Tuberculosis biomarker pipeline: what will it take to come up with a rapid test?
Imagine examining 4,470 publications, manually extracting and curating data from 443 articles to identify 641 unique biomarkers, and lastly rating the quality of the pertinent data. This was the reality of our group’s efforts to catalyze translation of a biomarker into an urgently needed rapid test to diagnose tuberculosis (TB).
Our goal, through the generous support of the Stop TB Partnership’s, was to map the research and development (R&D) landscape of TB biomarker research and synthesize the current level of evidence of promising biomarkers for test development.
Activities in TB biomarkers research have generated a vast number of reports. Our group took a big breath and dived into the data to capture the existing biomarkers and evaluate the level and quality of evidence around them.
This herculean task was worth the effort, as TB continues to take more than 1 million lives each year while affecting millions of individuals in the world’s poorest countries1. It could be argued that effective diagnostic and treatment solutions exist if we look through the lens of high-resource settings. However, many patients in low- and middle-income countries currently remain undiagnosed because diagnostic tools are unaffordable or unavailable in areas where they first present for care. Tuberculosis cannot be tackled effectively in these countries without new tools; thus the World Health Organization (WHO) and TB community are calling for a simple, affordable non-sputum-based rapid test2 (similar to a pregnancy test).
Our work aims to identify the most promising biomarkers for the development of a point-of-care test for standalone detection or triage for active TB. A fingerstick blood-based TB test deployable at community health care settings, similar to an HIV rapid diagnostic test (see photo), would dramatically increase the number of people having access to TB testing and treatment. [Photo credit: John Rae]
The search for accurate biomarkers that can be used to develop such tools is the topmost priority for test developers or catalyzers of development (such as FIND or BMGF), but despite decades of TB biomarker research, significant investment, and hundreds of reports on biomarker candidates, only two biomarkers have translated so far into commercial tests3. What happened to all the other markers that have been identified and published? Our review highlighted that only a small number of well-designed studies exist that would inspire confidence in a marker that can be translated into a test. A large proportion of studies did not use a clinically relevant negative control population (i.e. patients presenting with TB symptoms in relevant epidemiological settings), which often substantially overestimates the performance of a marker, particularly of host origin. Also, few studies assessed diagnostic performance in light of WHO criteria, and even fewer studies were undertaken to validate previously discovered biomarkers. This explains, among other factors, why progress towards a clinically relevant, simple, non-sputum-based test has been slow. The TB community can do better than this and part of our paper outlines recommendations for future research.
Innovative tools and coordinated, collaborative international efforts are necessary to overcome the challenges in developing effective and accessible TB diagnostic products. Bm2Dx.org is designed as a dynamic systematic review to bring together a research community around the database and to ensure that TB biomarker research translates into collaboration, development of novel tests and health impact. [Photo credit: John Rae]
Certainly, the TB community could achieve more through collaboration. Discovery research studies are too often repeated unnecessarily using scarce resources. Promising biomarkers are not systematically nominated for follow-up confirmation work in non-biased clinical studies. This dismal patchwork of fragmented research would benefit from “a coordinated big science approach”, of which this systematic review could constitute the first building block. Working in that direction, FIND is now using the data from the work published in Nature Microbiology as the foundation for a web platform that aims to overcome the disconnects in the biomarker R&D pipeline. This platform, called(biomarkers to diagnostics), will be a one-stop-shop for TB biomarker research and will include standardized study reporting and grading frameworks, with the goal of facilitating collaborative and coordinated efforts to validate promising markers and develop novel tests. Bm2Dx.org will link users to numerous resources (e.g. or FIND’s
Authors: Romain Wyss, Emily MacLean, Tobias Broger, Claudia M. Denkinger
1. Global Tuberculosis Report 2018 (World Health Organization, 2018).
2. High-priority target product profiles for new tuberculosis diagnostics: report of a consensus meeting (World Health Organization, 2014).
3. Yerlikaya, S., Broger, T., Maclean, E., Pai, M. & Denkinger, C. M. A tuberculosis biomarker database: the key to novel TB diagnostics. International Journal of Infectious Diseases 56, 253–257 (2017).