bismnt, Vlad
Campbell,Joe Connolly, Katia Desnoyer, Fred Larsen, Susan Myerow, Wallace Neluati, Bayer Diagnostics, Walpole, MA; Janet Waldron, Susan Bromley, Johan Surtihadi, Bayer Diagnostics, Emeryville, CA The new VERSANTTM HCV RNA 3.0 Assay (bDNA) incorporates several recent improvements in branched DNA (bDNA) technology. The Capture and Target Probes, directed to the 5' untranslated and core regions of the HCV genome, specifically capture the viral RNA to the microwells and initiate formation of the bDNA complex. Completion of bDNA amplification is accomplished by sequential addition of Preamplifier and Amplifier probes. Alkaline phosphatase Label Probes are used with dioxetane substrate to generate chemiluminescent signal. The signal produced is directly related to the amount of target HCV RNA present in the sample, and is quantified by reference to a standard curve. The assay is semi-automated and is performed using the Bayer System 340 analyzer. The addition of preamplifier molecules to the bDNA complex increases the extent of signal amplification produced. The use of cruciform target probe configurations and incorporation of non-natural nucleosides iso5MEC and isoG in bDNA probes reduces non-specific hybridization. The effect of these improvements is a substantial increase in assay sensitivity, specificity and precision compared to the HCV 2.0 version. Up to 84 specimens cart be tested in singlicate on a 96 well microtiter plate for a maximum of 168 specimens per run, with HCV quantification values reported in copies/mL and international units IU/mL over a four log dynamic range. The VERSANTTM H C V RNA 3.0 Assay (bDNA) is currently available for Research Use Only in the US, not for use in diagnostic procedures. Key assay performance characteristics have been evaluated in predinical studies at Bayer Diagnostics. The studies used multiple kit lots that included component diversity. HCV-positive patient samples and an eight member HCV dilution panel were used to assess total assay precision, linearity, specificity and sensitivity. Total imprecision was less than 30% CV over the entire range of the assay. Assay linearity was within 0.1 log of expected values over the entire dynamic range of the assay. Specificity was at least 95% at the Reporting Threshold and sensitivity was at least 95% at the Limit of Quantitation. The assay quantified HCV genotypes la, lb, 2a, 2b, 3a, 4a, 5a and 6a equivalently. The potential interference of commonly prescribed drugs and pathogens likely to be present in patients with hepatitis C was assessed by spiking these agents into HCV positive and negative samples. No significant effect on assay performance was seen. Hemoglobin up to 200mg/dL, lipids up to 100mg/dL, bilirubin up to 60mg/dL and protein up to 9g/dL were also found to have no significant effect on assay performance when spiked into HCV negative and positive samples. These experimental observations will be further tested in clinical trials.
Ratziu, Groupe Hospitalier Pitie-Salpetriere, Paris France; Sylvie Chevret, Biostatistics HOpital St Louis, Paris France; Claude Jardel, Joseph Monssalli, Djamila Messons, Gronpe Hnspitalier Pitie-Salpetriere, Paris France; Francoise Degos, Hopital Beaujon, Ctichy France Background A liver fibrosis index was recently prospectively validated in a crosssectional study where patients infected by HCV had only one biopsy and no longitudinal follow-up. The aim of this study was to retrospectively assess the diagnostic value of this index in patients included in a randomized trial of interferon using repeated measurements, two biopsies and hyaluronic acid as a comparative reference. Methods 165 patients who had had two interpretable liver biopsies and at least one stored serum sample before interferon treatment were selected. 78 patients received interferon alpha 3 MU TIW for 24 weeks and 87 followed a reinforced regimen for 48 weeks. A fibrosis index combining five biochemical markers (alpha2macroglobulin, haptoglobin, apolipoproteinA1, gammaglutamyI transpeptidase, and total bilirubin adjusted for gender and age) as well as hyaluronic acid was assessed on 461 samples avaiIable at baseline, at the end of treatment and at the end of follow-up (72 weeks). Findings : There was a significant decrease of the fibrosis index score among the 17 sustained virologic responders, from 0.33 -+ 0.06 (mean + se) at baseline to 0.18 + 0.06 at 72 weeks in comparison with 92 non-responders (from 0.41 -+ 0.03 at baseline to 0.44 -+ 0.03 at 72 weeks; p<0.001) and in comparison with 56 relapsers (from 0.36 + 0.03 at baseline to 0.32 --_ 0.03 at 72 weeks; p=0.05). No significant differences were observed for hyaluronic acid. Interpretation: This fibrosis index could be used as a surrogate marker of the anti-fibrotic effect of treatments in patients with chronic hepatitis C.
