The GeneXpert HBV Viral Load test is a simplified tool to scale up screening and HBV monitoring in resource-limited settings, where HBV is endemic and where molecular techniques to quantify HBV DNA are expensive and scarce. However, the accuracy of field diagnostics compared to gold standard assays in HBV-endemic African countries has not been well understood. We aim to validate the diagnostic performance of the GeneXpert HBV Viral Load test in freshly collected and stored plasma and dried blood spot (DBS) samples to assess turn-around-time (TAT) for sample processing and treatment initiation, to map GeneXpert machines and to determine limitations to its use in The Gambia. Freshly collected paired plasma and DBS samples (n = 56) were analyzed by the GeneXpert test. Similarly, stored plasma and DBS samples (n = 306, n = 91) were analyzed using the GeneXpert HBV test, in-house qPCR and COBAS TaqMan Roche. The correlation between freshly collected plasma and DBS is r = 0.88 with a mean bias of −1.4. The GeneXpert HBV test had the highest quantifiable HBV DNA viremia of 81.4% (n = 249/306), and the lowest was detected by in-house qPCR at 37.9% (n = 116/306) for stored plasma samples. Bland–Altman plots show strong correlation between GeneXpert and COBAS TaqMan and between GeneXpert and in-house qPCR with a mean bias of +0.316 and −1.173 log10 IU/mL, respectively. However, paired stored plasma and DBS samples had a lower mean bias of 1.831 log10 IU/mL, which is almost significant (95% limits of agreement: 0.66–3.001). Patients (n = 3) were enrolled in the study within a TAT of 6 days. The GeneXpert HBV test displayed excellent diagnostic accuracy by detecting HBV viremia in less than 10 IU/mL.
Read full abstract