December 12, 2023 | An international research consortium led by scientists from LMU University Hospital Munich has tested a new simple, rapid tuberculosis test that uses a blood sample from the fingertip. The candidate test, called the Cepheid Mycobacterium tuberculosis Host Response prototype cartridge (MTB-HR), works by measuring a three-gene transcriptomic signature. More specifically, it analyzes the expression of the GBP5 and DUSP3 genes, which are upregulated in tuberculosis, and the KLF2 gene, which is downregulated in this disease.
The researchers evaluated the performance of this test as part of a study that includes a large, prospective cohort of children from several low and middle-income countries. Findings appeared in The Lancet Infectious Diseases (DOI: 10.1016/S1473-3099(23)00491-7). According to the authors, this paper “presents the first estimates on diagnostic accuracy of the MTB-HR on a fingerstick blood sample within a large, well defined, and geographically diverse pediatric cohort.”
Furthermore, it “has the potential to provide an urgently needed new approach to improve tuberculosis diagnosis for children globally,” they wrote. Support for the study came from the European and Developing Countries Clinical Trials Partnership, Swedish International Development Cooperation Agency, UK Medical Research Council, Bundesministerium für Bildung und Forschung, German Center for Infection Research, and Beckman Coulter. In addition, Cepheid provided test kits and GeneXpert platforms at no cost.
There remains “a critical need” for new diagnostics and triage tools that are focused on this disease and geared toward children, “so it is great for a multi-center study to assess a new three-gene host signature for this group,” Devan Jaganath, a pediatric infectious diseases physician at the University of California, San Francisco, told Diagnostics World. This particular tool is attractive because of the sample type it relies on, added Jaganath, who was not involved with the research.
Jeff Fischer, president and co-founder of Longhorn Vaccines & Diagnostics, a company whose diagnostic tools have been used in disease areas including tuberculosis, called the study “very promising,” pointing out that it “further demonstrates that new techniques and sample types can be used” to enhance the diagnosis of tuberculosis in children.
Challenges of Diagnosing Tuberculosis in Pediatric Populations
Tuberculosis—a curable and preventable infectious disease caused by a form of bacteria—occurs all across the world and in all age groups, although more than 8 in 10 cases and fatalities occur in low- and middle-income countries, according to a World Health Organization (WHO) fact sheet. The disease was responsible for over one million deaths in 2022 alone and remains a significant cause of death in children under the age of five.
One of the reasons is that this disease can be especially challenging to diagnose in children. To date, the most commonly used tests have relied on the analysis of sputum, but these samples can be tricky to gather from children and infants. Non-specific symptoms and a low bacterial load can also hinder diagnosis.
Beyond that, most tuberculosis test samples must be sent to labs for analysis, which tacks added wait time onto the process. As a result, the disease is regularly misdiagnosed or diagnosed too late, especially in parts of the world with limited resources.
“To diagnose children with tuberculosis, one of the biggest challenges isn’t necessarily cost or supply, but getting ahead of testing,” noted Fischer. “In doing so, government and health systems can reduce the time to early interventional treatment, which is key to reducing the spread.”
Brittany Hilker, who lived in Haiti for over a decade, has worked with nurses and doctors who are faced with children suffering from tuberculosis on a daily basis. She explained that, in many cases, tuberculosis testing is controlled by the government. While this reliance on government health programs may be a good way to track the distribution of tests and related statistics, problems can arise when government clinics are understaffed and underfunded—and some children in need of help are inevitably missed.
Another sizable challenge is that many patients live in rural areas far from any hospital that has diagnostic tests and capabilities, noted Hilker, who went on to found a nonprofit organization called Partner Haiti upon her return to the United States several years ago. Nations with widespread tuberculosis may also face issues like a lack of health care facilities, closed clinics, transportation issues, or gang-controlled roads, among others, added Hilker, who was not involved with The Lancet Infectious Diseases study.
Study Tests Performance of New Tool
New testing options are needed to address these types of diagnostic challenges, according to Laura Olbrich, lead author of the Lancet paper and a medical scientist at the Division of Infectious Diseases and Tropical Medicine at LMU University Hospital Munich.
Although WHO has approved tuberculosis culture and molecular rapid diagnostic tests, she and her co-authors indicated these options have less than optimal sensitivity. Furthermore, collection methods of specimens can be invasive and can require skilled staff and infrastructure.
One of the main advantages of the MTB-HR tool, on the other hand, is that it is both quick and non-invasive. The blood sample can be easily taken from the fingertip, and the results are available in roughly an hour.
