Transcript of the FDA Prescription Drug User Fee Act Information Technology Strategic Planning Public Meeting

Rockville, Maryland
October 19, 2007

Public Meeting
9:00 a.m.
FDA Office of the CIO
5630 Fisher's Lane, Room 16B44
Rockville, Maryland 20857

A P P E A R A N C E S

Panel Members:
Margo Burnette, FDA Moderator
Malcolm Bertoni, M.S., Director
Planning Staff
FDA Office of Planning
Timothy Stitely, Chief Information Officer, FDA
Mark Gray, PDUFA IT Program Director
FDA Office of the
Chief Information Officer

P R O C E E D I N G S

MS. BURNETTE: Let's go ahead and get started. It's a few minutes after 9. Today is either a really lucky day or a really challenging day. One or the other. We found out on Tuesday that this room had been flooded and we weren't going to be able to be here today. The good news is we are here. The bad news is things like microphones and so forth aren't here. So we're going to proceed and we'll try to take care of that as we're going through.

So thank you all for coming. The other good thing is I brought my umbrella, because it's supposed to rain for the first time after 34 days of no rain. So it's a good day.

I'm Margo Burnette, Executive Director of the Bioinformatics Board, and I'll be the moderator for today's meeting. I just want to thank you all for coming today. I want to remind you the purpose of the public meeting is to solicit views and information on issues concerning how the agency can best plan and apply IT resources to support the process of review of human drugs, and also we're really looking for views and information on -- to identify and prioritize IT solutions that will support this process.

The scope of this is not an RFP meeting or soliciting bids or anything of that sort. We really want to hear from the stakeholders what's the content of the IT plan that's going to be most useful to you. You may be aware that we are going to be developing a strategic IT plan the first of the year and we want input on what you think would be most useful to be included in that plan.

Of course, also data standards and guidance are big issues here at FDA, so what are the data standards and guidance that best support the available IT capabilities and any implementation concerns; and lastly, how the agency architecture and IT solutions can best be applied to support the public health needs. So that's the scope and the purpose of our hearing for today.

There are a few ground rules. I think everybody has found -- this is housekeeping. I think everybody has found the rest rooms outside, right across the hall. There's a vending room next to that. And then there's a cafe across the street if someone wants to go out for lunch and stuff.

The way this will work today is we do have an FDA panel here, made up of Tim Stitely, who is the Chief Information Officer; Malcolm Bertoni, who is the Acting Assistant Commissioner for the Office of Planning. I never get his title right. And Mark Gray, who is the PDUFA IT director.

Each of them will give a little presentation in a few moments and then we do have the regularly scheduled speakers. The speakers are limited to 30 minutes and I will give you a two-minute warning if you get to your 30 minutes. If you don't take the full 30 minutes, nobody will ever complain.

One of the other rules surrounding these public meetings is that questions are only allowed from the panel. So this is not a dialogue kind of meeting. So the panel can ask questions of the presenters. We will be accepting comments 15 days after, and the Website is on the agenda sheet; is that correct?

MR. GRAY: Yeah. Where you can access the docket.

MS. BURNETTE: Where you can access the docket and submit your comments there. And the transcript and all presentations will be made available after the meeting on the docket. Okay?

Those are our overall. I'm going to introduce, Malcolm Bertoni is going to go first and he's going to talk about -- I'm going to introduce all three of you at once. Malcolm is going to talk about the overall strategic and performance planning context for the PDUFA IT program. Malcolm.

MR. BERTONI: Good morning, everyone. I'm just getting over a cold, so I hope I'll get through this with my loud voice. But I'm really delighted to be here and I really want to thank all of you for taking time out of your busy schedules to come here and provide us with your ideas about how we can improve information management at FDA. I'm really here primarily to listen. That's why we're here to ask questions. I'm really looking forward to what you have to say.

However, before we start hearing from you, we wanted to offer some overview and kind of set the stage, set the context for the PDUFA IT program.

So I'm going to start off with some FDA-wide perspectives and try to link it a little bit to the program and, of course, it will be followed by Tim Stitely, who will talk about the specific approach that we're taking from an information technology standpoint, and Mark will get into the details of the actual IT commitments in the newly authorized Prescription Drug User Fee Act, PDUFA. I'm sure you've heard that acronym before.

So what we wanted to do was start off with really with the 10,000-foot view to look at what does the agency actually do. Okay?

If you look at kind of a very top level strategic view of the functions that we do to protect and promote the public health, we like to view it as a sort of life-cycle view, where you kind of start on the product development or pre-market review side. We're actually looking at reviewing applications that try to demonstrate the safety and effectiveness of new medical products.

Now on the food side there's a much more minor part of the pre-market, but we're here primarily to talk about human drugs today, and so that's probably the most relevant. Then, of course, if you do get approval you can start manufacturing and distributing these products and so there's another phase which we call sort of the product quality and safety, where it's really the inspection and making sure that manufacturers are following the FDA standards for quality and for safety of the products.

And then, of course, if you manufacture that product safely, get it out on the market, consumers, health care providers make their choices about using the products, and that's where we have our post-market consumer and patient safety where we're doing pharmaco vigilance and looking and monitoring and doing surveillance of how the products are being used, what kinds of unexpected adverse events are happening.

We're also providing information to the public, to health care providers on how to use the products safely. Another key part in that part. So you can see that we have this sort of life-cycle view and we learn from that experience and that feedback into the product development as well as how we handle the inspection.

And you can see we have -- very soon we'll be publishing a strategic action plan that's an update to the plan that Dr. McKellan put out, I guess four years ago. And in that you're going to see four strategic goals. And here you see three of the four are lined up with that life cycle view that we just showed you; improved patient and consumer safety, the second box in there is related to the post-market safety. You see increased access to new medical and food products lines up with the pre-market goal. Improve the quality and safety of manufactured products in the supply chain, of course, lines up with that product quality. And then in the very beginning we have sort of the interior and infrastructure view of strengthening FDA for today and tomorrow.

So looking at some of the commissioner's initiatives to strengthen the agency and looking to workforce and the infrastructure and some of the initiatives that Tim will be talking about also address that particular area.

Now, supporting that, those broad strategic goals, we have a number of objectives, long-term outcome goals, strategies, we have annual performance goals that are part of the Government Performance and Results Act, that if you ever want to weigh through our large telephone book of the performance budget every year, you will see a lot of information about annual goals in there. And then we have short-term action items that are more initiatives moving us more toward our vision of the future and some of those will be detailed in the strategic action plan that we will be releasing very soon.

Now, here is an eye chart here. The purpose is to show you that the Prescription Drug User Fee Act has had a number of performance goals that have changed over time. And, of course, from the very beginning the PDUFA program came in about the time of the Government Performance and Results Act, GPRA, so there was some parallel efforts of helping drive government towards more performance-based activities.

And I think PDUFA has been a really important driver to help us learn how to track our performance much better and we've done I think a very good job of learning how to improve our business processes, and track how we're doing and actually accomplish these PDUFA goals. And I just wanted to put this up here because it's really important to understand that the information management requirements of tracking all this information is really quite intense and so that's another one of the drivers for some of the information technology improvements that we will be talking about today.

Of course it's not just about tracking the work. It's actually looking at the information we're taking in. As you can imagine and many of you are contributors to this, there is an incredible volume of information that we have to process as part of protecting and promoting the public health and doing our job. You just look at the evaluation of new product, you see there are 240 new drug applications and licensing applications each year and, you know, there's literally the case where when a new application comes in sometimes there's a big semi-truck backing up to load in all of the paper. Right? But you're supposed to say, no, that's the old way.

What we're going to be doing and what we're going to be talking about is the new way where we're actually developing applications and submitting the applications electronically according to standards. So that's an incredible amount of information. You can see some of the numbers there not just on the pre-market side but looking at the oversight of manufacturing facilities that are really all around the world now, but there's still quite a number in the United States that we have to go out and inspect and manage the information there.

Then, of course, there's monitoring the product safety while it's out there being used and there are quite a number of different efforts. Event reports and other kinds of safety information that we have to review when it's submitted and process and analyze and also archive.

Now, of course, making this transition from a paper world to an electronic world is particularly important. We have a little metaphor here, it's not intended to be any product endorsement. It's merely just a, sort of a metaphor for the kind of transformation that the agency is going through.

If you look at what's going on in the world and you look at the increase in imports and increase in complexity of products that we're regulating, you can see that our workload has really been going up in a number of different dimensions for several years now. And we also know that there are a lot of competing demands in the federal government for resources.

So as responsible stewards of your tax dollars, we need to make sure that we're looking at about how we can become more productive and that's really what we're trying to do, is to become much more productive and one of the best ways we can do that is to sort of get more miles out of a given gallon of gas by being more efficient with our business processes, and that really means modernizing to the standardized electronic E-government type of solutions so we're not managing paper and electronic. That's the path that we're on and that we're going to talk about today.

Here's another illustration. We've got incredible amounts of paper that we're still managing and yet you can see when we go to an electronic document room, we've got a lot less space and a lot more efficient operations.

And we've been thinking about this hard and trying to make some hard choices about this. This is sort of a decision tree that represents some of the decisions that the agency has made and in strategically about how we're moving forward.

You know, the first box here, automated standards-based submission and review environment. Are we willing to commit to building that infrastructure. Because if we don't do that, then we're going to be stuck with partial automation, a lot of manual processes. Well, if the answer is yes, then we have to ask ourselves, are we willing to move toward having all submissions be in a standardized electronic format and there have been a number of forums where we've asked public and stakeholders about this question. Because if we don't do that, then we're going to have to continue with these parallel paper and electronic systems, which is not very efficient.

