COBAS TAQMAN HBV TEST

{0} # Summary of Safety and Effectiveness Data ## I. GENERAL INFORMATION: Device Generic Name: Nucleic acid assay for detection of Hepatitis B Virus (HBV) DNA Device Trade Name: COBAS TaqMan HBV Test For Use With The High Pure System Name and Address of Applicant: Roche Molecular Systems, Inc. (RMS) 4300 Hacienda Drive Pleasanton, CA 94588 Date of Panel Recommendation: None Premarket Approval Application (PMA) Number: P050028 Date of Notice of Approval to the Applicant: September 04, 2008 ## II. INDICATIONS FOR USE: The COBAS TaqMan HBV Test For Use With The High Pure System is an in vitro nucleic acid amplification test for the quantitation of Hepatitis B Virus (HBV) DNA in human serum or plasma (EDTA), using the High Pure Viral Nucleic Acid Kit for manual specimen preparation and the COBAS TaqMan 48 Analyzer for automated amplification and detection. The test is intended for use as an aid in the management of patients with chronic HBV infection undergoing anti-viral therapy. The assay can be used to measure HBV DNA levels at baseline and during treatment to aid in assessing response to treatment. The results from the COBAS TaqMan HBV Test must be interpreted within the context of all relevant clinical and laboratory findings. Assay performance characteristics have been established for individuals treated with adefovir dipivoxil. Assay performance for determining the state of HBV infection has not been established. The COBAS TaqMan HBV Test is not intended for use as a screening test for blood or blood products for the presence of HBV or as a diagnostic test to confirm the presence of HBV infection. {1} Page - 2 Summary of Safety and Effectiveness Data III. CONTRAINDICATIONS: None known. IV. WARNINGS AND PRECAUTIONS: For *in vitro* diagnostic use only. The warnings and precautions for the COBAS TaqMan HBV Test For Use With The High Pure System are stated in the respective product labeling. V. DEVICE DESCRIPTION: *Kit Configurations and Components* The COBAS TaqMan HBV Test For Use With The High Pure System consists of the following kits: - High Pure System Viral Nucleic Acid Kit - COBAS TaqMan HBV Test Kit Configuration Each kit contains labeled reagents assembled according to storage temperature requirements and controlled room temperature. *A High Pure System Viral Nucleic Acid Kit (HPS)* is a generic manual specimen preparation kit, for the isolation and purification of nucleic acids. Control, amplification, and detection reagents are provided separately in the COBAS TaqMan (CTM) HBV Test kit. The High Pure System Viral Nucleic Acid Kit contains the reagents and kit-specific disposables necessary for specimen and control processing. The following reagents and disposables are supplied in this kit: LYS (Lysis/Binding Buffer): A Tris buffered solution containing guanidine hydrochloride, urea, and Triton X-100; 2 x 25 mL CAR (RNA, lyophilized): Poly (A) carrier RNA is a lyophilized powder consisting of single stranded synthetic polymer, polyadenylic acid; 2 x 2 mg IRB (Inhibitor Removal Buffer): A Tris buffered solution containing guanidine hydrochloride to which is added 80 mL of ethanol; 1 x 33 mL WASH (Wash Buffer): A Tris buffered solution containing sodium chloride to which is added 20 mL of ethanol; 1 x 20 mL ELB (Elution Buffer): PCR grade water; 1 x 30 mL Disposables: RS (High Pure system Viral Nucleic Acid Rack Set) 4 x each; WR (High Pure System Viral Nucleic Acid Waste Rack) 8 x each *The CTM HBV Test Specimen Preparation and Control Reagents* consist of an HBV Quantitation Standard (QS) DNA, two HBV DNA positive controls, and a negative control. The HBV QS and HBV positive controls have assigned HBV DNA expected ranges. The positive controls are prepared in a matrix of negative human plasma. The HBV QS is added, at a known quantity, to each specimen at 9 {2} Page - 3 Summary of Safety and Effectiveness Data the beginning of the specimen preparation procedure. The addition of HBV QS allows for the calculated titer of HBV target DNA to be adjusted, accordingly, if the fluorescence of the HBV QS is delayed due to inhibition or poor sample recovery. The following Specimen Preparation and Control Reagents are provided in the CTM HBV Test kit: HBV QS (HBV Quantitation Standard): Tris-HCl buffer, EDTA, &lt;0.001% linearized, double stranded plasmid DNA containing an insert. The DNA insert contains HBV primer binding sequences and a unique probe binding region. Amaranth dye, &lt; 0.005% Poly rA RNA (synthetic), 0.05% Sodium azide, urea, and Triton X-100; 2 x 1.0 mL HBV H(+)C [HBV High (+) Control]: &lt;0.001% linearized, double stranded plasmid DNA containing HBV sequences. Negative Human Plasma, non-reactive by US FDA licensed tests for antibody to HCV, antibody to HIV-1/2, and HBsAg; HIV-1 RNA, HCV RNA and HBV DNA not detectable by PCR methods, 0.1% Proclin 300; 2 x 1.0 mL HBV L(+)C [HBV Low (+) Control]: &lt;0.001% linearized, double stranded plasmid DNA containing HBV sequences. Negative Human Plasma, non-reactive by US FDA licensed tests for antibody to HCV, antibody to HIV-1/2, and HBsAg; HIV-1 RNA, HCV RNA and HBV DNA not detectable by PCR methods, 0.1% Proclin 300; 2 x 1.0 mL CTM (-) C [COBAS TaqMan Negative Control (Human Plasma)]: Negative Human Plasma, non-reactive by US FDA licensed tests for antibody to HCV, antibody to HIV-1/2, and HBsAg; HIV-1 RNA, HCV RNA and HBV DNA not detectable by PCR methods, 0.1% Proclin 300; 4 x 1.0 mL The COBAS TaqMan HBV Test Amplification and Detection Reagents include reagents necessary to perform PCR amplification and detection of HBV and HBV QS target DNA. Additionally, these reagents include the manganese solution required as a co-factor for the DNA polymerase activity of the ZO5 DNA polymerase. The following Amplification and Detection Reagents are provided in the CTM HBV Test kit: HBV MMX (COBAS TaqMan HBV Master Mix): Tricene buffer, Potassium hydroxide, Potassium acetate, Glycerol, &lt; 0.001% dATP, dCTP, dGTP, dUTP, &lt; 0.001% Upstream and downstream primers to the Pre-core/core region of HBV, &lt; 0.001% Fluorescent-labeled oligonucleotide probes specific for HBV and the HBV Quantitation Standard, &lt; 0.001% Oligonucleotide aptamer, &lt;0.05% ZO5 DNA Polymerase (microbial), &lt; 0.1% AmpErase (uracil-N-glycolase) enzyme (microbial), 0.09% Sodium azide; 2 x 1.4 mL HBV Mn2+ (COBAS TaqMan Manganese Solution): &lt; 0.2% Manganese acetate, Glacial acetic acid, 0.09% Sodium azide; 2 x 1.0 mL ## Assay Principle and Format The COBAS TaqMan HBV Test is based on two major processes: (1) manual specimen preparation to obtain HBV DNA by lysis of viral particles followed by precipitation of the HBV DNA with alcohol and adsorption of the HBV DNA to {3} Page - 4 Summary of Safety and Effectiveness Data glass fibers; (2) automated PCR amplification of target DNA using HBV specific complementary primers, and detection of cleaved dual fluorescent dye-labeled oligonucleotide detection probes that permit quantitation of HBV target amplified product (amplicon) and HBV Quantitation Standard (QS) DNA, which is processed, amplified, and detected simultaneously with the specimen. The HBV QS is incorporated into each individual specimen and control at a known copy number and is carried through the specimen preparation, PCR amplification and detection steps along with the HBV target. The COBAS TaqMan 48 (CTM 48) Analyzer calculates the HBV DNA titer in the test specimen by comparing the HBV signal to the HBV QS signal for each specimen and control. The Master Mix reagent contains primer pairs and probes specific for both HBV DNA and HBV QS DNA. The Master Mix has been developed to ensure equivalent quantitation of genotypes A through G of HBV. The detection of amplified DNA is performed using target-specific and QS-specific dual labeled oligonucleotide probes that permit independent identification of HBV amplicon and HBV QS amplicon. The quantitation of HBV viral DNA is performed using the HBV QS. The HBV QS is a non-infectious, linearized, double stranded plasmid that contains the identical primer binding sites as the HBV DNA target and a unique probe binding region that allows HBV QS amplicon to be distinguished from HBV target amplicon. The HBV QS is incorporated into each individual specimen and control at a known copy number and is carried through the specimen preparation, PCR amplification and detection steps along with the HBV target. The CTM 48 Analyzer calculates the HBV DNA titer in the test specimen by comparing the HBV signal to the HBV QS signal for each specimen and control. The HBV QS compensates for effects of inhibition and controls for the preparation and amplification processes to allow the accurate quantitation of HBV DNA in each specimen. The CTM 48 Analyzer is a flexible bench top, batch analyzer that automates the incubation, timing, thermal cycling, and real time photometric measurement of the PCR process into a single analytical system. The primary operational components of the analyzer are the thermal cycler, incubator, and photometric detection systems. The COBAS TaqMan HBV Test provides quantitative results over a very wide dynamic range since the monitoring of amplicon is performed during the exponential phase of amplification. The higher the HBV titer of a specimen, the earlier the fluorescence of the reporter dye of the HBV probe rises above the baseline fluorescence level. Since the amount of HBV QS DNA is constant between all specimens, the fluorescence of the reporter dye of the HBV QS probe should appear at the same cycle for all specimens. In cases where the QS amplification and detection is affected by inhibition or poor specimen recovery, the