← Product Code [MZP](/submissions/MI/subpart-d%E2%80%94serological-reagents/MZP) · P020022

# BAYER VERSANT (TM) HCV RNA 3.0 ASSAY (BDNA) (P020022)

_Siemens Healthcare Diagnostics, Inc. · MZP · Mar 28, 2003 · Microbiology · APWD_

**Canonical URL:** https://fda.innolitics.com/submissions/MICROBIOLOGY/subpart-d%E2%80%94serological-reagents/MZP/P020022

## Device Facts

- **Applicant:** Siemens Healthcare Diagnostics, Inc.
- **Product Code:** [MZP](/submissions/MI/subpart-d%E2%80%94serological-reagents/MZP.md)
- **Decision Date:** Mar 28, 2003
- **Decision:** APWD
- **Regulation:** 21 CFR 866.3170
- **Device Class:** Class 2
- **Review Panel:** Microbiology
- **Attributes:** Expedited Review

## Intended Use

The VERSANT® HCV RNA 3.0 Assay (bDNA) is a signal amplification nucleic acid probe assay for the quantitation of human hepatitis C viral RNA (HCV RNA) in the serum or plasma (EDTA and ACD) of HCV-infected individuals using the Bayer System 340 bDNA Analyzer. Specimens containing HCV genotypes 1-6 have been validated for quantitation in the assay. The VERSANT HCV RNA 3.0 Assay (bDNA) is intended for use as an aid in the management of HCV-infected patients undergoing anti-viral therapy. The assay measures HCV RNA levels at baseline and during treatment and is useful in predicting non-sustained virological response to HCV therapy. The results from the VERSANT HCV RNA 3.0 Assay (bDNA) must be interpreted within the context of all relevant clinical and laboratory findings. Assay performance characteristics have been established only for individuals treated with interferon alfa-2b plus ribavirin. No information is available on the assay’s predictive values when other therapies are used.

## Device Story

The VERSANT HCV RNA 3.0 Assay (bDNA) is an in vitro diagnostic test for quantifying hepatitis C viral RNA in human serum or plasma. The device uses a sandwich nucleic acid hybridization procedure; viral RNA is released from virions and captured to microwells via synthetic oligonucleotide probes. Target probes hybridize to viral RNA and pre-amplifier probes, forming a branched DNA (bDNA) complex. Alkaline phosphatase-labeled probes bind to this complex; chemiluminescent substrate (Lumi-Phos® Plus) is added, and light emission (RLUs) is measured by the Bayer System 340 bDNA Analyzer. Light intensity is proportional to HCV RNA concentration, determined against a standard curve. The assay is used by laboratory technicians to monitor viral load at baseline and during treatment. Results assist clinicians in predicting non-sustained virological response (NSVR) to anti-viral therapy, allowing for potential adjustment of treatment regimens. The device benefits patients by providing objective data to guide therapy management.

## Clinical Evidence

Clinical study of 351 subjects treated with interferon alfa-2b plus ribavirin. Evaluated baseline and during-treatment viral loads (weeks 4, 8, 12, 24) as predictors of sustained virological response (SVR) and non-sustained virological response (NSVR). Viral load thresholds (100,000 IU/mL at week 4; 10,000 IU/mL at weeks 8, 12, 24) and log-drop rules (1-log at week 4; 2-log at weeks 8, 12, 24) demonstrated negative predictive values (NPV) >95% for NSVR across genotypes. Analytical studies confirmed linearity (615–7,690,000 IU/mL), reproducibility, and specificity.

## Technological Characteristics

Sandwich nucleic acid hybridization assay; chemiluminescent detection. Components: Lysis reagent (proteinase K), synthetic oligonucleotide capture/target/amplifier probes, polystyrene microwells, alkaline phosphatase-labeled probes, Lumi-Phos® Plus substrate. Requires Bayer System 340 bDNA Analyzer. Standardized against WHO HCV RNA International Standard (conversion factor 5.2 copies/IU). Storage: 2-8°C and -60 to -80°C.

## Regulatory Identification

A nucleic acid-based hepatitis C virus (HCV) ribonucleic acid (RNA) test is identified as an in vitro diagnostic device intended for prescription use as an aid in the diagnosis of HCV infection in specified populations, and/or as an aid in the management of HCV-infected patients including guiding the selection of genotype-specific treatment in individuals with chronic HCV infection. The test is intended for use with human serum or plasma. The test is not intended for use as a donor screening test for the presence of HCV antibodies in blood, blood products, or tissue donors.

## Special Controls

*Classification.* Class II (special controls). The special controls for this device are:(1) For all nucleic acid-based HCV RNA tests, the labeling required under § 809.10(b) of this chapter must include:
(i) A prominent statement that the test is not intended for use as a donor screening test for the presence of HCV RNA from human cells, tissues, and cellular and tissue-based products.
(ii) A detailed explanation of the principles of operation and procedures for performing the assay.
(iii) A detailed explanation of the interpretation of results.
(iv) Limitations, which must be updated to reflect current clinical practice and disease presentation and management. These limitations must include, but are not limited to, statements that indicate:
(A) The specimen types for which the device has been cleared and that use of this test kit with specimen types other than those specifically cleared for this device may result in inaccurate test results.
(B) When applicable, that assay performance characteristics have not been established in populations of immunocompromised or immunosuppressed patients or, other populations where test performance may be affected.
(C) Test results are to be interpreted by qualified licensed healthcare professionals in conjunction with the individual's clinical presentation, history, and other laboratory results.
(2) For all nucleic acid-based HCV RNA tests, the design verification and validation must include:
(i) Detailed device description, including the device components, ancillary reagents required but not provided, and an explanation of the device methodology. Additional information appropriate to the technology must be included such as design of primers and probes, rationale for the selected gene targets, specifications for amplicon size, and degree of nucleic acid sequence conservation.
(ii) For devices with assay calibrators, the design and nature of all primary, secondary, and subsequent quantitation standards used for calibration as well as their traceability to a standardized reference material that FDA has determined is appropriate (
*e.g.,* a recognized consensus standard). In addition, analytical testing must be performed following the release of a new lot of the standard material that was used for device clearance or approval, or when there is a transition to a new calibration standard.(iii) Documentation and characterization (
*e.g.,* determination of the identity, supplier, purity, and stability) of all critical reagents (including nucleic acid sequences for primers and probes) and protocols for maintaining product integrity.(iv) Detailed documentation of analytical performance studies conducted as appropriate to the technology, specimen types tested, and intended use of the device, including, but not limited to, limit of detection (LoD), upper and lower limits of quantitation (ULoQ and LLoQ, respectively), linearity, precision, endogenous and exogenous interferences, cross reactivity, carryover, matrix equivalency, and sample and reagent stability. Samples selected for use in analytical studies or used to prepare samples for use in analytical studies must be from subjects with clinically relevant circulating genotypes in the United States. Cross-reactivity studies must include samples from HCV RNA negative subjects with other causes of liver disease, including autoimmune hepatitis, alcoholic liver disease, chronic hepatitis B virus, primary biliary cirrhosis, and nonalcoholic steatohepatitis, when applicable. The effect of each claimed nucleic-acid isolation and purification procedure on detection must be evaluated.
(v) Risk analysis and management strategies, such as Failure Modes Effects Analysis and/or Hazard Analysis and Critical Control Points summaries and their impact on test performance.
(vi) Final release criteria to be used for manufactured test lots with appropriate evidence that lots released at the extremes of the specifications will meet the claimed analytical and clinical performance characteristics as well as the stability claims.
(vii) Multisite reproducibility study that includes the testing of three independent production lots.
(viii) All stability protocols, including acceptance criteria.
(ix) Final release test results for each lot used in clinical studies.
(x) Analytical sensitivity and specificity of the test must be the same or better than that of other cleared or approved tests.
(xi) Lot-to-lot precision studies, as appropriate.
(3) For devices intended for the qualitative detection of HCV RNA, in addition to the special controls listed in paragraphs (b)(1) and (2) of this section, the design verification and validation must include detailed documentation of performance from a multisite clinical study. Performance must be analyzed relative to an FDA cleared or approved qualitative HCV RNA test, or a comparator that FDA has determined is appropriate. This study must be conducted using appropriate patient samples, with appropriate numbers of HCV positive and negative samples in applicable risk categories. Additional genotypes must be validated using appropriate numbers and types of samples. The samples may be a combination of fresh and repository samples, sourced from within and outside the United States, as appropriate. The study designs, including number of samples tested, must be sufficient to meet the following criteria:
(i) Clinical sensitivity of the test must have a lower bound of the 95 percent confidence interval of greater than or equal to 95 percent.
(ii) Clinical specificity of the test must have a lower bound of the 95 percent confidence interval of greater than or equal to 96 percent.
(4) For devices intended for the quantitative detection of HCV RNA, the following special controls, in addition to those listed in paragraphs (b)(1) and (2) of this section, apply:
(i) Labeling required under § 809.10(b) of this chapter must include a prominent statement that the test is not intended as a diagnostic test to confirm the presence of active HCV infection, when applicable.
(ii) Design verification and validation must include the following:
(A) Detailed documentation of the following analytical performance studies conducted as appropriate to the technology, specimen types tested, and intended use of the device, including but not limited to: LoD, ULoQ and LLoQ. LoD, LLoQ, and linearity studies must demonstrate acceptable device performance with all HCV genotypes detected by the device.
(B) Detailed documentation of clinical performance testing from either:
(
*1* ) A multisite clinical study with an appropriate number of clinical samples from chronically HCV infected patients in which the results are compared to an FDA-cleared or approved quantitative HCV RNA test, or a comparator that FDA has determined is appropriate. This study must include a sufficient number of HCV positive samples containing an analyte concentration near the LLoQ to describe performance at this level. Clinical samples must cover the full range of the device output and must be consistent with the distribution of these genotypes in the U.S. population. Clinical samples may be supplemented with diluted clinical samples for those viral load concentrations that are not sufficiently covered by natural clinical specimens, or(
*2* ) A clinical study with prospectively collected samples demonstrating clinical validity of the device.(C) Detailed documentation of a qualitative analysis near the lower end of the measuring range demonstrating acceptable performance when used as an aid in diagnosis.
(5) For devices intended for HCV RNA genotyping, in addition to the special controls listed in paragraphs (b)(1) and (2) of this section, design verification and validation must include the following:
(i) Detailed documentation of an analytical performance study demonstrating the LoD for all HCV genotypes detected by the device.
(ii) Detailed documentation, including results, of a multisite clinical study that assesses genotyping accuracy (
*i.e.,* the proportion of interpretable results that match with the reference method result) and the genotyping rate (*i.e.,* the proportion of results that were interpretable).(6) For any nucleic acid-based HCV RNA test intended for Point of Care (PoC) use, the following special controls, in addition to those listed in paragraphs (b)(1) and (2) of this section, apply:
(i) Clinical studies must be conducted at PoC sites.
(ii) Additional labeling must include a brief summary of the instructions for use that are appropriate for use in a PoC environment.

