Optimizing cryopreservation strategies for scalable cell therapies: A comprehensive review with insights from iPSC‐derived therapies

Abstract

Off‐the‐shelf cell therapies hold significant curative potential for conditions, such as Parkinson's disease and heart failure. However, these therapies face unique cryopreservation challenges, especially when novel routes of administration, such as intracerebral or epicardial injection, require cryopreservation media that are safe for direct post‐thaw administration. Current practices often involve post‐thaw washing to remove dimethyl sulfoxide (Me₂SO), a cytotoxic cryoprotective agent, which complicates the development and clinical translation of off‐the‐shelf therapies. To overcome these obstacles, there is a critical need to explore Me₂SO‐free cryopreservation methods. While such methods typically yield suboptimal post‐thaw viability with conventional slow‐freeze protocols, optimizing freezing profiles offers a promising strategy to enhance their performance. This comprehensive review examines the latest advancements in cryopreservation techniques across various cell therapy platforms, with a specific case study of iPSC‐derived therapies used to illustrate the scalability challenges. By identifying key thermodynamic and biochemical phenomena that occur during freezing, this review aims to identify cell‐type independent approaches to improve the efficiency and efficacy of cryopreservation strategies, thereby supporting the widespread adoption and clinical success of off‐the‐shelf cell therapies.

1. INTRODUCTION

Off‐the‐shelf cell therapies have the potential to revolutionize cell and gene therapy. Currently, patient‐derived (autologous) cell therapies face significant scalability challenges and high costs, reducing their accessibility. Due to being patient‐specific each autologous cell therapy can only serve one individual. The average cost of cell therapy is over $400,000 and can be as high as $1 million USD per patient. ¹ , ² In contrast, donor‐derived (allogeneic) off‐the‐shelf cell therapies can be mass‐produced and potentially treat millions of patients. This scalability promises to reduce costs, expanding their availability and access. One of the key bottlenecks in realizing off‐the‐shelf cell therapies is achieving effective cryopreservation. Current cryopreservation protocols used in preclinical and clinical cell therapy candidates are not conducive to an off‐the‐shelf approach. This limitation hinders the translation of these therapies from the clinic to the market.

Cryopreservation involves freezing sensitive materials at ultra‐low temperatures to extend their shelf life. Protocols for cryopreserving living cells have been developed over several decades and typically involve using 5%–10% dimethyl sulfoxide (Me₂SO) as a cryoprotective agent (CPA). ³ The cell suspension is frozen at a rate of 1°C per minute to a temperature of −80°C ⁴ and then stored in the vapor phase of liquid nitrogen at approximately −130°C. When ready for use, the cells are rapidly thawed in a 37°C water bath or with an automated thawing device. Post‐thaw, Me₂SO is removed due to its cytotoxicity at temperatures above 0°C. This is typically achieved by diluting the Me₂SO with a neutral buffer, centrifuging the cell suspension to separate and remove the supernatant, and then resuspending the cells in their final formulation.

However, cell therapy presents a unique challenge: the cells are administered to patients post‐thaw, meaning the CPAs in the cryopreservation media must be safe for administration. While the intravenous administration of Me₂SO is common practice for chimeric antigen receptor (CAR)‐T therapies and hematopoietic stem cell (HSC) transfusions, it has been associated with adverse events. Most adverse events are minor, such as nausea and headaches. However, severe cases and fatalities, though rare, have been reported. ⁵ , ⁶ , ⁷

Although the risks associated with intravenous Me₂SO administration are well documented, novel cell therapy candidates are exploring novel administration routes, such as direct injections into the heart, ⁸ spine, ⁹ brain, ¹⁰ and eye. ¹¹ There is limited in vivo safety data for Me₂SO administration at these sites, and in vitro studies indicate potential cytotoxicity. Me₂SO concentrations as low as 1% have decreased viability in rat retinal ganglion neurons, ¹² and 0.5% concentrations resulted in a 50% viability loss in rat hippocampal neurons. ¹³ Cell viability is a critical quality attribute (CQA) for cell therapies. ¹⁴ Both in vivo data from intravenous administration of Me₂SO and in vitro findings from relevant cell types highlight risks to patient safety and product quality. Additionally, the lack of understanding surrounding these novel administration routes heightens the risk of using Me₂SO as a cryoprotectant. For this reason, preclinical and clinical cell therapy candidates remove the Me₂SO post‐thaw and wash and resuspend the cell product in a saline buffer before administration at the point‐of‐care. This introduces significant risks including contamination through adventitious agents due to an open process ¹⁵ and potential product damage through pipetting‐induced shear stress. ¹⁶ As such, it also poses risks to patient safety.

