PREVER framework for specification of related substances based on in silico prediction and forced degradation studies

Medicines are not static objects; they are chemical entities that can change over time, much like food that spoils or metal that rusts. When a drug breaks down, it can form new substances called impurities or degradation products. Even in tiny amounts, these new chemicals can alter how well a medicine works or, worse, make it unsafe to take. For scientists developing new drugs, the challenge is to predict exactly what these unwanted byproducts might be before the medicine ever reaches a patient. This is especially difficult in the early stages of development, when there is not yet enough real-world data on how the drug behaves over months or years. To solve this, researchers are increasingly turning to a strategy called "Safe and Sustainable by Design," which uses computer models and careful testing to anticipate risks and build safety into the product from the very beginning.

A team of researchers at Farmanguinhos in Brazil applied this forward-thinking approach to a common medication used to treat schistosomiasis, a parasitic disease that affects millions of people worldwide. The drug in question is praziquantel, often sold as 600-milligram tablets. The team developed a new framework called PREVER to figure out exactly what impurities might appear in these tablets and to set strict limits on how much of them is allowed. Their goal was to create a safety net that combines computer predictions with real-world experiments, ensuring that the medicine remains pure and effective throughout its shelf life.

The process began with a digital prediction phase. The researchers used computer software to simulate how the praziquantel molecule might break apart under various stressful conditions, such as exposure to acid, heat, or light. Based on these simulations and a review of existing scientific literature, they identified six potential impurities that could theoretically form. These included a substance called praziquanamine, a compound known as cyclohexane carboxylic acid, and several other molecules with specific chemical weights. At this stage, these were just possibilities, theoretical ghosts of chemicals that might exist if the drug were to degrade.

To see if these ghosts were real, the team moved to the laboratory for the reproduction phase. They took actual praziquantel tablets and subjected them to extreme stress conditions designed to force the drug to break down faster than it would in a normal medicine cabinet. They exposed the tablets to strong acids, bases, and oxidizing agents, and heated them to high temperatures. Using advanced chemical analysis tools, they watched to see which of the predicted impurities actually appeared. The experiment confirmed that three of the predicted substances did indeed form: praziquanamine and two other degradation products the team labeled DP-1 and DP-2. Although the team expected cyclohexane carboxylic acid to form alongside praziquanamine, it was not detected experimentally, likely because it lacks the chemical structure needed to be seen by their specific testing equipment.

With the list of real impurities in hand, the researchers then had to determine if they were dangerous. Instead of testing each one on animals, which is time-consuming and raises ethical concerns, they used a computer-based toxicological assessment. This method, known as in silico testing, compares the structure of the new impurities to known chemicals with established safety profiles. The computer models predicted that none of the impurities were likely to cause genetic mutations or other serious health issues. Based on these safety calculations, the team established temporary safety limits for each impurity. For example, they decided that praziquanamine should not exceed 0.2 percent of the total tablet weight, while DP-1 could be allowed up to 0.5 percent. These numbers provided a provisional safety guardrail while the product was being finalized.

The final and most critical step was verification. The team stored the praziquantel tablets in a controlled environment that mimicked the hot and humid conditions of a tropical climate for two full years. They tested the tablets regularly to see if any of the impurities they had predicted and temporarily limited actually appeared during normal storage. The results were remarkably clean. After 24 months, none of the five major impurities they had been watching for were detected in the tablets. The drug remained stable, and the predicted breakdown products simply did not form under real-world storage conditions.

Because the long-term study showed no degradation, the researchers adjusted their final safety rules. Instead of keeping the different limits for each specific impurity, they established a single, conservative rule: the total amount of any single impurity, and the total of all impurities combined, must not exceed 0.2 percent. This decision was based on the fact that the drug was more stable than the worst-case scenarios suggested, yet the team still wanted to maintain a high safety margin. The study concluded that the PREVER framework successfully guided the development of the medicine, using a mix of computer predictions and experimental proof to ensure the final product was safe, stable, and ready for patients who rely on it to fight a devastating disease.