## Submission Summary (Full Text)

> This content was OCRed from public FDA records by [Innolitics](https://innolitics.com). If you use, quote, summarize, crawl, or train on this content, cite Innolitics at https://innolitics.com.
>
> Innolitics is a medical-device software consultancy. We help companies design, build, and clear FDA-regulated software and AI/ML devices, including [a PMA](https://innolitics.com/services/regulatory/), [a 510(k)](https://innolitics.com/services/510ks/), [a SaMD](https://innolitics.com/services/end-to-end-samd/), [an AI/ML medical device](https://innolitics.com/services/medical-imaging-ai-development/), or [an FDA regulatory strategy](https://innolitics.com/services/regulatory/).

{0}

8

# Summary of Safety and Effectiveness Data

## I. General Information

Device Generic Name: Assay hybridization and/or nucleic acid

Device Trade Name: Bayer VERSANT® HCV RNA 3.0 Assay (bDNA)

Applicant Name and Address: Bayer HealthCare LLC
Division Business Group Diagnostics
Street P.O. Box 2466
Berkeley, CA 94702-0466

Premarket Approval Application (PMA) Number: P020022

Date of Panel Recommendation: None

Date of Notice of Approval to Applicant: March 28, 2003

## II. Indications for Use

The VERSANT® HCV RNA 3.0 Assay (bDNA) is a signal amplification nucleic acid probe assay for the quantitation of human hepatitis C viral RNA (HCV RNA) in the serum or plasma (EDTA and ACD) of HCV-infected individuals using the Bayer System 340 bDNA Analyzer. Specimens containing HCV genotypes 1-6 have been validated for quantitation in the assay.

The VERSANT HCV RNA 3.0 Assay (bDNA) is intended for use as an aid in the management of HCV-infected patients undergoing anti-viral therapy. The assay measures HCV RNA levels at baseline and during treatment and is useful in predicting non-sustained virological response to HCV therapy. The results from the VERSANT HCV RNA 3.0 Assay (bDNA) must be interpreted within the context of all relevant clinical and laboratory findings.

Assay performance characteristics have been established only for individuals treated with interferon alfa-2b plus ribavirin. No information is available on the assay’s predictive values when other therapies are used.

## III. Contraindications: None

## IV. Warnings and Precautions

Warnings and precautions for users of the Bayer VERSANT HCV RNA 3.0 Assay (bDNA) are stated in the product labeling.

{1}

Page 2 of 21

Summary of Safety and Effectiveness Data

## V. Device Description

### Kit Configuration and Components

For detection of HCV RNA, the VERSANT HCV RNA Assay (bDNA) contains the following major components:

|  Component | Description  |
| --- | --- |
|  Lysis Reagent | proteinase K solution with sodium azide (< 0.1%) and other preservatives  |
|  Capture Probes | synthetic oligonucleotides in buffered solution with sodium azide (< 0.1%) and other preservatives  |
|  Target Probes | synthetic oligonucleotides in buffered solution with sodium azide (< 0.1%) and other preservatives  |
|  Capture Wells | polystyrene microwells coated with synthetic oligonucleotides  |
|  Pre-Amplifier Probes | synthetic oligonucleotides in buffered solution with sodium azide (< 0.1%) and other preservatives  |
|  Amplifier Probe | synthetic oligonucleotides in buffered solution with sodium azide (< 0.1%) and other preservatives  |
|  Substrate | chemiluminescent substrate (Lumi-Phos® Plus)  |
|  Label Probe | enzyme-labeled synthetic oligonucleotides in buffered solution with sodium azide (< 0.1%) and other preservatives  |
|  High Positive Control | human plasma containing recombinant single-stranded phage DNA with sodium azide (< 0.1%) and other preservatives  |
|  Low Positive Control | human plasma containing BPL-treated HCV with sodium azide (< 0.1%) and other preservatives  |
|  Negative Control | human plasma nonreactive for HCV with sodium azide (< 0.1%) and other preservatives  |
|  Standards A, D | human plasma containing recombinant single stranded phage DNA with sodium azide (< 0.1%) and other preservatives  |
|  Standards B, C | human plasma containing recombinant single stranded phage DNA with sodium azide (< 0.1%) and other preservatives  |
|  Standard E | human plasma containing recombinant single stranded phage DNA with sodium azide (< 0.1%) and other preservatives  |

In addition, the following component is required:

Bayer System 340 bDNA Analyzer with Data Management Software (DMS), version 6.11.1 or greater, for the VERSANT HCV RNA 3.0 Assay (bDNA)

## VI. Principle of Device Methodology

The VERSANT HCV RNA 3.0 Assay (bDNA) is a sandwich nucleic acid hybridization procedure for the quantitation of hepatitis C viral (HCV) RNA in human serum and plasma. After HCV genomic RNA is released from the virions, the RNA is captured to a microwell by a set of specific, synthetic oligonucleotide capture probes. A set of target probes hybridizes to both the viral RNA and pre-amplifier probes, which form the backbone of the bDNA complex. The capture probes and the target probes bind to the 5' untranslated and core regions of the HCV genome. Amplifier probes subsequently hybridize to the pre-amplifier probe, forming a branched DNA (bDNA) complex.

{2}

Page 3 of 21

Summary of Safety and Effectiveness Data

Multiple copies of an alkaline phosphatase (AP) labeled probe then hybridize to the immobilized bDNA complex. Detection is achieved by incubating the AP-bound complex with a chemiluminescent substrate. Light emission is proportional to the amount of HCV RNA present in each sample, and results are recorded in relative light units (RLUs) by the analyzer. A standard curve is prepared using light emission from standards containing known concentrations of recombinant single-stranded bacteriophage DNA. Concentrations of HCV RNA in specimens are determined from the standard curve.