The authors conducted a meta‐analysis of clinical trials involving induced pluripotent stem cell (iPSC)‐based therapies in an attempt to identify trends in cryopreservation. iPSCs were chosen due to their suitability for allogeneic approaches, which facilitate the manufacturing of off‐the‐shelf cell therapies, by providing an ethical and theoretically limitless source of cells with limitless differentiation potential, as opposed to using embryonic stem cell‐derived cell therapies, or terminally differentiated cells. This mini‐review serves as a case study to highlight specific issues and the broader need for optimized cryopreservation strategies that can be applied across various cell therapy platforms.

1.1. Case study 1: Minireview of iPSC‐based cell therapy clinical trials

Clinical studies from two systematic reviews of iPSC‐based clinical trials were analyzed ¹⁷ , ¹⁸ ; with additional studies added from the literature. For detailed methodology refer to Ref. [17]. A total of 57 clinical trials were included in the analysis. 32% (18/57) of trials disclosed use of Me₂SO; 9% (5/57) describe a post‐thaw wash step prior to administration in publications associated with the clinical trial (JPRN‐JMA‐IIA00384, ¹⁹ JPRN‐JMA‐IIA00385, ¹⁹ JPRN‐jRCTa031190228, ⁹ NCT06394232, ¹¹ and NCT03763136. ²⁰ ) 5% (n = 3/57) of clinical trials administer the iPSC‐based cell product fresh after being cultured for up to 96 h post‐thaw and stored at 4–8°C prior to administration (JPRN‐JMA‐IIA00384, ¹⁰ JPRN‐JMA‐IIA00385, ¹⁰ and JPRN‐jRCTa031190228 ⁹ ). Both post‐thaw washing and culture of the cell product are performed at the point‐of‐care, and are logistically challenging as they are examples of postprocessing, and make the adoption of off‐the‐shelf cell therapies difficult. Use of a safe to administer Me₂SO‐free cryopreservation media would eliminate the need for post‐thaw washing. A limitation of this meta‐analysis is that few of the clinical trials (22%, n = 13/57) in the sample disclose the cryopreservation protocol employed. To elucidate some trends, an analysis of cryopreservation protocols for preclinical studies involving iPSC‐based cell therapy products was conducted.

1.2. Case study 2: Minireview of cryopreservation practices in iPSC‐based cell therapies

A minireview was conducted by the authors to illustrate cryopreservation practices in iPSC‐based cell therapies. The terms “iPSC” and “cryopreservation” were searched in pubmed.com on June 9th 2024 for publications between 2017 and 2024, yielding 74 articles. Only preclinical animal or first‐in‐human studies that use iPSC‐based cell therapies with therapeutic intent were selected for the analysis, resulting in 12 articles (Table 1).

TABLE 1.

Preclinical animal and human studies using iPSC‐based cell therapies.

Cell type	Disease area	Administration	Species	References
Hepatocytes	Inherited liver disease	Intrasplenic injection	Rat	[21]
Corneal endothelial cell	Corneal edema	Intraocular injection	Monkey	[22]
Dopaminergic neurons	Parkinson's	Intraventricular injection	Rat	[23]
Dopaminergic neurons	Parkinson's	Intraventricular injection	Rat	[24]
Mid‐brain dopamine neurons	Parkinson's	Intraventricular injection	Rat	[25]
Neural progenitor cell	Spinal cord injury	Spinal injection	Pig	[26]
mDA neurons	Parkinson	Intraventricular injection	Mouse	[27]
Retinal pigment epithelial cell	Vision restoration	Intraocular injection	Rat	[28]
Megakaryocytes	Thrombocytopenia	IV transfusion	Mouse	[29]
CAR macrophage	Pancreatic cancer	IV transfusion	Mouse	[30]
NK cells	Cancer	IV transfusion	Mouse	[31]
Neural stem/progenitor cell	Spinal cord injury	Spinal injection	Human	[9]

Note: The table summarizes preclinical studies exploring iPSC‐derived cell therapies across various species and disease models. The listed studies involve different administration routes tailored to the target disease area, with a significant portion of studies administering the cell product directly into the affected organ/tissue (n = 9/12, 75%).

Cryopreservation protocols of the above studies were analyzed, the findings of which are presented in Figure 1 below. As can be seen in the figure, 100% (n = 12/12) of the preclinical iPSC‐based cell therapy candidates use Me₂SO as a cryoprotectant. Uniform freeze rates of 1°C/min are employed by 67% (8/12) of candidates, and this number may be higher in practice as the remaining 33% (4/12) do not disclose the freeze rate used. A 100% (n = 12/12) use a post‐thaw wash to remove Me₂SO, which requires point‐of‐care postprocessing and poses an obstacle to off‐the‐shelf cell therapies. Despite these trends, both the clinical and preclinical mini‐reviews underscore the lack of comprehensive cryopreservation protocols optimized for iPSC‐based cell therapies. This highlights a broader and more pressing need for innovation in cryopreservation methodologies tailored to off‐the‐shelf cell therapies.

