Objective: To evaluate the quantification performance of the new Cepheid GeneXpert HIV-1 viral load assay, on a wide panel of HIV-1 variants.
Methods: Clinical performance was evaluated relative to the Abbott RealTime HIV-1 assay on 285 HIV-1 seropositive samples selected to cover the assays quantification range (40 copies/mL-10,000,000 copies/mL), and included RNA undetectable or detected seropositive samples. The panel comprised 120 subtype B, 150 non-B, and 15 nontypable clinical samples; serial dilutions of 18 viral supernatants representative of the divergent viruses of HIV-1 groups N, O, and P were also tested.
Results: Based on samples selected according to the Abbott assay viral loads (VL), the Cepheid assay detected or quantified 222/285 (78%) samples and the Abbott assay 240/285 (84%). Xpert yielded VLs for 162 (76%) of the 213 quantifiable samples with Abbott. This difference corresponded to 51 samples with VL >40 copies/mL by the Abbott assay (all below 200 copies/mL) but detected (n = 40) or undetectable (n = 11) by the Cepheid assay. VL of samples quantifiable by both assays (n = 162) showed very strong correlation, with a Spearman correlation coefficient of 0.985 and a Bland-Altman's mean of differences of -0.01. Performance for quantification of the non-M samples showed very good correlation, with significantly higher values with Cepheid for the group N and 2 group O samples.
Conclusions: Our study showed that the Xpert HIV-1 VL assay offered very good performance for detection and quantification of the current HIV-1 genetic diversity; differences reported at the threshold could be an issue and requires further evaluations. The practicability of this new assay makes it suitable for low-income countries, where it could facilitate and improve follow-up of patients, as well as for high-income regions.