Abstract Rationale Drug-resistant tuberculosis (DR-TB), particularly rifampicin-resistant TB (RR-TB) and multidrug-resistant TB (MDR-TB), remains a major public health challenge in Thailand. Early and accurate diagnosis is crucial for achieving optimal treatment outcomes and preventing disease transmission. Molecular diagnostic tools, such as the Xpert MTB/Rifampicin (RIF) Ultra assay, have substantially improved the rapid detection of DR-TB; however, discrepancies between molecular results and phenotypic drug susceptibility testing (DST) have occasionally been observed. Therefore, this study aimed to evaluate the concordance between genotypic testing using the Xpert MTB/RIF Ultra assay—including probe-specific rpoB gene findings—and phenotypic DST results. Methods A retrospective study was conducted at the Central Chest Institute of Thailand between October 1, 2021, and September 30, 2025. Sputum test results were collected from 217 patients with RR-TB and MDR-TB. The study included sputum specimens that were detected as rifampicin-resistant using the Xpert MTB/RIF Ultra assay and had corresponding phenotypic DST results confirming rifampicin resistance or susceptibility. Concordance between genotypic and phenotypic DST results, along with descriptive statistics, was analyzed using STATA software version 15.1. Results A total of 217 RR-TB and MDR-TB isolates were identified during the study period. Of these, 87 isolates with both Xpert MTB/RIF Ultra and phenotypic DST results available were included in the analysis. Among these, 48 isolates were phenotypically rifampicin-resistant, while 6 were rifampicin-susceptible. There was good agreement between the two DST methods, with an overall agreement of 88.88%. The most common mutation pattern occurred in probes rpoB3 combined with rpoB4 (44 isolates, 50.57%), followed by rpoB3 alone (19 isolates, 21.84%). The concordance between Xpert MTB/RIF Ultra and phenotypic DST for most probe-specific rpoB genes showed moderate to excellent agreement except for rpoB1, where the overall agreement was 50% (Table 1). Conclusions This study demonstrates good consistency between genotypic and phenotypic DST results. The use of genotypic DST in place of conventional growth-based phenotypic DST for detecting RR-TB appears promising, given its shorter turnaround time and potential to enable prompt treatment decisions. For probe-specific detection, overall concordance for rpoB2, rpoB3, and rpoB4 was moderate to excellent, while further and larger studies are needed to clarify concordance for rpoB1. This abstract is funded by: None
Chiracharasporn et al. (2026) studied this question.