Regulatory Highlights Cite This: Org. Process Res. Dev. XXXX, XXX, XXX−XXX
pubs.acs.org/OPRD
Regulatory Highlights
Downloaded via NEW MEXICO STATE UNIV on July 5, 2018 at 22:45:07 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
■
INTRODUCTION This article seeks to examine areas of regulatory guidance and practice that can be considered to be of highest current interest and impact on the activities of chemists and engineers in process research and development of pharmaceuticals, either because of ongoing evolution of new guidance or, in the case of existing guidelines, clarification through supplementary publications aligned to a specific guideline. In this review we look at four areas. First is the publication of the widely anticipated ICH Q12 Step 2 document. Related to this in some ways is the topic of prior knowledge, and here we examine recent discussions aimed at defining how to effectively utilize such knowledge. Finally, we look at the status of two recently finalized guidelines: ICH M7 on mutagenic impurities1 and ICH Q3D on elemental impurities.2
An approach of particular importance that is included in the guideline is the “post-approval change management protocol” (PACMP), which allows for specific changes to be predescribed to regulators and agreement to be reached on the scientific approach and data expectations that will support the change. This ability to predefine how to successfully make a change will bring great clarity and predictability to the planning and prosecution of, particularly, complex change types (often viewed as major changes needing “prior approval” in current regulatory change systems). Furthermore, the predetermination of data necessary to support the change allows for the final communication of the change to be a simple matter of confirming the suitability of the change with the expected data and for the regulatory change class to be reduced on the basis of the prior agreement of the change management approach. Importantly, a PACMP can be either agreed for a single change for a single product or constructed and agreed in a more wideranging manner to support multiple similar changes to be conducted on more than one product. This is of immense potential value to industry and regulators alike. Annex II of the draft guideline provides illustrative examples of different types of PACMPs, giving an example of a PACMP for a single change (to a manufacturing site for a drug substance) and an example of the more general management of such a site change. In a section of the guideline on supporting post-approval changes for marketed products, where considerable manufacturing experience has been accrued, important approaches are given for the management of changes in analytical procedures and discussing how data requirements for changes (for stability data) can be impacted by product and process understanding. In addition, the guidance seeks to provide an approach to differentiate the levels of regulatory oversight of particular changes on the basis of known impact and criticality of the potential change to product quality. The ability to differentiate change expectations on the basis of actual product understanding is a natural extension of the approaches taken in ICH Q8 and Q11, where for example product and process understanding can establish a “Design Space” for manufacturing and control within which changes are not seen as requiring regulatory oversight. In the draft of Q12, this concept is further developed by the concept of “Established Conditions” (ECs), with discussion of how investment in understanding can impact submission expectations (with Appendix I of the draft guideline providing an illustration of CTD sections that contain ECs and Annex I suggesting illustrative examples of ECs for both chemical products and biological products) and post-approval change management expectations. Importantly, the guidance discusses how this approach could be used for existing products, where the manufacturing process may have been described without any differentiation of change management expectations, leading to inefficient use of both industry and regulatory resources. The draft guideline also includes a suggested system for the collation of such “agreed” regulatory change mechanisms for a
■
ICH Q12: GUIDELINE ON TECHNICAL AND REGULATORY CONSIDERATIONS FOR PHARMACEUTICAL PRODUCT LIFECYCLE MANAGEMENT Recent ICH quality guidelines (Q8−Q11)3−6 have focused on providing guidance on the development and manufacture of drug substances (Q11)6 and drug products (Q8),3 showing “baseline” and “enhanced” scientific approaches, and utilizing quality risk management tools (Q9) within the pharmaceutical quality management system (Q10). To further support the implementation of these development and manufacturing approaches, ICH recognized the value in providing tools and approaches for the management of post-approval chemistry, manufacturing, and controls (CMC) changes based on product and process understanding that could be employed by all ICH participants. Several useful tools had been established in different regions, and it was recognized that pharmaceutical innovation and continuous improvement would be optimally supported if best practices could be employed in similar ways across the regions. Achieving this harmonization would result in more efficient manufacture and change and would also increase the value of the pharmaceutical quality system and support continued optimization of the utilization of valuable resources within regulatory agencies and inspectorates (e.g., toward oversight of critical rather than noncritical changes, incentivizing industry’s understanding and management of manufacturing). The ICH Concept Paper for the development of this guidance was endorsed in 2014.7 The drafted consensus document is now available for public comment (step 2 of the ICH process),8 with comments being collected by the regions during 2018 (with various comment deadlines). The draft guidance includes some potentially very important approaches for future CMC change management, and importantly, the tools and approaches being developed are seen as usable across the range of pharmaceutical product types (including drug−device combinations) and applicable to existing products as well as newly approved products. © XXXX American Chemical Society
