Establishing Successful Non-GMP Material Supply Chains for Outsourced API Development and Manufacturing – Evolution of a Phase-Dependent Strategy

• SGL Chemistry Consulting, LLC

Abstract

The development of successful non-GMP material supply chain strategies is the result of sound planning. The ability to devise a plan for the supply of raw materials and Registered Starting Materials (RSMs) depends on an understanding of how the need for these materials evolves and changes over the development cycle, the regulatory requirements that need to be addressed, and the role of sponsors in defining and justifying the non-GMP materials used in their API manufacturing processes.

Introduction

Determining practical sources of raw materials and regulatory starting materials (RSMs) that are not produced under good manufacturing practices (GMP) is often far from a trivial pursuit. In recent years, a number of factors have further compounded the task of sourcing of these compounds:

• The needs and objectives for sourcing in early development differ from those in later development.
• In early development, the objective is to find and secure sufficient materials of the required quality to produce the necessary quantity of API, as quickly as possible.
• In later development, in addition to the basic requirements described above, the ability to establish a sustained supply of non-GMP materials becomes a priority.
• Establishment and demonstration of control over these materials is an essential part of late stage sourcing.
• The concepts and definitions of pharmaceutical raw materials and regulatory starting materials are evolving, due to the increasing complexity of APIs.
• Raw materials do not tend to be as “raw” as they used to be. A steadily decreasing proportion of raw materials are commodity chemicals (commercially available compounds used for purposes other than drug development and manufacturing). If available at all, many of the chemicals currently used as raw materials are small volume catalog items, which, if required in larger quantities, require custom production.
• In addition to the fact that multiple steps are used to produce an RSM, a given RSM supplier is more likely to prepare or source raw materials that require several steps for their preparation.
• There is a more explicit regulatory requirement for a clear strategy for management of risk in the selection of key input materials in API manufacturing. This trend has been apparent in the last 5-10 years, and can be correlated with the two factors cited above.
• In the past, sponsor companies were left more or less on their own to select, control and use the RSMs proposed for a given API manufacturing process, until an End of Phase II (EOPII) regulatory meeting or a meeting prior to submission of a New Drug Application (NDA) provided feedback on the acceptability of the strategy. This is still generally the case, but the amount of preparation to ensure that proposed RSMs have the highest likelihood of being accepted by regulatory agencies has increased.

A discussion of the significance of these factors, and how they impact creation and execution of non-GMP sourcing and supply chain strategies is discussed in the remainder of this article.

Sourcing for Early Development – From Candidate Nomination through Preclinical and Phase I Supply

Sourcing of raw materials and intermediates in early development tends to be expedient and tactical, as opposed to strategic. In many cases, a medicinal chemistry synthetic route to the API is all that is available, and if possible and practical, this synthesis is adapted and taken forward as the early enabling route for the first stage of development, including Phase I clinical supply. A key question to be addressed in early API development is the availability of starting materials and key reagents.

The trend in NCE APIs is one of increasing structural complexity, with regard to scaffold backbones, combinations and placement of functional groups and stereochemistry. Complexity traces upstream in a synthesis to the raw materials for a given API. There is an inverse relationship or trade-off between structural simplicity in raw materials and RSMs and increasing molecular structural complexity. If the materials used to manufacture a complex API are structurally uncomplicated, then the necessary chemical transformations are more demanding and sophisticated. Conversely, if complexity is purchased, in the form of advanced raw materials or RSMs, the GMP process is simpler. Purchase of raw materials and RSMs that are structurally complex means that there is heavier reliance on third party suppliers (i.e., other than the GMP API manufacturer) to shoulder the technical demand of providing these materials with the required quality. As a result, more oversight of third party suppliers of complex non-GMP materials is necessary.

A historically relevant example of the use of starting materials possessing sufficient complexity to act as templates for the synthesis of complex final molecules is the Chiron¹ or chiral synthon approach, which was formalized and advanced by Hanessian. In the Chiron approach, a “chiral pool” of optically pure small molecules, including amino acids, carbohydrates, hydroxy-acids and other naturally occurring compounds are used as synthetic starting materials. Although initially directed primarily toward the total synthesis of natural products, the Chiron approach was and is also adopted and implemented, where applicable and practical, by industrial medicinal and process chemists.

