Lyophilization Cycle Optimization of Cell-Derived Products

The ability of a biological formulation to maintain its physical, chemical, and therapeutic properties is often put to the test during transportation and storage. Because many cell-derived therapeutics are administered by infusion or injection, lyophilization is a common method used to keep the product viable. There are, however, many factors that can influence the behavior of a drug within a batch of vials, and issues with maintaining biological activity and stability during formulation. This article provides an overview of the factors that can influence protein behavior during lyophilization of a pharmaceutical product and the ways in which manufacturers can reduce processing costs and timelines.

Special considerations for cell-derived products
Because proteins are prone to chemical and physical degradation, aggregation events and potency loss of the product are common challenges associated with the lyophilization of biologics. “While lyophilization is done to achieve long-term stability, the process itself can be quite destabilizing for the molecule,” says Mathew Cherian, PhD, director and senior fellow, pharmaceutical development at Hospira. “Finding the optimal cryoprotectant levels; the optimal rate and extent of freezing; the optimal pressure during primary and secondary drying; and measuring the extent of primary and secondary drying” are crucial to ensuring a good-quality final product. Changes to the product, such as freezing, concentration and pH shifts, decreasing temperatures, and desorption can cause irreversible denaturation, adds Edward H. Trappler, president of Lyophilization Technology.

Protein stress. Although primary freezing is believed to cause the most stress to a biologic product, degradation and the formation of aggregates can occur at any stage in the lyophilization process. According to Martin Gonzalez, senior group leader, One-2-One R&D at Hospira, primary freezing can cause pH shifts, super-concentration of protein species, and exclusion of solvents, which can result in problems in protein thermodynamic stability. This instability can cause protein unfolding, denaturation, or aggregation. Specifically, says Cherian, the formation of a solid-liquid interface due to the advancing freezing front can generate aggregates.

Dehydration stresses during drying can also cause problems. An emerging theory is that drying increases surface area, and as a result, the protein molecules on the surface of the dried solute (i.e., those that are under mechanical stress within the solid matrix) are at a “greater risk for aggregation or other negative consequences upon reconstitution,” notes Jim Searles, PhD, technical fellow, global manufacturing science and technology, Hospira. Proteins in dried products, therefore, are not as protected by the amorphous environment of the inner matrix. According to Kevin Ward, PhD, director of R&D, Biopharma Technology, the risks of protein instability can be minimized by “intelligent formulation design and the use of excipients with specific cryo- and lyo-protective qualities.” Searles of Hospira points out that the “addition of a surfactant to the formulation can mitigate interfacial damage, and emerging science is showing that post-drying, super-Tg [glass-transition temperature] annealing can allow structural relaxations in the lyophilized matrix that result in stability improvements.”

To ensure optimal storage conditions and preservation of structure for a freeze-dried biologic, molecular mobility must be minimized and suitable excipients selected, according to Trappler. A sound formulation is the first step, and then the formulation must be protected from temperature cycling, light, and oxygen. Vials should be stored upright, packed with vibration-absorbent secondary packaging, and handled with care during shipping, notes Cherian.

Other concerns lie in the post-reconstitution risks that exist with biologics. “Aside from the all-important retention of activity, as clinical demand for high-concentration formulations (> 100 mg/mL) continues, the formulation scientist must be cognizant of increased aggregation propensity as well as managing viscosity after reconstitution,” says John G. Augustine, PhD, principal development scientist, analytical and formulation development at CMC Biologics.

The most important step to control instability during storage is to carry out thorough preformulation work as early as possible during the drug-development process. This information can help inform the downstream processing cycle, aid future purification efforts, and help ensure that the proper excipients to minimize degradation and aggregation have been selected, says Augustine.

Scale-up: Moving from the laboratory to a commercial facility
One of the common mistakes in lyophilization scale-up is to assume that a laboratory-scale freeze dryer containing a few hundred vials behaves the same way as a production-scale freeze dryer containing thousands of vials (1). Commercial equipment capabilities can vary widely depending on their design specifications and their age. Equipment differences and batch uniformity under new temperature profiles must be considered during processing when switching from lab lyophilization experiments to commercial lyophilization systems, says Trappler.

