Evaluation of Changes in Pharmaceutical Formulation on NIR and Raman Drug Content Calibration Models

To evaluate the effect of changes in a pharmaceutical formulation on NIR and Raman drug content calibration models. This study is important since the development of calibration models in NIR spectroscopy is a time consuming effort and formulations are often varied during product development. Training set samples were prepared with different Ibuprofen concentrations that ranged from 1.0% to 5.0% (w/w). The formulation contains spray dried hydrous lactose, colloidal silicon dioxide and microcrystalline cellulose. The Concentrations of Lactose and MCC were varied from ±3 to ±7 to break the correlations of excipient. A formulation of 3% (w/w) ibuprofen was prepared with croscarmellose sodium instead of microcrystalline cellulose. Spectra for each formulation were taken with a Hololab 500 Raman RXN System from Kaiser Optical Systems (Ann Arbor, MI) with a non-contact probe System™. Near infrared transmittance spectra were obtained with a Bruker MPA instrument. NIR and Raman spectra were analyzed using Pirouette™ 3.11 software to develop the multivariate calibration model.The NIR calibration model predicted the ibuprofen drug concentration in tablets with acceptable accuracy from 0.05 – 5.0% (w/w). A similar calibration model will be developed using Raman spectroscopy. Raman and NIR results were compared, including their ability to quantify the drug content in the croscarmellose formulation. A qualitative analysis shows the specific narrow bands of Raman and the broad bands of NIR. Calibration models may be developed with both Raman and NIR systems. In NIR, the calibration models with Crosscarmelose Sodium require more calibration factors to account for the variation. In Raman, additional work is required to determine the effect of excipient.