cGMPs / QRM Process

Ciurczak Interviews Nasr on QbD and the Need for Robust Processes

In this interview with Emil Ciurczak during IFPAC 2007, Dr. Nasr discusses Agency activities and challenges, and what we might see in the future.

Click here to download an audio file of Part 1 of this interview. To hear Part 2 of this interview, click the Download Now button at the end of the transcript that follows.



Emil Ciurczak: We’re talking with Dr. Moheb Nasr, director of CDER’s office of new drug assessment. You may have heard some of his remarks at the plenary session, but there are a few things that we’d also want to talk about.

We were just mentioning before we went "on the air," that Dr. Nasr was "ahead of the curve." At San Diego’s IFPAC meeting in 1999, he had already spoken at the session that I put together on process analysis, long before the FDA subcommittee was formed.

When a director at FDA calls someone at a drug company, the only questions you ask are: "How high, and when do you want me to come down?" And so Ajaz [Hussain] hooked me for that, but long before then, you were already preaching the virtues of process and trying to figure out what was going on. To give you your street creds, you were leading the fight before anyone even heard of Ajaz.

You’ve been championing QbD. Do you think that these recent industry layoffs and cutbacks (such as those recently announced at Pfizer) may hurt this effort? Also, PAT and QbD call for patience money and time. But often when you have a squeeze for profits and time, none of those three are given you by management. What’s your take?

Moheb Nasr: I think it may or may not hurt the effort. It will all depend on where the cuts are made. If the cuts are in development, it may hurt efforts at Pfizer and other companies that do submission. If the cuts are in manufacturing, it may have an impact on process improvement and post-marketing changes, innovations and changes of that sort. It depends on where the cuts are. Any time there are cuts in industry it hurts the economy. If cuts are in marketing it will not hurt.

EC: This question is a close parallel: A number of people say, "Why do we hear about PAT in meetings?" My experience with NIR back in the 1980s we had, for example, at a conferences, e.g. in New Orleans, we had so many people in some sessions that the fire marshals asked us to open the door and make more room, because people were standing five and six deep along the back we were overflowing with people.

Then suddenly in the late '90s it petered away. Not that people lost interest, but lawyers for the companies suddenly decided, "Hey, this is a useful technique."

Now I know, from talking to people, that J&J and Hoffman-Laroche are doing PAT but you never see them at meetings. It is possible that this is a good sign that we’re not hearing, that people are so busy doing PAT that they don’t have time for "PowerPoint PAT?"

MN: I think that’s an excellent point. I do IR with many industry sponsors in a formal way through the drug review process. In meetings like IFPAC and a lot of public workshops. I also get invited to speak at corporate meetings at companies like Pfizer, Merck, Wyeth ... I know myself there are other folks who are not present at these meetings who are doing as much and doing very good work on PAT.

Last October, I visited a company that had not been a vocal PAT proponent. They shared with me what they are doing in different aspects of development and mfg, and how PAT was being used and impacting their whole quality operations. I was most impressed. I had no idea, before I went to visit them in New Jersey, because they are not as vocal.

Every company has its own strategy. When I went to that company I met with the CEO and I asked him point blank how come they had not been supportive of discussion. He said “On the contrary. We have different business strategies. We try to see where [new approaches for us such as PAT] fit within existing business needs rather than creating a new additional program, where the need might be questionable.”

Some companies with an abundance of resources create a new program, then they think later on where that program can help. With other companies where resources aren’t abundant, where they implement more business practices, they find issues first and then find the solutions.

EC: There are also a number of people who are still looking for the second shooter in the Kennedy Assassination, people who believe in the conspiracy theory. They don’t want to give aid and comfort to either their competitors or to generics manufacturers. They’re just waiting until major products go offline to suddenly drop prices because they have a PAT process, to try to blindside a generics company. There’s probably some truth in it. Obviously any price reduction will help the general population.

At ISPE you talked about adopting technology for continuous processing – there are validaton concerns, but also questions of process recalls. When you’re running a process for 5-6 weeks you make 80 million tablets and have a recall, how do you define batch?

MN: Your question has several parts. First, do we need new technologies to fully implement QbD? Yes. With existing infrastructure and platforms, QbD fits. It can be done, but it works much better with new platforms.

To be fully realized, new manufacturing platforms need to be there, when you think about a higher level of development, the design space, you cannot model what you don’t understand. You can do scale up experiments with equipment and it will help you, but it doesn’t really get you to a mechanistic understanding. In summary, I think new technologies are needed.

Now, with new technologies, there will be some associated regulatory challenges. Because of that, we at the Agency are currently revisiting our process validation guidance where the traditional "three batches" approach may not be appropriate and it will not be appropriate for these types of new technologies.

We are looking into different ways to verify the quality of the batch, and using approaches such as continuous process verification and the level of validation/verification that is proportional to process understanding.

What do I mean by that? If you have fully developed your design space, and you have full knowledge and good predictability of the batch, the validation requirement that will be expected of you will be considerably less than if you’re using empirical techniques to assess quality of the batch and the only time you scaled up was to do validation. In other words, the level of validaton and verification is proportional to the level of understanding of the development effort at the sponsor.

