Conference Summary: Gaps, Lessons Learned, and Areas for Improvement. K. Brorson, G. Miesegaes, O. Tounekti, et al.

Transcription

1 Downloaded from on November 19, 2018 Conference Summary: Gaps, Lessons Learned, and Areas for Improvement K. Brorson, G. Miesegaes, O. Tounekti, et al. PDA J Pharm Sci and Tech 2014, Access the most recent version at doi: /pdajpst

2 CONFERENCE PROCEEDING ARTICLE Conference Summary: Gaps, Lessons Learned, and Areas for Improvement K. BRORSON 1, G. MIESEGAES 1, O. TOUNEKTI 2, J. SKENE 2, and J. BLUMEL 3 1 Center for Drug Evaluation and Research (CDER), U.S. Food and Drug Administration (FDA), Bethesda, MD; 2 Health Canada, Ottawa, Ontario, Canada; and 3 Paul-Ehrlich-Institut, Federal Institute for Vaccines and Biomedicines, Langen, Germany PDA, Inc Gaps and Action Items from 2009 Meeting At the 2009 meeting in Indianapolis, gaps and areas for improvement for individual unit operations were identified during an interactive closing session with the participants. A set of action items to address these gaps was drafted, and they were listed in the 2009 conference White Paper (1) and were reiterated in the 2011 meeting invitation letter. Follow-up was left optional and dependent on firms and individuals to take initiative(s). In an interactive session on the second day of the conference, a status check of these items was taken. It was found that many had been addressed to some extent, while others are still pending. Gaps and Action Items with Follow-Up During the discussion, it became clear that some action items that have since been addressed, at least partially, included the following: Determination of the isoelectric point (pi) for mammalian viruses Bovine Viral Diarrhea Virus (BVDV), reovirus type 3 (Reo3), Pseudorabies virus (PRV), poliovirus, human papillomavirus (PPV), and others. Characterization of virus preparations. Implementation of polymerase chain reaction (PCR)- based Chinese hamster ovary (CHO) RVLP (retrovirus-like particle) assays as a replacement for protein A validation studies using murine leukemia virus (MuLV). The status of each of these action items is summarized below. Virus Isoelectric Points (pis) doi: /pdajpst Determining the pis of mammalian viruses of concern would be helpful to better design ion exchange chromatography (IEX) operations to better partition viruses. This is an area that has not been systematically studied, although examples of virus pi measurements exist in various literature sources, many dating to the 1960s and 1970s. Few of these were performed on viruses typically used in industrial clearance studies. Since the 2009 meeting, some efforts using a method called chromatofocusing have been reported. Chromatofocusing is a chromatography-based method to measure the pi of macromolecular complexes. It relies on creation of an internal ph gradient when ion-exchange columns are run with certain buffers. Analytes will focus within the ph gradient depending on their native pi and the surrounding ionic conditions. The sample is run in an IEX column under conditions that will force virus to bind media. After careful buffer selection, the focusing behavior creates a ph gradient in the column. The ph at which the virus elutes is noted, and the experiment is repeated under various salt conditions. The salt (typically sodium chloride, NaCl) alters the chemistry within the column, and the virus elution ph shifts. The natural pi can then be mathematically extrapolated. It does not require large amounts of starting material, but it can involve complicated mathematical transformations of elution peak behavior. Nonetheless, it is a reasonable alternate to more problematic and tedious approaches (i.e., electrokinetics). It was previously used in a collaboration between CDER/FDA and the University of Maryland Baltimore County to measure the isoelectric point of three bacteriophages (2). In 2009 D. Strauss of Genentech Inc. (3) used chromatofocusing to measure the pi of three key viruses used in clearance studies: mice minute virus (MMV) at 6.2, Simian virus 40 (SV40) at 5.4, and MuLV at 5.8. At this meeting some follow-up data from CDER/FDA was reported regarding chromatofocusing PPV, a porcine parvovirus used as a surrogate for the more troublesome MMV murine version. These studies found that the pi of PPV seems somewhat more acidic than MMV, probably Vol. 68, No. 1, January-February