An illustrative example of the Chiron approach as applied to industrial chemistry is the sequence used to generate Fragment B, the side chain precursor, for the first generation Pfizer commercial process for Atorvastatin (Lipitor)², shown below:

In the last 20-25 years, advances in asymmetric synthetic methodology, including chemocatalysis and biocatalysis, have progressed considerably beyond the manipulation of chirons. In addition to the considerable advances made in asymmetric synthesis, catalytic functional group interconversion, transition metal-mediated cross coupling reactions, C-H activation and ring closing olefin metathesis, among other synthetic methodologies, have enabled an expansion in the diversity of chemical space explored.³ This expansion has expectedly given rise to an increase in the structural novelty and complexity of drug candidates advanced from discovery into development.

As more development candidates are nominated that bear novel scaffolds and combinations of functionalities, a challenge arises – when the required quantity of a complex material jumps from a few grams to a kilogram, how is it secured? Some precursors are available in small quantities from catalog companies, but often they need to be custom produced to ensure sufficient quantities and quality. Almost as often, unconventional and/or expensive reagents are needed to prepare them, and these reagents may also need to be custom-prepared. The increasing inability to secure raw materials “off the shelf” adds time and cost to the synthesis of even modestly scaled up batches of complex APIs. The additional time and cost is associated with practicality and quality. Practical, scalable syntheses of complex raw and starting materials must be developed. The materials need to be adequately characterized in order to determine attributes (purity, impurity profile), and use-tested to correlate those attributes with material performance. This challenge is usually not insurmountable at early stages of chemical development and manufacturing. However, as the need arises to develop a process capable of providing increasing quantities of material, corresponding requirements for increases in process understanding, control, a reliable supply chain, and cost of goods (COGs) are factors that must be confronted and addressed.

Sponsor Engagement in Sourcing of Raw Materials and RSMs

When it is clear that sourcing of materials for a given API manufacturing process may not be routine, a sponsor has several options. The first option is to allow the CDMO to handle identification and qualification of suppliers, and then procurement of the necessary materials. Many sponsors will initially defer to the CDMO to procure raw materials and RSMs. All CDMOs in the API manufacturing space have experience in this area, however their resources may be limited for particularly difficult to source materials.

Going forward from early development, the most convenient situation is that the CDMO, by itself, is able to find reliable sources that are able to produce materials in the required quantities and quality. However, supply planning has become more complex, and supply strategy evolves over the course of development. It is up to the sponsor to monitor the situation as development proceeds from early to mid and late stages, in order to gauge if and when a point comes where the limits of the CDMO to source key materials are being approached. At this point, it is necessary for the sponsor, themselves, or with the aid of a sourcing specialist, to do the work necessary to complement the capability of the CDMO in order to establish a reliable supply chain. This is always a collaborative effort, since a critical aspect of establishment of a supply chain is qualification of the suppliers by the CDMO, and this entails analytical, QC and use-testing of materials from suppliers under consideration.

Whereas many raw materials and reagents are available from third party suppliers, the RSM is often custom-produced in-house at the CDMO, given that they are often initially intermediates in the GMP API manufacturing process, and there is the need for significant process and analytical R&D to optimize the chemistry for scale-up and production at the scales required. Once developed, optimized and scaled, the chemistry for the production of RSMs may be transferred to a more cost-effective manufacturing site. Proceeding in this way affords the sponsor more control over the third party production of RSMs, since the processes are better understood, and this can be clearly communicated to potential suppliers in a technical package. Increased process understanding also enhances understanding of the quality requirements for RSMs, and this provides more clarity on the viability of a given precursor as an RSM.

Sourcing for Mid- and Late-Stage Development

As a drug development program transitions from Phase I to Phase II clinical trials, the number and size of the clinical trials tends to increase significantly. An increase in the quantity of API required usually necessitates an increase in process capability, as well as process controls to ensure consistent delivery. It is well understood that, from a quality point of view, the majority of risk associated with a given API manufacturing process resides in its final steps, in which the penultimate intermediate and its impurity profile, along with how the final API is isolated and purified, have the most significant impact on quality. For this reason, the early steps of a given API process are the most changeable, since their expected direct impact on the quality of the final API is low.

