Background <p>Pleural tuberculosis (pTB) poses a diagnostic challenge due to its non-specific clinical symptoms and paucibacillary form of disease. Aligning with the WHO's End TB goals, we developed antigen-detection assays for pTB on both pleural fluid (PF) and paired urine samples.</p> Methods <p>A panel of 4 <i>Mycobacterium tuberculosis</i> antigens, namely Ag85A, CFP-10, BfrB, and ESAT-6, were evaluated in PF and corresponding urine samples. For assay development, we enrolled 131 patients with pleural effusion who presented with clinical symptoms such as fever, chest pain, breathlessness, and other related manifestations. Patients were classified using a composite reference standard as ‘Definite’ pTB (n = 16; Xpert MTB/RIF and/or culture positive), ‘Probable’ pTB (n = 46; with&#xa0;favourable response to anti-tubercular therapy), ‘Possible’ pTB (n = 25; radiological findings suggestive of pTB) and ‘non-TB’ groups (n = 44; patients with a confirmed alternative diagnosis). The ‘Probable’ and ‘Possible’ pTB groups showed exudative effusion by Light’s criteria, and in addition, exhibited either elevated adenosine deaminase levels (&gt; 40&#xa0;IU/L) and/or PF protein concentrations exceeding 2.5&#xa0;g/dL. For urine-based assays, urine of all 131 patients was used, whereas for PF-based ELISAs, n = 117 corresponding PF samples were used [categorised as ‘Definite’ pTB (n = 11), ‘Probable’ pTB (n = 43), ‘Possible’ pTB (n = 23) and ‘non-TB’ (n = 40)]. ROC curves were generated, and cut-off values were selected using ELISA results of ‘Definite’ pTB versus ‘non-TB’ groups.</p> Results <p>In PF-based ELISAs, Ag85A and BfrB antigens showed a sensitivity of 63.6% (95% CI:30.8–89.0) and 54.5% (95%CI:23.4–83.2), with a specificity of 97.5% (95%CI:86.8–93.3) and 90.0% (95%CI:76.3%-97.2%), respectively, in the ‘Definite’ pTB group. In urine-based ELISAs, BfrB antigen had 75% (95%CI:47.6–92.7) sensitivity and 93.2% (95%CI:81.3–98.5) specificity, while Ag85A had 68.7% (95%CI:41.3–88.9) sensitivity with 93.2% (95%CI:81.3–98.5) specificity, in the ‘Definite’ pTB group.</p> Conclusion <p>Ag85A and BfrB emerged as promising antigens for the diagnosis of pTB in both PF and urine samples. The results for urine and PF-based assays did not differ significantly (<i>p</i> &gt; 0.05), indicating the potential use of urine-based ELISA as a non-invasive surrogate test for pTB diagnosis. However, further studies on a larger cohort are needed to validate the potential utility of urine-based assays as an alternative diagnostic test for pTB.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Development of immunoassays targeting Mycobacterium tuberculosis antigens Ag85A, CFP-10, BfrB, and ESAT-6 in paired pleural fluid and urine samples for the diagnosis of pleural tuberculosis

  • Adarsh Mishra,
  • Manisha Dass,
  • Ritu Singhal,
  • Puneet Arora,
  • Rishika Bukhariya,
  • Pratibha Sharma,
  • Neha Jindal,
  • Anjali Gola,
  • Ravindra Dewan,
  • Manpreet Bhalla,
  • Sagarika Haldar

摘要

Background

Pleural tuberculosis (pTB) poses a diagnostic challenge due to its non-specific clinical symptoms and paucibacillary form of disease. Aligning with the WHO's End TB goals, we developed antigen-detection assays for pTB on both pleural fluid (PF) and paired urine samples.

Methods

A panel of 4 Mycobacterium tuberculosis antigens, namely Ag85A, CFP-10, BfrB, and ESAT-6, were evaluated in PF and corresponding urine samples. For assay development, we enrolled 131 patients with pleural effusion who presented with clinical symptoms such as fever, chest pain, breathlessness, and other related manifestations. Patients were classified using a composite reference standard as ‘Definite’ pTB (n = 16; Xpert MTB/RIF and/or culture positive), ‘Probable’ pTB (n = 46; with favourable response to anti-tubercular therapy), ‘Possible’ pTB (n = 25; radiological findings suggestive of pTB) and ‘non-TB’ groups (n = 44; patients with a confirmed alternative diagnosis). The ‘Probable’ and ‘Possible’ pTB groups showed exudative effusion by Light’s criteria, and in addition, exhibited either elevated adenosine deaminase levels (> 40 IU/L) and/or PF protein concentrations exceeding 2.5 g/dL. For urine-based assays, urine of all 131 patients was used, whereas for PF-based ELISAs, n = 117 corresponding PF samples were used [categorised as ‘Definite’ pTB (n = 11), ‘Probable’ pTB (n = 43), ‘Possible’ pTB (n = 23) and ‘non-TB’ (n = 40)]. ROC curves were generated, and cut-off values were selected using ELISA results of ‘Definite’ pTB versus ‘non-TB’ groups.

Results

In PF-based ELISAs, Ag85A and BfrB antigens showed a sensitivity of 63.6% (95% CI:30.8–89.0) and 54.5% (95%CI:23.4–83.2), with a specificity of 97.5% (95%CI:86.8–93.3) and 90.0% (95%CI:76.3%-97.2%), respectively, in the ‘Definite’ pTB group. In urine-based ELISAs, BfrB antigen had 75% (95%CI:47.6–92.7) sensitivity and 93.2% (95%CI:81.3–98.5) specificity, while Ag85A had 68.7% (95%CI:41.3–88.9) sensitivity with 93.2% (95%CI:81.3–98.5) specificity, in the ‘Definite’ pTB group.

Conclusion

Ag85A and BfrB emerged as promising antigens for the diagnosis of pTB in both PF and urine samples. The results for urine and PF-based assays did not differ significantly (p > 0.05), indicating the potential use of urine-based ELISA as a non-invasive surrogate test for pTB diagnosis. However, further studies on a larger cohort are needed to validate the potential utility of urine-based assays as an alternative diagnostic test for pTB.