3 in the vicinity of 5.5, although at the time of this meeting a more formal analysis was pending (Figure 1). Characterization of Virus Preparations Both H. Dehghani (Amgen) and G. Miesegaes (CDER/ FDA) presented on this subject (Session IV). In 2010, Parenteral Drug Association (PDA) Technical Report 47 Preparation of Virus Spikes Used for Virus Clearance Studies was released (4). This document expanded on important quality attributes and controls needed for these preparations to be used in scale-down studies that are realistic models of large-scale processing. Both talks in the symposium elaborated on studies of a subset of viruses used in clearance studies, and one has since been published for public dissemination (5). CHO RVLP Assay Implementation This was first reported in 2009 by M. Zhang of Genentech Inc. (6) and expanded upon at the 2011 Symposium by M. Zhang (Session IV). It was reported that RVLP removal by protein A chromatography is comparable to that measured from spiked MuLV, and that RVLP removal determination can use full-scale samples. A follow-up article for publication is under way (manuscript in preparation). Ongoing or Unresolved Gaps and Action Items The discussion identified 2009 meeting action items where there is ongoing dialog but no clear resolution. These topics include virus retentive filtration strategies and development of low-ph inactivation guidelines. Validation of Virus-Retentive Filtration One question arose related to the best validation models for virus-retentive filtration: Since no breakthrough of retrovirus has been observed through small virus retentive filters... decide whether to (1) just claim MMV log reduction value (LRV) (for retrovirus) instead, (2) use phage as models, (3) do study with MuLV. In principle either (1) or (2) seems Figure 1 Representative preliminary data resulting from topics discussed at previous Viral Clearance Symposia. Here, studies to identify the isoelectric point (pi) of various model virus species (PPV shown) were conducted based on the isoelectric focusing technique. See Strauss et al 2009 for other virus types whose pi were identified using this approach. 84 PDA Journal of Pharmaceutical Science and Technology

4 logical, as filters have size-based mechanism of clearance and both phage and parvovirus are smaller than retrovirus. The regulatory follow-up has apparently been ambiguous. The audience was asked if any had submitted an Investigational New Drug/Biologic License Application proposing this. The answer was yes, with mixed responses from regulatory authorities. The one factor that seemed to be very important for successful acceptance was to raise the question proactively that is, to ask the regulators before doing the study with an alternative approach. Development of Low-pH Inactivation Guidelines A second action item was to develop a detailed protocol for low-ph inactivation studies that can guide firms and contract testing organizations (CTOs). Ideally, the protocol should outline proper instructions and checks for incubation, as well as important ancillary considerations like instrument calibration. After a short discussion, it was decided that this endeavor represented low-hanging fruit a defined task that could be accomplished in a timely manner yet would have a large positive impact. A small committee was formed under the leadership of D. Vacante, a symposium participant representing Janssen Inc. Over the course of the next year, the protocol was developed by a group of subject matter experts. The final version is in this special symposium proceedings issue of the PDA Journal. Remaining Knowledge Gaps and Action Items Some action items from 2009 seem to have not yet been addressed, and they will be addressed in subsequent symposia. These include the following: Potential impact of lipids on the effectiveness of detergent treatment as well as its effectiveness for a broad range of virus inactivation Investigating possible anion exchange chromatography (AEX) buffer effects on LRV (e.g., resinspecific effects?) Investigating product and column resin nonspecific binding of virus during protein A chromatography Conclusion of Follow-Up from the 2009 Meeting In conclusion, some progress on the action items was evident, and more was presented at this meeting. The question was raised whether a more systematic approach was warranted (i.e., establish a consortium versus reliance on individuals and firms to take initiative and address them on a product-specific basis). It was held that for certain paper-based exercises, like the low-ph study standard operating procedure, a consortium or team approach was more feasible. How to best coordinate regulatory follow-up was touched on. While no consortium was initiated at the 2011 meeting, it was agreed that a follow-up meeting, probably on the East Coast, should be held. Hopefully, some of the action items identified in the 2011 meeting (see below) will be addressed and the potential need for a consortium can be revisited. Lessons Learned from the 2011 Meetings: Virus-Retentive Filtration The closing of the 2009 Symposium focused on a discussion of hypothesis-driven research projects. Examples of high-priority research interests derived from