How much do you know about the composition of the excipients you are using? How much does anyone know about the composition of the excipients they work with? If we do not know about the composition of our excipients, can we efficiently design and develop robust formulations with an adequate design space? My point is that we really do not understand enough about the excipients we use. This is not meant as a criticism, so much as a plea; a plea to share the information we have about our formulations, and in particular about our excipients, that can be disclosed to the public domain (work on placebo formulations, work on formulations of drugs that never made it through development, work on formulations of drugs that have been withdrawn from the market, etc.). No one knows it all, and I suspect that formulation scientists are working in isolation, continually discovering what has already been discovered elsewhere. This is wasteful of resources at a time when our industry is under pressure to reduce the prices of medicines, and when industry is looking to cut costs.
This was the premise for the development of the Katalog der Pharmazeutische Hilfstoffen by the major Swiss pharmaceutical companies in the 1970s, and which, in turn, inspired the development of the Handbook of Pharmaceutical Excipients, the 6th Edition of which was published recently. The Handbook is not perfect, but it is a good start. However, it is only as good as the information that is available, and to emphasize my earlier point, I suspect a considerable amount of information is not available in the public domain.
In this column, I want to explore several questions concerning excipient composition. I use the word explore because there are no hard and fast answers to many of the questions. In part, the answers will depend on whether or not we really understand an essential difference between excipients and active pharmaceutical ingredients (APIs). This difference is that APIs are in the formulation to treat the patient, excipients are there to help convert the API into a medicine that the patient can use; they bring functionality or performance to the formulation. Unformulated, most APIs are quite inappropriate for patient use.
An essential element of Quality-by-Design (QbD) is that we are able to show increased understanding of our formulations. Part of that increased understanding must relate to excipients since excipients are one of the three components of a pharmaceutical formulation (along with the API and the processing). The important questions are therefore; “How does excipient performance arise?” “What is the composition of a particular excipient?” and “How does excipient composition influence excipient performance?” I use the word performance because, like Greg Amidon (University of Michigan, and Chair of the USP Excipient General Chapters Expert Committee), I think that ‘functionality’ is a horrible word! We all know what it means, but I much prefer the term ‘performance’ which is the term I shall use in this article where possible. I also want to explore the question of impurities in pharmaceutical excipients.
Excipients are a very diverse group of materials. They comprise all the different states of matter; gas, liquid, semi-solid and solid, many different chemical types, such as: inorganics, carbohydrates, hydrocarbons, amino-acids, oligopeptides and proteins, synthetic polymers, natural polymers and other materials, and they can be of animal, vegetable, mineral or synthetic origin. There is also now an excipient/adjuvant manufactured using recombinant technology. Excipients may be harvested in some of the least developed areas of the world, or they may be manufactured in large, modern chemical plants using comparatively sophisticated chemical technology, and everything in between. This is part of the problem with excipients; they are impossible to categorize simply, and there are often as many exceptions as there are examples that prove the rule.
How Does Excipient Performance Arise?
In answer to this question, my reply is that excipient performance must derive, in part, from the chemical composition of the material, and in part from its physical structure (including polymorphic forms). At first glance, this sounds quite straightforward, but it isn’t, and this is what can catch people out. If there is one message for you to take away from reading this article; it is that, for the most part, excipients work because they are not pure, but are in fact mixtures containing different minor components which are necessary for their performance. Hiroto Miyamoto, formerly of JPEC (Japanese Pharmaceutical Excipients Council) has termed these components functional components. The USP refers to them as concommitant components. I much prefer Miyamoto-san’s terminology. Although having regard to Greg Amidon’s preference, we should probably call them performance components.
A further important point to understand is that these functional components or concomitant components are not impurities in the API sense. (In my opinion, ‘impurities’ is a term that should be reserved solely for APIs.) These functional components are very necessary to achieve optimum excipient performance in the formulation. Let me give you a couple of examples to illustrate this point.
Dibasic calcium phosphate dihydrate is a common pharmaceutical excipient, and the coarse or un-milled grade may be used in the manufacture of tablets by direct compression. This material comprises monoclinic crystals/crystal fragments which deform by brittle fracture during tablet compaction. It is possible to make very pure dibasic calcium phosphate dihydrate today, using a precipitated calcium source and very high purity phosphoric acid. However, this material does not work as well in direct compression. Quite simply, without other ions to disrupt the crystal lattice arrangement, the individual crystals of the ultra pure dibasic calcium phosphate dihydrate do not fracture in the same manner as the regular material during compaction. It appears that dibasic calcium phosphate dihydrate requires some dislocations in its crystal lattice to acts as point defects, and encourage the fracture of the particles, and that foreign ions provide these dislocations. Thus we need a certain quantity of foreign ions for optimum performance; not too many, and not undesirable ions such as lead, but not too few either.
