(343g) Application of Fluidized Bed Drying Modeling to the Development and Scale-up of a Continuous Wet Granulation Line

Continuous wet granulation has gained attention in the last few years for the continuous manufacturing of high drug load products. Continuous manufacturing is associated with a reduction of time and cost for drug development and manufacture using more sustainable processes, with minimal scale-up and technology transfer [1] and hence has made a significant progress and won researchers’ attention [2] in the last decade. The simplest set up is direct compression route, consisting of powder feeding, blending, and tableting. In the case of wet granulation, this is extended by incorporating intermediate steps such as wet granulation, drying, and milling. This improves the material flow, compaction, stability, and reduces sticking, among others, which is especially relevant in high drug load blends. To maintain reduced development times, process modelling is key to identify, reduce, and control risk in the different process stages. A part of the process where modelling can significantly contribute is the drying of the wet granules. In the continuous wet granulation technology ConsiGma-25 [3], which is part of the Pfizer Continuous Manufacturing Module (PCMM) [4], the drying process takes place in a segmented fluidized bed dryer (FBD). Particularly, a twin-screw system with liquid injection generates the wet granules, which are distributed sequentially in six cells that dry and discharge a specific load of material, targeting a final granule moisture. The resulting granules are then conditioned and fed at constant throughput to the granule blending and tablet compression.

There are several aspects of the drying stage design and de-risking that can be addressed with digital tools, including: (i) the schedule of the segmented six-cell dryer, (ii) potential drying constraints for high throughputs due to the semi-continuous nature of the drying operation, (iii) defining the discharge condition that guarantees the target loss on drying (LOD) is reached in the final dried granules for new formulations, e.g. typically using the so-called delta-T approach, (iv) the estimation of drying times depending on multiple factors like initial granule moisture, drying air conditions, cell load, other material properties, and their potential variability in the process, (v) the tracking of moisture content in downstream granule feeding, mixing and tableting operations, and (vi) understanding the effect of moisture content in the extra-granule mixing behavior.

In this work, several process modelling tools have been applied in the development of the drying process in a continuous wet granulation line at clinical and commercial scale. This includes the scale-up from a lab-scale process to one-cell fluidized bed drying in a simpler ConsiGma-1 system, and full-scale segmented drying in ConsiGma-25, together with the technology transfer from development to manufacturing. Additionally, digital tools are used to understand the impact of cell-to-cell variability, reduce the number of required experiments for new drugs, and improve the process design for continuous drying. Process simulations allow making better predictions, optimize the drying process, and reduce the development time of new compounds.

This way, a first principles FBD model has been applied, which captures the most relevant experimental conditions, like the liquid/solid ratio, cell filling time, and drying air flowrate, temperature, and relative humidity. The model predicts the drying time required to achieve granules with moisture content below a given target consistently in all cell loads. It is based on material and energy balances, related through the drying rate. The model accounts for two differentiated phases, where the free-bound and the close-bound moisture is evaporated respectively, referred to as the falling rate kinetics [5]. It was developed using APAP ConsiGma-25 data with different initial moisture content. Drying parameters and material properties were measured using DVS (Dynamic Vapor Sorption) and MAL (Material Assessment Lab) analysis. Material transference parameters were estimated using measured dryer cell temperature and moisture content from experiments. Experimental data was used for model validation. This model allows us to predict the drying curve and drying time for reaching a target moisture content for different initial granule water contents (Figure 1). Additionally, the model is directly applied to optimize the drying process using virtual Design of Experiments (DoE) and to evaluate the impact of the variability in different material and processing conditions, including granule moisture, particle size, drying air temperature, flow rate and humidity, and cell load, using Global Systems Analysis. Residence time distribution approaches have been used for material tracking. The six-cell scheduling at different throughputs and liquid/solid ratios, as well as the minimum fluidization condition [6] based on Ergun equation are also analyzed.

[1] K. Plumb (2005). Continuous Processing in the Pharmaceutical Industry: Changing the Mind Set. Chem. Eng. Res. Des. 83, 730–738. https://doi.org/10.1205/cherd.04359.

[2] D. Blackwood, (2018). A Check-in on our Journey Towards Creating a Continuous Direct Compression Platform Technology. IFPAC Annual Meeting, 2018, Maryland, March 11-14.

[3] ConsiGma-25 reference

[4] R. Steiner, M. Jornitz (2017). Continuous processing in the pharmaceutical industry: status and perspective. In: Kleinebudde, P., Khinast, J., Rantanen, J. (Eds.), Continuous Manufacturing of Pharmaceuticals. John Wiley & Sons Inc, pp. 369–403. https://doi.org/10.1002/9781119001348.ch11.

[5] J. Burgschweiger, E. Tsotsas (2002). Experimental investigation and modelling of continuous fluidized bed drying under steady-state and dynamic conditions, Chemical Engineering Science 57 (24), 5021-5038.

[6] S.U. Pradhan, J.W. Bullard, S. Dale, P. Ojakovo, A. Bonnassieux (2022). A scaled down method for identifying the optimum range of L/S ratio in twin screw wet granulation using a regime map approach, International Journal of Pharmaceutics 616, 121542