Equipment differences can create drying rate differences and variations in the final water content in the products. “Industrial freeze dryers have larger chambers with larger shelf dimensions,” says Ward. “This leads to a reduction in radiative heating and often an increase in intra-shelf temperature profiles.”

As a result, sublimation rate capability of the production equipment should be carefully tested, note representatives from Hospira. Product quality testing on at least one full shelf of vials should be done on products sampled from the edge as well as the interior of the shelf on the drying unit, they note. Additionally, vial handling, washing, and depyrogenation characteristic of full-scale manufacturing can weaken glass, making vials susceptible to potential vial breakage.

Cost-saving effortsShortened lyophilization times. Shortening lyophilization times is economical and reduces cost-per-unit. “Reducing the cycle time reduces the energy costs per run, and also increases the potential throughput of a facility by maximizing the number of runs that can be processed in a similar timeframe,” comments Ward.

Oftentimes, altering the design space can facilitate shorter drying times and help with product output. This type of process alteration, however, should occur while the product is still in the lab, observes Hospira’s Gonzalez. “It’s better to achieve cycle time reduction before going into full commercial operation, perhaps by spending a little bit more time developing a robust cycle that yields more in the long run,” he says.

The design space influences shelf temperature, chamber pressure, and hold time for each step of the lyophilization cycle, notes Mark Nachtigall, PhD, scientist, global manufacturing science and technology, Hospira. The ideal design space-one that will generate optimal primary drying times-will allow a product to remain as warm as possible without collapsing. A shorter drying time will also rely on the sublimation rate capability of the lyophilizer, says Nachtigall. “Experimental exploration of these parameters in combination with statistical evaluation of the effects of changes in the parameters will help determine the optimal process conditions,” he says.

According to Ward, the single biggest influence on drying time is the critical temperature of the formulation itself. As a result, designing a formulation with a high critical temperature is the best way to obtain better lyophilization cycle times. Ping Ma, PhD, senior group leader, global pharmaceutical R&D at Hospira, however, says that increasing shelf temperature would be the most efficient way to reduce cycle time. Robert Stoner, associate research scientist, Global Pharmaceutical R&D at Hospira agrees, saying, “Designing cycles that run at warmer shelf temperatures and higher pressures reduces equipment stress with greater facility/utilities savings.”

Another option would be to lyophilize a more concentrated solution (i.e., increasing API concentration to reduce the amount of solvent), because less water to sublimate is correlated with shorter lyophilization cycle times, note Stoner and Ma. Manufacturers can also reduce lyophilization cycle time by reducing freezing hold time, says Ma.

Related cost-saving measures. Other ways to create shorter lyophilization cycle times include increasing the vial size. Using taller vials will allow more vials to fit into the lyophilization chamber per batch, notes Stoner. This method, however, may have to be coupled with longer drying times, as there would be smaller shelf contact area and taller lyophilization cakes. Lastly, reformulating with fewer excipients can reduce cost as well, suggests Gonzalez.

Influences on drying timeForm and function. The physical properties set during the freezing phase can influence drying time greatly, says Nachtigall. These properties include the size of the crystals in the matrix and crystal form of the frozen product. In general, slower cooling produces larger, better networked crystals, says Ward, and the larger pores and more “open structure” that is produced present less of an impedance to vapor migration during drying. Primary drying and sublimation is accelerated with larger crystal sizes. Ward warns, however, that control is needed to balance safety and efficacy, as some biologic products are sensitive to slow cooling. Additionally, total process time may be adversely affected because of larger crystal sizes, notes Trappler. Larger crystals represent a reduction in the surface area and a decreased desorption rate, which could mean the product would require longer secondary drying times.

A primary drying rate can be significantly influenced by an additional annealing step, which can improve the appearance and homogeneity of the cake of the final drug product and ameliorate the non-uniform nature of freezing and drying. A post-freezing annealing process has been shown to increase the rate of primary drying (1). “This is because annealing eliminates the smallest ice crystals through Ostwald ripening,” asserts Nachtigall. “These ice crystals leave channels as they dry that water vapor lower in the cake will travel through; larger channels equal less resistance, and therefore, faster drying times.” Crystal forms matter as well, and annealing can promote crystallization for those products predispositioned to crystallize, he adds. “A crystalline product matrix typically has a higher collapse temperature than an amorphous one,” which allows for higher drying temperatures, and subsequently, faster drying times.

