• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Vaccines, Blood & Biologics

  • Print
  • Share
  • E-mail

Summary of Safety and Effectiveness - VITROS Immunodiagnostics Products Anti-HIV 1+2 Calibrator VITROS Immunodiagnostics Products Anti-HIV 1+2 Reagent Pack

1   General Information

Name and Address of Applicant

Ortho-Clinical Diagnostics, Inc
100 Indigo Creek Drive
Rochester NY 14626-5101

Device Trade Name

VITROS Immunodiagnostic Products Anti-HIV 1+2 Reagent Pack
VITROS Immunodiagnostic Products Anti-HIV 1+2 Calibrator

Classification (Generic) Name of Device

Antibody to Human Immunodeficiency Virus types 1 and/or 2 (Anti-HIV 1+2) assay

PMA Number: BP050051

Date of Panel Recommendation: N/A

Date of Notice of Approval To Applicant: N/A

2   Device Description

2.1 Principle of Device Methodology

The VITROS Immunodiagnostic System (VITROS Analyzer) allows for the determination of analytes in human samples (for example, serum and plasma). All assays on the VITROS Analyzer employ an enhanced chemiluminescence detection reaction. The Analyzer is fully automated with a refrigerated on board assay storage system. All standard bar code symbologies are supported by the Analyzer, which has a throughput of up to 90 assays per hour. The Analyzer also provides menu driven software, which can be accessed from a high-resolution touch screen monitor.

The VITROS Anti-HIV 1+2 assay is performed using the VITROS Anti-HIV 1+2 Reagent Pack and the VITROS Anti-HIV 1+2 Calibrator on the VITROS Immunodiagnostic System. An immunometric bridging technique is used; this involves a two-stage reaction. In the first stage HIV antibody present in the sample binds with HIV recombinant antigen coated on the wells. Unbound sample is removed by washing. In the second stage horseradish peroxidase (HRP)-labeled recombinant HIV antigens are added in the conjugate reagent. The conjugate binds specifically to any human anti-HIV 1 or anti-HIV 2 (IgG and IgM) captured on the well in the first stage. Unbound conjugate is removed by washing.

The bound HRP conjugate is measured by a luminescent reaction. A reagent containing luminogenic substrates (a luminol derivative and a peracid salt) and an electron transfer agent is added to the wells. The HRP in the bound conjugate catalyzes the oxidation of the luminol derivative, producing light. The electron transfer agent (a substituted acetanilide) increases the level of light produced and prolongs its emission. The VITROS Analyzer reads the light signals. The amount of HRP conjugate bound is indicative of the level of anti-HIV 1+2 present.

2.2 Kit Configuration and Components

For detection of anti-HIV 1+2, the VITROS System is comprised of the following:

  • VITROS Immunodiagnostic Products Anti-HIV 1+2 Reagent Pack (VITROS Anti-HIV 1+2 Reagent Pack) and VITROS Immunodiagnostic Products Anti-HIV 1+2 Calibrator (VITROS Anti-HIV 1+2 Calibrator) together comprise the VITROS Anti-HIV 1+2 assay.

     

The VITROS Anti-HIV 1+2 Reagent Pack is composed of three components:

  • Conjugate reagent [HRP-HIV 1 and HRP-HIV 2 recombinant antigens derived from yeast, with fetal calf serum in buffer with anti-microbial agent (Kathon)]
  • Assay reagent [buffer with and anti-microbial agent (Kathon)]
  • Coated microwells [HIV 1 and 2 recombinant antigens derived from yeast].

The VITROS Anti-HIV 1+2 Calibrator contains:

  • Anti-HIV 1+2 positive plasma in anti-HIV 1+2 negative human plasma with anti-microbial agent (Kathon). The Calibrator is supplied ready for use.

     

In addition, the following components are required:

  • VITROS ECi Immunodiagnostic System (VITROS Analyzer) - dedicated instrumentation, cleared by the FDA as an immunodiagnostic analyzer (K962919/S1), which provides automated analysis of the VITROS assays.

     

    In addition, the most current documentation supporting the use of this analyzer with 5 previously approved hepatitis assay PMAs has been reviewed by CDRH as part of these approvals. The most recent approval was obtained March 2004 (P030026). The documentation was provided following "Guidance for Content of Premarket Submissions for Software Contained in Medical Devices (May 1998) at a MODERATE level of concern.

     

  • VITROS Immunodiagnostic Products Signal Reagent and VITROS Immunodiagnostic Products Universal Wash Reagent are Common Reagents used in all VITROS System assays, which were cleared as part of the VITROS Immunodiagnostic Products Total T3 510(k) Premarket Notification (K964310).

     

2.3 Contraindications, Warnings and Precautions

For in vitro diagnostic use only.

Warnings and precautions for users of the VITROS Anti-HIV 1+2 Reagent Pack and Calibrator are stated in the respective product labeling.

2.4 Alternate Practices and Procedures

Determining the presence of anti-HIV 1+2 in patients may be achieved by using a variety of commercially available, FDA licensed serological tests. Additionally, when test results are used in conjunction with a physician's assessment and other laboratory test results, infection with HIV can be identified.

2.5 Marketing History

Below is a table describing the countries where the VITROS Anti-HIV 1+2 Reagent Pack and Calibrator are currently available through September of 2007.

Argentina Japan
Armenia Korea
Australia Kyrgystan
Azerbaijan Lebanon
Bangladesh Lichtenstein
Belarus Malaysia
Bermuda Maldives
Bolivia Martinique
Brazil Moldova
Brunei New Zealand
Burma ( Myanmar) Nicaragua
Canada North & Central Africa
Chile Oman
Colombia Panama
Costa Rica Paraguay
Croatia Peru
Dominican Republic Philippines
Ecuador Reunion
Egypt Romania
El Salvador Russia
European Union (CE) 27 countries Saudi Arabia
French Antilles Singapore
French Guayana South Africa
Georgia Sri Lanka
Guadeloupe Taiwan
Guatemala Tajikistan
Haiti Thailand
Honduras Trinidad and Tobago
Hong Kong Turkey
Iceland Turkmenistan
India UAE ( Arabian Gulf)
Indonesia Uruguay
Israel Venezuela
Jamaica Vietnam

2.6 Potential Adverse Effects of the Device on Health

Since the VITROS Immunodiagnostic Products Anti-HIV 1+2 Reagent Pack and VITROS Immunodiagnostic Products Anti-HIV 1+2 Calibrator are 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.

Antibodies against HIV 1+2 if falsely detected may lead to an erroneous Anti-HIV 1+2 result and confusion in the diagnosis and treatment of HIV and subsequently AIDS. A false reactive result may lead to a diagnosis of HIV infection, and administration of immunosuppressive drugs potentially harmful to the patient.

If an Anti-HIV 1+2 reactive patient is not identified, then there is a safety concern for the patient and a public health concern. A false negative result could delay the diagnosis of infection with HIV, which requires early aggressive therapy to prevent its progression to AIDS. A false negative result could also prevent a patient from being alerted to avoid high-risk behaviors that may facilitate viral transmission to others and become a public health concern.

2.7 Summary of Non-Clinical Studies

Instrumentation

Software and hardware verification testing was performed for the Vitros ECi/ECiQ Immunodiagnostic System (VITROS Analyzer). Appropriate information and study results were furnished demonstrating that the VITROS Analyzer hardware and software, used with the VITROS Immunodiagnostic Products Anti-HIV 1+2 Reagent Pack and VITROS Immunodiagnostic Products Anti-HIV 1+2 Calibrator, functioned as described and had appropriate safeguards.