## VII. Alternate Practices and Procedures

Prediction of response or nonresponse to HCV antiviral therapy has been based on the qualitative detection of HCV RNA during and at the end of anti-viral therapy. There has not been an adequate method commercially available to predict nonresponse to antiviral therapy during the course of treatment. In addition, the infecting HCV genotype, individual’s gender, previous therapy for HCV infection has been used as predictors of response or nonresponse to therapy.

## VIII. Potential Adverse Effects of the Device on Health

Since the Bayer VERSANT HCV RNA 3.0 Assay (bDNA) 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 event of an erroneous prediction of nonresponse to HCV antiviral therapy, there is a possibility that a patient would have anti-HCV treatment terminated or changed when unnecessary.

## IX. Marketing History

The VERSANT HCV RNA 3.0 Assay (bDNA) is currently approved in several countries, including France, Canada, Australia, Switzerland and Portugal. The product has not been withdrawn from the market in any country for any reason relating to the safety and effectiveness of the device.

## X. Summary of Non Clinical Studies

### Instrumentation

Software and hardware verification testing was performed for the Bayer System 340 bDNA Analyzer with Data Management Software (DMS) for the VERSANT HCV RNA 3.0 Assay (bDNA). Appropriate information and study results were furnished demonstrating that the Bayer System hardware and software, used with the VERSANT HCV RNA 3.0 Assay (bDNA), functioned as described and had appropriate safeguards.

{3}

Page 4 of 21

Summary of Safety and Effectiveness Data

## Device Stability

The stability of the VERSANT HCV RNA 3.0 Assay (bDNA) was assessed on three manufactured device lots. Evaluations of stability include testing the kits at periodic intervals through 14 months for all 3 lots. Throughout the study, assay results were compared to established specification ranges to evaluate potential assay deterioration. Results of these real time stability studies indicate that the VERSANT HCV RNA 3.0 Assay (bDNA) is stable for at least 12 months when stored at its labeled storage conditions of 2 to 8 °C (Box 1) and -60 to -80 °C (Box 2). (Note: Reagent components are shipped in two boxes. Box 1 contains reagent components that require storage between 2 and 8 °C, whereas the Box 2 reagent components require storage at -60 to -80 °C. See the devices labeling for a breakdown of the various components contained in Box 1 and Box 2.)

## Assay Standardization

The VERSANT HCV RNA 3.0 Assay (bDNA) is anchored to a reference standard (RNA transcript), which is evaluated by three independent, validated analytical methods (absorbance at 260nm, hyperchromicity and phosphate analysis). The quantity of RNA in the transcript (in copies/mL) is determined from its phosphate content. The phosphate assay utilizes dilutions of the United States National Institute of Standards and Technology (NIST) phosphate standard reference material and is, therefore, directly traceable to the US NIST.

Multiple kit lots were used to determine the conversion factor between the VERSANT HCV RNA Assay (bDNA) HCV RNA copies/mL and the World Health Organization's (WHO) HCV RNA International Units (IU). Dilutions of the WHO International Standard for hepatitis C virus (HCV) RNA¹ were tested in this evaluation. The conversion factor for changing HCV RNA copies/mL to WHO IU was determined to be 5.2 HCV RNA copies/IU.

## Assay Quantitation Limits

In the scientific literature it has is reported that the VERSANT HCV RNA has a LoD of 3,200 HCV RNA copies/mL. Bayer furnished information showing that at the level of 3,200 HCV RNA copies/mL at least 95% (lower bounds 95% confidence interval) of specimens not containing HCV RNA were designated as no HCV RNA present. However, only 39.5% of specimens containing HCV RNA were reported as HCV RNA present. It was determined that a more appropriate LoD would be 5,200 HCV RNA copies/mL. At this level at least 95% (lower bounds of 95% confidence interval) of specimens contain HCV RNA were detected. The value of 3,200 HCV RNA copies/mL was redesignated the detection limit (DC) denoting that below this level there is a high probability that HCV RNA would not be detected. The following was added to the device's labeling to educate users on the difference between DC and LoD.

- The Limit of Detection (LoD) is the lowest concentration of virus that yields an assay result at or above the Detection Cutoff in 95% of replicate determinations. The LoD for this assay is 1,000 IU/mL (5,200 HCV RNA copies/mL).
- The Detection Cutoff (DC) is the point on the assay quantitation scale such that 95% of negative specimens produce results below this cutoff with 95% confidence. The DC for this assay is 615 IU/mL (3,200 HCV RNA copies/mL).

¹ Hepatitis C virus RNA, 1st International Standard, 1997, NIBSC code 96/790

{4}

Page 5 of 21

Summary of Safety and Effectiveness Data

- The Upper Quantitation Limit (UQL) for this assay has been determined to be 7,690,000 IU/mL (40,000,000 HCV RNA copies/mL).
- The quantitation range of the assay is from 615 IU/mL (3,200 HCV RNA copies/mL) to 7,690,000 IU/mL (40,000,000 HCV RNA copies/mL).

## Assay Linearity

The assay was shown to be linear from 615 to 7,690,000 IU/ml (3,200 to 40,000,000 HCV RNA copies/mL). In addition, assay linearity was confirmed using serial dilutions of patient specimens that spanned the range from approximately 481 to 4,800,000 IU/ml (2,500 to 25,000,000 HCV RNA copies/mL) that were collected as serum and in EDTA and ACD anticoagulants. The following graph is representative of the assay linearity shown, figure 2.

![img-0.jpeg](img-0.jpeg)
Figure 2. Linearity of the VERSANT HCV RNA 3.0 Assay (bDNA)

## Comparison of Fresh Serum/Plasma Samples

Specimens were collected in serum (plastic serum separator tubes –SST-on and off the separator gel), EDTA (plastic K₂EDTA Vacutainer® tubes and plastic K₂EDTA PPT tubes with plasma stored on the gel), and ACD (glass ACD, Solution A) from 45 anti-HCV RNA nonreactive blood donors who were antibody negative for HIV, HCV, and HBsAg and from 25 HCV RNA reactive individuals with recent viral load assay results of &gt;962 IU/mL (5,000 HCV RNA copies/mL). RLUs and HCV RNA quantitations of HCV RNA nonreactive and HCV reactive serum specimens were compared to the matched anticoagulated plasma specimens. In order to establish equivalence between the test and reference conditions, the upper 95% confidence limits for the RLU ratios for HCV RNA nonreactive specimens had to be &lt;1.2 and the 90% confidence limits for the quantitation ratios for HCV RNA reactive specimens had to fall within the interval [0.70, 1.43].

The results of the study showed that there was minimal matrix effect when the anticoagulants were compared to serum (reference matrix), see table below for results containing HCV RNA.

{5}

Page 6 of 21

Summary of Safety and Effectiveness Data

## Comparison of Fresh Serum/Plasma Samples

|  Matrix | Geomean Quantitation (HCV RNA copies/mL) | Ratio Compared to Reference | Lower 90% Confidence Limit of Ratio | Upper 90% Confidence Limit of Ratio  |
| --- | --- | --- | --- | --- |
|  Serum (Reference) | 4,634,164 | -- | -- | --  |
|  Serum (On Gel) | 4,718,033 | 1.02 | 0.94 | 1.10  |
|  K_{2} EDTA Vacutainer | 5,210,881 | 1.12 | 1.04 | 1.22  |
|  K_{2} EDTA PPT (On Gel) | 5,383,630 | 1.16 | 1.07 | 1.26  |
|  ACD | 4,415,869 | 0.95 | 0.88 | 1.03  |

Slightly lower results were obtained with the specimens in ACD. This may be due to the dilutional effect of liquid anticoagulants. It is not believed that this effect will have a detrimental effect on the final assay results if all testing from an individual is performed with specimens collected in ACD. Essentially no differences were noted for the HCV RNA negative specimens tested. The above information shows that the assay should furnish relevant results for the matrices listed.

