Regulatory Strategy for Starting Material Designation

Introduction

The purpose of this paper is to lay out the regulatory considerations related to the choice of starting materials and synthetic route for an API. From a regulatory perspective there are no “right” or “wrong” choices in this regard. The difference comes in the regulators’ expectations of the amount of data needed to support long versus short synthetic routes and the corresponding regulatory commitments.

What are regulators concerned about as they review the chemistry sections of clinical trial and marketing applications? Their concerns can be summed up in a single question: Do you know what is in the medication that will be given to patients? The questions that flow from that are related to the drug’s strength, impurities, and release in vivo.

It is the question of impurities that underlies most of the issues that have been raised over the years regarding starting material choice for active pharmaceutical ingredients (APIs). In the global marketplace that currently exists, it is possible to obtain an API from a supplier on literally every continent except Antarctica. While a drug company might choose a component supplier based, in part, on business concerns (i.e., price, reliability of supply), regulators go back to the primary question: Do you know what’s in the final drug product? Regulators want assurance of consistent purity for API starting materials.

Good Manufacturing Practices (GMPs) are required from the starting material forward in the manufacture of an API [1]. Far less regulatory oversight is present in the manufacture of starting materials. In order to ensure consistent quality, regulators would prefer that GMPs extend as far back as possible, thus preferring longer syntheses using simpler starting materials [2]. A longer synthetic route also provides the potential for multiple purification steps which also has the effect of limiting impurities in the final API. Further, regulators seek to limit a firm’s choice of suppliers and the ability to switch suppliers of starting materials in order to ensure a consistent and known level of quality.

Any change in the manufacture of a starting material such as a different synthetic route or altered manufacturing conditions may lead to a different impurity profile for the compound. Novel impurities or increased amounts of previously known impurities can cause changes in the impurity profile of the API either because the impurity passes through the synthesis and purification of the API or because it leads to the formation of novel impurities by reacting with the synthetic reagents along the way.

Impurities in a starting material are likely to be structurally related to the starting material, having arisen through side reactions in the preparation of the starting material and passed through purification steps in the drug substance synthesis due to that similarity. This scenario creates two problems: 1) Detecting a structurally related impurity in the presence of much larger quantities of the API and 2) Removing the similar compound(s) by some type of purification process.

Analytical methods developed to detect and measure known impurities may not detect the novel impurities in the API that result from a change in the purity profile of starting materials. In general, there is a tradeoff between the cost and time to perform analytical testing such as HPLC and the ability of that test to discriminate among a number of structurally related compounds; i.e., rapid screening methods tend to be less discriminating.

Similarly, purification techniques may not be very effective in removing such impurities because they have physicochemical properties close to that of the API. Again, there is a tradeoff between the effectiveness of a purification step in ridding API of a particular impurity and the cost, speed, and yield of that process.

Regulators are well aware of the analytical and purification tradeoffs and so strive to limit the risk of such impurity problems by the following strategies:

1. Seeking synthetic routes to APIs that are several steps long. This has the effect of requiring GMPs further back in the synthetic sequence. There is also the expectation that multiple steps, including isolation and purification would remove most impurities. Finally, the sequence of chemical transformations may cause impurities resulting from reaction of starting material impurities to be sufficiently different in their properties that they can be detected and removed.
2. Desiring to know the synthetic route for the preparation of the starting material and to limit a manufacturer to starting material produced only by one or more approved routes. Knowledge of the route makes it possible to predict the impurities that might be in the starting material and to ensure that analytical methodology is in place to detect them.
3. Limiting a manufacturer to a single or limited number (e.g., 1-3) of suppliers of a starting material. This approach seeks to take advantage of any due diligence performed by the manufacturer on this supplier and to prevent going elsewhere to a starting material supplier who may use a different synthetic route which could bring in novel impurities that existing analytical methods can’t detect and current purification processes can’t remove.

This approach enables regulators to fully understand the quality of the starting material and jointly assess with the sponsor acceptable downstream control strategies.