And if we do decide that that's the direction we need to go, then the question is are we willing to harmonize our business processes across different parts of the organization that are working on the same programs. You all know that the Center for Drug Evaluation and Research works alongside the Center for Biologics Evaluation and Research on the PDUFA program and they've gone up as separate strategic business units, if you will, several business organizations that have their own business processes and systems, so the question is can we harmonize those business processes. And I can say that we've been doing some work in that area and feel very encouraged by the good work of the folks in though two centers to move in that direction. Because if we can't do that, then we're going to continue with the separate systems that we have to keep trying to link together and that's not as efficient as we would like to be.

But if we can do that, then we can move towards a common enterprise IT system, which is really the goal, not necessarily uniformly, but where it makes sense and where there's a good business case for it. Those are some of the things that I think Tim will talk to you about in a few minutes.

So this is sort of the true confessions slide. If you look at where the agency is today, we really do not have, across the board, enough standardized format for how we're taking in information, you know. We get things in all kinds of media; not only just paper. We are getting some electronic submissions on physical media, like CDs and DVDs and we're also getting electronic submissions through the electronic gateway, which, of course, is the preferred way.

When you look at all the information, some people have said, you know, we have, in many cases more like a data landfill than a database. That's improving all the time. But when you look at how things have been in the past, you know, that's just the honest situation, and I'm sure many of you can probably point to your organizations that have worked through the same kinds of issues.

And we're trying very hard to fill gaps and to really clean up a lot of the duplication or disconnects in our databases when it comes to tracking facilities and product. We're making some excellent headway in that regard as well.

And when it comes to making the information available to the public, we're still trying to get a better handle on that. But you can see the opposite, where are we going, but we are moving towards standardized formats and terminology. We have been working in international standards, development organizations and work withing with other federal government standards organizations to try to make some great improvements there. We are trying to make this all electronic and secure and reliable and we've had some important developments there when it comes to our move to the White Oak Campus, the development of a data center there.

We want to do this because it's not just about being efficient. When we can get the information into the standardized databases, we're going to be able to do better science. And that's really the Holy Grail of this, improving public health. Being able to look across different products, product areas to do some meta analysis to really learn better how to make sure that products are safe and effective as they can be. And, of course, just being able to manage our information when it comes to facilities and products and being able to make that information appropriately redacted, of course, to, available to the public in a more timely way.

So how are we going to do this? It's a large undertaking. It involves both the engineers and technologists, but also the people who are responsible for running the programs. The program managers, the lead reviewers, the people that have the process expertise.

So what we've done is we've stood up what we call the FDA Bioinformatics Board and it's a governance board that cuts across all the different organizational units and it also represents a lot of different disciplines that you see, the supporting disciplines at the bottom of the chart, looking at all these issues from policy standpoint, from a performance metrics standpoint, looking at the business processes and the mechanics of those. Looking at data standards and technology and the financial aspects.

And how we've organized this is not according to center or organization, but we've organized it around those strategic functions that cut across all of the organizations. So you see there's a pre-market review business review board. I happen to be serving as a chair of that right now and we have these really outstanding people who represent the centers and know the pre-market review process very well, who are participating in how we try to organize this new way of thinking about information management.

There is also a board for product quality, for post-market safety and then we have some cross-cutting ones. Administrative services and a new one, scientific computing and computational sciences, trying to look at some of the research or some of the more intense computing needs, the true bioinformatics, if you will, that is becoming so important in this deal.

That's really all I wanted to do in terms of sort of setting the big picture stage to provide the context for where the agency is going, from the information management standpoint and now I will turn it over to Tim to talk more about the technology.

Thank you for your attention.

MR. STITELY: Thank you, Malcolm. I would also like to extend my welcome to everybody and thank everybody for coming here regardless of their busy schedules to provide us feedback.

I want to give you a little background on myself first because this is the first time that I've been with this group. I've been in the Federal Government for just over 17 years now. I started at NIH for 13 years, I spent two years at CMS, and I've transitioned here in March. And I am very happy and excited to be here. There's a lot of changing things going on here and I look forward to the feedback and input that we get from you today. Because stakeholder input and feedback is very important in this process.

I've met with various groups across the industries that we regulate to understand the differences in the regulatory processes and I know there are uniquenesses there as well as in our centers because of the uniqueness in each of the regulatory processes.

So as Malcolm said, as we go forward to an agency-wide approach to IT there still needs to be the flexibility to have that uniqueness to meet the regulatory processes. So we will standardize and come into common systems where it makes good business sense and it's the right thing to do and there will still be uniqueness where that is required as well.

This is my one slide. I'll be talking to it and hopefully touch the points that are up there. The PDUFA IT plan is a very important step in this process. It's providing the stakeholders, yourselves, the opportunity to the provide us input and direction as we move forward to meet the PDUFA IT goals. And we will be able to get that feedback both through this meeting and in an open comment period as we put the plan out for review.

From an agency approach we are definitely moving towards an enterprise IT approach. We're doing that in several ways. As Malcolm said, the Bioinformatics Board and Business Review Boards are a very important part of that and from an IT perspective we're supporting those across the board to make sure that we understand where the business is going and that we can enable the business, for that is what IT is here to do, to enable the business and not just do IT to do IT.

To accomplish this, starting this fiscal year, the beginning of October, all IT resources, both FTEs and funding, has come under the management of the CIO, which is a first tier at this agency, and so that again will allow us to have the breadth and look into what is going forward and make sure that we do align, where possible, for a common approach.

The -- in support of the PDUFA IT, I think Malcolm touched on this a bit, the harmonization that is going on already at CDER and CBER and across all the BRBs and across the product lines that we do. But specifically in CDER and CBER the initiative to create a tracking system and work flow system, to manage and process regulatory submissions is underway and is making good progress.

Going forward it's very important to get to automated, to an automated and standards-based IT environment. As we stated in the PDUFA commitment letter, the agency's long term going is a standards-based electronic submission process that will encompass the life cycle of the products that we regulate. And this goal was not only for the PDUFA side of things, but we're doing that as an agency across the agents to make sure that we have an end-to-end from discovery through, from discovery through consumption across the product lines that we regulate.

Standards are very critical to meeting this approach and we are very active across the federal government and federal health care IT initiatives as well as internationally to make sure that we align to those standards as they become available and to make sure that we can receive and exchange information going forward.

The PDUFA IT plan will be a very important mechanism to communicate our strategy and priorities in moving towards a standards-based environment. As Margo or Malcolm or somebody mentioned this morning, that this plan will be coming available later this year and again today is an opportunity for us receive input and feedback from our stakeholder base.

So with that, I will turn it back to -- well, I'll pass it onto Mark, and I do look forward to the feedback as we work through this process and I want to thank you again for your interest and participation in this meeting today.

MR. GRAY: All right. I guess I'll stand out front here. I thought I was going to have the security of the podium, but I guess I'll just go ahead.

What I'm going to do is talk a little bit about PDUFA IV, what's in our commitment letter, like what we say we're going to do in the next five years, talk about one of the mechanisms and areas of how we're going to communicate our progress and how some of the measures we're going to put around that.

So one of the things which was already touched on is, our goal is to move towards a standards-based environment. Under PDUFA we want to speed our progress towards that.

We've asked for additional funding and resources so we can move forward with that, but before we jump right into PDUFA IV, I wanted to talk about the progress we've clearly made in this area under PDUFA III -- all of these things that are listed right here were goals that we had or commitments that we agreed to under PDUFA III. We implemented the eCTD standards which put a standard format around submissions. We've seen a great uptake in the number of submissions we have received over the last few years. It started in 2003.

I'll talk about some of the measurements we're going to take to tell you the progress we're making. But for the first three or four years that we've had eCTD in production we've received 6,000 submissions. This year alone we received another 8,000. So we're encouraged by the uptake.

As Malcolm pointed out, we receive a lot more than 8,000 submissions a year. We receive hundreds of thousands of submissions when you look across both CBER and CDER. We've implemented the electronic submissions gateway that provides the opportunity for you to send in your commissions electronically over the Internet securely for us to take those in, process those submissions and get them to the review desk quicker.

We've actually had a number of companies or a couple companies present at meetings like the DIA, talk about the advantages of the electronic submissions gateway even for the larger submissions, because you don't have to then reload or download or however you want to look at it, load up your submissions of like 30 DVDs or tapes, because you know where they go to, they go to the document room and then those guys that Malcolm showed you sitting there. The room does get a little more crowded as some of those submissions come in and they've to load them in automatically.

Both centers have taken steps and if you've seen the production right now, have the ability to receive that submission, provide feedback on that submission that they have in the center and they automatically process those submissions so it gets to the reviewer's desk quicker.

We've seen, the gateway has been in production for a little over 15 months now, maybe up to a year and a half. I have some numbers just to let you know. We incorporated the errors stuff in there so we are receiving error reports before that, but when we first got the gateway we received about 7,000 a month. We now receive about 15,000 a month, the last two or three months so we doubled the amount of errors reporting through the gateway.

And then in the pre-market area, we're now seeing, we've received over 10,000 submissions, pre-market submissions at CDER and CBER that we're announcing on average the last two or three months of over a thousand submissions. So we're making some progress. We've put things in place. We know we have to move forward, we need to work together on how we move forward with the uptake of the electronic submissions. But those are two things that we can build upon as we move forward.

The progress towards a consolidated infrastructure -- we said that we would do analysis in PDUFA III, take a look at opportunities. One of the things we did just from an organizational point of view, we have a single organization that takes a look at the infrastructure across the board.

One of things that Tim touched on, one of the projects we do have on right now is looking across the board at our overall infrastructure, something we called the bioinformatics platform and that is really to take a look at our overall needs as we look forward and try to anticipate the additional information requirement that we'll have based on the new sciences that's happening.

And then we want to continue to improve our communications and technical interactions. Under PDUFA III we met quarterly with industry to talk about our progress and provide them updates on different projects we had going on. And we also had a PDUFA five-year plan there.

