Abstract
As the complexities of the pharmaceuticals needed to prevail over serious diseases continue to grow, the need for technologies to enable their efficient storage and delivery are as essential as ever. Lately, drugs such as vaccines, proteins, and stem cells are increasingly requiring frozen storage to maintain their efficacies before use. Notably, the advent of cellular therapy products has invariably elevated the need for cryopreservation and frozen storage of cellular starting materials, intermediates, and/or final product. The container closure integrity (CCI)—which is a major requirement for aseptic or sterile packaging systems—at these extremely low temperatures has not been fully understood. For vial-based systems particularly, the commonly used rubber stoppers are expected to lose their elastic properties below their glass transition temperatures, suggesting a potential temporary loss of sealability under frozen storage conditions and posing a risk to CCI. The measurement of the CCI at these conditions such as −80°C is therefore critical; a process that can be very challenging. Previous works had explored the use of Oxygen Headspace Analysis to measure CCI at low temperatures. Here, we present the evaluation of the CCI of rubber-stoppered aluminosilicate glass vials (Valor®) and plastic vials (Crystal Zenith®) using the helium leak CCI test method at −80°C, with correlation to residual seal force (RSF). The results and their implications are then discussed with regard to the suitability of certain packaging components as frozen storage container closure systems.
- Frozen storage
- Helium leak
- Container closure integrity test
- Corning Valor® glass
- Crystal Zenith® plastic
- Residual seal force
1. Introduction
The design and development of a robust sterile packaging system dictates the prevention of product loss and the maintenance of sterility throughout its life cycle, ensuring product quality and patient safety (1). Container closure integrity (CCI) is one of such requirements, and for vial-based systems consisting of a glass or plastic vial, a rubber stopper, and an aluminum crimp seal, this implies creating, ensuring, and maintaining an adequate seal with acceptable CCI. The maintenance of CCI at all conditions cannot be overemphasized. Drug products can be sensitive to varying environmental factors, such as oxygen, temperature, light, and pressure (2), to which they can be exposed when CCI is lost.
Lately, low-temperature storage (e.g., −80°C or on dry ice) therapeutic drugs or vaccines have become more frequent (3) and are expected to increase with the emergence of cellular and personalized therapy products (4). Furthermore, the increasing complexities of typical drugs such as vaccines to address progressively complex diseases are dictating colder than normal storage conditions to maintain their efficacies before use. Indeed, the multiple efforts being deployed to address the current COVID-19 pandemic will require vaccines that have to be stored in low-temperature conditions (5).
Beyond the typical after-effects of CCI described earlier, drug products that require low-temperature storage present an additional unique set of challenges. The available materials that can withstand storage and shipment at such conditions are limited. Also, the halobutyl-based rubber stoppers (terpolymer of isobutylene, isoprene, and brominated or chlorinated isoprene) that are frequently used in pharmaceutical packaging lose their elastic properties in these conditions, as these temperatures are below their glass transition temperature (Tg, usually around −65°C), posing a sealability risk (6, 7). Furthermore, the temporary loss of CCI under these frozen conditions could lead to an overpressure in the vial, which can have extensive and sometimes potentially dangerous implications (6, 7). For example, Zuleger et al. (7) reported the presence of overpressures in vials storing live viral vaccine at −80°C, an incidence that was previously absent in other viral products at −20°C. Finally, the temporary loss in closure integrity (and potential leaks in or out of vials) may go unnoticed, as the stopper regains its elastic properties and thus reseals the vial once the temperature is brought to room temperature for use.
Therefore, it is critical to identify a way to measure the CCI at these frozen temperatures and to determine the applicability of certain primary packaging components. Generally, the CCI test (CCIT) methods described in the literature (8⇓⇓–11) are not useful to identify failures in CCI at −80°C. Because they quantify the leak present at the time of measurement, it is impractical to use those methods at low temperatures. Thus, vial systems are brought to room temperature, at which point the temporary leak cannot be determined as resealing would have occurred. The excellent work by Zuleger et al. (7) utilized the headspace analysis method to determine temporary leaks at these low temperatures.
