(344b) Continuous Production Technology for Pharmaceutical Semi-Solid and Liquid Formulations: Processability and Influence of Process Parameters on the Product Quality

¹Laboratory of Pharmaceutical Process Analytical Technology, Ghent University, Ottergemsesteenweg 460, 9000 Ghent, Belgium

²Janssen Global Technical Operations, Pharmaceutical Mfg (PM) Platform, Turnhoutseweg 30, 2340 Beerse, Belgium

³Laboratory of Pharmaceutical Technology, Ghent University, Ottergemsesteenweg 460, 9000 Ghent, Belgium

INTRODUCTION

Pharmaceutical production today mainly consists of batch processes. These are time-consuming, expensive and sub-optimal processes. Continuous processes on the other hand, allow for more flexibility, an easier scale-up and better quality control. The aim of this study is to investigate whether the Contimix system, a modular continuous production system, initially developed for the continuous production of semi-solid and liquid cosmetics, can be employed for pharmaceutical applications as well. A commercially available pharmaceutical cream, which is originally produced via a batch process, was manufactured with the Contimix system. In a first step, the processability of the cream with the Contimix system was investigated. Furthermore, the influence of the process parameters on the product quality was studied through statistical Design of Experiments (DoE), where the different process parameters were varied simultaneously.

MATERIALS AND METHODS

Materials

A pharmaceutical cream (Janssen Pharmaceutica N.V., Beerse, Belgium) was selected as model formulation.

Methods

All creams were produced with a continuous production system (Contimix, Hebold systems, Cuxhaven, Germany). A two level fractional factorial design was applied to study the influence of five process parameters on the product quality: process temperature, throughput, mixing speed in mixing vessel 1, mixing speed in mixing vessel 2 and water phase temperature. The studied quality attributes of the cream (DoE responses) were Active Pharmaceutical Ingredient (API) concentration (at the beginning, middle and end of a DoE run), yield point, viscosity of the yield point, viscosity (at low, intermediate and high shear rate) and pH. The data was analyzed using the MODDE Pro 11.0 software (MKS Umetrics, Umeå, Sweden). The API concentration of the creams was measured using a Raman Rxn2 spectrometer (Kaiser Optical Systems, Ann Arbor, MI, USA), equipped with a fiber-optic PhAT probe. The laser wavelength was 785 nm. Spectra were collected off-line with an exposure time of 20 s and a resolution of 5 cm^-1. The spectral range of the system was from 150-1890 cm^-1. Rheological measurements were conducted with a Haake Mars III rheometer (Thermo Fisher Scientific, Karlsruhe, Germany). The experiments were carried out at a fixed temperature (25°C), using a 60 mm parallel plate geometry. A stress ramp and viscosity curve were obtained from each produced cream. The pH of the creams was measured with a glass HI 3220 probe (Hanna instruments, Temse, Belgium).

RESULTS AND DISCUSSION

Prior to the experiments, the raw materials of the cream were divided into three pre-phases, which were batch mixed before the actual continuous production. This additional step was necessary since the Contimix system does not feature as many pumps as there are components in the cream. Furthermore, there is no possibility of introducing powders directly into the process. The results of the DoE show that the influence of the process parameters on the pH was negligible. The API concentration at the three time points between the experimental runs was not influenced by the process parameters. However, when the API concentration within the runs was evaluated, a downward trend in API concentration was detected for runs with a high throughput. This was caused by an inhomogeneity of pre-phase 3, which contained the API. Even for runs produced at a lower throughput, a similar trend was observed, suggesting that the homogenization of pre-phase 3 was not optimal, leading to an inhomogeneity of the API in the final product. The stress ramp experiments showed an influence of the process temperature. A higher process temperature was linked with a lower viscosity. The viscosity curve experiments showed an influence of the process temperature and throughput, but both influences were minimal. However, when the viscosity curves were investigated, a difference between the runs at low and high process temperature was visible. A possible explanation for the influence of the process temperature on the yield point and viscosity can be found in the short time between manufacturing the creams and conducting the rheological experiments. To draw conclusions about this effect, a number of runs at low and high process temperature should be executed and the viscosity of these creams should be measured directly (1 day) and a longer period (4 weeks) after production. Despite achieving processability, further improvements are necessary to transfer this continuous production technology to the pharmaceutical industry. The batch pre-step should be replaced by a continuous step. In order to achieve this, a way of feeding powder directly into the system is necessary. This would make the production system completely continuous, with all the advantages this entails. The original batch produced cream was measured with the same techniques as the creams of the DoE runs and compared with a center point run (run 18). The pH and API concentration were similar for both creams. From the stress ramp experiment, it can be concluded that the yield point of run 18 was slightly higher than that of the original product, indicating that both creams have similar rheological properties. This was confirmed by the viscosity curves, where the viscosity decrease with increasing shear rate was similar.

CONCLUSION

Processability of a pharmaceutical cream with the Contimix, a continuous production system, was achieved. From the screening design, it can be concluded that the process temperature had a significant influence on the viscosity of the creams, i.e. higher process temperature resulted in creams with a lower viscosity. Changing the other process parameters had no significant impact on the measured quality attributes. Furthermore, the Contimix was capable of producing a commercially available pharmaceutical cream in a continuous way with a similar quality as the original formulation.