The goal in PDUFA III was that we had to produce it after six months. So we produced one after six months. We hit the goal. We actually updated it once since then so that was about maybe another year later. But there hasn't been an update to that plan since 2004. So one of the things we're going to change in PDUFA IV is to provide more of a rolling IT plan.

And then the last bullet really just builds on top of what Tim and Malcolm talked about. This is moving us towards our vision for a standards-based review and submissions environment and we want to move forward with that, because in a lot of ways PDUFA leads us in moving in that direction.

So what did we commit to under PDUFA? There are particular goals, and you can read the actual wording in the commitment letter, but this puts it into a little bit of a more, hopefully readable English and a little bit more understandable sometimes. We agreed that by the end of PDUFA IV applicants and sponsors will be able to submit their information in electronically with automatic links to information that they've already submitted.

And that ties into the second bullet point. Of course we need those tools so the reviewers can access it and do their analysis, but we feel this is one of the ways we can help streamline the process. You don't have to submit information again. You can automatically link back to it and it also, if you will, gives the reviewers a little comfort. Because whether you're sending it in paper or electronic, even though you say I have sent it in already, don't worry about that section over there, there's always -- it's just human nature to say, you know, I need to go take a look at that and see if that's really changed aren't or not. If you know there's an automatic link, you've already taken a look at it, you've already reviewed it, that helps streamline the process and helps make us more productive also.

Under structured product labeling, which we require right now, we know everybody is sending in right now, for labeling we want to, with the electronic capabilities we have, we want to make it a little bit more streamline to, do it at a sectional or module level. So, again, we can streamline the process. You're just sending in the section or the part of the label that you're changing and we can build our processes around there and streamline it so we can have that interaction and we know right away how the label has changed. That's very important.

Just talking about it from an electronic submission point of view. Right now somebody has to sit there and look at it before the SPL tells us, bearing in mind, what changed. So these are the types of directions we want to commit to looking at it from a sectional level.

Then I talked about the electronic submissions gateway. So we've asked you to send your submission in. We sent you back acknowledgments and thanks for having done that, but we want to expand that and if we are really going to make this a totally electronic environment, we want to look at ways of communicating back to you.

So in a lot of ways some of that capability is already there. We'll have to take a look at those processes. An example I would use is a lot of people that send in their electronic submissions through the gateway, it may go back to a person like me, the IT guy. You probably don't want your information requests going back to them for their approval. Oh, now we got approval for something today. I wonder who I should tell?

We will have to look at the process and make sure that it's in place and also internally, we want to make sure that we have the process in place, but we should make the whole thing electronic and that's really our goal.

And then we didn't talk too much about PDUFA overall, but a big change for PDUFA IV is in the area of drug safety and post-marketing. So I'm not here to talk about all the things we're going to do from a business end, but one of the advantages we have now, we have resources and funding to allow us to look at external databases, link to those outside databases so that we can do additional analysis, additional research and do some target surveillance around drug safety after the drugs hit the market.

And we're also going to update and modernize our adverse event systems. The AERS system has been very good up to this point in time, it's been in production for number of years, but we want to standardize that, provide some more analysis capabilities through that application, and additional reporting activities.

So those are the things we said we will do under PDUFA IV. So the other thing is process and measures. How are we going to communicate with our stakeholder base, how are we going to measure ourselves? Again, this is written into the commitment letter and I'll talk about the plan in a, on the next slide, but we're having a rolling five-year plan and we've agreed to provide updates to the plan as we move forward.

We plan to continue to have the quarterly meeting and to expand those a little bit and make them a little bit more operational at this point in time. At this point they've been mostly updates. I think most people agree that they've been -- I know it's been very beneficial from an FDA point of view, we've received positive feedback from industry on those meetings and we want to continue those and expand some of the discussions that we've had as we move forward.

And then we've talked about, well, how are we going to measure this? We've talked about electronic submissions and we want to move towards this. One of the measurements is really almost a joint measurement between us, our sponsors and our applicants. How many electronic submissions are we receiving? I can tell you overall that we get about 15 percent of our applications or submissions electronically.

That sounds -- you know, if I just give you that number, which we've given the public a number of times, that sounds pretty discouraging. But if I start saying well, for original submissions though, we're starting to see a 40 percent on those or supplements which would be more like 50 percent. Well, that changes the demand.

Now, we're getting more content in electronically. So we're going to look at those numbers and then as we report out on those, we're going to see progress. But also if there's areas that we can target. So for example, amendments, a lot of people, that simple amendment is just a process that we get a lot of those and continue to get those in paper. Maybe there's some way to determine how we can get everything electronically.

Then around complying with standards, one of the things we heard was, well, for example, with the eCTD, we've implemented, but we want feedback on that. So we're putting things in place, validation tools. So we'll communicate back to the individual sponsor or applicant, well, how are you doing, what kind of errors do you see? We also want to accumulate statistics around that so we can report out and tell people here's the common errors. Here's the type of errors that we're seeing. So for somebody who's just preparing the eCTD or has some questions, we can provide you with information on some of the pitfalls, if you will.

And then we keep talking about we're going to move towards this common environment. Malcolm mentioned we have an agency governance process, which we do. Tim has talked about some of the investments we have from an agency perspective, which we do. One of the ways is, well, how do we know we're moving towards that? Well, there's some proof in how we're moving towards that. We have a electronic submissions gateway that wasn't only built for PDUFA, that was built for the agency. For example, CDM has implemented a process, that's a veterinary drug.

It's taken advantage of the electronic submissions gateway. But we're going to provide information on our spending that says here's our money, we're moving to using this electronic submission. So the legacy IT system, just to let you know, is not just a system that's maintenance. What would go in that bucket is if for some reason we decided to do development on a center only system, CDER not CBER, that goes into that bucket. We will put for all common development, we'll report out how we're progressing and you should see that line moving up as we move to four.

Next slide.

So for the five-year plan, why we're here today, and one of the things we wanted to do is provide an opportunity for our stakeholders to provide us some feedback through this public meeting.

Through the plan, I'll skip the first bullet point for a second. But the full process is that we're working on the plan right now, we're going to take your input and figure out how we can change some of our communication in the plan. But we'll publish the plan by the end of this year for public comment.

So it's really a draft plan, if you will. This is our best thinking, this is how we think we want to communicate with you, but provides an opportunity for everyone to provide comments back to us. I believe we agreed that it would be at least a 45-day comment period. We'll take those comments in, take them into consideration and then develop our public final plan by the end of May.

We also agreed to put an annual assessment out within two months after each fiscal year. So we'll talk about the projects we're talking about and the things we say we're going to do. We'll put an assessment, here's where we're at, here's how we're doing, whether possibly there was a delay in a schedule or whether there was some change in direction. The annual assessment will do that. But very importantly we said we would update the plan periodically.

A lot of people thought we said annually. It may work out to be annually. We may see in the first part of PDUFA that we updated it after 10 months because there's a lot of activity. Maybe later in PDUFA it only gets updated after 15 months, but the key to this is that we'll provide a plan. So when we're talking about PDUFA V, there will actually be a plan out there talking about the direction we want to take as we go into PDUFA V. So we're not sort of -- it's not that we're starting from scratch, but at least we've had that communication and there's really a document that we can work from.

And then through the governance process and Malcolm mentioned that there's an agency governance process in place. Well, before, what we've done, what we did in PDUFA III is CBER and CDER got together, we produced a plan and said this is what we're going to do in PDUFA. I don't know what the rest of you guys are doing, but this is what we're going to do in PDUFA. We still have the PDUFA groups and we need to put the plan together.

But it goes through, for example, the pre-market BRB is going to look at our plan. They're going to have an opportunity internally to say -- we're going to have an opportunity internally to say this is the direction we're going in and have them provide feedback to us so that we can be in alignment with the direction they're going in too and they understand the direction that PDUFA is going in, because we do understand that this in many ways, PDUFA does lead the effort toward using, toward electronic submissions.

And then from the content, not to talk too much about that, but this is also again within the commitment letter. We'll talk about our vision, which we talked about a little bit already today. But more importantly, how do we want to get to that vision? So we'll talk about the business processes, talk about automation, that means at the project level. We're developing projects to, or systems to meet a particular process need. Talk about those changes, what we're doing, not just from a system point of view, but some of the processes that are involved.

And then electronic data standards. Everyone knows that's very important because that's how we're going to get to this more efficient environment.

So we said we would put in the plan, which makes perfect sense, the electronics standards we're considering for adoption and development if we see a need to develop it, but mostly adoption. What systems are we implementing, how are we implementing them? What processes are we targeting? And then how does that affect you. What are our time lines or many times, what's our strategy? We might not always be able to put time lines in if with your going through a standards development organization, but you can say our strategy is to get this to point, this point, this point and talk about different things, how you plan on approaching that.

Then from an overall perspective, what's our process around developing, adopting, considering, evaluating and deploying electronic data standards. What's the approach we want to take? For example, with the electronic gateway we had an approach, we put it out, we asked for industry partners to come in and pilot the gateway and most people thought it was a pretty successful model. Those are the kind of things we want to describe on an overall level and then we tailor it, of course, to each project. There can be a different need depending on what standard we're talking about or process we're talking about.

Next slide, please.

So this is my last slide, and I see we were supposed to go to 10:10. I will not spend 25 minutes on this slide.

Anyway, the goal here though and the goal for all of this, as was, Malcolm alluded to, is really to improve our processes, and our information analysis and the public health. No matter who I talk to, whether it's internal to the FDA or the people I've met in industry, that's always been our mission. Our mission is not to develop life or new shiny things for IT people so they can say I worked on the latest and greatest. You know, isn't that fun? Our overall mission is to make us more efficient.