Helium leak rate, a different and excellent deterministic method endorsed in USP <1207> (8) also enables the quantifiable detection of leaks. Recently, a proprietary equipment that allows the use of modified commercial tracer gas (helium) leak detection for real-time CCI measurement and assessment of seal integrity at very low temperatures was developed (12). Previously, Nieto and colleagues (13) also used a modified helium leakage test to directly assess CCI at frozen conditions. Their study revealed the helium leakage method to be more sensitive in detecting the CCI effect than gas headspace, enabling selection of an optimum container closure system. Other work by Nieto and colleagues (14) and Mehta et al. (15) also revealed the sensitivity of CCI to other factors such as manufacturing processes and transportation. This is outside the scope of this work.
Here, we present the low-temperature CCI data for vials that were capped under various sealing conditions. The sealing stress resulting from the force exerted by the compressed elastomeric closure flange on a vial land sealing surface after the application of an aluminum seal (crimping), called the residual seal force (RSF), is an indirect estimation of the elastomeric closure compression and ultimately, the sealability (16⇓⇓–19). Thus, container closure systems were created with targeted RSF values to investigate the correlation between the maintenance of CCI during frozen storage and the quality of the vial sealing process as measured by RSF. This is important because CCI can depend critically on the quality of the vial sealing (7), meaning that a holistic approach needs to be taken to ensure CCI: It is not just the performance of the primary packaging components, but quantitative data also needs to be generated on the vial sealing process.
The RSF values were subsequently correlated with the helium leak rate at −80°C for both glass and plastic container systems
2. Materials and Methods
This section provides details on the materials and various methods used in the study.
2.1. Materials
The components and targeted conditions used in this study are shown in Table I and Figure 1. These components are standard parts within the pharmaceutical industry and are available commercially.
Pictures of the packaging components used in the study.
Packaging Components, Sample Size, and Targeted Test Conditions Used in the Study
2.2. Methods
The procedures, equipment, and methods employed in the creation and evaluation of the samples are described below.
2.2.1. Sample Creation:
All the samples were prepared at time, t = 0 for each targeted RSF value. The RSF values for each component configuration were previously determined through capping optimizations. These capping optimizations are nontrivial as they are responsible for generating the desired seal quality. Briefly, compression analysis was first performed to determine the optimized target compression for the stopper in the study, which can range between 14% and 20% of a stopper noncompressed thickness depending on the durometer and compression set of the elastomer. Here, we determined different compressions the stopper is subjected to during the sealing cycle. This analysis requires reliable, repeatable equipment—we utilized a Mitutoyo dial indicator integrated with a surface plate mount. The difference between the measured combined height of the components before and after sealing were then used to calculate the percent stopper compression. Although playing a secondary role to compression, esthetic appeal is nonetheless an important criterion when developing proper capper settings for a specific package. As a result, each package was sealed at the target compression with acceptable esthetics. Once an acceptable stopper compression, and by extension its capping parameters, was identified, appropriate sample sizes were created using these conditions and the RSF was measured. Capping was done using a small-scale Genesis RW-50 capper at Genesis Packaging Technologies. The Genesis RW-50 capper simulates the sealing capabilities of the Genesis RW-600 model used in production. Subsequently, ranges of RSF values were obtained by varying the levels of stopper compression and capping at those parameters.
Once capped, the RSF value of all component configurations (with the exception of the initial time samples) was measured twice for every condition (after capping and before helium leak rate measurement). All tested samples were filled with only helium gas. Briefly, the vials were filled with helium inverted on a fixture for a specified flow rate and time. Due to size differences, the 2 mL vials were filled at a flow rate of at least 200 cc/min for a minimum of 15 s, whereas the 10 mL vials were filled at the same flow rate for a minimum of 1 min. Subsequently, the vials were then either analyzed for helium leak at t = 0 or placed on stability at −80°C for timed studies.
