(171d) Leveraging Mechanistic Models for Scale up and Optimization of Lyophilization

Leveraging mechanistic models for scale up and optimization of Lyophilization.

Abstract

Generic pharmaceutical industry has to operate in tight timelines of product development and scale up from lab to plant where time to market makes or breaks the viability of the business case. Getting it right the first time demands detailed product & process understanding and implementing appropriate mechanistic modelling and Quality-by-design (QbD) approaches in the product life cycle, to minimize downstream risks. This work describes example cases of such efforts involving the freeze/ vacuum drying unit operation with solvents.

Lyophilization or freeze drying with aqueous/ solvent systems is a routine, but expensive, unit operation carried out across a range of industries (pharmaceuticals, food, paper, etc). It’s especially relevant for temperature sensitive materials (for ex., peptides, liposomes, and onco molecules) where traditional high temperature drying unit operations cannot be deployed. The process involves loading the solutions of the required products in vials or trays and drying to obtain typically a free flowing powder (without meltback or collapse or oiling out). Typical challenges in this process involve: defining the optimal drying conditions (temperature, pressure vs. time) for a given material on a certain equipment so that product of an acceptable quality is obtained; ensuring uniformity of this quality across the vials in a given batch; and translation of such conditions during scale-up from lab scale to plant scale. For both the aqueous and solvent based systems, this demands a quantitative understanding of the impact of the complex interactions (Figure 1) of the relevant material properties, equipment characteristics, process conditions on the quality of the output

Figure 1 – Lyophilization- CMAs, CQAs, and CPPs.

product. This is especially true as the cost of failure at plant scale is prohibitive as this is typically the last step in the manufacturing of these products. Substantial literature exists (for example, M.J. Pikal, et al 1984, Pisano et al 2013) in capturing the vial-level physics with quasi-steady state assumptions to describe aqueous systems and demonstrating the applicability. These are solved using one-dimensional coupled heat and mass transfer equations.

The present work focuses on leveraging vial and equipment level physics to help with the definition and scale-up of lyo cycles in aqueous and solvent based systems. The first part focuses via case studies on how thermal characterization of lyophilizers can be leveraged to: bring predictability in scale-up (from 100s of vials to 1000s of vials) and; optimize lyo cycles using a novel algorithm. Example scenarios show such optimization lowering the drying time by as much as 20% compared to the traditional cycle development strategy. The second part presents cases of how such efforts can be extended to solvent systems by incorporation of additional physics. The higher vapor pressure of the solvent based systems demands more attention to the downstream utilities such as condenser and vacuum pump during scale up.