We feel if we have information coming in that's standards-based that provides us opportunities to build tools, better analysis, do cross-analysis, will make us more productive and efficient, provide greater consistency across the FDA to talk some of the business process, about the harmonization we've already started in the plan.

Through the plan and through our discussions, hopefully you understand the direction that we're going in so that you can be in alignment, so that you can go back to your bosses and go this is what the FDA is doing, I think we should really put some funds towards that. And I've heard that before. If we communicate more, it makes your job easier when you go back to whoever you need to report to get the funding you need for your project. And we understand that this is a global environment and we need to look at harmonization across the whole globe.

And so this is something that's very important. So we want to improve how we're doing and a lot of that is through our mechanisms and communications. Most people understand that the U.S. Government is going in a certain direction, so the FDA has to be in alignment with that, but we want to improve so that you understand if we have to implement something based on what the government says, we're going to implement that, but you understand the direction and you understand the decision making around that process.

So that's what I had for today. I appreciate everybody showing up. As Margo alluded to, we had some issues with the room so I'm really happy we didn't have to come up with an overflow room, because we didn't have one. So this is the perfect amount of people so far and I'll turn it over to Margo. Thank you.

MS. BURNETTE: How are we doing without the microphone? We requested one, but those in the back, can you hear okay? Yeah? Okay. Good.

We're ahead of schedule. One comment I do want to make is I know we have a schedule here that says this person will present at this certain time and that's assuming each individual took the maximum amount of time. So please note that we're going to keep on rolling. So if you're scheduled to present at 2 o'clock, you might not want to come running in the door at 2.

One of the things, too, is I want to thank Malcolm and Tim and Mark for their comments and hope that was helpful to put it all in broad context for you. One thing that Mark did is reverted into a few of those acronyms without explaining, so I wanted to explain what he meant when he said it would be approved by the BIB, we're not a bunch of babies. That's what we call the Bioinformatics Board. That's just an abbreviation. And the BRBs, that's the Business Review Boards that Malcolm mentioned. So the BIB is the Bioinformatics Board.

Is Liquent ready to proceed? Okay. I would like to go ahead and introduce somebody from Liquent, which is a Thomson Scientific business.

MR. NICHOLS: I'll use the podium. If you can't hear me. Let me know. First of all, I want to thank you for having us here. You had posted a list of questions or areas you wanted some feedback on and so what we thought we would do is give you our thoughts from a vendor's perspective, who also has access to a pretty large selection of sponsors and applicants who we collect feedback from through a number of means.

So I did want to give the caveat that the areas that I'll be addressing are the areas that, where we do interact with the stakeholders involved. There are a couple of questions that we probably would not find applicable for us.

So on the question about what would help to improve the agency's, the quality of electronic submissions to the agency, there are three main areas where we thought you had a great influence on quality. You can put the whole slide up.

Standards definitions, consistency of standards within the FDA and communication with stakeholders. So if I talk about standards definitions first, it's our understanding, our belief that global standards are the most desirable things to have as well as submissions validation criteria standards. An example there would be the use of the three letter prefix in the folders being used for storing electronic submissions and how does that relate to cross-application references.

Also within SPL we think that there would be a desire for some clarification on your thoughts regarding PLR and the use of MedDRA and SNOMED coding. We've heard that there are some questions around that. And then finally, and you began addressing this in your earlier presentations, some improved feedback mechanism for communicating the non-critical technical issues and a process for quickly correcting those. I don't know if that's necessarily getting into bi-directionality, but a least a more clear process around that.

So relative to consistency of standards within the FDA, we think that there is a desire for standard rules for electronic submissions across the different FDA centers, which would include the animal health and combination products, et cetera, as well as the incorporation of DDMAC and advertising and the promation labeling submissions into the eCTD process.

And again, finally increased communication between reviewers and the Office of Business Process that would allow you to provide some instructions on how to handle requests for information outside of the typical eCTD specifications, such as a request for data.

In terms of the last point, communications with stakeholders, you've already mentioned this a few times today. I'm messing up my own slide, sorry.

Some guidance clarification. We annually conduct a survey of our, of the marketplace and one of the things that came out as one of the biggest reasons why people were not quickly moving towards electronic submissions was a lack of clarity and guidance from the authorities around some key issues.

I believe we're on the wrong slide.

So included in that I would say is that some guidance clarification around folks who have already been using study tagging files and how that should be carried forward, cross-application references, specifically, especially within the eCTDs, the hyperlinking issue and this is one that was brought to us specifically by one of our clients knowing we were coming here today.

And then finally, some clarification around INDs in the eCTD format. I think it would also be important that industry understands some lessons learned from the submissions that you have received. Is there a scorecard process or benchmarking process that you're looking to put in place that helps guide industry to understand what did a good eCTD look like, for example, and what made it more reviewable than another one, perhaps?

And understanding the validation criteria that you're using across those submissions. I think an example of that that has been helpful for us has been some of the activity in Europe with the e-Ticks process that helps us understand across multiple scenarios how the information is processed and validated.

Also expectations around documentation. Again, this is another pointer at the hyperlinking and cross-referencing question, but they can turn into document links, especially across and including life-cycle information.

And also what do you expect around document level tables of contents?

People are always asking for some clarification on life-cycle management operators. When should one be used versus another. What are your expectations of their usage, if that could be clearly documented and communicated that would be helpful.

And then finally a reviewer training which I think was already mentioned as well, but making that a somewhat clearer process that we understand how reviewers are being trained so people can also comment on that process and we will understand how they're doing their review process.

So what would help increase the quantity of electronic submission to the agency? Basically some incentive innovation and broadening of the acceptance of those.

Some feedback we've heard and it came from some of the other standards in Europe where there's some incentive to have faster action dates or temporarily reducing fees to encourage people to get through this process and then increasing the acceptance cross within the FDA. I mentioned the DDMAC and advertising and promation submissions and eCTD would be helpful. And would it be okay to move towards a process of providing graphical representations of packaging the requests for a physical sample as an option.

In terms of next question about prioritization, the first priority that we would like to see is the consistency of standards within the FDA. After that I think the communication is the next biggest item and the rest you can see here. But again, consistency and communication should be a theme that you're hearing here in terms of our feedback.

The next slide. What data standards are needed to implement these improvements? We think a standard for exchange of promotional materials, a communication exchange, the clinical and non-clinical exchange of information with STPM and send. And there's some considerations for XML document exchange beyond what you're currently receiving in terms of structured product labeling, but where are you in considering this protocol and in terms of stability information and perhaps others.

We think that SPL is still being approached as pretty much a tactical process and once, we've got to kick the volume over to more content being delivered in a structured process where people will start committing to that as a delivery mechanism.

How should the FDA engage stakeholders? Public meetings like this, the quarterly meetings sound like a great idea. Increased ability to comment on these processes and the formation of work groups or testing teams we think would also be a valuable process to engage in and those teams could also follow through to post-mortem implementation to look for opportunities for ongoing improvement.

In terms of lead time for the stakeholders, there's basically three categories that we would put this in. Minor changes, the complexity of the change really drives the lead time filter. The minor change, something like a valid value for an attribute, probably 3 to 9 months from the final standard. This isn't necessarily only to adapt or adjust for new technology, but a lot of times it more often is the process that goes around that needs to be adjusted within a sponsor's organization so that they can get to that implementation.

We think moderate changes such as the implementation or move to a new version, a major version of a standard, probably about a year, and then major changes such as the regulated product submission standard implementation, probably two years from final standard in terms of opinion or adoption.

The data standard areas what could provide the greatest challenge. I think anything that's associated with currently unstructured content moving to a structured format. Such as protocols and labeling is a great first example of that. I think that will be the area for greatest challenge because from an applicant or sponsor's perspective, moving from a very comfortable unstructured Microsoft Word type environment to something highly structured is a huge hurdle to overcome from a process and technology perspective inside.

I think any standards associated with content level also have a continual life cycle over many years, whether structured or unstructured, will be a great challenge because people will need to find a format or a method or a process for keeping track of those changes for the 10, 15 or 20 years.

And those that I think -- that would be most difficult for an industry to implement or for the agency to manage would be the ones that obviously created the greatest challenge.

Approaching that we think will facilitate the most effective and efficient adoption. Early involvement from agency and industry and software vendors. Getting all the players involved as we're dealing with these technology standards, involved as quickly and early as possible. Looking for global input in early standards development.

We saw that, what happened with the first go around with RPS that the word global was sort of pulled out of that because enough people didn't feel like they had input from a global perspective. Anything that we're hearing that has a more global ability to be implemented is something that's easier to adopt because that's more usable across multiple agencies. A commitment to harmonized standard is what we're hearing through PDUFA IV.

And then lastly, we think there's also some value in considering the assignment of a business process manager to the sponsors to assist in those questions on the development and maintenance of their eCTDs. Who do they go to with questions as they're trying to work on an eCTD that can help shepherd through the process within the agency to get answers to them.

The key areas required, new or expanded electronic commission guidance. There's a quite a list there with eCTD, including as I mentioned, life-cycle management, clarification of INDs, study tagging files, migration from one version to another. When we go to version 3.3.3, what would be some recommendations from you regarding migration?

Case report forms from an eCTD process. And then on the regulated products submission standard it would be helpful for to us know the current agency thinking on its usage and implementation. We get asked that a lot when this comes around, how quickly will the agency be in a position to start accepting that or want to accept that. And then some of the other centers, what are their expectations around some of these electronic submissions?

Lessons learned. I think some suggestions here would be to look at the revision to the EU telematic strategy to see if there's possible recommendations that could be implemented here.

Some other possibilities would be in looking at other enterprise standards, a consistent process across centers. Your training program for reviewers and making that information publicly available so that sponsors and applicants understand how reviewers are trained in the process and obviously the implementation of scalable and open technologies.