2.2.2. System Calibration:
The helium leak instrument used here is a mass spectrometer that is tuned to detect only helium. Prior to each sample analysis, the Head Space Analysis Module (HSAM) probe of the instrument was calibrated with 100% helium, 75% helium, and 50% helium, with system suitability performed using NIST traceable leak standards on each day of analysis. System suitability includes analysis of E-5 std cc/s, E-6 std cc/s, E-7 std cc/s, and E-8 std cc/s leak standards, a helium background analysis, and a concentration measurement of 75% helium.
2.2.3. Sample RSF and Helium Leak Rate Measurements:
Measuring the helium leak rate at −80°C on the test day involves a few steps for practical purposes. First, the stored samples were pulled out and allowed to come to room temperature (for at least 1 h). This is to enable the measurement of the RSF. RSF was measured using a Genesis RSF tester (Model # AWG, Serial # 173) in which a slow, constant rate of strain is applied to the top of a capped vial and resistance to package compression is monitored thereby generating a stress-strain profile (19).
Once the RSF value was obtained, the vials were refrozen (−80°C ± 10°C freezer) for a minimum of 3 h before analysis. Afterwards, the vials were placed in a special fixture, called the Vacuum Test Fixture Module (VTFM), connected to the helium leak detection instrument. The VTFM is temperature-controlled and continuously monitored while the samples are being analyzed. For our studies, it was set to −80°C and monitored before the analysis of each vial.
The principle of the helium leak rate measurement with this system was previously described (20). Briefly, the instrument places the sample under high vacuum and measures the helium as it escapes from the sample vial. The helium partial pressure present in the leak detector is measured quantitatively by the mass spectrometer as a leak rate. As the vial is brought to room temperature, the HSAM measures the remaining helium concentration in the vial. The actual helium leak rate, which is proportional to the size of the leak through which the helium escapes, is calculated from the helium concentration and the measured helium leak rate. This is the corrected and reported value.
3. Discussion of Results
This section provides discussion of the results obtained from thevarious measurements.
3.1. RSF Measurements
Multiple optimizations were conducted to identify the right capping conditions necessary to achieve the targeted RSFs. It is important to ensure that, once identified, these targeted “locked-in” stopper compression, and by extension, RSFs, can be reliably and consistently achieved. Indeed, multiple publications have shown that RSF can be achieved repeatedly for a set of packaging components, capping machines, and conditions (16, 21⇓–24). Figure 2a–d displays the results of the target and the measured RSF values for the samples used in the study. It indicates that the high stopper compression (or RSF up to 20 lbs) can be achieved with these component configurations. It should be noted that the 10 mL Crystal Zenith vial was only able to achieve a maximum RSF value of 15 lbs. It might be possible to achieve a higher RSF value with a different capping machine or associated components. This is outside the scope of this work and has been well described previously (24).
Relationship between the targeted and the measured initial RSF values for (a) 2 mL Crystal Zenith vial, (b) 2 mL Corning Valor vial, (c) 10 mL Crystal Zenith vial, and (d) 10 mL Corning Valor vial.
As the rubber stopper relaxes over time, the RSF value will gradually decrease; although the maximum drop is expected within the first 24–48 h after which it becomes relatively constant (22). This is also corroborated by the plots of Figure 3a–d.
Average RSF value over time for (a) 2 mL Crystal Zenith vial, (b) 2 mL Corning Valor vial, (c) 10 mL Crystal Zenith vial, and (d) 10 mL Corning Valor vial.
3.2. Container Closure Integrity
The results of the helium leak rates are presented in Figure 4. As observed from the plots, the glass vials generally showed a lower leak rate in comparison with the plastic vials. A number of competing factors might be at work, such as the higher permeability and lower rigidity (25) of the plastic (leading to potential micromovement post capping). Moreover, the difference in the shrinkage rate between the rubber stopper and the vial may play a role. The plastic vial and the rubber stopper typically have similar shrinkage rates whereas the glass vial shrinks at up to 20 times slower (6). This can help promote a tighter seal for the plastic vial during and up to frozen temperatures such as −80°C. It should be noted that the glass vial is fitted with a European blowback, whereas the plastic vial has none.
