Feed factor profile prediction model for two-component mixed powder in the twin-screw feeder.

In continuous pharmaceutical manufacturing processes, it is crucial to control the powder flow rate. The feeding process is characterized by the amount of powder delivered per screw rotation, referred to as the feed factor. This study aims to develop models for predicting the feed factor profiles (FFPs) of two-component mixed powders with various formulations, while most previous studies have focused on single-component powders. It further aims to identify the suitable model type and to determine the significance of material properties in enhancing prediction accuracy by using several FFP prediction models with different input variables. Four datasets from the experiment were generated with different ranges of the mass fraction of active pharmaceutical ingredients (API) and the powder weight in the hopper. The candidates for the model inputs are (a) the mass fraction of API, (b) process parameters, and (c) material properties. It is desirable to construct a high-performance prediction model without the material properties because their measurement is laborious. The results show that using (c) as input variables did not improve the prediction accuracy as much, thus there is no need to use them.

Greedy design space construction based on regression and latent space extraction for pharmaceutical development.

Implementation of the design space (DS) is a scientific concept for ensuring quality to be submitted as a part of the regulatory filing of a drug product for approval to market. An empirical approach is constructing the DS based on the regression model whose inputs are process parameters and material attributes over the different unit operations, i.e., a high-dimensional statistical model. While the high-dimensional model assures quality and process flexibility through a comprehensive process understanding, it has difficulty visualizing the feasible range of input parameters, i.e., DS. Therefore, this study proposes a greedy approach to constructing the extensive and flexible low-dimensional DS based on the high-dimensional statistical model and the observed internal representations that satisfies both comprehensive process understanding and the DS visualization capability. Introducing the observed correlation structure enabled the dimensionality reduction of the DS. The non-critical controllable parameters were fixed to the target values in visualizing the low-dimensional DS as a function of critical parameters. The expected variation of non-critical non-controllable parameters was considered the source of variation in prediction. The case study demonstrated the proposed approach's usefulness for developing the pharmaceutical manufacturing process.

Evaluation of infrared-reflection absorption spectroscopy measurement and locally weighted partial least-squares for rapid analysis of residual drug substances in cleaning processes.

The usefulness of infrared-reflection absorption spectroscopy (IR-RAS) for the rapid measurement of residual drug substances without sampling was evaluated. In order to realize the highly accurate rapid measurement, locally weighted partial least-squares (LW-PLS) with a new weighting technique was developed. LW-PLS is an adaptive method that builds a calibration model on demand by using a database whenever prediction is required. By adding more weight to samples closer to a query, LW-PLS can achieve higher prediction accuracy than PLS. In this study, a new weighting technique is proposed to further improve the prediction accuracy of LW-PLS. The root-mean-square error of prediction (RMSEP) of the IR-RAS spectra analyzed by LW-PLS with the new weighting technique was compared with that analyzed by PLS and locally weighted regression (LWR). The RMSEP of LW-PLS with the proposed weighting technique was about 36% and 14% smaller than that of PLS and LWR, respectively, when ibuprofen was a residual drug substance. Similarly, LW-PLS with the weighting technique was about 39% and 24% better than PLS and LWR in RMSEP, respectively, when magnesium stearate was a residual excipient. The combination of IR-RAS and LW-PLS with the proposed weighting technique is a very useful rapid measurement technique of the residual drug substances.

Estimation of active pharmaceutical ingredients content using locally weighted partial least squares and statistical wavelength selection.

Development of quality estimation models using near infrared spectroscopy (NIRS) and multivariate analysis has been accelerated as a process analytical technology (PAT) tool in the pharmaceutical industry. Although linear regression methods such as partial least squares (PLS) are widely used, they cannot always achieve high estimation accuracy because physical and chemical properties of a measuring object have a complex effect on NIR spectra. In this research, locally weighted PLS (LW-PLS) which utilizes a newly defined similarity between samples is proposed to estimate active pharmaceutical ingredient (API) content in granules for tableting. In addition, a statistical wavelength selection method which quantifies the effect of API content and other factors on NIR spectra is proposed. LW-PLS and the proposed wavelength selection method were applied to real process data provided by Daiichi Sankyo Co., Ltd., and the estimation accuracy was improved by 38.6% in root mean square error of prediction (RMSEP) compared to the conventional PLS using wavelengths selected on the basis of variable importance on the projection (VIP). The results clearly show that the proposed calibration modeling technique is useful for API content estimation and is superior to the conventional one.