Finally, in terms of the specific concerns I'm not sure that this is necessarily an issue. We would expect all third parties used by the FDA to abide by confidentiality agreements and I think third parties being used by sponsors would be held to that same standard. But there was one area you would want to make sure you specifically looked at that I know people worry about a lot is the security areas in terms of the access and transfer of information, and ensuring that the third party computers are secure as well.

That's it.

MS. BURNETTE: Anybody have any questions?

MR. BERTONI: Just one question. Could you clarify when you -- I think you mentioned something about surveys or something, the results provided here, were they based on the survey or just your company's opinions? What did they represent?

MR. NICHOLS: A blend of that information. The question is: What is the basis for our positions here. And I had mentioned the survey. We do on an annual basis conduct an industry-wide regulatory compliance or regulatory survey in terms of trends and what people are worrying about in terms of certain key areas or IT submissions, outsourcing.

So that is in large way a contributing source to our position as well as conversations that we have directly with clients where we solicited complaints prior to coming here today.

MR. BERTONI: So would it be fair to say this is your firm's sort of integration of your opinions and your survey results and conversations and really reflects --

MR. NICHOLS: Yes. The question was is this an integration of all that information. Yes, it is.

MR. BERTONI: Very helpful.

MS. BURNETTE: Any other questions? All right. Thank you. We will go ahead and take a break and we have 20 minutes for the break. We will give you 10 extra. Come back at 10:30. We'll have our first speaker back at 10:30 and could the next two speakers see me, introduce me to -- BIO PhRMA and the SAFE-BioPharma. Whoever is going to be doing those presentations, come and see me so I'll have names. Thank you. See you back at 10:30.

- - -
(Recessed at 10:02 a.m.)
(Reconvened at 10:26 a.m.)
- - -

MS. BURNETTE: Okay. We're going to go ahead and get started on time here. One thing is we are a little ahead of schedule and we've looked at -- if you want to go ahead and move the 1 o'clock speakers up to 11:30. And what we're going to do is we're going to go ahead and break for lunch at 11:30 and have from 11:30 to 1. We only had an hour allowed, but that would be very tight for you to get lunch and come back in. So we will be breaking about 11:30 and we'll start back right at 1. So be sure and be here. We will get started right then. Okay?

Our next presentation is from BIO PhRMA and we have two presenters and -- where's Rachel? There she is. Okay. So we have, from BIO PhRMA we have Steve Ward, Allen Jones and Rachel Carle, who will be presenting.

MR. JONES: I'm Al Jones from GlaxoSmithKline and this is Rachel Carle from Genzyme and Steve Ward from Eli Lilly are here with me. We want to take the opportunity to thank the agency. I think we're at a good place as we move toward automation. Thank you.

The Biotechnology Industry Organization and the Pharmaceutical Research and Manufacturing of America represent over a thousand companies worldwide and they spend over 60 billion dollars a year in the research, development and manufacturing marketing of products for human health.

I think the first point we want to make is we're in this together to partnership and it's not our role to work to divergent purposes. We are enablers to support business process. I think that's something that we heard from the FDA presenters earlier and from Jim as well, that what we want to do at this juncture as we move forward with PDUFA IV is make sure that the business process is working as efficiently as they can.

If we can go to the next slide. We are going to do a few things today. We are going to answer a few of the questions that were raised today, and we are also going to talk about a survey that we have begun. We haven't finished it so the survey results are incomplete. We will teach you a little bit of some of the results we have got, but we will be providing those a bit later and we will talk about that as we go forward.

Quality is always an issue. What we would like to talk about this morning is not just the quality between individual sponsor and agency, but what does the community look like from the quality perspective. How can the community affect the better quality of the applications of the agency?

We would like to see the agency post common errors on its Website. We would like to see the FDA's validation tool made available publicly so that we can do that validation ourselves and what we get is a better quality to save you that time and effort.

Particularly when we look at the numbers that were presented I think by Malcolm for how many submissions that you receive. We think if we create a forum of best practices, concerns and common issues where we got together and talked about what does a good document look like, what does a good submission look like. It would be a benefit to the broader community. What's a QH hit list? How could we improve the quality -- how could we improve the quality of the submissions that we are building. Let's get together, do that if we can.

Technical discussions offer us an opportunity to improve the quality of the submissions that we have. As our company has moved to eCTD, we've utilized the services of the Office of Business Process support at CDER quite a bit to hold outside of the pre-NDA application meetings, conversations about format and eCTD. Those have been extremely valuable and we think that's the best practice. We would like to see that happen across the agency and the other centers as well.

Often there are times when at the pre-NDA meetings and the pre-application meetings technical discussions should be held. And so let's make sure that we include, if there's a statistical package that's going to be prepared and submitted so that the reviewer can do in situ evaluation, let's include those at the time.

We've heard the global drum beat. We want to make sure we continue that. The global environment is critical. There's a safety event in Thailand this morning. There's a safety event that's a worldwide event. We must evaluate it globally, if it's appropriate to be submitted to the FDA, we want to make sure we do that. It's got a quality and a quantity component.

We would like to see the establishment of end points for legacy formats. Our submission publishing managers actually keep a spread sheet of all the various formats we have submitted, submissions in, to make sure that we do life cycle in the right format. I can only imagine what it must be like for the agency as well. So that's an area we would like to see some improvement.

Implementation and the execution of the implementation embedding is something that's critical from a quality standpoint. It's not just good enough to establish the standard. It's not good enough to set it out. To return the investment that we're taking on these IT initiatives we really need to make sure that we embed the implementation.

Next slide. What can we do from a quantity perspective? Facilitate business process, make it easy to do data work electronically. Often I see us do double work, I call it, such as having a phone call with someone from the agency and then we will document it -- okay. It works.

Often I'll see us have a telephone call or a fax event that's unofficial and we have to follow it with an official event. Let's see if we can do that one time. What if we were to build a submission capability that online you could have that event occur officially and only one time. When we re-engineer processes, let's make sure we do an electronic process and not the preparation -- I'll hold it now. Thank you very much.

Often, when we re-engineer process, let's make sure we do it electronically and not in paper. It wouldn't do, for example, to take a CD and create a PDF form and send that PDF form. We need to do that as a database. Let's register an investigator's CD in the database and send you the pointer today. Let's do the same thing for the 1572, let's hold that in a dual electronic capability. Electronic standards can be incentives or added burdens. So let's make sure when we do that it's actually going to save us money and time.

What's the benefit of paper -- of electronic over paper? I think one of the most compelling presentations that I heard was a reviewer from the agency at an FDIA meeting last year talk about the fact that she had access to an eCTD submission and she had a question on a data set. From within the margin of her PDF file she was able to do an analysis and in ten minutes satisfy the situation that she was concerned about and continue with the review.

She contrasted that to the paper model where she would have to call the sponsor, wait a week to get that, write the review cycle. That's an incredibly powerful statement of the value of electronic submissions. So let's try to put a cost to that. Cost is time, time is money and money is speed to patient and improvement for the betterment of human patient.

Harmonize across the various agencies. There's a quantity component to that. Every time we have to customize a submission for one of our agencies, it's costing us time, it's taking time away from us to create another submission that we can submit to you in electronic format.

Implement consistently across centers. Again, virtual preferences or center preferences are very difficult for us to maintain. It's important to engage the vendor community throughout the development and implementation process. It's a partnership. It's a multi-wave partnership. In order for us to move to the next generation of what we want to do, we need them in the room with us. So it can be tricky at times, but it's the right thing to do. Next slide please.

What data standards are needed to implement these improvements? While that's clearly one piece of the pulse we really think a road map is critical. So we are extremely excited about the five-year plan. What we would like to do is invite you on a helicopter ride. We're going from Washington to San Francisco or we're going from Rockville to Oakland, we're not exactly sure, but we've advocated the development of a standards-based road map and end-to-end process.

From initial dialogue with the agency prior to the AND, through to the full life cycle of the product, its safety model, which may include -- if we follow the safety events that might occur with a product, and we want to make sure that that road map continues. We prepared and submitted and presented at the meetings. We need to understand that there are resources with which we can apply to this. It needs to be a business process driven investment initiative. So what brings the biggest value to our work, our effort? How do we go about meeting that demand?

Now, to elucidate what we've done within the industry, we've created a survey that we've sent out to all our constituent members. We've gotten just a little feedback. We will get more over time and we would like to share a little bit of information with you about that.

First a little bit about the survey. Next slide, please. The data are incomplete and there is a lot of adversity at this point, but there are some trends we would like to share with you. Number one, it's clear to us that terminology is going to be a big issue. So we spent over a decade gathering electronic content and we have done a very good job of describing it. So the ability to interrogate it and to derive some of the cross-application information that Malcolm mentioned earlier is really not a capability that we've got right now. So we think that's going to be one area that we need to focus on.

We like the idea of a shared service model, public/private partnerships that's way for us to share costs. That's a way for us to ensure that the business process flows effectively and it's something we think is a key enabler as we move forward.

We need to make sure that our vision includes that business process return investment, embedding the capability beyond just setting it out there is going to be critical.

I would like to give you some of the parameters. You don't have this as a slide, but you can leave that up if you would like to. How we've conducted the surveyor or what the survey contains.

So we've asked companies to give us input on 15 cross-functional capabilities and 22 initiatives. Fifty-seven process oriented capabilities for current initiatives in the discovery, treatment, regulatory and post-marketing areas. So we looked across the whole spectrum of the research, development and marketing areas.

We asked them to write on value, impact on the advancement of human health care, management of pre-information marketing, safety, efficiency of drug application, the evaluation process, promotion of global harmonization and the support of a consistent and technical architecture.

We've also asked them to write on implementation impact and specifically on people related to retaining the stacking requirements, impact on process, impact on technology.