## Comparison of Stability of Serum/Plasma Samples

Specimens prepared in the same manner as section 8.6 were used to evaluate the recommended storage conditions on final assay results. The reference condition was considered to be centrifugation, removal of plasma or serum, and freezing at -60 to -80 °C within 4 hours of collection except for the PPT tubes which were centrifuged within 2 hours. This was compared to conditions in which the serum and plasma specimens were held at 2 to 8 °C for 8, 24, or 48 hours before freezing at -60 to -80 °C, see table below for results obtained with samples containing HCV RNA.

{6}

Page 7 of 21

Summary of Safety and Effectiveness Data

Comparison of Stability of Serum/Plasma Samples

|  Matrix | Storage Time | N | Geomean^{2} Quantitation (HCV RNA copies/mL) | Ratio Compared to Reference  |
| --- | --- | --- | --- | --- |
|  Serum (Off Gel) | 0 (Reference time) | 25 | 4,634,164 | ---  |
|   |  8 | 25 | 4,626,440 | 1.00  |
|   |  24 | 25 | 4,712,195 | 1.02  |
|   |  48 | 25 | 4,651,777 | 1.00  |
|  EDTA | 0 (Ref) | 25 | 5,210,881 |   |
|   |  8 | 25 | 5,286,557 | 1.01  |
|   |  24 | 25 | 5,287,621 | 1.01  |
|   |  48 | 25 | 5,357,185 | 1.03  |
|  ACD | 0 (Ref) | 25 | 4,415,869 |   |
|   |  8 | 25 | 4,387,779 | 0.99  |
|   |  24 | 25 | 4,382,524 | 0.99  |
|   |  48 | 25 | 4,418,249 | 1.00  |
|  SST (On Gel) | 0 (Ref) | 25 | 4,718,033 |   |
|   |  8 | 25 | 4,703,233 | 1.00  |
|   |  24 | 25 | 4,654,383 | 0.99  |
|   |  48 | 25 | 4,812,202 | 1.02  |
|  PPT (On Gel) | 0 (Ref) | 25 | 5,383,630 |   |
|   |  8 | 25 | 5,186,217 | 0.96  |
|   |  24 | 25 | 5,110,336 | 0.95  |
|   |  48 | 25 | 5,348,198 | 0.99  |

It appears that all sample matrices may be stored in accordance with Bayer's recommendations. Similar results were obtained with HCV RNA negative specimens.

## Evaluation of Assay Reproducibility and Linearity across Matrices

Linearity, precision, and sensitivity across matrices were assessed using a six-member panel created from pooled specimens collected in serum, K₂EDTA, and ACD from two HCV RNA-positive individuals. The pooled undiluted specimen was pre-tested using the VERSANT HCV bDNA Assay to determine its concentration. The following table shows the reproducibility, expressed as percent correlation coefficient, of the assay across matrices:

Reproducibility

|  Reproducibility Matrix:Anti-Coagulant | Level 1 (neat) | Level 2 (2 x 10^{6} HCV RNA copies/mL) | Level 3 (2 x 10^{4} HCV RNA copies/mL) | Level 4 (10^{4} HCV RNA copies/mL) | Level 5 (5 x 10^{3} HCV RNA copies/mL) | Level 6 (2.5 x 10^{3} HCV RNA copies/mL)  |
| --- | --- | --- | --- | --- | --- | --- |
|  Serum | 13.1% | 11.6% | 22.0% | 24.9% | 34.5% | 40.0%  |
|  EDTA | 14.4% | 11.3% | 22.7% | 25.3% | 29.3% | 47.4%  |
|  ACD | 12.7% | 12.3% | 27.4% | 27.2% | 34.6% | 41.1%  |

² Geomean = The geomean of n numbers is the n<sup>th</sup> root of the product, e.g., geomean for two numbers a1 and a2 is $\sqrt{a1 \cdot a2}$ for three numbers a1, a2, a3 it is $\sqrt[3]{a1 \cdot a2 \cdot a3}$, etc.

{7}

Page 8 of 21

Summary of Safety and Effectiveness Data

Linearity was evaluated based on log differences from the expected concentration. Log differences for all matrices were within ± 0.1 log.

The above studies demonstrated that the VERSANT HCV RNA Assay has reasonable reproducibility and linearity across the matrices claimed.

## Potentially Cross-Reacting and Interfering Substances

Cross-reacting and interfering substances were studied in HCV RNA-positive and HCV RNA-negative specimens to determine if these substances had an effect on assay specificity or HCV quantitation. The various substances or conditions were evaluated in 25 HCV RNA-positive and 35 to 45 HCV RNA-negative specimens, depending on the number of conditions tested. The following table shows all substances and conditions tested and their results:

Effect of Potentially Interfering Substances or Conditions on Quantitation of HCV RNA

|  Interfering Substance/Condition | Levels tested | Effect  |
| --- | --- | --- |
|  Hemoglobin
(upper limit of normal range 0.5 mg/dL) | 500 mg/dL | No effect  |
|  Conjugated Bilirubin
(upper limit of normal range = 0.3 mg/dL) | 40 mg/dL | No effect  |
|  Unconjugated Bilirubin
(upper limit of normal range = 1.1 mg/dL) | 10, 20 and 40 mg/dL | Unconjugated bilirubin at 20 and 40 mg/dL had decreased quantitation in HCV RNA-positive specimens  |
|  Triglycerides
(upper limit of normal range = 250 mg/dL) | 3000 mg/dL | No effect  |
|  Protein (upper limit of normal is ~8 g/dL) | <8, 8-9, 9-10, 10-11 g/dL | Protein at >9 g/dL had decreased quantitation in HCV RNA-positive specimens  |
|  Liver Enzymes (ALT, AST, GGT) | ALT 2x UNL; AST and GGT > UNL | No effect  |
|  Cryoprecipitates | Precipitate present, not quantified | No effect  |
|  Drugs* | 5 X peak therapeutic concentration except as noted in footnote | No effect  |
|  Disease Pathogens** | 10^{7} CFU for bacteria and yeast. Concentrations of CMV, HIV-1, HAV, EBV, and HBV noted in footnote | No effect  |

* Drugs Tested = Prednisone, ganciclovir, indinavir, lamivudine (tested at 1/2 C<sub>max</sub>), Pegylated interferon alpha 2b (PEG-INTRON), cyclosporin, acyclovir, AZT, ritonavir, Ribavirin, alpha interferon (Intron A), tacrolimus, DDI, nelfinavir, D4T, Mycophenolate mofetil, rapamycin, azathioprine, saquinavir, amantadine, trimethoprim/sulfamethoxazole
** Disease Pathogens Tested = E. coli, P. aeruginosa, K. pneumoniae, H. influenzae, E. cloacae, P. fluorescens, S. aureus, S. marcescens, S. pneumoniae, S. epidermidis, Streptococcus group B, C. albicans, HIV-1 (10<sup>6</sup> HIV-1 RNA copies/mL), HAV (10<sup>6</sup> TCID<sub>50</sub>), CMV (7 x 10<sup>5</sup> DNA copies/mL) and HBV (10<sup>6</sup> HBV DNA copies/mL)

For the substances and disease conditions tested there does not appear to be significant interference or cross-reactivity with the VERSANT HCV RNA Assay.

{8}

Page 9 of 21

Summary of Safety and Effectiveness Data

## Genotype Detection

Two studies were performed to evaluate assay performance of HCV genotypes, 1, 2, 3, 4, 5, and 6 relative to genotype 1. The first study evaluated the ability of the VERSANT HCV bDNA Assay to quantitate non-1 genotypes. The second study evaluated the precision, linearity and analytical sensitivity of the VERSANT HCV bDNA Assay when measuring non-1 genotypes.

In the first study, RNA transcripts were prepared from the 5' untranslated region of cloned examples of HCV 1, 2, 3, 4, 5, and 6. Each transcript representing an HCV genotype was diluted to produce HCV RNA quantitations at approximately 384,615 (high concentration) and 1,346 (low concentration) IU/mL (2,000,000 and 7,000 HCV RNA copies/mL) relative to the VERSANT HCV bDNA Assay.

The relative recovery was calculated for each non-1 genotype as the ratio of the percent recovery of the non-1 genotype divided by the percent recovery for 1.

All of the non-1 genotype transcripts had quantitations that were within 1.5-fold of the genotype 1 transcript and of each other.

