FDA's Janet Woodcock on PAT and The Critical Path

Click the Download Now button below to hear an audio file of Part 1 of Dr. Woodcock's speech. Click here to access Part 2 of her speech. A transcript of Dr. Woodcock's presentation follows.

Good morning. Ajaz asked me to talk about FDA’s Critical Path Initiative, and how that fits in with the goals that all of you are here to discuss, which really focus on optimizing pharmaceutical manufacturing, particularly through the application of process analytical technologies (PAT).

I’d like to explain how the Critical Path Initiative and other FDA initiatives such as Pharmaceutical Quality for the 21st Century relate to PAT, and talk about implications that Critical Path and quality initiatives have for PAT.

I’d also like to discuss how we can, in the future, link critical product attributes to clinical outcomes, to gain a better understanding of what we’re trying to control, in other words, to determine the attributes that determine a product’s good clinical performance. At this point, this is something that we know very little about, except in the very broadest sense.

So the basis thesis behind FDA’s Critical Path Initiative, which really resonates across manufacturing, is that investment and progress in basic medical science has far surpassed investment and progress in the medical product development process.

And the analog in the manufacturing world is the fact that R&D investment by companies has far surpassed actual interest and investment in how to make product and actually get it out the door.

The development process is becoming a serious bottleneck. Our theory at FDA, and we’ve said this publicly, across the board, is that we’re using the evaluation tools and infrastructure of the last century — and in some cases, tools from very early in the last century — to develop this century’s medical advances.

FDA would like to change this.

But all of this must be viewed within the context of this huge R&D investment, which has driven heightened public expectations that we’re going to have all these novel therapies available based on all new science.

There’s been an explosion in new science. There are ventures and labs all over the world that have all this great new stuff that’s going to cure more patients and meet medical needs.

Novel drug submissions, however, have been flat…approvals last year were very low relative to performance in the '80s and '90s, and much less than what came out in the mid-'90s.

Although investment in research and discovery has gone up tremendously, we’re not seeing this reflected in output of new products, even though the public is expecting this result. As a result, they’re getting mad at NIH and the pharmaceutical industry.

There’s been a definite plateau in the pharma development pipeline, and a lot of discussion and speculation in the literature as to its causes. "Maybe we invested too much in genomics and it wasn’t ready yet," we hear, and so forth and so on. You’ve heard all this speculation.

But what we need to do is to work on things we can address and correct. So the Critical Path focuses more on the development process

At the same time we’re that drug development today has a lower chance of success. That can’t be good news. So despite all the answers you have from science, as you move new products into preclinical and clinical development, their success rate is actually declining. Today, new compounds going into Phase I have an 8% chance of success reaching the market, compared with 14% a decade and a half ago.

And, more ominously, the Phase III failure rate is now about 50% vs. 20% a decade ago. I say "ominous" because, by the time you get to Phase III you’ve sunk a lot of cost into the compound, and invested a lot of money. It’s very painful. It also causes your stock value to go down, because you’ve told everyone that you have this great new compound going into clinical trials.

And of course you know, you read about it in the papers all the time, "this program had to be abandoned, that program had to be stopped, etc." and all the disappointment that results. So the predictability of the process is not improving right now.

So we looked at all this at FDA a number of years ago. We knew that these are all big problems, but we saw that the societal investment on the R&D side has been huge. Investment was now needed in the development process; at FDA, we need to improve our standards and to use a lot of science to help us improve standards for manufacturing.

There’s a huge private and public investment in basic research, and huge investment in the development of specific products, but we don’t see the same kind of investment in “how” to develop an evaluation process. Academia has not been funded, and this is true in clinical trials, believe it or not, as well as manufacturing.

There’s been a little more funding for research in toxicology because of environmental issues. FDA has done this type of work over the years, but Congress has not funded FDA to do any of this work.

In the private sector where many of you work, these efforts are often viewed as proprietary and confidential, and so knowledge is not generalized. As a group, we cannot advance our development science as we should, all the way from manufacturing to clinical.

We need to explain this to the public. They’ve never heard about this before. They think that when one of your scientists has a brilliant idea, in two years they should be able to hand a new drug using that idea over to the patient. There should be no problem, the public reasons, and if there is, then you’ve failed. And your scientists think that too, by the way.

It’s never been explained that basic research is a long way away from having a product. We’ve adopted the term Clinical Path to explain the fact that this is a long and extremely technically challenging process.

And there’s research that we need to do to make this better. Some people are talking about translational research and Phase I research along with Critical Path research.