There have always been advantages associated with performing some chemical process development work on custom produced raw and starting materials. This work can improve quality of these materials, which also can raise the yields for production of the RSMs. An improvement in the economy of the production of RSMs reduces the volumes of raw materials, solvents and reagents required, resulting in a reduction of cost and cycle time. Process R&D and optimization for raw materials and RSMs also develops understanding that provides a means of control over these materials, including control of impurity profiles and carryover of impurities into the GMP API manufacturing process.

In the past, impact of the quality of the RSMs on the API was addressed by defining them far upstream in the synthesis, and preferably, as made commercially on a large scale, for multiple products and uses, outside of pharmaceutical applications. As a result, there was significant latitude given to the sponsors from regulatory agencies in how this was done. In recent years, with custom-produced RSMs often only a few steps from the API, oversight by regulatory agencies of sponsor proposal and justification of RSMs has become more stringent,⁴ and this has had the effect of moving these considerations to earlier points in the development cycle.

In particular, the European Medicines Agency (EMA) reflection paper on requirements for selection and justification of starting materials of chemical active substances issued in September 2014 has signaled a shift in formal expectations for the rigor and thoroughness with which syntheses of proposed RSMs need to be justified. The problem stated by the reflection paper relates to the perceived increase in risk to the quality of APIs from RSMs in new drug applications, due to three major factors:

• NDAs submitted with very short API manufacturing processes having complex, custom-produced RSMs, which increase the risk of the RSMs affecting the quality of the API.
• The prevalence of API manufacturing outsourced to CDMOs, with RSMs produced by third party suppliers.
• The previous perceived openness of regulatory agencies to interpretation of ICH Q115 on the part of applicants (sponsors) was based on an era when a higher proportion of RSMs were commercial commodity chemicals, as opposed to RSMs requiring multistep custom production.
• This seems to have been misinterpreted and generalized on the part of a significant number of sponsors who have not made the distinction between simpler and more complex RSMs, or considered this distinction when they justified more complex RSMs to regulatory agencies in their regulatory submissions.

The above three factors are compounded by insufficiency of information provided by some sponsors to justify their selection of RSMs in their applications, particularly a lack of assessment of criticality of the synthetic steps for the RSMs, relative to those of the GMP API manufacturing process. Criticality is determined by the potential to impact the quality of the final API. What is required is a data-based justification of the distinction between critical steps that could potentially affect the quality of the final API, and non-GMP steps used to produce a proposed RSM, which are considered non-critical, because they are sufficiently controlled and distant (as regards number of transformations and purifications) from the API, and thereby do not possess significant potential to affect final API quality.

The current regulatory position on starting materials for API manu-facturing processes does not mean that complexity and RSMs are mutually exclusive. But it does mean that there are explicit expectations that, if complex starting materials are to be proposed, requiring multiple steps for their production, then it must be established by the sponsor that the demarcation between critical (i.e., GMP) and non-critical (i.e., non-GMP) steps can be adequately justified, from a scientific point of view.

The practical effect of current regulatory thinking on RSMs is the need for planning and execution that goes well beyond finding a supplier and placing an order for material. As an API process with complex starting materials is scaled up, sponsors need suppliers with the capacity to produce the foreseeable quantities of RSMs, and the skill to demonstrate that the quality of these materials is adequate, and they meet the criteria for being designated as distinct from GMP intermediates in the API manufacturing process. This adds an aspect to non-GMP material sourcing activities that has analogy to outsourcing of chemical development and manufacturing of APIs. Prior to ordering of bulk materials or executing a supply agreement, it needs to be established that each complex RSM is produced with at least a rudimentary understanding of formation, fate and purge of impurities, and of impurity profiles and carryover. This entails evaluation and optimization of synthetic routes, scale up, and analytical method development and validation. In addition to making possible the ability to consistently produce materials with the necessary quality, the rudimentary understanding and control of production of non-regulated materials also provides data that will be submitted in the NDA and is the basis for more in depth justification of RSMs, should the need arise due to push back by regulatory agencies.

Comparison of Priorities in Early and Late Stage Raw Material and RSM Sourcing

For the purposes of this article, there are five elements of a successful material sourcing effort that would lead to a supply chain that enables late stage development and commercial launch:

• Production process capable of supplying adequate inventory for supply needs.
• Timing/logistics/coordination.
• Risk management.
• Technical support for quality/regulatory compliance.
• Cost management.