In the second study, precision, linearity and analytical sensitivity were assessed using panels made with specimens from three individuals infected with genotypes 1a, 1b, 2a, 2b, 3a, 4a, and 6a and one individual infected with genotype 5a. Each genotype specimen³ was serially diluted to target concentrations of 385,000, 1,900, 960, and 480 IU/mL (2,000,000; 10,000; 5,000; and 2,500 HCV RNA copies/mL) as determined in the VERSANT HCV bDNA Assay. The lower concentration levels were chosen to bracket the estimated limit of detection, lower quantitation limit, and detection cutoff of the VERSANT HCV RNA 3.0 Assay (bDNA). Evaluation of the total %CVs for all the genotypes and dilution levels above the detection cutoff (615 IU/mL) showed that the assay quantitates all genotypes with acceptable precision (&lt; 40%).

The estimated LoDs of all of the genotypes tested were within 1.3-fold of genotype 1.

Since the VERSANT HCV RNA 3.0 Assay (bDNA) assay has been standardized against the WHO HCV RNA Standard, which is HCV genotype 1⁴ this would account for the variance seen with the other HCV genotypes. It is considered that the variance between HCV subtypes will not be an effectiveness issue if the device is used in accordance with its indications for use.

## Stability

The stability of the VERSANT HCV RNA 3.0 Assay (bDNA) was assessed on three manufactured kit lots. Evaluation of stability included testing the kits at periodic intervals through 14 months for all three lots. Throughout the study, assay results were compared to established specification ranges to evaluate potential assay deterioration. Results of these real-time stability studies indicate that the VERSANT HCV RNA 3.0 Assay (bDNA) is stable for at least 12 months when stored at its labeled storage conditions of 2 to 8 °C (Box 1) and -60 to -80 °C (Box 2).

---

³ One patient specimen containing genotype 4 had an HCV RNA concentration of 6,480 IU/mL and therefore could not be tested at the high target concentration of 385,000 IU/mL.

⁴ Saldanha J, Lelie N, Heath A, et. al., Establishment of the First International Standard for Nucleic Acid Amplification Technology (NAT) Assays for HCV RNA, Vox Sang 1999;76:149–158

{9}

Page 10 of 21

Summary of Safety and Effectiveness Data

Studies were conducted to ensure the shipping container would maintain the proper transport temperatures for up to 24 hours. Bayer states their shipping plan requires that domestic shipments arrive within 24 hours of shipping. The conditions tested were “summer” and “winter” conditions when the device would be shipped by overnight shipment. The shortest period of time found for transport temperatures to remain within specification was with “summer conditions” for 2 to 8 °C. This shipment went out of specification a mean of 41.8 hours. There was an equilibrium time for the containers when they were initially packed. This equilibrium period caused temperatures on the 2 to 8 °C container side to drop below 2 °C, e.g., -12 to +3 °C. Bayer states that they have internal data showing the DNA probes may be frozen and thawed multiple times.

The information provided supports a device expiration time of 12 months and a shipment time of 24 hours.

## Reproducibility and Linearity

Assay reproducibility was established using a panel with HCV target concentrations between 615 and 7,690,000 IU/mL (3,200 and 40,000,000 HCV RNA copies/mL). Two panel members, QC1 and QC3, were constructed using recombinant single-stranded bacteriophage DNA as target material, diluted in human plasma matrix. The remaining panel members were beta-propiolactone (BPL)-treated HCV-positive patient specimens spiked into HCV RNA-negative human plasma. Varying concentrations were constructed by dilution. Two operators at each of 3 external trial sites performed the testing, with each operator performing 4 assay runs with each of the 3 clinical reagent lots. Each operator used a different instrument, i.e., different serial number.

The following shows, for each panel member, the number of replicates, the concentration, total log standard deviation (SD), Within-plate percent correlation coefficient (%CV) and total %CV derived from the above study.

{10}

Page 11 of 21

Summary of Safety and Effectiveness Data

Table 5. Total Log SD, Within-Plate %CV, and Total %CV for the HCV RNA Reproducibility Panel Members

|  Panel Member | N | Concentration in IU/mL (copies/mL) | Total log SD | Within Plate %CV | Total %CV  |
| --- | --- | --- | --- | --- | --- |
|  QC1 | 216 | 6,820,720
(35,467,746) | 0.07 | 13.9% | 17.0%  |
|  QC2 | 216 | 607,756
(3,160,333) | 0.07 | 9.1% | 17.0%  |
|  DT^{1} | NA | 100,000
(520,000) | 0.07 | NA | 17.0%  |
|  QC3 | 216 | 68,207
(354,677) | 0.06 | 10.2% | 15.0%  |
|  QC4 | 216 | 60,776
(316,033) | 0.07 | 9.3% | 16.4%  |
|  DT | NA | 10,000
(52,000) | 0.09 | NA | 21.0%  |
|  QC5 | 216 | 6,078
(31,603) | 0.08 | 10.9% | 18.8%  |
|  QC6 | 216 | 1,519
(7,901) | 0.10 | 18.9% | 24.2%  |
|  QC7 | 216 | 760
(3,950) | 0.12 | 24.2% | 28.2%  |
|  DC^{2} | NA | 615
(3,200) | 0.14 | 26.9% | 32.4%  |
|  QC8 | 216 | 380
(1,975) | 0.18 | 33.1% | 42.2%  |

1 DT= Decision Threshold Data from this reproducibility study were combined with data from reproducibility studies across specimen matrices and genotypes to obtain an estimate of the assay precision profile. A quadratic mode was filled o the precision profile to determine total reproducibility across sites, lots, operators, matrices, and genotypes at the decision thresholds.
2 NA = Not applicable
3 For the Detection Cutoff (DC), the within-plate, total % CV and total log SD were obtained by interpolation.

The Limit of Detection (LoD), defined as the HCV RNA concentration that is detected in 95% of replicate determinations with respect to the detection cutoff (DC) of 615 IU/mL, was estimated to be 1,000 IU/mL (5,200 HCV RNA copies/mL).

The assay has a quantitation range of 615 to 7,690,000 IU/mL (3,200 HCV RNA copies/mL to 40,000,000 HCV RNA copies/mL). Results reported between the DC of 615 IU/mL and the LoD of 1,000 IU/mL must be interpreted with caution because the detection rate of the assay gradually decreases from 95% at 1,000 IU/mL to 39.5% at 615 IU/mL and the variability of the assay increases from 26.6% CV at 1,000 IU/mL to 32.4% CV at 615 IU/mL.

The assay is linear from 615 IU/mL to 7,690,000 IU/mL (3,200 to 40,000,000 HCV RNA copies/mL). The accuracy of the assay, defined as the percent recovery (ratio of observed mean quantitation to expected concentration), was from 92% to 108% across the quantitation range. In addition, the linearity of the assay was confirmed using serially diluted individual patient specimens collected as serum and either EDTA or ACD anticoagulants (anticoagulants claimed for use).

{11}

Page 12 of 21

Summary of Safety and Effectiveness Data

![img-1.jpeg](img-1.jpeg)

The above information demonstrates that the device has adequate reproducibility and linearity for the indication for use claimed.

## Analytical Specificity for Limit of Detection and Detection Cutoff

Analytical specificity of the VERSANT HCV RNA 3.0 Assay (bDNA) was determined using 999 anti-HCV negative serum specimens from men and women of which the majority were repeat volunteer blood donors. The specimens were determined to be negative for antibodies to HCV using an FDA approved HCV ELISA assay. Samples were tested across four testing sites using three kit lots of the VERSANT HCV RNA 3.0 Assay (bDNA). Of the 999 samples tested, 992 (99.3%) produced results below the limit of detection (LoD) (&lt;1,000 HCV RNA IU/mL or 5,200 HCV RNA copies/mL) with a lower 95% confidence limit of 98.7%. Alternatively, of the 999 samples tested, 987 (98.8%) produced results below the detection cutoff (&lt;615 IU/mL or 3,200 HCV RNA copies/mL) with a lower 95% confidence limit of 98.1%.

