Clinical performance of Determine HBsAg 2 rapid test for Hepatitis B detection

Hepatitis B virus (HBV) infection is estimated to affect 292 million people worldwide, 90% of them are unaware of their HBV status. The Determine HBsAg 2 (Alere Medical Co, Ltd Chiba Japan [Now Abbott]) is a rapid test that meets European Union (EU) regulatory requirements for Hepatitis B surface antigen 2 (HBsAg) analytical sensitivity, detecting the 0.1 IU/mL World Health Organization (WHO) International HBsAg Standard. This prospective, multicentre study was conducted to establish its clinical performance. 351 evaluable subjects were enrolled, 145 HBsAg‐positive. The fingerstick whole blood sensitivity and specificity were 97.2% and 98.5% (15′ reading, reference assay cut‐off 0.05 IU/mL), sensitivity increasing to 97.9% with the prespecified cut‐off 0.13 IU/mL (EU regulations). The venous whole blood, serum and plasma sensitivity was 97.2%, 97.9%, and 98.6%, respectively (15′ reading); reaching 99%, 99.5% and 100% specificity. A testing algorithm following up an initial positive fingerstick test result with plasma/serum test demonstrates 100% specificity. The Determine HBsAg 2 test gives 15‐minute results with high sensitivity and specificity, making it an ideal tool for point‐of‐care testing, with the potential to enable large‐scale population‐wide screening to reach the WHO HBV diagnostic targets. The evaluated test improves the existing methods as most of the reviewed rapid tests do not meet the EU regulatory requirements of sensitivity.

an initial positive fingerstick test result with plasma/serum test demonstrates 100% specificity. The Determine HBsAg 2 test gives 15-minute results with high sensitivity and specificity, making it an ideal tool for point-of-care testing, with the potential to enable large-scale population-wide screening to reach the WHO HBV diagnostic targets. The evaluated test improves the existing methods as most of the reviewed rapid tests do not meet the EU regulatory requirements of sensitivity. (HBsAg) positivity for 6 months or longer, is a risk factor for serious liver diseases, such as cirrhosis and hepatocellular carcinoma. [1][2][3][4] As CHB infection may be asymptomatic, and screening programs are not widely implemented, a diagnosis of CHB is often obtained when latestage symptoms appear. Approximately two-thirds of CHB infected patients in Europe are unaware of their HBV status. [5][6][7][8][9] Published estimates show that 96000 people die each year in EU/EEA countries from HBV and hepatitis C virus (HCV)-related liver disease. 10,11 The Global Health Sector Strategy to eliminate viral hepatitis by 2030, approved by the 69th World Health Assembly in 2016, includes the diagnosis of 90% of HBV infected people by 2030. It was estimated in 2016 that only 10% of 292 million HBV infected individuals were diagnosed. Hence a significant increase in HBV screening programs is required. The emergence of directly acting antivirals for the treatment of HCV has increased the emphasis on screen and treatment programs to meet World Health Organization (WHO) targets 12 ; thus parallel screening of HBV could conveniently be incorporated into these new programs.
HBsAg is the outer coat of HBV, which is produced in excess during the course of infection and can be detected easily in blood.
HBsAg, together with HBV DNA, is the earliest indicator of acute infection and may be present before symptoms appear. It is also present in patients with chronic infection and is used to screen for and detect hepatitis B at different stages of the disease. 13,14 Identification of infected individuals and initiating treatment in a timely manner, before progression to significant liver disease, is imperative for maximizing treatment efficacy. 5,15 Currently, HBV screening in Europe is largely restricted to laboratory testing. However, rapid tests present several advantages for patient screening, such as easily obtainable results, testing convenience, the potential to facilitate the diagnosis earlier in the disease stage as well as to conduct testing outside established healthcare settings and to reach high-risk groups. 6 (whole blood only). The Abbott Architect quantitative HBsAg assay ("Architect"), with the cut-off 0.05 IU/mL, was used as the primary reference method. Prespecified analyses were also conducted using the cut-off 0.13 IU/mL, which is the requirement for analytical sensitivity for an HBsAg test in the EU. 23 2 | MATERIALS AND METHODS