Fibrosis Index 0.400'50I
o
0~0 ~
:"
020 j
-~i
000 ~
"~
c,
o Non Re~onders
'.
i Relapsers
~
!~
f ......t
Sustained Responders
Baseline 6too 12rno 24mo
Foil~-up
1565
1566
PERSISTENTLY NORMAL ALT DOES NOT PREDICT BENIGN HISTOLOGY IN METHADONE PATIENTS W I T H ACTIVE HEPATITIS C. Diana L
SERUM ALT LEVELS AS PREDICTORS OF CHRONIC HEPATITIS C IN URBAN PRIMARY CARE CLINIC POPULATIONS. Milton G Mutchnick,
Sylvestre, UCSF, Organization to Achieve Solutions in Substance-Abuse (OASIS,), Oakland, CA; Barry Clements, OASIS, Oakland, CA
Wayne State University School of Medicine, Detroit, MI; Paul H Naylor, Immuno-Rx, Inc, Farmingdale, NY;Joseph R Merline, MarkJ Upfal, Wayne State University School of Medicine, Detroit, MI
Background: Active hepatitis C patients with persistently normal ALT values are generally considered to have less severe hepatitis and less histologically advanced disease. Study Population: 50 HCV-exposed methadone maintenance patients with a positive PCR and liver biopsy within 3 years. 23/50 with at least 4 normal ALT values within a 1 year period were designated as having a persistently normal ALT. Results: Patients with persistently normal ALT values were more likely to be African-American (48% vs 22%), and somewhat more likely to be female (48% vs 41%) and older (52 yr vs 47 yr). They had similar estimated length of disease (28 yr vs 27 yr) and years of hea W alcohol consumption (12 yr). The average inflammation grade was 2.20 in the normal ALT subpopulation vs 2.30 in the elevated ALT group, and the average fibrosis stage was 2.0 (normal ALT) vs 2.25 (abnormal ALT). Twenty two percent of patients with a persistently normal ALT had a fibrosis stage of 3 or higher vs 37% in the abnormal ALT cohort. Conclusion: Methadone maintenance patients with active HCV exhibit relatively advanced liver disease overall, and nearly a quarter of those with persistently normal ALT and active HCV will present with severe fibrosis or cirrhosis. Caution should be used in interpreting liver disease on the basis of ALT in these patients, particularly those who are African-American, female, and older.
BACKGROUND: Chronic Hepatitis C (CHC) eludes detection in part because it damages the liver gradually with minimal impact on liver enzymes. As a consequence, many patients with hepatitis C have "normal" ALT values during the course of their infection. Currently, the utility of measuring serum levels of this enzyme as a predictor of CHC remains to be determined. AIMS: The goal of this study was to define ALT levels in CHC patients identified in an urban primary care setting and to compare them to levels seen in non-infected patients, as well as the overall population in the clinic for whom an ALT test was ordered. METHODS: Existing laboratory data on all patients in the urban general medicine practice at Wayne State University School of Medicine were examined to review all ALT and Hepatitis C antibody tests that had been performed during a three-month period. In our laboratory, vatues greater than 65 units/dl are flagged as elevated. RESULTS: At least one test for serum ALT was performed in 2725 individuals with a range of 1-6000 units/dl (median 57.1). A hepatitis C antibody test was performed in 259 patients with a positive rate of 32.8%. Of those positive patients (n = 85), the average ALT was 95.3 units/dl as compared to the anti-HCV negative individuals (n = 174) whose average was 244.6 units/dl (p = 0.03). Using the laboratory cut-off of 65 units/dl, 44.7% of the anti-HCV positive patients had elevated ALT. This compared to 45.4% for the anti-HCV negative patients and 21.4% for the total population (n = 2725). IfALT of 50 units/dl was used to define elevated levels, 57.6% of the anti-HCV positive patients had at least one elevated ALT compared to 30.9% of the total patient population. If the ALT was 40 nnits/dl or above, then 39.9% of the tested patients were positive for Hepatitis C compared to 21.8% of the total patient population. CONCLUSIONS: In this patient population, auti-HCV was two fold more common among those whose ALT levels exceeded 40 units/dl as compared to all patients tested for ALT and who had levels less than 40 units/dl. These data appear to support hepatitis C testing for any patient for whom an ALT is greater than 40 units/dl and may in fact support testing in all urban patients for whom an ALT is ordered. Additional investigation of larger populations, with simultaneous examination of the interaction between ALT screening thresholds and risk factors, would assist in the development of best practices in screening for Hepatitis C.