these 2009 discussions included Utility of claiming phage or parvovirus LRV as a substitute for MuLV Characterization of virus spike preparations, with intent for optimizing quality At the 2011 Symposia, these areas had been addressed at least partially, while new lessons were identified. Firms presented on many of the topics that were discussed previously, which included (1) generic validation with a single virus; (2) robustness studies, including design of experiment (DoE)-based studies, demonstrating LRV consistency across multiple conditions and monoclonal antibody (mab) products; and (3) further defining the performance of different filter types. Others presented work not previously discussed, such as the use of protective guard columns and their influence on viral clearance. In summary, lessons learned at the 2011 meeting included the following: Performing filter-by-filter comparisons for your specific process may prove fruitful during development. For example, process-specific factors can be present that may affect different filter brands in different ways. The potential impact of spike impurities on virusretentive filtration is a key consideration, for example, with respect to the scale-down model. Vol. 68, No. 1, January-February

5 Transitions in pressure, a trend with virus breakthrough, may be considered worst-case. This pressure change induced breakthrough, however, is seemingly sporadic in nature, though has been shown across multiple filter types and is likely to occur during the buffer flushing procedures commonly employed. Although supportive data is accumulating, to unequivocally use parvovirus data for bypassing retrovirus validation requirements would require prospective discussion with regulatory authorities. Not all critical process parameters (CPPs) are known as of yet; however, concerns can be reduced with identification of a reasonable and appropriate design space. Symposium: Virus-Retentive Filtration In order to better understand what might be considered as a mechanistically simplistic but still often a complex operation, the symposium members decided the following to be the most relevant future steps: To better understand pressure release issues, one approach would be to investigate the fluid dynamics of various virus species. With respect to the variations in viral clearance across filter types, try working with or communicating with the various firms. This may prove useful in terms of better understanding the quality of their filters at small scale, the integrity testing procedures/requirements, and so on. Concerns with plugging and breakthrough could be assessed using microscopic techniques, for instance, to identify and track viruses across the various filter types, pore sizes, and perhaps transport mechanisms. Use of your firm s database records on parvovirus clearance in order to gain insight on the causes of breakthrough in your process. Possibilities may include overloading, flux decay, and use of new versus old virus-retentive filters. Lessons Learned from the 2011 Meetings: AEX AEX is a widely used polishing step that is capable of clearing multiple virus species under a variety of process conditions. It is effective when run in either bind-and-elute or flow-through modes, has a reasonable recycling lifetime when appropriately cleaned, and does not have any technical drawbacks such as the need to monitor for ligand leeched off the resin, as with protein A. Whether the recently introduced adsorptive chromatography membrane technologies are as capable of viral clearance remains to be formally addressed. The following was generally agreed to during the closing session regarding AEX: It is largely robust; outliers tend to be rationally explained. Impurity load can be considered an additional CPP, along with ph and conductivity. Conductivity tends to have a greater impact on viral clearance than does ph. Although the specifics are still being defined, DNA and host cell proteins levels can influence viral clearance, likely by competing for ligand binding sites. The DNA molecular weight (MW) spectrum is relevant, as well as the position of AEX in the overall process (polishing step #1 versus #2), as higher impurity loads may influence performance. For flow-through mode, it is likely most appropriate to define capacity based on impurities and not the mab intermediate per se. ph and conductivity can affect this too, however. Symposium: AEX AEX is one of the best understood unit operations regarding its mechanism, impact on product quality attributes, and viral clearance. It was also noted that adsorptive membrane technologies that are currently available are increasingly being used in the bioprocessing industry. They may have certain advantages that make them worth considering either as a change to an existing platform or when designing a new scheme altogether. As more data are gathered we will be better able to assess the extent of comparability between the two technologies. Future research efforts that should prove useful regarding AEX in general include 86 PDA Journal of Pharmaceutical Science and Technology