Another point that many people may not be aware of is that dibasic calcium phosphate dihydrate, although somewhat stable at room temperature, is unstable at elevated temperature; even quite modest elevated temperatures (<100ºC). In practice, the surface of the dibasic calcium phosphate dihydrate crystals is converted to calcium pyrophosphate to stabilize the material. Dibasic calcium phosphate dihydrate will dehydrate to form the anhydrate. The surface of the anhydrate appears to be more acidic than the dihydrate [1], and dehydration to the anhydrate will release a lot of water of crystallization. There is an accompanying change in the crystal habit from monoclinic to triclininc. This can have implications for film coating, particularly modified release coatings, and packaging.
Microcrystalline cellulose is a very popular excipient and has many uses in formulation science. It is prepared by the acid hydrolysis of wood pulp. But what does it contain, because it is not all α-cellulose (also known as cellulose-I)? Some of the minor components include cellulose-II, hemicelluloses, sugar residues (from the hydrolysis), formic acid residues and ammonia residues. In addition, different pulps seem to have a different optimum degree of polymerization value (indicative of polymer chain length). If we over hydrolyze or under hydrolyze the pulp we will not get optimum performance. What this all suggests to me is that we do not know enough about the composition of microcrystalline cellulose, and that degree of substitution may be a poor surrogate for performance.
The more important point is that we do not know enough about the composition of microcrystalline cellulose to be able to say which component is important for maintenance of performance in any pharmaceutical application. And that is more of a concern.
What is the Composition of a Particular Excipient?
It should be apparent from the dibasic calcium phosphate and microcrystalline cellulose examples described in the preceding paragraphs that the composition of excipients, and its implications for formulation performance and stability, can be complex; not just for polymers, but also for supposedly simple molecules. In addition to the components derived as a consequence of the raw materials and processing, there may be other components. For example there can be processing aids and additives. Processing aids are used to improve some aspect of the manufacture or isolation of the excipient, for example antioxidants to suppress oxidative side reaction, or surfactants to improve the removal of oily residues from a raw material. Additives, by contrast, are added after final isolation to improve the storage or handling of the excipient, for example anti-caking agents.
The International Pharmaceutical Excipients Council (IPEC) has recently published a guide to Excipient Composition. They have been working to address these, and other issues, to try and provide some much needed understanding of excipient composition. In a QbD world, we do need to understand excipient composition better, and in particular we need to understand the composition profile of our excipient. (This is analogous to an impurity profile for an API.)
To paraphrase this IPEC Guide, excipients can include several different components, including: the nominal component, concomitant components, additives, processing aids, degradants, residual solvents, unreacted starting materials, residual catalysts or metallic reagents, reaction by-products or raw material components. Some of these will be present at very low levels in the final excipient, but can we state categorically that they do not influence functionality? – I don’t believe so. What we can state is that there are certain components, such as toxic heavy metals, that are undesirable and should be kept to a minimum.
How Does Excipient Composition Influence Performance?
The examples of dibasic calcium phosphate dihydrate and microcrystalline cellulose given above show that excipient composition does influence excipient performance, and these are not isolated instances.
Polyethylene glycol is available in a wide range of different pharmaceutical grades; some are liquid, some are semi-solid. It is manufactured by the reaction of ethylene oxide and water at elevated pressure, and in the presence of a catalyst. However, all grades can undergo autoxidation and may give rise to peroxides and free radicals [2]. There are two ways to counteract this; the addition of an antioxidant or the use of an inert atmosphere, such as nitrogen, to exclude oxygen. In this case, the antioxidant is a processing aid, not an additive. A company wished to validate an alternate source of a particular grade of polyethylene glycol for one of their products. Imagine their dismay when the preliminary batches failed on stability. The upshot of the investigation was that the original supplier used an antioxidant as a processing aid, but the alternate supplier used the inert atmosphere method. Unfortunately for the excipient user, the residual antioxidant from the excipient was stabilizing the whole product, hence the stability disaster during the preliminary investigation of the alternate source material.
This raises another point that is often misunderstood by excipient manufacturers and users alike. The USP-NF does not permit additives in materials stated to conform to the relevant monograph, unless specifically permitted in the monograph [3]. However, since few people read the USP-NF General Notices, the use of additives has traditionally not been declared, but they have been used in some excipients for many years, even predating the development of the monograph. The USP has made a concerted effort in recent years to update such monographs to include the presence of an additive, but also to include a labeling requirement that the additive be declared. The current NF monograph for Polyethylene glycol (NF 27, 2009) permits a suitable antioxidant to be included, but less than 10 years ago (NF 19, 2000) there was no such statement.
