A New Definition for API Starting Materials

May 31, 2004

The FDA guidelines that stipulate the conditions for starting materials are currently being superseded. PhRMA and several European pharmaceutical manufacturers have already added their voices to the ongoing discussion and, as this editorial explains, the debate is far from over. Rules are important, but industry is looking to a future when health care authorities are less inclined to impart total control over every aspect of production.

The discussion surrounding the definition of active pharmaceutical ingredient (API) starting materials (SMs) for registration documentation has recently come to the fore again with the appearance of a new draft of the "Guideline for Submitting Supporting Documentation in Drug Applications for the Manufacture of Drug Substances, "originally published in 1987.1 The issue has been debated within the industry and, more specifically, at a recent conference in Budapest (6th APIC/CEFIC European Conference on Active Pharmaceutical Ingredients, Budapest, Hungary, 5-7 November, 2003), where members of the US Food and Drug Administration (FDA) and representatives of both the North American and European sector were present. Interestingly enough, when the views of all three parties were examined, there was a considerable degree of agreement.2-4 This coincidence, however, was only superficial. The pharmaceutical industry on both sides of the Atlantic is concerned about some of the criteria being used in this document to establish the nature of starting materials for drug substance manufacture.

The defining criteria, contained in the new draft as "Attachment I," is the topic that this editorial wishes to address. As the old guideline states: "What constitutes the 'starting material' may not always be obvious." Because this is indeed the case and discussions mainly focus on this point, FDA has tried to remodel the old definition of a starting material - with less rather than more fortune. The importance of specifying the starting material at some point of the development process is indeed evident. From a regulatory point of view, the action of defining (industry) and accepting (agency) a defined starting material stipulates the starting point of the drug substance registration documentation. From a good manufacturing practice (GMP) stance, and according to the globally accepted International Conference on Harmonization (ICH) guideline ICH Q7a, from this step of the synthesis onwards it is the obligation of the API manufacturer to apply GMPs to the production and control of the API in question, with few exceptions.

It is useful to recall that there is a well thought out definition of SMs in the ICH Q7a guideline. This states that "An API starting material is a raw material, intermediate or an API that is used in the production of an API and that is incorporated as a significant structural fragment into the structure of the API." And, it adds, as typical characteristics, "An API starting material can be an article of commerce, a material purchased from one or more suppliers under contract or commercial agreement, or produced in-house. API starting materials normally have defined chemical properties and structure."

Precision dilemmas A more precise definition, however, has proven extremely difficult to establish - like any intent to define the multiplicity of reality. The options are, it would seem, either to define an SM as something with so many criteria that it becomes impractical or too difficult to comprehend, or to allow so much freedom in the definition that its existence would be meaningless. Thus, the question arises: Is there a real need to define a starting material a priori in a form that could add precision to the well known and widely accepted definition given by ICH Q7a? Or dare we give industry the chance to define an SM on a case by case basis according to the characteristics of the particular process in question, considering risk and science as contributing factors in the decision?

The latter would mean an exact classification, putting the burden on industry to define a starting material for API synthesis on an ad hoc basis, avoiding any prejudgement derived from formalistic or non-scientific reasons. The starting material could be the final intermediate - or indeed any other intermediate - right back to the benzene ring if it proves necessary. But the decision would have to be based on scientific rationale and not on arbitrarily predetermined, peculiar-to-the-process criteria. In some instances, this might make sense, but has proven inadequate in the majority of them.

An additional advantage of this proposal would be to accept, for regulatory purposes, a definition and sum of characteristics of an object given in a GMP document such as ICH Q7a. This would have the bonus of eliminating two statements on the same topic. There is a trend in these discussions to separate - perhaps artificially - the GMP needs from the filing ones. The only acceptable criterion should be the only really essential one: the quality of the API. An adequate starting material could then be defined as something that allows the production of a substance, distinctively showing an appropriate, clearly specified and reproducible quality.

This approach would also place the concept in line with the widely accepted idea of validation. The point at which validation activities are normally initiated is not defined in a general way for API synthesis, but on a case by case basis, for it is very clearly a function of the process itself. Finally, it would comply with the ideas that FDA has been promoting during the last 2 years, establishing a risk- and science-based approach as an effective way of fixing and controlling manufacturing processes.

Perhaps now we can answer the rhetorical question posed earlier. Attachment I of the Guideline should be reduced to two of the sentences included in it, which read: "The starting material should be selected and controlled so that the risk from future changes in the quality of the starting material affecting the identity, quality, purity or potency of the drug substance is minimized. A proposed starting material should be chosen so that sufficient information will be available to FDA on the manufacturing process to evaluate the safety and quality of the drug substance."

The pharmaceutical industry needs rules to be able to ensure that its products are safe and of good quality. Yet, the myriad production and analytical techniques developed by scientists during the last few decades, combined with sophisticated equipment and electronic monitoring devices, allow for thorough process design and controls such that health agencies are now able to place their trust in the pharmaceutical industry. The tendency to regulate every single aspect of the industry's activities should, therefore, become a thing of the past. Many of the current health agency guidelines were written in the 'spirit of authority,' with the attitude that they may and should control everything - otherwise there is no guarantee of quality. This attitude has often produced a climate of mistrust on both sides. The governmental health care institutions of developed countries have begun to realize that this kind of control is simply no longer possible (FDA gives a good example with its new risk- and science-based approach, which industry has unanimously welcomed). One reason for this is the ubiquitous lack of resources but, on another much more important level, it is just no longer adequate. We don't need a guardian, we need a partner. The etymology of the word 'authority' is significant in this environment. It is derived from the Latin 'auctoritas' and this from 'auctor' (creator, historical source). It stems originally from the verb 'augere,' and this, surprisingly enough, means "let something progress." Such health authority, triggering progress through an open attitude, is the dream of our highly regulated industry.

References 1. Food and Drug Administration, Guidance for Industry, Drug Substance, Chemistry, Manufacturing and Controls Information (US Food and Drug Administration, 5600 Fishers Lane, Rockville, Maryland, USA, 9 January 2004).

2. S. Miller, "Status of BACPAC II and of the Revision of the 1987 FDA Drug Substance Guidance," presented at the 6th APIC/CEFIC European Conference on Active Pharmaceutical Ingredients (Budapest, Hungary, 5-7 November 2003).

3. W. Mavroudakis, "Revision of the 1987 FDA Drug Substance Guidance - The US Pharma Industry View," presented at the 6th APIC/CEFIC European Conference on Active Pharmaceutical Ingredients (Budapest, Hungary, 5-7 November 2003).

4. C. Hemmingsen, "Revision of the 1987 FDA Drug Substance Guidance - The EU API Industry View," presented at the 6th APIC/CEFIC European Conference on Active Pharmaceutical Ingredients (Budapest, Hungary, 5-7 November 2003).