Accuracy of Tuberculosis Molecular Bacterial Load Assay to Diagnose and Monitor Response to Anti-Tuberculosis Therapy: A Longitudinal Comparative Study with Standard-of-Care Smear Microscopy, Xpert MTB/RIF Xpert-Ultra, And Culture

Abstract

Background: Tuberculosis (TB) takes long to treat, and thus requires effective tools to monitor treatment response and guide clinical decision making. Tuberculosis-Molecular Bacterial Load Assay (TB-MBLA), a ribosomal RNA based test was evaluated for accuracy to diagnose and monitor treatment response.

Methods: Presumptive TB cases were enrolled into diagnostic arm and tested for TB by TB-MBLA in comparison with Xpert MTB/RIF Ultra (Xpert-Ultra) and smear microscopy (SM) using Mycobacteria Growth Indicator Tube (MGIT) culture as a reference test. Xpert-Ultra-positive cases were enrolled on anti-TB therapy and monitored for treatment response at weekly-monthly intervals until month-6 and then three months after treatment.

Findings: 210 participants, median age 35 years (IQR: 27-44),135 (64%) males, 72 (34%) HIV positive were enrolled. The sensitivities of TB-MBLA and Xpert-Ultra were similar, 99% but different specificities, 91% and 78% respectively. Ten participants were Xpert-Ultra trace-positive, 80% of whom were negative by TB-MBLA and MGIT culture. SM had lower diagnostic sensitivity, 75% but higher specificity 98%. Among SM-negative and HIV-positive participants, TB-MBLA and Xpert-Ultra sensitivities were 92% and 100% respectively. 129 participants were enrolled into the treatment follow-up arm. Positivity for TB dropped with treatment in all tests, but the rate was slower with Xpert-Ultra. Consequently, 33% of participants were still Xpert-Ultra positive at the end of treatment but were clinically well at 3-months post-treatment. Rate of conversion-to-negative of DNA-based Xpert-Ultra was 3·3-fold slower than the rRNA-based TB-MBLA. TB-MBLA-measured bacillary load, among month-2 SM-positive cases who received extra month of intensive treatment was not different from SM negative cases.

Interpretation: TB-MBLA is consistent with Xpert-Ultra for pre-treatment diagnosis of TB but is more accurate for treatment response monitoring than Xpert-Ultra and standard-of-care SM.

Funding Information: Emmanuel Musisi’s doctoral research was supported by the European and Developing Countries Clinical Trial Partnership (EDCTP)-funded PanACEA II studentship (grant number TR1A2015-1102) and the University of St Andrews St Leonards scholarship. Funding from Makerere University Research and Innovation Fund (MAKRIF) by the Government of Uganda to Emmanuel Musisi and Samuel Wamutu supported collection and processing of specimens. Enrolment was funded by NIH R01 HL128156 and NIH R01 HL143998 grants.

Declaration of Interests: Wilber Sabiiti and Stephen Gillespie provide a pro bono advice for a company that is developing TB-MBLA for clinical use. All other members declare a no conflict of interest. No contributing author (s) declared him/herself to be (a) medical writer(s) or editor(s).

Ethics Approval Statement: The project was approved by the the University of St Andrews Teaching and Research committee (UTREC) approval [Approval code: MD14702], Makerere University School of Medicine Research and Ethics committee [REC REF No. 2006-017] and Makerere University School of Biomedical Sciences Research and Ethics Committee [REC REF No: SBS 529] respectively. Before enrolment, participants consented to utilising their biological samples and clinical data for this study. All study related activities were conducted according to the guideline outlined in the Good Clinical and Laboratory Practice manual.

Keywords: Tuberculosis, TB-MBLA, Xpert MTB/RIF Ultra, Smear microscopy, Culture, Diagnosis, Treatment response monitoring, Accuracy