Are Current Critical Concentrations for Mycobacterium tuberculosis Drug Susceptibility Testing in Egg-Based Media Adequate? Insights from Phenotypic and Genotypic Analyses

Tuberculosis is an ancient enemy that still claims millions of lives each year, primarily because the bacteria that cause it can learn to ignore the medicines designed to kill them. When a patient is infected, doctors must determine which drugs will work and which will fail, a process known as drug susceptibility testing. For decades, the most reliable way to do this has been to grow the bacteria in a laboratory and watch how they react to different antibiotics. However, this traditional method is slow, often taking weeks to yield an answer, and it requires specialized equipment and expensive ingredients that are not always available in the places where the disease is most common. In recent years, scientists have turned to faster genetic tests that look for specific changes in the bacteria's DNA, hoping to predict resistance without waiting for the bugs to grow. Yet, these genetic tests are not perfect; they can miss subtle forms of resistance or flag harmless genetic variations as dangerous threats. The core question facing researchers is whether the older, slower methods can be improved to remain the gold standard, or if they are simply too flawed to be trusted in the modern fight against tuberculosis.

A team of researchers in Brazil set out to answer this question by re-examining a classic laboratory technique that uses egg-based media to grow the bacteria. This medium, known as Ogawa-Kudoh, has been used for decades, particularly in Japan, but its ability to accurately measure resistance to modern drugs had not been thoroughly tested against the newest international standards. The scientists gathered sixty-five different strains of the tuberculosis bacteria, some of which were known to be resistant to various drugs and others that were susceptible. They then subjected these strains to a rigorous comparison, testing them side-by-side using the egg-based method, a highly precise liquid-based method considered the current reference standard, and several genetic tests that scan for known mutations. The goal was to see if the egg-based approach could reliably distinguish between bacteria that were truly resistant and those that were still vulnerable, and to determine if the genetic tests were missing anything that the growing bacteria could reveal.

The results of this comparison were strikingly clear. When the researchers measured the exact amount of drug needed to stop the bacteria from growing, the egg-based method and the liquid reference standard agreed with each other almost perfectly. Out of hundreds of measurements taken, only two instances showed a difference between the two growing methods. This level of agreement suggests that the egg-based medium is just as capable as the more complex liquid systems at detecting resistance to key drugs like rifampicin, isoniazid, and several newer antibiotics. The team was able to establish specific thresholds, or cutoff points, for each drug within the egg medium. For example, they determined that a concentration of twenty milligrams per liter of rifampicin was the point where the bacteria in the egg medium began to show resistance, a finding that aligns closely with what is known about the bacteria's behavior in other environments.

However, the study also highlighted a significant gap between what the bacteria showed in the lab and what the genetic tests predicted. While the two growing methods agreed with each other, they often disagreed with the genetic analysis. In several cases, the genetic tests failed to detect resistance that was clearly visible when the bacteria were grown in the presence of the drugs. This happened because the genetic tests look for specific, known changes in the DNA, but the bacteria can sometimes develop resistance through other, less common mechanisms that the tests do not scan for. Conversely, the genetic tests sometimes flagged bacteria as resistant when they were actually susceptible, often because they detected harmless genetic variations that did not affect the bacteria's ability to survive the drugs. This discrepancy was particularly noticeable with the bacteria's response to isoniazid and rifampicin, where the genetic tests missed mutations or misinterpreted harmless ones, leading to potential errors in treatment decisions if relied upon alone.

The researchers also explored how well these methods worked for newer drugs like pretomanid and bedaquiline. For pretomanid, the egg-based method performed well, clearly separating the resistant strains from the susceptible ones. However, the situation was different for bedaquiline. In the egg-based medium, the bacteria seemed to survive much higher concentrations of the drug than they did in the liquid tests, making it difficult to determine a reliable cutoff point. The scientists suspect this is because the proteins found in the egg medium interact with the drug, making it less effective in that specific environment. This finding suggests that while the egg-based method is excellent for many drugs, it may not be suitable for testing bedaquiline without further adjustments.

Ultimately, the study confirms that the old egg-based method remains a powerful and reliable tool for detecting drug resistance in tuberculosis. It offers a practical alternative for laboratories that may not have access to the most expensive liquid culture systems, provided they use the newly established thresholds for each drug. The research underscores that while genetic tests are fast and useful, they cannot yet replace the need to actually grow the bacteria and watch how they respond to treatment. The genetic maps are incomplete, and the living bacteria often reveal secrets that the DNA alone cannot show. By validating the egg-based medium against modern standards, the researchers have provided a pathway for more accurate and accessible testing, ensuring that patients receive the right treatment even in resource-limited settings where the disease is most prevalent.