Performance of the BED-CEIA in population infected with HIV1 recombinant viruses CRF06cpx and CRF06A
Abstract number: P1266
Huik K., Avi R., Sadam M., Karki T., Denks K., Krispin T., Ustina V., Rüütel K., Uusküla A., Murphy G., Lutsar I.
Objectives: The Calypte HIV-1 BED Incidence EIA (BED-CEIA) was developed to overcome the HIV subtype dependence associated with many of the assays used in the identification of recent HIV infection. Indirectly, the assay determines the proportion of HIV-1 specific IgG to the total human IgG and has been used in populations containing subtypes B, D, E and C. However, the use of this assay in populations infected with recombinant viruses has not been described. To evaluate the assay performance in a population infected with HIV-1 recombinant viruses CRF06 cpx and CRF06A and to determine the prevalence of recent infections among newly diagnosed subjects.
Methods: The blood sera were collected in subjects with the first positive HIV test but unknown negative test in the past (group A; n = 122), in those with the history of positive HIV test of more than 6 months ago (group B; n = 65) and in subjects with first positive HIV test and known HIV negative test within last 3 months (group C; n = 6); all nave to antiretroviral agents. The BED-CEIA was performed as per manufacturers instructions with the optical density (OD) values of test specimens normalised by a ratio using a calibrator to minimise interrun variation. As per instructions samples were tested in singleton but those with a normalised OD (ODn) 1.2 were re-tested in triplicate and the median values were used for evaluation. An ODn value <0.8 was considered to indicate recent infection. The receiver operating curve (ROC) was constructed using ODn data of group B and C. The performance of curve was considered fair if the AUCROC value was between 0.7 and 0.8.
Results: The mean SD ODn values in subgroup A, B and C were 0.70 0.69, 0.86 0.68, and 0.32 0.19, respectively. A total of 87/122 (71%) subjects in group A, 31/65 (48%) in group B and all 6 in group C were categorised into recent infections. The calculated sensitivity and specificity with the NPV and PPV were 100% and 52%, 100% and 16%, respectively. The ROC curve constructed using ODn values of subgroup B and C showed an AUCROC value of 0.77 SD 0.08. No correlation between viral load and ODn value was observed.
Conclusion: In subjects infected with CRF06 cpx or CRF06A viruses the BED-CEIA has good sensitivity but lacks required specificity. Whether the poor specificity of the assay is due to the viral subtype or associated with the frequent reinfections commonly seen in IDUs, or another reason remains to be identified in future trials.
|Session name:||19th European Congress of Clinical Microbiology and Infectious Diseases|
|Location:||Helsinki, Finland, 16 - 19 May 2009|
|Back to top|