Original title: Narrow Therapeutic Index Drugs: Regulatory Frameworks, Bioequivalence, and Model-Informed Precision Dosing
Mengmeng Liu, Zhe Wang, Yongbo Chen, Yuqi Jia +4 · September 2026
Some medicines have a narrow therapeutic indexA drug where the dose that works is very close to the dose that causes harm.. This means the gap between a dose that heals and a dose that poisons is very small. Because of this tiny margin, researchers reviewed global rules for these drugs and checked how computers can help guide safe dosing. Different countries make generic versions pass different tests before patients can switch between brands. Computers help solve this by modeling how medicines move through the human body. The researchers found that global rules and drug lists vary widely. Several agencies tighten their test limits, while China and the United States require fully replicated crossover studies for certain drugs. Meanwhile, population-based and Bayesian computer tools are the most mature methods to track and personalize doses. Newer machine learning and hybrid tools help predict drug exposure, doses, and toxicity risks, but most evidence for them is retrospective and needs broader testing.
You are baking a delicate cake where one extra gram of salt makes it inedible. Different recipe books give different instructions for measuring that salt, and some allow a wider margin of error than others. In this paper, the cake is a narrow therapeutic index drug, the salt is the dose, and the recipe books are the regulatory frameworks from different countries. The paper compares these rules and explores smarter ways to measure the dose precisely for each patient.
Patients on drugs like warfarin or lithium rely on precise dosing. Yet different countries have different rules for approving generic versions, creating risks when patients switch products. Better models could personalize doses, but they are not yet proven in real-world settings.
This is a review, not a new study. The evidence for machine learning and hybrid models is mostly retrospective and lacks external validation. The paper does not propose a specific harmonized framework, only calls for one. The review may not cover all regional differences exhaustively.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
Narrow Therapeutic Index Drugs: Regulatory Frameworks, Bioequivalence, and Model-Informed Precision Dosing — Mengmeng Liu, Zhe Wang, Yongbo Chen et al. — Pharmaceutics — doi:10.3390/pharmaceutics18091190
Abstract licence: CC BY
Explanation generated by Synthesis from the abstract — it is not the original paper.