Frozen Assets

There is an increasing push to make everything in biopharma production faster and more efficient – but some things just take time. Cryogenic freezing is one process where, despite efforts to reach extremely low temperatures as quickly as possible, faster is not always better, particularly for cell-based products where freezing too rapidly can damage cell membranes.

Cryogenic freezing is required for a large number of biotherapeutics, as well as stem cells and tissues. However, the freezing protocols for individual products are numerous and highly specific – and many companies struggle because of a lack of controls in freezing equipment. Temperature curves should be individualized based on the specific product type, volume, and concentration to best preserve structural and functional integrity (1).

Broadly speaking, there are two main approaches for cryogenic freezing. Fast uncontrolled freezing is a simple (but far from ideal) cryogenic freezing method that is usually performed via direct contact with liquid nitrogen. This approach can quickly freeze large volumes, but there is little control over freezing rates, which makes it difficult to follow the suggested freezing requirements for a given biopharmaceutical product. For example, CHO cells cannot reestablish equilibrium in such a short time, which may result in intracellular ice formation damaging cellular structures.

Controlled-rate freezing uses comparatively slow freezing rates based upon the product's characteristics. Did you know that CHO cells have an optimum freeze rate of about -1 °C/min. For mRNA, it is a little faster. Controlled-rate freezing reduces the osmotic stress in cells and increases cell survival. The downside of this approach is that it demands advanced equipment with precise temperature monitoring and controls that can reproduce freezing rates at different scales.

Finding the balance between slow and fast freezing is difficult because it can limit applicability under certain circumstances. Fast uncontrolled freezing can have a negative impact on the cell membranes because of the high level of osmotic stress and formation of intracellular ice. On the other hand, slow rates can come with a higher risk of other freeze-related damage, such as cryoconcentration, cell dehydration, or osmotic injuries, if the cooling rates are not ideally met.

The industry needs more controlled-rate freezing processes that allow the optimal freezing rate to be selected, especially during the phase transition (solidification) (2). Fortunately, new innovations are emerging. Plate freezing, for example, allows for controlled freezing rates even when handling larger volumes of biologics. There is efficient heat transfer because of the direct contact with the primary packaging, allowing for more homogenous freezing processes and reducing the risk of freeze-related damages. With plate freezing techniques, biologics may be cooled to temperatures as low as -80 °C, and then further cooled using a cryogenic freezer. This is possible because, below -47 °C, switching to rapid freezing has been shown to produce no negative impact on cell viability after thawing (3, 4). 

Also emerging are novel liquid nitrogen-based cryogenic freezers with fully controlled freezing rates. Despite using liquid nitrogen as a means to lower temperatures, the biologics themselves are not directly exposed because they can be covered in protective packaging, and then frozen by the dosed addition of liquid nitrogen (using injection systems) to help customize the freezing process.

We recently studied the optimal freezing time for CHO-K1 cells and found that the best rates were achieved using such a novel cryogenic freezer at cooling rates of approximately 1 °C/min, with cell viability reaching levels of more than 90 percent (5).

These new systems are not the final frontier of freezing science; technology will need to continue to improve – and accessibility needs to be addressed. Nevertheless, I believe we will see a rising adoption of more controlled and reproducible cryogenic freezing processes.