Freezing processes are a well-established unit operation in the biopharmaceutical industry to increase the shelf-life of protein-based drugs. While freezing reduces degradation reaction rates, it may also exert stresses such as freeze concentration. Macroscopic freeze concentration in large-scale freezing processes has been described thoroughly by examination of frozen bulk material, but the transient process leading to such freeze concentration profiles has not been monitored yet for biopharmaceutical solutions. In this study, Raman spectroscopy as a process analytical technology is demonstrated for model formulations containing monoclonal antibodies (mAbs) or bovine serum albumin (BSA) in varying concentrations of sucrose and buffer salts. Therefore, a Raman probe was immersed into a bulk volume at different heights, monitoring the freeze concentration in the liquid phase during the freezing processes. Partial least square regression models were used to quantitatively discriminate between the protein and excipients simultaneously. The freeze concentration profiles were dependend on freezing temperature and formulation with freeze concentrations up to 2.4-fold. Convection currents at the bottom of the freezing container were observed with a maximum height of 1 mm. Furthermore, freeze concentration was correlated with the sucrose concentration in a formulation. Analysis of the freeze concentration slope indicated diffusion from the bottom to the top of the container. In summary, Raman spectroscopy is a valuable tool for process validation of freeze concentration simulations and to overcome scale-dependent challenges.

中文翻译：

---

拉曼光谱作为研究生物制药冷冻浓缩过程的过程分析技术

冷冻工艺是生物制药行业中一项成熟的单元操作，可延长蛋白质类药物的保质期。虽然冷冻会降低降解反应速率，但它也可能会产生压力，例如冷冻浓度。大规模冷冻过程中的宏观冷冻浓度已经通过冷冻散装材料的检查得到了彻底的描述，但导致这种冷冻浓度曲线的瞬态过程尚未对生物制药溶液进行监测。在这项研究中，拉曼光谱作为一种过程分析技术被证明适用于含有不同浓度蔗糖和缓冲盐的单克隆抗体 (mAb) 或牛血清白蛋白 (BSA) 的模型配方。因此，将拉曼探针浸入不同高度的体积中，在冷冻过程中监测液相中的冷冻浓度。偏最小二乘回归模型用于同时定量区分蛋白质和赋形剂。冷冻浓度曲线取决于冷冻温度和冷冻浓度高达 2.4 倍的配方。在冷冻容器底部观察到最大高度为 1 mm 的对流。此外，冷冻浓度与制剂中的蔗糖浓度相关。冷冻浓度斜率的分析表明从容器底部到顶部的扩散。总之，拉曼光谱是一种有价值的工具，可用于冷冻浓度模拟的过程验证和克服与规模相关的挑战。