AASLD ABSTRACTS
563A
1563
1564
PRECLINICAL EVALUATION OF THE VERSANT HCV RNA 3.0 ASSAY (BDNA). Andy Worlock, John Arruda, Tonya Black, Susan Bonapace, Julie
BIOCHEMICAL MARKERS OF LIVER FIBROSIS IN PATIENTS INFECTED BY HEPATITIS C VIRUS: LONGITUDINAL VALIDATION IN A RANDOMIZED TRIAL. Thierry Poynard, Francoise Imbert-Bismnt, Vlad
Campbell,Joe Connolly, Katia Desnoyer, Fred Larsen, Susan Myerow, Wallace Neluati, Bayer Diagnostics, Walpole, MA; Janet Waldron, Susan Bromley, Johan Surtihadi, Bayer Diagnostics, Emeryville, CA The new VERSANTTM HCV RNA 3.0 Assay (bDNA) incorporates several recent improvements in branched DNA (bDNA) technology. The Capture and Target Probes, directed to the 5' untranslated and core regions of the HCV genome, specifically capture the viral RNA to the microwells and initiate formation of the bDNA complex. Completion of bDNA amplification is accomplished by sequential addition of Preamplifier and Amplifier probes. Alkaline phosphatase Label Probes are used with dioxetane substrate to generate chemiluminescent signal. The signal produced is directly related to the amount of target HCV RNA present in the sample, and is quantified by reference to a standard curve. The assay is semi-automated and is performed using the Bayer System 340 analyzer. The addition of preamplifier molecules to the bDNA complex increases the extent of signal amplification produced. The use of cruciform target probe configurations and incorporation of non-natural nucleosides iso5MEC and isoG in bDNA probes reduces non-specific hybridization. The effect of these improvements is a substantial increase in assay sensitivity, specificity and precision compared to the HCV 2.0 version. Up to 84 specimens cart be tested in singlicate on a 96 well microtiter plate for a maximum of 168 specimens per run, with HCV quantification values reported in copies/mL and international units IU/mL over a four log dynamic range. The VERSANTTM H C V RNA 3.0 Assay (bDNA) is currently available for Research Use Only in the US, not for use in diagnostic procedures. Key assay performance characteristics have been evaluated in predinical studies at Bayer Diagnostics. The studies used multiple kit lots that included component diversity. HCV-positive patient samples and an eight member HCV dilution panel were used to assess total assay precision, linearity, specificity and sensitivity. Total imprecision was less than 30% CV over the entire range of the assay. Assay linearity was within 0.1 log of expected values over the entire dynamic range of the assay. Specificity was at least 95% at the Reporting Threshold and sensitivity was at least 95% at the Limit of Quantitation. The assay quantified HCV genotypes la, lb, 2a, 2b, 3a, 4a, 5a and 6a equivalently. The potential interference of commonly prescribed drugs and pathogens likely to be present in patients with hepatitis C was assessed by spiking these agents into HCV positive and negative samples. No significant effect on assay performance was seen. Hemoglobin up to 200mg/dL, lipids up to 100mg/dL, bilirubin up to 60mg/dL and protein up to 9g/dL were also found to have no significant effect on assay performance when spiked into HCV negative and positive samples. These experimental observations will be further tested in clinical trials.
Ratziu, Groupe Hospitalier Pitie-Salpetriere, Paris France; Sylvie Chevret, Biostatistics HOpital St Louis, Paris France; Claude Jardel, Joseph Monssalli, Djamila Messons, Gronpe Hnspitalier Pitie-Salpetriere, Paris France; Francoise Degos, Hopital Beaujon, Ctichy France Background A liver fibrosis index was recently prospectively validated in a crosssectional study where patients infected by HCV had only one biopsy and no longitudinal follow-up. The aim of this study was to retrospectively assess the diagnostic value of this index in patients included in a randomized trial of interferon using repeated measurements, two biopsies and hyaluronic acid as a comparative reference. Methods 165 patients who had had two interpretable liver biopsies and at least one stored serum sample before interferon treatment were selected. 78 patients received interferon alpha 3 MU TIW for 24 weeks and 87 followed a reinforced regimen for 48 weeks. A fibrosis index combining five biochemical markers (alpha2macroglobulin, haptoglobin, apolipoproteinA1, gammaglutamyI transpeptidase, and total bilirubin adjusted for gender and age) as well as hyaluronic acid was assessed on 461 samples avaiIable at baseline, at the end of treatment and at the end of follow-up (72 weeks). Findings : There was a significant decrease of the fibrosis index score among the 17 sustained virologic responders, from 0.33 -+ 0.06 (mean + se) at baseline to 0.18 + 0.06 at 72 weeks in comparison with 92 non-responders (from 0.41 -+ 0.03 at baseline to 0.44 -+ 0.03 at 72 weeks; p<0.001) and in comparison with 56 relapsers (from 0.36 + 0.03 at baseline to 0.32 --_ 0.03 at 72 weeks; p=0.05). No significant differences were observed for hyaluronic acid. Interpretation: This fibrosis index could be used as a surrogate marker of the anti-fibrotic effect of treatments in patients with chronic hepatitis C.