Now, what is a batch? The definition of the batch will change dramatically. I don’t have a definition for you today. But the whole concept of batch is changing. Traceability will not be that different. With continuous monitoring, traceability will be much easier than it is today.

EC: Good answer, you should be in politics. Now about your program for training investigators. I don’t call them inspectors anymore, because that conjures up negative images, as if they’re looking for criminal action.

I think the other good thing as a side note, is that FDA has gone from bad guy to partner. That’s what I try to preach in my consulting and my courses. I came into the industry in the 1970s at Ciba, and, at the time [for any drug company] the atmosphere was “see what you can hide.” Or “Mom’s coming home. Hide the beer.”

Or, “Just don’t talk to them and don’t help them.” And it’s turned around to the point where what FDA is doing is almost analogous to what the USDA does to help the farmers improve. They’re helping us keep our business. In my courses, I keep pointing out that if it weren’t for the industry, they wouldn’t have jobs, so, in some sense, we’re all in this together. How is the program coming, for getting all investigators trained in PAT?

MN: We have a training program. That resulted in the development of the FDA Guidance in September 2004. Since that time, we’ve had additional training. What’s critical is creation of the pharma inspectorate, the PI program, where we create investigators who are trained in all facets [of science-based manufacturing and development], not only PAT but in development in science and QBD. This is going very well, and we’ve made good progress.

Part of the challenge that we have with this program is that some of the people we train leave the Agency and others are near retirement so we have to continue this effort. This is going well. However, I think that some of the resource issues that Agency currently faces may have a negative impact on the program. I don’t know how negative that impact will be, but it won’t speed these efforts. As far as our reviewers are concerned , they’re being trained in many facets of PAT.

Within my group I created a manufacturing science branch, and courted people who were leading PAT in industry, brought people with academic perspective to this group. It encompasses manufacturing challenges in PAT… I have members of manufacturing science branch participate. Having these two groups and good training for investigators and reviewers by themselves is not enough. What’s needed is to bring them together.

For the CMC pilot program, we have review inspection teams where reviewers and investigators together attend meetings with sponsors.

EC: I recall back around 1971 or thereabouts, one my old classmates, a chemistry major, came in with an inspector, and when I asked what he was doing, my colleague replied, “I have to tell him what he is looking at.”

Everybody forgets that the F in FDA stands for food and your agency is also responsible for food. Because FDA inspectors at that time could spend all morning evaluating food safety at the Fulton Fish Market, making sure that the sanitarty conditions were okay, and then in the same afternoon, he’s expected to be an expert in drugs, and to inspect a pharmaceutical facility. It’s not an enviable job and a little sympathy is good.

In the Reagan years, they cut back funding and jobs. It’s one of the few things that as a citizen, I don’t mind my tax dollars going to. Also because I don’t consider you guys “the enemy” anymore.

I don’t think most people recognize the quality and number of scientists who are working at FDA who could be at any institution. People only see the regulatory face, but they forget that you’re doing a lot of your own science and projects.

I was always impressed to meet the people there, and now you have a new facility where you’re not spread out.

Now my next question. Is there a chance that FDA could put a little gentle pressure on the industry to make more public the fact that they’re doing PAT. At lot of it is still hand-waving and "trust us we’re doing PAT." But I think if more people knew, perception is important, if more companies were to give case histories, "I’ve got ABC drug, and we’re using PAT and saving X millions per year," I think it would spur other people to do more with PAT.

MN: It’s up to the individual company, to make its own business case and share. I don’t think extra publicity is needed at this time. Some people think so. I don’t.

What we do need to share are success stories. People on the sidelines need to know that if you use these approaches for manufacturing, how beneficial it will be to them. Sharing these success stories at workshops like IFPAC is very beneficial.

At the Agency, through the Advisory committee of Pharmaceutical Sciences and Manufacturing Sciences subcommittee, we bring sponsors to give input to the public to discuss success. We ask them, specifically: What can I do to facilitate the implementation of good science.

EC: Now, here’s a question that I often hear from students regarding PAT. What is in it for FDA? As I understand it, the Agency is legally required to inspect pharmaceutical facilities every so often. Now, if a company has a fully operating PAT program, wold you ever consider, if a company gave you permission, using your computer to observe a PAT process for half day. Could that ever count as a site visit?

MN: I don’t have oversight over this part of the Agency, but the value of using QbD and PAT to the Agency is tremendous. Using these approaches provides a much higher assurance of product quality, and we’re responsible for that. It will enhance the efficiency of the inspection process, and hasten the development of new drugs to meet unmet medical needs.

It will also allow us, as an Agency, to focus our resources where we have higher assurance, provide public as high a level of assurance of products. In short, it provides a lot of benefits to Agency, in terms of resources, innovation, and bringing new medication to the public.

Could we have inspection remotely? I see it as quite possible at some point in the future, and that may fit in very well with our risk-based inspection program.