Comparison of Plasma Collected in Glass or Plastic with Serum Collected in Glass.

For assessing clinical equivalence of the VITROS assay sensitivity using samples collected in alternative matrix/plastic tube combinations compared with serum samples collected in glass, differences between plastic and glass S/C values greater than 1.1022 or less than 1.1022 were considered clinically relevant. Twenty-five HIV antibody spiked samples collected in the study comparing fresh serum and plasma were used to establish acceptance or rejection of the hypothesis that alternative matrix/plastic tube combinations are equivalent when compared to Serum/Glass. The S/C value difference between Serum/Glass compared to an alternative matrix/tube combination for each individual positive sample was calculated (plastic minus glass) and the difference data were analyzed. The mean and 98.75% confidence limits were calculated. As there were four comparisons (Serum/Glass compared to Serum/Plastic, EDTA/Plastic, Heparin/Plastic, or Citrate/Plastic), 98.75% confidence limits were calculated for maintaining the overall 5% type I error rate.

The maximum allowable S/C ratio shift for each specimen in the antibody negative sample population was calculated based on the negative samples in the VITROS Anti-HIV 1+2 assay Clinical Study data. The maximum allowable S/CO ratio shift for each specimen in antibody positive sample population was based on clinical trial precision estimates for a panel member near the cutoff and is limited to --------.

The following table summarizes the comparison data used to assess clinical equivalence of the four matrix/tube combinations with Serum/Glass for anti-HIV positive samples.

VITROS Anti-HIV 1+2 Assay Blood Collection Matrix/Tube Comparison Summary - Positive Samples

VITROS Anti-HIV 1+2 Assay Serum/Plastic
vs.
Serum/Glass
Heparin/Plastic
vs.
Serum/Glass
EDTA/Plastic
vs.
Serum/Glass
Citrate/Plastic
vs.
Serum/Glass
Mean of S/C Ratio Difference

0.4444

0.1375

-0.1486

-0.1584

Standard Error of S/C Ratio Difference

0.0224

0.0114

0.0134

0.0151

98.75% Confidence Interval 1 of the Mean Difference

(0.3870, 0.5018)

(0.1084, 0.1666)

(-0.1830, -0.1142)

(-0.1970, -0.1198)

Acceptance limit

±1.1022

±1.1022

±1.1022

±1.1022

Acceptance (Yes or No)

Yes

Yes

Yes

Yes

Variance of S/C Ratio Difference

0.0377

0.0097

0.0136

0.0170

Power for Equivalence Testing

100%

100%

100%

100%

1. With multiplicity adjustment.

The following table summarizes the comparison data used to assess clinical equivalence of the four matrix/tube combinations with Serum/Glass for anti-HIV negative samples.

VITROS Anti-HIV 1+2 Assay Blood Collection Matrix/Tube Comparison Summary - Negative Samples

VITROS Anti-HIV 1+2 Assay Serum/Plastic
vs.
Serum/Glass
Heparin/Plastic
vs.
Serum/Glass
EDTA/Plastic
vs.
Serum/Glass
Citrate/Plastic
vs.
Serum/Glass
Mean of S/C Ratio Difference

-0.0178

0.0054

0.0075

0.0051

Standard Error of S/C Ratio Difference

0.0005

0.0006

0.0007

0.0006

98.75% Confidence interval 1 of the Mean Difference

(-0.0191, -0.0164)

(0.0039, 0.0070)

(0.0058, 0.0092)

(0.0035, 0.0067)

Acceptance limit

±0.0542

±0.0542

±0.0542

±0.0542

Acceptance (Yes or No)

Yes

Yes

Yes

Yes

Variance of S/C Ratio Difference

0.000021

0.000027

0.000034

0.000030

Power for Equivalence Testing

100%

100%

100%

100%

1.With multiplicity adjustment.

The data support that specimens collected as serum in glass or plastic tubes as well as EDTA, heparin or citrate plasma are suitable for use in the VITROS Anti-HIV 1+2 assay.

Comparison of Stability of Serum/Plasma Samples

Twenty fresh blood samples (10 unspiked and 10 spiked with anti-HIV 1 or HIV 2 to give a target result of 2.0 ± 1.0 s/c), were collected and aliquoted into a variety of serum and plasma collection tubes. The following anticoagulants were evaluated in this study: heparin, EDTA, and citrate. Testing with the Vitros Anti- HIV 1+2 assay was conducted on the same day blood was drawn, and again after 5 and 7 days storage at 2-8°C and after a minimum of 28 days at -20°C with up to 3 freeze/thaw cycles.

The acceptance criteria for the negative samples is that they should not be misclassified on any of the occasions on which they were determined.

The acceptance criteria for the spiked samples is that no sample should be misclassified (i.e. gives a negative result rather than a positive result) on any of the occasions on which they are determined.

Comparison of Storage

None of the storage conditions tested had clinically significant effects on negative samples. All anti HIV 1 and anti-HIV 2 spiked samples were correctly classified as reactive.

These data show that storage of serum or plasma (heparin, EDTA, Citrate) samples for 5 and 7 days at 2-8°C and after a minimum of 28 days with up to 3 freeze / thaw cycles would not have a significant effect on the test results with the VITROS Anti-HIV 1+2 assay.

Potentially Cross Reacting Subgroups

A total of 236 patient samples from the following 14 potentially cross-reacting sub-groups were tested in the VITROS Anti-HIV 1+2 assay: HCV infection, HBV infection, HTLV I and II antibody positive, EBV infection, Influenza vaccine recipients, multiply transfused patients, multiparous females, dialysis patients, hemophilia patients, autoimmune disease patients, high rheumatoid factor, yeast (Candida) reactive, SOD reactive samples, and cord blood. An additional 15 samples from patients who had not received the Influenza vaccine were tested as a control for the Influenza vaccine recipients group. In all, 251 patient samples were tested for this study.

In these 251 samples tested, 3 were reactive in the VITROS Anti-HIV 1+2 assay. In the 14 clinical categories, 1 sample from the hemophilia group was found to give a reactive result in the VITROS Anti-HIV 1+2 assay yielding results just above the cutoff. This sample yielded negative results with an FDA approved method. The second reactive sample was from the pre-Influenza vaccine group. This sample was also reactive with an FDA licensed anti-HIV 1/2 assay. A third sample in the SOD reactive subgroup tested reactive on initial determination and negative upon repeat testing.

Summary of VITROS Anti-HIV 1+2 Assay Results with Potentially Cross-Reacting Specimens

Sample Category No. Samples Tested No. Negative No. Reactive No. Confirmed Positive
HCV Infection

16

16

0

0

HTLV 1+2 Positive

16

16

0

0

EBV Infection

15

15

0

0

Multiparous Females

16

16

0

0

Pre-Influenza Vaccine

15

14

1

0*

Post-Influenza Vaccine

15

15

0

0

Rheumatoid Factor

16

16

0

0

Autoimmune Disease

16

16

0

0

Multiply Transfused Patients

16

16

0

0

HBV Infection

16

16

0

0

Hemophilia

16

15

1

0

Dialysis

16

16

0

0

Yeast Reactive

20

20

0

0

SOD Reactive

22

21

1

0

Cord Blood (Neonates)

20

20

0

0

*Pre-Influenza specimen was tested and found reactive with an FDA licensed anti-HIV 1/2 assay. There was insufficient sample to perform a Western blot.

