Design of experiments and multivariate analysis approach to study dissolution stability of a modified-release drug...

The purpose of this paper is to present the use of Design of Experiments and multivariate analysis for evaluation of a modified-release drug product stability to support post-approval lean stability approaches. The focus of the paper was to investigate potential root-causes for acceleration of dissolution upon stability.

For statistical evaluation of stability data, multiple linear regression statistics was used. The design space of the stability study was modeled using MODDE 12.1 software. For experimental set-up, parameters such as Temperature, Time, Packaging, Batch, and Active Pharmaceutical Ingredient supplier were selected.

With multiple linear regression modeling of the all generated stability data until six months, we were able to identify or confirm the Stability-related quality attributes and Shelf life limiting attributes. From the multiple linear regression correlation coefficients, we have evaluated that decrease of an antioxidant upon stability could cause potential shift in dissolution. However, main factors for accelerated dissolution can be attributed to other material and process variables. In the last part of the study, we have shown the usefulness of these methodologies for supporting lean stability approaches. With enhanced drug product knowledge, we designed two reduced long-term stability studies and showed that with 'One-half' reduced design, we would still be able to confirm 24-month shelf life.

Implementing Quality by design approaches on stability studies could reduce the need for excessive analytical testing, help to evaluate meaningfulness of the data and set a risk-based stability testing strategy.