FIGURE 1.

Cryopreservation trends in iPSC‐based preclinical studies.

This figure demonstrates the trends in cryopreservation protocols across n = 12 preclinical studies involving the administration of iPSC‐derived cell products (Me₂SO, dimethyl sulfoxide).

The analysis of clinical studies was limited by the scarcity of publicly available manufacturing protocols, likely due to proprietary concerns. Consequently, preclinical protocols may not fully represent clinical practices. However, if preclinical sponsors introduce modifications to cryopreservation processes during the clinical phase, this could pose significant challenges in establishing product comparability. According to the ICH Q5 guideline adopted by the FDA in 2005, any changes in the manufacturing process that could affect product quality may necessitate additional preclinical studies if comparability cannot be established through existing data. ³² Furthermore, post‐thaw washing of cell therapy products to remove Me₂SO is not scalable, as it introduces risks to product quality and patient safety. These risks can be avoided if sponsors chose to explore alternative Me₂SO‐free CPAs that are safe to administer to patients early in the development process.

In practice, however, alternative CPAs yield lower post‐thaw viabilities than Me₂SO when coupled with standard freezing protocols using uniform freezing rates of 1°C/min. Post‐thaw viability is a critical quality attribute (CQA) for cell therapies. This is especially true for cell therapies that treat neurodegenerative diseases, as approximately 90% of the cells die within 24 h of being administered. ³³ , ³⁴ Addressing this challenge requires a deep understanding of both the formulation composition's impact on cell viability and the influence of the freezing and thawing profile. Comprehensive knowledge of these factors is essential for optimizing cryopreservation strategies to maximize cell viability and therapeutic efficacy post‐thaw.

The broader review presented in this paper goes beyond iPSCs and explores cryopreservation strategies across various cell therapy platforms. The focus is on optimizing freezing profiles to enhance the performance of Me₂SO‐free cryopreservation methods, ultimately aiming to improve post‐thaw viability and facilitate the widespread adoption of off‐the‐shelf cell therapies. This approach seeks to identify universal principles by examining the biochemical and physical phenomena that mammalian cells experience during freezing.

Optimizing freeze profiles involves multistep freezing profiles instead of a single‐step uniform freezing rate. The advantage of multistep freezing profiles is that they account for different freeze‐rate‐dependent damage at different stages of the freezing process. This was put forward as the two‐factor hypothesis, ³⁵ demonstrating that slow freezing rates amplify certain forms of cryoinjury, while fast ones amplify other types of cryoinjury.

In this review, the authors propose to categorize the freezing profile into early‐stage, mid‐stage, and late‐stage freeze events. Early‐stage freeze events are supercooling, ice nucleation, and latent heat phase; mid‐stage is cryoconcentration effect and intracellular ice formation; and late‐stage is intracellular and extracellular glass transition and end temperature (see Figure 1). Each stage and event have optimal temperatures and freezing rates, which reduces damage and improves post‐thaw viabilities. This literature review details these phenomena and outlines strategies that different research groups have employed. Finally, a framework for optimizing freezing profiles for Me₂SO‐free cell therapy cryopreservation is presented, using variables and ranges identified by analyzing 21 cell therapy cryopreservation protocols in the literature. While most protocols in the analysis do not emphasize Me₂SO‐free cryopreservation, they serve as a proof‐of‐principle that freeze profile optimization can improve cryopreservation outcomes, which can be applied to Me₂SO‐free cryopreserved cell therapies to facilitate the adoption of off‐the‐shelf cell therapies.

This figure presents an overview of events occurring during the freezing cell suspensions, including CPA incubation, ice nucleation, supercooling, latent heat release, cryoconcentration effect, intracellular ice formation, intracellular glass transition, and extracellular glass transition.

2. METHODOLOGY

To conduct this literature review on cryopreservation of cell therapies, the Abstract Sifter macro ³⁶ was used to search for relevant articles in PubMed. The main search terms used were “cell therapy AND cryopreservation,” resulting in 323 articles retrieved.

To refine the search and focus on freeze optimization, we incorporated additional keywords “freeze” and “optimization,” which the macro identified in the abstracts of 29 articles. Of these, five articles were selected based on the inclusion criteria, which required a disclosed freeze protocol in the study. Additionally, snowballing was employed to identify related articles. This process led to the discovery of 16 more relevant articles.

For the selected articles, freezing profiles were thoroughly examined and broken down into individual steps. Ranges for each step in the freezing profile were recorded. The reported post‐thaw viability for the specific cell line under investigation was also noted. The data collected from the selected articles were then analyzed to identify trends and insights related to the impact of freeze protocols on post‐thaw cell viability.

