Home Search Abstracts View Session E-mail Abstract Author


Session 183 Poster Abstracts
Detecting Acute/Incident Infection
Session Day and Time: Monday, 1-2:30 pm
Poster Hall


992
A Clinical Study of Antigen/Antibody Rapid Testing for Acute HIV Infection
Christopher Pilcher*1, B Louie2, S Keating3, M Pandori2, F Fish4, T Keren4, M Lebedeva3, S Liska2, M Busch3, and F Hecht1
1Univ of California, San Francisco, US; 2San Francisco Publ Hlth Lab, CA, US; 3Blood Systems Res Inst, San Francisco, CA, US; and 4Inverness Med Innovations

Background:  Existing approaches to diagnosing acute HIV infection (e.g., RNA pooling) are unsuitable for point-of-care screening and cannot be used in resource-limited settings. The newly introduced Determine HIV-½ Ag/Ab Combo is the first simple-to-use, rapid (20 to 30 minute) HIV test to be introduced that detects both p24 antigen and anti-HIV antibodies. p24 antigen and antibody are detected as separate lines on the assay. 

Methods:  We evaluated the diagnostic performance of novel, rapid test-only acute HIV screening algorithms in the University of California–San F rancisco Options acute HIV infection screening program. Eligible subjects had symptoms consistent with an acute retroviral syndrome or a high risk exposure to HIV, with a negative HIV test result within 12 months. For this study we tested all subjects using the Determine rapid Ag/Ab test, and the Clearview StatPak rapid Ab test; we defined HIV status using a series of gold-standard tests: a third-generation high-sensitivity HIV antibody enzyme immunoassay (Vitros ECi); individual specimen HIV RNA (Roche); HIV WB (Biorad); and a less-sensitive modification of the Vitros assay to define recent HIV infection (LS-Vitros; cutoff S/C<14).  

Results:  Among 168 subjects tested, 87 were HIV and 81 were HIV+ (9 acute; 59 recent; 13 long-term infections by LS-Vitros assay). Used as a single screening assay, the Determine rapid Ag/Ab test identified 79 of 81 (97.5%) HIV infections, including 7 of 9 (77.8%) StatPak rapid Ab-nonreactive acute HIV infections. The 2 specimens that were missed by both the Determine rapid Ag/Ab and StatPak rapid Ab test had >500,000 copies HIV RNA/mL and also very low level antibodies (detectable only by the Vitros ECi assay, but undetectable by the LS-Vitros assay (S/C <0.1). The Determine rapid Ag/Ab test gave one false positive result.

Algorithm

Sens (95CI)

Spec (95CI)

Rapid Ag/Ab Combo (Determine)

97.5 (92.4 to 99.6)

98.9 (94.3 to 99.9)

Rapid Ag/Ab + Ab (Determine+StatPak)

97.5 (92.4 to 99.6)

100 (96.4 to 100)

Rapid Ab test (StatPak)

88.9 (80.9 to 94.5)

100 (96.4 to 100)

Non-rapid, sensitive Ab test  (Vitros ECi)

95.1 (88.8 to 98.5)

100 (96.4 to 100)

Conclusions: The Determine rapid HIV Ag/Ab test appears to have sufficient clinical performance to merit full scale trials of acute HIV screening.  Successful introduction of an effective point-of-care screening test for acute HIV infection would have implications for HIV testing and HIV prevention programs, and for the conduct of pathogenesis and prevention trials.