Risk assessment and transmission of fluoroquinolone resistance in drug-resistant pulmonary tuberculosis: a retrospective genomic epidemiological study

Fluoroquinolone is widely used to treat various bacterial infectious diseases. However, misuse of fluoroquinolone without proper prescription has led to a significant increase in fluoroquinolone resistance. Fluoroquinolone is essential for the treatment of multidrug-resistant tuberculosis (MDR-TB), and fluoroquinolone resistance is associated with poor treatment outcomes.¹⁴.Global prevalence data on FQ resistance are limited due to inadequate FQ testing facilities in many tuberculosis-endemic areas. Therefore, localized data are urgently needed to understand FQ resistance in MDR and non-MDR TB, which is critical to determine the feasibility of introducing a standardized shorter MDR TB regimen. In our study, the overall genotypic resistance rate of fluoroquinolones was 33% in MDR TB, 5.4% in non-MDR, and 16.5% in rifampicin-monoresistant tuberculosis patients. Sethi et al. (2020)¹⁵ reported 38.6% in RR isolates in India, and Li et al. (2024)¹⁶ reported 34.7% FQ resistance in MDR-TB patients in China. In an Indian study, Sharma et al. (2019)¹⁷ and Mamatha et al. (2018)¹⁸ reported fluoroquinolone resistance rates of 27.3% and 32%, respectively, which are lower than the rates reported in our study. Our results suggest a lower resistance rate than previous reports, but the 33% FQ resistance in MDR-TB in our study was significantly higher than the global rate (20.0%) reported by WHO (2020).¹⁹ Report. The high FQ resistance rate among MDR-TB patients in South India suggests that incorporation of FQ into treatment might result in ineffective treatments and worsening treatment outcomes. These findings highlight the urgent need for FQ resistance testing before initiating MDR-TB treatment. In our study, the detection rate of extensively drug-resistant tuberculosis (XDR-TB) was 1.33% (2 isolates), which is relatively lower than the reported global prevalence of XDR-TB and the 8.6% reported in India by Sethi et al., 2020¹⁵. The resistance rate to all FQs in our cohort among non-MDR-TB was 5.4%, which is relatively higher than the 0.8% reported by Kim et al., 2018¹⁴indicating a trend that warrants comparison with the recent spread of FQ resistance in tuberculosis-affected countries such as India.

Among 91 fluoroquinolone (FQ) resistant isolates, the frequency of gyrA Mutations were higher than gyrBwhich is consistent with the findings of Kabir et al.⁹The most common mutations in the gyrA Gene associated with FQ resistance in M. tuberculosis are S91P, A90V and D94A/N/Y/G/H. Most FQ-resistant isolates harbored a mutation at codon D94G, with A90V being the most common. Our study found that 49% of FQ-resistant isolates carried the D94G mutation, a significantly higher number than that reported by Tania Matsui et al. (2020).²⁰ report of 44% in Brazil. Our results are further supported by a recent study in Ethiopia, which found a gyrA/D94A gene mutation (2%) in FQ-resistant TB isolates²¹In addition, our study identified a rare gyrA Mutation at codon D94H in nine isolates, a mutation not frequently reported in other studies¹⁴. Among the 91 FQ-resistant isolates, 26 showed resistance to levofloxacin and low resistance to moxifloxacin, while 62 showed resistance to levofloxacin and high resistance to moxifloxacin. In addition, alanine, asparagine and serine are nonessential amino acids that promote brain functions, remove toxins and synthesize blood cells. Any functional alteration of these nonessential amino acids as a result of mutations may lead to difficulties in the production of proteins necessary for cell growth, maintenance and repair mechanisms.²²In our study, we observed a heteroresistant mutation pattern in the gyrA Gene shown by 39 (42.9%) isolates and characterized by the expression of the MUT probe and all WT probes, including codons 90, 91 and 94. This is consistent with a recent report by Dixit et al. (2023)²³which showed a mutation pattern of 39.3% heteroresistance in the gyrA Gene in India.

Heteroresistance (HR) is common in M. tuberculosis and is considered one of the most important steps in the development of drug resistance in bacterial isolates (Ye et al., 2021)²⁴. It can be a mixed infection when both resistant and susceptible strains infect a person simultaneously, or when a single clone changes from a susceptible strain to a resistant strain due to a genetic mutation under antibiotic pressure. It is worth noting that heteroresistance has been associated with limited treatment options and an increasing rate of adverse treatment outcomes, as reported by Rigouts et al.²⁵.

