Influence of physical factors on the accuracy of calibration models for NIR spectroscopy

The quality of pharmaceutical drugs is strongly influenced by a number of physical and chemical factors that require careful control during the production process in order to ensure that the end-product will meet the specifications. Near infrared spectroscopy has proved effective for monitoring changes in such factors and is currently the most widely used technique for controlling drug manufacturing processes. In this work, the authors determined an active pharmaceutical ingredient (API) throughout its production process. The influence of particle size, galenic form, compaction pressure and coating thickness on NIR spectra was evaluated with a view to developing effective methodologies for constructing simple, accurate calibration methods affording API quantization at different stages of a drug production process. All calibration models were constructed from data for laboratory samples alone and NIR calibration models for determining the API determination by using product weights as reference values. The proposed models were validated by application to samples obtained at three stages of a drug manufacturing process and comparison of the predicted values with HPLC reference values. The RSEP values thus obtained never exceeded 1.5%.