This methodology allowed for a comprehensive examination of various freeze optimization strategies across different cell therapy platforms, not limited to iPSC‐derived products. By utilizing the Abstract Sifter macro and implementing snowballing, the literature search maximized the inclusion of relevant studies related to cryopreservation and freeze optimization of diverse cell therapies. This review's methodology provides an objective and evidence‐based analysis of freeze protocols, contributing to a broader understanding of best practices in cryopreservation for a wide range of cell therapies.

2.1. Prefreezing treatment: CPA incubation time and temperature

Before cells are cryopreserved, many protocols suggest incubating them in the cryopreservation media so the CPAs permeate the cell membrane. This creates an osmotic balance between the intracellular and extracellular concentration of the CPA, reducing osmotic damage during freezing. When selecting incubation temperature and time, it is also essential to consider potential CPA‐related toxicities at warmer temperatures and longer incubation times, as with Me₂SO‐based cryopreservation media. ³⁷

For this reason, most protocols select an incubation temperature of 4°C (n = 14/20). One group ³⁸ compared incubating at 4 and 22°C and found that the lower temperature resulted in higher post‐thaw viabilities (75% vs. 70%, respectively). Several studies use incubation temperatures of 0°C. ³⁹ , ⁴⁰ , ⁴¹

It is important to consider the freezing point of the cryopreservation media when selecting an incubation temperature. Freezing point depends on the CPA and other components of the freezing media. The concentration of cells may also impact the freezing point, however this requires further investigation as there is a gap in the literature. Incubation should be performed at temperatures above the freezing point, to ensure that incubation does not trigger freezing, as this reduces the control over freezing process. Control strategies for freezing are explored in further sections.

CPA incubation times range between 0 and 60 min; however, the majority of protocols fall in an 8–15 min range (n = 9/13). Three studies compared shorter 10–15 min incubation times to longer 45–60 min incubation times and found no significant impact on post‐thaw viability when performed at 4°C. ²⁴ , ³⁸ , ⁴²

Based on these trends, incubation at 4°C for 8–15 min should allow the CPA to permeate the cell membrane sufficiently without exposing the cells to toxic CPA‐related effects that occur at temperatures above 4°C. This approach is grounded in fundamental cell biology principles that govern all cells with a membrane and is applicable to various cell therapy platforms. By ensuring an osmotic balance between the intracellular cytoplasm and the extracellular cryoprotective solution, this strategy reduces osmotic damage and optimizes cell viability post‐thaw.

2.2. Early‐Stage freeze events: Super cooling, ice nucleation, and latent heat phase

2.2.1. Super cooling and shock freezing

The first event during freezing is supercooling. During freezing, liquids experience supercooling, where the internal temperature drops below the freezing point while remaining in a liquid state due to the external environment's decreasing temperature. ³⁸ An ice nucleation event triggers rapid ice formation throughout the sample, marking the supercooling point. After ice nucleation, the temperature rapidly rises to the freezing point (henceforth referred to as rewarming phase), entering the latent heat phase, where the solution's energy dissipates to complete the transition from liquid to solid state. ⁴³

In uniform cell cryopreservation protocols, ice nucleation is spontaneous and highly variable, resulting in supercooling points ranging between −8 and −15°C. ⁴⁴ , ⁴⁵ Lower supercooling points lead to a more rapid rewarming phase. For instance, 15°C of supercooling can expose cells to a temperature shift at a rate of 900°C/min, potentially damaging the cryopreserved cells due to the formation of sharper and more damaging ice crystals. ³⁵ These thermodynamic phenomena, which influence the freezing of liquids and formation of ice crystals, are universally relevant to a wide range of cell therapy platforms.

The impact of supercooling on post‐thaw viability has been demonstrated in liver spheroids, where manipulating ice nucleation through the use of ice nucleating agents showed significant differences in post‐thaw viability. Samples supercooled by 1°C showed 98% post‐thaw viability, while those supercooled by 7°C had 40% post‐thaw viability. ⁴⁶

Controlling ice nucleation can minimize supercooling. Various strategies are used to control ice nucleation, such as liquid nitrogen spray, ³⁷ , ⁴⁷ , ⁴⁸ ice nucleating agents, ⁴⁹ mechanical agitation, ⁴⁶ , ⁴⁹ electrical current, and shifts in pressure. ⁴⁴ Shock freezing is the most commonly employed method for controlling ice nucleation in cell therapy. ²⁴ , ²⁷ , ⁴⁰ , ⁴¹ Notably, shock freezing is integral in the cryopreservation of Kymriah (Kite Pharma, Gilead Sciences), the first approved CAR‐T cell therapy product. ⁵⁰

2.2.2. Shock freezing

Shock freezing involves rapid reduction of freezer temperature, followed by quick rewarming to a temperature near the solution's freeze point. ³⁹ The rapid temperature reduction creates cold spots on the container's surface where cells are suspended, triggering ice nucleation. As a result, ice nuclei form and spread throughout the sample. Rapid rewarming ensures that the freezer temperature remains near the freezing point, discouraging further temperature changes in the sample until the latent heat phase is complete.