A
DOI: 10.1021/acs.oprd.8b00186 Org. Process Res. Dev. XXXX, XXX, XXX−XXX
Organic Process Research & Development
Regulatory Highlights
The case studies and outputs, including a detailed meeting report drafted by the EMA, are available on the EMA Web site and highlight the conclusions described below. What is Prior Knowledge? The source of prior knowledge source can be internal knowledge from a company’s proprietary development and manufacturing experience (e.g., historical experience based on similar compounds, products, and processes; application of “platform technologies”; knowledge from previous filing) or external knowledge such as reference to scientific and technical publications. It was noted that publication of “generizable” internal knowledge could be a way to increase transparency and scientifically validate the knowledge. Common textbook knowledge is not considered to be prior knowledge in the context of the discussions and examples explored in this workshop, as it is publicly available and generally accepted knowledge that does not normally require further elaboration or justification for use. How Can Such Prior Knowledge Be Used in Development? Examples of how prior knowledge can be used that were discussed during the workshop include risk assessments, platform approaches to development, and lifecycle management. Risk Assessments. Prior knowledge can be used to inform risk assessments during product and process development and in defining a control strategy. For example, prior knowledge can be employed in defining critical quality attributes (CQAs) and critical process parameters (CPPs) and informing an experimental plan and process ranges. Platform Approaches to Development. Prior knowledge can be used in synthetic process design, prediction of drug substance physical properties, and platform formulations (i.e., dosage forms, excipients, manufacturing process). For APIs, elements of prior knowledge include the following: • Use of prior knowledge to select the manufacturing process and to design chemical quality • Prediction of mutagenic impurity risks (QSAR “expert system” supported) and their purge (e.g., Lhasa Mirabilis) • Development of platform control strategies for specific reaction types • Use of prior knowledge to select and develop the manufacturing process controlling physical properties and how that links to the complexity and BCS classification of the drug product dosage form • Process design to control crystallization kinetics and particle size reduction steps (milling, etc.) • Use of prior process knowledge to support scale-up • Utilization of accelerated stability data and models to address risks and set retest dates. Lifecycle Management. There is a clear use for prior knowledge in lifecycle management, including linking risk to variation classification, using post-authorization tools such as PACMPs, using existing shelf-life data to assess a changed product, and managing changes in analytical methods. Prior knowledge should become increasingly useful following the implementation of ICHQ12. How Should Its Use in Regulatory Submissions Be Justified? The EMA has stated that an applicant should justify the relevance and applicability of prior knowledge for each specific new product. The relevance and applicability of the prior knowledge to the product under assessment must be justified. Prior knowledge is currently often used implicitly, and it is necessary to identify the prior knowledge more explicitly if and when it is used. The intended purpose of including the
product via use of a product lifecycle management (PLCM) approach, wherein the agreed changes can be clearly collated alongside the manufacturing commitments and the agreed (lesser) change reporting category for the changes. Annex III of the draft documentation provides an example of a PLCM document. The guideline also contains content describing the pharmaceutical quality system (PQS) change management expectations (with Appendix II of the guideline providing further illustration of principles of change management) and the relationship between industry and regulators and importantly between regulatory assessment and inspection needed to support strong implementation of the approaches within Q12. The draft guideline clearly already provides tools and approaches for change management of immense potential value. Nevertheless, the opportunity to comment on the draft is always an important step in the development of an ICH guideline, and it is important to ensure that comments assist in providing the clearest possible final guidance that will be readily and consistently implemented to mutual industry and regulator benefit. It is noteworthy that the current draft of the guideline includes wording suggesting that some concepts may not be implementable at the current time across every region. It will be of greatest benefit if the tools and approaches as described and agreed in the finalized guidance will be available for use on as wide a global basis as possible, in line with the ongoing vision of ICH for science-based, harmonized, and efficient regulation of pharmaceuticals.
