Evaluation of fingerstick blood point-of-care testing of hepatitis B DNA for enhanced hepatitis B treatment decision making: a diagnostic accuracy study

Abstract

Hepatitis B virus (HBV) DNA testing is essential for the management of HBV infection. Routine HBV DNA tests in central laboratories are expensive and require processed venous blood, limiting accessibility. This study is the first published assessment of the point-of-care Xpert HBV DNA assay performance using fingerstick capillary blood compared with standard-of-care venous blood testing. Participants with chronic HBV infection were enrolled from six hospitals. Fingerstick capillary blood was tested using Xpert HBV Viral Load assay (quantification lower limit: 100 IU/mL). Venipuncture whole blood was tested with COBAS AmpliPrep/COBAS TaqMan HBV DNA Test (gold standard). The sensitivity and specificity of the Xpert were evaluated for identifying HBV DNA ≥100 and >2,000 IU/mL. Agreement between quantitative measurements of assays was assessed. A total of 246 participants were included (median age 45, 46% female, 18% HBeAg positive, 48% on HBV treatment, 6% with cirrhosis). For HBV DNA ≥100 IU/mL, the sensitivity and specificity of the Xpert were 97.0% (95% CI: 94.9, 99.1) and 90.3% (86.6, 94.0), respectively. For HBV DNA >2,000 IU/mL, the sensitivity and specificity were 95.3% (92.7, 98.0) and 95.0% (92.4, 97.8), respectively. Viral load differences in non-concordant samples ranged from 0.1 to 1.1 log IU/mL. Overall, the Xpert viral loads were a mean 0.12 log IU/mL higher than gold standard (95%CI: -0.43, 0.67). In conclusion, minimal differences in HBV DNA levels were identified between the Xpert and gold standard assays, with differences in non-concordant results unlikely to impact clinical decisions. This evidence supports developing a dedicated Xpert HBV DNA fingerstick assay for decentralized care, crucial for remote, resource-limited settings and hard-to-reach populations, including prenatal care for women with HBV.IMPORTANCEThis study represents the first assessment of a point-of-care hepatitis B virus (HBV) DNA assay using fingerstick capillary blood (Xpert HBV Viral Load assay). Our findings demonstrated high sensitivity and specificity for the point-of-care test, with close agreement between the point-of-care and standard-of-care assays across the full quantitative spectrum of HBV viral load measurements. Importantly, the differences between the assays in participants with non-concordant results were not substantial enough to alter clinical management, suggesting that this point-of-care method is both accurate and reliable for clinical use. By highlighting the potential for decentralizing HBV care, our research provides compelling evidence to support the development of a dedicated Xpert HBV DNA point-of-care test. Such a development could greatly benefit patients in remote and resource-limited settings, where access to laboratory-based testing is limited.

Keywords: HBV; HBV DNA; diagnostics; linkage to care; sensitivity; specificity; viral load.

Fig 1

Overview of the participants’ flow.

Fig 2

(A) Bland–Altman bias plot of differences in HBV viral loads detected by the point-of-care Xpert HBV viral load assay and the standard-of-care assay (n = 130, bias: 0.118 log IU/mL, 95% limits of agreement: 0.430 to 0.666 log IU/mL); (B) linear correlation between HBV viral loads detected by the point-of-care Xpert HBV viral load assay and the standard-of-care assay (n = 130, r = 0.9824, P < 0.001).