## Within Subject Variability

Since this assay is quantitative and will be used to test sequential specimens from the same individual it was necessary to establish within subject variability. Twenty-nine subjects (16 males and 13 females) with clinically stable, chronic HCV infection were enrolled in this study. Serum specimens were drawn from subjects weekly for 6 to 8 weeks with an average of 7 weeks. Enrollment and continued eligibility criteria for the study were: 1) No antiviral therapy or clinical events within the 3 months prior to the start of the study and 2) No initiation of antiviral therapy or clinical events during the 8-week study period. The ultimate objective of the study was to determine the minimum fold-change (ratio) between two successive measurements that is unlikely to occur by chance. The minimum fold-change was determined using the estimated total variability which is composed of assay and biologic variability. The fold-change for the VERSANT HCV RNA 3.0 Assay (bDNA) was estimated to be 2.6-fold or 0.41 logs. The biologic variability alone was 2.1 fold. The largest ratio between the minimum and maximum viral load within any subject was 1.6 logs and the median change across all subjects was 0.27 logs

## XI. Summary of Clinical Studies

To determine the distribution of HCV RNA in chronically infected individuals, HCV RNA levels were measured by the Bayer VERSANT HCV RNA (bDNA) Assay in 650 chronically infected patients who were not undergoing therapy at the time of specimen collection. In 431 patients, quantification values were obtained from baseline specimens just before the initiation of anti-HCV therapy; the remaining 219 patients were not subsequently put on therapy. The figure below shows the distribution of HCV RNA log quantitations in HCV RNA IU/mL for the combined study population. The median quantitation was 5.8 log IU/mL with 5th and 95th percentiles of 4.5 log

{12}

Page 13 of 21

Summary of Safety and Effectiveness Data

IU/mL and 6.6 log HCV RNA IU/mL in the log scale, or a median of 694,566 HCV RNA IU/mL with 5th and 95th percentiles of 33,570 and 3,681,828 HCV RNA IU/mL.

Figure 1. Distribution of log HCV RNA in IU/mL (N = 650)
![img-2.jpeg](img-2.jpeg)
DC = Detection cutoff
UQL = Upper quantitation limit.

Bayer conducted a study in a group of individuals undergoing HCV antiviral therapy (interferon alfa-2b plus ribavirin – Rebetron®). The objectives of the study were to determine if HCV RNA quantitations or changes in quantitations were predictive of sustained virological response (SVR) and non-sustained virological response (NSVR) at post-treatment follow-up (PTF).

Retrospectively collected specimens from 351 subjects treated with interferon alfa-2b plus ribavirin were assayed across six sites in North America (5) and Europe (1). Specimens were collected at baseline, weeks 4, 8, 12, 24 (for subjects who had 48 weeks of treatment), end of treatment (EOT), and post-treatment follow up (PTF). SVR was defined as having no detectable HCV RNA at PTF using a sensitive qualitative HCV RNA test. The population included previously treated and treatment-naive subjects. Table 4 shows the demographics for the Bayer study population.

{13}

Page 14 of 21

Summary of Safety and Effectiveness Data

Table 4. Description of Study Population

|  Characteristic | Number of Subjects (N = 351) | Percent of Subjects  |
| --- | --- | --- |
|  Race  |   |   |
|  Asian | 14 | 4.0% (14/351)  |
|  African-American | 4 | 1.1% (4/351)  |
|  Caucasian | 235 | 67.0% (235/351)  |
|  Hispanic | 16 | 4.6% (16/351)  |
|  Other^{1} | 82 | 23.4% (82/351)  |
|  Age Range  |   |   |
|  20-24 | 2 | 0.6% (2/351)  |
|  25-34 | 29 | 8.3% (29/351)  |
|  35-44 | 135 | 38.5% (135/351)  |
|  45-54 | 130 | 37.0% (130/351)  |
|  55-59 | 28 | 8.0% (28/351)  |
|  60-64 | 16 | 4.6% (16/351)  |
|  65-74 | 11 | 3.1% (11/351)  |
|  Genotype^{2}  |   |   |
|  1 | 239 | 68.1% (239/351)  |
|  2 | 50 | 14.2% (50/351)  |
|  3 | 58 | 16.5% (58/351)  |
|  4 | 6 | 1.7% (6/351)  |
|  5 | 0 | 0.0% (0/0)  |
|  6 | 2 | 0.6% (2/351)  |
|  6-Month Treatment Duration | 162 | 46.1% (162/351)  |
|  Genotype 1 | 85 | 52.5% (85/162)  |
|  Genotype non-1 | 77 | 47.5% (77/162)  |
|  1-Year Treatment Duration | 189 | 53.8% (189/351)  |
|  Genotype 1 | 154 | 81.5% (154/189)  |
|  Genotype non-1 | 35 | 18.5% (35/189)  |
|  Previous Therapy | 76 | 21.6% (76/351)  |
|  Males | 54 | 71.0% (54/76)  |
|  Females | 22 | 29.0% (22/76)  |
|  No Previous Therapy | 275 | 78.4% (275/351)  |
|  Males | 184 | 66.9% (184/275)  |
|  Females | 91 | 33.1% (91/275)  |

1 Subjects treated and tested at the site in France did not have race recorded for legal reasons.
2 Percentages add up to greater than 100% because 4 subjects had mixed infections.

Table 5 shows the percent of sustained virological responders and odds ratios for prediction of SVR using baseline characteristics and treatment duration variables for the study population. The variables that were the most predictive of SVR (i.e., had the highest odds ratio) were genotype and baseline HCV RNA with a threshold of 100,000 (10⁵) IU/mL (520,000 HCV RNA copies/mL).

{14}

Page 15 of 21

Summary of Safety and Effectiveness Data

Table 5. Percent Response and Odds Ratios for Prediction of SVR Using Baseline Characteristics¹

|  Predictor | Variable | N | Percent with SVR | Odds Ratio (95% Confidence Interval)  |
| --- | --- | --- | --- | --- |
|  Genotype | Non-1 | 112 | 70.5 | 4.3 (2.6, 7.1)  |
|   |  1 | 239 | 36.0  |   |
|  Baseline HCV RNA | ≤10⁵ IU/mL | 63 | 66.7 | 2.7 (1.5, 5.0)  |
|   |  >10⁵ IU/mL | 288 | 42.7  |   |
|  Treatment Duration | 48 Weeks | 189 | 54.0 | 1.8 (1.2, 2.9)  |
|   |  24 Weeks | 162 | 38.9  |   |
|  Liver Disease Stage² | NC | 288 | 49.3 | 1.7 (0.9,3.1)  |
|   |  C | 63 | 36.5  |   |
|  Age | <40 | 77 | 57.1 | 1.7 (1.0, 2.9)  |
|   |  ≥40 | 274 | 44.2  |   |
|  Previous Treatment | Naive | 275 | 48.7 | 1.4 (0.8, 2.4)  |
|   |  Experienced | 76 | 40.8  |   |
|  Gender³ | Female | 113 | 49.6 | 1.2 (0.7, 1.9)  |
|   |  Male | 238 | 45.8  |   |

1 Data shown are from the Bayer Clinical Study
2 NC = non-cirrhotic, C = cirrhotic
3 Early studies reported that gender was a statistically significant predictor of SVR. However, later studies have shown that gender and body weight are confounded and, when ribavirin dose is based on body weight, gender is no longer predictive of SVR.

For results obtained with the VERSANT HCV RNA 3.0 Assay (bDNA) in this study population, viral load decision thresholds ranging from 615 to 400,000 IU/mL and log drop in viral load (baseline to Week 4, 8, 12, or 24) thresholds ranging from 1 to 3 logs were evaluated for their ability to predict SVR and NSVR. Optimal viral load and log drop thresholds were selected at each time point based on the largest number of predicted NSVR and negative predictive values (NPV) ≥ 95%. Table 6 defines the final prediction rules selected.

Table 6. Definition of Prediction Rules.

|  Week | Type of Rule | Prediction Rule | Predicted Non-sustained Virological Responders | Predicted Sustained Virological Responders¹  |
| --- | --- | --- | --- | --- |
|  4 | Viral Load Threshold | 100,000 IU/mL² | ≥ 100,000 IU/mL | < 100,000 IU/mL  |
|   |  Log Drop¹ | 1 Log Drop | ≤ 1 log and ≥ 1,000 IU/mL² | > 1 log or < 1,000 IU/mL  |
|  8, 12 and 24 | Viral Load Threshold | 10,000 IU/mL² | ≥ 10,000 IU/mL | < 10,000 IU/mL  |
|   |  Log Drop | 2 Log Drop | ≤2 log and ≥1,000 IU/mL | > 2 log or < 1000 IU/mL  |

1 Log drops are calculated relative to pretreatment baseline viral load
2 100,000 IU/mL = 520,000 HCV RNA copies/mL, 10,000 IU/mL = 52,000 HCV RNA copies/mL, 1,000 IU/mL = 5,200 HCV RNA copies/mL

{15}

Page 16 of 21

Summary of Safety and Effectiveness Data

Table 7 shows the above rules applied to the study population, the NPV and positive predictive values (PPV) of the viral load and log drop rules for Weeks 4, 8, 12, and 24 (for subjects receiving 48 weeks of treatment) and the corresponding adjusted and unadjusted odds ratios for SVR. The adjusted odds ratios are based on multivariate logistic regression models. The final model selected adjusted for the following variables: genotype, liver disease stage, length of therapy, previous treatment, and age.