Shock freezing is the method of choice for cell therapy cryopreservation in a good manufacturing practice (GMP) environment due to several advantages. It eliminates the requirement for additional excipients in the final formulation buffer, avoiding the need for toxicology studies or seeking regulatory approval for novel excipients. Many industry‐standard control rate freezers, like CryoMed (Thermo Fisher), support shock freezing, while freezers with pressure changes, mechanical agitation, or electric impulses are less prevalent. Once optimized, shock freezing can be easily programmed and documented, potentially leading to more consistent results and reduced variability in supercooling.

Four variables can be optimized in shock freezing—the rate of freezing, the shock freeze end temperature, the rewarming rate, and the rewarming end temperature. In shock freeze protocols that utilize multiple steps, ³⁸ , ³⁹ , ⁴⁶ , ⁵¹ the authors selected the fastest shock freeze and rewarm rate, lowest shock freeze end temperature, and highest shock freeze rewarm temperature. This selection was made to facilitate comparison with profiles that use only these four variables. The ranges for the four variables are presented in Table 2 below.

TABLE 2.

Shock freeze parameters and their range and effect from literature.

Shock freeze parameter	Range	Effect/conclusion
Shock freeze cooling rate	25 to 300°C/min	Rates >25°C/min and < 300°C/min lead to post‐thaw viabilities >80%
Shock freeze temperature	−80 to −30°C	Temperatures > −40°C lead to higher post‐thaw viabilities
Shock freeze rewarming rate	2.5 to 150°C/min	No direct comparisons available
Shock freeze rewarming temperature	−35 to −5°C	No direct comparisons available

Note: This table represents the shock freeze variables identified in literature, along with their associated effects. References [38, 39, 40, 41, 46, 51, 52].

It is worth noting that most of the analyzed studies use different cell lines and cryopreservation media, so conclusions should be seen as guidelines for optimization, as the effects are likely specific to the cell line and media used. However, shock freezing is likely a universal variable, as it impacts supercooling, a thermodynamic phenomenon, rather than a cell type‐specific factor.

Ranges for shock freezing rates are 25–300°C/min. Shock freezing rates above 35°C/min appear to be favorable, as they tend to have higher post‐thaw viabilities ³⁹ , ⁴⁰ , ⁴¹ , ⁴⁶ , ⁵¹ compared with studies using rates below 35°C/min. ²⁴ , ²⁷ Thus, faster shock freezing rates above 35°C/min are likely favorable, as they more reliably create cold spots on the cryogenic container, which trigger ice nucleation.

Regarding shock freeze temperature, studies show no significant impact on post‐thaw viability when using temperatures above −40°C. ²⁴ , ²⁷ , ⁴⁶ However, nearly a two‐fold increase in post‐thaw viability has been demonstrated when using temperatures below −40°C, especially at −70°C. ⁵³ Lower shock freeze temperatures potentially provide more effective ice nucleation control than warmer temperatures.

Studies used a range rewarming rates, from 2.5°C/min ⁴⁶ , ⁵¹ to 150°C/min. ³⁸ At this stage, comparing single factors between studies becomes challenging, as only some use the same combination of shock freezing factors. There is likely a dependence between variables. For instance, studies using the lowest rewarm rates of 2.5°C/min also use warmer shock freeze temperatures of −33°C. ³⁸ , ⁴⁶ , ⁵¹ Thus, when using lower shock freeze temperatures, it may be favorable to use faster rewarm rates to avoid a disparity in the freezer temperature and the temperature of the sample as it enters the latent heat phase.

No study compared different rewarming temperatures. Several use rewarming temperatures closer to the freezing point of the solution, between −5 and −12°C, ²⁴ , ²⁷ , ³⁸ , ⁴¹ while others use lower temperatures ranging from −20 to −35°C. ³⁹ , ⁴⁰ , ⁴⁶ , ⁵¹ , ⁵² However, a rewarm end temperature near the freezing point may be favorable, as the amount of control over the latent heat phase reduces if there is a disparity between the temperature of the freezer and the latent heat phase. ⁵³

In summary, shock freeze variables have compounding variables. Using faster shock freezing rates above −35°C/min, reducing the freezer to a temperature of −40°C or lower appears favorable based on the above research. However, those may need to be combined with faster rewarm rates to a temperature close to the solution's freezing point. This avoids disparities between freezer and sample temperatures during the latent heat phase. Not confined to specific cell types, these thermodynamic principles are applicable to a broad range of cell therapy scenarios. The duration of the latent heat phase is another variable that may be optimized during the early freezing stage.