| Study design
This study was conducted at multiple study sites in Europe. The applicable Ethics Committee for each participating site approved of the study and all subjects provided written informed consent.
The intended population included individuals of all ages who presented to participating clinical sites for HBsAg screening (highrisk population) or for routine follow-up of confirmed HBV, as well as relevant control populations, including Hepatitis A-, Hepatitis C-, and HIV-positive patients. Subjects who had already participated in this study at a previous date, or who were enrolled in a study evaluating an investigational drug and had started taking this drug (ie, had progressed past the screening stage), or who belonged to a vulnerable population as deemed inappropriate for the study by the study investigator(s), were not eligible for study enrolment.
The subjects provided one ethylenediaminetetraacetic acid (EDTA) whole blood tube, one tube for Serum separator tubes (SST) obtained by venipuncture, and a fingerstick blood sample. The fingerstick blood sample was obtained using an EDTA capillary blood collection tube (n = 223) or a Microsafe capillary blood collection tube (SAFE-TEC Clinical Products LLC, PA) (n = 125). After an aliquot of the EDTA whole blood venipuncture sample had been removed for Determine HBsAg 2 testing, the tube was centrifuged to obtain a 3404 | plasma specimen. Each sample type was evaluated using Determine HBsAg 2, and each sample type of each patient was measured once only. Fingerstick testing was conducted by nonlaboratory study staff members such as nurses and doctors; venous whole blood, plasma, and serum testing were conducted by nonlaboratory or laboratory study staff members according to local protocol. All testing personnel were blinded to the subject's clinical HBV status. To ensure blinding between sample types, each study staff member tested only one sample type from each subject during each testing occasion, except during batch testing sessions, where the samples were coded. Serum and plasma samples were frozen and batch-tested. A Determine HBsAg 2 test result is intended to be interpreted between 15 and 30 minutes after sample application, and to validate this time interval in this study every Determine HBsAg 2 test was interpreted at 15 minutes, and again at 30 minutes by the same operator. The Abbott ARCHITECT quantitative HBsAg assay ("Architect"), with the cut-off 0.05 IU/mL, was used as the primary reference method. In the EU, an analytical sensitivity of 0.13 IU/mL is required for an HBsAg diagnostic test. 23 Hence prespecified statistical analyses were also conducted using the cut-off 0.13 IU/mL. For the resolution of discrepancies and disease reclassification (see reference methods) testing was also performed using the Elecsys HBsAg II assay and Elecsys HBV core antibody testing. Basic demographic information and a brief medical history were collected from each subject. Results from the sites' standard of care HBV diagnostic testing were also recorded.

| Investigational Determine HBsAg 2 test
The Determine HBsAg 2 test is an in vitro, visually read, qualitative, immunochromatographic assay for the detection of HBsAg in human serum, plasma, fingerstick whole blood or venous whole blood.
The Determine HBsAg 2 test consists of single-use test strips and a chase buffer bottle, both stored at room temperature and requires no maintenance. The assay requires 50 µL specimen for a test, and the test result can be interpreted at 15 minutes and no later than at 30 minutes.
Details of the test can be found in the package insert. 24

| Study reference methods
The reference method for the study, the Abbott ARCHITECT quantitative HBsAg assay, uses a chemiluminescent microparticle immunoassay and was conducted using the Architect i2000 analyzer (Abbott Diagnostics, IL). A positive result was automatically repeated.
For a subset of the subjects who were positive on Architect (above 0.05 IU/mL) without an existing diagnosis of HBV, further testing for potential disease reclassification from HBV disease positive to HBV disease negative was conducted with the Elecsys HBsAg II assay, and with the Elecsys II confirmatory assay where applicable. This testing was performed using the MODULAR E170 analyzer (Roche Diagnostics GmbH, Mannheim, Germany), which uses an electrochemiluminescence immunoassay, and was also conducted for subjects with two or more sample types that showed discrepant results between the Determine HBsAg 2 test and Architect. As part of disease reclassification and discrepant result resolution, total HBV core antibody (immunoglobulin G + immunoglobulin M) analysis using the MODULAR E170 platform was also conducted. EDTA plasma or serum from each subject was used for the reference testing. The assays were conducted according to the manufacturers' instructions.

| Study population
This study enrolled 365 subjects across five clinical sites in the UK and one clinical site in Spain. In the UK, two sites recruited patients at hepatology clinics, one site recruited patients at a gastroenterology clinic, one at a digestive diseases clinic and a paediatric hepatology clinic, and one at a sexual health clinic. The clinical site in Spain was affiliated with five recruiting hospitals, of which four recruited patients at digestive disease clinics and one at an infectious disease clinic. Eleven subjects had no reference result available or improper sample handling and were excluded from the analyses.
The median age of all evaluable subjects was 49 years (range: 2 to 86 years). Eight children or adolescents (age <18 years) were enrolled into the study. One hundred sixty-three (46.4%) evaluable subjects were female.  To increase the diagnostic specificity, the Determine HBsAg 2 performance for fingerstick and venous whole blood was also calculated using an algorithm where a positive Determine HBsAg 2 fingerstick or venous whole blood test result is followed up with a Determine HBsAg 2 test using plasma or serum to confirm results (Table 3). This algorithm resulted in 100% specificity. The fingerstick results with quantitative values are illustrated in Table 4.