So those surveys are out in our constituency now for evaluation.

Here are the preliminary survey results. Again, these are broad stroke. What can you can see from a process specific standpoint, eClinical is one of the high areas. What we mean by that is the whole space of designing this study, conducting the study, gathering the data, interpreting the data, reporting the data. So that whole process, really the IND phase of the regulatory process.

From a cross-functional standpoint, I mentioned these earlier, terminology, the public/private partnership. We think critics and the work they've done so far has been tremendously valuable. We think it's a great way forward. We would like to see those types of things continue, and data aggregation. SDTM has been a tremendous activity. What happened with CDISC is the right way forward. We would like to see that through to conclusion. We would like to see that brought to maturity before we move to the next stage. For example, XML, although we understand XML will come at a later stage.

In the medium category, eCTD, investigator portal and structured-product label is seen as the process specific medium category, while the eGateway and eSignature is in the medium. We don't understand that trend, but we're wondering since we are engaged in those it's seen as on the books. It's not necessarily forward looking as much as an embedding type of technology.

And scoring low at this point, but maybe coming up fast, we don't know, stability and financial disclosure.

There are some opportunities beyond this, however, that we will be reporting on as we come forward.

I want to take this opportunity to thank the agency for speaking with you today and now I would like to have Steve take the microphone. I'll hold it for you if you don't mind.

MR. WARD: I think I can handle that. As we went on to address question 5 that was put forward in the public notice, we really thought there was a little more important way to frame the question. We would like to rephrase it as how should FDA engage stakeholders to achieve the strategy plans that are enabled by IT.

So that's why we framed up there we thought it important that we talk about driving change that we frame what those business opportunities are and get that into a discussion between the business leadership at the agency and the business leadership in the industry and service providers or vendors that can get that alignment. We understand that oftentimes FDA will have to drive prescriptively. But where there is an information opportunity to inform or drive that alignment it will help ensure that that happens before we begin the development of IT standards, that we can get the right time lines and resources aligned and more in sync with what the FDA's time line is driving and make sure that will work for all of us as well. Then we can dive into implementation, we recognize that while the agency is committed already to working with all the organizations, that there is a challenge because the --

Because of that, we really need to continue to promote this broad constituency that would be active in participating with development of standards, taking those business process goals and translating them into IT capability enablers. We think that an implementation working group that does have broad expectation of the membership has the means to go there.

And we really believe that we can leverage the trade groups to get that effort started. We recognize the feedback that's been given before about not wanting to work solely with a trade group, but we also belief that there is opportunities such as having them get the working group where the trade associations heard the need to get a multi-agency group together, started that effort and really have responded to it in a group now that probably does not even recognize it was started by PhRMA and the leadership.

So it has developed into that broad contingency, but coming to the PhRMA group first and saying there is a need, we need to get people engaged from across the industry into this and we will be glad to work with the agency to get that participation.

We think that in addition we need to take a look at efforts that have gone on for the last decade or so with the conference on harmonization and learn from those and help move those processes forward, recognizing that really what that represents in the broadest terms is the form for regulatory, industry, leadership to come together and talk about alignment and opportunities for harmonization, but perhaps not be the best organization for the next step which is to develop IT that will be used in those systems.

So we are working to try to move that process forward to recognize the value of working with other internationally recognized standards organizations and believe there is a model that we can meet with leadership to get alignment on process changes and those requirements off to an institution that has expertise in developing technical standards, bring those results back to that forum to test them to ensure they achieve their purpose and drive that work forward into implementation.

We believe that as Alan mentioned previously, the opportunity of a public/private partnership is probably an ideal way to try to drive forward a consistent business model. That model is exercised through the NCI's efforts. If we think of that as one turn of the crank and turn that two or three more times and the agency were to agree to participate in some form of public/private partnership, I think we could find it becomes much more effective in driving forward. And we have to ensure all needs are being met and there is a win win proposition for others to align their resources with the same time line.

That leads us into talking about a life cycle policy for version standards and that really dovetails into the next slide that talks about question 6, what topics would be most useful to include in an IT plan. The elements we have listed here map really well to the slide that Mark shared from FDA's commitment of what to put in an IT plan.

We talk about a plan that presents the linkage between the business and the regulatory process, how you drive that business process and needs into required capability and then finally to enabling solutions. So you take those IT plans and goals and drive from them the information and the capability that satisfies that business need, and identify where and how you are working to develop those standards and systems.

So we think that FDA is on the right path to build that plan that will help the industry line up our resources to work within them and have the transparency to see where FDA is moving.

As we talk about the life cycle and timing, it's probably the one area we heard the most feedback from in the past year or two where it's difficult to predict the timing of change coming forward, the timing of regulation and guidance. We understand that when we talk about trying to approach this plan of the life cycle is that sense of having a routine life cycle, that was also presented in the previous slides, where you think of going from that idea stage where you create this sort of initial awareness and alignment on the business process change. Frame your plans into a development stage where you are perhaps working to identify the requirements, work with standards organizations to build the standards, draft standards and are developing initial systems into testing, following the implementation and system retirement.

So while the exact time frames may not be accurate always in a five-year plan, understanding the scale of the change and what the agency perspective is will help industry scale appropriately as well our plans.

Unfortunately the more we shortchange those or have a lack of transparency, the greater risk we have that we won't be aligned to those efforts, have costs or not be prepared to meet those terms.

Next slide please. Then we have question 12 which was asking what lessens are learned and best practices that we could include here. I apologize for the business slide with the bullet point list, but I'll pull out just a few to highlight and that is talking about the initial capabilities of the highest value and making sure that we get the stakeholders involved in the aspects there. Recognizing that it's not just one entity and to implement these would take a consensus or implementation effort across industry, service providers and vendors.

We really understand that the agency has embraced the concept in terms of engineering a new vision and not simply automating legacy processes. How do we really achieve an optimal business process? And we recognize oftentimes that optimization will span well beyond the FDA's control arena.

The FDA had perhaps gone out and tried to optimize their existing system for collecting registration information that is submitted responses. We have come up with a different solution, having brought clinicians together, research institutions and PhRMA companies and really talking about can we redesign the process as it flows across all of them for the benefit of every one of those parties and what best promotes better science and public health.

When the opportunity is there, we believe that getting that consensus can result in value and still meet all the FDA's true requirements, but also offer opportunities that would help fund the implements of those efforts.

Next talking about efficient and effective solutions. We are supportive of a public/private partnership as a means to do that. I believe that offers an alternative, using CRADA as a way to try to provide solutions to the FDA at reduced costs, knowing that everyone is cost sensitive and needs to spend our investments wisely.

The last point gets to framing. Understanding this is a five-year rolling plan with the periodic updates that may come from that probably would be most helpful as the industry sits down in our budgeting and business process to say where do we believe, and time frames are never exact, but where do we believe the FDA is trying to drive change and implement and we will need engagement from the industry so that we can still align our resources and thoughts.

Oftentimes in the early stages that would only require a few companies to be involved and a representative of different aspects, large companies and small companies, to partner in those initial pilots or proof of concept exercises. As you see, from the life cycle, it's scaling up to broader and full implementations. And what that does is it really says that change then is really no longer disruptive to industry if it's dropped in and unannounced, but based in the five-year plan and having a transparency that comes with that, it could be part of the planning cycles of industry and aligned appropriately.

So that leads us to the last question that we came prepared to address today about specific concerns that might exist for a third party entity. I would frame this overall by saying we have provided feedback to this question and this response, but actually much more detailed feedback was provided at the December public meeting that specifically focused on the role and whether FDA should engage in public/private partnerships and we would refer back to more detailed feedback.

But paramount at the top of the list would be one with regard to protection of intellectual property and the ability to secure it and assure that confidentiality of information that's done and then a standing service providing would expect reliability and accountability and affordability play in and not be different from anybody looking to bring in a third party service, but going through that.

The last point would be one that we would maintain and might be misinterpreted if not explained a little further, but in terms of maintaining the set up for capital investment by innovators. We are talking about the service providers, the software vendors to help provide solutions to our industry so we don't have to develop everything from the ground up. We believe there is a way forward to use a shared service provider that can still leverage commercial interests and provide an opportunity for them to realize value from their innovation and we would encourage that as a framework going forward.

So in conclusion, we fully support FDA's commitment to public and maintain this five-year plan. We believe they are aligned with the industry's desires and direct going forward is appropriate. We believe this IT plan, a major benefit will be the prioritization and ability for all sides to be on target with our budget and resources that have the most valor or the most urgent needs.

We believe that this is an effective vehicle because it provides a single point of reference for standards and systems and initiatives associated with them and allows for all the stakeholders to have that transparency and resources to meet the FDA's demands. It also enabled collaboration for greater transparency.

We realize that we have not addressed all the questions today. PhRMAs do intend to address the questions that were not framed today as well as provide some additional perspective and a much more detailed analysis coming from our survey results, which are still being analyzed.

We thank you for your efforts to develop an IT plan that helps link us to the strategic goals and thanks for bringing us together in this room.

MS. BURNETTE: Anyone have questions for Steve or Alan or Rachel? No. Thank you. You did perfect on your timing. Our next speaker is Cindy Cullen, a CTO from SAFE-BioPharma Association.

MS. CULLEN: Thank you. Anybody who is interested, I can summarize my talk in two words. First, standards and standardization and incentives. If you really don't want to hang around, it's like the old 70's movie. They would give the last scene of the movie and then they would start the movie. So that's how I'm doing this talk basically. That's the end.

And so this morning I'll give you a quick overview of the SAFE-BioPharma stand. Maybe we could do hands of who is actually familiar with SAFE-BioPharma? Oh, I'm very impressed. It warms my heart to see so many people are familiar with it.