appearance of fluorescence will be delayed, thereby enabling the calculated titer of HBV target DNA to be adjusted accordingly. The appearance of the {4} Page - 5 Summary of Safety and Effectiveness Data specific fluorescent signal is reported as a critical threshold value (Ct). The Ct is defined as the fractional cycle number where reporter dye fluorescence exceeds a predetermined threshold (the Assigned Fluorescence Level), and starts the beginning of an exponential growth phase of this signal. A higher Ct value indicates a lower titer of initial HBV target DNA. A 2-fold increase in titer correlates with a decrease of 1 Ct for target HBV DNA, while a 10-fold increase in titer correlates with a decrease of 3.3 Ct. As the concentration of the virus increases, the growth curves shift to earlier cycles. Therefore the leftmost growth curve corresponds to the highest viral titer level whereas the rightmost growth curve corresponds to the lowest viral titer level. Target Growth Curves for a Dilution Series of Virus Spanning over a 5-log₁₀ Range:  The amount of Quantitation Standard added to each specimen is constant for each reaction. The Ct value of the Quantitation Standard is similar regardless of the viral titer. Quantitation Standard Growth Curves for a Dilution Series of Virus Spanning over a 5-log₁₀ Range: {5} Page - 6 Summary of Safety and Effectiveness Data  The fractional cycle number (Ct) is calculated where the fluorescence signal crosses the Assigned Fluorescence Level. Fluorescence Values at Every Cycle are Normalized for Each Growth Curve:  HBV DNA Quantitation: The COBAS TaqMan HBV Test quantitates HBV viral DNA by utilizing a second target sequence (HBV Quantitation Standard) that is added to each test specimen at a known concentration. The HBV Quantitation Standard is a non-infectious, linearized, double stranded plasmid DNA construct, containing fragments of HBV sequences with primer binding regions identical to those of the HBV target sequence. The HBV Quantitation Standard also generates an amplification product of the same length and base composition as the HBV 13 {6} Page - 7 Summary of Safety and Effectiveness Data target DNA. The detection probe binding region of the HBV Quantitation Standard has been modified to differentiate HBV Quantitation Standard amplicon from HBV target amplicon. During the annealing phase of the PCR on the COBAS TaqMan 48 Analyzer, the specimens are illuminated and excited by filtered light and filtered emission fluorescence data are collected for each specimen. The readings from each specimen are then corrected for instrumental fluctuations. These fluorescence readings are sent by the instrument to the AMPLILINK software and stored in a database. Pre-Checks are used to determine if the HBV DNA and HBV Quantitation Standard DNA data represent sets that are valid, and flags are generated when the data lie outside the preset limits. After all Pre-Checks are completed and passed, the fluorescence readings are processed to generate Ct values for the HBV DNA and the HBV Quantitation Standard DNA. The lot-specific calibration constants provided with the COBAS TaqMan HBV Test are used to calculate the titer value for the specimens and controls based upon the HBV DNA and HBV Quantitation Standard DNA Ct values. The COBAS TaqMan HBV Test is standardized against the WHO International Standard for Hepatitis B Virus DNA for Nucleic Acid Technology (NAT) Assays Testing NIBSC 97/746 and titer results are reported in International Units (IU/mL). ## Results The COBAS TaqMan 48 Analyzer automatically determines the HBV DNA titer for the specimen or control. The HBV DNA titer is expressed in International Units (IU)/mL. The conversion factor between HBV copies/mL and HBV IU/mL is 5.82 copies/IU based on linkage to the WHO International Standard for HBV DNA Nucleic Acid Amplification Technology (NAT) Assays, NIBSC Code 97/746. ## The COBAS TaqMan 48 Analyzer: - Determines the Cycle Threshold value (Ct) for the HBV DNA and the HBV Quantitation Standard DNA. - Determines the HBV DNA titer based upon the Ct values for the HBV DNA and HBV Quantitation Standard DNA and the lot-specific calibration coefficients. - Determines that the calculated IU/mL titers for HBV L(+)C and HBV H(+)C fall within the assigned ranges stated on the Controls Value Card supplied with the kit. Run Validation: The run is valid if no flags appear for the COBAS TaqMan HBV Controls. ## Interpretation of Results For a valid run, each individual specimen should be checked for flags or comments on the result printout. A valid run may include both valid and invalid specimen results depending on whether flags and/or comments are obtained for the individual specimens. Specimen results are interpreted as follows: {7} Page - 8 Summary of Safety and Effectiveness Data | Titer Result | Interpretation | | --- | --- | | Target Not Detected | No Ct value for HBV obtained. Report results as "HBV DNA not detected". | | <2.9E+01 IU/mL | Below 2.9E+01 IU/mL (lower limit of quantitation, LLoQ). HBV DNA is not quantifiable. | | ≥2.9E+01 IU/mL and ≤1.1E+08 IU/mL | Calculated results greater than or equal to 29 IU/mL and less than or equal to 1.1E+08 IU/mL are within the Linear Range of the Assay. | | > 1.10E+08 IU/mL | IU/mL are above the Linear Range of the assay. Report results as "greater than 1.10E+08 HBV DNA IU/mL". If quantitative results are desired, the original specimen should be diluted 1:100 with HBV-negative human plasma or serum depending upon the matrix of the original specimen (plasma samples must be diluted in plasma and serum samples must be diluted in serum), and the test repeated. Multiply the reported result by the dilution factor. | $1.1E+08\ IU/mL = 1.1 \times 10^{8}\ IU/mL$ Note: In some rare instances, specimens with very high titers can produce an invalid result with a flag “QS_INVALID.” The result for these samples is not valid and must be repeated. If the sample continues to produce an invalid result the user can dilute the sample 1:100 in HBV negative plasma (for plasma samples) or HBV negative serum (for serum samples) in order to try to obtain a valid result. Multiply the reported result by the dilution factor. ## VI. ALTERNATIVE PRACTICES AND PROCEDURES Currently, methods for following the progress of antiviral therapy include immunoassay (serological tests, EIA), biochemical (ALT), and histological (liver biopsy - fibrosis, inflammation). There has not been an adequate molecular method commercially available to follow HBV DNA response to antiviral therapy during the course of treatment. ## VII. MARKETING HISTORY The COBAS TaqMan HBV Test For Use With The High Pure Viral Nucleic Acid System has not been withdrawn from the following markets for reasons related to safety or effectiveness. The test is currently available in the following countries: | Argentina | Iceland | Pakistan | | --- | --- | --- | | Australia | India | Paraguay | | Austria | Indonesia | Philippines | | Belarus | Ireland | Poland | | Belgium | Israel | Qatar | | Canada | Japan | Russia | | Chile | Korea | Saudi Arabia | | Cyprus | Kuwait | Slovenia | {8} Page - 9 Summary of Safety and Effectiveness Data | Czech Republic | Lebanon | South Africa | | --- | --- | --- | | Denmark | Luxembourg | Spain | | Ecuador | Malaysia | Sweden | | Egypt | Malta | Switzerland | | Finland | Mexico | Syria | | France | Netherlands | Turkey | | Germany | New Zealand | United Arab Emirates | | Greece | Norway | United Kingdom | | Hong Kong | Oman | Venezuela | | Hungary | | | ## VIII. POTENTIAL ADVERSE EFFECTS OF THE DEVICE ON HEALTH To aid in the management of patients with chronic HBV infection undergoing anti-viral therapy, the results from the COBAS TaqMan HBV Test must be interpreted within the context of all relevant clinical and laboratory findings. Since the COBAS TaqMan HBV Test is for *in vitro* diagnostic use, there is no direct adverse effect on the health of the patient. However, failure of the product to perform as indicated or human error in use of the product may lead to a false result in the assessment of a response to HBV antiviral therapy, erroneously indicating the need for unnecessary change in treatment. The assay is not intended for use as a screening test for blood or blood products for the presence of HBV or as a diagnostic test to confirm the presence of HBV infection. Assay performance characteristics have been established for individuals treated with adefovir dipivoxil. Assay performance for determining the state of HBV infection has not been established. ## IX. SUMMARY OF NONCLINICAL STUDIES ### A. Laboratory Studies #### Traceability to the WHO Standard Several standards and controls have been used to provide traceability to the WHO Standard. This includes the Eurohep R1 standard, the WHO Standard, and RMS HBV Calibrators. The Eurohep R1 standard, and the CTM HBV test calibrators behave similarly relative to the WHO standard with a deviation not more than 0.1 $\log_{10}$. {9} Page - 10 Summary of Safety and Effectiveness Data Traceability of the COBAS TaqMan HBV Test For Use With The High Pure System HBV Calibrators (2-7 on $\log_{10}$) to the WHO Standard (range 2-5 on $\log_{10}$) and Eurohep Standard R1 (range 2-7 on $\log_{10}$):  ## Linear Range The linear range study was evaluated in accordance with the methods defined in the CLSI Guideline EP6-A, “Evaluation of the Linearity of Quantitative Measurement Procedures: A Statistical Approach; Approved Guideline.” The COBAS TaqMan HBV Test For Use With the High Pure System was found to give a linear response from 2.9E1 ($\log_{10} = 1.46$) HBV DNA IU/mL to 1.10E8 ($\log_{10} = 8.02$) HBV DNA