Potentially Cross-Reacting Sub-Groups - Microbiological Studies

The potential for bacterial contamination to affect the performance of the VITROS Anti-HIV 1+2 assay was evaluated further by testing samples spiked with Staphylococcus aureus, Escherichia coli and Pseudomonas aeruginosa. The samples were tested with and without a spike of anti HIV 1+2.

Of the samples that were tested none of the anti-HIV 1+2 unspiked (negative) samples were found to be false reactive and none of the anti-HIV 1+2 spiked samples were observed to be false negative in the VITROS Anti-HIV 1+2 assay.

Potentially Interfering Substances

The potentially interfering effects of hemoglobin, bilirubin and triolein were evaluated using samples from 30 patients. The results demonstrate that hemoglobin (up to 500 mg/dL), bilirubin (up to 20 mg/dL) and triolein (up to 3000 mg/dL) cause no misclassification of results. Samples spiked with anti-HIV-1 and anti-HIV-2 reactive plasma were tested near the cut-off (cut off s/c=1.00) and were observed to remain reactive at all levels tested with each potential interferent. Similarly, no interference was observed in samples not spiked with anti-HIV-1 and anti-HIV-2 reactive plasma, with results remaining below 1.00 s/c.

HIV-1 Mean Result at 0 Interferent Level Mean Result at Maximum Interferent Level
Test Substance Sample Maximum level Tested s/c Classification s/c Classification

Hemoglobin

HIV-1 Spiked sample

500 mg/dL

1.11

Reactive

1.16

Reactive

Negative sample

500 mg/dL

0.06

Negative

0.09

Negative

 

 

 

 

 

 

 

Bilirubin

HIV-1 Spiked sample

20 mg/dL

1.53

Reactive

1.51

Reactive

Negative sample

20 mg/dL

0.08

Negative

0.08

Negative

 

 

 

 

 

 

 

Triolein

HIV-1 Spiked sample

3000 mg/dL

1.34

Reactive

1.36

Reactive

Negative sample

3000 mg/dL

0.06

Negative

0.06

Negative

 

HIV-2 Mean Result at 0 Interferent Level Mean Result at Maximum Interferent Level
Test Substance Sample Maximum level Tested s/c Result s/c Result

Hemoglobin

HIV-2 Spiked sample

500 mg/dL

1.21

Reactive

1.29

Reactive

Negative sample

500 mg/dL

0.06

Negative

0.09

Negative

 

 

 

 

 

 

 

Bilirubin

HIV-2 Spiked sample

20 mg/dL

1.39

Reactive

1.40

Reactive

Negative sample

20 mg/dL

0.17

Negative

0.18

Negative

 

 

 

 

 

 

 

Triolein

HIV-2 Spiked sample

3000 mg/dL

1.55

Reactive

1.55

Reactive

Negative sample

3000 mg/dL

0.06

Negative

0.06

Negative

Additional Potential Interferents

A total of 60 samples were tested from the following subsets: patients with Cholesterol <200 mg/dL, patients with Cholesterol >300 mg/dL, patients with Total Protein between 6 and 8 g/dL, patients with Total Protein > 9 g/dL, normal patient samples (assumed normal IgG concentration) and patient samples spiked with Human IgG to achieve a concentration of >2620 mg/dL. Samples for Cholesterol and Total Protein testing were naturally occurring. Samples for IgG testing were normal patient samples that were spiked with a purified IgG preparation to achieve a level above the normal range.

All samples in each subset were spiked with anti-HIV-1 or anti-HIV-2 reactive plasma to evaluate performance with positive samples. All samples in each subset were also tested after being spiked with negative plasma to evaluate performance with negative samples.

All samples spiked with anti-HIV-1 or anti-HIV-2 reactive plasma yielded reactive results. Among samples spiked with negative plasma 47 of 60 yielded negative results. A total of 13 samples, all in the high total protein group, yielded reactive results. Twelve of 13 reactive samples also tested reactive with an FDA approved method.

Patient samples spiked with cholesterol up to 415 mg/dL or IgG up to 2620 mg/dL do not interfere with the clinical interpretation of results.

Summary of VITROS Anti-HIV 1+2 Assay Data from Potentially Interfering Sample Conditions

Sample Category No. Samples Tested No. Negative VITROS Anti-HIV 1+2 Assay Reactive No. Reactive in FDA Licensed assay
Cholesterol <200 mg/dL

10

10

0

0

Cholesterol >300 mg/dL*

10

10

0

0

Total Protein between 6 and 8 g/dL

10

10

0

0

Total Protein >9 g/dL

20

7

13

12

Serum spiked with Human IgG**

10

10

0

0

*Maximum cholesterol level in the tested sample was 415 mg/dL.
**Maximum IgG concentration was 2620 mg/dL.

A total of 22 additional samples were tested from patients with Total Protein >9 g/dL. Samples were tested to determine if the presence of high total protein would cause negative samples to yield reactive assay results. A total of 21 of the 22 samples yielded negative results with the VITROS Anti-HIV 1+2 assay. A single sample yielded reactive results with the VITROS Anti-HIV 1+2 assay. This sample was determined to be reactive in an FDA approved assay and confirmed HIV-1 antibody positive by Western blot.

Summary of VITROS anti-HIV 1+2 Data from Additional Study of Samples with High Protein

Sample Category No. Samples Tested No. Negative VITROS
Anti-HIV 1+2 Assay Reactive
No. Reactive in FDA Licensed assay No. Confirmed HIV antibody positive in Western blot

Total Protein >9 g/dL

22

21

1

1

1

Patient samples containing protein >9 g/dL do not consistently interfere with the clinical interpretation of results.

Genotype Detection

Genotype detection was assessed using the Boston Biomedica, Inc. Worldwide HIV Performance Panel. This panel consists of 25 naturally occurring plasma specimens originating from diverse geographic locations. Twenty three of these specimens have been characterized to be anti-HIV reactive, while two are anti-HIV nonreactive. The reactive specimens represent HIV Group M (subtypes A, B, C, D, E, F, and G) Group O, and HIV-2 genotypes. All 23 of the anti-HIV reactive panel members were also reactive in the VITROS Anti-HIV 1+2 assay, while the two anti-HIV nonreactive panel members were negative in the VITROS Anti-HIV 1+2 assay. Two lots of VITROS Anti-HIV 1+2 Reagent Packs and Calibrators were included in this study.

BBI Worldwide HIV Performance Panel Test Results with VITROS Anti-HIV 1+2 Assay

Panel ID Number

VITROS Results

 

Genotype

Result (s/c)

Classification

WWRB302(M)-01

O

9.45

Reactive

WWRB302(M)-02

A

55.3

Reactive

WWRB302(M)-03

G

85.3

Reactive

WWRB302(M)-04

G

56.4

Reactive

WWRB302(M)-05

A

54.6

Reactive

WWRB302(M)-06

G

65.0

Reactive

WWRB302(M)-08

G

60.0

Reactive

WWRB302(M)-09

A

82.6

Reactive

WWRB302(M)-10

NEG

0.14

Negative

WWRB302(M)-11

HIV-2*

26.3

Reactive

WWRB302(M)-12

C

89.4

Reactive

WWRB302(M)-14

D

61.2

Reactive

WWRB302(M)-15

D

59.2

Reactive

WWRB302(M)-16

D

64.1

Reactive

WWRB302(M)-17

D

73.0

Reactive

WWRB302(M)-19

C

55.2

Reactive

WWRB302(M)-21

B'

81.4

Reactive

WWRB302(M)-22

E

72.7

Reactive

WWRB302(M)-24

E

78.4

Reactive

WWRB302(M)-25

HIV-2*

40.6

Reactive

WWRB302(M)-26

B

74.1

Reactive

WWRB302(M)-27

B/D

54.1

Reactive

WWRB302(M)-28

F

79.6

Reactive

WWRB302(M)-29

B

86.7

Reactive

WWRB302(M)-30

NEG

0.10

Negative

* The HIV-2 status of these specimens was determined by serological testing.

