A Novel, Rapid, Accurate and Non-Invasive Mean to Diagnose Tuberculosis

diagnose tuberculosis

Tuberculosis is an infectious disease caused by the bacteria Mycobacterium Tuberculosis for which reliable and rapid diagnostic methods are crucially needed to improve treatment efficacy. Fortunately, a group of investigators recently published a very promising approach based on mass spectrometry analysis of blood specimens to diagnose tuberculosis.


Tuberculosis (TB) is a life-threatening infectious disease caused by the bacteria Mycobacterium tuberculosis (MT). Clinically, the disease is characterized by a chronic and severe cough with bloody sputum sometimes expelled, fever and weight loss among others. Epidemiologically, TB ranks among the top 10 deadliest diseases in the world and its prevalence is particularly alarming in the developing world, where 95% of all cases occur.

Although current antibiotic treatments for TB show a relatively good efficacy, novel, sensitive, reliable, and rapid diagnostic methods are much needed. Indeed, current gold-standard medical practices involve bacterial isolation and culture, which is a long, non-quantitative and error-prone process. These limitations can have major implications for TB patients, especially for two subgroups: 1) HIV-positive patients for which disease progression is remarkably fast and for whom sputum samples are often of limited usage and 2) patients infected with Extra-Pulmonary Tuberculosis (EPTB) for which accurate diagnosis often requires more invasive procedures.

To circumvent the limitations associated with these procedures, a team led by Dr. Ye Hu in Houston recently developed and published new research in PNAS. They used a mass spectrometry-based approach to quantitatively assess TB progression using a simple blood specimen. The method relies on the detection of two MT-specific antigens secreted at all stages of infection, thereby enabling diagnosis of almost all non-latent cases. In this procedure, the first step consists of conjugating silicon nanoparticles to antibodies recognizing these antigens. Second, the sample is deposited on the matrix-antibody conjugate to allow affinity-binding of these antigens and, finally, MALDI-TOF (a mass spectrometry technique enabling to detect the antigen) is performed. When combined with internal standards for which the antigen concentration is known, this method allows determining the exact concentration of the antigen and, thus, the extent to which a patient is infected.

In many ways, this method is very promising. First, the authors convincingly demonstrate the quantitative aspect of their technique with a very strong correlation (R2=0.99) observed between the signal and the actual concentration of the antigen. Second, with a detection threshold nearly 100 times lower than the average concentration of the antigen in any given patient, the method is very sensitive. Finally, as a proof-of-principle of the clinical reliability of their method, the authors tested TB patient’s samples, many of which belonged to subgroups such as latent TB and HIV-positive samples. Overall, they were able to accurately diagnose 88% of all detectable TB cases, greatly outperforming culture-based methods (~60%). Even more impressively, as opposed to culture-based methods, their method proved to be just as accurate for EPTB and HIV-positive patients and does not require any invasive procedure. Therefore, in addition to being faster, non-invasive and more reliable, their method also shows a remarkable efficacy for patient cohorts in which current methods suffer the most from the aforementioned limitations.

Although promising, this method has some limitations that are somewhat minor compared to those of current diagnostic methods. First, a non-negligible source of false-positive is patients infected with other closely related bacterial species, collectively known as nontuberculous mycobacteria (NTM), among which ~10% tested positive for MT. Second, only one of the two MT antigens measured, CFP-10, shows detectable levels in most TB patients, suggesting the other one, ESAT-6, might be of limited use. Finally, although the authors show that samples from patients undergoing antibiotic treatment almost all show a decreased signal over time, this data is not corroborated by an orthogonal method to monitor disease progression, raising questions about the real disease progression in these patients. In summary, despite these limitations, their method still holds the promise to greatly improve the accuracy and speed of TB diagnosis. In addition, the application of such process in a clinical setting is not inconceivable in the near future since most modern hospitals are now equipped with mass spectrometry equipment. However, considering the financial resources required, this might pose a major challenge in the developing world where almost all TB cases are diagnosed.


Written By: Samuel Rochette M.Sc.