The prevalence rate of fluoroquinolone resistance (FQ) among non-MDR-TB in this study is 5.4%, which is higher than the global estimate of 0.8%, suggesting that caution is needed in the widespread use of FQ in the population. In this study, FQ resistance among MDR-TB and RR-TB was 32.7% and 16.5%, respectively, which is also higher than the global estimates (Dixit et al., 2023).²³FQ resistance in MDR/RR-TB in this study is 24.9%, higher than the global estimate of 18.0%²⁶Among newly diagnosed cases of H resistance, FQ resistance was 6.9%, lower than the 9.8% reported in a recent study in Pakistan.²⁷Similarly, FQ resistance in previously treated cases was 3.81% compared to 44.6% in the previous study report by Sethi et al.¹⁴ Among newly diagnosed MDR/RR-TB cases, FQ resistance was 23.4%, which was higher than the 21.82% reported in the recent study conducted in China.¹⁵ and 14.2% in an Indian study²⁸Similarly, FQ resistance in previously treated cases was 74.24% compared to the previous study report by Sethi et al. of 44.6%.¹⁵ Another study in India reported 72.8% FQ resistance¹⁰. The high FQ resistance was observed in previously newly diagnosed MDR/RR-TB cases, which might be due to the high transmission of the drug-resistant strains. The high FQ resistance rate in MDR and non-MDR-TB patients (H-resistance) might lead to ineffective treatment of H-monoresistance and unfavorable outcomes.

Of the 289 MDR/RR-TB isolates, 53% had a mutation at codon S450L of the rpoB Gen, which is lower than the rate of 77% previously reported by Tania Matsui et al.²⁰Among 1230 H-monoresistant isolates, 65% showed a mutation at codon S315T of the katG gene, resulting in high isoniazid resistance – a lower rate than the 72% previously reported. It is noteworthy that of the 53% with a mutation at the S450L codon of the rpoB gene and the 65% with a mutation at the S315T codon of the katG 27.9% and 5.9% of patients, respectively, had an increased risk of FQ resistance, a previously unreported finding. This information is valuable for policymakers as it provides timely evidence. In addition, this study provides important insights for physicians in their daily treatment practice and serves as important baseline information for researchers.

The study found that the rate of adverse outcomes was 42.9% in MDR/RR tuberculosis patients and 20.4% in non-MDR TB patients. Our study showed a higher rate of adverse outcomes (42.9%) in MDR/RR tuberculosis patients compared to a previous study in Ethiopia by Bogale et al. (2023).²⁹which reported a rate of 23.68%. Nair et al. (2017)³⁰ reported a 40% unfavorable outcome rate in India. Aaina et al. (2022)²² reported rates of 29.4% for MDR-TB and 14.5% for non-MDR-TB cases in India. The unfavorable outcome for non-MDR patients (H resistance) in our study was 20%, lower than the 29.4% reported by Aaina et al. (2021)²² in India. These unsuccessful treatment outcomes were significantly associated with FQ resistance. The increasing percentage of unfavorable outcomes could pose a risk for the transmission of tuberculosis-resistant forms and have a negative impact on the country's GDP.

The study has significant strengths, such as recruiting a large sample and using a variety of diagnostic methods. However, our study has several limitations. We relied on secondary data for patient characteristics and included only patients with a laboratory diagnosis of tuberculosis. Our study used a cross-sectional design and identified only associated factors, not risk factors, for FQ-resistant transmission. We were unable to determine whether FQ resistance was due to previous treatment because data on previous FQ use before DR-TB diagnosis were not available. Furthermore, drug susceptibility testing for drug-susceptible tuberculosis is not routinely performed, although FQ resistance in DS-TB could lead to adverse treatment outcomes. FQ resistance is higher in regions where these drugs are commonly prescribed and sometimes misused as fluoroquinolones.

This text highlights the worrying rise in FQ resistance in India due to unregulated prescription of these drugs. The study found a higher proportion of MDR/RR-TB cases with FQ-resistant genotypes, even among isolates with resistance to a single drug. The high FQ resistance rate found in the study is alarming for the National Tuberculosis Elimination Program and highlights the need for judicious use of fluoroquinolone drugs. This report also describes specific mutations associated with high FQ resistance in TB patients from India that could help develop a new algorithm for rapid diagnosis of drug-resistant TB, leading to better treatment outcomes. About one-third of FQ-resistant cases were classified as transmissible, indicating an urgent need for policymakers to address the higher FQ resistance rate in India among DR-TB patients. The results suggest that the diagnosis of FQ resistance should preferably be performed at the initial diagnosis stage to identify all resistance-promoting mutations and to ensure effective treatment and resistance control.