What to inspect and where to inspect are closely related to three factors:

  1. The facility itself and its history of compliance;


  2. What are the products’ associated risks, i.e., the risks that producing these products pose to the public?


  3. The complexity of the manufacturing process and the degree of understanding of that process. This may result in new approaches.

EC: This type of approach could make it easier to comply with frequency of inspection.

MN: Yes, but you must understand that it could never be used for every product and process.

EC: This question is as personal as it is professional. I do a fair amount of work in Europe. We hear that, at best, the EMEA is neutral regarding PAT. I’ve heard, off the record from friends that their attitude is “why would anyone be bothered to do this? It it too expensive.”

Their position is “show me,” but I don’t see them doing the pioneering and leadership that FDA is doing. Is there a chance that they can be won over? Most companies are multinational, and it’s very difficult to convince management to go into QbD just for one division, it would save money to have entire corporation embrace these programs, ie in Canada, Europe, etc and to show them that it is going to save corporate a large chunk of money...

MN: There are steps we’re taking to facilitate this. That’s where we’re putting our energy: through ICH Q8-10 we have shared harmonized understanding of the direction where we’re going. I was very pleased at the last ISPE-PDA meeting in December in Washington; EMEA presented on QbD, we made progress.

ICH must be the cornerstone, and that’s where I see the greatest opportunity for realizing harmonized approach.

The second area is bilateral dialogue among regulators. I have discussion with colleages in Europe and Japan, as we move into QbD. Debate is healthy, and vigorous.

The third is through public discussion and debate. And part of ISPE is its international leadership forum, where we bring leaders from regulator agencies around the world to answer questions. People working within the industry challenge us and inquires about our initiatives to address industry challenges. In the US and Europe we have participated in these venues, we have public workshiops and invite colleages We have an upcoming one later this month, in February, and I have invited European colleagues, Emil Cook, head of inspectional program at EMEA, Susanne Heitl, quality lead, will be here. So, I invite my colleagues in Europe and Japan to come and work with us on this.

EC: One more topic: Dissolution. It’s a complex issue, and some people in the Agency are suggesting using mechanical calibrations rather than testers. Simultaneously a number of people are using chemometric predictive methods for dissolutions. But I see dissolution, at best, as a tenuous predictor of in vitro and in vivo correlation. And now we want to predict a predictor using chemometrics. So it's like a third level... Is there any uneasiness about using a 13-factor calibration model of a PLS to predict what the granulation might do as a tablet in the human body?

MN: I gave a presentation last May on how I see the future of dissolution on Quality by Design. In summary, I think we try to oversimplify and generalize. I see a great value of dissolution and surrogate in vitro release methodologies during drug development, to establish the clinical linkage between clinical trial materials and the products to be produced and commercialized.

There is a great value in looking at drug release characteristics. Under QbD, since we have better understanding of product and process, if we can identify the critical attributes and factors that impact dissolution, e.g. particle size, it may wind up being better to control particle sizing, which can be done on-line with NIR. No one answer will fit all. Depends on product and process.

Dissolution is very valuable, as well as surrogate testing during drug development, but once we establish a corerelation, then it is best to look at factors by themselves, rather than surrogates.

EC: The Agency brought sanity to dissolution back in the '70s. I spent a lot of time simulating what happened in the stomach, adding pepsin, chymotripsin to dissolution. At some point, at the behest of Agency, we said, "Wait. Let’s try just water and a paddle at 50 rpm — same water, same paddle, same speed." The method certainly became a good quality tool to show if things were setting up right regarding stability. We weren’t trying to predict blood levels, but just wanted to that make sure drug product doesn’t change over tme. So now, we’re bringing common sense to an interesting and strange technique that’s peculiar to the pharma industry.

MN: In my way of thinking, dissolution by itself, as a surrogate to drug release, needs to be evaluated properly. The drug release is a complex attribute that results from the combination of different things: formulation, manufacturing and physico-chemical characteristics of actives, and analytical methodologies.

So you have the analytical piece, you have the biopharm piece, the formulation and manufacturing. All of these pieces are needed in order to set appropriate specs and methodologies. We started to look into this in depth at my office a few months ago, and I have just hired an associate director of biopharmaceuticals, to ensure that dissolution spec for QC is being done right.

We’re ... looking at the complex factors, not only the biopharm but formulation, but the manufacturing, physico-chemical characteristics, and analytical metholodologies I now chair an internal working group that brings all this expertise together.

I’m also aware of a program going on right now aided by Pfizer looking at issue in a similar manner.

EC: So it’s not your father’s FDA anymore. The early days, when, if you added HPLC to a new drug application, it was automatically rejected, are clearly over. Things have come a long way.

In the past, it was easy for the industry to call the FDA the bogeyman, but now the Agency is seen as helping to improve the safety and quality of the final product. Thank you.

Free Subscriptions

Pharma Manufacturing Digital Edition

Access the entire print issue on-line and be notified each month via e-mail when your new issue is ready for you. Subscribe Today.

pharmamanufacturing.com E-Newsletters

A mix of feature articles and current new stories that are critical to staying up-to-date on the industry, delivered to your inbox. Choose from an assortment of different topics and frequencies. Subscribe Today.