The goal of working with regulators in the area of starting material selection, then, should be to provide a solid, data-driven case that the choice of starting material is backed up by sufficiently discriminating analytical methodology, appropriately set starting material acceptance criteria, and effective purification processes that we are certain that we know what is in the medicine we will be giving to patients.

The scope of this paper focuses on what documentation we need to provide to regulators in a briefing document to support our starting material designation decision and the strategy involved in the negotiations with regulators around that designation.

Regulatory Environment

U.S. FDA

The FDA’s Guideline for Submitting Supporting Documentation in Drug Applications for the Manufacture of Drug Substances [3] lists the following criteria for defining a starting material:

• It is incorporated into the new drug substance as an important structural element.
• It is commercially available.
• It is a compound whose name, chemical structure, chemical and physical characteristics and properties, and impurity profile are well defined in the chemical literature.
• It is obtained by commonly known procedures (this applies principally to starting materials extracted from plants and animals, and to semisynthetic antibiotics).

The FDA acknowledges that this is not a strict definition by asserting that a starting material will frequently meet several of the criteria, and that generally starting material agreement is reached with the FDA chemist before submission of the NDA, such as during an IND End-of-Phase 2 meeting or pre-NDA meeting.

The FDA’s concerns about impurity control led to a temporary attempt to place rigid constraints on a single element of overall control strategy, starting materials, by forcing control through strict definition (FDA Draft Guideline, Guidance for Industry: Drug Substance: Chemistry, Manufacturing, and Controls Information, Jan. 2004, withdrawn Fed. Regist. Notice June 1, 2006) [4]. The draft guideline was withdrawn partly because of its overly prescriptive selection principles for starting materials including:

• Propinquity – A starting material should be separated from the final intermediate by several reaction steps that result in isolated and purified intermediates.
• Carryover of Impurities – A chemical proposed as a starting material should not be the source of significant levels of impurities in the drug substance. A significant level is considered to be greater than 0.10 percent in the drug substance of the starting material itself, impurities in the starting material, or synthetic derivatives of impurities in the starting material.
• Complexity of Structure – If advanced techniques suitable for complex structures (1H-NMR, 13C-NMR, 2D NMR, mass spectrometry, elemental analysis, X-ray crystallography, chiral HPLC) are needed to distinguish the proposed starting material from potential isomers and analogs, the chemical is not an appropriate candidate for designation as a starting material.
• Although the agency had worthy intentions for developing these selection principles, such dogmatic rules regarding starting materials represented an obstacle for sponsors desiring to discuss alternative strategies for impurity control.

EMEA

The CPMP Guideline on the Chemistry of New Active Substances [5] notes the following concerning a starting material:

• It is incorporated as a significant structural fragment into the structure of the active substance.
• It should be fully characterized to ascertain suitability for intended use and complete specifications should be provided, including an impurities profile.
• The name and address of starting material supplier(s) should be provided.

Furthermore, the guideline states that starting materials should be justified, and that a reaction scheme showing their synthesis may be useful to evaluate the suitability of their specifications. The possibility that impurities present in a starting material may be carried through the synthesis/process unchanged or as derivatives should be discussed in the application.

ICH

ICH Q7 Good Manufacturing Practice Guide for Active Pharmaceutical Ingredients [1] includes the following points regarding a starting material:

• It is a raw material, intermediate, or an API that is used in the production of an API.
• It is incorporated as a significant structural fragment into the structure of the API.
• It can be an article of commerce, a material purchased from one or more suppliers under contract or commercial agreement, or produced in-house.
• It has defined chemical properties and structure.
• The guideline maintains that the company should designate and document the rationale for the point at which production of the API begins.

Although ICH Q8 Pharmaceutical Development [6] does not specifically address starting materials, it provides for the concept of quality-by-design where an overall control strategy, including starting material selection and justification, can be discussed. An appropriate design space may provide regulatory flexibility for the sponsor while assuring regulators that drug substance quality is not compromised by changes within the design space.

Additional guidance on starting materials is expected to be contained in ICH Q11 Development and Manufacture of Drug Substances, which is currently at an early drafting stage and has not yet been published as a draft.

The current guidelines all contain requirements for the starting material to be fully characterized and to contribute structurally to the API. Other starting material conditions outlined by the guidances provide additional clarity for the sponsor to consider holistically rather than as separate conflicting interpretations.

