Calibration and validation of an oral fluid-based sensitive/less-sensitive assay to distinguish recent from established HIV-1 infections.

Sensitive/less-sensitive (S/LS) serum-based serologic methods have been developed to measure human immunodeficiency virus (HIV) incidence by distinguishing recent from established infections. Such methods require venipuncture. The goal of this study was to develop an alternative to serum-based S/LS testing using oral fluid (OF) as the testing medium. Serum/OF pairs were collected from 342 patients attending 15 Adolescent Trials Network (ATN) clinical sites. The sera were tested with the use of the dilutional Vironostika (DV; Biomerieux, Durham, NC) S/LS assay (DV(SOD=1.0)) as the reference against which an OF LS assay was calibrated using 40 of the OF pairs. Receiver operating characteristic (ROC) curve analyses pinpointed the OF LS test parameters that maximized concordance with the serum-based DV. Validation of the calibrated OF LS included testing of the remaining 302 serum/OF pairs. During calibration the maximum concordance with the DV was 95.2% and 89.5% for 21 recent and 19 established samples, respectively, at a 1:50 OF sample dilution and an optical density (OD) cutoff of 0.280. When applied to the validation sample set (N=302), the concordance was 73.6% for the recent samples and 89.6% for the established samples. The OF LS assay showed a good concordance with the serum-based reference S/LS assay. It presents an alternative to invasive specimen collection, and has the potential for increasing test compliance in young subjects. However, because of the uncertainty of the performance characteristics of the serum-based S/LS assay with which it was compared, further validation of the OF LS using seroconversion sample pairs is needed. (c) 2007 Wiley-Liss, Inc.