The reasoning behind the need to set specifications for drug substance and drug product lies in the fact that they are part of the control strategy that guarantees the appropriate quality of each lot (See Box 3.3.4). In other words, specifications serve to confirm the quality, and not to characterize the product. The manufacturer is expected to provide the rationale and justification for including and/or excluding certain tests for specific quality attributes. This justification needs to be based on scientifically sound considerations.

The following points should be taken into consideration when establishing scientifically justifiable specifications.

• Specifications are linked to a manufacturing process.
  Specifications are used to demonstrate manufacturing consistency, specifically for product-related substances, product-related impurities and process-related impurities. Heterogeneity patterns other than those specified during preclinical and clinical development need investigation (more 3.3.7).
• Specifications should account for the stability of drug substance and drug product.
  Specifications should take into account changes in the quality due to degradation during storage. Since bioproducts are so diverse, their stability characteristics cannot be assessed using a universal stability indicator. The manufacturer should propose a product-specific stability-indicating profile (more 3.3.8).
• Specifications are linked to preclinical and clinical studies.
  Specifications should be based on data obtained for lots used in pre-clinical and clinical studies (more 3.3.9).
• Specifications are linked to analytical procedures.
  There are multiple analytical procedures used for critical quality attributes (e.g. potency, the nature and quantity of product-related substances, product-related impurities and process-related impurities). The analytical technology may evolve in the course of product development, so the data obtained during development should correlate with the data obtained at the time the marketing application is filed (more 3.3.10).

The battery of tests that get included in the specifications depends on each specific product. The rationale behind the definition of acceptable range of acceptance criteria should be justified based on data obtained from lots used in preclinical and/or clinical studies, data from lots used for demonstration of manufacturing consistency, and data from stability studies, and relevant development data. In some cases, testing at production stages may be sufficient (more 3.3.11).

The tests and acceptance criteria that apply to drug substance specifications are outlined in Table 3.4.1 . Pharmacopoeial tests (e.g., endotoxin detection) should be performed on the drug substance, where appropriate. Additional drug-substance-specific acceptance criteria may also be necessary.

The tests and acceptance criteria that apply to drug product specifications correspond to those that apply to drug substances (Table 3.3.7).

Pharmacopoeial tests (sterility, endotoxins, microbial limits, volume in container, particulate matter, uniformity of dosage units, moisture content etc.) apply to the relevant dosage forms. The tests for uniformity of dosage units may be done as in-process controls and corresponding acceptance criteria set. Some unique dosage forms may need additional tests other than those outlined in (Table 3.3.7).

With reference to quality assurance of food control laboratories, special attention is paid to ensure that the consumers get a safe, wholesome food product free from adulteration, correctly labelled and presented. A Codex standard for any food or foods should be drawn up in accordance with the Format for Codex Commodity Standards (2). The Codex Alimentarius Commission has also adopted Harmonized Guidelines for Internal Quality Control in Analytical Chemistry Laboratories. They are particularly relevant for the application of molecular and immunological analytical methods which are currently recognized as tools for determination of specific DNA and protein analytes in foods.