Cytotoxicity Risks in Single-use Media and Bioreactor Containers

Disposable biocontainers used for media preparation and storage, and for single-use bioreactors, have become well-established equipment for research, development, and production of preclinical and clinical batches of bioprocess drug APIs and vaccines. They are now increasingly being implemented for full-scale approved product manufacturing. Numerous studies have been published or presented at conferences demonstrating comparable performance characteristics for cell-line growth and productivity relative to media stored in traditional stainless steel or glass tanks and cultured in stainless steel or glass bioreactors.

In the shadow of these broad successes, however, there have been several reports of diminished cell growth properties associated with some biocontainer or bioreactor films, generally during initial qualification, but on rare occasion, attributed to a change in film formulation or manufacturing/transport conditions with a previously qualified and historically suitable film. One of the earliest reported cases of reduced cell growth in a single-use bioreactor was associated with adsorption of cholesterol in a defined culture medium supplemented with a cholesterol-lipid complex that was coupled to methyl-β-cyclodextrin.  In this case, the bioreactor film had previously been qualified for culture of the same NSO cell line in the media supplemented with serum. Reduced cell growth in the bioreactor with the defined culture media was attributed to extraction of cholesterol from the cell membranes in the presence of excess methyl-β-cyclodextrin and subsequent adsorption to the low density polyethylene (LDPE) film contact layer surface. The problem was resolved by modifying the formulation of the culture medium to remove the excess methyl-β-cyclodextrin (1).

More recently, Eibl et al. reported a study of six different cell lines cultured in different media formulated with water that had been stored in nine different polymer film biocontainers from various suppliers (2).  In this study, unidentified aqueous extractable compounds (cytotoxic leachables) from two of the nine biocontainers tested appeared to reduce growth properties of some cell line/media combinations, while others were not affected.

Water extraction can result in solubilization of polar (ionic) compounds from the biocontainer film. Storage of culture media supplemented with serum or a lipid-based additive can be expected to leach these polar compounds (salts), but also extract (leach) moderately non-polar (non-ionic) compounds due to the non-polar solubilization properties of these more non-polar additives. Further study of cell line growth in supplemented media previously stored in single-use biocontainers is underway.

A related study using culture media as the extractant was reported by Wood et al., who described a survey of 13 different film biocontainers from eight different suppliers using different cell culture media warmed to 37 ˚C as the contact fluid, prior to culture of different Chinese hamster ovary (CHO) cell lines (3). Among all the biocontainers tested, only one was reported to show some cytotoxic effect under these conditions, for a single cell line.  Root-cause investigation of the problematic biocontainer suggested a volatile or air-quenched compound was associated with the observed negative growth parameters (cytotoxicity).  This volatile leachable was associated with water or media and concentration dependent on temperature and exposure time (as all leachables are). The cytotoxic effect could be eliminated by aeration, allowing the volatile compound to dissipate prior to filling with culture media. Growth rates of a model cell line in non-irradiated (non-sterile) bags and glass bottles were comparable and higher, suggesting the volatile leachable was a degradant generated by gamma irradiation. Based on their test data, the authors suggested cell-culture specific screening of single-use components for cytotoxic effect in addition to extractables studies with model solvents (generally assessed versus potential leachables levels and potential systemic toxicity effects in the final dosage).

Cytotoxic Effect Involving Antioxidants
The most controversial case was reported by Hammond et al., published in the PDA Journal of Pharmaceutical Science and Technology (4). This study has garnered an extraordinary amount of attention due to its investigation into a root cause of observed biocontainer-associated cytotoxicity.  The Parenteral Drug Association (PDA) has anecdotally reported this paper to be its most downloaded in history. The authors reported a cytotoxic effect in two sensitive cell lines with a particular single-use biocontainer, attributable to a degradation product of tris[2,4-di-tertbutylphenyl] phosphite (Irgafos 168, BASF), a commonly used antioxidant compound formulated into many biocontainer films and other plastics approved for use in drug and food manufacturing.  

This study has led to calls by some users for biocontainer suppliers to reformulate their products either with a different antioxidant or even no antioxidant at all.  Such suggestions, however, are ill advised. For one, other biocontainers formulated with Irgafos 168 have shown no cytotoxic effect in numerous studies. Being a widely used antioxidant in LDPE films, as well as other single-use plastic components—perhaps the most widely used—experience with potential cytotoxicity of degradants of alternative antioxidants is far more limited. Changing to a different antioxidant, therefore, provides no assurance that similar cytotoxicity effects will not be seen, and perhaps even to a greater degree.  Avoiding all antioxidants is also undesirable as many plastics will be rapidly degraded by oxidation under the heat of processing or the energy of gamma irradiation.  Antioxidants, often a component of extractables and leachables, are required to enable plastics to maintain their functional properties throughout shelf life.  They protect plastics polymers from oxidative degradation under the heat of processing (formulation and extrusion or molding) and prolong shelf life. Antioxidants preferentially degrade by scavenging free radicals that would otherwise break covalent polymer bonds, causing the plastic to become brittle and generate particles or potentially more undesirable leachables.