IU/mL in both EDTA plasma and serum with deviation from linearity not more than $0.20 \log_{10}$, in both matrices. The study was performed using two lots of COBAS TaqMan HBV Test For Use With the High Pure System reagents and serial dilutions of an HBV positive genotype A specimen that was assigned relative to the WHO Standard. Twenty-seven replicates were tested per level in EDTA plasma and in serum. {10} Page - 11 Summary of Safety and Effectiveness Data Linearity of the COBAS TaqMan HBV Test For Use With The High Pure System in EDTA Plasma for Genotype A:  Linearity of the COBAS TaqMan HBV Test For Use With The High Pure System in Serum for Genotype A: {11} Page - 12 Summary of Safety and Effectiveness Data  Linearity for genotypes other than genotype A was not evaluated. The analytical measurement range of analyte values that can be directly measured on a specimen without any dilution using the COBAS TaqMan HBV Test is 29 to $1.1\mathrm{E} + 08\mathrm{IU / mL}$. The clinically reportable range of analyte values that can be measured on a sample with a maximum dilution of one to one-hundred using the COBAS TaqMan HBV Test is 29 to $1.1\mathrm{E} + 10\mathrm{IU / mL}$. ## Inclusivity (Genotype Titer Quantitation) The performance of the COBAS TaqMan HBV Test on HBV genotypes was evaluated by analysis of 23 purified, linearized and quantitated plasmid DNAs containing representative sequence inserts from HBV genotypes A through G). In addition, a plasmid representing the common pre-Core mutation G:A at nucleotide position 1896 was tested. Each plasmid DNA was diluted to concentrations of 5.2E1, 5.2E2, 5.2E5 and 5.2E8 IU/mL. Each dilution was co-amplified with HBV QS DNA and analyzed in triplicate with the COBAS TaqMan HBV Test. The titers for all plasmids were compared with that of a control plasmid DNA. COBAS TaqMan HBV Test Inclusivity Testing - Typed Plasmid DNA Tested: | Plasmid Designation | Genotype | Parent Specimen Origin | | --- | --- | --- | | p8423-c1 | A | India | | p1115-c1 | A | Burundi | | p3952-c1 | A | Cameroon | | p4199-c2 | A | Norway | {12} Page - 13 Summary of Safety and Effectiveness Data | p1764-c1 | B | China | | --- | --- | --- | | p1767-c1 | B | China | | p3958-c1 | B | East Asia | | p830-c1 | B | Societe Island | | p3982-c1 | B | Vietnam | | p1786-c1 | C | China | | p11549-1 | C | Bangladesh | | p3872-c1 | D | Iran | | p1103-c1 | D | Tunisia | | p3953-c2 | D | North Africa | | p18-c1 | D | Sweden | | p30893-5 | D | Sweden | | p4244-c1 | D | Denmark | | p3217-c1 | E | Senegal | | p3963-c2 | E | Nigeria | | p9203-c1 | F | Colombia | | p479-c1 | F | Venezuela | | p1009-c1 | F | Spain | | p00042975-4 | G | United States | | pIT1896 | Pre-Core | Italy | The COBAS TaqMan HBV Test performance on all 23 plasmids and DNA copy numbers for all genotypes and the pre-Core mutation, and agreement with each other and the control plasmid DNA, can be seen from the of the COBAS TaqMan HBV Test results on HBV Genotypes A through G and a Pre-Core Mutant:  20 {13} Page - 14 Summary of Safety and Effectiveness Data # Limit of Detection Using Clinical Specimens Across All HBV Genotypes The limit of detection was determined in accordance with the methods defined in the CLSI Guideline EP17-A, "Protocols for Determination of Limits of Detection and Limits of Quantitation; Approved Guideline." The LoD was determined using two lots of reagents for seven clinical specimens of HBV representing genotypes A through G diluted into both EDTA plasma and serum. One representative clinical sample of each genotype was tested. The HBV titer for each parent specimen was provided by the vendor or determined in-house. Dilution panels were generated from these final titer assignments. Each panel consisted of six members representing input levels at 15, 10, 8, 6, 4, and 1 IU/mL. Each level of each dilution was tested with 16 replicates split across two runs for each of two reagent lots for each genotype specimen in each matrix across eight days. A total of 32 replicates of each panel member were tested for each genotype in each matrix. The hit rate was determined at each input level and the Limit of Detection is defined as the lowest level demonstrating a ≥95% hit rate and where all higher input levels have ≥95% hit rate. The LoD for the various HBV genotypes for the EDTA plasma and serum is summarized in the table below. The data shown represents the combined results from testing with two lots of reagents: | Genotype | EDTA Plasma | | Serum | | | --- | --- | --- | --- | --- | | | LOD, IU/mL | Hit Rate (95% Confidence Limits) | LOD, IU/mL | Hit Rate (95% Confidence Limits) | | A | 10 | 100% (89.1 – 100%) | 4 | 100% (89.1 – 100%) | | B | 4 | 97% (83.8 – 99.9%) | 4 | 100% (89.1 – 100%) | | C | 4 | 97% (83.8 – 99.9%) | 4 | 100% (89.1 – 100%) | | D | 4 | 100% (89.1 – 100%) | 4 | 97% (83.8 – 99.9%) | | E | 4 | 97% (83.8 – 99.9%) | 4 | 100% (89.1 – 100%) | | F | 4 | 100% (89.1 – 100%) | 4 | 100% (89.1 – 100%) | | G | 6 | 97% (83.8 – 99.9%) | 4 | 100% (89.1 – 100%) | The LoD for the COBAS TaqMan HBV Test, detecting any of the seven genotypes tested, was determined to be 10 IU/mL. No significant difference between the seven genotypes was observed. In addition, there was no significant difference between plasma and serum or between the two lots of reagents tested. Considering that COBAS TaqMan HBV Test does not differentiate between HBV genotypes, the overall LoD of the assay to detect HBV in clinical specimens is determined to be 10 IU/mL. # Limit of Detection Using the WHO International Standard The WHO Standard was freshly diluted into each of five unique EDTA plasma specimens and each of five unique serum specimens. Each level of each dilution {14} Page - 15 Summary of Safety and Effectiveness Data was tested with six replicates split across two runs for each of two reagent lots for each matrix. A total of 10 runs were conducted over five days for each reagent lot for each matrix to give a total of 60 replicates for each level for each matrix. These studies demonstrate that the COBAS TaqMan HBV Test can detect HBV DNA in EDTA plasma and serum at concentrations as low as 10 IU/mL with a positivity rate greater than 95%. The concentration of HBV DNA using the WHO international standard in EDTA plasma and serum that can be detected with a positivity rate of greater than 95% as determined by Probit Analysis, is 3.5 IU/mL and 3.4 IU/mL, respectively. Limit of Detection in EDTA Plasma of the COBAS TaqMan HBV Test For Use With the High Pure System using the WHO International Standard (Genotype A): | WHO Standard Based Concentration (HBV DNA IU/mL) | No. Valid Replicates | No. Positives | Positivity Rate | | --- | --- | --- | --- | | 29.0 | 59 | 59 | 100% | | 22.0 | 59 | 59 | 100% | | 15.0 | 60 | 60 | 100% | | 12.0 | 59 | 59 | 100% | | 10.0 | 59 | 59 | 100% | | 8.0 | 60 | 59 | 98% | | 6.0 | 60 | 59 | 98% | | 4.0 | 60 | 59 | 98% | | 2.0 | 60 | 51 | 85% | | 1.0 | 60 | 37 | 62% | | 0.5 | 60 | 23 | 38% | | 0.0 | 60 | 0 | 0% | | Probit 95% Hit Rate | 3.5 IU/mL [95% confidence limits of 2.8 – 4.7 IU/mL] | | | Limit of Detection in Serum of the COBAS TaqMan HBV Test For Use With the High Pure System using the WHO International Standard (Genotype A): | WHO Standard Based Concentration (HBV DNA IU/mL) | No. Valid Replicates | No. Positives | Positivity Rate | | --- | --- | --- | --- | | 29.0 | 59 | 59 | 100% | | 22.0 | 59 | 59 | 100% | | 15.0 | 60 | 60 | 100% | | 12.0 | 60 | 60 | 100% | | 10.0 | 59 | 59 | 100% | | 8.0 | 59 | 58 | 98% | {15} Page - 16 Summary of Safety and Effectiveness Data | 6.0 | 60 | 59 | 98% | | --- | --- | --- | --- | | 4.0 | 60 | 58 | 97% | | 2.0 | 59 | 52 | 88% | | 1.0 | 59 | 39 | 66^ | | 0.5 | 60 | 19 | 32% | | 0.0 | 60 | 0 | 0% | | Probit 95% Hit Rate | 3.4 IU/mL [95% confidence limits of 2.7 – 4.6 IU/mL] | | | ## Limit of Quantitation The limit of quantitation was determined using the WHO International Standard in accordance with the methods defined in the CLSI Guideline EP17-A, "Protocols for Determination of Limits of Detection and Limits of Quantitation; Approved Guideline." The WHO Standard was freshly diluted into each of five unique EDTA plasma specimens and each of five unique serum specimens. Each level of each dilution was tested with six replicates split across two runs for each of two reagent lots for each matrix. A total of 10 runs were conducted across five days for each reagent lot for each matrix to give a total of 60 replicates for each level for each matrix. The studies demonstrated that the COBAS TaqMan HBV Test can determine the concentration of HBV DNA in EDTA plasma and serum at concentrations as low as 29 IU/mL with an acceptable level of accuracy: | Matrix | Expected Concentration | Expected log_{10}Concentration | Observed Avg. log_{10}Concentration | Absolute Bias | SD log_{10}Concentration | Total Analytical Error^{1} | | --- | --- | --- | --- | --- | --- | --- | | EDTA Plasma | 29 IU/mL | 1.462 | 1.25 | 0.21 | 0.24 | 0.69 | | Serum | 29 IU/mL | 1.462 | 1.247 | 0.22 | 0.17 | 0.56 | ¹ The Total Analytical Error, or TAE, is defined as Bias + 2SD; the TAE was 0.69 for plasma, and 0.56 for serum. At this concentration, the difference between two measurements of more than 1.0 log₁₀ IU/mL is statistically significant. ## Sample Handling and Collection ## Sample Types (Serum and Plasma) The COBAS TaqMan HBV Test is for use with serum or EDTA plasma specimens only. To demonstrate the equivalency of the use of EDTA plasma or serum, 60 matched clinical specimen sets (each set is EDTA plasma and serum drawn from a single HBV-infected or HBsAg-positive individual) were tested to demonstrate plasma and serum equivalency. Each sample