Evaluation of the virtues and pitfalls in an HIV screening algorithm based on two fourth generation assays – A step towards an improved national algorithm

Background: Fourth-generation immunoassays used for HIV screening, simultaneously detect anti-HIV antibodies and HIV-1 P24 antigen, but are prone to false-positive results. Usually, they are followed by highly specific third-generation assay, able to differentiate between HIV-1/2 infections. In Israel, screening algorithm is based on consecutive testing by two fourth-generation assays and confirmation by a third-generation test. Objectives: To evaluate the performance of this algorithm. Study design: Architect HIV1/2 Combo (Combo) reactive results were tested by Vidas HIV Duo Ultra (VD). Confirmation was by INNO-LIA HIV 1/2 or Geenius assays. Five-year results were retrospectively analyzed. HIV true positives (TPs), acute infected (AI), false-positives (FPs) and HIV negatives, were as defined by the algorithm. Results: 501,338 individuals were screened, of which 956 were TPs, 64 AI and 30 F Ps. Specificity was almost 100% and positive predictive value 97%. VD was negative in 94% of confirmed Combo false-reactive individuals. The Combo results in the first tested sample differed substantially between TPs, AI and FPs, enabling the determination of a cutoff value that distinguished 94% of TPs and AI from FPs. Conclusions: An algorithm is suggested that will use a single sample collection. HIV negative diagnosis will be based on Combo unreactive or Combo reactive/VD negative results. HIV positive diagnosis will be based on Combo reactive/ VD positive results, given a Combo value above a designated cutoff. Below this cutoff samples will be tested by a molecular assay. Since HIV-2 rarely occurs in Israel, the use of a third-generation confirmation assay should be discussed.