Improvement in the Diagnosis of Tuberculosis Combining Mycobacterium Tuberculosis Immunodominant Peptides and Serum Host Biomarkers

Abstract

Pulmonary tuberculosis (PTB) is a public health problem with 10.4 million new cases reported in 2017 (1). According to the World Health Organization (WHO), accurate diagnostic tests based in serum biomarkers to detect new cases of tuberculosis are necessary. To evaluate antibodies against Mycobacterium. tuberculosis (Mtb) peptides (Ab-Mtb) and three soluble host biomarkers by ELISA serial multiple test in sera from non-infected controls (NIC, n=31), latent tuberculosis (LTB, n=37) and PTB (n=28) patients in a diagnosis tuberculosis assay. Levels of four Ab-Mtb peptides derived from Mtb and three host response molecules in serum from NIC, LTB and PTB were evaluated by ELISA as tuberculosis biomarkers. Multiple comparisons tests, determination of diagnostic values and ROC curves were performed. Serial and parallel multiple tests were performed with the biomarkers with the highest discriminatory capacity to improve diagnostic values of the test. We found significant differences between biomarkers levels in PTB comparing LTB and NIC to all candidate biomarkers; peptides P12033, P12037, and serum biomarkers such as sCD14 and chemokine CXCL9 showed the best sensitivity and specificity, the highest discriminatory power, and the best area under the curve (AUC) individually. In serial multiple tests, P12037 and sCD14 together have 92% of sensitivity and 91% of specificity, with positive and negative likelihood ratios greater than10. Ab-Mtb peptide P12037 and sCD14 could be applied in a diagnostic test for suspected PTB to improve accuracy and time to diagnosis and could be implemented in a POCT device which can be affordable.