We tell Congress, the public and medical scientists that the science needed to enable efficient manufacturing is different from basic science, but it’s just as worthwhile, just as necessary to get new products out. There are three types of science:

• Safety –
How to predict whether product will be safe enough.


• How do you predict how it will look?
Obviously, we’re really bad at that. That’s why we have such a high Phase III failure rate.


• Industrialization –
How to turn an early scientist’s idea into something that can actually be manufactured, consistently, at commercial scale. That’s something that the public can try to understand. If you put it this way, people get it.

We say that each one of these areas faces many challenges. We’re trying to build a case for everyone here, that we have to do this, we have to invest. And some of the people we’re trying to convert include the CEOs of big companies.

Patient groups are very interested, but just can’t figure out why their diseases — whether MS, ALS or cancer — aren’t getting cured. We’ve invested billions every year in basic research, and people can’t figure out why we don’t have products in hand to help their patients.

We explain how complicated the Critical Path is, how many disciplines are involved, and how all of these need to be improved. This Critical Path initiative we’re doing has a number of principles.

We’ve had no funding for this at all, specifically from Congress, so we’ve focused on establishing collaboration between government, academia, industry and patient groups, and we focus on developing new infrastructure and tools, not specific products.

Everyone wants to focus on specific product development, which is viewed as glamorous. We need, instead, to focus on the tools that enable product development. We’re trying to build support for an academic science base.

In manufacturing, we talked to NAS and academia. This is a major industry important to public health and important to biomed science and needs to be supported.

We need to build opportunities to share knowledge and data instead of keeping it. If one company implements PAT and nobody hears about it, that’s not good.

We need to develop enabling standards … standards for clinicals and for genomics. This is a very broad field.

We published an initial report and had open comment period. We’ve also have multiple partnerships. We’re pleased that industry has stepped up to the plate.

We’ve published a list of projects that need to be done, including major projects in manufacturing.

Recently published a list of collaborators and we’re continuing to work…we have a major article that will be published in Clinical Pharmacology Pharmacopeia on the Critical Path Initiative, and we continue to crest along.

The Critical Path list published last spring names manufacturing as one of its major opportunities. You may not hear that much about this.You may read about Critical Path in the papers, or hear about it on TV, but most of that will focus on clinical.

These articles or sound bites won’t talk about manufacturing, because most people aren’t familiar with it, but that doesn’t mean that we’re not focusing on it. We are, to the extent that we can, since we don’t receive any direct funding.

The goal is to incorporate modern quality principles into how we regulate manufacturing, and into how manufacturers think about what they’re doing. Currently some manufacturers seem to think their job is satisfying FDA.

Instead, they ought to be satisfying their customers by producing high quality product, not responding reactively to regulators. That was the goal of the initiative.

This type of thinking became the prototype for the Critical Path initiative.

We could see across the FDA the need to modernize and to incorporate modern science. We have to incorporate that into regulation. The “modernizing manufacturing” theme was then incorporated into the bigger Critical Path umbrella when we came out with that.

So when you hear about CP, “modernizing manufacturing” is part of that. But it’s part of the broader 21st century Quality Initiative.

We’re moving along with the Quality 21st Century Initiative. So what does this have to do with PAT?

Well, if the Critical Path is about accelerating the pace of introduction of new science and technology into regulations and into the regulated industries, we find that these industries, in all phases, are often reluctant to incorporate new scientific advances rapidly. Fear of regulatory consequences impedes adoption of new sciences. But if we’re to get over these problems, we need to stop that.

Because these problems are like a train wreck. We have the productivity problem, going this way, and as Ajaz said, we have the healthcare crisis on the same track moving toward it from the opposite direction. Society is bent out of shape over the cost of pharmaceuticals. Something is going to have to give in this equation.

What we need to do is to start modernizing everything. That’s the way out. We won’t be able to improve that rapidly, and our healthcare system has its own problems, which Congress and all the upcoming presidential candidates will be dealing with.

But PAT is a dramatic new way of thinking of pharma manufacturing. It’s one of our standard bearers for the Quality initiative. Its something that other industries have had t for a long time. This industry hasn’t. PAT has also been successful in other settings, so its proven. It’s not emerging technology.

But what we especially like about it is that it’s also emblematic of the need to move from empirically derived trial and error methods to rigorous, mechanistically based and statistically controlled processes. And believe me, this is not exclusive to manufacturing. The development process is empirical in its entirety.

Animal testing is entirely empirical. Once a protocol has been validated, we expose animals to increasing doses of the compound then do body count to determine whether it might be safe enough to give to people.

