Normally, I do not attend all sessions on every day of a meeting (of any kind). I would like to state that this one was a rare exception. I have been associated with PAT since I was a member of the PAT sub-committee for the FDA. Most pre- and post-guidance meetings have been concerned with the actual measurements, hardware, and software. As such, the majority of the audience has been scientifically trained (chemists and engineers). Unfortunately, scientists and engineers dont write the checks in an organization.
This PAT meeting was a break with tradition: a syllabus mostly concerned with the economic and personnel requirements of PAT. The format was a little different from many meetings I have attended over the years, so I will be using a somewhat different outline from my usual reports. I cannot describe in detail all the sessions and talks, so I will relate some highlights of the program (and hope not to offend speakers not mentioned, but, there were soooo many good talks )
The first day consisted of three workshops:
- Quantitative Process Characterization: a First and Important Step of a Successful PAT Program,
- Developing Manufacturing Excellence Through the Implementation of PAT, and
- Perceived Barriers and Facilitators to PAT Adoption.
The group used these tools to analyze manufacturing problems and characterize key drivers of variation in a control release drug product. The brainstorming was a classic demonstration of using various disciplines (a staple of PAT) to attack a multilayered problem. The groups presented their solutions after a suitable conferencing period. The dual purposes were to show the value of differing views and the sheer number of approaches that could be considered.
The second session was run by Gary Ritchie (USP), who was a last-minute substitute (but you would never know it from the finished product). Gary used the FDA guidance on PAT as the framework of the interactive session. Adding the work being performed by the ASTM E55 committee on process control, the lively session was a model of participation. Gary stressed communication; communication within an organization and constant communication with the FDA. Basically, the FDA does not work well with surprises. It is far better to work on PAT with them from day one.
The third workshop, chaired by Nancy Mathis (Mathis Instruments) had the students/participants brainstorming factors which (they believed) will aid or impede implementation of PAT in a company. It was interesting that some participants placed some agents in both columns. The results of this session were the kickoff for the second days talks. The most interesting point was how participants viewed Quality Assurance and/or the FDA as potential aids or barriers to PAT.
I had the pleasure of chairing this days session, so was able to interact and ask questions freely. Indeed, seldom have so many people wished to ask questions or comment on the content of talks. Fortunately, breaks were one-half hour and the crowd was small enough to mingle at the group lunches. Few questions were left unanswered.
As I mentioned under Day 1, Nancy Mathis began Day 2 by moderating a panel consisting of PAT experts from GSK, Wyeth, (late of) Aventis, Invensys and Patheon. There were similarities between their experiences and different outlooks, as well. [Since there were no recording devices to encourage candid remarks, I will not attribute quotes to any one member of the panel.]
One point of consensus was that all levels of the company had to be on board for any PAT project to succeed. It was agreed that QA had to be included in the preliminary discussions or they would be the biggest defender of the status quo. Since the mission of QA is to keep the company from violating protocol, it is understandable that a QA kept from the project will, on principle, oppose changes in current practices.
It was interesting to see that the member from Puerto Rico believed that (for critical products) the goal is 100% analysis of dosage forms. The others, while not openly disagreeing, were not pushing the point. All did agree that there were antibodies within any organization. That is, ANY new idea will have some person who will attempt to block it. Careful planning, proper personnel (include ALL departments), and thoughtful arguments against the nay-sayers are needed to go forward. All agreed that a knowledgeable champion within the company, of at least vice-president level, is the key to pursuing the dream of PAT. The least successful approach is to have PAT arise from any single department, then try to sell the concept to all the other interested parties.
Other speakers addressed concepts not normally stated openly. Jean-Marie Geoffroy (GPO Manufacturing Sciences), in addition to encouraging us to build the business case, made another important point: PAT can reduce the oversight opportunities of the FDA. In other words, showing control allows the Agency to visit other sites where the old ways still hold sway. That was a different concept: more control by the company means less control by the Agency.
Quality up, costs down = ROI
The economic points were well articulated by Philippe Cini (Tunnell). The areas of financial value can be created and quantified are: Quality, Inventory, Capital Investment, Time to Market, and Regulatory Risk. Some of these have been discussed previously and are somewhat intuitive. For instance, the quality of the product will improve as we understand and control the manufacturing process. And, of course, the inventory will diminish as the OOS (out of spec) batches end. There will also be fewer warehouses cluttered with raw materials and intermediates as the process becomes more controlled and there are fewer failures.
Capital investment will slow down as we move from 10% efficiency to even 50% (still far, far below potato chips); equipment will approach full use and fewer pieces of manufacturing hardware will be needed as well as the buildings to house them. Time to market will decrease simply because new products will be understood from the beginning. There will, by the process of elimination alone, be many similarities between current products and new products. Understanding current products will lead to understanding new products. This, in turn, will speed up submissions and, more importantly, acceptances by the FDA.
Regulatory risk isnt always obvious, but it, too, is there. The more control we have over a process, the less chance for a recall. As we control the process, the FDA slows down its oversight; the agency shifts from law-enforcement to partner in progress.
Some of the other interesting points made by speakers include the well thought-out talk by Krishan Kumar (Merial, Ltd.) on technology transfer. Being able to run cutting edge technology in the pilot plant isnt enough; these tools must be able to transfer to production as well.
Kevin Bynum (Novartis) showed the flow of data/information in a complex PAT setting. He allowed that a mountain of data that cannot easily be mined is as good as an old storeroom full of papers. Chemical and operational data can and must be correlated for PAT to have a chance of working.
I really liked the fact that generic companies were well represented, not only in the program, but in the general audience. I have suspected that, after the biggest innovators (e.g., Pfizer, Aventis, Merck) have adopted PAT, the generics will be in the forefront. By dint of the fact that the first generic gets the largest piece of the pie, speed to market is at least as critical. Recalls and delays of shipments could cost a generic supplier the leadership position for a particular product. That position is seldom regained once lost.
In all, anyone interested in learning more about these wonderful speakers can click the Download Now button below to view the event brochure (PDF format), get the names of the speakers, then contact them and get their talks. The effort would be well worth it. Trust me.
(Editor's Note: For more information on IQPC's PAT Business Strategies Summit and maybe some assistance in obtaining copies of presentations contact Kerri Hughes, Conference Production Team Leader, at IQPC, 535 5th Ave., 8th Floor, New York, N.Y. 10017. Phone: 212-885-2760; fax: 212-885-2762; e-mail: email@example.com. She's looking forward to talking with you.)