6 Defining the pi of mammalian viruses beyond those already described (e.g., Reference 3) Continue to determine whether other CPPs exist Better understanding impact of impurities on viral clearance More strictly define capacity Develop a strategy for adequate monitoring of relevant parameters Lessons Learned from the 2011 Meetings: Protein A Chromatography Protein A chromatography has been the gold standard for mab capture for many years. Still, this approach has its shortcomings, and in some cases may not be the ideal method for product capture. Studies have continued to better characterize this operation, given its extensive use, and while simultaneously understanding that it continues to remain rather ill-defined from a mechanistic standpoint. Other operations considered as viable alternatives for example, mixed-mode chromatography, cation exchange chromatography (CEX), and the rare occurrences of custom affinity platforms have been increasingly investigated, including their extent in contributing as a viral clearance unit operation. Nevertheless, protein A remains the most commonly employed unit operation and under most conditions can often produce reliable clearance of MuLV. The removal of other viruses, however, tends to vary; a definitive explanation as to why these variances arise does not exist at this time. Further awareness/insight into the mechanism of viral clearance by protein A continues to develop. Because of this knowledge gap, only limited generalizations could be made at this time. Observations discussed and agreed upon by the group included the following: A definitive mechanism of viral clearance for protein A is not yet complete; further studies are warranted. Issues brought up at the initial symposium remain elusive. For example, although known feedstock effects exist for this unit operation, the specific component(s) within the feedstock, and the exact nature of its impact on viral clearance, remains elusive. Additional factors that can influence viral clearance, as presented and discussed at both (2009 and 2011) meetings, include load volume and elution conductivity. Symposium: Protein A Chromatography The group unanimously decided that little progress can be made regarding protein A chromatography until a more complete mechanistic understanding is achieved. Attempts to identify specific approaches to address this were proposed, although no clear-cut response was generated. It was proposed that revisiting one s in-house viral clearance database, including the FDA s, may provide useful insight to initiate hypotheses and drive future laboratory study. An example of this proposed at the meeting was to mine the database for evidence of LRV trends between Fc fusion proteins versus full-length mabs. Lessons Learned from the 2011 Meetings: CEX Presentations at this session of the symposium confirmed previous data that virus clearance on CEX is due primarily to electrostatic interactions of the virus with the resin and could also be due to hydrophobic interactions. Virus binding efficiency depends on the physicochemical properties of the product such as the pi, the type of resin, and the type of virus. DoE studies showed that CEX can be effective for removal of viruses over a narrow range of operational parameters. Participants concluded that CEX viral clearance effectiveness is not linked to yield or the presence of impurities and that conductivity and ph represent the key CPPs. When performed in bind and elute mode to polish monoclonal antibodies, CEX was shown to be less amenable to developing modular approaches. Symposium: CEX Follow up studies using different platforms (other CEX resins, products, viruses) may be needed to establish the design space for the identified CPPs. Lessons Learned from the 2011 Meetings: Detergent Inactivation Treatment with detergents is a very robust and effective method for inactivation of lipid-enveloped vi- Vol. 68, No. 1, January-February

