NIPTE, IPEC Work to Standardize Excipient Properties Database

April 14, 2011
Excipients, many of them naturally derived, are a source of variability in drug manufacturing, as differences in processing and other factors can lead to significant differences in the properties and performance of materials of the same grade, not only between different vendors but from the same vendor.

Pharma is a relatively minor customer for commodity excipients, reportedly accounting for roughly 0.02% of the cellulose market, so it has been impossible to get most bulk vendors to tailor product to pharma needs.

As Prabir Basu, head of the National Institute for Pharmaceutical Technology and Education (NIPTE) puts it, “We can send unmanned vehicles to Mars, yet cannot fully understand and control pharmaceutical excipients and predict their performance in the final product.” He believes that industry needs something akin to former president Kennedy’s ambitious space plan, to drive investigation of physical and mechanical properties of materials, to allow users to account for potential impact of interactions with APIs and other ingredients, and process equipment.

Several years ago, professors within the NIPTE consortium of universities began to discuss the idea of developing a database that would allow formulators to understand and predict excipient properties and functionality in pharmaceutical processes. The database is the first step in developing an excipient information management system that can be used to guide formulation development. It would serve as a tool, like the references routinely available to process developers in petrochemicals and chemicals industries.  FDA began to fund this effort about two years ago. Last month, representatives from NIPTE, IPEC, the excipient manufacturers’ trade group, and industry met to review the project. The database would be hosted by Purdue’s PharmaHub and publicly available.

Steve Hoag, professor at the University of Maryland’s (Baltimore) School of Pharmacy described the effort.  Despite references such as the Handbook of Excipients, there is little help available, in any standardized way, to help drug formulation scientists understand the correlation between raw material inputs such as crystal form, particle size distribution and bulk density, to such critical quality attributes as hardness, disintegration time, dissolution, or mixing pattern.

In addition, Hoag described a lack of standard analytical methods. One can measure excipients on a molecular level, via physicochemical tests, on the particulate level, via size distribution, by volume methods such as flow and viscosity. Yet, such connections are essential if pharmaceutical Quality by Design is to take shape. As Hoag concluded in his presentation, drug manufacturers are currently “on thin ice” in terms of their ability to predict any excipient’s performance.

As NIPTE’s Basu has put it, success in manufacturing pharmaceutical dosage forms depends heavily on being able to accommodate the variability in the physical properties of both the active ingredient and excipient raw materials to yield reproducibly high quality products.  The petrochemical industry, which also uses naturally derived raw materials, is able to do this, he said,

Despite a general awareness of the importance of physical properties, formulation scientists face a knowledge gap as they attempt to develop pharmaceutical products and manufacturing processes. Basu explained that certain material properties can be linked to functionality through the establishment of “functionality related characteristics.” These are more easily measured surrogates for the desired functionality, which often can’t be measured at all. Examples would be the degree of gelatinization of start and its impact on dissolution, he said, or Carr Index, Hausner Ratio, or Heistand Index, which can be related to flowability, compression or bonding characteristics.

In addition to clarifying the connection between desired product attributes and mechanical and physical material properties, this database would also allow formulation scientists to understand mixing rules, and be able to assess tradeoffs in performance resulting from varying the levels of different excipients in a formula.

At this point, the database is open access (read only) and contains data from literature and commissioned academic studies. As Hoag summarizes results from the March 31 meeting at Purdue, there is some flexibility in how the data is organized to provide a structure that can grow into a full information management system. To start the process, stakeholders agreed to focus on direct compression excipients. Once the database structure defined, other classes of excipients will be added. Microcrystalline cellulose (MCC) was chosen as the lead excipient for further database development, as it is the most commonly used, high-volume, continuously produced DC excipient.

Hoag says the key features of the database will include:
1. Traceability of information to a test method, equipment and person or organization that supplied the data. It is hoped that the structured entry of information into the database will help to standardize test methods, which will  facilitate comparisons.
2. A structure centered around the listing of compendial excipients, products containing those excipients and data on selected lots of excipients.
3. A setup that will make it easy for one to find key characters or attributes of an excipient and compare by excipient type, different grades or products and functional categories.

At the March 31 meeting, an online demonstration highlighted the key features of the database. Discussions addressed features, ease of use, incorporation of pre-existing user/supplier databases, and possible mechanism to fund the future development of the database.

A key discussion point was that many excipient-related effects on finished product performance are a function of the specific application (formula, process, scale). In other words there may be no general correlation between a specific excipient attribute and finished product performance. However in certain circumstances the variability of the same excipient attribute may be highly significant to finished product performance.

Data on excipient attributes combined with finished product performance data in the NIPTE database would enable multivariate data mining to identify such ephemeral interactions and guide research into their mechanisms. It was noted that PQRI is also looking at this area and the NIPTE database could be the ideal clearing house for such a data meld. On the excipient side this would require access to supplier data, which may raise problems of confidentiality, but the database can be configured in such a way to allow the blinded incorporation of such data, if eventually made available.

For continuous-production high-volume excipients (e.g. MCC) a finer level of scrutiny is required beyond Certificate of Analysis (CoA) data, which tends to average out underlying variability. The traceability of a particular excipient needs to link user reference numbers, through supplier “batch” numbers, to intra-“batch” references. As MCC “batches” may represent hundreds to thousands of tones, over weeks to months of continuous manufacture, access to the most relevant in-process data will give a more accurate assessment of the impact of the excipient variability on finished product performance. One way of doing this is sequential numbering of containers within a given “batch”.

There is sufficient flexibility in the structure of the database to link to pre-existing databases or simply import data via spreadsheets.

The database initiative is currently funded by FDA and in the future it is likely that such monies will switch more to scientific research on excipients, rather than maintenance of the database itself. NIPTE cannot solicit commercial funds directly so one idea would be for NIPTE to partner with IPEC on future database development.

Prabir Basu of NIPTE proposed further discussion at the May 24 IPEC QbD meeting in Washington, a subsequent meeting with FDA, with representation from IPEC’s QbD committee, and a two-day workshop on excipients for 2012 in collaboration with FDA/AAPS.

If enough people are interested, NIPTE would be willing to host a demonstration interactive webinar before the 24th May IPEC QbD Committee meeting. PharmaManufacturing.com will continue to cover this effort as it evolves.

About the Author

Agnes Shanley | Editor in Chief