Impurities decision tree

Q6A Specifications Test Procedures and Acceptance Criteria for New Drug Substances and New Drug Products Chemical Substances (including Decision Trees) Specification setting/ impurity testing... 

The process of qualifying an individual impurity or a given impurity profile at a specified level(s) is summarized in Table 1.1. Safety studies can be conducted on the drug containing the impurity or on the isolated impurity. Several decision trees have been proposed describing threshold levels... 

The profile of impurities in a new drug substance may change for a variety of reasons, such as process scale-up changes, synthetic route changes, and changes made to key intermediates. ICH decision trees help classify, qualify, and select limits for new molecular entities (NMEs). If an impurity exceeds the qualification threshold listed below in Table 3 (ICFI Q3A(R)), studies are needed to qualify that impurity in drug substances. 

Adequate data may be available in the scientific literature to qualify an impurity. Additional safety testing becomes necessary when safety data from the literature is not available and decreasing the level of the impurity below the threshold is not possible. Figure 1 is an illustration of the decision tree for consideration of Safety Studies for drug substances in NDAs. 

Tests for New Drug Substances That Are Optically Active. Chiral impurities had been excluded from ICH guidelines on impurities because of the practical difficulties of measuring them. However, in Q6A, a decision tree describes if, and when, chiral identity tests, impurity tests, and assays may be needed for both new drug substances and products, according to criteria such as the following ... 

A decision tree is provided that enables a logical sequence to be followed, culminating, if necessary, in toxicology studies to assign safe levels to impurities. 

A decision tree for qualification to establish the biological safety of an impurity is described. 

Organic and inorganic impurities and residual solvents are included in this category. Scheme 4 shows a decision tree that can be used to decide the disposition of impurities in the drug substance. 

Scheme 4 Decision tree for disposition of impurities in the drug substances. Source. Courtesy of Pharmquest Corporation, Mountain View, California, U.S.A.

The impurity of a cluster can be formulated in a number of ways as we shall see in Section 15.1.9. Here even a simple measure will do, for example the ratio of word-1 over word-2. The stopping criteria usually involves specifying a minimum decrease in impurity. The decision tree gets round the problem of modelling all the feature combinations by effectively clustering certain feature combinations together. This is not always ideal, but it certainly does allow for more accurate modelling than naive Bayes, as we can see that the same feature can appear in different parts of the tree and have a different affect on the outcome. 

Stopping criteria usually involve specifying a minimum decrease in impurity, and that clusters should have a minimum occupancy (say 10 data points). The process of examining and splitting clusters is shown in Figures 15.9, 15.10, and 15.11. A trained decision tree is shown in Figure 15.12. 

---