7 ruses. Temperature, time, and detergent concentration constitute the key CPPs that influence virus inactivation effectiveness. Other platform-specific CPPs were discussed, but it was evident that further study would be needed to confirm them, or to identify additional ones. Also decided was that DoE-based experiments, as with other operations, would aid in supporting the use of bracketed operating conditions, and for assessing results generated across different platforms and with multiple viruses. Symposium: Detergent Inactivation The use of detergents by industry may be limited by their toxicity, by the need to reduce residual levels in the final product, and by challenges associated with disposal requirements. Development of new generation biodegradable and non-toxic detergents was discussed. Presentations at this symposium showed promising results for several new generation detergents. However, work is needed in collaboration with detergent manufacturers to establish the safety of their ingredients. Lessons Learned from the 2011 Meetings: Emerging Unit Operations Ceramic hydroxyapatite chromatography, mixed-mode chromatography, AEX membranes, UV-C viral inactivation, and precipitation of virus with caprylate are all emerging technologies that may play a significant role in bioprocessing going forward. Although most of the methods presented have been implemented in some manufacturing processes, it has been on a caseby-case basis and is process-specific. Further experimentation is needed for almost all of these emerging unit operations to determine the parameters (CPPs) and to develop a better understanding of the action (mode of action, MoA). The mixed-mode chromatography and AEX membranes both demonstrated characteristics similar to that observed for chromatography (AEX), namely the impact of impurities (DNA and HCP) on the robustness of the process step. Further work is needed to develop rules for deciding if mixed-mode chromatography is orthogonal to hydrophobic interaction chromatography and/or AEX. The presentation on UV-C inactivation demonstrated much promise due to the well understood and documented MoA and the potential for continuous processing. It is yet to be determined if robust LRV can be obtained with adequate protein quality. How to define the scale-down process remains unresolved. Lessons Learned from the 2011 Meetings: Integration and Risk Management The risk management session featured multiple presentations, all discussing the application of risk management to the development of a viral clearance control strategy. Key presentations included a proposed risk assessment strategy for viral safety from the FDA (see Xu et al. in this issue of the PDA Jounal), a discussion on using criticality assessment based on impact and uncertainty of each unit operation, and setting the overall virus clearance strategy limits as the sum of all stages. The integration session saw new approaches to improve viral clearance studies. Presentations included the development of a quantitative PCR (QPCR) RVLP assay allowing for measurement of actual virus RVLP (not model virus) at scale (not scale-down model). The use of in-line spiking of virus into the scaled-down process to better simulate actual commercial manufacturing conditions was discussed. Data was presented demonstrating that variations in the quality of virus stock preparations may affect the overall viral clearance capabilities one can achieve. During this session an ongoing effort to develop an ASTM standard for a low-ph retroviral clearance unit operation was presented. Several in attendance expressed enthusiasm regarding this proposal and volunteered to review the ASTM low-ph standard as experts from industry. The ASTM standard E was released in late 2012 (7). The session ended with a call for examples of the use of platform data in regulatory submissions during the next symposium in 2013, in an effort to identify issues related to the acceptability of platform data. Lessons Learned from the 2011 Meetings: Quality of Virus Spike Preparations In addition to feedstock-derived components, impurities present in the virus spikes can impact clearance capability. In 2010, the Parenteral Drug Association (4) released a technical report that outlines current approaches regarding the preparation, characteriza- 88 PDA Journal of Pharmaceutical Science and Technology

8 tion, and storage of the viruses used in viral clearance studies. This is a useful reference, given that further insight presented at the symposium that demonstrated the variability one may experience solely as a result of variations in virus spike quality attributes. As the session closed, it was decided that, as part of a firm s initial process screening, assessing the extent of virus spike impurities that exist in the preparation(s) to be used would likely offer benefits such as increase in process control. Nonetheless, considerations agreed upon by the group included the following: It is possible to make better quality virus spikes, including for those firms generating their own preparations in-house. Characterization and quantification of spike quality attributes may aid in reducing variations in LRV and ultimately yield increased process control. The choice of using standard purification approaches or additional techniques to generate a highly pure virus spike (e.g., use of pre-filters, passing