Exploring novel assay formats for Indoleamine 2,3-dioxygenase as a Tuberculosis biomarker

Heena Ranchod*1, 2, Heather Hong*1, Neil Martinson3 and Melinda Suchard1, 2 *Co-first authors
1 Centre for Vaccines and Immunology, National Institute for Communicable Disease, Johannesburg, South Africa
2 Department of Chemical Pathology, School of Pathology, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa
3 Perinatal Health Research Unity, DST/NRF Centre of Excellence for Biomedical TB Research, and Soweto Motlosana Collaborating Centre for HIV/AIDS and TB, University of the Witwatersrand, Johannesburg, South Africa

Abstract
Current methods available for diagnosing Tuberculosis (TB), a disease caused by Mycobacterium tuberculosis, are limited in their sensitivity, specificity, ease of obtaining test specimens as well as the time taken to obtain results. Indoleamine 2,3-dioxygenase (IDO) is the rate-limiting enzyme in the conversion of tryptophan (Tryp) to kynurenines (Kyn), that have immune suppressive functions. IDO can be measured by the ratio of Kyn to Tryp concentrations by ELISA or Real Time quantitative PCR (qPCR) from peripheral blood. The Kyn/Tryp ratio has been previously proposed as a TB biomarker. Here we compared the Kyn/Tryp ratio measured by ELISA, and IDO gene expression measured by qPCR in comparison to Beta-actin (an internal control gene), in patients with active TB versus healthy household contacts from the South Africa - Hopkins TB (SoHoT) collaboration study. The SoHOT study had used culture and GeneXpert Ultra (Cepheid) for TB diagnosis.
In HIV-uninfected individuals, there was a significantly higher Kyn/Try ratio in patients with active TB (median = 0.115, n=22) compared to the healthy control group (median = 0.08, n=128, p = 0.0011). Thus, a Kyn/Tryp ratio at a cut-off of 0.10, could be used to diagnose TB in HIV negative individuals, yielding sensitivity of 72.7% and specificity of 65.6%. Similarly, a significantly higher ΔCt value was obtained in TB infected patients (median = 11.58, n=9) compared to the healthy control group (median = 8.58, n=66, p < 0.0001).
In contrast, in HIV-infected individuals, no significant difference in the Kyn/Tryp ratio or ΔCt values could be found between TB infected and healthy controls when considering all Xpert Ultra and culture positive results as the gold standard.
Interestingly, limiting analysis to TB culture-confirmed patients, in the HIV-infected group the Kyn/Tryp ratio was significantly higher in patients with confirmed TB (median = 0.22, n=18) compared with those without TB (median = 0.12, n=25, p = 0.0324). Using TB culture as the gold standard improved performance of the Kyn/Tryp ratio (at a cut off of 0.11) in HIV infected individuals, however, the sensitivity was only 72.2% and specificity of 40% for TB diagnosis.
Thus, measuring IDO activity by ELISA or qPCR may indicate mycobacterial load and severity of active tuberculosis, and warrants deeper exploration particularly for treatment monitoring.

Heena Ranchod PhD

National Institute for Communicable Diseases/University of the Witwatersrand

Appearances