2.2.3. Latent heat phase

Controlling the latent heat phase can potentially have a significant impact on post‐thaw viability. The range for the latent heat phase is 0–30 min; however, the most common duration is 10–15 min. ³⁷ , ⁴⁶ , ⁴⁸ , ⁵² , ⁵⁴ A hold time of 30 min led to a 20% improvement in post‐thaw viability when compared with no hold time. ⁵⁵

Only one study systematically tested different durations of the latent heat phase. ⁵² This study demonstrated that a 2 min latent heat phase time can improve post‐thaw viability by 20% compared with no holding time. It also demonstrated a curvature effect in the latent heat phase, with a 10 min holding time being ideal and 20 min leading to reduced viability. ⁵² This suggests that both short and long holds are superior to no hold time, however that shorter holds outperform longer holds. The authors recommend testing a range between 2 and 15 min when optimizing this variable. While specific cell types may require tailored hold times for optimal results, testing within the 2 to 15‐minute range should be broadly applicable across various cell types relevant to therapy.

2.2.4. Mid‐stage freeze events: Cryoconcentration effect and intracellular ice formation

The mid‐stage of freezing results in two types of cryoinjury: cryoconcentration effect and intracellular ice formation. The cryoconcentration effect occurs when a solution freezes, causing solutes to be expelled by the ice front into the unfrozen liquid, increasing solute concentration. ⁵⁶ Slower freezing rates amplify this effect, while faster rates reduce it. On the other hand, intracellular ice formation is amplified by faster freezing rates and reduced by slower rates. This effect involves the formation of ice crystals inside the cell. These two phenomena are potentially interconnected and were initially termed the two‐factor hypothesis. ³⁵

The cryoconcentration effect creates a concentration gradient; regions freezing first having lower solute concentrations than those freezing later. ⁵⁷ Slower freezing rates intensify this gradient, potentially impacting cell quality due to unequal CPA distribution. Cells in areas with lower CPA concentration may not benefit from the CPAs' cryoprotective effects, whereas higher CPA concentrations may harm cells in the areas with higher CPA concentration due to solute toxicity. Slower freezing rates can lead to cellular dehydration, which is beneficial in reducing intracellular ice formation, but excessive dehydration may cause cell rupture during thawing due to osmotic shock. ³⁵ , ³⁸

In contrast, intracellular ice formation is more pronounced with faster freezing rates. ⁵⁸ Rapid freezing prevents sufficient cellular dehydration, increasing intracellular water and creating more intracellular ice during freezing, which can damage the cell. Faster freezing rates also lead to sharper ice crystals and a tighter ice matrix, potentially harming cells. ⁵⁹

The freezing rate must be optimized to address both phenomena and balance the two sources of damage. There is no consensus in literature regarding which temperature zones these effects are more prevalent. The temperature range between −0 and −10°C has been defined as the zone in which cryoconcentration is more prevalent and −50 to −60°C as the zone in which intracellular ice formation is more prevalent. ⁶⁰ This was determined by freezing induced pluripotent stem cells (iPSCs) at varying uniform rates, observing them using cryomicroscopy, and analyzing the data using a computer algorithm.

Several studies investigate the impact of freezing rates on post‐thaw viability, testing freezing rates of 0.1–10°C/min. The majority of protocols analyzed (n = 15/21) use multistep freezing profiles; They use a relatively slower freezing rate in the temperature range of −4–−60°C. ²⁴ , ⁴⁸ , ⁶⁰ , ⁶¹ , ⁶² This range will be considered the intermediate freeze rate, characterized as the portion of the freeze profile between the end of the latent heat phase and the intracellular ice formation in Figure 2.

FIGURE 2.

Overview of the events occurring during freezing.

The highest post‐thaw viabilities are achieved using 1°C/min. ²⁴ , ⁶⁰ , ⁶¹ , ⁶³ Freezing rates below 1°C/min yield relatively lower post‐thaw viabilities regardless of cell type. ²⁴ , ⁵⁵ , ⁶¹ , ⁶⁴ , ⁶⁵

In the range of 5–10°C/min, the study trend leans toward lower post‐thaw viabilities at faster freezing rates. ⁴² , ⁴⁸ Cryopreserved NDHF cells showed no significant difference between 5 and 10°C/min; however, both rates had lower post‐thaw viability than at a freezing rate of 1°C/min. ⁶³ Rates of 1°C/min have long been used because they allow the intracellular water to escape the cells through CPA‐induced osmosis, which reduces the incidence of intracellular ice crystal formation.

Optimizing the freezing rate is crucial to balancing these damaging effects. Studies have shown that post‐thaw viability is highest at 1°C/min, with rates below this yielding relatively lower viabilities regardless of cell type. ²⁴ , ⁵⁵ , ⁶⁰ , ⁶¹ , ⁶⁵ In the 5–10°C/min range, post‐thaw viabilities generally decrease with faster freezing rates. ⁴² , ⁴⁸ , ⁶⁰ , ⁶³

However, freezing rate responses vary depending on the cell type and maturation stage. Different cell types may exhibit unique sensitivities to freezing rates. For example, iPSC‐derived sensory neurons showed varying responses to freezing rates at different maturation stages. ³⁷ Based on findings of the above protocols, the authors recommend testing freezing rates in the range of 1–5°C/min for this stage of the freezing profile, as they are likely applicable to a wide range of cell types.