The new tool was tested as part of the RaPaed-TB study, a prospective diagnostic accuracy study led by Norbert Heinrich at LMU University Hospital Munich and carried out along with collaborators in India, South Africa, Mozambique, Tanzania, and Malawi.
The study included 975 children under 15 years old who were suspected of having tuberculosis. Participants were enrolled between early 2019 and the mid-point of 2021. In all of the children, the MTB-HR was utilized at baseline and at one month; it was then repeated at both three and six months in those receiving treatment for tuberculosis.
To explore the accuracy of the MTB-HR, the researchers studied the tuberculosis status of the participants using a standardized reference test based on sputum and bacterial cultures. “MTB-HR was reliably able to differentiate children with culture-confirmed tuberculosis from those unlikely to have tuberculosis, both in the overall cohort and in key subgroups of interest,” the authors noted, such as young infants, children with HIV, and children with severe malnutrition.
“The results were encouraging,” Olbrich said in a news release. “Compared to detection in culture, the test identified almost 60% of children with tuberculosis, with 90% specificity. This makes the test comparable with or better than all other tests that work with biomarkers.” Although bacterial culture is the reference because it produces the most stable results, those results can take weeks or months to obtain, she added, and this option is often unavailable where pediatric patients present.
She and her co-authors noted in the study that MTB-HR has the potential to be used as a treatment decision and monitoring tool because a positive result could point to a more severe form of the disease and offer insight into treatment timeframe. This could address what they call “an important gap” among the tests currently available to diagnose pediatric tuberculosis.
Limitations and Room for Improvement
However, Olbrich and colleagues also acknowledged several limitations that should be taken into consideration. “Because we performed a rigorous diagnostic study in a well-defined population of children investigated for tuberculosis disease, large proportions had confirmed and unconfirmed tuberculosis,” they noted. “This over-representation of true tuberculosis cases might overestimate MTB-HR performance in the target population of children presenting at primary health care.”
In some subgroups, conventional diagnosis can be especially tricky. This includes populations such as young infants, children with severe malnutrition or HIV, and children living in settings with limited resources. Notably, some of these subgroups included in the study were small, which limits “the interpretability of accuracy estimates,” the authors pointed out. In addition, to facilitate the assignment of clinical case definitions, some children were not included in the study—mainly because of an unclear clinical trajectory.
Yet another caveat is that the study dataset is made up of samples that were examined immediately after collection, as well as those analyzed after samples were stored. “We observed that there was a difference in median scores between the stored and the fresh samples, probably due to RNA degradation in biobanked samples,” they explained.
Messenger RNA (mRNA) testing is expected to play a crucial role in monitoring tuberculosis treatment, Longhorn’s Fischer told Diagnostics World, but the use of mRNA can be limited because it degrades quickly. For this reason, he explained that the use of mRNA for detecting tuberculosis in pediatric populations (using the protocol described in the Lancet study) would likely be limited to hospitals or other health care facilities where samples could be tested shortly after blood is drawn from a patient.
“Further optimization and the possible use of nucleic acid stabilization technologies may improve sensitivity and allow for widespread testing using this technique,” he added.
The authors of The Lancet Infectious Diseases study “found that it may be useful to support the diagnosis of children when combined with molecular testing of respiratory samples,” explained UCSF’s Jaganath. Further validation is needed, but this highlights the notion that childhood tuberculosis diagnosis “may require a combination of tools with clinical assessment to improve case detection.”
He also pointed out that the accuracy of the MTB-HR was lower when including children with a clinical diagnosis of tuberculosis, which makes up a large proportion of children with this disease—especially young children.
Because children experience a different immune response to tuberculosis than adults do, “host biomarkers specific for this group may be needed to improve performance,” added Jaganath, whose research program is focused on improving the care of children with tuberculosis by pursuing new biomarkers and tools for both diagnosis and treatment—and evaluating their accuracy and implementation via clinical research.
Considering this is the first study on the use of this test in children, the study co-authors indicated that additional research is needed to replicate their results. And since the reference signature of the test was arrived at using adult samples, they believe its accuracy can be further improved after altering the calculation of the signature for children.
“The use of an age-adjusted or child-specific cutoff needs to be explored, which should be prospectively tested, as well as the optimal placement within the health-care cascade, including within recently recommended treatment decision algorithms published by WHO,” they noted.
Because trained lab staff members used the MTB-HR (as opposed to routine clinical staff), implementation studies are needed to evaluate test performance at different levels of care and to study any differences that may emerge when healthcare workers use the test.
**
Paul Nicolaus is a freelance writer specializing in science, nature, and health. Learn more at www.nicolauswriting.com.