Zhongkang Ji,1,* Sheng Bi,1,* Bin Lu,1,2 Lin Zheng,1 Xiuyuan Jin,1 Shujuan Huang,1 Liangxiu Jiang,1 Yuping Wang,1 Cheng Ding,1 Kaijin Xu1 

1State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, National Medical Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, People’s Republic of China; 2Department of Infectious Diseases, Affiliated Dongyang Hospital of Wenzhou Medical University, Dongyang, Zhejiang, People’s Republic of China

*These authors contributed equally to this work

Correspondence: Kaijin Xu; Cheng Ding, State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, National Medical Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, People’s Republic of China, Tel +86-13750870030, Email zdyxyxkj@zju.edu.cn; dingchengzju@zju.edu.cn

Objective: To assess the diagnostic performance of the blood-based Cepheid 3-gene Host Response test (MTB-HR), urine-based Lipoarabinomannan (LAM), and a combination of MTB-HR and LAM (MTB-HR & LAM) for detecting active tuberculosis (ATB).
Methods: All participants were recruited from the First Affiliated Hospital, Zhejiang University School of Medicine, between June 8, 2023 and September 13, 2023. Subsequently, the participants were classified into the ATB group or non-active tuberculosis (non-ATB) group based on microbiological evidence. MTB-HR and LAM tests were performed using fingerstick blood and urine samples from each participant, respectively. The diagnostic performance of the tests was evaluated based on the sensitivity, specificity, Youden index, and Kappa value. Pairwise comparisons of the areas under the receiver operating characteristic curves (AUROCs) between different tests were conducted using nonparametric methods.
Results: A total of 297 participants were included. The MTB-HR test demonstrated diagnostic efficacy with a sensitivity of 77.37% (95% CI: 70.37– 84.38) and a specificity of 85.63% (95% CI: 80.19– 91.06). The LAM test demonstrated a high specificity of 97.50% (95% CI: 95.08– 99.92), albeit with a lower sensitivity of 54.74% (95% CI: 46.41– 63.082). The sensitivity and specificity of the MTB-HR & LAM were 83.21% (95% CI: 76.95– 89.47) and 83.13% (95% CI: 77.32– 88.93), respectively. Only MTB-HR & LAM exhibited higher values of area under the receiver operating characteristic curve than the LAM test (MTB-HR & LAM vs LAM: 0.83 vs 0.76, P=0.0031).
Conclusion: In this study, although both non-sputum-based triage MTB-HR and LAM do not meet the WHO diagnostic target currently, they show possible values for triage and diagnosis in ATB. Compared to single MTB-HR or LAM test, the combined MTB-HR & LAM does not demonstrate advantages.

Keywords: active tuberculosis, diagnostic, three-gene host response, Lipoarabinomannan, AUROC