2.3. Late‐stage freezing events—Intra and extracellular glass transition and end temperature

Late‐stage freezing events are intracellular and extracellular glass transitions. Glass transition, also known as vitrification, is a phenomenon solids experience at low temperatures. In glass transition, a flexible solid transitions to an amorphous, glass‐like state. ⁶⁵ For Me₂SO‐based cryopreservation media, this occurs at approximately −120°C for extracellular glass transition. ⁶⁶ , ⁶⁷ Before the transition, the ice crystal lattice is flexible and can shift its orientation. However, after the transition, the crystal lattice is fixed and immobile. This slows the metabolism of cryopreserved cells to a minimum, as only small molecules, such as ions, can pass through the ice matrix. The cells are at their most protected during this phase, which is why after the cryoprotocol is completed, the cells are transferred to secondary storage, typically a liquid Nitrogen dewar, and placed in vapor phase liquid Nitrogen at approximately −135°C. ⁶⁷

Glass transition is a reversible process, and if the temperature rises above the glass transition point, devitrification can occur. This is harmful to cells, as shifts in the ice lattice can disrupt their integrity. ⁶⁷ As such, the end temperature of the cryoprotocol is crucial, as the cells potentially rewarm while being transported from the initial controlled rate freezer (CRF) to the secondary cold storage container. Lower‐end temperatures mitigate this risk, as elaborated below.

It is hypothesized that the intracellular ice undergoes glass transition before the extracellular compartment, at approximately −50 to −60°C. Intracellular ice transition is established and accepted in prokaryotes. ⁶⁸ , ⁶⁹ Moreover, there is compelling evidence of glass transition in Jurkat cells, a model T‐Cell line. ⁶⁶ Given that intracellular glass transition has been observed in both prokaryotes and eukaryotes, it is likely a universal phenomenon impacting virtually all cell types. This broad applicability underscores its relevance for a wide range of cell therapy platforms. Intracellular glass transition is attributed to the intracellular compartment having a higher protein density than the extracellular solution, which is further amplified by cellular dehydration during freezing. This increases the viscosity of the intracellular solution, causing it to undergo glass transition at a relatively higher temperature than the extracellular compartment. It is further hypothesized that the cells are less susceptible to freeze‐rate‐dependent cryoinjuries once the intracellular glass transition is complete, allowing faster freezing rates. ⁶⁶

Interestingly, it is common practice in cell therapy cryopreservation protocols to increase the freezing rate to 10°C/min after a temperature of −40 to −60°C is reached. ²⁴ , ⁴⁰ , ⁴⁵ , ⁴⁷ , ⁶² As mentioned previously, freezing rates of 10°C/min at earlier stages of the freezing process can have a potentially damaging effect. However, protocols that increase the freezing rate after potential intracellular glass transition has occurred still report viabilities in the range of 80%–95% across various cell types. ⁴¹ , ⁴⁷ , ⁴⁸ , ⁶² The authors define this as the final freezing rate, seen in Figure 2 as the rate between the end of the intracellular glass transition and the final freezing temperature.

One article stands out in particular, which found the increase in freezing rate further in the freezing process to be beneficial on post‐thaw viability. A cryoprotocol that increases the freezing rate to 5°C/min after reaching −30°C had a nearly two‐fold increase in rat mesenchymal stem cell (MSC) post‐thaw viability (80% vs. 40%) compared with a uniform 1°C/min freezing profile. ⁶⁵ While −30°C is relatively higher than the hypothesized intracellular glass transition temperature range of −50 to −60°C, it can be explained by the intracellular glass transition temperature being cell type and cryopreservation media dependent. These findings warrant further investigation.

Lastly, the freezer end temperature can also be optimized. As mentioned, the final temperature ensures a devitrification event does not occur during transportation of the cells from the initial CRF to the secondary cold storage container. This is a critical part of the cryoprotocol, as cells can rewarm at approximately 5 to 10°C/min. ⁶⁴ Samples could significantly rewarm and devitrify depending on how far the initial CRF is located from the liquid Nitrogen dewar.

Aside from damage by the shifting ice matrix, ice recrystallization can also occur during warming. Ice recrystallization is when thawed liquid from the extracellular solution enters cells by osmosis to rehydrate and restore the osmotic balance. As this liquid comes in contact with intracellular ice crystals, it can refreeze, causing the ice crystals to grow and expand. This is potentially harmful to the cell, as the increased volume of ice can cause it to rupture. ⁷⁰ Thus it is vital to have a low‐end temperature of the freezing profile to ensure the cells do not rewarm during transportation from the CRF to the liquid nitrogen dewar.

