Can identifying recent Mycobacterium tuberculosis infection support faster tuberculosis vaccine trials?

Tuberculosis remains one of the world's most persistent infectious diseases, caused by a bacterium that infects a vast portion of the human population yet rarely makes people sick. In most cases, the immune system keeps the bacteria in a dormant state, and the infected person carries the germ for years or even a lifetime without developing active disease. Only a small fraction of those infected will eventually fall ill, and the risk of becoming sick is highest in the first few years after the initial infection. This biological reality creates a massive hurdle for scientists trying to create a new vaccine. To prove a vaccine works, researchers must wait to see if the people who receive it get sick compared to those who do not. Because so few infected people develop the disease, trials must enroll thousands of participants and run for many years just to gather enough cases to draw a conclusion. This slow, expensive process has stalled the development of better vaccines for decades.

A team of researchers at the University of Colorado Boulder, the University of Cambridge, and Emory University has proposed a way to speed this process up by changing who they ask to join the trial. Currently, most vaccine studies use a blood test called an interferon-gamma release assay to screen participants. This test is excellent at detecting if someone has ever been infected with the tuberculosis bacterium, but it cannot tell when the infection happened. A person who was infected twenty years ago and a person infected last week will both test positive. Since the risk of developing active disease is much higher for those infected recently, including people with old, remote infections dilutes the pool of participants with the highest risk. The researchers asked a simple question: what would happen if they could identify and enroll only those with a recent infection?

To answer this, the team built a statistical model to simulate how vaccine trials would play out in three high-burden countries: South Africa, Kenya, and India. They compared the standard approach, which enrolls anyone who tests positive for any past infection, against a new strategy that would use a hypothetical test capable of distinguishing recent infections. This new test would only remain positive for about two years after a person is infected, effectively filtering out those with old infections. The researchers also looked at other methods, such as enrolling only young people or only the family members of people who are currently sick, to see how these strategies compared.

The simulations revealed that using a test for recent infection would dramatically shrink the size of the trial needed to get a clear answer. In their model, finding enough disease cases to evaluate a vaccine required four to six times fewer participants if the trial enrolled only those with recent infections, compared to the current method of enrolling anyone with a positive test. This reduction is significant because it means fewer people need to be followed over time, which saves years of research and vast amounts of money. The researchers noted that this approach was even more efficient than enrolling household contacts of sick patients, a strategy often used to find high-risk individuals.

However, this efficiency comes with a trade-off. Because the new test only identifies a small window of recent infections, researchers would have to screen a much larger number of people to find the few who qualify for the trial. The model estimated that for every single disease case needed to prove the vaccine works, scientists would need to test roughly 600 people using the new recency test, compared to only about 130 people using the standard test. Despite this increase in screening, the researchers calculated that the overall cost would likely still be lower. They reasoned that the expense of running a clinical trial for a participant is so high that it outweighs the cost of the extra testing, provided the new test does not cost more than five to ten times as much as the standard test.

The study does not claim that this method is ready to be used immediately. The researchers emphasize that the test for recent infection they modeled is still in development, and its actual performance characteristics are not yet fully known. They assumed in their model that the test would be perfect, but real-world tests might miss some infections or give false alarms. Furthermore, there is still uncertainty about whether a vaccine would work equally well on people with very recent infections compared to those with older ones. If the vaccine works differently depending on how long a person has been infected, a trial focused only on the recently infected might give results that are hard to interpret for the broader population.

The authors conclude that while diagnostics for recent infection hold great promise for making vaccine trials faster and more efficient, more work is needed to validate these tools. They outline the specific qualities these tests must have, such as how long they stay positive and how accurately they detect the infection, before they can be trusted in a clinical setting. By identifying the right group of people to study, science may finally be able to overcome the natural history of tuberculosis and bring a new vaccine to the world sooner.