through membrane adsorbers) can be debated. Of likely more relevance is whether consistency in preparation quality is achieved. The overall impact across unit operations is currently unknown, although those that have been studied thus far (virus retentive filtration, chromatography) suggest a general influence of impurities is possible. What are the relevant operations in which to appropriately use pure spikes? For virus retentive filtration, yes. For other operations, and if so, how are they affected by spike preparation quality? Which are the relevant attributes to consider? Symposium: Virus Spikes One action item agreed upon by the group was to work toward developing a certificate of analysis with regard to establishing virus spike preparation quality procedures and standards. This will be a large-scale and ongoing endeavor. However, it is clear that some contract testing organizations are now offering more highly purified virus spike preparations in response to customer requests and market forces. Acknowledgements All authors thank Audrey Jia, Joanna Qing Zhou, Michael Laska, Narahari Pujar, and Colette Ranucci for their careful and timely review of the manuscripts in this conference proceedings issue of the PDA Journal. Logistical and financial support for the 2011 South San Francisco Viral Clearance Symposium was provided by Amgen Inc. (Thousand Oaks, CA) and Genentech Inc., a Member of the Roche Group (South San Francisco, CA). References 1. Miesegaes, G.; Bailey, M.; Willkommen, H.; Chen, Q.; Roush, D.; Blümel, J.; Brorson, K. Proceedings of the 2009 Viral Clearance Symposium. Dev. Biol. (Basel) 2010, 133, Brorson, K.; Shen, H.; Lute, S.; Perez, J. S.; Frey, D. D. Characterization and purification of bacteriophages using chromatofocusing. J. Chromatogr., A 2008, 1207 (1), Strauss, D. M.; Lute, S.; Tebaykina, Z.; Frey, D. D.; Ho, C.; Blank, G. S.; Brorson, K.; Chen, Q.; Yang, B. Understanding the mechanism of virus removal by Q sepharose fast flow chromatography during the purification of CHO-cell derived biotherapeutics. Biotechnol. Bioeng. 2009, 104 (2), PDA. Technical Report 47. Preparation of Virus Spikes Used for Virus Clearance Studies. Parenteral Drug Association: Bethesda, MD, Miesegaes, G.; Lute, S.; Dement-Brown, J.; Kaushal, S.; Strauss, D.; Chen, D.; Brorson, K. A survey of quality attributes of virus spike preparations used in clearance studies. PDA J. Pharm. Sci. Technol. 2012, 66 (5), Zhang, M.; Lute, S.; Norling, L.; Hong, C.; Safta, A.; O Connor, D.; Bernstein, L. J.; Wang, H.; Blank, G.; Brorson, K.; Chen, Q. A novel, Q-PCR based approach to measuring endogenous retroviral clearance by capture protein A chromatography. Biotechnol. Bioeng. 2009, 102 (5), ASTM. E : Standard Practice for Process for Inactivation of Rodent Retrovirus by ph. ASTM International: West Conshohocken, PA, Vol. 68, No. 1, January-February

9 An Authorized User of the electronic PDA Journal of Pharmaceutical Science and Technology (the PDA Journal) is a PDA Member in good standing. Authorized Users are permitted to do the following: Search and view the content of the PDA Journal Download a single article for the individual use of an Authorized User Assemble and distribute links that point to the PDA Journal Print individual articles from the PDA Journal for the individual use of an Authorized User Make a reasonable number of photocopies of a printed article for the individual use of an Authorized User or for the use by or distribution to other Authorized Users Authorized Users are not permitted to do the following: Except as mentioned above, allow anyone other than an Authorized User to use or access the PDA Journal Display or otherwise make any information from the PDA Journal available to anyone other than an Authorized User Post articles from the PDA Journal on Web sites, either available on the Internet or an Intranet, or in any form of online publications Transmit electronically, via or any other file transfer protocols, any portion of the PDA Journal Create a searchable archive of any portion of the PDA Journal Use robots or intelligent agents to access, search and/or systematically download any portion of the PDA Journal Sell, re-sell, rent, lease, license, sublicense, assign or otherwise transfer the use of the PDA Journal or its content Use or copy the PDA Journal for document delivery, fee-for-service use, or bulk reproduction or distribution of materials in any form, or any substantially similar commercial purpose Alter, modify, repackage or adapt any portion of the PDA Journal Make any edits or derivative works with respect to any portion of the PDA Journal including any text or graphics Delete or remove in any form or format, including on a printed article or photocopy, any copyright information or notice contained in the PDA Journal