NIR spectroscopy for monitoring of the critical manufacturing steps and quality attributes of paliperidone prolonged release tablets

Abstract

The present work aimed to develop near-infrared (NIR) methods for the in-line monitoring of the powder blending process and off-line prediction of paliperidone tablets quality attributes.The powder blending monitoring method was developed for a Y-shaped mixing equipment, with the spectrometer attached to the filling port and remotely controlled, assuring that the NIR sensor was covered in powder every time spectra were recorded. The method was calibrated by using a set of five different formulations corresponding to a paliperidone content ranging from 60 to 140% of the expected product content. Based on the recorded data, a Partial Least Squares method has been developed taking into account the domain between 1350 and 1550nm of the 1st Savitzky-Golay derivatised spectra, with a low root mean square error of cross-validation of 6.7, offering a real time overview of the homogenisation process.The second part of the study was focused on the calibration and validation of NIR methods for the prediction of paliperidone content and resistance to crushing (hardness) of tablets. Both calibration strategies involved using samples with five levels of variation, 80 to 120% of the target drug concentration and hardness, respectively. The methods were validated by using three level datasets, 90 to 110% of the property of interest. The obtained results were promising, with measured vs. predicted recoveries of over 97.2% for paliperidone content and over 96.3% for tensile strength, values well within the tolerance limits.The in-line blending monitoring allowed a good process overview, enabling the establishment of the process end point, while the noninvasive assay of the tablets quality attributes granted an adequate end product.