Strategy and Negotiation

The CMC team will eventually reach a point in development where a discussion of starting materials and impurity control strategy with regulators is necessary. The timing of the discussion should ensure that any new proposals or strategy changes will allow sufficient time before MAA submission to fully develop analytical and process data to support the change.

Typically, this discussion takes place with the U.S. FDA at an End-of-Phase-2 CMC meeting. The much later timing of a Pre-NDA meeting would likely not allow enough time to incorporate strategy changes without delaying the planned submission of the marketing application. Meetings with European or Japanese regulators to discuss these issues are not commonly requested (or granted), however, agreements with the FDA have generally proven to be acceptable to other regulators as well. Furthermore, the FDA encourages sponsors to discuss these issues during development, whereas access to other agencies is more restricted and advice is less binding.

A briefing document containing proposed starting materials and impurity control strategy must be assembled and shared with the FDA prior to the meeting. The briefing document should clearly explain how the starting material proposal meets the expectations of relevant guidelines and/or addresses the agency’s concerns. After reviewing the document the FDA may provide a written assessment of the issues, possibly with counter-arguments supporting a more conservative approach. The CMC team must decide whether to accept the agency’s suggestions or to proceed with the meeting to defend the original proposals or possibly suggest alternative strategies.

If the CMC team decides to move forward with the planned meeting, it should prepare for the meeting by gaining an understanding of the FDA’s explicit and implicit concerns. As previously discussed, the agency will typically attempt to restrict impurities in the drug substance by requiring a longer API synthetic route, specifying the route of starting material synthesis, and/or identifying a single starting material supplier. The meeting discussion can only focus on the briefing document contents. The FDA will not typically discuss new information provided in response to any FDA comments.

The CMC team must therefore anticipate these challenges and be prepared to justify its impurity control strategy proposal by providing data supporting the following three primary control aspects [7,8]:

1. Analytical capability
2. Process purification capability (impurity rejection)
3. Starting material acceptance criteria

Ideally, the briefing document should contain such information to support the robustness of the overall control strategy. For example, data from the following may be considered for inclusion in the briefing document:

• Starting materials from several sources should be analyzed for impurity identification/content and forward-processed to manufacture API. The API and intermediates en route should also be analyzed.
• Starting materials manufactured via multiple synthetic routes should be analyzed and forward-processed as above.
• Other potentially viable synthetic routes not utilized should be discussed. Specifically, an assessment of the commercial viability of suppliers (e.g., ability to reliably produce at commercial scale) should be discussed as part of the overall decision,
• Analytical method capability should be discussed. HPLC stacked chromatograms are particularly illustrative of impurity discriminating abilities of methods. It may be helpful to synthesize potential impurities and to calculate response factors to demonstrate appropriate impurity quantitation. Detailed method procedures and/or validations should not be included in a briefing document, but rather, should be available upon request.

The CMC team should be prepared for close scrutiny of any data in the briefing document; experience has shown that regulators may attempt to limit impurities by proposing a single starting material supplier or a single route that results in the cleanest API. Rarely do different vendors use the same synthetic route for a moderately complex starting material; hence the CMC team may need to augment its proposal by pointing out the importance of providing for multiple starting material suppliers/routes in order to ensure a continuity of supply. Prior to the meeting the CMC team should agree on backup positions or counter-proposals in the event the regulators remain unconvinced of the merits of the initial proposal. The team should avoid economic- or price-based discussions of starting materials/vendors, as that is not a concern of the regulators.

While it is impossible to predict the exact discussion flow of a meeting with regulators, the justification of any robust control strategy proposal should emphasize the aforementioned concepts of analytical capability, process purification capability, and starting material acceptance criteria. At times, the regulators may just need a discussion of existing data to show that their actual concern of protecting the public is in fact met by the original proposal.

