Performance Evaluation of the Aptima Assays in Comparison with the cobas 6800 Assays for the Detection of HIV-1, HBV, and HCV in Clinical Samples

Abstract

Background: Accurate and consistent viral load (VL) quantitation of HIV type 1 (HIV-1), hepatitis B virus (HBV), and hepatitis C virus (HCV) is important for diagnosis and clinical monitoring. Assay results have to be concordant and compatible across laboratories. We evaluated the performance of three Aptima assays (Hologic, San Diego, CA, USA) and compared their VL values with corresponding cobas 6800 assay (Roche Diagnostics, Mannheim, Germany) results, using 840 clinical samples.

Methods: The correlation between VL results obtained using the two assays was evaluated in terms of analytical sensitivity, precision/reproducibility, linearity, and cross-reactivity. Agreement rates were determined using kappa statistics. The overall agreement of VL values was examined using Passing-Bablok regression analysis.

Results: All CVs were within 5%; the assays had good precision for detecting all three viruses. The linearity of quantitation assessed using three AccuSpan linearity panels (Seracare, Milford, MA, USA), was excellent for the Aptima assays. For HIV-1 and HCV, the results of both assays showed excellent agreement (κ=0.89 and 0.90, respectively) while for HBV, the results showed good agreement (κ=0.69). For analytical sensitivity, the VLs required for a 100% detection rate of HIV-1, HBV, and HCV were 20 copies/mL, 7.5 IU/mL, and 5.0 IU/mL, respectively. The results for HIV-1, HBV, and HCV obtained using both assays correlated strongly (R²=0.97, 0.93, and 0.95, respectively).

Conclusions: The cobas 6800 and Aptima assays, with fully automated and high-throughput molecular platforms for HIV-1, HBV, and HCV VL measurements, show good analytical performance and a strong correlation between results. The study results suggest that the assays can be used interchangeably for long-term monitoring of chronic infections.

Keywords: Agreement; Analytical sensitivity; Aptima assay; HIV; Hepatitis B virus; Hepatitis C virus; Performance; Viral load; cobas 6800 assay.

Fig. 1

Aptima results obtained with the AccuSpan Linearity Panels. The values are presented as log copies/mL for HIV-1 (A) and log IU/mL for HBV (B) and HCV (C) (N=2 per target concentration level).

Fig. 2

Method comparison for HIV-1 (N=196), HBV (N=182), and HCV (N=126) VL assessments. (A) Passing–Bablok regression results for 196 quantitated HIV-1 clinical samples that were assayed with the Aptima and cobas 6800 assays. (B) Bland–Altman plot of the difference between the Aptima and cobas 6800 HIV results vs. the mean. The mean bias was –0.27 log copies/mL, with 95% CIs ranging from –0.84 to 0.29 log copies/mL. (C) Passing–Bablok regression for 182 quantitated HBV clinical samples that were assayed with both the Aptima and cobas 6800 assays. (D) Bland–Altman plot of the differences between the Aptima and cobas 6800 HBV results vs. the mean. The mean bias was –0.02 log copies/mL, with 95% CIs ranging from –0.75 to 0.71 log copies/mL. (E) Passing–Bablok regression results for 126 quantitated HCV clinical samples that were assayed with the Aptima and cobas 6800 assays. (F) Bland–Altman plot of the differences between the Aptima and cobas 6800 HCV results vs. the mean. The mean bias was –0.14 log copies/mL, with 95% CIs ranging from –1.00 to 0.72 log copies/mL. Abbreviation: CI, confidence interval.