Then we look at how it works in people, and see how they react to increasing doses. We look and see what happens, and then if it doesn’t work we do another trial and if that doesn’t work, we do another. And so on. We do hypothesis testing, etc. We do regression to determine whether it works well or not.

The result of this, as you all know, is a really expensive drug development process. And people are not at all satisfied because, at the end of the day, for example, with Vioxx, we discover things that we couldn’t have discovered earlier in the process.

So this focus and fixation on empiricism throughout the entire drug development process is the reason why it’s so long, cumbersome and expensive.

And we’re not crazy. It’s just that biology is fantastically complicated. Until recently, we haven’t been able to apply the mechanistic approach. But that’s changing, and manufacturing and PAT really lead the way.

We have all the tools at hand, it’s just a matter of implementing them.

And so we need to do a lot of things in the quality area. Manufacturers need to change their approaches. They need to invest in new technology, which, of course, is expensive. And this is not a great time to ask for big investment.

Then, you have to break down silos between R&D and production so that there can be a more seamless transition between the two.

You have to be less conservative, which is tough at a time when there are pressures to be more efficient and you’re having problems with productivity.

The human response to all this is to become more conservative and “cling to what you know,” but that’s not the correct way forward.

At the same time, internationally, regulators will need to accept and encourage new ways of doing business, and harmonize regulatory approaches.

Those of you who are familiar with ICH know that FDA has been diligent, and Mohab and his colleagues (and Ajaz, when he was with the FDA) have been very diligent, in working with global regulatory community in trying to see that changes are made. We’re doing extremely well, I think, given the pace of international negotiations.

It’s may be moving slower than global warming, yet we’re getting there. But we can’t just do it in the U.S. It needs to be an internationally accepted change.

For PAT, manufacturers will need to invest in technology for product, equipment vendors will need to focus on pharma market, and global regulators will need to be more accepting and encouraging for PAT, and to modify regulatory processes to accommodate PAT.

There’s a need to do training and academic research. All these things are coming together.

The biggest barriers to PAT today are risk aversion and reluctance to invest in manufacturing. Currently companies are even more conservative due to issues with the bottom line. It’s hard to change.

There’s a need for standards specific to pharma. But these are all addressable and we’re working on them.

That’s how PAT and manufacturing and quality fit into the overall Critical Path.

Now I’d like to talk about linking critical attributes to clinical performance. One problem is uncertainty. We live in a world that’s full of uncertainty as to what causes what.

That’s why regulatory standards are often highly conservative.

I’ve been involved in legislative discussions, e.g. changing statutes on oversight and manufacturing, and people wrote in comments that are beyond what would be imaginable. And some proposed changes were almost unimaginable.

For instance, once a company made what they considered to be a "minor change" and failed to inactivate a live polio vaccine. That was a major important function.

Then we often see the results of procedural failures, process failures, not assay failures. For example, people out in ethylene glycol or another toxic compound… this usually happens outside the U.S., but then people get extremely ill.

These are the reasons for tight control over all aspects of manufacturing.

If you ask different review staff, "What would be the clinical effect of minor variations, say, in content uniformity?" a clinical reviewer would say, "I don’t know, but I’m really worried about it." A reviewer would say, "It’s very significant." An investigator would say, "It's a terrible clinical problem." I would say, "Clinically undetectable in most cases," and I’d say it with absolute confidence. Here’s why.

When you look at exposure in solid dosage forms, how many people are getting exposed and at what levels… based on liver and renal status, size, these variations swamp minor variations. A few people’s metabolisms act differently.

That’s the kind of variability we’re talking about at clinical, and most people don’t take their drugs most of the time. Some people take medicine more often, thinking more is better.

The effect of minor variations of content is negligible from a clinical standpoint. An exception is stability over expiration period, and what is very important, what is critical, is how large the fluctuation and how much variability.

Pharma is multivariable; there’s no simple rule that can be applied. Most content variations are clinically negligible.

Thinking like this is painful, but provides us with a major opportunity for better performance. And the actual quality of the drug must be ensured. These are the things that are important.

So what is the future of Critical Path and Quality initiative? Some folks were concerned when Ajaz left.

But FDA’s Commissioner is very committed to the whole Critical Path Initiative.

Manufacturing and everything are viewed as very significant.

Manufacturing is mature; companies have been working together for quite a number of years. It's well established; it may not be so flashy right now.

We’re engaged in changing the way the review process works and other functions, but we’re moving forward in all these areas, and you can expect progress.