Table 7. Predictors of Sustained and Non-Sustained Virological Response at Weeks 4, 8, 12, and 24 and Corresponding Odds Ratios for SVR

|   | Negative Predictive Value^{1} | Positive Predictive Value^{2} |   |   |   |   |   |
| --- | --- | --- | --- | --- | --- | --- | --- |
|   |   |   |   |  Odds Ratio^{3}  |   |   |   |
|  Week | Prediction Rules | NPV (95% CI) | N | PPV (95% CI) | N | Unadjusted | Adjusted  |
|  4 | 100,000 IU/mL | 96.6 (88.3-99.6) | 57/59 | 55.5 (49.2-61.7) | 142/256 | 35.5 | 26.4  |
|   |  1 Log Drop^{4} | 94.0 (86.7-98.0) | 79/84 | 60.2 (53.5-66.5) | 139/231 | 23.9 | 22.3  |
|  8 | 10,000 IU/mL | 98.7 (92.8-100.0) | 74/75 | 66.5 (59.6-73.0) | 135/203 | 146.9 | 117.8  |
|   |  2 Log Drop | 97.5 (91.2-99.7) | 77/79 | 67.3 (60.3-73.8) | 134/199 | 79.4 | 69.1  |
|  12 | 10,000 IU/mL | 97.1 (90.1-99.7) | 68/70 | 59.1 (52.6-65.3) | 143/242 | 49.1 | 29.8  |
|   |  2 Log Drop | 97.4 (90.9-99.7) | 75/77 | 60.9 (54.3-67.1) | 143/235 | 58.3 | 43.1  |
|  24^{5} | 10,000 IU/mL | 100 (81.5-100) | 18/18 | 64.5 (54.6-73.5) | 69/107 | NA | NA  |
|   |  2 Log Drop | 100 (82.4-100) | 19/19 | 65.1 (55.2-74.1) | 69/106 | NA | NA  |

1 NPV = Percent of non-sustained virological responders out of the patients who are predicted non-sustained virological responders; i.e. HCV RNA positive 6 months post-therapy.
2 The positive predictive value (PPV) of HCV RNA levels is not intended to be used to guide therapy. PPV = Percent of sustained virological responders out of the patients who are predicted sustained virological responders; i.e. HCV RNA negative 6 months post-therapy.
3 Odds ratios cannot be calculated if the NPV is 100%. Those cases are indicated by NA. All odds ratios were significant (p &lt; .001).
4 None of the log drop rules at Week 4 attained an NPV of 95%. The 1-log drop had the highest NPV.
5 For subjects receiving 48 weeks of treatment

To determine if there was a difference for the VERSANT HCV RNA 3.0 Assay (bDNA) to predict SVR or NSVR in individuals where the HCV genotype was known Bayer evaluated the study data broken out by HCV genotype (1 and non-1). Table 8 shows prediction for individuals infected with HCV genotype 1 for NSVR and SVR.

{16}

Page 17 of 21

Summary of Safety and Effectiveness Data

Table 8. HCV Genotype 1: Predictors of SVR and NSVR at Weeks 4, 8, 12, and 24

|  Week | Prediction Rules | Negative Predictive Value^{1} |   | Positive Predictive Value^{2}  |   |
| --- | --- | --- | --- | --- | --- |
|   |  | NPV (95% CI) | N | PPV (95% CI) | N  |
|  4 | 100,000 IU/mL | 96.5
(87.9-99.6) | 55/57 | 46.6
(38.7, 54.6) | 75/161  |
|   |  1 Log Drop^{3} | 93.3
(85.1-97.8) | 70/75 | 50.3
(41.9, 58.8) | 72/143  |
|  8 | 10,000 IU/mL | 98.5
(92.1-100.0) | 67/68 | 57.6
(48.2, 66.7) | 68/118  |
|   |  2 Log Drop | 97.2
(90.3-99.7) | 70/72 | 58.8
(49.2, 67.9) | 67/114  |
|  12 | 10,000 IU/mL | 96.9
(89.3-99.6) | 63/65 | 49.0
(40.9, 57.2) | 75/153  |
|   |  2 Log Drop | 97.1
(90.1-99.7) | 68/70 | 50.7
(42.3, 59.0) | 75/148  |
|  24^{4} | 10,000 IU/mL | 100
(81.5-100) | 18/18 | 59.3
(48.2, 69.8) | 51/86  |
|   |  2 Log Drop | 100
(82.4-100) | 19/19 | 60.0
(48.8, 70.5) | 51/85  |

1 NPV = Percent of non-sustained virological responders out of the patients who are predicted non-sustained virological responders; i.e. HCV RNA positive 6 months post-therapy.
2 The positive predictive value (PPV) of HCV RNA levels is not intended to be used to guide therapy. PPV = Percent of sustained virological responders out of the patients who are predicted sustained virological responders; i.e. HCV RNA negative 6 months post-therapy.
3 None of the log drop rules at Week 4 attained an NPV of 95%. The 1-log drop had the highest NPV.
4 For subjects receiving 48 weeks of treatment

Table 9 shows prediction for individuals infected with HCV genotype non-1 for NSVR and SVR.

{17}

Page 18 of 21

Summary of Safety and Effectiveness Data

Table 9. HCV Genotype Non-1: Predictors of SVR and NSVR at Weeks 4, 8, 12, and 24

|   |   | Negative Predictive Value¹ |   | Positive Predictive Value²  |   |
| --- | --- | --- | --- | --- | --- |
|  Week | Prediction Rules | NPV (95% CI) | N | PPV (95% CI) | N  |
|  4 | 100,000 IU/mL | 100
(15.8, 100) | 2/2 | 70.5
(60.3, 79.4) | 67/95  |
|   |  1 Log Drop | 100
(66.4, 100) | 9/9 | 76.1
(65.9, 84.6) | 67/88  |
|  8 | 10,000 IU/mL | 100
(59.0, 100) | 7/7 | 78.8
(68.6, 86.9) | 67/85  |
|   |  2 Log Drop | 100
(59.0, 100) | 7/7 | 78.8
(68.6, 86.9) | 67/85  |
|  12 | 10,000 IU/mL | 100
(47.8, 100) | 5/5 | 76.4
(66.2, 84.8) | 68/89  |
|   |  2 Log Drop | 100
(59.0, 100) | 7/7 | 78.2
(68.0, 86.3) | 68/87  |
|  24³ | 10,000 IU/mL | NA | NA | 85.7
(63.7, 97.0) | 18/21  |
|   |  2 Log Drop | NA | NA | 85.7
(63.7, 97.0) | 18/21  |

1 NPV = Percent of non-sustained virological responders out of the patients who are predicted non-sustained virological responders; i.e. HCV RNA positive 6 months post-therapy.
2 The positive predictive value (PPV) of HCV RNA levels is not intended to be used to guide therapy. PPV = Percent of sustained virological responders out of the patients who are predicted sustained virological responders; i.e. HCV RNA negative 6 months post-therapy.
3 For subjects receiving 48 weeks of treatment.

The NPVs for genotype non-1 for both viral load and log-drop rules were 100% at Weeks 4, 8, and 12 but sample sizes for subjects with NSVR were small, ranging from 2 to 9 patients. It is believed that the lower sample sizes for subjects with NSVR reflect the fact that subjects infected with non-1 genotypes have a better response to therapy than subjects infected with genotype 1.