■
PRIOR KNOWLEDGE Prior knowledge has always been an important tool in designing manufacturing processes and control strategies and is highlighted in ICH Q8−Q11 and other regulatory guidelines as a core consideration supporting enhanced development. The role of prior knowledge in both enhancing and streamlining development has long been accepted. However, there has never been specific guidance on how to apply prior knowledge. As enhanced approaches to development and regulatory submissions have matured, industry and regulators have become more accustomed to using platform and prior knowledge during development and in regulatory files, and it has become recognized that opportunities exist to use prior knowledge in a more consistent and harmonized manner. For this reason, in November 2017 the European Medicines Agency (EMA) Biologics Working Party (BWP) and Quality Working Party (QWP) hosted a joint workshop on the use of prior knowledge. The scope included active pharmaceutical ingredients (APIs) and drug products and covered chemical and biological products. The workshop was attended in person by 51 regulators from the national competent authorities/EMA and 49 industry representatives and was broadcast live on the EMA Web site. The aim of the workshop was to address what prior knowledge entails and how it can be used to support product development, manufacturing, and control strategies. General discussions were further elaborated through a number of specific industry case studies and a discussion of experiences to date of accelerated access schemes. The workshop addressed the following: 1. What is prior knowledge? 2. How can prior knowledge be used in product development? 3. How should prior knowledge be used and justified in regulatory submission? B
DOI: 10.1021/acs.oprd.8b00186 Org. Process Res. Dev. XXXX, XXX, XXX−XXX
Organic Process Research & Development
Regulatory Highlights
Figure 1. Purge factor decision tree for use under ICH M7.
prior knowledge should be made very clear, i.e., which routine data collection or controls does it replace? The extent of data provided in support of prior knowledge claims should be commensurate with the risk and extent of intended use in order to allow for an efficient benefit−risk evaluation by regulators. If in-house knowledge from related products is used, the data and source should be identified as appropriate and differentiated from product-specific data. Tabular or graphical presentations are preferred for data that are amenable to such an approach. Prior knowledge should be presented in supportive dossier sections (3.2.S.2.6, 3.2.A, 3.2.R) to support the control strategy established for the product. The EMA encouraged submission of prior knowledge in scientific advice applications and marketing authorization application (MAA) files and encouraged applicants to discuss the appropriateness and extent of use of prior knowledge with the EMA early in development in order to agree on the best way to present prior knowledge in the MAA dossier. Conclusions. Prior knowledge is an established tool that is explicitly or implicitly used to inform decisions during pharmaceutical development and lifecycle management. As ICH Q8− 11 approaches to development mature, more opportunities for the use of prior knowledge to support consistent, streamlined development and justification of established control strategies in regulatory submissions are becoming available. Equally, it is essential that all regulatory regions, particularly the U.S., Japan, and the EU, take a harmonized view of how prior knowledge can be effectively used to fully realize the potential of such approaches.
■
In relation to purge calculations and the use of in silico tools, an industry consortium has continued the development of Mirabilis. Parallel to the evolution of the software, the consortium recently published9 a set of guiding principles seeking to clearly and consistently show how purge factors align directly to control options 1, 2, 3, and in particular 4, as described in ICH M7.1 These principles set out a clear process defining how the purge ratiopredicted versus required or excess purge capacity of a manufacturing process for a specific potential mutagenic impuritycan be used to determine what control option is most appropriate and the extent of additional data required to support the selected control option. This is illustrated in Figure 1. When the purge ratio is greater than 1000, it proposes that no further experimental data are required to justify the control strategy for the impurity being assessed. As the ratio decreases, understandably the burden of proof in terms of supporting experimental data increases. For example, for purge ratios between 10 and 100 for a potential mutagenic impurity in a commercial API route, detailed experimental fate and purge studies are expected to support a commercial option 4 control strategy. It is hoped that this publication will further augment the consistent and appropriate use of the purge factor approach throughout the development lifecycle of the manufacturing route to inform control strategy selection and planning for efficient experimental data collection. In 2017, an addendum table was added to ICH M71 in which specific limits for a series of mutagenic carcinogens were published. Early in 2018, an industry paper was accepted for publication10 that provided definitive limits for an additional 24 commonly used compounds, in many cases directly addressing previously misassigned concerns over mutagenicity (e.g., acetaldehyde, acetamide, and formaldehyde). The actual permissible daily exposures (PDEs) established differed by several orders of magnitude from the default threshold of toxicological concern (TTC) of 1.5 μg/day (e.g., proposed PDE for acetamide = 7.1 mg/day.) Sharing knowledge such as this openly in the literature is vital to ensure that attention is focused correctly on compounds of highest toxicity concern, and the authors encourage the scientific community to continue sharing this knowledge openly.