It is not just the presence of other components that can cause problems. Sometimes a change in polymorphic form of the excipient can present problems. Lactose is a very common excipient and available for pharmaceutical use in different forms and different grades. Lactose is also a reducing sugar and will undergo a Maillard-type reaction with primary or secondary amines. The Maillard reaction has been known for many years. An interesting twist was reported in the early 1970s by Blaug and Huang [4]. These researchers showed quite clearly that spray-dried lactose was more reactive than crystalline lactose monohydrate. It is believed that spray-dried lactose comprises lactose monohydrate crystals stuck together by a thin layer of amorphous lactose. The enhanced reactivity should not be a surprise because the amorphous form is a high energy form. The formation of the amorphous form during spray-drying should also not be a surprise since the removal of the water is very quick and such rapid drying would favor the formation of amorphous material from the lactose present in solution prior to drying. Even today, I am surprised just how many formulation scientists are not aware of or do not understand such very basic chemistry.
Are There Impurities in Excipients?
I would argue no, there are no impurities in excipients! For example, in the case of dibasic calcium phosphate dihydrate cited above, the foreign ions required to disrupt the crystal lattice during deformation could include heavy metal ions such as lead. I suggest that excipient components can be divided quite simply into desirable components and undesirable (potentially toxic) components. In my opinion, the term ‘impurity’ should be reserved for active pharmaceutical ingredients (APIs) where it does have a place. This is possibly an extreme view, but I believe it is justified by our understanding of excipients, and the facts. I will accept that undesirable components should be controlled to below acceptable safe levels. I will also accept that is would be desirable to control certain acceptable components to within a specified range to ensure a consistent performance from the excipient. However, I do not know of any one instance, where I can state categorically, that if we control component ‘x’ between these limits, we will have an excipient giving a consistent performance.
Thus, our lack of understanding of the link between excipient composition and functionality forces us to search for surrogate tests that can be used to try and predict whether or not a particular batch of excipient will be acceptable for a certain application. There is also the question of variability; both between batches and within batches, as discussed in Part IV of this column [5], and we have not even begun to address this. There are plans and proposals for different projects to investigate aspects of excipient performance and variability. But this is not something that can be developed as a short-term project. As I explained at the beginning of this column, we do not know enough about our excipients, and what makes them perform the way they do. In a QbD world, is this acceptable? I do believe that there is plenty of data that could be useful and should not compromise intellectual property, if made public. As I stated above, I would like to see such information published for the benefit of all of us working in pharmaceutical formulation development. A pipe-dream? Maybe! But if we do not ask, we will never get there!
Excipient composition is a complex issue. I do not think there will be any quick fixes, but if we do not start to investigate, we will never find the answers. I hope this column has provided you with some food for thought. The next article in this series will address aspects of risk management as they relate to excipients and QbD.
References
1. Glombitza BW, Oelkrug D, Schmidt PC. Surface Acidity of Solid Pharmaceutical Excipients I. Determination of the Surface Acidity. Eur J Pharm Biopharm (1994), 40 (5), 289-293.
2. Wallick D. Polyethylene Glycol, in The Handbook of Pharmaceutical Excipients, 6th Ed. Rowe RC, Sheskey PJ and Quinn ME (eds.), (2009) American Pharmaceutical Association, Washington, DC and Pharmaceutical Press, London, pp. 517-522.
3. General Notices and Requirements; 5.20.10. Added Substances, Excipients, and Ingredients in Official Substances. USP 32-NF27. United States Pharmacopeia Convention, Inc., Rockville, MD. Volume 1, p. 5.
4. Blaug SM, Huang W. Interaction of dextroamphetamine sulfate with spray-dried lactose. J Pharm Sci (1972), 61, 1770–1775.
5. Moreton RC, Functionality and Performance of Excipients in a Quality by Design World, Part 4: Obtaining Information on Excipient Variability for Formulation Design Space. Am Pharm Rev (2009), 12 (5) July/August, 28-33.
Dr. Moreton has over thirty years’ experience in the pharmaceutical industry. He has worked as a formulation scientist developing a variety of different dosage forms, and has experience in the design, development, scale-up, technical transfer and validation of drug products and associated analytical methods, both during clinical development and eventual transfer into commercial manufacture, and working with licensing partners and contractors. He has also worked in QA/QC, Regulatory Affairs and Technical Support in excipients and drug delivery.
He is a past Chair of the AAPS Excipients Focus Group, and of IPEC-Americas. He is a member of the International Steering Committee of the Handbook of Pharmaceutical Excipients, and of the USP Expert Committee—Excipient Monograph Content 2. He has authored and co-authored scientific papers and book chapters, and lectured extensively in the areas of excipients, drug delivery and formulation at universities, training courses and symposia in the U.S. and Europe.