👤 Rashmi Ratnam

Fluoroquinolones (FQs) are key components of World Health Organization (WHO)-recommended regimens for multidrug-resistant tuberculosis (MDR-TB). Accurate detection of FQ resistance is essential for optimizing treatment. This study evaluated the concordance between the Second-Line Line Probe Assay (SL-LPA) and Liquid Culture Drug Susceptibility Testing (LC-DST) for detecting FQ resistance in Mycobacterium tuberculosis isolates. In this retrospective study, 1402 non-duplicate clinical isolates of MDR TB were tested using SL-LPA and LC-DST at a reference laboratory. Genotypic resistance was identified through mutations in the gyrA and gyrB genes identified by SL-LPA, while phenotypic resistance was determined using MGIT-based LC-DST at critical concentrations for fluoroquinolones. Targeted nanopore sequencing was performed on a subset of isolates with discordant molecular and phenotypic results to investigate resistance-associated mutations. SL-LPA detected FQ resistance in 907 (64.7%) isolates, whereas LC-DST identified resistance in 852 (60.8%) isolates. Using LC-DST as the reference standard, SL-LPA showed a sensitivity of 93.2%, specificity of 98.6%, positive predictive value of 99.2%, and negative predictive value of 88.7%. Overall concordance between the two methods was observed in 1292 (92.2%) isolates. Discordant results occurred in 110 (7.8%) isolates, mainly involving low-level resistance mutations or inferred resistance due to missing wild-type bands on SL-LPA. Nanopore sequencing of 15 discordant isolates identified high-confidence mutations (Asp94Tyr, Asp94Gly, Asp94Asn) and interim or low-confidence mutations (Ala90Val, Ser91Pro, Asp94Ala, gyrB Asn499Asp, Asp461Asn). SL-LPA demonstrates excellent specificity and positive predictive value for detecting FQ resistance; however, discordance associated with low-confidence mutations and heteroresistance highlights the importance of integrating molecular assays with phenotypic DST and sequencing to improve MDR-TB resistance detection and guide treatment decisions. Show less

Background & objectives Central TB division facilitated development of a line probe assay (LPA) artificial intelligence (AI) tool. The tool was developed, trained, and validated for performance by collecting more than 18,000 LPA strips across culture and drug susceptibility Testing (C&DST) laboratories. The Indian Council of Medical Research (ICMR)-National Institute for Research in Tuberculosis (NIRT) evaluated the LPAAI tool independently. The objective was to establish and verify an AI-driven system for automatically interpreting LPA strips, which are employed in tuberculosis drug resistance screening, to improve accuracy, consistency, and scalability across diverse laboratory settings. Methods The AI system integrates faster regions convolutional neural network (FR-CNN) for strip detection, detection transformer (DETR) for band localisation, and a hierarchical neural network (HNN) for classification of bands, loci, and drug labels. Independent validation was conducted by ICMR-NIRT using 2810 first-line (FL)-LPA and 241 reflex second-line (SL-LPA) across ten intermediate reference laboratories (IRLs). Results AI comparative models demonstrated an accuracy range of 92-100 per cent, with sensitivity between 80-100 per cent and specificity from 86-100 per cent for the tub, rpoB, katG, InhA, gyrA/gyrB,rrs, and eisgenes. The overall F1 score varies from 0.81 to 1.00, indicating perfect precision and recall. Interpretation & conclusions This AI system offers a novel, modular architecture capable of expert-level interpretation of LPA strips. The AI tool performs at par with expert readers and offers a reliable, scalable solution for LPA interpretation.AI tool adoption can reduce interpretation time, enhance result uniformity, and improve treatment delivery across India's TB programme, supporting national goals for TB elimination. Show less