While the reported cytotoxicity was attributed to a gamma radiation-induced degradant of Irgafos 168, the absence of cytotoxic effect in other biocontainers also formulated with Irgafos 168 suggests it is not the Irgafos 168 or its primary degradant, Oxidized Irgafos 168, that was the problem per se. Furthermore, while radiation dosage was also shown to be a contributory factor in the levels of the cytotoxic degradant present, as well as its effect, the other biocontainers containing Irgafos 168 were also subjected to similar high radiation doses without showing cytotoxic effect. Clearly, some factor other than radiation dose and the mere presence of Irgafos 168 is at play here.  

Following publication of the Hammond et al. paper, the supplier of the implicated biocontainer came forward and presented a joint explanation during a webinar and at the BPI Conference in Boston (5). They disclosed that the problem was associated with an excess amount of Irgafos 168 and that the solution, arrived jointly with the user, was to reformulate the film with a reduced level of Irgafos 168. Studies subsequently showed that the cytotoxic degradant and its effects on select cell lines was eliminated by reducing the Irgafos 168 concentration in the polymer formulation, presumably without adversely affecting polymer properties (though impact on shelf life was not reported).  That solution also provides a possible explanation why other films with Irgafos 168, receiving similar radiation dosages, did not demonstrate cytotoxic effects. They likely already had lower, optimized levels of Irgafos 168 that do not generate sufficient levels of the degradant leachable to cause cytotoxicity.

Recommendations
As suggested by some of these authors, end users and suppliers might collaborate to develop an industry acceptable standardized test by identifying a cell line and culture medium formulation that would serve as a representative worst case and satisfy a majority of the end-user requirements.  As with the current screening test, United States Pharmacopeia (USP) <87> Biological Reactivity Test, in vitro (MEM Cytotoxicity), this new “worst case” cytotoxicity test could be conducted by biocontainer suppliers (or their contract testing laboratories), and results could be published in the supplier’s bio-container validation data packages,potentially reducing the need for user-specific qualification tests.  Efforts are underway to achieve this, however at present, no user has come forward with a sufficiently sensitive cell line they are willing to declassify. It, therefore, remains prudent for end users to screen all new cell lines or media formulations with their intended or potential biocontainer(s) and bioreactor film(s) prior to initiating production. Suppliers may provide evaluation kits to facilitate such testing.

Other key factors in avoiding cytotoxicity events are the change control, change notification, and traceability practices of the biocontainer supplier, as well as their understanding of the materials science. The last case discussed here purportedly did not occur in a new biocontainer screening, but rather was the result of an unannounced formulation change (increasing the Irgafos 168 concentration to excess). As seen here, supply chain-related changes in formulation, processing, component manufacturing, or transport and storage that do not impact form, fit, or general function--and even perhaps improve productivity of the film--could still potentially be harmful to cell-culture processes. Biocontainer and single-use system suppliers need to apply best efforts to avoid changes, and, where imposed in the absence of a suitably sensitive generic bioreactor cytotoxicity screening test, need to communicate change notifications with biocontainer/bioreactor customers in advance of any anticipated changes to enable users to qualify and accommodate any unavoidable changes.

References
1. J. Okonkowski et al., “Cholesterol Delivery to NSO Cells: Challenges and Solutions in Disposable Linear Low-Density Polyethylene-Based Bioreactors,” J. BioSci. and BioEng., 103, No. 1, 50-59 (2007).
2. N. Steiger and R. Eibl, “Interlaboratory Test for Detection of Cytotoxic Leachables arising from Single-Use Bags,” Chemie Ingenieur Technik 2013, 85, No. 1-2, 26-28 (2012).
3. J. Wood et al., “Strategy for Selecting Disposable Bags for Cell Culture Media Applications Based on a Root-Cause Investigation,”  Biotechnology Progress, 29, No. 6, 1535-1549 (November/December  2013).
4. M. Hammond et al., “Identification of a Leachable Compound Detrimental to Cell Growth in Single-Use Bioprocess Containers,”  PDA Journal of Pharma. Sci. and Technol., 67, No. 2, 123-134 (March-April 2013).
5. S. Klein and M. Goodwin, “New Approaches to Managing Single-Use Systems,” presented at BioProcess International Conference, Boston, MA, October, 2013.

Jerold Martin

Jerold Martin is Sr. VP Global Scientific Affairs at Pall Life Sciences, Port Washington, NY and Chairman of the Board and Technology Committee Bio-Process Systems Alliance (BPSA) jerold_martin@pall.com.