was tested in duplicate {16} Page - 17 Summary of Safety and Effectiveness Data and the mean titer for each sample was calculated. A Deming linear regression analysis was also performed using the calculated mean titers. Fifty matched sets were used for data analysis which included 41 matched sets testing within the linear range of the test and nine matched sets with titers above the linear range of the assay upon initial testing and retested following 1:100 dilution to obtain the titer of the original sample. The results from ten sets (five negative for HBV DNA and five sets positive for HBV DNA but with titers too low to quantify) were not included in the final analysis. The individual log titer difference (log titer EDTA plasma – log titer) for 49 of the 50 matched sets was ≤ 0.30 with the one set having a difference of -0.37. The mean difference was -0.05 log (95% CI: -0.036, 0.025), indicating that the results between serum and EDTA plasma were not significantly different. The result from the linear regression analysis which demonstrates the effect of matrix type on HBV DNA results from patient specimens is shown in the figure below, with slope=1.0131 (95% confidence interval is [1.0045 – 1.0216]) with an intercept of -0.0605 (95% confidence interval is [-0.1065 to -0.0144]). Regression Analysis of Matched Serum – EDTA Plasma Samples (n=50):  The pooled standard deviation estimates for the EDTA plasma and serum samples in the matrix equivalency study are shown in the table below. | Matrix | Mean log Titer | Pooled SD | | --- | --- | --- | | EDTA Plasma | 4.621 | 0.060 | | Serum | 4.616 | 0.093 | {17} Summary of Safety and Effectiveness Data # Specimen Collection Blood should be collected in BD SST Serum Separator Tubes or in tubes using EDTA (lavender top) as the anticoagulant. Whole blood can be stored at $2 - 25^{\circ}\mathrm{C}$ for no longer than 1 day. Serum or plasma should be separated from whole blood within 1 day of collection by centrifugation at $800 - 1600\mathrm{xg}$ for 20 minutes at room temperature, and transferred to a sterile polypropylene tube. The figure below illustrates the data from specimen collection studies. The largest observed difference between the EDTA plasma conditions was not more than $\pm 0.12\log_{10}$ and the largest observed difference between the serum conditions was not more than $\pm 0.16\log_{10}$ . HBV Stability in Whole Blood With EDTA Anticoagulant or in Serum Separator Tubes Before Separation into Plasma or Serum:  # Specimen Transport Transportation of whole blood, serum or plasma must comply with country, federal, state and local regulations for the transport of etiologic agents. Whole blood must be transported at $2 - 25^{\circ}\mathrm{C}$ and processed within 6 hours of collection. # Specimen Storage {18} Page - 19 Summary of Safety and Effectiveness Data Serum or plasma specimens may be stored at room temperature for up to 3 days, at 2-8°C for up to 7 days or frozen at -20°C to -80°C for up to six weeks. The largest observed difference between the EDTA plasma conditions was not more than ±0.11 log₁₀ and the largest observed difference between the serum conditions was not more than ±0.05 log₁₀ across the tested conditions. It is recommended that specimens be stored in 800-900 μL aliquots in sterile, 2.0 mL polypropylene screw-cap tubes. Figure below shows the HBV stability data in EDTA-plasma or serum from these specimen storage studies:  Serum and plasma specimens may be frozen and thawed up to five times without a loss of HBV DNA. The largest observed difference between the EDTA plasma conditions was not more than ±0.19 log₁₀ and the largest observed difference between the serum conditions was not more than ±0.10 log₁₀. Figure below illustrates the HBV results after up to five freeze-thaw (F-T) cycles data from these freeze-thaw studies:  **Analytical Specificity** {19} Page - 20 Summary of Safety and Effectiveness Data # Cross-reactivity The analytical specificity of the COBAS TaqMan HBV Test was evaluated by adding cultured virus into HBV-negative human EDTA plasma or analyzing specimens from subjects positive for other viral agents (listed in table below). Except for one CMV infected specimen and HPV strain 18, none of the non-HBV DNA or RNA viruses tested were positive for HBV DNA. Subsequent testing of the CMV infected specimen did not consistently confirm the initial result. Subsequent testing of HPV strain 18 indicated that no positive results for HBV were detected at HPV concentrations less than $2.0\mathrm{E} + 09~\mathrm{cp / mL}$ Analytical specificity specimens tested: | Virus Added Into Plasma | Specimens from Infected Patients (n) | | --- | --- | | Adenovirus type 7 | Cytomegalovirus infected patients (2)2 | | Cytomegalovirus AD-169 | Epstein-Barr Virus infected patients (2) | | Epstein-Barr Virus (RAJI Burkitt's Lymphoma cells) | Hepatitis A Virus infected patients (2) | | Hepatitis A Virus PA21 | Hepatitis C Virus infected patient genotype 4 (1) | | Herpes Simplex type 1, MacIntyre | Hepatitis C Virus infected patient genotype 6a (1) | | Herpes Simplex type 2, MS | HIV-1 infected patients (2) | | Human Papilloma Virus Strain 181 | | | Influenza A virus A/Hong Kong/8/68 | | | Influenza B virus B/R75 | | | Varicella-Zoster Ellen | | | West Nile Virus | | ${}^{1}$ HPV strain 18 returned a positive HBV result at an HPV concentration of ${2.0E} + {09}\mathrm{{cp}}/\mathrm{{mL}}$ . Subsequent testing indicated no HBV positive results for HPV concentrations less than ${2.0E} + {09}$ cp/mL. 2One of the two CMV specimens from infected patients returned a positive result which was not consistently confirmed in subsequent testing. # Interfering Substances Clinical specimens with elevated levels of triglycerides, bilirubin, albumin and hemoglobin were tested in the absence and presence (approximately $150~\mathrm{IU / mL}$ ) of HBV and did not interfere with the quantitation of HBV DNA in the ranges tested: | | Range of Specimens Tested | Normal Range | | --- | --- | --- | | Triglycerides | 655 – 1,378 mg/dL | 45 – 190 mg/dL | | Bilirubin | 3.7 – 8.2 mg/dL | 0.25 – 1.2 mg/dL | | Albumin | 5,100 – 6,600 mg/dL | 2,800 – 5,000 mg/dL | {20} Page - 21 Summary of Safety and Effectiveness Data The following drug compounds were tested at $1 \times C_{\max}$ and $3 \times C_{\max}$ for each drug in the absence and presence (approximately 150 IU/mL) of HBV and did not interfere with the quantitation of HBV DNA by this test: | Nucleotide DNA Polymerase Inhibitors Adefovir dipivoxil Tenofovir disoproxil fumarate | Nucleoside Reverse Transcriptase and DNA Polymerase Inhibitors Lamivudine Zidovudine Zalcitabine Stavudine Abacavir | | --- | --- | | HIV Protease Inhibitors Indinavir Ritonavir Nelfinavir Saquinavir Amprenavir Lopinavir/Ritonavir | Non-nucleoside HIV Reverse Transcriptase Inhibitors Nevirapine Efavirenz | | Immune Modulators Interferon alpha-2a Interferon alpha-2b | HIV Fusion Inhibitor Enfuvirtide | | CMV Treatment Compounds Ganciclovir Valganciclovir hydrochloride Acyclovir Valacyclovir hydrochloride | | ## Within-Laboratory Precision Within-Run, Run-to-Run and Total Precision were evaluated in accordance with the methods defined in the CLSI (formerly NCCLS) Guideline EP5-A2, "Evaluation of Precision Performance of Quantitative Measurement Methods; Approved Guideline – Second Edition." This procedure permits the determination of both Within-Run and Total Precision through the performance of a single multiple-day and multiple operator study. A run, consisting of three replicates of each of ten panel members diluted from an HBV genotype A clinical specimen, was performed daily for 15 days. Each panel member was taken through the entire COBAS TaqMan HBV Test procedure, including specimen preparation, amplification and detection, representing all aspects of the test procedure. The study was performed for three lots of COBAS TaqMan HBV Test reagents, and {21} Page - 22 Summary of Safety and Effectiveness Data the combined results are presented in HBV DNA IU/mL and HBV DNA $\log_{10}$ IU/mL. Precision of the COBAS TaqMan HBV Test (in IU/mL): | Specimen | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | Average Observed HBV DNA Titer (IU/mL) | 1.07E8 | 5.33E7 | 1.04E7 | 8.52E5 | 9.28E4 | 1.21E4 | 1200 | 111 | 49 | 14 | | Within-Run CV | 19% | 10% | 12% | 7% | 14% | 16% | 14% | 22% | 27% | 50% | | Run-to-Run CV | 11% | 14% | 12% | 14% | 16% | 18% | 18% | 26% | 17% | 22% | | Total CV | 23% | 17% | 17% | 16% | 22% | 24% | 22% | 34% | 32% | 54% | | Total No. Replicates | 132 | 134 | 134 | 135 | 135 | 134 | 135 | 135 | 135 | 135 | Precision of the COBAS TaqMan HBV Test (in $\log_{10}$ IU/mL): | Specimen | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | Average Observed HBV DNA Titer ($\log_{10}$ IU/mL) | 8.02 | 7.72 | 7.01 | 5.92 | 4.94 | 4.06 | 3.07 | 2.03 | 1.67 | 1.05 | | Within-Run Standard Deviation | 0.07 | 0.04 | 0.05 | 0.03 | 0.21 | 0.18 | 0.06 | 0.09 | 0.11 | 0.59 | | Run-to-Run Standard Deviation | 0.05 | 0.06 | 0.05 | 0.06 | 0.07 | 0.07 | 0.08 | 0.10 | 0.07 | 0.00 | | Total Standard Deviation | 0.08 | 0.07 | 0.07 | 0.07 | 0.22 | 0.19 | 0.10 | 0.13 | 0.13 | 0.59 | ## Reproducibility The reproducibility of the COBAS TaqMan HBV Test For Use With the High Pure System was evaluated by two operators at each of three external clinical sites. Each operator performed three days of testing on each of three lots of reagents with each panel. Each run comprised a single panel with each panel member tested in triplicate. The results of the reproducibility study are summarized in tables below for EDTA plasma and serum. Components of Variance (Percentage of Total Variance) of