BACKGROUND: Roche Diagnostics recently announced that the COBAS® AmpliPrep/COBAS® TaqMan (CAP/CTM) system, widely used across low-and middle-income countries, will be phased out by early 2024 and replaced by the Roche cobas® 5800 System (c5800). The analytical performance of the c5800 for HIV-1 Quantitative Nucleic Acid Test for viral load (VL) using plasma and HIV-1/HIV-2 qualitative testing using dried blood spots (DBS) was independently evaluated for implementation consideration in the President's Emergency Plan for AIDS Relief (PEPFAR)-supported countries.
METHODS: Analytical evaluations of the cobas® HIV-1 Quantitative Test and the cobas® HIV-1/HIV-2 Qualitative Nucleic Acid Test were conducted. HIV-negative plasma or whole blood samples were spiked with WHO 4^th HIV-1 International Standard, 2^nd HIV-2 International Standard, or cultured virus. Testing was performed using the HIV-1 Quantitative or HIV-1/HIV-2 Qualitative Nucleic Acid Test workflow. Analytical performances, including precision, linearity, subtype detection, and cross-contamination, were evaluated. For the qualitative assay, reproducibility, cross-contamination, and subtype coverage for HIV-1 A, B, C, D, CRF02-AG, and HIV-2 were determined. For both assays, the limit of detection (LOD) was calculated using PROBIT analysis, and error rates were assessed.
RESULTS: The LOD for the HIV-1 Quantitative Test was 37.1 copies/mL, the LODs for HIV-1/HIV-2 Qualitative Test were 299 copies/mL and 1425 copies/mL for HIV-1 and HIV-2, respectively. Testing of forty or fifty replicates of HIV-1 plasma or HIV-1/HIV-2 DBS samples over five days, by two testers with different reagent lots, showed 100% reproducibility. The five major HIV-1 subtypes evaluated were all detected. No cross-contamination was detected. The error rate was 0% for HIV-1 Quantitative Test from 435 tests and 0.48% for the HIV-1/HIV-2 Qualitative Test from 415 tests. The correlation between the nominal and actual VL concentration of five subtypes was extremely high with the R² correlation coefficients were all 0.996 or higher.
CONCLUSIONS: We report here the first independent evaluation for HIV-1 Quantitative and HIV-1/HIV-2 Qualitative workflows on the new c5800. The manufacturer’s claims were verified. The c5800 combines the capacity to conduct HIV-1 VL and HIV infection with differentiation of HIV-1 and HIV-2, which will prove to be a useful tool in HIV diagnosis and treatment monitoring.

Download the e-Poster (PDF)