Assessing the accuracy of the updated Fujifilm SILVAMP TB LAM II assay between lot numbers.

Abstract

Background: The Fujifilm SILVAMP TB LAM (FujiLAM) assay, a point-of-care test detecting urine-based lipoarabinomannans, has demonstrated notable diagnostic accuracy for TB. However, FujiLAM exhibits lot-to-lot variability, limiting its clinical use. A new version of the product, Fujifilm SILVAMP TB LAM II (FujiLAM II), was developed to address this issue. This study evaluated the lot-to-lot variability and diagnostic accuracy of the FujiLAM II.

Methods: We assessed three lots of FujiLAM II using 158 biobanked, frozen urine specimens from people living with HIV (PLHIV): 72 TB-positive and 86 TB-negative, according to the microbiological reference standard.

Results: Independent proportions of the results among the three FujiLAM II lots did not differ (Cochrane Q test: P > 0.37). Overall, sensitivities of the three FujiLAM II lots were 78.3%, 79.7% and 79.7%, and specificities were 92.9%, 93.0% and 93.0%, respectively. In patients with CD4 cell counts <200 cells/µL, sensitivities of the three FujiLAM II lots were 97.8%, 100% and 100%, and specificities were 91.7%, 91.8% and 91.8%, respectively.

Conclusion: The FujiLAM II demonstrated no lot-to-lot variability and exhibited high sensitivity and specificity for TB diagnosis in patients with CD4 cell counts <200 cells/µL. The FujiLAM II enhanced reproducibility as a TB diagnostic tool in PLHIV.