The following three scenarios present the types of discussion with FDA that are likely to take place at an End of Phase II meeting, either face to face or by telephone. Each case provides a successful negotiating strategy based on actual discussions with FDA. Note that the strategy is the same in every case: 1) Seek clear understanding of the FDA position, 2) Bring the discussion to the data presented on the analytical capabilities, process purification capability, and the effectiveness of the starting material acceptance criteria, and 3) Seek agreement that the data package is adequate to support the choice of starting material while understanding that the FDA must review the actual NDA submission to come to a final determination.

Regulators may seek to limit the number of suppliers of starting materials to a few or even only a single supplier. It is important in this case to be able to demonstrate that possible impurities have been considered and synthesized, if necessary, in order to demonstrate that the method for acceptance of the starting material can detect them. Similarly, potential reaction products of such impurities that might contaminate the drug substance should be prepared so that a related substances test performed on the API can detect and quantify them. In the same way it is important to test the purification processes to be used in the drug substance synthetic route proposed. In this way it can be shown that lots of starting materials with higher levels of impurities can be used because the process is capable of removing them down to acceptable levels. A useful way of demonstrating this process capability is to deliberately add excess amounts of likely impurities to the starting material to determine whether the purification process was capable of removing them and any resulting reaction products.

Another approach by regulators is to request/require a longer synthesis. By extending the route, the number of operations performed under GMPs is increased and, in the regulators opinions, the assurance of quality of the resulting API is increased. Here again, providing data which demonstrate the capability of the analytical methods proposed is an important first step in responding. If potential impurities have been prepared and shown to be reliably detected by the methods, the pressure to begin GMP operations further back is reduced. If, in addition, the purification capability of the proposed manufacturing process has been shown to be able to address these potential impurities, even when added in excess, then it is harder to argue for a longer synthetic route.

Another concern that may be raised by regulators is that once the starting material(s) is agreed upon, it may be manufactured by any route including some that may not have been considered in the development of the analytical methods, purification processes, and starting material acceptance criteria. Since a novel synthesis of the starting material may present unexpected impurities, the analytical methods may not be able to detect them, the purification process may not remove them, and the starting material acceptance criteria may not account for them. In such a case, a reasonable fallback position may be to agree that only a limited number of synthetic routes to the starting materials will be employed by suppliers: those for which the analytical methods, purification process, and starting material acceptance criteria have been demonstrated to be adequate. Such a commitment would mean that any change of synthetic route by a supplier would have to be reported by the supplier to the sponsor and a post-approval manufacturing supplement (variation) would need to be filed with data qualifying the new route. This is a compromise that retains flexibility for the sponsor to let suppliers use any route that has been qualified, but ensures regulators that any new route would be submitted for their review.

Conclusion

The choice of starting materials and synthetic route for an API is a critical decision that must take into account Cost of Product Sold, Quality, and continuity of supply. Of those concerns, regulators are focused primarily on Quality, leaving the others for industry to manage. This paper has presented the regulatory considerations that should be taken into account when selecting starting material and synthetic routes. From a regulatory perspective there are no “right” or “wrong” choices, only differing amounts of data necessary to support long versus short synthetic routes.

If we can demonstrate that we have the following:

• The analytical capability to detect impurities which may arise from different synthetic routes to a starting material or from the reaction products of impurities in the starting material at levels that ensure the resulting API will be of high purity,
• The process purification capability to remove such impurities to ensure that the resulting API is of high purity,
• Robust acceptance criteria for the starting material(s) so that any impurities which may arise from supplier or synthetic route changes will be detected and controlled to appropriate levels in the final API,

then we will be in a strong negotiating position with regulators.

The FDA’s willingness to meet at the end of Phase 2 to discuss the proposed synthetic route and starting materials for the API makes this an appropriate time to address these issues. A successful negotiating strategy was presented. The strategy uses the following elements: 1) Seek clear understanding of the FDA position, 2) Bring the discussion to the data presented on the analytical capabilities, process purification capability, and the effectiveness of the starting material acceptance criteria, and 3) Seek agreement that the data package is adequate to support the choice of starting material while understanding that the FDA must review the actual NDA submission to come to a final determination. We can then use the results of the FDA end of Phase 2 meeting to help us plan our submission to the EU, Japan, and the rest of the world.