Fibrosis Index 0.400'50I
o
0~0 ~
:"
020 j
-~i
000 ~
"~
c,
o Non Re~onders
'.
i Relapsers
~
!~
f ......t
Sustained Responders
Baseline 6too 12rno 24mo
Foil~-up
1565
1566
PERSISTENTLY NORMAL ALT DOES NOT PREDICT BENIGN HISTOLOGY IN METHADONE PATIENTS W I T H ACTIVE HEPATITIS C. Diana L
SERUM ALT LEVELS AS PREDICTORS OF CHRONIC HEPATITIS C IN URBAN PRIMARY CARE CLINIC POPULATIONS. Milton G Mutchnick,
Sylvestre, UCSF, Organization to Achieve Solutions in Substance-Abuse (OASIS,), Oakland, CA; Barry Clements, OASIS, Oakland, CA
Wayne State University School of Medicine, Detroit, MI; Paul H Naylor, Immuno-Rx, Inc, Farmingdale, NY;Joseph R Merline, MarkJ Upfal, Wayne State University School of Medicine, Detroit, MI
Background: Active hepatitis C patients with persistently normal ALT values are generally considered to have less severe hepatitis and less histologically advanced disease. Study Population: 50 HCV-exposed methadone maintenance patients with a positive PCR and liver biopsy within 3 years. 23/50 with at least 4 normal ALT values within a 1 year period were designated as having a persistently normal ALT. Results: Patients with persistently normal ALT values were more likely to be African-American (48% vs 22%), and somewhat more likely to be female (48% vs 41%) and older (52 yr vs 47 yr). They had similar estimated length of disease (28 yr vs 27 yr) and years of hea W alcohol consumption (12 yr). The average inflammation grade was 2.20 in the normal ALT subpopulation vs 2.30 in the elevated ALT group, and the average fibrosis stage was 2.0 (normal ALT) vs 2.25 (abnormal ALT). Twenty two percent of patients with a persistently normal ALT had a fibrosis stage of 3 or higher vs 37% in the abnormal ALT cohort. Conclusion: Methadone maintenance patients with active HCV exhibit relatively advanced liver disease overall, and nearly a quarter of those with persistently normal ALT and active HCV will present with severe fibrosis or cirrhosis. Caution should be used in interpreting liver disease on the basis of ALT in these patients, particularly those who are African-American, female, and older.
BACKGROUND: Chronic Hepatitis C (CHC) eludes detection in part because it damages the liver gradually with minimal impact on liver enzymes. As a consequence, many patients with hepatitis C have "normal" ALT values during the course of their infection. Currently, the utility of measuring serum levels of this enzyme as a predictor of CHC remains to be determined. AIMS: The goal of this study was to define ALT levels in CHC patients identified in an urban primary care setting and to compare them to levels seen in non-infected patients, as well as the overall population in the clinic for whom an ALT test was ordered. METHODS: Existing laboratory data on all patients in the urban general medicine practice at Wayne State University School of Medicine were examined to review all ALT and Hepatitis C antibody tests that had been performed during a three-month period. In our laboratory, vatues greater than 65 units/dl are flagged as elevated. RESULTS: At least one test for serum ALT was performed in 2725 individuals with a range of 1-6000 units/dl (median 57.1). A hepatitis C antibody test was performed in 259 patients with a positive rate of 32.8%. Of those positive patients (n = 85), the average ALT was 95.3 units/dl as compared to the anti-HCV negative individuals (n = 174) whose average was 244.6 units/dl (p = 0.03). Using the laboratory cut-off of 65 units/dl, 44.7% of the anti-HCV positive patients had elevated ALT. This compared to 45.4% for the anti-HCV negative patients and 21.4% for the total population (n = 2725). IfALT of 50 units/dl was used to define elevated levels, 57.6% of the anti-HCV positive patients had at least one elevated ALT compared to 30.9% of the total patient population. If the ALT was 40 nnits/dl or above, then 39.9% of the tested patients were positive for Hepatitis C compared to 21.8% of the total patient population. CONCLUSIONS: In this patient population, auti-HCV was two fold more common among those whose ALT levels exceeded 40 units/dl as compared to all patients tested for ALT and who had levels less than 40 units/dl. These data appear to support hepatitis C testing for any patient for whom an ALT is greater than 40 units/dl and may in fact support testing in all urban patients for whom an ALT is ordered. Additional investigation of larger populations, with simultaneous examination of the interaction between ALT screening thresholds and risk factors, would assist in the development of best practices in screening for Hepatitis C.