Detection of HIV-1 Group O Specimens

Thirteen confirmed HIV 1 Group O antibody positive samples were tested with the VITROS Anti-HIV 1+2 assay. All 13 samples gave a reactive result. All samples were confirmed as HIV O reactive using Western blots developed at the Institut Alfred Fournier. The results are shown in the following table.

HIV-1 Group O Sample ID

VITROS Anti-HIV 1+2 Assay Result (s/c)

R407/937

25.4

R407/939

23.3

R407/941

5.85

R407/942

27.1

R407/943

10.1

R407/946

2.73

R407/947

4.08

R407/948

11.1

R407/949

26.1

R407/950

5.18

R407/951

8.69

R407/952

19.0

R407/953

36.6

One additional sample from the Worldwide HIV Performance Panel obtained from Boston Biomedica, Inc. (sample WWRB302(M)-01) was tested with the VITROS Anti-HIV 1+2 assay, and gave a reactive s/c result of 9.45. Thus, all 14 HIV-1 Group O samples gave reactive results with the VITROS Anti-HIV 1+2 assay.

Dilutional Sensitivity

An assessment of assay dilutional sensitivity was performed by testing known anti-HIV-1 and anti-HIV-2 reactive samples that had undergone serial 2-fold dilutions with a negative matrix. Results were compared between the VITROS Anti-HIV 1+2 assay and an FDA licensed anti-HIV 1/2 assay. Two lots of VITROS Anti-HIV 1+2 Reagent Pack and Calibrator were included in this study.

For HIV-1 specimens, the VITROS Anti-HIV 1+2 assay dilutional sensitivity ranged from --------- to ---------- and the Reference assay ranged from ---------- to -----------. For HIV-2 specimens, the VITROS Anti-HIV 1+2 assay dilutional sensitivity ranged from ---------- to ---------- and the Reference assay ranged from ------- to -------.

Linearity of the diluted samples produced high correlation coefficients (R2 >0.98) through the region of the VITROS Anti-HIV 1+2 assay cutoff.

Stability

VITROS Anti-HIV 1+2 Reagent Packs, Calibrator and Controls that were subjected to a period of simulated transport to mimic effects of shipment were tested at various time points up to 52 weeks after storage at 2-8°C (36-46°F). All results obtained were within acceptability limits, and overall no trends were evident.

In addition, a commercially obtained performance panel was tested using transported, stored materials at week 0, week 26 and week 52. Materials stored for 52 weeks yielded results that indicated no change in the classification of the samples (antibody positive or antibody negative) from the classifications obtained at the initial time point.

This data supports the storage of the VITROS Anti-HIV 1+2 Reagent Pack and Calibrator for 52 weeks at 2-8°C (36-46°F).

Open On-Board Storage for the VITROS Anti-HIV 1+2 Reagent Pack

VITROS Anti-HIV 1+2 Reagent Packs that were subjected to a period of simulated transport to mimic effects of shipment were opened and placed in an environmental chamber for a period of 8 weeks to simulate the storage on board the VITROS Analyzer. These Reagent Packs were tested at various time points within the 8-week time period. In addition, a single transported, opened Reagent Pack from each Kit Lot was removed from the chamber on 6 different occasions, and brought to room temperature over the 8 week period to simulate typical customer usage. Results of testing were within acceptability limits and overall no trend was observed between Reagent Packs stored at 2-8°C (36-46°F) and freshly opened, and Reagent Packs stored opened on board for 8 weeks. This data support the on board storage of Reagent Packs for up to 8 weeks.

Open Off-Board Storage for VITROS Anti-HIV 1+2 Calibrators

VITROS Anti-HIV 1+2 Calibrators that were subjected to a period of simulated transport to mimic effects of shipment were opened, pooled, sub-aliquoted and stored at 2-8°C (36-46°F) and -20°C (-4°F) for 13 weeks. Results of testing these Calibrators at various time points up to 13 weeks indicated no observable trends and met all acceptance criteria.

The data support the unopened storage of the calibrators at 2 8 °C (36 46 °F). After opening store for up to 13 weeks at 2 8 °C (36 46 °F) or 13 weeks at 20 °C (-4 °F) (with no more than 1 freeze-thaw cycle).

VITROS Universal Wash Reagent Study

VITROS Anti-HIV 1+2 Reagent Packs, Calibrators and Controls that were subjected to a period of simulated transport to mimic the effects of shipment were tested with 3 lots of VITROS Universal Wash Reagent at weeks 0 and 26 to determine the effect of aged VITROS Universal Wash Reagent.

The data presented in this report show that the performance of the VITROS Anti-HIV 1+2 assay is acceptable when used with VITROS Universal Wash Reagent that is either fresh or 26 weeks old.

VITROS Signal Reagent Study

VITROS Anti-HIV 1+2 Reagent Packs, Calibrators and Controls that were subjected to a period of simulated transport to mimic the effects of shipment were tested with 3 lots of VITROS Signal Reagent to determine the effect of aged VITROS Signal Reagent.

The data presented in this report show that the performance of the VITROS Anti-HIV 1+2 assay is acceptable when used with VITROS Signal Reagent that is either fresh or 6 months old.

Temperature Stressing Study (-20oC / -4°F)

VITROS Anti-HIV 1+2 assay Reagent Packs and Calibrators were subjected to 2 freeze/thaw cycles and the performance compared with Reagent Packs and Calibrators stored at 2-8°C (36-46°F). Although freezing and thawing the VITROS Anti-HIV 1+2 Reagent Pack had no adverse effect on calibration quality parameters or control results, as a precaution freezing VITROS Anti-HIV 1+2 Reagent Pack is not recommended.

Temperature Stressing Study (30oC and 37oC / 86oF and 99oF)

VITROS Anti-HIV 1+2 Reagent Packs and Calibrators were subjected to 5 days at 30°C (86oF) or 1 day at 37°C (99oF) and the performance compared with Reagent Packs and Calibrators stored at 2-8°C (36-46°F).

Exposing VITROS Anti-HIV 1+2 Reagent Packs and Calibrators to a temperature of 30°C (86oF) for 5 days or 37°C (99oF) for 1 day had no adverse effect on calibration quality parameters or control results. Inadvertent exposure of the Reagent Packs or Calibrator up to these temperatures for the times stated would not significantly compromise the performance of the VITROS Anti-HIV 1+2 assay.

Microbiology

VITROS Anti-HIV 1+2 reagents are formulated with anti-microbial agents (Kathon) that provide protection against adventitious contamination by microorganisms. Evaluation of the microbial load of each reagent (Assay Reagent, Conjugate Reagent and Calibrator) post-dispensing and at 26 and 52 weeks demonstrated that the total aerobic count is generally on the order of ≤10CFU/mL. In addition, the levels of preservative in each reagent were determined over a period of 52 weeks. Results for the Assay Reagent, Conjugate Reagent and Calibrator demonstrated that the preservative concentrations were above the minimum inhibitory concentration throughout 52 weeks of testing.

