Plasma Metabolomics in the Diagnosis of Pulmonary Tuberculosis and Early Identification of Multidrug Resistance Open Access

Collins, Jeffrey (2017)

Permanent URL: https://etd.library.emory.edu/concern/etds/gt54kn797?locale=en
Published

Abstract

Introduction: The lack of peripheral blood biomarkers for active tuberculosis (TB) disease is a major obstacle to development of point of care tests for diagnosis and treatment monitoring. Plasma high resolution metabolomics (HRM) is an innovative method to discover and evaluate peripheral blood biomarkers such as those derived from the cell envelope of Mycobacterium tuberculosis (Mtb).

Methods: We compared plasma HRM from 17 patients with active TB disease to 16 household contacts without active TB. We used a suspect screening approach to identify metabolites matching known lipids from the Mtb cell envelope based on retention time and accurate mass matches. Mtb lipid suspects were tested for association with disease status using the linear model for microarray data. To identify biomarkers of treatment response, we performed plasma HRM on 61 patients at the time of diagnosis and after 4 weeks and 8 weeks of anti-TB therapy. We included 17 patients later found to have multidrug resistant (MDR) TB and 44 patients with drug susceptible TB. A mixed effects model was used to evaluate the effect of MDR status on metabolite intensity and the change in intensity over time.

Results: Plasma HRM identified four Mtb lipid suspects significantly increased in active TB patients and undetectable in most household contacts: phosphatidylglycerol (PG), monoacylglycerophosphoinositol (Lyso-PI), monoacylated diacylglycerophosphoinositolmonomannoside (Ac1PIM1), and monoacylglycerophosphoinositolmonomannoside (Lyso-PIM1) (p<0.001 for all). These metabolites provided excellent classification accuracy for active TB disease (AUC 0.95). During treatment, the intensity of Lyso-PIM1 was significantly lower in patients with MDR TB compared to those with drug susceptible TB disease (p=0.01). Of the other 29 Mtb lipid suspects significantly different between MDR TB and drug susceptible TB patients receiving treatment (raw p≤0.05), 27 were also lower in MDR TB patients.

Conclusions: Lyso-PIM1, Lyso-PI, PG and Ac1PIM1 provided excellent classification accuracy for the diagnosis of active TB disease. Additionally, the intensity of Lyso-PIM1 was significantly lower in MDR TB patients during treatment. If confirmed in large studies, further development of these biomarkers could lead to a rapid point of care test for diagnosis and treatment response in active TB disease.

Table of Contents


Table of Contents

Introduction……………………………………………………………………………1-2

Background…………………………………………………………………………….3-7

Methods………………………………………………………………………………...8-13

Results……………………………………………………………………….…….....14-17

Discussion…………………………………………………………………………….18-25

References…………………………………………………………………………...26-28

Tables.…………….…………………………………………………………….......29-32

Figures……………………...……………………………………………………..….33-38

List of Tables

Table 1…..………………….…………………………………….………………………29

Table 2……………………………………..……………………………………………..30

Table 3…………………………………………..……………………………………..…31

Table 4…………………………………….…….…………….…………………........32

List of Figures

Figure 1…..……………………………………………………….………………………33

Figure 2…………………………………………………………………………………...34

Figure 3………………………………………………………………………………...…35

Figure 4……….………………………………………………………………….........36

Figure 5…………………………………………………………………………………...37

Figure 6………………………………………………………………………………...…38

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
School
Department
Degree
Submission
Language
  • English
Research Field
Keyword
Committee Chair / Thesis Advisor
Committee Members
Last modified

Primary PDF

Supplemental Files