Clinical Validation of a Point-of-Care Multiplexed In Vitro Immunoassay Using Monoclonal Antibodies (the MSD Influenza Test) in Four Hospitals in Vietnam

Abstract

ABSTRACT Point-of-care (POC) diagnostic tests for influenza can considerably shorten the time to clinical decision making. An investigational POC test based on a multiplexed immunoassay was developed by Meso Scale Diagnostics, LLC (MSD), with the objective to make a more sensitive rapid test that can also subtype influenza A viruses (1977 H1, H3, and H5). Between February and November 2010, we conducted a prospective multicenter study at four hospitals in Vietnam and compared the performance of this test to that of the WHO/CDC real-time reverse transcriptase PCR (RT-PCR) on nasal and throat swab specimens from patients presenting with influenza-like illness. Five hundred sixty-three adults and children with a median age of 25 months were enrolled. Sensitivity and specificity of the test with combined results from nasal and throat swab samples were 74.0% (131/177) and 99.7% (351/352), respectively, compared to RT-PCR. The POC test was as sensitive for influenza virus B as for influenza virus A (74.4% [64/86] versus 73.6% [67/91]). The positivity rate was associated with lower cycle threshold values (a marker for higher viral loads), sample type (73.6% for nasal swab versus 52.4% for throat swab), and younger age. A total of 210 (18.7%) out of 1,126 MSD tests failed, and for 34 (6%) of patients, both test samples failed (these were excluded from the performance analysis). Subtyping could be assessed only for influenza virus A/H3N2, as 1977 H1N1 was not circulating at the time and no H5N1-infected patients were enrolled, and was successful only in 9/54 patients infected with H3 influenza virus who had a positive POC test result for influenza virus A. This novel POC test provided highly sensitive detection of influenza viruses A and B compared to the reported sensitivities of other rapid tests. However, 18.7% of tests failed for technical reasons and subtyping for H3 was poor. Drawbacks to the technology include the requirement for a dedicated reader instrument and the need for continual updating of subtyping antibodies within the test array.