A study conducted according to US Pharmacopoeia (USP) 23/NF 18, general chapter 51, assessed the ability of the reagents to withstand or control microbial contamination. Results indicate that the preservative systems for Assay Reagent, Conjugate and Calibrator met the requirements of the USP 23 at 0, 26 and 52 weeks.

Precision

Precision was evaluated on a different VITROS ECi/ECiQ Immunodiagnostic System at three external sites, using one Reagent Pack and Calibrator kit lot. At least two replicates each of a four member panel were assayed on a single occasion per day on 20 different days. The data shown in the table were rounded following all calculations.

Clinical
Site
Mean VITROS
Anti-HIV 1+2 Assay
Results (s/c)
Repeatability* Between Day** Total*** No.
of
Obs.
No.
of
Days
S.D. C.V.(%) S.D. C.V.(%) S.D. C.V.(%)

Site 1

0.07

Negative

0.007

9.9

0.005

6.9

0.008

12.1

40

20

6.06

HIV-1

0.071

1.2

0.172

2.8

0.186

3.1

40

20

4.12

HIV-2

0.046

1.1

0.163

3.9

0.169

4.1

40

20

1.32

HIV-1

0.023

1.8

0.046

3.5

0.052

4.0

40

20

Site 2

0.08

Negative

0.006

7.6

0.003

4.3

0.007

8.7

40

20

6.66

HIV-1

0.169

2.5

0.163

2.5

0.235

3.5

40

20

4.39

HIV-2

0.060

1.4

0.100

2.3

0.117

2.7

40

20

1.39

HIV-1

0.045

3.2

0.035

2.5

0.057

4.1

40

20

Site 3

0.07

Negative

0.003

4.4

0.004

5.6

0.005

7.1

40

20

6.22

HIV-1

0.085

1.4

0.131

2.1

0.156

2.5

40

20

4.42

HIV-2

0.033

0.7

0.158

3.6

0.162

3.7

40

20

1.34

HIV-1

0.028

2.1

0.034

2.5

0.044

3.3

40

20

* Repeatability: Variability of the assay performance from replicate to replicate.
** Between Day: Variability of the assay performance from day to day.
*** Total: Variability of the assay combining the effects of repeatability and between day.

Precision was further evaluated incorporating between site and between lot variation. The study was performed at three external sites using three reagent lots. At least three replicates each of a four member panel were assayed on a single occasion per day on six different days. The between site, between lot, and total precision estimates CV (%) were derived from a variance component analysis. The data shown in the table were rounded following all calculations.

Mean VITROS
Anti-HIV1+2 Assay
Results (s/c)
Between Site * Between Lot ** Total *** No.
of
Obs.
S.D. C.V. (%) S.D. C.V. (%) S.D. C.V. (%)

0.10

Negative

0.004

3.9

0.031

30.0

0.034

32.0

162

1.09

HIV-2

0.000

0.0

0.152

14.0

0.164

15.0

162

1.34

HIV-1

0.000

0.0

0.017

1.3

0.070

5.2

162

3.76

HIV-1

0.049

1.3

0.210

5.6

0.275

7.3

162

* Between Site: Variability of the assay performance from site to site.
** Between Lot: Variability of the assay performance from lot to lot calculated using data across all sites.
*** Total: Variability of the assay incorporating factors of site, lot and day.

Calibration Interval

The performance of the VITROS Anti-HIV 1+2 assay within and beyond one calibration interval (28 days) was evaluated at three sites by testing a three-member panel with one kit lot. One panel member was close to the VITROS Anti-HIV 1+2 assay cut-off. Additional testing was performed on Days 29 and 30 of the calibration cycle to show that the Analyzer would still yield valid results beyond the end of a 28-day cycle. Two replicates of each panel member were run per day at each clinical site. Appropriate calibration was performed and verified on Day 0 of the study, and the testing was performed for a total of 20 study days over a 28-day period.

Least squares regression analyses were performed within site and across sites. For analyses within site, although the slopes were statistically significant for two panel members in at least one site, the changes in S/C ratios over the entire testing period were either not clinically relevant or were too small to have any clinical implications. For analyses across sites, the mean slope was not statistically significant for any of the panel members.

In conclusion, the VITROS Anti-HIV 1+2 assay demonstrated adequate performance throughout the entire calibration interval, and continued to perform successfully two days beyond the expiration of calibration, as per the study design.

Seroconversion Panels

Twenty commercially available seroconversion panels were tested. Results for the twenty panels are summarized in the following table. The table presents the days elapsed from the date of the initial bleed to the last negative sample and first reactive sample. Data are presented for both assays for each of the seroconversion panels.

Days to Evidence of HIV Infection

Panel ID Licensed
Anti-HIV 1/2 Assay
VITROS
Anti-HIV 1+2 Assay
Difference in Days to
Anti-HIV Reactive Result

-*

+** -*** +**** Licensed Assay minus VITROS Anti-HIV 1+2 Assay

PRB904

49

92

49

92

0

PRB910

14

26

14

26

0

PRB916

15

30

15

30

0

PRB923

37

47

37

47

0

PRB924

26

33

26

33

0

PRB925

22

44

22

44

0

PRB926

9

27

9

27

0

PRB927

0

28

28

33

-5

PRB929

21

25

18

21

4

PRB931

15

28

15

28

0

PRB933

0

21

0

21

0

PRB934

0

7

0

0

7

PRB935

28

43

28

43

0

PRB940

7

11

0

7

4

PRB941

9

18

9

18

0

PRB944

9

14

9

14

0

PRB945

7

13

7

13

0

PRB947

0

9

0

9

0

PRB952

14

17

10

14

3

PRB959

7

9

0

7

2

* Post bleed day of last non-reactive result, usually denotes previous bleed from first repeatedly reactive result.
** Post bleed day of first repeatedly reactive result.
*** Post bleed day of last non-reactive result, usually denotes previous bleed from first reactive result. **** Post bleed day of first reactive result.

The VITROS anti-HIV 1+2 and reference anti-HIV 1/2 assays were in agreement for 14 of the 20 panels. The VITROS ANTI-HIV 1+2 ASSAY became reactive one bleed (from two to seven days) earlier for five of the twenty panels. The reference assay became repeatedly reactive one bleed (five days) earlier for the final panel.

2.8 Summary of Clinical Studies

A multi center study was conducted to establish the performance characteristics of the VITROS Anti-HIV 1+2 assay using samples obtained in the U.S. and internationally from individuals at low or high risk for HIV infection, or known to be HIV antibody positive. Statistical testing was performed to ensure that the distribution of VITROS Anti-HIV 1+2 s/c values was homogeneous across the three testing sites participating in the study. For the clinical studies, all initially reactive specimens were retested in duplicate. Repeat testing did not alter the performance characteristics of the assay, therefore, repeat testing of initially reactive specimens is not recommended.

The specificity of the VITROS Anti-HIV 1+2 assay was evaluated among individuals at low risk for HIV infection. The sensitivity of the VITROS Anti-HIV 1+2 assay was evaluated among individuals known to be HIV antibody positive, and by testing serially collected samples from individuals with HIV infection (seroconversion panels). Assay performance was further evaluated among individuals with signs or symptoms of HIV infection and among individuals belonging to groups recognized to be at risk for HIV infection due to lifestyle, behavior, occupation or known exposure event.

