Host-pathogen Interactions: Honing in on enteric fever
Enteric fever, also known as typhoid, is a disease that affects about 22 million people and causes about 200,000 deaths every year, according to conservative estimates (Buckle et al., 2012). Enteric fever is spread by bacteria belonging to the Salmonella genus, with two sub-species—Salmonella Typhi and Salmonella Paratyphi A—being responsible for most cases of the disease. And although the number of cases of enteric fever has fallen significantly over recent decades, there is a clear need for a diagnostic test for Salmonella that is rapid, affordable and accurate. Moreover, it is important to be able to distinguish between enteric fever caused by Salmonella Typhi and enteric fever caused by Salmonella Paratyphi A in order to ensure that the correct drugs are prescribed and to combat the development of antibiotic resistance.
The ideal diagnostic test offers high levels of sensitivity (that is, it is able to detect a high proportion of the people tested who have a particular disease), and also high levels of specificity (that is, it returns a ‘No’ for a high proportion of people without the disease). The current gold standard diagnostic tool for enteric fever is blood culture, but this has a low specificity and it is also challenging to perform consistently in regions with limited resources. As a consequence, misdiagnosis and mistreatment of enteric fever is common, which leads to poor clinical outcomes and the potential spread of antibiotic resistance. The development of nucleic acid and antigen-detection tests to diagnose enteric fever has faced similar challenges, with lower than expected sensitivity and specificity. One explanation for this could be that the levels of Salmonella in the blood tend to be very low at all stages of the disease (Darton et al., 2014).
Now, in eLife, Elin Näsström and co-workers at Umeå University, Oxford University (including clinical research units in Ho Chi Minh City and Kathmandu), and the London School of Hygiene and Tropical Medicine apply a promising new approach to this challenge (Näsström et al., 2014). Instead of trying to detect Salmonella in the blood during infection, they used a technique called metabolomics. The basic idea of this approach is that infection leads to metabolic changes, such that a person with enteric fever (or any infection) could have a profile of metabolites in their blood that is different to the metabolite profile of a healthy person. The challenge, therefore, is to identify a ‘metabolic fingerprint’ that can be used to detect enteric fever with high levels of sensitivity and specificity.
The application of metabolomics is relatively new in infectious diseases research compared to the application of genomics and proteomics. Despite this, screening the metabolome in blood plasma has identified useful prognostic profiles of several diseases, including sepsis (Langley et al., 2013). One of the major benefits of this technique is that it utilizes a pattern of biomarkers (that is, the various metabolites), as opposed to relying on just one host biomarker, as has been the focus of previous approaches.
Näsström et al. collected blood samples from 75 participants from South East Asia: 25 had enteric fever caused by S. Typhi, 25 had enteric fever caused by S. Paratyphi A, and 25 did not have enteric fever. Using a combination of gas chromatography and mass spectrometry, Näsström et al. identified 695 distinct peaks that were associated with different metabolites: from these they selected six peaks that had significantly different heights in the three groups of patients. This meant that they were able to tell if the patient had S. Typhi, S. Paratyphi A, or neither. That this mass spectrometric analysis was able to distinguish two Salmonella groups that share many similarities is remarkable. Moreover, in addition to its diagnostic potential, this new approach might also provide insights into the antigenic and physiological differences between the two strains.
There are, however, several issues that need to be addressed before this work will reach the clinic. The sample size in the study was relatively small (although it was randomly collected from a clinical trial repository). Näsström et al. also acknowledge that they did not perform any external validation, so it is not clear that the six biomarkers they identified will be relevant in other populations. It will be necessary to address these two points in future trials. Lastly, while negative controls were included, an expanded set of all-cause fever samples (that is, samples in which the enteric fever was caused by other organisms) will be needed to determine how well this panel of six biomarkers will perform in a real clinical setting.
Nevertheless, these findings represent an important step in the right direction. Näsström et al. acknowledge that mass spectrometers are too complex and expensive for routine use, but other more affordable technologies may be able to detect the six-metabolite fingerprint that they have identified. It might also be possible to use a similar approach to diagnose other bacterial infections.
Finally, given that antibiotic resistance is now a major public health challenge, and that we might be entering a ‘post-antibiotic era’ (Woolhouse and Farrar, 2014), these findings could represent a substantial technological advance and an important first step to addressing the problem of diagnosing and properly treating enteric fever and other blood-borne infections.
References
-
Typhoid epidemiology, diagnostics and the human challenge modelCurrent Opinion in Gastroenterology 30:7–17.https://doi.org/10.1097/MOG.0000000000000021
-
An integrated clinico-metabolomic model improves prediction of death in sepsisScience Translational Medicine 5:195ra95.https://doi.org/10.1126/scitranslmed.3005893
Article and author information
Author details
Publication history
Copyright
© 2014, McKinnon and Abdool Karim
This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.
Metrics
-
- 659
- views
-
- 22
- downloads
-
- 4
- citations
Views, downloads and citations are aggregated across all versions of this paper published by eLife.
Download links
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Further reading
-
- Medicine
- Microbiology and Infectious Disease
- Epidemiology and Global Health
- Immunology and Inflammation
eLife has published articles on a wide range of infectious diseases, including COVID-19, influenza, tuberculosis, HIV/AIDS, malaria and typhoid fever.
-
- Epidemiology and Global Health
- Genetics and Genomics
Alzheimer’s disease (AD) is a complex degenerative disease of the central nervous system, and elucidating its pathogenesis remains challenging. In this study, we used the inverse-variance weighted (IVW) model as the major analysis method to perform hypothesis-free Mendelian randomization (MR) analysis on the data from MRC IEU OpenGWAS (18,097 exposure traits and 16 AD outcome traits), and conducted sensitivity analysis with six models, to assess the robustness of the IVW results, to identify various classes of risk or protective factors for AD, early-onset AD, and late-onset AD. We generated 400,274 data entries in total, among which the major analysis method of the IVW model consists of 73,129 records with 4840 exposure traits, which fall into 10 categories: Disease, Medical laboratory science, Imaging, Anthropometric, Treatment, Molecular trait, Gut microbiota, Past history, Family history, and Lifestyle trait. More importantly, a freely accessed online platform called MRAD (https://gwasmrad.com/mrad/) has been developed using the Shiny package with MR analysis results. Additionally, novel potential AD therapeutic targets (CD33, TBCA, VPS29, GNAI3, PSME1) are identified, among which CD33 was positively associated with the main outcome traits of AD, as well as with both EOAD and LOAD. TBCA and VPS29 were negatively associated with the main outcome traits of AD, as well as with both EOAD and LOAD. GNAI3 and PSME1 were negatively associated with the main outcome traits of AD, as well as with LOAD, but had no significant causal association with EOAD. The findings of our research advance our understanding of the etiology of AD.