Temperatures employed in the cryoprotocols part of this analysis range between −70 and −196°C, however −80°C (n = 8/29) is the most common end temperature, closely followed by −100°C (n = 7/29). End temperatures of −100–−160°C were compared with no significant impact on post‐thaw viability. ⁴⁶ This indicates that while temperatures below a certain point do not provide additional benefits, they are potentially not harmful. This may not necessarily be true for end temperatures below a certain threshold. A freezer end temperature of −70°C has led to a post‐thaw viability of 60% in iPSC‐derived dopaminergic neurons (DA). ⁵⁴ While other factors may have influenced this, it is a potential indication that end temperatures below a specific range may have harmful effects.

3. CONCLUSION

In conclusion, off‐the‐shelf iPSC‐based cell therapies hold immense promise for regenerative medicine, with clinical candidates in pipelines worldwide generating excitement. However, cryopreservation poses a significant barrier to unleashing the full potential of these therapies. CPAs, like Me₂SO, pose potential risks to patients when administering post‐thaw. Post‐thaw removal of Me₂SO, while commonly employed, introduces risks and potential damage to the final cell therapy product, raising concerns about its safety and quality.

To address these challenges, researchers are exploring alternative Me₂SO‐free CPAs. Their current suboptimal results may be enhanced by optimizing the freezing profile. Freezing rate optimization, particularly shock freezing, has been shown to control ice nucleation and reduce the extent of supercooling, thereby reducing variability in cryopreservation protocol and mitigating potential damage to the cells. Controlling the duration of the latent heat phase may also be a potential avenue to increasing post‐thaw viability.

By manipulating the freezing rate at different stages of the freezing profile, intracellular ice formation and cryoconcentration effect can be balanced. Both damage cells through different mechanisms. However, cryoconcentration is mitigated by fast freezing rates, while slow freeze rates mitigate intracellular ice formations. Additionally, understanding the impact of glass transition at different freezing stages can further optimize post‐thaw viabilities. Investigating the hypothesized intracellular glass transition may yield additional benefits to post‐thaw viability. Finally, ensuring a sufficiently low end temperature of the freezing profile may protect cells from damage through rewarming, such as ice recrystallization. These variables can be used to optimize freezing profiles systematically.

This literature review highlighted the importance of a comprehensive approach to cryopreservation and proposed a framework through which cryoprotocols can be optimized. The strength of the proposed framework lies in its reliance on universal thermodynamic and biochemical principles, such as osmotic cell membrane dynamics and the behavior of liquids as they freeze and transition to solid states. These principles are applicable across a wide range of cell therapy platforms. The variables and recommended ranges displayed in Table 3 are a starting point for researchers to optimize Me₂SO‐free cryoprotocols tailored to their cell type and freezing media. Researchers can develop safer and more efficient cryopreservation protocols for cell therapies through rigorous experimentation and analysis. The proposed framework for optimizing freeze profiles for Me₂SO‐free cell therapy cryopreservation serves as a guide to advance the field and realize the full potential of cell therapies in transforming disease treatment.

TABLE 3.

Recommended ranges for freeze profile variables.

Freeze Profile Variable	Recommended Range
CPA incubation temperature	4°C
CPA incubation time	8 to 15 min
Shock freezing cooling rate	35 to 200°C/min
Shock freezing temperature	−80 to −40°C
Shock freezing rewarming rate	20 to 150°C/min
Shock freezing rewarming temperature	−15 to −5°C
Intermediate freezing rate	1 to 5°C/min
Intracellular glass transition temperature	−40 to −60°C
Final freezing rate	1 to 10°C/min
Final temperature	−80 to −150°C

Note: This table indicates the recommended ranges for the freeze profile variables identified in the review based on analyzing 21 cryopreservation protocols for cell therapies the literature.

The authors plan to leverage this framework to optimize a Me₂SO‐free cryopreservation protocol for a cell therapy. The variables identified in the literature and presented in Figure 2 are going to be evaluated through the design of experiment (DOE) methodology. The ranges presented in Table 3 establish the boundaries of the design space. Overcoming the cryopreservation barrier will pave the way for broader access to these innovative cell therapies and improved patient outcomes.

AUTHOR CONTRIBUTIONS

Michael Dobruskin: Conceptualization; methodology; writing – review and editing; project administration; writing – original draft; investigation. Geoffrey Toner: Resources; supervision; project administration; funding acquisition; validation. Ronald Kander: Supervision; resources; validation; project administration.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

Dobruskin M, Toner G, Kander R. Optimizing cryopreservation strategies for scalable cell therapies: A comprehensive review with insights from iPSC‐derived therapies. Biotechnol. Prog. 2024;40(6):e3504. doi: 10.1002/btpr.3504

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no new data were created or analyzed in this study.