2.8.1 Results by Specimen Classification

Samples from subjects at high or low risk for HIV infection were tested with an FDA-licensed anti-HIV 1/2 assay, and with the VITROS Anti-HIV 1+2 assay at the three testing sites. The HIV antibody status (HIV Antibody Positive, HIV Antibody Negative or HIV antibody status Not Determined) of the individual subject was defined according to the following licensed and supplemental assay testing algorithm. In those instances where the licensed assay was negative but the VITROS Anti-HIV 1+2 assay was reactive, supplemental testing was performed to determine the HIV antibody status of the sample.

Licensed Anti-HIV 1/2 Assay Result Supplemental Testing Result(s) HIV Antibody Status

Negative

Not Applicable

HIV Antibody Negative

Reactive

Western blot (WB) Negative

HIV Antibody Negative*

Reactive

Western blot Positive

HIV Antibody Positive

Reactive

Western blot Indeterminate
Indirect Immunofluorescence assay (IFA) Negative

HIV Antibody Negative*

Reactive

Western blot Indeterminate
IFA Positive

HIV Antibody Positive

Reactive

Western blot Indeterminate
IFA Indeterminate

HIV Antibody Status
Not Determined **

* Samples from high risk subjects whose anti-HIV 1/2 assay results were discordant were tested with an HIV-2 EIA/IFA. The HIV antibody status remained "Negative" if the HIV-2 EIA was negative. If the HIV-2 EIA was repeatedly reactive and the HIV-2 IFA was negative or indeterminate, the HIV status was "Not Determined". The HIV status was "Positive" if the HIV-2 IFA was positive.
** These samples were tested with an HIV-2 EIA/IFA. HIV antibody status remained "Not Determined" if the HIV-2 EIA was negative, or if the HIV-2 EIA was repeatedly reactive but the HIV-2 IFA was negative or indeterminate. The HIV antibody status was "Positive" if the HIV-2 IFA was positive.

The specificity/negative percent agreement of the VITROS Anti-HIV 1+2 assay was calculated as the percentage of the combined HIV Antibody Negative and status Not Determined subjects that tested negative with the VITROS assay. The sensitivity/positive percent agreement of the VITROS Anti-HIV 1+2 assay was calculated as the percentage of HIV Antibody Positive subjects that tested reactive with the assay.

2.8.2 Specificity in Individuals at Low Risk for HIV Infection

Samples from 1444 subjects at low risk for HIV infection were tested with the VITROS anti-HIV 1+2 and licensed anti-HIV 1/2 assays (with supplemental testing as required). These samples were obtained from pregnant women in the U.S. (N=297), pregnant women in the U.S. in the period around labor and delivery (N=49), from insurance applicants in the U.S. for whom HIV testing was required (N=999), and from pediatric subjects ages 2-17 years (N=99).

Demographics of Seroreactivity in the Low Risk Pediatric Population

There were 99 unlinked samples from low risk pediatric subjects tested with the VITROS Anti-HIV 1+2 assay. The group was 49.5% male and 50.5% female and ranged in age from 2 to 17 years. None of the 99 samples was reactive with the VITROS Anti-HIV 1+2 assay. The distribution of VITROS Anti-HIV 1+2 assay negative results among the low risk pediatric subjects by age and gender is presented in the following table:

Demographics of Seroreactivity for the VITROS Anti-HIV 1+2 Assay in Low Risk Pediatric Subjects (N=99)

Age Range Gender VITROS Anti-HIV 1+2 Assay Results Total
Reactive Negative
N Percent N Percent

2–4

Female

0

0.0

13

100.0

13

Male

0

0.0

12

100.0

12

5–9

Female

0

0.0

11

100.0

11

Male

0

0.0

14

100.0

14

10–14

Female

0

0.0

10

100.0

10

Male

0

0.0

14

100.0

14

15–17

Female

0

0.0

16

100.0

16

Male

0

0.0

9

100.0

9

Total

0

0.0

99

100.0

99

The results of the low risk populations are summarized in the following table:

VITROS Anti-HIV 1+2 and Licensed Anti-HIV 1/2 Assay Results in Low Risk Populations (N=1444)

Population Description Number
Tested
Licensed
Anti-HIV 1/2 Assay
VITROS
Anti-HIV 1+2 Assay
Total WB
Positive
NR IR RR
(WB+)
NR Reactive (WB+)
Pregnancy
(Low Risk–U.S.)

297

295

2

2 (0)

294

3 (1)

1

Labor & Delivery
(Low Risk–U.S.)

49

49

0

0 (0)

48

1 (0)

0

Insurance Applicants
(Low Risk–U.S.)

999

993

6

6 (2)

991

8 (5)

5

Pediatric
(Low Risk–U.S.)

99

99

0

0 (0)

99

0 (0)

0

Total

1444

1436

8

8 (2)

1432

12 (6)

6

NR = non reactive (negative); IR = initially reactive; RR = repeatedly reactive; WB = licensed HIV-1 Western blot

Six of the 1444 low risk samples were reactive with the VITROS Anti-HIV 1+2 assay and positive on HIV-1 Western blot positive (6/6; 100% detection). Only two Western blot positive samples (2/6; 33.3% detection) were repeatedly reactive with the licensed assay.

The performance of the VITROS Anti-HIV 1+2 assay compared with HIV antibody status in low risk populations is summarized in the following table:

Agreement of the VITROS Anti-HIV 1+2 Assay with HIV Antibody Status in Low Risk Populations (N=1444)

VITROS Anti-HIV 1+2 Assay Results HIV Antibody Status Total
Positive Negative Not Determined*

Reactive

6

5**

1

12

Negative

0

1432

0

1432

Total

6

1437

1

1444

*This sample remained HIV status "Not Determined" following VITROS, licensed and supplemental testing for anti-HIV-1 and anti-HIV-2 (EIA, WB, IFA). This sample was considered anti-HIV negative when calculating specificity.
**Three samples were initially reactive with the VITROS assay during the clinical study and were retested in duplicate: both duplicate results for the three samples were negative. All three were initially non reactive with the reference assay. HIV-1 Western blot results are not available for the three samples

The specificity of the VITROS Anti-HIV 1+2 assay was calculated as the percentage of the combined HIV antibody Negative and status "Not Determined" subjects that tested negative with the VITROS assay.

The specificity of the VITROS Anti-HIV 1+2 assay in the low risk populations was 99.58% (1432/1438 in this study with a 95% exact confidence interval (CI) of 99.09% to 99.85%) compared with 99.58% (1432/1438) for the licensed anti-HIV 1/2 assay. The six low risk samples that were HIV-1 Western blot positive (HIV antibody positive) were excluded from the specificity calculation.

2.8.3 Sensitivity in Individuals Positive for Antibodies to HIV-1 and HIV-2

Sensitivity in Individuals Positive for Antibodies to HIV-1

Samples from 1121 HIV-1 infected adults were tested with the VITROS Anti-HIV 1+2 assay. These subjects were enrolled in Florida (72.8%; N=816), California (16.3%; N=183), New Jersey (8.9%; N=100) and Texas (2.0%; N=22). Clinical and laboratory documentation of HIV-1 infection and HIV antibody positive status were obtained from medical records for each of the 1121 individuals. In addition, 40 unlinked residual samples from HIV-1 antibody positive pediatric subjects, ages 1-16 years of age, were also tested with the VITROS Anti-HIV 1+2 assay.

CD4+ counts were available for 1094 of the 1121 HIV-1 infected adults. As shown in the following table, the VITROS Anti-HIV 1+2 assay was reactive with all 1121 samples regardless of the subjects' CD4+ counts.