Helium leak rate vs average RSF value for (a) 2 mL vials, (b) 10 mL vials, (c) 2 mL and 10 mL Crystal Zenith vials, and (d) 2 mL and 10 mL Corning Valor vials (vertical axis is in log form).
In Figure 4a, all 2 mL Corning Valor vials and 99% of the 2 mL Crystal Zenith vials (at all ranges of RSF and time) allowed leak rates below 6 × 10−6 STD*cc/s, a value below which is associated with substantially reduced risk of microbial ingress as shown by Kirsch et al. (20). Kirsch demonstrated the relationship between the leak rate and the probability of microbial ingress (24). The Crystal Zenith vials passed with only one failure with an RSF value of 4.80 lbs. The failed sample was created at low capping settings, which resulted in the inability to pull a vacuum during the testing process, representing gross leaks.
In Figure 4b, all 10 mL Crystal Zenith vials and 96% of the 10 mL Corning Valor vials (at all ranges of RSF and time) allowed leak rates below the Kirsch criterion of 6 × 10−6 STD*cc/s. The four failures observed with the 10 mL Corning Valor vials had RSF values of 3.8 lbs, 4.5 lbs, 5.0 lbs, and 5.7 lbs. As expected, failures were observed when the RSF was low, particularly below 6 lbs. This value is in alignment with suggested lower limit of around 25 N (about 5.6 lbs) from previous work by Ovadia et al. (23); they developed a statistical approach to identifying a lower limit of RSF that enable a passing rate of CCI.
Figure 4c and d compare the leak rates based on vial sizes for each type of vial. It is difficult to determine which vial size allows more helium leak, regardless of vial type. Indeed, as previously indicated, many factors such as surface area, initial seal tightness, and lip strength could play a role in the observed overall helium leak rate.
4. Implications
The implications of these results are quite significant. First, this work describes the use of a deterministic method (helium leak) to measure leak rates at frozen storage conditions for both glass and plastic. Also, the studies provided supporting evidence that both types of components, under the right conditions, can be adequate container closure systems for −80°C packaging and storage. Specifically, Corning Valor glass vials and Daikyo Crystal Zenith plastic vials, despite the material differences (Tg) with the vial rubber stopper (the shrink rate of the rubber is similar to that of the Crystal Zenith plastic and is higher than that of glass), maintained container closure using the highly sensitive helium leak method provided that the vial sealing quality was good (quantified by a minimum RSF value of 6.0 lbs, which aligns with another study [23]). This underscores the criticality of both the specific characteristics of the primary packaging components and the quality of the vial sealing process for deep cold storage CCI.
It is further noted that this work showed a higher risk of CCI failure with lower RSF values, a fact that is well documented in the literature. Moreover, our work indicates the compatibility of the standard Daikyo halobutyl stoppers and West seal with these vials. Indeed, high RSF values can be achieved through necessary optimizations with these container combinations.
Finally, the results of this work have direct applications to many current and new therapeutics. In particular, these data can directly be used in identifying and selecting the right packaging systems for vaccines under development.
5. Concluding Remarks
In this study, we presented the results of frozen temperature storage (−80°C) with four model vial container closure systems in two sizes (2 mL and 10 mL). We found that the Corning Valor glass and Crystal Zenith plastic vials can provide adequate CCI at that temperature. The data also support the application of a deterministic helium leak method at such temperatures. Finally, this work provides a framework on the use of correlation between RSF and CCIT for vial-based systems at frozen storage conditions.
Although this work focused on CCI, there are other risks during frozen storage such as cryo-concentration, loss of homogeneity, pH shifts, difficulties in thaw (and subsequent inhomogeneity with potential impact for dose variations), dry ice shipment-related issues like CO2 permeation, and so forth (26, 27). Furthermore, there might be potential variations between −20°C and −80°C particularly across stopper formulations and vial types. These are indeed suggestions for future work.
Conflict of Interest Declaration
The authors declare that they have no competing interests.
Acknowledgments
The authors would like to thank West Pharmaceutical Services, Corning Inc., and Genesis Inc. for their contributions to the sample creation, measurements, and discussions of results.
- © PDA, Inc. 2023