A comparison of these elements during early and later stage development is provided in Table 1.

Table 1. Summary Comparison of 5 Elements of Successful Sourcing for Custom Raw Materials and RSMs – Early vs. Later Stage Development

Summary

Sourcing of raw materials and RSMs has always been more demanding than assumed by the uninitiated. The challenge of procurement and establishment of a supply chain has been compounded in recent years by:

• Increasing complexity of non-GMP materials, requiring more R&D for process and analytical method development, characterization, scale-up and production.
• This adds time and cost to the work needed to establish a supply chain leading to the delivery of RSMs at mid and late stages of API development.
• The evolution of procurement strategy as a drug candidate is developed.
• The early steps of an API manufacturing process are likely to change as the shift is made away from the early stage enabling process to a mid or late stage process capable of meeting demands for increased quantities, reliable quality and lower cost of API.
• Increased regulatory oversight and stringency of expectations for justification of RSMs.
• Based on the combined effect of increased structural complexity and changeability on the risk to the GMP API manufacturing process, and a perceived lack of information provided in regulatory submissions to allow proper regulatory agency evaluation of whether risk has been adequately managed by the sponsor.

The result of existing and evolving challenges of sourcing and procurement of non-GMP materials for API manufacturing is that more work is required by suppliers, CDMOs and sponsors. This work consists of sufficient understanding of non-GMP production processes and sufficient characterization of non-GMP materials to justify their designation as being distinct from the GMP API manufacturing process. It is ultimately incumbent on the sponsor to ensure that adequate preparatory work is done by their suppliers to support the strategies they have chosen for proposal of RSMs in their regulatory submissions.

Acknowledgements

SGL thanks Sue Wollowitz (Wollowitz and Associates), Danny Levin (Norac Pharma) and Greg Reid (ChemDev Solutions) for helpful discussions and suggestions on content in this article.

References

1a) S Hanessian, Total Synthesis of Natural Products: The ‘Chiron’ Ap-proach, Pergamon Press, Elmsford, NY, 1983, and subsequent editions
b) S Hanessian, Design and Implementation of Tactically Novel Strategies for Stereochemical Control Using the Chiron Approach, Aldrichimica Acta, 1989, 22,3

2a) PL Browser, et. al., The synthesis of (4R-cis)-1,1-dimethylethyl 6-cyanomethyl-2,2-dimethyl-1,3-dioxane-4-acetate, a key intermediate for the preparation of CI-981, a highly potent, tissue selective inhibitor of HMG-CoA reductase, Tetrahedron Lett., 1992, 33, 2279
b) BD Roth,The Discovery and Development of Atorvastatin, A Potent Novel Hypolipidemic Agent, Prog. Med. Chem., 2002, 40

3) O. Mendez-Lucio, J.L. Medina-Franco, The many roles of molecular complexity in drug discovery, Drug Discovery Today, August, 2016

4) European Medicines Agency, Reflection paper on the requirements for selection and justification of starting materials for the manufacture of chemical active substances, September, 2014

5) ICH guideline Q11 on development and manufacture of drug substances (chemical entities and biotechnological/biological entities), Step 3, September, 2011

Stuart G. Levy is currently Principal Consultant at SGL Chemistry Consulting, LLC, a sole practitioner practice founded in 2010, devoted to strategic and technical support of chemical and pharmaceutical development undertaken by start-up and small pharmaceutical companies. Stuart’s practice focuses on the transition of small molecule discovery leads into candidate nomination, preclinical development and early clinical development, but also supports projects at later stages of development, including process validation and NDA submission. Troubleshooting and problem-solving are specialties of the practice, but the preference is to anticipate and avoid as many predictable obstacles as possible. Stuart has 21 years of experience in industry, including 10 years at the bench as a process chemist and 5 years of experience in two permanent positions at the director level, with full responsibility for all CMC activities. In addition to innovators (sponsors), he has worked at two service organizations, a chemistry CRO and a CDMO. Stuart’s expertise in the evaluation, selection, engagement and management of vendors used for outsourced chemical and pharmaceutical R&D and manufacturing enhances his ability to serve his clients Stuart obtained a BS degree in biochemistry and a PhD in chemistry, both from the University of Illinois, Chicago. He was a postdoctoral fellow in medicinal chemistry in the School of Medicine at the University of California, San Diego.