4.1 A significant amount of data is being generated during pharmaceutical development and manufacturing activities. The interpretation of such data is becoming increasingly difficult. Individual examination of the univariate process variables is relevant but can be significantly complemented by multivariate data analysis (MVDA). Such methodology has been shown to be particularly efficient at handling large amounts of data from multiple sources, summarizing complex information into meaningful low dimensional graphical representations, identifying intricate correlations between multivariate datasets taking into account variable interactions. The output from MVDA will generate useful information that can be used to enhance process understanding, decision making in process development, process monitoring and control (including product release), product life-cycle management and continual improvement.

4.2 MVDA is a widely used tool in various industries including the pharmaceutical industry. To generate a valid outcome, MVDA should contain the following components:

4.2.1 A predefined objective based on a risk and scientific hypothesis specific to the application,

4.2.2 Relevant data,

4.2.3 Appropriate data analysis techniques, including considerations on validation,

4.2.4 Appropriately trained staff, and

4.2.5 Life-cycle management.

4.3 This guide can be used to support data analysis activities associated with pharmaceutical development and manufacturing, process performance and product quality monitoring in manufacturing, as well as for troubleshooting and investigation events. Technical details in data analysis can be found in scientific literature and standard practices in data analysis are already available (such as Practices E1655 and E1790 for spectroscopic applications, Practice E2617 for model validation and Practice E2474 for utilizing process analytical technology).

1.1 This guide covers the applications of multivariate data analysis (MVDA) to support pharmaceutical development and manufacturing activities. MVDA is one of the key enablers for process understanding and decision making in pharmaceutical development, and for the release of intermediate and final products.

1.2 The scope of this guide is to provide general guidelines on the application of MVDA in the pharmaceutical industry. While MVDA refers to typical empirical data analysis, the scope is limited to providing a high level guidance and not intended to provide application-specific data analysis procedures. This guide provides considerations on the following aspects:

1.2.1 Use of a risk-based approach (understanding the objective requirements and assessing the fit-for-use status),

1.2.2 Considerations on the data collection and diagnostics used for MVDA (including data preprocessing and outliers),

1.2.3 Considerations on the different types of data analysis and model validation,

1.2.4 Qualified and competent personnel, and

1.2.5 Life-cycle management of MVDA.

1.3 This standard does not purport to address all of the safety concerns, if any, associated with its use. It is the responsibility of the user of this standard to establish appropriate safety and health practices and determine the applicability of regulatory limitations prior to use.