From the information above it can be seen that the NPVs of the viral load prediction rules were &gt;95% at all four timepoints, independent of genotype. The highest NPV of the log drop in viral load rule was 93.3% at Week 4, using a 1-log drop. The NPVs for a 2-log drop at weeks 8, 12, and 24 were &gt;95% at all three timepoints, independent of genotype. While the PPVs were lower than the NPVs, ranging on average from 55% to 67% for both viral load threshold and log drop, they were slightly higher than the expected percent of SVR of 47% at baseline than if other predictive factors, e.g., HCV genotype, baseline viral load, were taken into consideration.

The data from this study using a 2-log drop rule are consistent with those reported in the National Institutes of Health (NIH) Consensus Development Conference Statement, Management of Hepatitis C: 2002.⁵ The NIH statement was based primarily on study populations treated with pegylated-interferon plus ribavirin.

Based on the clinical studies mentioned above Bayer established the following interpretation of result table showing probabilities for treatment failure.

⁵ National Institutes of Health Consensus, Management of Hepatitis C: 2002, June 10-12, 2002; Preliminary Draft Statement, June 12, 2002; Final Statement, Revisions made September 12, 2002

{18}

Page 19 of 21

Summary of Safety and Effectiveness Data

## Interpretation rules using absolute viral load:

- At week 4, if HCV RNA is ≥100,000 IU/mL, then there is a 97% chance of treatment failure.
- At week 8, if HCV RNA is ≥10,000 IU/mL, then there is a 99% chance of treatment failure.
- At week 12, if HCV RNA is ≥10,000 IU/mL, then there is a 97% chance of treatment failure.
- At week 24, if HCV RNA is ≥10,000 IU/mL, then there is a 100% chance of treatment failure.

## Interpretation rules using decrease in viral load compared with pretreatment baseline level:

- At week 4, if HCV RNA is ≥10% (≤1 log drop) of its pretreatment baseline level and ≥1,000 IU/mL, then there is a 94% chance of treatment failure.
- At week 8, if HCV RNA is ≥1% (≤2 log drop) of its pretreatment baseline level and ≥1,000 IU/mL, then there is a 97% chance of treatment failure.
- At week 12, if HCV RNA is 1% (≤2 log drop) of its pretreatment baseline level and ≥1,000 IU/mL, then there is a 97% chance of treatment failure.
- At week 24, if HCV RNA is ≥1% (≤2 log drop) of its pretreatment baseline level and ≤1,000 IU/mL, then there is a 100% chance of treatment failure.

It is recommended that the above interpretations be included with the patient’s report.

From the above information it appears that the Bayer VERSANT HCV RNA (bDNA) Assay should be safe and effective when used in the population indicated and according to its directions for use.

## XII. Conclusions Drawn from Studies

The described preclinical and clinical studies demonstrated the following:

- The detection cutoff of 615 IU/mL (3,200 HCV RNA copies/mL) and a limit of detection of 1000 IU/mL (5,200 HCV RNA copies/mL) were confirmed, as shown in the Analytical Specificity and Reproducibility Studies
- The assay is linear and reproducible from 615 IU/mL to 7,690,000 IU/mL (3,200 c/mL to 40,000,000 copies/mL), which defines the quantitation range of the assay.
- The endogenous substances hemoglobin, conjugated bilirubin, triglycerides, elevated liver enzymes, and cryoprecipitates appear not to interfere with assay performance. Unconjugated bilirubin did not appear to interfere with assay results in HCV RNA-negative specimens, but levels &gt;10 mg/dL resulted in lower quantitations in HCV RNA-positive specimens. A level of 10 mg/dL unconjugated bilirubin is well outside the upper limit of the normal range (~1.1 mg/dL) and it is believed would rarely be encountered. Protein levels above 9 g/dL were associated with decreased HCV quantitation in HCV-positive specimens.
- None of the therapeutic drugs tested interfered with assay performance, nor did viruses, bacteria, or yeast that were tested.

{19}

Page 20 of 21

Summary of Safety and Effectiveness Data

- Specimens undergoing up to 4 freeze-thaw cycles appeared to have no change in their final result.

- Serum and K₂ EDTA and ACD Solution A anticoagulants can be used for specimen collection for the VERSANT HCV bDNA Assay. Serum and plasma from EDTA and ACD Vacutainer tubes can be held at 2-8 °C for up to 48 hours before freezing at −60 to −80 °C.

- RNA transcripts from HCV genotypes 2, 3, 4, 5, and 6 quantitated within 1.5-fold of genotype 1. Dilutional panels prepared from specimens collected from patients with each of these different genotypes were found to have reproducibility, linearity, and analytical sensitivity similar to those of genotype 1.

## Clinical Studies

- Study results indicated that virtually all subjects with chronic HCV infection have HCV RNA concentrations that are quantifiable in the VERSANT HCV bDNA Assay. Greater than 99% of the baseline specimens from subjects enrolled in HCV treatment had viral loads &gt; 615 IU/mL (3,200 copies/mL), and 94% of a population of subjects with chronic HCV infection, not in treatment, had quantifiable levels of HCV RNA. There were no apparent differences in quantitations between males and females that would place limitations on the use of the assay.

- The specificity study results showed there were no false reactive results for the diseases tested.

- The within-subject variability data, when combined with estimates of variance from the reproducibility study, showed that a fold-change of 2.6 (0.41 log) or greater was unlikely to occur by chance and could be used as an indicator of a true change in HCV RNA viral load. The biologic variability alone was 2.1 fold.

- The clinical study showed that both baseline and during-treatment viral loads were statistically significant predictors of non-sustained viral response. At assessment weeks 4, 8, 12, and 24, during-treatment viral loads provided negative predictive values of &gt; 95% for subjects who were not sustained virological responders. Log drops in viral loads also provided NPVs of &gt; 95% at weeks 8, 12, and 24.

## Safety

As a diagnostic test, the VERSANT HCV RNA 3.0 (bDNA) assay involves removal of blood from an individual for testing purposes. The test, therefore, presents no more safety hazard to an individual being tested than other tests where blood is removed.

## Benefit/Risk

The submitted clinical studies demonstrate that the VERSANT HCV RNA 3.0 (bDNA) Assay, when compared to current methods for assessing response to anti-viral therapy, has a similar ability to predict nonresponse to antiviral for individuals infected with HCV. It has been shown that the device has no demonstrable cross-reactivity with viruses or organisms that may cause clinical hepatitis. Therefore, this device should benefit the physician in predicting antiviral response in individuals chronically infected with HCV.

{20}

Page 21 of 21

Summary of Safety and Effectiveness Data

The results of the preclinical and clinical laboratory studies the VERSANT RNA 3.0 (bDNA) Assay, provide reasonable assurance the device is safe and effective when used in accordance with the device labeling.

## XIII. PANEL RECOMMENDATION

In accordance with the provisions of section 515(c)(2) of the act as amended by the Safe Medical Devices Act of 1990, this PMA was not referred to the Microbiology Advisory Panel, an FDA advisory committee, for review and recommendation because the information in the PMA substantially duplicates information previously reviewed by this panel.

## XIV. CDRH Decision

FDA issued an approval order on March 28, 2003.

The applicant’s manufacturing facility inspected on February 20, 2002 and was found to be in compliance with the Quality Systems Regulation (21 CFR 820).

## XV. Approval Specifications

Directions for use: See Labeling

Hazards to Health from Use of the Device: See Indications, Contraindications, Warnings, precautions and Adverse Events in the labeling.

Postapproval Requirements and Restrictions: See approval order.

---

**Source:** [https://fda.innolitics.com/submissions/MICROBIOLOGY/subpart-d%E2%80%94serological-reagents/MZP/P020022](https://fda.innolitics.com/submissions/MICROBIOLOGY/subpart-d%E2%80%94serological-reagents/MZP/P020022)

**Published by [Innolitics](https://innolitics.com)** — a medical-device software consultancy. We help companies design, build, and clear FDA-regulated software and AI/ML devices. If you're preparing [a PMA](https://innolitics.com/services/regulatory/), [a 510(k)](https://innolitics.com/services/510ks/), [a SaMD](https://innolitics.com/services/end-to-end-samd/), [an AI/ML medical device](https://innolitics.com/services/medical-imaging-ai-development/), or [an FDA regulatory strategy](https://innolitics.com/services/regulatory/), [get in touch](https://innolitics.com/contact).

**Cite:** Innolitics at https://innolitics.com