HBV DNA Concentration ($\log_{10}$ IU/mL) — EDTA Plasma: {22} Page - 23 Summary of Safety and Effectiveness Data | Geno-type | Mean of HBV DNA Concentration (log_{10} IU/mL) | Geometric Mean of HBV DNA Concentration (IU/mL) | No. of Tests | Lot | Site/Instru-ment | Operator | Day/Run | Within-Run | Total Precision Variance of log_{10} HBV DNA Concentration | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | A | 1.32 | 21 | 162^{a} | 0.0015 (3%) | 0.0000 (0%) | 0.0008 (1%) | 0.0044 (7%) | 0.0521 (89%) | 0.0588 (100%) | | 2.34 | 220 | 162 | 0.0004 (5%) | 0.0000 (0%) | 0.0008 (11%) | 0.0015 (20%) | 0.0048 (64%) | 0.0074 (100%) | | 3.36 | 2,314 | 162 | 0.0003 (4%) | 0.0000 (0%) | 0.0019 (30%) | 0.0014 (22%) | 0.0027 (44%) | 0.0062 (100%) | | 4.35 | 22,369 | 162 | 0.0010 (12%) | 0.0002 (2%) | 0.0019 (24%) | 0.0019 (23%) | 0.0032 (39%) | 0.0081 (100%) | | 5.19 | 154,752 | 162 | 0.0014 (24%) | 0.0003 (5%) | 0.0011 (20%) | 0.0011 (20%) | 0.0017 (31%) | 0.0056 (100%) | | 7.29 | 19,444,058 | 162 | 0.0000 (1%) | 0.0000 (0%) | 0.0005 (14%) | 0.0007 (21%) | 0.0022 (64%) | 0.0035 (100%) | | C | 1.38 | 24 | 162 | 0.0020 (9%) | 0.0006 (3%) | 0.0006 (3%) | 0.0059 (25%) | 0.0143 (61%) | 0.0235 (100%) | | 2.34 | 219 | 160^{b} | 0.0022 (34%) | 0.0004 (6%) | 0.0005 (7%) | 0.0000 (0%) | 0.0035 (54%) | 0.0064 (100%) | | 3.43 | 2,686 | 162 | 0.0004 (7%) | 0.0016 (29%) | 0.0009 (17%) | 0.0008 (15%) | 0.0019 (33%) | 0.0057 (100%) | | 4.39 | 24,479 | 161^{c} | 0.0004 (8%) | 0.0010 (21%) | 0.0008 (16%) | 0.0011 (22%) | 0.0015 (32%) | 0.0047 (100%) | | 5.24 | 172,515 | 162 | 0.0007 (17%) | 0.0009 (20%) | 0.0008 (19%) | 0.0006 (13%) | 0.0014 (31%) | 0.0044 (100%) | | 7.38 | 23,791,720 | 162 | 0.0013 (21%) | 0.0001 (1%) | 0.0004 (8%) | 0.0008 (14%) | 0.0033 (56%) | 0.0059 (100%) | a 14 out of 162 (8.6%) results were “HBV DNA detected, less than 10.00 HBV DNA IU/mL.” b 2 results were invalid. c 1 result was invalid. Standard Deviation Components of HBV DNA Concentration (log₁₀ IU/mL) — EDTA Plasma: | Geno-type | Mean of HBV DNA Concentration (log_{10} IU/mL) | Geometric Mean of HBV DNA Concentration (IU/mL) | No. of Tests | Lot | Site/Instru-ment | Operator | Day/Run | Within-Run | Total Standard Deviation of log_{10} HBV DNA Concentration | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | A | 1.32 | 21 | 162^{a} | 0.0391 | 0.0000 | 0.0277 | 0.0662 | 0.2283 | 0.2425 | | | 2.34 | 220 | 162 | 0.0188 | 0.0000 | 0.0291 | 0.0381 | 0.0691 | 0.0860 | | | 3.36 | 2,314 | 162 | 0.0161 | 0.0000 | 0.0431 | 0.0368 | 0.0522 | 0.0787 | | | 4.35 | 22,369 | 162 | 0.0311 | 0.0128 | 0.0441 | 0.0431 | 0.0564 | 0.0900 | {23} Page - 24 Summary of Safety and Effectiveness Data | Genotype | Mean of HBV DNA Concentration (log_{10} IU/mL) | Geometric Mean of HBV DNA Concentration (IU/mL) | No. of Tests | Lot | Site/Instrument | Operator | Day/Run | Within-Run | Total Standard Deviation of log_{10} HBV DNA Concentration | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | 5.19 | 154,752 | 162 | 0.0368 | 0.0161 | 0.0329 | 0.0337 | 0.0416 | 0.0748 | | | 7.29 | 19,444,058 | 162 | 0.0051 | 0.0000 | 0.0222 | 0.0268 | 0.0474 | 0.0592 | | C | 1.38 | 24 | 162 | 0.0449 | 0.0251 | 0.0246 | 0.0770 | 0.1196 | 0.1533 | | | 2.34 | 219 | 160^{b} | 0.0466 | 0.0190 | 0.0212 | 0.0000 | 0.0589 | 0.0800 | | | 3.43 | 2,686 | 162 | 0.0205 | 0.0402 | 0.0307 | 0.0286 | 0.0431 | 0.0755 | | | 4.39 | 24,479 | 161^{c} | 0.0196 | 0.0319 | 0.0274 | 0.0327 | 0.0391 | 0.0686 | | | 5.24 | 172,515 | 162 | 0.0270 | 0.0292 | 0.0290 | 0.0242 | 0.0368 | 0.0663 | | | 7.38 | 23,791,720 | 162 | 0.0354 | 0.0079 | 0.0212 | 0.0285 | 0.0574 | 0.0768 | a 14 out of 162 (8.6%) results were “HBV DNA detected, less than 10.00 HBV DNA IU/mL.” b 2 results were invalid. c 1 result was invalid. Reproducibility results summary: Total %CV for HBV panel members — EDTA Plasma: | Genotype | Mean of HBV DNA Concentration (log_{10} IU/mL) | Geometric Mean of HBV DNA Concentration (IU/mL) | No. of Tests^{1} | Total Precision Variance of log_{10} HBV DNA Concentration | Total Precision Standard Deviation of log_{10} HBV DNA Concentration | lognormal CV (%)^{2} | | --- | --- | --- | --- | --- | --- | --- | | A | 1.32 | 21 | 162^{a} | .0588 | 0.24 | 60 | | | 2.34 | 220 | 162 | .0074 | 0.09 | 20 | | | 3.36 | 2,314 | 162 | .0062 | 0.08 | 18 | | | 4.35 | 22,369 | 162 | .0081 | 0.09 | 21 | | | 5.19 | 154,752 | 162 | .0056 | 0.07 | 17 | | | 7.29 | 19,444,058 | 162 | .0035 | 0.06 | 14 | | C | 1.38 | 24 | 162 | .0235 | 0.15 | 36 | | | 2.34 | 219 | 160^{b} | .0064 | 0.08 | 19 | | | 3.43 | 2,686 | 162 | .0057 | 0.08 | 17 | | | 4.39 | 24,479 | 161^{c} | .0047 | 0.07 | 16 | | | 5.24 | 172,515 | 162 | .0044 | 0.07 | 15 | | | 7.38 | 23,791,720 | 162 | .0059 | 0.08 | 18 | a 14 out of 162 (8.6%) results were “HBV DNA detected, less than 10.00 HBV DNA IU/mL.” b 2 results were invalid. c 1 result was invalid {24} Page - 25 Summary of Safety and Effectiveness Data Components of Variance (Percentage of Total Variance) of HBV DNA Concentration (log₁₀ IU/mL) — Serum: | Genotype | Mean of HBV DNA Concentration (log₁₀ IU/mL) | Geometric Mean of HBV DNA Concentration (IU/mL) | No. of Tests | Lot | Site/Instrument | Operator | Day/Run | Within-Run | Total Precision Variance of log₁₀ HBV DNA Concentration | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | A | 1.04 | 11 | 162^{a} | 0.0054 (5%) | 0.0127 (13%) | 0.0010 (1%) | 0.0044 (4%) | 0.0767 (77%) | 0.1001 (100%) | | | 2.10 | 127 | 160^{b} | 0.0000 (0%) | 0.0022 (16%) | 0.0007 (5%) | 0.0043 (30%) | 0.0072 (50%) | 0.0144 (100%) | | | 3.34 | 2,194 | 162 | 0.0000 (1%) | 0.0022 (28%) | 0.0009 (11%) | 0.0016 (21%) | 0.0031 (39%) | 0.0078 (100%) | | | 4.34 | 21,749 | 161^{c} | 0.0020 (20%) | 0.0029 (29%) | 0.0008 (8%) | 0.0023 (23%) | 0.0019 (19%) | 0.0099 (100%) | | | 5.17 | 147,146 | 162 | 0.0024 (35%) | 0.0011 (16%) | 0.0007 (10%) | 0.0016 (24%) | 0.0011 (16%) | 0.0069 (100%) | | | 7.27 | 18,732,744 | 162 | 0.0000 (0%) | 0.0003 (6%) | 0.0001 (4%) | 0.0003 (7%) | 0.0033 (83%) | 0.0039 (100%) | | C | 1.38 | 24 | 162 | 0.0005 (2%) | 0.0000 (0%) | 0.0032 (9%) | 0.0068 (19%) | 0.0244 (70%) | 0.0349 (100%) | | | 2.34 | 218 | 161^{c} | 0.0010 (11%) | 0.0000 (0%) | 0.0005 (5%) | 0.0040 (48%) | 0.0030 (35%) | 0.0084 (100%) | | | 3.43 | 2,664 | 162 | 0.0003 (5%) | 0.0003 (5%) | 0.0009 (15%) | 0.0029 (46%) | 0.0018 (29%) | 0.0062 (100%) | | | 4.39 | 24,555 | 162 | 0.0009 (12%) | 0.0004 (6%) | 0.0005 (6%) | 0.0040 (53%) | 0.0018 (24%) | 0.0076 (100%) | | | 5.22 | 167,232 | 162 | 0.0009 (13%) | 0.0006 (9%) | 0.0005 (7%) | 0.0032 (46%) | 0.0017 (24%) | 0.0070 (100%) | | | 7.37 | 23,676,552 | 161^{c} | 0.0006 (8%) | 0.0000 (0%) | 0.0004 (5%) | 0.0041 (51%) | 0.0030 (37%) | 0.0082 (100%) | a 65 out of 162 (40.1%) results were “HBV DNA detected, less than 10.00 HBV DNA IU/mL.” b 2 results were invalid. c 1 result was invalid Standard Deviation Components of HBV DNA Concentration (log₁₀ IU/mL) — Serum: | Genotype | Mean of HBV DNA Concentration (log₁₀ IU/mL) | Geometric Mean of HBV DNA Concentration (IU/mL) | No. of Tests | Lot | Site/Instrument | Operator | Day/Run | Within-Run | Total Standard Deviation of log₁₀ HBV DNA Concentration | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | {25} Page - 26 Summary of Safety and Effectiveness Data | Genotype | Mean of HBV DNA Concentration (log_{10} IU/mL) | Geometric Mean of HBV DNA Concentration (IU/mL) | No. of Tests | Lot | Site/Instrument | Operator | Day/Run | Within-Run | Total Standard Deviation of log_{10} HBV DNA Concentration | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | A | 1.04 | 11 | 162^{a} | 0.0737 | 0.1127 | 0.0312 | 0.0660 | 0.2769 | 0.3164 | | | 2.10 | 127 | 160^{b} | 0.0000 | 0.0473 | 0.0260 | 0.0654 | 0.0850 | 0.1200 | | | 3.34 | 2,194 | 162 | 0.0067 | 0.0470 | 0.0292 | 0.0406 | 0.0553 | 0.0883 | | | 4.34 | 21,749 | 161^{c} | 0.0451 | 0.0539 | 0.0281 | 0.0482 | 0.0435 | 0.0995 | | | 5.17 | 147,146 | 162 | 0.0489 | 0.0332 | 0.0262 | 0.0403 | 0.0328 | 0.0831 | | | 7.27 | 18,732,744 | 162 | 0.0000 | 0.0160 | 0.0121 | 0.0164 | 0.0573 | 0.0624 | | C | 1.38 | 24 | 162 | 0.0232 | 0.0000 | 0.0566 | 0.0823 | 0.1562 | 0.1868 | | | 2.34 | 218 | 161^{c} | 0.0309 | 0.0000 | 0.0215 | 0.0636 | 0.0547 | 0.0917 | | | 3.43 | 2,664 | 162 | 0.0181 | 0.0179 | 0.0306 | 0.0536 | 0.0421 | 0.0787 | | | 4.39 | 24,555 | 162 | 0.0297 | 0.0210 | 0.0214 | 0.0634 | 0.0424 | 0.0872 | | | 5.22 | 167,232 | 162 | 0.0302 | 0.0251 | 0.0221 | 0.0569 | 0.0413 | 0.0837 | | | 7.37 | 23,676,552 | 161^{c} | 0.0249 | 0.0000 | 0.0204 | 0.0644 | 0.0549 | 0.0906 | a 65 out of 162 (40.1%) results were “HBV DNA detected, less than 10.00 HBV DNA IU/mL.” b 2 results were invalid. c 1 result was invalid Reproducibility Results Summary: Total %CV for HBV Panel Members — Serum: | Genotype | Mean of HBV DNA Concentration (log_{10} IU/mL) | Geometric Mean of HBV DNA Concentration (IU/mL) | No. of Tests^{1} | Total Precision Variance of log_{10} HBV DNA Concentration | Total Precision Standard Deviation of log_{10} HBV DNA Concentration | lognormal CV (%)^{2} | | --- | --- | --- | --- | --- | --- | --- | | A | 1.04 | 11 | 162^{a} | .1001 | 0.32 | 84 | | | 2.10 | 127 | 160^{b} | .0144 | 0.12 | 28 | | | 3.34 | 2,194 | 162 | .0078 | 0.09 | 21 | | | 4.34 | 21,749 | 161^{c} | .0099 | 0.10 | 23 | | | 5.17 | 147,146 | 162 | .0069 | 0.08 | 19 | | | 7.27 | 18,732,744 | 162 | .0039 | 0.06 | 15 | | C | 1.39 | 24 | 162 | .0349 | 0.19 | 45 | | | 2.34 | 218 | 161^{c} | .0084 | 0.09 | 21 | | | 3.43 | 2,664 | 162 | .0062 | 0.08 | 18 | | | 4.39 | 24,555 | 162 | .0076 | 0.09 | 20 | | | 5.22 | 167,232 | 162 | .0070 | 0.08 | 19 | | | 7.37 | 23,676,552 | 161^{c} | .0082 | 0.09 | 21 | a 65 out of 162 (40.1%) results were “HBV DNA detected, less than 10.00 HBV DNA IU/mL.” b 2 results were invalid. {26} Page - 27 Summary of Safety and Effectiveness Data *1 result was invalid* The table below summarizes the results for HBV Negative Panel Members from the reproducibility study for each matrix. Eleven false positives were observed in the study and all were below the LOD. Specificity was 100% in EDTA plasma [95% CI = (0.98, 1.00)] and 97% in serum [95% CI = (0.94, 0.99)]: | Matrix | Total Valid Results | Target Not Detected | Target Detected but Below LOD^{1} | ≥10 and <29 IU/mL^{2} | Within Linear Range^{3} | | --- | --- | --- | --- | --- | --- | | EDTA Plasma | 324 | 323 | 1 | 0 | 0 | | Serum | 322 | 312 | 10 | 0 | 0 | 1 The limit of detection (LOD) for the assay is 10 IU/mL. Results &lt; 10 IU/mL are below the LOD. 2 Results 10 IU/mL to &lt; 29 IU/mL are above the LOD, but below the linear range. 