(299a) Tswg Development and Scale up from Laboratory Benchtop Experiments

Tablets represent most of the commercially available dosage forms of active pharmaceutical ingredients (API. Broadly, three routes exist for tablets manufacturing – direct compression, dry granulation, and wet granulation. In many cases, direct compression is not feasible due to poor blend flowability, and therefore a granulation step is usually required.

The industry has been moving from batch to continuous manufacturing, and in this sense, twin-screw wet granulation (TSWG) is increasingly being considered for continuous processing of pharmaceutical products owing to its flexibility in process settings, wide range of throughput, compact size, and short residence time. The overall TSWG process can be divided into five steps: (i) feeding; (ii) blending; (iii) granulation with two co-rotating screws assembled in a barrel top with inlet ports for powder and liquid feeds (iv) fluid bed drying in which moisture content is reduced to a target value; and (v) milling to reduce oversized granules which compromise the tabletability.

The current study aimed to demonstrate the development of a TSWG process in a scaled down, batch lab setup leveraged with models to predict conditions of full-scale continuous manufacturing. To this end, dried granules from multiple formulations were produced and characterized in a batch lab setup consisting of Thermo Sientific™ Pharma 11 extruder, MiniGlatt fluid bed dryer (FBD) and FreWitt-lab conical mill, or in a GEA Pharma Systems ConsiGma™‑1 system (Collette™, Wommelgem, Belgium). Previous work [1] has concluded that the ConsiGma™-1 system (mobile lab unit) is equivalent to ConsiGma™‑25 systems.

The experimental work in the granulation step at both process scales included a three-factorial design varying excipients ratio, liquid-to-solid (L/S) ratio, and screw speed. While screw configurations, formulation (excipient ratio range), and L/S ratio ranges were maintained between process scales, the screw speed levels were adjusted so that respective dimensionless powder feed numbers (PFN) are similar at both process scales. Thus, comparable fill levels between process scales are obtained, given that PFN is proportional to the granulator fractional fill level [2].

To guide the range of L/S ratios chosen, torque measurements were done in different equipment: Mixer Torque Rheometer (MRT), IKA Magic Plant and FT4 Powder Rheometer^®. In the literature it is noted that the torque variation as a function of the L/S ratio can be used to determine a preliminary L/S design space [4]. Preliminary results showed that the IKA Magic Plant provided more accurate results than FT4, directly comparable to the data acquired with the Mixer Torque Rheometer.

The design of experiments for the drying step varied the dryer air flow rate, dryer air inlet temperature, and drying time while maintaining the wet material mass loads. Here, the levels of dryer air flow rate used were adjusted so that comparable fluidization regimes are obtained in both scales. For this purpose, a model was developed predicting the fluidization maps (including minimum gas velocity for fluidization and entrainment gas velocity) given the granule properties and process parameters. In addition, an FBD model was developed resembling the one in Gavi [3] to predict the drying curves and temperature profiles in MiniGlatt and ConsiGmaâ„¢-1 system. The normalized single-particle drying kinetics and sorption isotherms were obtained from the dynamic vapor sorption technique.

The results of the granulation step obtained at both process scales were compared using process maps. These are useful to better understand the influence of scale-independent parameters (PFN, specific mechanical energy, L/S) on granule quality attributes (e.g., porosity, particle size) and find equivalences and differences in terms of granulation mechanisms between process scales. Similarly, the drying curves obtained in MiniGlatt and ConsiGmaâ„¢-1 system show similarities, indicating that it is possible to approximate the fluidization regimes between scales by judiciously adjusting the dryer air flow rate and mass load of wet materials. Moreover, the FBD model was able to provide reasonable predictions of the granule residual moisture content and temperature over time, making it an important tool defining process operating ranges of interest.

References

1. Vercruysse J, Peeters E, Fonteyne M, Cappuyns P, Delaet U, Van Assche I, et al. Use of a continuous twin screw granulation and drying system during formulation development and process optimization. European Journal of Pharmaceutics and Biopharmaceutics. 2015; 89:239–47.
2. Osorio JG, Sayin R, Kalbag AV, Litster JD, Martinez‐Marcos L, Lamprou DA, et al. Scaling of continuous twin screw wet granulation. AIChE Journal. 2017; 63:921–32.
3. Gavi E. Application of a mechanistic model of batch fluidized bed drying at laboratory and pilot scale. Drying Technology. 2020; 38:1062–78.
4. Junnila A, Wikstrom H, Megarry A, Gholami A, Papathanasiou F, Blomberg A, Ketolainen J, Tajarobi P, Faster to First-time-in-Human: Prediction of the liquid solid ratio for continuous wet granulation. European Journal of Pharmaceutical Sciences. 2022; 172.