MUTAGENIC IMPURITIES
With the advent of ICH M7,1 which was finalized in 2014, as an industry we entered a phase of maturity regarding this topic, with the guideline providing a very effective framework for practical management of mutagenic impurities. The focus now is on support of the process, and to further this we have seen a series of helpful refinements that focus on key areas such as flexible control limits (i.e., limits linked to duration of exposure based on Haber’s law), the use of purge calculations, and in silico assessments and specific limits. C
DOI: 10.1021/acs.oprd.8b00186 Org. Process Res. Dev. XXXX, XXX, XXX−XXX
Organic Process Research & Development
■
Regulatory Highlights
ELEMENTAL IMPURITIES Efforts in this area are currently focused on three fronts: • Finalization of risk assessments to ensure compliance with the ICH Q3D guideline for all products supplied to those markets having implemented ICH Q3D and to the date for implementation • Continued development of ICH Q3D dermal limits • Removal of the heavy metals limit test USP Marketed Product Compliance. When it was published at the end of 2014, ICH Q3D2 provided a 3 year moratorium in relation to established products, meaning that all such products would have to demonstrate compliance with the guideline at the end of 2017. Many involved will testify to the Herculean effort required to complete this within large organizations where hundreds if not thousands of products were within scope. What has been the outcome? Informal feedback within the industry is that aside from a small number of products, organizations have found that the vast majority of products assessed require no additional control measures because they already have appropriate quality control measures. Elemental Impurities within Excipients. The ICH Q3D guideline describes how a risk-based approach to the control of elemental impurities in drug products can be taken, highlighting within this that assessments should be data-driven. Options in terms of data include both data generated specific to a drug product and published data. In 2015 the U.S. Food and Drug Administration (FDA) and the European International Pharmaceutical Excipient Council (IPEC) jointly published the outcome of a focused study on some 200 excipient samples covering a range of excipients. This concluded that the overall risk associated with excipients, including those that are mined, was relatively low, especially when typical proportions in formulated drug products were considered. With the express aim of building upon this initial study, a consortium of pharmaceutical companies has established a database to collate the results of analytical studies of the levels of elemental impurities within pharmaceutical excipients. This database currently includes the results of over 25 000 elemental determinations for over 200 different excipients and represents the largest known, and still rapidly expanding, collection of data of this type. A recently published analysis of the database11 examined a series of aspects, including data coverage as well as impurity levels and variability (across supplier/grade, etc.). The database includes results from multiple analytical studies for many of the excipients and thus can give a clear indication of both excipient supplier and batch-to-batch variability as well as any variability associated with the different testing organizations and methods employed. The results are telling. Critically, the data confirm the findings of earlier, smaller FDA−IPEC studies showing that elemental impurity concentrations in excipients, including mined excipients, are generally low and when used in typical proportions in formulated drug products are unlikely to pose a significant patient safety risk. The database is now in active use within member organizations, providing real evidence in support of holistic ICH Q3D risk assessments and in the future potentially significantly reducing the need for testing. However, it is necessary to recognize that there was a sense that mined excipients could still present a risk over the long term. That variability in elemental impurity levels within mined excipients will vary over time, and further data will be required. There is therefore a need for continued collaboration between the pharmaceutical industry and excipient manufacturers.