3 Linear range (&gt;29 IU/mL to 1.10E+8 IU/mL). ## Performance of COBAS TaqMan HBV Test with HBV-Negative Samples The performance of the COBAS TaqMan HBV Test with HBV-negative samples was determined by analysis of HBV-negative serum and EDTA plasma from blood donors. A total of 220 specimens (110 individual EDTA plasma and 110 serum specimens) that were non-reactive for HBsAg and anti-HBc were tested. All specimens were noted as Target Not Detected for HBV DNA for both EDTA plasma and serum (100%, or 110/110, with a 95% confidence interval of 96.6% to 100%). ## Carryover / Cross-Contamination Studies The objective of this study was to verify that any cross contamination potential occurring when the entire process (sample processing and amplification / detection) is performed as prescribed by the manufacturer. Testing consisted of at least five runs consisting of alternating high titer positive and negative samples. Each run consisted of twelve replicates of high positive sample at ~2E7 IU/mL and twelve replicates of HBV-negative human EDTA plasma in an alternating pattern. Each run also included four replicates of COBAS TaqMan Negative Control (NC), two replicates of COBAS TaqMan HBV Low Positive Control (LPC) and two replicates of COBAS TaqMan HBV High Positive Control (HPC). All 60 replicates of HBV-negative EDTA plasma produced a result of “Target Not Detected” and all 60 replicates of high titer sample were valid. All of the individual controls were valid and within the specified range. Therefore, the carryover contamination rate was 0% across 5 runs consisting alternating high positive HBV specimen (~2E7 IU/mL) and HBV-negative plasma. ## Recommended Storage Stability {27} Summary of Safety and Effectiveness Data Results of real-time stability studies indicate that the COBAS TaqMan HBV Test is stable for 12 months when stored at its labeled storage conditions of 2 to 8°C. ## Open Bottle Reagent Stability The study was designed to confirm the stability of the opened, reconstituted and unused portion of reagents between the first and second use. Three levels of clinical HBV Proficiency Sample Panel in EDTA Plasma and three levels of clinical HBV Proficiency Sample Panel in Serum were used to assess the stability of opened vials of reagents. The panel descriptions: | HBV Proficiency Panel (Serum) | HBV Proficiency Sample Panel (Plasma) | Titer | | --- | --- | --- | | Proficiency Sample 1 (PS1) | Proficiency Sample 1 (PS1) | ~150 | | Proficiency Sample 2 (PS2) | Proficiency Sample 2 (PS2) | 1 x 10E5 | | Proficiency Sample 3 (PS3) | Proficiency Sample 3 (PS3) | 5 x 10E8 | The stability of the following twelve reagents were assessed after the storage under the conditions indicated below: | Reagent | | Storage Temp | Storage Time (weeks) | | --- | --- | --- | --- | | COBAS TaqMan HBV MMX | HBV MMX | 2 - 8°C | 2, 3, 4, 5, 6, 7 | | COBAS TaqMan Mn2+ | CTM Mn2+ | 2 - 8°C | 2, 3, 4, 5, 6, 7 | | COBAS TaqMan HBV QS | HBV QS | 2 - 8°C | 2, 3, 4, 5, 6, 7 | | COBAS TaqMan Negative Control1 | CTM (-) C | -20°C | 2, 3, 4, 5, 6, 7 | | COBAS TaqMan HBV Low Positive Control | HBV L(+)C | -20°C | 2, 3, 4, 5, 6, 7 | | COBAS TaqMan HBV High Positive Control | HBV H(+)C | 2 - 8°C | 2, 3, 4, 5, 6, 7 | | Reconstituted PolyA carrier RNA | CAR | 2 - 8°C | 2, 3, 4, 5, 6, 7 | | Reconstituted Proteinase K | PK | 2 - 8°C | 2, 3, 4, 5, 6, 7 | | Lysis Buffer | LYS | 15 - 25°C | 2, 3, 4, 5, 6, 7 | | Working Wash Buffer (following Ethanol addition) | WASH | 15 - 25°C | 2, 3, 4, 52 | | Working Inhibitor Removal Buffer (following Ethanol addition) | IRB | 15 - 25°C | 2, 3, 4, 52 | | Elution Buffer | ELB | 15 - 25°C | 2, 3, 4, 5, 6, 7 | (1) Although the stability of CTM (-) C was evaluated, the CTM (-) is a single use reagent. Once opened, any unused portion of CTM (-) C must be discarded. (2) Wash Buffer and Inhibitor Removal Buffer was prepared freshly on weeks 2 and 3. At each time point, testing was performed using 3 replicates of each panel member and 3 and 1 replicate of each kit control (NC, LPC and HPC). {28} Page - 29 Summary of Safety and Effectiveness Data Once opened, ELB and LYS are stable for 30 days when stored at 15-25°C in original vial. After addition of ELB to reconstitute the CAR and PK, unused reconstituted CAR and PK are stable for 30 days when stored at -15 to -25°C in original vial. After addition of ethanol, IRB and WASH Working Solutions are stable for 30 days when stored at 15-25°C. Once opened, unused portions of HBV MMX, HBV L(+), HBV H(+), HBV QS and CTM Mn²⁺ are stable for 30 days when stored at 2-8°C in the original vial. ## X. SUMMARY OF CLINICAL STUDIES ### Study Population and Baseline Parameters The clinical performance of the COBAS TaqMan HBV Test For Use With The High Pure System was evaluated by assessing the antiviral therapy response in chronic HBV-infected subjects undergoing treatment with adefovir dipivoxil. The HBV DNA data were obtained from testing patient samples previously collected under two study protocols, one of which evaluated patients with chronic HBeAg+ HBV infection and compensated liver function and one that evaluated patients with presumed precore mutant (HBeAg-/ HBV DNA+) chronic HBV infection with compensated liver function. Marcellin et al.⁴ and Hadziyannis et al.⁶ previously described subject selection for both studies. The study population consisted of 407 chronic HBV infected patients enrolled in double-blind, randomized, placebo-controlled studies of Adefovir Dipivoxil. Demographic data, drug dosing data, HBV genotype, HBeAg and anti-HBe results, ALT results, and Baseline (pre-treatment) and end-point liver biopsy results were available for each patient. Viral load testing was performed at Screening and at Weeks 4, 8, 16, 28, 44, and 48 (when available). Table below summarizes the study population at Screening: | Characteristic | Category | Summary Statistics | HBeAg+ | HBeAg- | Total | | --- | --- | --- | --- | --- | --- | | Total Number of Subjects | | N | 264 | 143 | 407 | | Placebo | | n (%) | 129 (49) | 50 (35) | 179 (44) | | 10 mg Adefovir Dipivoxil | | n (%) | 135 (51) | 93 (65) | 228 (56) | | Age (yr) | | Median (Min, Max) | 34 (16, 65) | 46 (18, 65) | 39 (16, 65) | | Weight (kg) | | Median (Min, Max) | 70 (41, 118) | 74 (46, 111) | 72 (41, 118) | | Sex | Male | N (%) | 191 (72) | 119 (83) | 310 (76) | | | Female | N (%) | 73 (28) | 24 (17) | 97 (24) | | Race | White | N (%) | 86 (33) | 90 (63) | 176 (43) | | | Asian | N (%) | 167 (63) | 48 (34) | 215 (53) | | | Other | N (%) | 11 (4) | 5 (3) | 16 (4) | | Genotype | A | N (%) | 73 (28) | 9 (6) | 82 (20) | {29} Page - 30 Summary of Safety and Effectiveness Data | Characteristic | Category | Summary Statistics | HBeAg+ | HBeAg- | Total | | --- | --- | --- | --- | --- | --- | | | B | N (%) | 49 (19) | 28 (20) | 77 (19) | | | C | N (%) | 111 (42) | 19 (13) | 130 (32) | | | D | N (%) | 25 (9) | 84 (59) | 109 (27) | | | Other | N (%) | 6 (2) | 3 (2) | 9 (2) | | HBV DNA < 1.72E+04 IU/mL | | N (%) | 3 (1) | 5 (3) | 8 (2) | | ALT <= ULN^{1} | | N (%) | 5 (2) | 7 (5) | 12 (3) | | Knodell Score | | N | 259 | 139 | 398 | | Total | | Mean (SD) | 9.5 (3.3) | 9.4 (3.4) | 9.5 (3.3) | | Necroinflammatory | | Mean (SD) | 7.7 (2.8) | 7.5 (2.8) | 7.6 (2.7) | | Fibrosis | | Mean (SD) | 1.7 (1.1) | 1.9 (1.2) | 1.8 (1.1) | $^{1}ULN =$ Upper Limit of Normal Range Screening samples were obtained six to 125 days before study start. On average, HBeAg+ patients were younger than the HBeAg- patients (median age = 34 years vs 46 years), predominantly Asian (63% vs 34%), were female (28% vs 17%) and were infected with primarily HBV genotypes A and C (70% vs 19%). HBeAg- patients were predominantly White (63%) and infected with HBV genotype D (59%). The Knodell necropsy scores for necroinflammation and fibrosis at Baseline (pre-treatment) were comparable for both populations. Patients included in the clinical performance analysis received either the standard 10 mg Adefovir dipivoxil dosing or placebo, as indicated in the following table summarizing available samples by treatment arm: | Population | No. Subjects — Placebo | No. Subjects — 10 mg Adefovir | No. Samples per Subject | Total No. Samples | | --- | --- | --- | --- | --- | | Chronic HBeAg+ | 129 | 135 | 7 | 1848 | | Chronic HBeAg- | 50 | 93 | 6 | 858 | | Grand Total | | | | 2706 | ## Within-Subject Variability in Absence of Treatment One hundred and seventy nine subjects were enrolled into the placebo arms of the HBeAg+ and HBeAg- studies of which 137 of 179 had results within the linear range of the assay for testing at Weeks 0, 4, and 8. These results were used to estimate within subject variability, which includes biological variability as well as total assay variability. The within subject variability from these results was estimated to be 0.58 