References

1. ICH Q7A Good Manufacturing Practice Guide for Active Pharmaceutical Ingredients (2000). http://www.ich.org/LOB/media/MEDIA433.pdf
2. Illing, GT, Timko, RJ, and Billet, L, Drug Substance Starting Material Selection, Pharmaceutical Technology, Volume 32, Issue 12, pp. 52-57 (2008) http://pharmtech.findpharma.com/pharmtech/article/articleDetail.jsp?id=570142&sk=&date=&pageID=5
3. Guideline for Submitting Supporting Documentation in Drug Applications for the Manufacture of Drug Substances (1987) http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm070632.pdf
4. (FDA Draft Guideline, Guidance for Industry: Drug Substance: Chemistry, Manufacturing, and Controls Information, Jan. 2004, withdrawn Fed. Regist. Notice June 1, 2006).
5. CPMP Guideline on the Chemistry of New Active Substances (2003) http://www.emea.europa.eu/pdfs/human/qwp/013096en.pdf
6. ICH Q8 Pharmaceutical Development http://www.ich.org/LOB/media/MEDIA4986.pdf
7. Gavin, PF, Olsen, BA, Wirth, DD, and Lorenz KT, A quality evaluation strategy for multi-sourced active pharmaceutical ingredient (API) starting materials, Journal of Pharmaceutical and Biomedical Analysis, Volume 41, Issue 4, 16 June 2006, Pages 1251-1259 http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6TGX-4JRVDX0-H&_user=499562&_rdoc=1&_fmt=&_orig=search&_sort=d&_docanchor=&view=c&_searchStrId=994037964&_rerunOrigin=google&_acct=C000024438&_version=1&_urlVersion=0&_userid=499562&md5=3b062dedd59d530e86754d2eee33bb34
8. Gavin, PF, Olsen, BA, A quality by design approach to impurity method development for atomoxetine hydrochloride (LY139603), Journal of Pharmaceutical and Biomedical Analysis, Volume 46, Issue 3, 13 February 2008, Pages 431-441 http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6TGX-4R466BF-1&_user=499562&_rdoc=1&_fmt=&_orig=search&_sort=d&_docanchor=&view=c&_searchStrId=994039439&_rerunOrigin=google&_acct=C000024438&_version=1&_urlVersion=0&_userid=499562&md5=0e42c2b684d489ddd2eedf36f4a4ee04 
   

William F. Kluttz, M.S. is a Principal Consultant, Global Regulatory Affairs (CMC) with Eli Lilly and Company, Indianapolis, Indiana. Bill has 18 years (the past 15 years in Regulatory Affairs) experience and his current responsibilities include establishing global regulatory CMC strategies for several compounds in the development portfolio, including CMC documentation and submissions to support clinical studies, as well as regulatory commercialization strategies for late-phase compounds.

Bob Seevers joined Eli Lilly and Company in January of 2002 after 8 years with the FDA. At Lilly his title is currently Principal Regulatory Consultant. He provides expertise in the areas of distribution and stability excellence as well as counterfeiting and knowledge and interpretation of regulatory and compliance expectations. Bob is a member of the steering committee of the Pharmaceutical Cold Chain Interest Group of PDA and served as an editor of Technical Report 39, “Cold Chain Guidance for Medicinal Products: Maintaining the Quality of Temperature-Sensitive Medicinal Products through the Transportation Environment.” While he was with the FDA, Bob was a chemistry reviewer in the Divisions of Medical Imaging, Surgical, and Dental Drug Products, Metabolism and Endocrine Drug Products, and Reproductive and Urologic Drug Products and chemistry review team leader in the Division of Neuropharmacological Drug Products. During this time Bob also served on the Stability Technical Committee for several years, including 2 as Vice-Chair and 2 years as Chair. In that capacity, he was the editor of the 1998 Draft Stability Guidance. Bob received his Ph.D. in Medicinal Chemistry from the University of Michigan in 1981.

This article was printed in the January/February 2010 issue of Pharmaceutical Outsourcing, Volume 11, Issue 1. Copyright rests with the publisher. For more information about Pharmaceutical Outsourcing and to read similar articles, visit www.pharmoutsourcing.com and subscribe for free.