And then -- but I'll still do the quick one for those hands I do not see raised. Then we will go into the business paradigm, changes that are happening within the industry and the importance of process improvement, some of what we are seeing our members as the industry needs and wants and then a little bit about the regulatory and legal implications as well as addressing directly some of the strategic questions.

So SAFE is basically a consortium that we have 8 founding members from the BioPharma industry and we have expanded into 12 members. We are primarily focused on the incentive and moving to an electronic environment. Our space is really the authentication, digital signature space, with -- so basically it's identity management. And we have developed a standard to facilitate that identity management, as well as the legal and contractual framework to support that.

It has actually many factors. So we have a multifactor as well. So we are directly linked into the standards from the federal bridge and that includes the basic need of assurance, levels of assurance.

In the business paradigm, the industry is completely changing. We have the changing business model. We have the in-sourcing, the outsourcing, the off-shoring, the near-shoring, the on-shoring, so all of that, all of the business processes are changing.

And that is directly impacting the legal framework. So if you are not doing your, all of your business onshore, then you are going to have to start dealing with the legal frameworks of around the world. And SAFE assists in that.

And then that is being facilitated by the changing technology. So we really want to be able to do all of the changing paradigms within business. If it hadn't been for the changing technology with, for example, the network infrastructure where you have the huge bandwidth available throughout the world. And that leads then to the need for a trust model, a ubiquitous trust model that can be, adapt to around the world so that you can have an identity that's basically understood and accepted in any part of the world.

And so the infrastructure is there to support the changing business model. And the changing business model is both within the IT as well as within the business itself. So we are doing the outsourcing of the business as well as the IT.

And this is being managed with the risk in mind so there's a distinct risk management component. And it is -- it's always towards the cost containment. We can become a very competitive business and it does require cost containment. And, of course, there is the regulatory impacts, not only from the FDA, but from around the world.

And these are a couple of the pain points that our members have basically said are specific pain points. The whole global regulatory issues. You know, the, that we don't -- we have to have, to deal with regulations from various industries and I will have a slide on this that really shows that around the world we have different regulatory bodies that we have to deal with as PhRMA companies and they sometimes have conflicting regulations.

The -- there is no standard at present in the identity management and the trust phase and if we link this back to our previous speaker, he was talking about the change across the whole industry. The one thing that we don't have is an identity that would follow someone from across all of the business process.

Inter-operability is always a huge issue, as well as the ability to legally enforce something. And the need for a business process improvement is primarily driven by the reduced cost, increased productivity, as well as reduced cycle times.

The process improvement, basically when you look at paper, you're talking about having to manage, store, archive, and retrieve the paper. And searching paper is a really challenge, and even searching digital data is a real challenge. If that were true then we would be one of the fastest growing companies in the country and searching is a huge issue. So the movement away from paper is very important.

If we look at the next bullet, it really shows how identity management is an expensive space. You've got 700,000 individuals in total in the clinical space, which is in fact a major reason why the critic's initiative is going after the fact of space. You have some individuals that are in need of one identity. What happens is that you have -- oftentimes each company will provide a laptop and then they will provide a user ID and a password for that laptop. And then, now that companies are moving away from the single factor into the multifactor authentication mechanism, that one individual will, in fact, often having multiple -- so they will have five or ten secure IDs or two secure IDs, a smart card, various other things that they have to manage as well as the passwords around that.

So it's a very onerous process and very costly. So in streamlining this process, there is the potential of huge savings in moving the business process away from paper. And there is a potential for eliminating data reentry. Because a lot of times now the CRO will enter it into their system and then the PhRMA will then have to enter it into their own system, and the potential advantages is that it's, is the speed of those across businesses.

What we found with incentive is when we were analyzing the business processes and attempting to integrate digital signatures into the business processes there were a lot of signatures that weren't really required, that a lot of the documents were using signatures that were really for approvals and authorization, when something like an initial would work rather than a full signature. So that there is the ability to streamline your business processes when you are moving those into the digital space.

Authentication standards are definitely important. Similar to the previous example where the person has the multiple laptops with multiple user IDs with passwords. If you could have one ID and one password, there would be so much cost savings with not having to manage the passwords as well as ease of use from the participant.

And then the industry has a lot of work rights and in-sigma processes, which is basically to optimize with the whole business process.

And here we have the pretty picture of all of the regulations around the world. And this is by no means a comprehensive list of regulations around the world. But it gives you the feel for how many regulations need to be followed if you are attempting to use drugs or medicines throughout the world.

And so the point of this slide is that, to try to get a pulse, a standardization around the whole regulatory environment so that we don't end up having the conflicts. Even within the United States, the pedigree space, the states are intending to run with that and it's causing major, major issues with the fact that the Florida standard doesn't match with the California standard and you are ending up having conflicts within the standards, and that if a drug is manufactured in Florida and sold in Alabama versus in California, you know, the manufacturers are going to have to manage all of that. So it would be wonderful if we could get that all cleaned up. I don't know if that's -- I won't comment beyond that.

So this is -- member implementations that are being done by the members of SAFE. I think there is a couple of interesting things that -- I'm not going to go over all of these in detail. But you will notice there are multiple projects being done where they are basically doing digital signatures within lab notebooks.

The scientists have basically said we don't care if you put in the digital signature. We are just so happy to have it in electronic form, that the moving to the digital signature was more of a cream than the actual meat of the process for the researchers. Because what they got was that search capability.

So they were -- previously it was difficult for them to find out what was happening, what the researchers were doing down the hall, much less the researchers that are around the world and with the eLab notebook projects they're actually able to know what's going on around the world. So you are not having so much rework done within the lab space.

And the other space is actually the, the other end of it, the ETC submissions, so there are a substantial number of companies that are using the electronic gateway and using digital signatures on those submissions.

So the -- the business process improvement. One of the very important areas is actually to do it from a standards-based process, to make sure that it's vendor agnostic and actually technology agnostic so that you can basically use the process across all vendors as well as across all technologies.

The space that SAFE is primarily involved in is the identity management stage and it's really about the ability to trust people's identities. It was easy when you worked within one building, you knew all the people on the staff, you could do voice recognition. Now when your staff is spread around the world, it becomes much more difficult to determine the true identity of that person. And then also from the perspective of when one person has multiple identities, it's hard to track all that information back to one particular person. So identities are integral to moving to a digital environment.

And then there is the whole concept of digitally signing documents. And it's actually -- potentially eliminates natural signatures within the infrastructure.

So an example of how that would really facilitate business processes you have some protocols that require up to five signatures. So you have -- in the old process that document would have to be shipped from one place to the next to get the signatures or in some cases if the facilities are close enough you have your researcher literally, who is taking this document, taking it to the next person on the list, and getting them to sign and literally wasting their time running the document from one researcher to the next in order to get the appropriate signatures.

In a digitally signed environment, basically you have a workflow so that the first person finds it, the next person gets an e-mail, they sign it, so it can be what was taking weeks to do in the past system can be done literally minutes.

And the whole concept of eliminating multiuser IDs and passwords is basically a cost savings.

So the advantage to doing the business process improvement is you end up with a faster, more secure, more productive environment that is also more accurate, and you can then have delivery of medicines in a faster state.

So this is diagrams of the existing trust model. This is the existing trust model and this is what some people have pointed out is that the right looks more complex than the left. Well, from a pictorial perspective that's true. However, if you look at the amount of work that's being done -- so if you have three companies that want to set up a collaboration, you basically have to set up three different arenas. So you would have this agreement, this agreement, and then this agreement.

Whereas in the next -- in the trust model, they basically set up a trust with SAFE and then each company would only set up that one trust with the SAFE environment and then you can have trust across all of these organizations with just that one agreement.

And to be honest, to be forthright here, basically you can do that essentially with the federal bridge. So SAFE is doing essentially the same thing with the federal bridge. We bought into the concept that, you know, the federal bridge has a solid standard, they have established what, a basic assurance identity is, what a media assurance identity is and what a high assurance identity is. So SAFE is meeting those levels of identity standards.

The other advantage to this model is that the FDA could then, with the digital signatures, automatically validate the validity of the signature so that they can easily prove that any document that comes to them is in fact a valid signature, the document hasn't been modified, and so it provides that level of validation.

And actually that, that level of validation, when you compare it to a wet signature, is much more quickly done, because with a wet signature you, you know, when you have a signature of Dr. Bob, it's very difficult to actually verify the wet signature, whereas in this model you can do that in a matter of seconds.

Now here is the specifics on the electronic submissions. So basically we are recommending that within the CFR part 11, which actually require the implementation of digital signatures, this -- I'm sure some people will stand up and go what are you saying, but actually, if you look at it from the perspective of the FDA has moved to a risk-based model, this makes perfect sense. So we are not suggesting that you move every signature that you have into a digital signature. Rather you do your risk analysis and where a wet signature today is required, that a digital signature would be required in its place.

The other thing is with whatever standard that the FDA would adopt, that it be compliant with the whole federal common policy and the Federal Bridge Certification Authority, basically the Federal Government has implemented that. And in fact, the FDA has started with that infrastructure.

The next is to require identity assurance at the medium assurance level and then to use multifactorial authentication. We think everyone realizes that the password or the single password authentication is relatively easy to compromise a multifactor identification and could carry a lot more weight.

Next we have the incentives component of this talk. We would recommend that the FDA actually provide incentives for us to do electronic environment. I hope I'm on the next slide. Sorry.

So in this one we have the FDA should interoperate with the -- it should be interoperable with standards that use a proliferation of digital signatures and identities, so it's usually acceptable, but also interoperable.

We have a high -- we would recommend that you use a high trust assurance model. Basically ensuring that you have non-refeud agents, security, record integrity, legal enforceability to the extent that it is possible. And then a major advantage for the FDA is that it's structured in a standard audit methodology around this whole identity management space.