Controlled nucleation. While obtaining larger ice crystals and a reduced surface area through the use of controlled nucleation has been shown to improve reconstitution times and reduce the primary drying time for concentrated proteins and antibodies, there is still some uncertainty as to whether controlled nucleation is truly a benefit to biologics, according to Trappler. He says, “Preliminary data show there is no detrimental effect in the initial critical quality attributes (CQA) of the protein preparation,” but that the “complete impact on CQA can only be confirmed from the results of long-term stability tests.” Even though controlled nucleation offers some process benefits, investigators may be able to better control for differences in freezing and drying with better glass vial “bottom geometry,” says Trappler. “Even with controlled nucleation, there are differences in ice crystal growth as well as consistency during drying due to vial bottom contour.”

With controlled ice nucleation-formed cakes, the rate of sublimation is greater than in cakes formed as a result of uncontrolled freezing. “From a clinical standpoint, for highly concentrated protein formulations, a reduction in reconstitution times can be observed in cakes produced under controlled nucleation,” says Augustine.

Regardless of what methods are used to control for differences in freezing behavior and prevent non-uniform drying of batches of biologics, there will always be some variation in product heterogeneity, says Ward. These differences can be due to differences in temperature control across a shelf or variations in vapor flow across a chamber, he points out.

Know your characters. Process engineering for lyophilization relies on adequate characterization of a formulation. To figure out the parameters for adequate solidification and the threshold temperature required to maintain product structure, a suite of tests and preformulation tests, such as studies of solubility, pH effect, stability in solution, and low-temperature analysis should be conducted, says Trappler. “Use of freeze-drying microscopy-an insightful method-coupled with low-temperature differential scanning calorimetry and electrokinetic or electrical resistance measurement” is essential to the formulation characterization process, he adds. Ward echoes these sentiments, saying that freeze-drying microscopy is the only method for determining collapse temperature, and thermal analysis can identify glass transitions and other events such as eutectic melting and crystallization. “Without this information,” he says, “cycle development is very much a trial-and-error process.”

Hospira’s Ma points out that characterization tools “enable [drug developers] to better understand each composition and its physicochemical interactions in formulation, optimize formulation process parameters, evaluate overall risk assessment, and establish in-process and final release specifications” for each formulation.

Future focus: Intelligent formulation design
Quality by design (QbD) is built into a well-characterized formulation. Thus, a product in development must be tested for collapse temperatures, sensitivity to freezing rates, and stability against excess moisture, among other factors, says Gonzalez. “For lyophilization, this [the QbD approach] usually means the designation of freezing rates, temperatures, pressures, and hold times as critical process parameters,” he says. “With so many parameters, one must fully leverage process understanding to decide which combinations of these to test in laboratory runs. It is also important to make use of process analytical technology, such as instruments that indicate completion of primary and secondary drying.”

Advances in protein characterization methods may allow greater assessment and insight in the development of formulations and processes for biologicals, says Trappler. He lists light scattering, size-exclusion high-performance liquid chromatography, dynamic light scattering, right-angle light scattering, infrared, and nuclear magnetic resonance as important analytical tools of the trade.

For biologics, the development of standard platform technologies may help companies optimize formulations and processing procedures. Companies are now developing such platforms for evaluating molecules with a risk-based approach, says Lisa Cherry, senior group leader, global pharmaceutical R&D at Hospira. Molecule formulation development and manufacturability are increasingly being driven by prior knowledge gleaned from similar molecules. “For formulation development of lyophilized biologics, there are relatively few combinations of excipients in currently licensed products,” notes Cherry. Starting within this formulation design space will offer a high probability of success, she says. “While formulations are being screened, one should use a conservative (and therefore long) lyophilization cycle, which can then be optimized once the final formulation candidate(s) are selected.”

Reference
1. B.S. Chang and S.Y. Patro, “Freeze-drying Process Development for Protein Pharmaceuticals,” in Lyophilization of Biopharmaceuticals, H.R. Constantino and M.J. Pikal, Eds. (2004), pp. 113-138. 

Article DetailsPharmaceutical Technology
Vol. 39, No. 3
Pages: 30-35
Citation: When referring to this article, please cite it as R. Hernandez, “Lyophilization Cycle Optimization of Cell-Derived Products,” Pharmaceutical Technology39 (3) 2015.