log₁₀ IU/mL for HBeAg- patients and 0.78 log₁₀ IU/mL for HBeAg+ patients. Biological variability was similar to the within-subject variability since the assay variability was negligible. The median change of viral load within a subject was estimated to be 0.30 log₁₀ IU/mL for HBeAg+ and 0.58 {30} Page - 31 Summary of Safety and Effectiveness Data $\log_{10}$ IU/mL for HBeAg- patients. Approximately 90% of the HBeAg+ patients’ and 80% of the HBeAg- patients’ change of viral load was less than 2.0 $\log_{10}$ IU/mL. ## Safety and Effectiveness Results ### Clinical Performance of the COBAS TaqMan HPS HBV Test in Patients on Therapy Assessment was performed at Screening and at Weeks 4, 8, 16, 28, 44, and 48 (when available). The primary objective was to determine the relationship between viral levels at various treatment time points compared with histological, serological, and biochemical responses to treatment. The results from testing with the COBAS TaqMan HBV Test For Use With The High Pure System were used to determine whether change (or absence of change) in HBV viral load at various time points may predict improvement (or lack of improvement) in a patient’s immune response marker or liver histology at different treatment time points. Statistical analysis of clinical data was used to assess whether viral response to treatment measured with COBAS TaqMan HBV Test For Use With the High Pure System is informative for assessing the response to treatment in HBeAg+ and HBeAg- patients with chronic hepatitis B. Observing changes in viral load in individual patients over time may help the clinician in the assessment of a patient’s response to therapy. ### HBeAg+ Patients A graph in the figure below illustrates the change of the median and inter-quartile range of change in HBV DNA from Screening in patients on adefovir dipivoxil and on placebo. It demonstrates efficacy of treatment of the HBeAg+ patients with chronic hepatitis B with adefovir dipivoxil compared to placebo.  {31} Page - 32 Summary of Safety and Effectiveness Data A useful monitoring tool is to observe whether there is an HBV viral load increase of more than $1\log_{10}$ after reaching a nadir. Results presented in the table below show that $60.7\%$ (82/135) of the patients reached a nadir in viral load by week 44 on treatment. Seventeen patients had more than $1\log_{10}$ increase in viral load by week 48 after achieving the nadir, a $12.6\%$ (17/135) of the total number of patients on treatment and $20.7\%$ (17/82) of the patients who achieved a nadir. Table below summarizes the data for the distribution of the $135\mathrm{HBeAg + }$ patients on adefovir dipivoxil by week on treatment and the viral load at which the nadir was reached: | Nadir Viral Load (IU/mL) | Number (%) of Patients With the Nadir Viral Load Achieved by Week | | | | | | Total By Viral Load | Cumulative By Viral Load | | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | 4 | 8 | 16 | 28 | 44 | 48 | | | | HBV DNA not detected | 0 | 0 | 0 | 0 | 1 (0.7) | 0 | 1 (0.7) | 1 (0.7) | | < 10 | 0 | 0 | 2 (1.5) | 3 (2.2) | 7 (5.2) | 6 (4.4) | 18 (13.3) | 19 (14.1) | | T 10 - < 100 | 0 | 0 | 0 | 0 | 9 (6.7) | 12 (8.9) | 21 (15.6) | 40 (29.6) | | h 100 - < 10³ | 0 | 0 | 2 (1.5) | 1 (0.7) | 0 | 10 (7.4) | 13 (9.6) | 53 (39.3) | | e 10³ - < 10⁴ | 1 (0.7) | 1 (0.7) | 2 (1.5) | 2 (1.5) | 8 (5.9) | 6 (4.4) | 20 (14.8) | 73 (54.1) | | 10⁴ - < 10⁵ | 0 | 0 | 1 (0.7) | 5 (3.7) | 3 (2.2) | 10 (7.4) | 19 (14.1) | 92 (68.1) | | t 10⁵ - < 10⁶ | 0 | 2 (1.5) | 1 (0.7) | 3 (2.2) | 10 (7.4) | 5 (3.7) | 21 (15.6) | 113 (83.7) | | s ≥10⁶ | 3 (2.2) | 5 (3.7) | 4 (3.0) | 4 (3.0) | 2 (1.5) | 4 (3.0) | 22 (16.3) | 135 (100) | | Total By Week | 4 (3.0) | 8 (5.9) | 12 (8.9) | 18 (13.3) | 40 (29.6) | 53 (39.3) | | | | Cumulative By Week | 4 (3.0) | 12 (8.9) | 24 (17.8) | 42 (31.1) | 82 (60.7) | 135 (100) | | | The results of the analysis of the associations between the responses to treatment at week 48 and baseline covariates for HBeAg+ patients are summarized in table below. Since the lower limits of the $95\%$ CIs of the odds ratios are smaller than 1 (0.26 - 0.96), there are no statistically significant associations: | Response to Treatment | Covariate | Category | N | Number of Patients With Response | Proportion (%) of Patients With Response | Unadjusted Odds Ratio (95% CI) | | --- | --- | --- | --- | --- | --- | --- | | Antigen Loss | Race | Asian | 84 | 23 | 27.4 | 0.75 (0.32, 1.80) | | | | Other | 45 | 15 | 33.3 | | | | Sex | Male | 99 | 28 | 28.3 | 0.79 (0.31, 2.14) | | | | Female | 30 | 10 | 33.3 | | | | Age | ≤30 | 60 | 15 | 25.0 | 0.67 (0.28, 1.54) | | | | >30 | 69 | 23 | 33.3 | | | | Genotype | B,C | 82 | 21 | 25.6 | 0.61 (0.26, 1.43) | | | | Non-B,C | 47 | 17 | 36.2 | | {32} Page - 33 Summary of Safety and Effectiveness Data | Histological | Race | Asian | 79 | 52 | 65.8 | 1.20 (0.52, 2.69) | | --- | --- | --- | --- | --- | --- | --- | | | | Other | 47 | 29 | 61.7 | | | | Sex | Male | 95 | 65 | 68.4 | 2.03 (0.81, 5.02) | | | | Female | 31 | 16 | 51.6 | | | | Age | ≤30 | 57 | 39 | 68.4 | 1.39 (0.63, 3.13) | | | | >30 | 69 | 42 | 60.9 | | | | Genotype | B,C | 77 | 52 | 67.5 | 1.43 (0.64, 3.21) | | | | Non-B,C | 49 | 29 | 59.2 | | | Biochemical | Race | Asian | 84 | 52 | 61.9 | 2.09 (0.96, 4.58) | | | | Other | 48 | 21 | 43.8 | | | | Sex | Male | 100 | 58 | 58.0 | 1.57 (0.65, 3.78) | | | | Female | 32 | 15 | 46.9 | | | | Age | ≤30 | 61 | 39 | 63.9 | 1.93 (0.91, 4.13) | | | | >30 | 71 | 34 | 47.9 | | | | Genotype | B,C | 83 | 51 | 61.4 | 1.96 (0.90, 4.26) | | | | Non-B,C | 49 | 22 | 44.9 | | Further data analysis of the HPS/CTM HBV Test results in HBeAg+ population was performed using two different definitions of the early virological response to treatment: (1) HBV viral load &lt; 2000 IU/mL (or approximately 104 cp/mL), (2) a decrease in serum HBV DNA from an initial Screening viral load value by ≥ 2 log₁₀. The statistical significance of the associations of the Race, Sex, Age and Genotype covariates with the viral response was studied by calculating odds ratios plus their exact 95% confidence intervals for both definitions of the viral response and summarized in two tables below. The logistic regression analysis of viral response as a function of the covariates showed no statistical significance of such associations for either definition of the viral response. Table below lists odds ratios for the association between viral response (&lt; 2000 IU/mL) and covariates, by week, for an HBeAg+ population: | Covariate | Category | Week | N | Number With ≥2-log₁₀ Decrease | Proportion (%) With ≥2-log₁₀ Decrease | Unadjusted Odds Ratio (95% CI) | | --- | --- | --- | --- | --- | --- | --- | | Race | Asian | 4 | 85 | 46 | 54.1 | 2.43 (1.10, 5.45) | | | Other | | 49 | 16 | 32.7 | | | | Asian | 8 | 84 | 55 | 65.5 | 1.10 (0.49, 2.44) | | | Other | | 49 | 31 | 63.3 | | | | Asian | 16 | 83 | 66 | 79.5 | 1.02 (0.38, 2.65) | | | Other | | 48 | 38 | 79.2 | | | | Asian | 28 | 84 | 68 | 81.0 | 1.09 (0.40, 2.85) | | | Other | | 49 | 39 | 79.6 | | {33} Page - 34 Summary of Safety and Effectiveness Data | Covariate | Category | Week | N | Number With ≥2-log_{10} Decrease | Proportion (%) With ≥2-log_{10} Decrease | Unadjusted Odds Ratio (95% CI) | | --- | --- | --- | --- | --- | --- | --- | | | Asian | 44 | 82 | 66 | 80.5 | 1.11 ( 0.41, 2.93) | | | Other | | 47 | 37 | 78.7 | | | | Asian | 48 | 84 | 65 | 77.4 | 1.51 ( 0.63, 3.58) | | | Other | | 49 | 34 | 69.4 | | | Sex | Male | 4 | 101 | 44 | 43.6 | 0.64 ( 0.27, 1.53) | | | Female | | 33 | 18 | 54.5 | | | | Male | 8 | 100 | 62 | 62.0 | 0.61 ( 0.23, 1.55) | | | Female | | 33 | 24 | 72.7 | | | | Male | 16 | 99 | 77 | 77.8 | 0.65 ( 0.17, 2.00) | | | Female | | 32 | 27 | 84.4 | | | | Male | 28 | 100 | 80 | 80.0 | 0.89 ( 0.26, 2.62) | | | Female | | 33 | 27 | 81.8 | | | | Male | 44 | 97 | 77 | 79.4 | 0.89 ( 0.26, 2.63) | | | Female | | 32 | 26 | 81.3 | | | | Male | 48 | 100 | 74 | 74.0 | 0.91 ( 0.32, 2.42) | | | Female | | 33 | 25 | 75.8 | | 41 {34} Page - 35 Summary of Safety and Effectiveness Data | Covariate | Category | Week | N | Number With ≥2-log_{10} Decrease | Proportion (%) With ≥2-log_{10} Decrease | Unadjusted Odds Ratio (95% CI) | | --- | --- | --- | --- | --- | --- | --- | | Age | ≤30 | 4 | 63 | 34 | 54.0 | 1.80 ( 0.86, 3.79) | | | >30 | | 71 | 28 | 39.4 | | | | ≤30 | 8 | 62 | 42 | 67.7 | 1.29 ( 0.59, 2.82) | | | >30 | | 71 | 44 | 62.0 | | | | ≤30 | 16 | 61 | 48 | 78.7 | 0.92 ( 0.36, 2.37) | | | >30 | | 70 | 56 | 80.0 | | | | ≤30 | 28 | 62 | 50 | 80.6 | 1.02 ( 0.40, 2.67) | | | >30 | | 71 | 57 | 80.3 | | | | ≤30 | 44 | 60 | 48 | 80.0 | 1.02 ( 0.39, 2.67) | | | >30 | | 69 | 55 | 79.7 | | | | ≤30 | 48 | 62 | 48 | 77.4 | 1.34 ( 0.57, 3.22) | | | >30 | | 71 | 51 | 71.8 | | | Genotype | B,C | 4 | 83 | 44 | 53.0 | 2.07 ( 0.95, 4.54) | | | Non-B,C | | 51 | 18 | 35.3 | | | | B,C | 8 | 82 | 53 | 64.6 | 1.00 ( 0.45, 2.20) | | | Non-B,C | | 51 | 33 | 64.7 | | | | B,C | 16 | 81 | 65 | 80.2 | 1.15 ( 0.43, 2.94) | | | Non-B,C | | 50 | 39 | 78.0 | | | | B,C | 28 | 82 | 67 | 81.7 | 1.23 ( 0.46, 3.18) | | | Non-B,C | | 51 | 40 | 78.4 | | | | B,C | 44 | 80 | 65 | 81.3 | 1.25 ( 0.47, 3.27) | | | Non-B,C | | 49 | 38 | 77.6 | | | | B,C | 48 | 82 | 64 | 78.0 | 1.63 ( 0.68, 3.85) | | | Non-B,C | | 51 | 35 | 68.6 | | Odds ratios for association between viral response (≥ 2 log₁₀ decrease from initial screening result) and covariates, by week, for HBeAg+ population: | Covariate | Category | Week | N | Number With ≥2-log₁₀ Decrease | Proportion (%) With ≥2-log₁₀ Decrease | Unadjusted Odds Ratio (95% CI) | | --- | --- | --- | --- | --- | --- | --- | | Race | Asian | 4 | 85 | 46 | 54.1 | 2.43 ( 1.10, 5.45) | | | Other | | 