VITROS Anti-HIV 1+2 Assay Results Among U.S. HIV Infected Adults with Documented CD4+ Counts (N=1121)

CD4+ Count N VITROS Anti-HIV 1+2 Assay
Reactive

< 200

149

149

200 - 499

429

429

> 499

516

516

Unknown

27

27

The VITROS Anti-HIV 1+2 assay results in the HIV-1 antibody positive population are summarized in the following table:

VITROS Anti-HIV 1+2 Assay Results in Known HIV Antibody Positive Subjects (N=1161)

Population Description VITROS Anti-HIV 1+2 Assay
N NR Reactive
Adult HIV-1 Positive
( U.S.)

1121

0

1121

Pediatric HIV-1 Positive ( U.S.)

40

0

40

Total

1161

0

1161

All 1161 samples were reactive with the VITROS Anti-HIV 1+2 assay. The sensitivity of the VITROS Anti-HIV 1+2 assay for U.S. subjects known to be positive for HIV-1 antibody was 100% (1161/1161; 95% CI = 99.68% to 100%) in this study.

Sensitivity in Individuals Positive for Antibodies to HIV-2

Sensitivity of the VITROS Anti-HIV 1+2 assay was also determined among 208 mono-infected HIV-2 antibody positive individuals from the Ivory Coast. Testing results are summarized in the following table:

VITROS Anti-HIV 1+2 Assay Results in Mono-Infected HIV-2 Antibody Positive Subjects (N=208)

Population Description VITROS Anti-HIV 1+2 Assay
N NR Reactive
HIV-2 Positive ( Ivory Coast)

208

0

208

All 208 anti-HIV-2 positive samples were reactive with the VITROS Anti-HIV 1+2 assay. The sensitivity of the VITROS Anti-HIV 1+2 assay in mono-infected HIV-2 antibody positive individuals was 100% (208/208; 95% CI = 98.24% to 100.0%) in this study.

Sensitivity in International Populations Known to be HIV Antibody Positive

Sensitivity of the VITROS Anti-HIV 1+2 assay was determined among 194 HIV antibody positive subjects from four geographic locations outside the U.S. Samples were obtained from archives in Africa (9.8%; N=19), Asia (53.1%; N=103), Europe (34.0%; N=66) and Latin America (3.1%; N=6). The VITROS Anti-HIV 1+2 assay results are presented in the following table:

VITROS Anti-HIV 1+2 Assay Results in International Populations Known to be HIV Antibody Positive (N=194)

Population Description VITROS Anti-HIV 1+2 Assay
N NR Reactive
HIV Positive ( Africa)

19

1*

18

HIV Positive ( Asia)

103

0

103

HIV Positive ( Europe)

66

0

66

HIV Positive ( Latin America)

6

0

6

Total

194

1*

193

*This archived sample was obtained from a rural clinic in Keffi, Nigeria. The VITROS Anti-HIV 1+2 assay gave an initial s/c result of 0.53, and repeat results of 0.55 and 0.56. Singleton testing with the licensed assay gave an s/c result of 1.93. An HIV-1 Western blot showed bands of 1+ intensity at the p17, p24 and gp160 positions. The sample is unlinked to the donor's identity. No clinical information or follow-up sample is available.

One sample was non-reactive and 193 samples were reactive with the VITROS Anti-HIV 1+2 assay.

2.8.4 Reactivity in Populations at High Risk for HIV-1 Infection and HIV-2 Infection

Demographics of Seroreactivity Study in High Risk Populations

Among the 2175 high risk subjects from the U.S. participating in the VITROS Anti-HIV 1+2 assay clinical study, 1975 (90.8%) reported no current signs or symptoms of HIV infection. Of these 1975 asymptomatic individuals, 14.7% (291) were enrolled in Florida, 11.2% (222) were enrolled in New Jersey, 1.1% (21) were enrolled in Texas and 73.0% (1441) were enrolled in California. The group was Caucasian (28.2%), African American (18.1%) Hispanic (49.6%), and Asian (1.1%) with the remaining 3.0% represented by other ethnic groups. The group was 50.3% male and 49.7% female and ranged in age from 14 to 82 years. All were at risk for HIV infection due to lifestyle, behavior, occupation or known exposure event, or belonged to groups at risk for HIV infection. The VITROS Anti-HIV 1+2 assay was reactive in 2.2% (43/1975) of the individuals in this group. The percent VITROS Anti-HIV reactive results observed in the asymptomatic population in each collection area was 3.8% (11/291) in Florida, 0.0% (0/21) in Texas, 3.6% (8/222) in New Jersey, and 1.7% (24/1441) in California. The distribution of VITROS Anti-HIV 1+2 assay reactive and negative results among the high risk subjects without signs or symptoms of HIV infection by age and gender is presented in the following table.

Seroreactivity for the VITROS Anti-HIV 1+2 Assay in High Risk Subjects Without Signs or Symptoms of HIV Infection (N=1975)

Age Range Gender VITROS Anti-HIV 1+2 Assay Results Total
Reactive Negative
N Percent N Percent
14–19

Female

0

0.0

28

100.0

28

 

Male

0

0.0

25

100.0

25

20–29

Female

1

0.4

233

99.6

234

 

Male

5

3.0

162

97.0

167

30–39

Female

1

0.4

284

99.6

285

 

Male

5

2.0

249

98.0

254

40–49

Female

7

2.4

289

97.6

296

 

Male

11

3.5

307

96.5

318

50–59

Female

2

1.7

116

98.3

118

 

Male

11

5.6

184

94.4

195

60–69

Female

0

0.0

14

100.0

14

 

Male

0

0.0

36

100.0

36

70–79

Female

0

0.0

1

100.0

1

 

Male

0

0.0

3

100.0

3

80–82

Female

0

0.0

1

100.0

1

 

Male

0

0.0

0

0.0

0

Total

43

2.2

1932

97.8

1975

Reactivity in Populations at High Risk for HIV-1 Infection

The performance of the VITROS Anti-HIV 1+2 assay was evaluated among 2175 individuals at high risk for HIV-1 infection prospectively enrolled in California (69.8%; N=1517), Florida (14.6; N=317), New Jersey (14.6%; N=317) and Texas (1.1%; N=24), and in 249 pregnant women at high risk for HIV infection enrolled in California (49.0%; N=122) and Florida (51.0%; N=127). Testing results are presented in the following table:

VITROS and Licensed Anti-HIV 1/2 Assay Results in U.S. High Risk Populations (N=2424)

Population Description Number Tested Licensed
Anti-HIV 1/2 Assay
VITROS
Anti-HIV 1+2 Assay
Total WB
Positive
NR IR RR (WB+) NR Reactive (WB+)

High Risk ( U.S.)

2175

2106

69

68 (53)

2116

59 (54)

54

Pregnancy
(High Risk - U.S.)

249

244

5

4 (4)

243

6 (5)

5

Total

2424

2350

74

72 (57)

2359

65 (59)

59

The VITROS Anti-HIV 1+2 assay was reactive in 65 samples; 59 of those were Western blot positive. The licensed assay was repeatedly reactive in 57 (57/59; 99.61% detection). All of the 57 Western blot positives that were repeatedly reactive in the licensed test were reactive in the VITROS Anti-HIV 1+2 assay.

