利用全基因定序預測結核病之抗藥性

Abstract

結核病是其中一個導致死亡的主要傳染病之一，且抗藥性菌株的傳播使問題更惡化，改善抗藥性結核病的診斷將能成功幫助全球結核病的預防與治療。此研究以培養為主的藥敏性測試為基準來評估全基因定序診斷到的表現。我們分析來自台灣高雄市於2019年1月至2021年7月之結核菌培養陽性通報個案中超過85％，將近2000隻TB菌株，發現全基因定序能達到接近90%的敏感性及高於95%的特異性及診斷正確率。我們也使用全基因定序來預測多種結核病藥物之盛行率，發現與傳統藥敏性測試之表現型結果有很高的一致性。整體而言，研究結果提供使用全基因定序診斷抗藥性結核病以及未來如何將其應用於臨床和公共衛生中的重要支持。

Tuberculosis (TB) is one of the largest contributors to mortality among all communicable diseases, and the spread of drug-resistant strains will only exacerbate this problem. Diagnostic methods in detecting drug-resistant tuberculosis (DR-TB) need to improve if global prevention and treatment efforts are to be successful. In this study, we evaluate the performance of whole genome sequencing (WGS) in detecting DR-TB compared to culture-based, drug susceptibility testing (DST). To do this, we prospectively analyzed nearly 2,000 TB isolates from Kaohsiung, Taiwan, between January 2019 to July 2021 and achieved a coverage rate of 85% for all TB cases in the city. We found sensitivity as high as 90% for certain anti-TB drugs and greater than 95% for specificity and diagnostic accuracy. Additionally, we used WGS to predict the resistance prevalence for various drugs and found great agreement with what was determined by phenotypic DST. Overall, our results provide greater support on the use of WGS in diagnosing DR-TB, and how it may be applied in both clinical and public health settings.