49 | 16 | 32.7 | | | | Asian | 8 | 84 | 55 | 65.5 | 1.10 ( 0.49, 2.44) | | | Other | | 49 | 31 | 63.3 | | | | Asian | 16 | 83 | 66 | 79.5 | 1.02 ( 0.38, 2.65) | | | Other | | 48 | 38 | 79.2 | | {35} Page - 36 Summary of Safety and Effectiveness Data | Covariate | Category | Week | N | Number With ≥2-log_{10} Decrease | Proportion (%) With ≥2-log_{10} Decrease | Unadjusted Odds Ratio (95% CI) | | --- | --- | --- | --- | --- | --- | --- | | | Asian | 28 | 84 | 68 | 81.0 | 1.09 (0.40, 2.85) | | | Other | | 49 | 39 | 79.6 | | | | Asian | 44 | 82 | 66 | 80.5 | 1.11 (0.41, 2.93) | | | Other | | 47 | 37 | 78.7 | | | | Asian | 48 | 84 | 65 | 77.4 | 1.51 (0.63, 3.58) | | | Other | | 49 | 34 | 69.4 | | | Sex | Male | 4 | 101 | 44 | 43.6 | 0.64 (0.27, 1.53) | | | Female | | 33 | 18 | 54.5 | | | | Male | 8 | 100 | 62 | 62.0 | 0.61 (0.23, 1.55) | | | Female | | 33 | 24 | 72.7 | | | | Male | 16 | 99 | 77 | 77.8 | 0.65 (0.17, 2.00) | | | Female | | 32 | 27 | 84.4 | | | | Male | 28 | 100 | 80 | 80.0 | 0.89 (0.26, 2.62) | | | Female | | 33 | 27 | 81.8 | | | | Male | 44 | 97 | 77 | 79.4 | 0.89 (0.26, 2.63) | | | Female | | 32 | 26 | 81.3 | | | | Male | 48 | 100 | 74 | 74.0 | 0.91 (0.32, 2.42) | | | Female | | 33 | 25 | 75.8 | | 43 {36} Page - 37 Summary of Safety and Effectiveness Data | Covariate | Category | Week | N | Number With ≥2-log_{10} Decrease | Proportion (%) With ≥2-log_{10} Decrease | Unadjusted Odds Ratio (95% CI) | | --- | --- | --- | --- | --- | --- | --- | | Age | ≤30 | 4 | 63 | 34 | 54.0 | 1.80 (0.86, 3.79) | | | >30 | | 71 | 28 | 39.4 | | | | ≤30 | 8 | 62 | 42 | 67.7 | 1.29 (0.59, 2.82) | | | >30 | | 71 | 44 | 62.0 | | | | ≤30 | 16 | 61 | 48 | 78.7 | 0.92 (0.36, 2.37) | | | >30 | | 70 | 56 | 80.0 | | | | ≤30 | 28 | 62 | 50 | 80.6 | 1.02 (0.40, 2.67) | | | >30 | | 71 | 57 | 80.3 | | | | ≤30 | 44 | 60 | 48 | 80.0 | 1.02 (0.39, 2.67) | | | >30 | | 69 | 55 | 79.7 | | | | ≤30 | 48 | 62 | 48 | 77.4 | 1.34 (0.57, 3.22) | | | >30 | | 71 | 51 | 71.8 | | | Genotype | B,C | 4 | 83 | 44 | 53.0 | 2.07 (0.95, 4.54) | | | Non-B,C | | 51 | 18 | 35.3 | | | | B,C | 8 | 82 | 53 | 64.6 | 1.00 (0.45, 2.20) | | | Non-B,C | | 51 | 33 | 64.7 | | | | B,C | 16 | 81 | 65 | 80.2 | 1.15 (0.43, 2.94) | | | Non-B,C | | 50 | 39 | 78.0 | | | | B,C | 28 | 82 | 67 | 81.7 | 1.23 (0.46, 3.18) | | | Non-B,C | | 51 | 40 | 78.4 | | | | B,C | 44 | 80 | 65 | 81.3 | 1.25 (0.47, 3.27) | | | Non-B,C | | 49 | 38 | 77.6 | | | | B,C | 48 | 82 | 64 | 78.0 | 1.63 (0.68, 3.85) | | | Non-B,C | | 51 | 35 | 68.6 | | All lower limits of the 95% confidence intervals in two tables above are smaller than 1, except for Race at Week 28 (when response defined as &lt; 2000 IU/mL) and Week 4 (when response defined as ≥ 2 log₁₀ decrease from initial screening result), which were both 1.1. This is in concordance with logistic regression analysis resulting in no statistically significant associations between the four covariates and viral load. Therefore, the virological responses at Weeks 4, 8, 16, 28 and 44 do not appear to be correlated with Race, Sex, Age, and HBV Genotype. {37} Page - 38 Summary of Safety and Effectiveness Data # Positive Predictive Value (PPV), Negative Predictive Value (NPV), and Odds Ratio (OR) Calculations for Week 48 Responses to Therapy with Respect to Viral Response at Various Times on Treatment in an HBeAg+ Population For each patient, three responses - Histological, Biochemical and HBeAg Loss - were measured at various times on treatment. These three responses were defined as positive at week 48 as follows: - Positive Histological response — improvement of histological status at Week 48 by at least 2 units of the Knodell necroinflammatory score without deterioration of the fibrosis score compared to the histological status at baseline - Positive Biochemical response — normalization of ALT test result at Week 48 compared to the biochemical status at the baseline - Positive HBeAg Loss response — HBeAg undetectable at week 48. Statistical analysis was performed to evaluate whether there is an association between each of the above Week 48 positive responses and a positive viral load response (defined as HBV DNA &lt; 2000 IU/mL or ≥ 2 log₁₀ decrease from screening) at Weeks 4, 8, 16, 28 or 44 on treatment. Conversely, statistical analysis was performed to evaluate whether there is an association between each of the above Week 48 negative responses and a negative viral load response (defined as HBV DNA ≥ 2000 IU/mL or &lt; 2 log₁₀ decrease from screening) at Weeks 4, 8, 16, 28 or 44 of treatment. Based on the information in two tables below, the viral response at Weeks 4, 8, 16, 28 and 44 is informative for predicting various responses at Week 48 (i.e., lower bound of the 95% Confidence Interval (CI) of OR exceeding 1). For the HBeAg+ population, the highest PPVs of the HPS/CTM HBV test take place with individual responses between 53.8% and 85.7%, while the highest NPVs of the HPS/CTM HBV test take place with the combination of all three responses being between 81.9% and 96.2%. This data indicates that between 81.9% and 96.2% of the HBeAg+ patients with negative early viral response (≥2000 IU/mL) are not expected to achieve all three responses to treatment at week 48 of treatment. Positive Predictive Value (PPV), Negative Predictive Value (NPV) and Odds Ratio (OR) for Three Individual Responses at Week 48 of Treatment Predicted by an Early Viral Response (&lt; 2000 IU/mL) in HBeAg+ Patients: | Week of Viral Response | Week 48 Response | NPV % (Proportion^{1}) | PPV % (Proportion^{2}) | OR (95% CI) | | --- | --- | --- | --- | --- | | 4 | Histology | 36.4 (44/121) | 66.7 (4/6) | 1.14 (0.16, 13.10) | | | Biochemical | 45.6 (57/125) | 71.4 (5/7) | 2.10 (0.33, 22.67) | | | HBeAg Loss | 73.0 (89/122) | 71.4 (5/7) | 6.74 (1.02, 72.78) | {38} Page - 39 Summary of Safety and Effectiveness Data | Week of Viral Response | Week 48 Response | NPV % (Proportion^{1}) | PPV % (Proportion^{2}) | OR (95% CI) | | --- | --- | --- | --- | --- | | 8 | Histology | 36.6 (41/112) | 71.4 (10/14) | 1.44 (0.38, 6.69) | | | Biochemical | 47.0 (55/117) | 73.3 (11/15) | 2.44 (0.67, 11.05) | | | HBeAg Loss | 75.4 (86/114) | 66.7 (10/15) | 6.14 (1.71, 24.52) | | 16 | Histology | 39.4 (39/99) | 76.0 (19/25) | 2.06 (0.71, 6.83) | | | Biochemical | 51.0 (53/104) | 80.8 (21/26) | 4.36 (1.44, 15.79) | | | HBeAg Loss | 76.2 (77/101) | 53.8 (14/26) | 3.74 (1.38, 10.11) | | 28 | Histology | 41.6 (37/89) | 78.4 (29/37) | 2.58 (1.00, 7.24) | | | Biochemical | 54.7 (52/95) | 81.1 (30/37) | 5.18 (1.96, 15.21) | | | HBeAg Loss | 85.7 (78/91) | 65.8 (25/38) | 11.54 (4.35, 31.03) | | 44 | Histology | 43.8 (32/73) | 75.5 (37/49) | 2.41 (1.02, 5.88) | | | Biochemical | 62.8 (49/78) | 85.7 (42/49) | 10.14 (3.79, 29.75) | | | HBeAg Loss | 93.2 (69/74) | 66.0 (33/50) | 26.79 (8.41, 97.69) | 1 The denominator is the number of patients predicted not to have an individual response to treatment at week 48 based on the lack of an early viral response of &lt; 2000 IU/mL; the numerator is the number of patients who had no respective week 48 response to treatment among the patients with the lack of an early viral response of &lt; 2000 IU/mL. 2 The denominator is the number of patients predicted to have an individual week 48 response to treatment based on an early viral response of &lt; 2000 IU/mL; the numerator is the number of patients who had a respective week 48 response to treatment among the patients with early viral response of &lt;2000 IU/mL. PPV, NPV and OR for a Combination of All Three Responses at Week 48 on of Treatment Predicted by an Early Viral Response (&lt; 2000 IU/mL) for an HBeAg+ Population: | Week of Viral Response | NPV % (Proportion^{1}) | PPV % (Proportion^{2}) | OR (95% CI) | | --- | --- | --- | --- | | 4 | 81.9 (104/127) | 16.7 (1/6) | 0.90 (0.02, 8.66) | | 8 | 83.9 (99/118) | 35.7 (5/14) | 2.89 (0.68, 10.85) | | 16 | 85.7 (90/105) | 36.0 (9/25) | 3.38 (1.09, 9.91) | | 28 | 91.6 (87/95) | 43.2 (16/37) | 8.29 (2.84, 25.10) | {39} Page - 40 Summary of Safety and Effectiveness Data | Week of Viral Response | NPV % (Proportion^{1}) | PPV % (Proportion^{2}) | OR (95% CI) | | --- | --- | --- | --- | | 44 | 96.2 (76/79) | 42.9 (21/49) | 19.00 (4.98, 104.34) | 1 The denominator is the number of patients predicted not to have the response to treatment at week 48 based on the lack of early viral response of &lt; 2000 IU/mL; the numerator is the number of patients who had no respective week 48 response to treatment among the patients with the lack of early viral response of &lt; 2000 IU/mL. 2 The denominator is the number of patients predicted to have week 48 response to treatment based on early viral response of &lt; 2000 IU/mL; the numerator is the number of patients who had all three responses to treatment at week 48 among the patients with early viral response of &lt; 2000 IU/mL. With the viral response defined as ≥ 2 log₁₀ decrease in value from the Screening viral load result, for the HBeAg+ population, the PPVs of HPS/CTM HBV Test are larger for the individual responses, 36.3% to 70.5%, while the NPVs are the largest for the combination of all three responses, ranging from 87.5% to 96.3%, as shown in two tables below. PPV, NPV and OR for Three Individual Responses at Week 48 on Treatment Predicted by Early Viral Response (≥ 2 log₁₀ Decrease from Screening) in HBeAg+ Patients: | Week of Viral Response | Week 48 Response | NPV % (Proportion^{1}) | PPV % (Proportion^{2}) | OR (95% CI) | | --- | --- | --- | --- | --- | | 4 | Histology | 43.1 (28/65) | 70.5 (43/61) | 1.81 (0.81, 4.05) | | | Biochemical | 58.6 (41/70) | 70.5 (43/61) | 3.38 (1.54, 7.48) | | | HBeAg Loss | 82.4 (56/68) | 43.3 (26/60) | 3.57 (1.49, 8…