It is interesting to reflect that had such studies been conducted ahead of finalization of ICH Q3D, it is possible that it would have allowed us to eliminate concerns about elemental impurities, at least for some low-risk excipients Another study could have achieved the same outcome for manufacturing equipment. Removal of Heavy Metals Testing. Perhaps our biggest challenge as an industry in this area relates to the potential to remove existing empirical testing for elemental impurities using the wet-chemistry heavy metals limit test because of differences in the global regulatory landscape. In the case of the United States Pharmacopeia (USP), this takes the form of the nowdeleted USP Chapter . On the basis of the time scale for implementation of ICH Q3D, most organizations are well-advanced in terms of the risk assessment of current products, as described above. In the clear majority of cases, this successfully demonstrates that the heavy metals test does not provide any additional control for elemental impurities. On this basis, it should therefore be possible to remove the heavy metals limit test, of which USP is the most prevalent example. The situation in the U.S. is that removal is relatively straightforward, as the test has already been removed from the USP. A statement to confirm completion of an elemental impurity risk assessment is then provided in the product annual update. Elsewhere, the situation is more challenging. In Europe there is no definitive position, but filing a simple show-and-tell type 1A variation seems to provide a pathway. Thereafter, the situation is considerably more complex. In Japan, the equivalent of the USP test has been retained in the Japanese Pharmacopeia (JP). Consequently, removing the test from an existing product (one where a monograph is published and it includes such a test) may require submitting a product-specific request to revise the individual monograph. It is also anticipated that removal of the test from approved but not monographed products will also require a post-approval change submission. In China, the Chinese Pharmacopeia (CP) will retain the test until at least 2020, and the indication is that the test should still be performed where registered. Outside of ICH regions, the situation is still more complicated. Given the prevalent position of the USP in many countries, API and product specifications often include USP . However, this test no longer exists! The challenge then concerns whether the test can be removed and the specification revised, and if so, how this should be done. The scale of this is significant, especially if a formal variations procedure is needed. One apparent option is to continue testing, but even this is complicated, as it is not clear how one could continue to use a test that no longer exists in the USP. Some organizations have even considered developing a “USP -like” test. Clearly, organizations do not want to continue to use an empirical test when a risk assessment has shown that it adds no value, but at present there is no obvious way to resolve this conundrum for globally marketed products until significant harmonization in compendial test requirements is achieved.
Andrew Teasdale*,† Matthew Popkin‡ Ron Ogilvie§ †
AstraZeneca, Macclesfield SK10 2NA, United Kingdom Chemical Development, GlaxoSmithKline, Tonbridge TN11 9AN, United Kingdom § Pfizer, Sandwich CT13 9ND, United Kingdom ‡
D
DOI: 10.1021/acs.oprd.8b00186 Org. Process Res. Dev. XXXX, XXX, XXX−XXX
Organic Process Research & Development
■ ■
Regulatory Highlights
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected].
REFERENCES
(1) Assessment and Control of DNA Reactive (Mutagenic) Impurities in Pharmaceuticals To Limit Potential Carcinogenic Risk M7(R1), Current Step 4 version, dated March 31, 2017. (2) Guideline for Elemental Impurities Q3D, Current Step 4 version, dated Dec 16, 2014. (3) Pharmaceutical Development Q8(R2), Current Step 4 version, dated August 2009. (4) Quality Risk Management Q9, Current Step 4 version, dated Nov 9, 2005. (5) Pharmaceutical Quality System Q10, Current Step 4 version, dated June 4, 2008. (6) Development and Manufacture of Drug Substances (Chemical Entities and Biotechnological/Biological Entities) Q11, Current Step 4 version, dated May 1 2012. (7) Final Concept Paper Q12: Technical and Regulatory Considerations for Pharmaceutical Product Lifecycle Management, dated July 28 2014, endorsed by the ICH Steering Committee on Sept 9, 2014. (8) Technical and Regulatory Considerations for Pharmaceutical Product Lifecycle Management Q12, draft version endorsed on Nov 16, 2017. (9) Barber, C.; Teasdale, A.; Antonucci, V.; et al. A consortiumdriven framework to guide the implementation of ICH M7 Option 4 control strategies. Regul. Toxicol. Pharmacol. 2017, 90, 22−28. (10) Bercu, J.; Galloway, S.; Teasdale, A.; et al. Potential impurities in drug substances: Compound-specific toxicology limits for 20 synthetic reagents and by-products, and a class-specific toxicology limit for alkyl bromides. Regul. Toxicol. Pharmacol. 2018, 94, 172− 182. (11) Boetzel, R.; Ceszlak, A.; Day, C.; et al. An Elemental Impurities Excipient Database: A Viable Tool for ICH Q3D Drug Product Risk Assessment. J. Pharm. Sci. 2018, DOI: 10.1016/j.xphs.2018.04.009.
E
DOI: 10.1021/acs.oprd.8b00186 Org. Process Res. Dev. XXXX, XXX, XXX−XXX
