A Statistical Evaluation on the Probability of USP Dissolution Test

溶離率指的是藥錠或膠囊類藥物在人體內溶解的百分率，此一特性會直接影響到藥物中的有效成分釋出及人體吸收率。故在每一批藥品出廠時，都必須檢驗其溶離率是否符合在藥典中所註冊之溶離率。USP/NF(United States Pharmacopeia and National Formulary)提供了一個三階段式的溶離率試驗流程，用以檢驗一批藥物的溶離率是否符合要求。藥物生產者在將藥物產品送檢之前，可先抽一份樣本，用以估計該批藥物通過溶離檢驗。但由於在USP/NF的溶離率試驗流程中，其三階段皆不獨立，且各階段中也同時要求平均值要保持在標準值之上以及淘汰劣質品的條件。這些彼此不獨立的檢驗條件，造成了通過機率的估算式難以用明確的式子表達。在過去的文獻中，除了對於通過機率給予一個下界估計值，也有在限定條件下，直接對通過機率給於一個近似的估計方法。在本論文中，借由蒙地卡羅的大量模擬法，將需要討論之參數範圍做適當的切割，把所有的通過機率精確的估算出來。並將模擬結果製成表格，提供使用者一個明確容易的方法來得到藥品通過溶離率檢驗機率之估計值。Dissolution is the rate that drug dissolves in human body from a solid dosage form(tablet, capsule) after oral administration. It directly influences the quantity of chemical compound releasing from drug. So dissolution is important to the strength of drugs. The dissolution rate of each drug product is then needed to be examined before it release to consumers. For testing whether a drug product has an enough dissolution rate, the United States Pharmacopeia and National Formulary (USP/NF) provides a three-stage dissolution testing procedure. The sponsors usually wants to establish in-house passing probability. The USP/NF dissolution testing procedure is a three- stages test and these stages are dependent. In addition, the criteria at each stage consists of the characteristic of individual units (individual requirements) and the distribution of the sample means (average requirements). It follows that the average requirement and individual requirement are not only dependent within the stage but also correlated between stages. As a result, approximated methods may provide over-estimation or under-estimation of the true passing probability. In this thesis, we use Monte Carlo simulation to provide accurate estimations. Tables of the passing probabilities are pro- vided for practical use. An example is given to illustrate the application of the proposed methods