The performance of the VITROS Anti-HIV 1+2 assay compared with HIV antibody status in the U.S. high risk populations is summarized in the following table:

Agreement of the VITROS Anti-HIV 1+2 Assay with HIV Antibody Status in U.S. High Risk Populations (N=2424)

VITROS Anti-HIV 1+2 Assay Results HIV Antibody Status Total
Positive Negative Not Determined*  

Reactive

59

5**

1

65

Negative

0

2359

0

2359

Total

59

2364

1

2424

*This sample remained HIV status "Not Determined" following VITROS, licensed and supplemental assay testing for anti-HIV-1 and anti-HIV-2 (EIA, WB, IFA). This sample was considered anti-HIV negative when calculating percent agreement.
** One sample was initially reactive with the VITROS assay during the clinical study and was retested in duplicate: both duplicate results were negative. The sample was initially non reactive with the reference assay. HIV-1 Western blot results are not available for the sample.

The positive percent agreement of the VITROS Anti-HIV 1+2 assay with HIV antibody status in U.S. high risk populations was 100% (59/59; 95% 95% CI = 93.94% to 100%) compared with 96.61% (57/59) for the licensed assay. The negative percent agreement was 99.75% (2359/2365; 95% CI = 99.45% to 99.91%) compared with 99.37% (2350/2365) for the licensed assay.

Reactivity in Populations at High Risk for HIV-2 Infection

The performance of the VITROS Anti-HIV 1+2 assay was evaluated among individuals at high risk for HIV-2 infection due to residence in an HIV-2 endemic area. The 488 subjects in this group were prospectively enrolled in the Ivory Coast. Testing results are presented in the following table:

VITROS and Licensed Anti-HIV 1/2 Assay Results in Individuals at High Risk for HIV-2 Infection (N=488)

Population Description Number Tested Licensed
Anti-HIV 1/2 Assay
VITROS
Anti-HIV 1&+2 Assay
Total WB
Positive
NR IR RR
(WB+)
NR Reactive (WB+)

High Risk
( Ivory Coast)

488

453

35

33 (26)

457

31 (26)

26

The VITROS Anti-HIV 1+2 assay was reactive in 31 samples; 26 of these were HIV-1 Western blot positive. The licensed anti-HIV 1/2 assay was repeatedly reactive in 33 samples; 26 of these were HIV-1 Western blot positive. All of the 26 HIV-1 positive s that were repeatedly reactive using the licensed test were reactive using the VITROS Anti-HIV 1+2 assay. The HIV-1 Western blot positive samples did not undergo further supplemental testing. Samples with discordant anti-HIV 1/2 EIA results that were negative or indeterminate by supplemental testing for antibodies to HIV-1 were tested for antibodies to HIV-2 (EIA and IFA). None were positive for anti-HIV-2.

The performance of the VITROS Anti-HIV 1+2 assay compared with HIV antibody status in the HIV-2 high risk population is summarized in the following table:

Agreement of the VITROS Anti-HIV 1+2 Assay with HIV Antibody Status in the HIV-2 High Risk Population (N=488)

VITROS Anti-HIV 1+2 Assay Results HIV Antibody Status Total
Positive Negative Not Determined*

Reactive

26

1

4

31

Negative

0

452

5

457

Total

26

453

9

488

*These samples remained HIV status "Not Determined" following VITROS, licensed and supplemental assay testing for anti-HIV-1 and anti-HIV-2 (EIA, WB, IFA). These samples were considered anti-HIV negative when calculating percent agreement.

The positive percent agreement of the VITROS Anti-HIV 1+2 assay with HIV antibody status in the HIV-2 high risk population was 100% (26/26; 95% CI = 86.77% to 100%) compared with 100% (26/26) for the licensed assay. The negative percent agreement was 98.92% (457/462; CI = 97.49% to 99.65%) compared with 98.48% (455/462) for the licensed assay.

2.8.5 Positive and Negative Percent Agreement of the VITROS Anti-HIV 1+2 Assay with HIV Antibody Status by Study Population

The positive and/or negative percent agreement of the VITROS Anti-HIV 1+2 assay with HIV antibody status in the high risk, low risk and HIV antibody positive populations tested in this clinical study are summarized in the following table:

Positive and Negative Percent Agreement of the VITROS Anti-HIV 1+2 Assay with HIV Antibody Status by Study Population (N=5919)

Population Positive Percent Agreement 95% Exact
Confidence
Intervals
Negative Percent Agreement 95% Exact
Confidence
Intervals

High Risk
( U.S.)*

100%
(59/59)

93.94%–100%

99.75%
(2359/2365)

99.45%–99.91%

High Risk
( Ivory Coast)

100%
(26/26)

86.77%–100%

98.92%
(457/462)

97.49%–99.65%

HIV Positive
( U.S.)**

100%
(1161/1161)

99.68%–100%

 

 

HIV Positive
(International)

99.48%
(193/194)

97.16%–99.99%

 

 

HIV-2 Positive
(Ivory Coast)

100%
(208/208)

98.24%–100%

 

 

Low Risk
( U.S.)***

100%
(6/6)

N/A****

99.58%
(1432/1438)

99.09%–99.85%

* Adult (N=2175); pregnant (N=249).
** Adult (N=1121); pediatric (N=40).
*** Pregnant (N=297); labor and delivery (N=49); insurance applicants (N=999); pediatric (N=99).
**** N/A=Not applicable. Confidence intervals calculated on small numbers are not meaningful.

2.8 Conclusions Drawn From Studies

These data demonstrate acceptable performance using the VITROS Anti-HIV 1+2 assay when testing specimens collected as serum or plasma (EDTA heparin and citrate) and processed according to product instructions for use.

The VITROS Anti-HIV 1+2 Reagent Pack and Calibrator can be stored for up to 52 weeks at 2-8ºC (36-46ºF). After opening, the Reagent Pack can be stored on-board the VITROS Analyzer (4-8ºC (39-46ºF), ≤40% relative humidity) for up to 8 weeks, and the Calibrator stored for up to 13 weeks at 2-8ºC (36-46ºF) or -20ºC (-4ºF) (with no more than one freeze-thaw cycle).

The preservative systems that the VITROS Anti-HIV 1+2 assay reagents are formulated and have been shown to meet USP 23 requirements at 52 weeks for Assay Reagent, Conjugate and Calibrator.

The VITROS Anti-HIV 1+2 assay demonstrated adequate precision estimates for within day and between day for each site as well as across all sites, and between replicates, between days, between sites and between lots when these variables were introduced.

The VITROS Anti-HIV 1+2 assay has been shown to perform adequately over a 28-day calibration interval.

The VITROS Anti-HIV 1+2 assay agrees with a reference anti-HIV 1+2 assay in 14 out of 20 seroconversion panels. Five (5) of the 20 panels were detected as being reactive one bleed sooner than determined by the reference assay. The reference assay became repeatedly reactive one bleed earlier for the final panel.

In addition, there was no unacceptable cross reactivity demonstrated when testing samples from individuals with other conditions or disease states or individuals vaccinated with the Influenza vaccine. There was no unacceptable interference demonstrated when testing samples spiked with hemoglobin, triolein, or bilirubin or samples with protein concentrations of 6-8 g/dL or high cholesterol. Samples with protein concentrations>9 g/dL may give falsely reactive results.

Based on the results of the clinical laboratory studies, the results of the VITROS Anti-HIV 1+2 assay, in conjunction with other serological evidence and clinical information, may be used as an aid in the diagnosis of infection with HIV-1 and/or HIV-2 in persons with signs or symptoms of, or at risk for, HIV infection.