CLINICAL%20TRIAL%20OF%20THE%20HEMA-STRIP%20HIV%20RAPID%20TEST%20USING%20FINGERSTICK%20BLOOD,%20WHOLE%20BLOOD,%20PLASMA,%20AND%20SERUM%20Niel%20T.%20Constantine1,%20Dan%20Bigg2,%20Daniel%20Cohen3,%20Joanne%20Court4,%20Alison%20Jones5,%20Fassil%20Ketema1,%20Thomas%20Koppes6,%20Vincent%20Marsiglia7,%20Harry

1
CLINICAL TRIAL OF THE HEMA-STRIP HIV RAPID TEST
USING FINGERSTICK BLOOD, WHOLE BLOOD, PLASMA,
AND SERUM Niel T. Constantine1, Dan Bigg2,
Daniel Cohen3, Joanne Court4, Alison Jones5,
Fassil Ketema1, Thomas Koppes6, Vincent
Marsiglia7, Harry Prince81University of
Maryland, Baltimore, MD, 2Chicago Recovery
Alliance, Chicago, IL, 3Fenway Community Health,
Boston, MA, 4MDC Associates, Boston,
MA,5Tapestry Health Services, Northhampton, MA,
6Northstar Medical Center Chicago, IL, 7BDC
Laboratory, Baltimore, MD, 8Focus Technologies,
Cypress, CA
Poster 2 HIV Diagnostics Meeting March
2005 Orlando, Florida
Background

• Insert your information here

RESULTS
RESULTS
HIV tests are indispensable for providing results
in a clinically-relevant time frame for
occupational exposures and women in labor without
known HIV status, and in public health clinics to
maximize the reporting of results to subjects in
a timely manner to minimize the effects of loss
to follow up. Moreover, it is increasingly
becoming important, especially in developing
nations where resources are scarce, that
alternate sample matrices/media (such as
fingerstick whole blood and saliva samples) be
available for testing by HIV assays.
Of the 1515 samples tested, 947 were classified
by the reference test(s) as being from
non-infected subjects, and 568 were from infected
persons.
The positive and negative predictive values among
all populations were estimated at 99.8 and
99.7, respectively. The sensitivities for the
different blood matrices were between 99.5 and
99.7, and the specificities were all 99.9.
Only one site produced sensitivity values below
99, and all sites produced specificity values
above 99.1.
Site summary with participant populations
(n1,515)
Overall Performance Characteristics (n1,515) -
Positive and Negative Predictive Value
Objectives
LR Low Risk, HR High Risk, Pos Positive
To evaluate the performance characteristics of a
novel rapid HIV test in a large number of persons
from several populations To determine the
concordance of results when using several
different blood matrices.
The Hema-Strip HIV test correctly identified 1513
of the 1515 samples producing 566 positive
results and 949 negative results. Overall, there
were 3 false negatives and 1 false positive. This
resulted in an overall sensitivity of 99.5 (95
CI 99.4, 100) and an overall specificity of
99.9 (95 CI 99.4, 100), thereby exceeding the
98 lower bounds of acceptable performance.
Performance Characteristics by Population Low
Risk - Positive and Negative Predictive Value
Methods
Sites Seven sites from 4 states (Maryland,
Illinois, Massachusetts, and Texas) were selected
to enroll study participants for the evaluation
of the rapid test. Prior to their participation
in the study, participants signed consent forms
approved by the Institutional Review Board at the
respective study sites. Samples A total of
1,515 participants were enrolled and tested in
this evaluation and included 504 at low risk for
HIV, 511 at high risk, and 500 persons known to
be infected with HIV. Fingerstick whole blood,
venous whole blood, plasma, and serum sample sets
were collected from each participant and tested
by the Hema-Strip HIV test, and results were
compared to FDA approved reference tests. Rapid
and Reference Tests The Hema-Strip HIV (SDS,
Beverly, MA) is a rapid, one-step, lateral flow
immunoassay that incorporates a protein A
colloidal gold conjugate, includes a procedural
control, and provides results within 15 minutes.
The reference assays included the Abbott HIVAB
HIV-1/HIV-2 (rDNA) EIA Kit (Abbott Laboratories,
Abbott Park, IL), the Genetic Systems HIV Types 1
and 2 (recombinant and synthetic peptide) EIA,
(Bio-Rad Laboratories, Redmond, WA), Genetic
Systems HIV-1 Western blot (Bio-Rad Laboratories,
Redmond, WA), and the Roche Amplicor HIV-1
Monitor Test (version 1.5) (Roche, Indianapolis,
IN). Algorithm Serum samples from each
participant were tested by the Abbott HIV 1/2 EIA
and the Hema-Strip HIV test. Samples initially
reactive by EIA or the Hema-Strip HIV test were
repeated in duplicate by the same assay. Western
Blots were performed on specimens exhibiting a
repeatedly reactive Abbott EIA and/or a reactive
result on the Hema-Strip HIV test. Western Blots
were not performed on samples from known HIV
Positive participants as long as the EIA was
reactive. Bio-Rad Genetic Systems HIV-1/2 EIA
testing (in triplicate) and the Roche Amplicor
NAT were performed on all discordant samples for
resolution. Interpretation of Results
Sensitivity, specificity, and positive and
negative predictive values of the Hema-Strip HIV
test were determined for each of the 4 sample
matrices. Overall, the sensitivity was based on
568 (37) true HIV positive samples and the
specificity was based on 947 (63) true negative
samples. All statistical tests were performed in
StatXact, and 95 confidence intervals were also
calculated for the performance characteristics of
the Hema-Strip HIV test.
Performance Characteristics by Population High
Risk - Positive and Negative Predictive Value
Performance Characteristics by Population Known
HIV Positive - Positive and Negative Predictive
Value
Conclusion
The Hema-strip HIV test has exhibited excellent
performance characteristics when evaluated at
multiple sites and using four blood matrices.
The test meets expectations for use in low risk,
high risk, and HIV positive populations. The
Hema-Strip HIV test is simple to perform,
requires minimal procedural steps, and is
suitable for point of care testing applications.
It is currently pending FDA PMA approval.