So now moving onto incentives. The way to, I guess, in general as well as the way to get people to do things is to provide them with incentives for doing that. The natural thing to do would be all things being equal, a paper submission comes in and electronic submission comes in, the electronic submission would be given priority. The next incentive would be to actually be -- for an electronic submission to take less time to process than a paper submission.

And then the other recommendation would be to actually set a specific date in which paper submissions would no longer be acceptable, where all submission would have to be done electronically and that would have to be, some length out given. I know a previous speaker had said two years for implementation. But for something like that it would have to, of course, be substantially longer than two years.

And so to basically facilitate the moving to all electronic submission. And actually the irony in this space is this actually works to everyone's advantage.

I know a lot of -- change is difficult and sometimes people need to be nudged a little bit. But from the PhRMA perspective, an all electronic submission is much more cost effective and a lot of it is simply moving the existing business processes into a more streamlined electronic process.

And such incentives would provide a clear time line for electronic implementation and then it would also send a specific clear message that this is the priority of the agency. And it would also provide for an orderly transition.

So in summary, we have the business process improvements, the move to the eSubmissions, you know, it's reduced costs, increased productivity, reduces cycle times and that's generally true with most business processing. Otherwise it's not really an improvement.

And the whole concept of having a global regulatory standard rather than one done at the company level or at the state level or the EU level would be excellent if we could get that type of standardization around the world as well as development of a trusted infrastructure or rather the taking advantage of the trust in the infrastructure that already exists.

So I think the basic concept is to use standards and to develop that would facilitate adoption. So thank you very much. I appreciate that you held this public info section and thanks for giving me the time.

MS. BURNETTE: Anyone have any questions? All right. Thank you very much.

Okay. So we will break for lunch. Gosh. You really get a long lunch. We will start back right at 1 o'clock. So be here at 1 o'clock and see you then.

- - -
(Recessed at 11:22 a.m.)
(Reconvened at 12:58 p.m.)
- - -

MS. BURNETTE: Let's get started. Our first speaker is Ed Helton from CDISC.

MR. HELTON: Good afternoon. I know this is a tough time to talk, once you all have had the post-cranial problem right now. I'm an experimental endocrinologist so we will be experimenting. That means for like 25 years I worked at the bench doing basic research on the endocrine system. I worked on that a lot. But being a certified chemist as well I quickly got into drug safety and after finishing an NIH fellowship I worked for WHO on the tumor genesis of the early -- and then from there I went to work for FDA. And FDA called me a metabolite chaser.

They brought me in and said this is a synthetic estrogen. It's been taken by moms and has caused cancer in the female offspring. And it was more like Ed, work out the mechanism. Look, we don't want to do this again.

Why am I saying this? I think the reason I'm doing it is the reason I'm with CDISC is it's a standards development organization. But more importantly, we don't have still yet a really good institutional memory or a data repository with standard data. I think you made an interesting comment about landfill. And there is so much information out there.

Like I think it was 1976 I wrote the Sentinel review on the toxicity of the oral contraceptive. I think I had 121 references. Probably today if I went on the Internet I would probably have 10,000. So I'm just trying to say that standards and standard data to have an institutional memory, and I mean a human institutional memory is very, very important.

The first question. What would help improve the quality of electronic submissions to the agency? Standard data collection. We are working on that with FDA and under the clinical path. Obviously, standards for content and terminology. A common standard for data transport of electronic submission to the agency. And the adherence to submission standards such as SDTM & ADAM. It's worth noting that the quality of the electronic data submitted relies on the quality of the electronic data collected. We are working hard on that with health care and PhRMA.

Next slide please. Next question. What would help increase the quantity of electronic submissions to the agency?

We hear many people say this, and it's simply a ruling or regulation from FDA that specifies the standards to be used for the electronic submission to the agency. We know there is rule making going on right now. Part of my duties with CDISC is we have worked for two and a half years on the business case for the application, the adoption, the implementation of data standards.

Believe it or not a lot of people are saying we are waiting for the regulation. We just don't think there is a strong enough statement yet to make the investment to do this even though it's in the eCTD, it's in the data specifications, in the study data specifications. I'm just saying I think when that rule making takes place, it will be a surge.

A standardized e-portal for eSubmission. We are always looking at doing the data models at CDISC, work on, develop or what have you, would it be compatible with an eSubmission? Would it work? Would we somehow in the transport or applying part 11 functions and requirement, would our standards maybe in some way do something that we didn't anticipate? Maybe it wouldn't do the correct version control or mess up the audit trail. We don't think that's happening. We are certain it isn't. But we do look at that. So it's an important concept. And clearly defined, an easy to use process for making electronic submissions.

I think the eCTD -- we think that was great document. But we also know that we, as a standards data organization, a consensus data standards organization, need to be looking at that and make sure that we are doing our part to support the submission.

The next question. How would you prioritize these quality and quantity improvements?

We say it seems to make best sense to work on the quality of the submissions prior to enable improvements and streamlining of the processes before going after increased quantity of submissions. To improve quality, continue to encourage the development of data acquisition standards. And to improve quantity, mandate eSubmission standards. That's the only way to do it. This is the way we are looking for it. We want it to always be this way and it's going to facilitate our Janus, our institutional memory, it's going to facilitate review, interaction with the sponsor, et cetera.

Question 4, what data standards are needed to implement these improvements? We are biased but we think the entire set of CDISC data standards, we are starting the second pilot with the FDA. And it's coming along very quickly. But again, having a standard protocol with standard elements that not only the sponsor could search, but you could search. If you said I want to go and look at all pivotal trials on the chronic treatment of Alzheimer's, you can do that out of your institutional memory.

The HL7 standards for health care. We think those are really important. And we really work a lot with HL7. Now I'm going to give you the wink, as I said. Sitting in this audience is quite a number of the people who sit on the HL7 RCRIM, the regulated clinical research information management technical committee. As a matter of fact, I'm the co-chair and that guy sitting there from Abbott, he's a co-chair, and the third co-chair, Randy Levin. It's what I consider to be a true interface of FDA, PhRMA, CDISC and other SDOs. The MEA shows up. SEND shows up. But we have a good number in this audience right now of the archon membership that is a very proactive, not always calm, you know, but it's a great forum.

We really say how are we going to do this? And sometimes people don't agree. And we have three full-week meetings a year to do this. And we are really pleased that FDA is there. Like I say, there is sometimes disagreement, but that's okay. It's an open consensus forum under HL7 with a lot of the players there.

Obviously the interface between CDISC standards and HL7 standards, including common concepts and terminology. And in effect, standards for the data content and data format are needed.

How should FDA engage stakeholders while developing, testing and implementing these solutions? I think CDISC was very happy with the work with FDA and we think that FDA should continue to engaging its positive manner with standards, development organizations and work with those organizations for the build and transport of eSubmissions and FDA can continue to engage stakeholders by running collaborative pilots to explore the mechanisms.

We have four pilots right now going on with FDA under ODM. How do you move EDC data standardly. We have a send pilot we have the new integrated safety data pilot. And although my colleagues in the audience denies under the RCRIM where we are working on the content, CDISC content to message the HL7 project, I almost consider that a pilot. And there is others.

What I'm saying is this continuation with HL7 and CDISC I think is a sentinel event. And in my 36 years of doing this most of this has only started in the last 10 years and I just think we need to keep going at all costs.

What topics are the most useful to include in IT plans? The interoperability of software, we hear that a lot.

How do you make solutions, tools, platforms be interoperable? Harmonization of intended standards. Particularly globally. EU and Asia, they are looking very hard. And the JPMA announced very recently that they were adopting CDISC. So that's good. So we are starting to get globalization of standards for good IT plans.

Technology systems obviously need to be compatible. They need to interface. I have a bias and we look a lot at CDISC and I think it's just that the open system requirements under part 11 are not bad requirements. To me I think they are quite robust. Guidance and time lines for the implementation of IT plans and the overall process of implementation, training, ongoing support for eSubmissions. And, of course, the regulatory mandate.

Training and implementation are so important. We work, as this lady knows, with the Office of Business Process Support, and we are learning that when we change a version of one of the CDISC standards like SPTM, we need to talk about it and make sure that are OBPS is ready to go on that and help. And we are doing a lot of training right now at the FDA. Frank is head of training. He is our COO as well at CDISC. And that training is being well received now.

The next question. What lead time is needed for stakeholders to respond to and be in alignment with FDA initiatives? I think it's fair to say that many stakeholders are already underway. Possibly there should be a two to three year transition period. I think some people are saying probably two years, when it's really in the realm of, we wouldn't have any choice in the matter. Although a lot of people are going in that direction.

And perhaps a three to five year lead time will be needed to transition to a new platform. We hear about the FDA having an ePlatform and we would think it would be wonderful if eventually your sponsor's ePlatform could somehow interface with your ePlatform and that there could be a two-way communication, real-time, continuous marketing application, rolling reviews, et cetera. So we are in great support of that.

Number 8. How should FDA coordinate with stakeholders on the adoption and implementation of the data standards? The FDA could engage stockholders by running collaborative pilots like is already described. The FDA can also make use of public hearings and discussions which you are doing today, which we think are really great and vitally important for FDA to have a clear, open communication with stakeholders in addition to well-trained personnel on staff who can operate against well-defined processes with the appropriate tools.

We think it's a great idea. We know it's getting better all the time and we want to support you and provide any training, any implementation. But we do think that we are all in this and we want to provide our support as you may need it or as it's of value to you. And -- well, it seems to be going rather well.

Question 9. What data standard areas provide the greatest challenge? Right now I think at CDISC, we think one of the greatest is the accurate transaction and transmission of health care data. Both public and private for examining and supporting therapy product safety and assessment.

Last spring I spoke at a similar hearing and the idea of having a public/private network of institutional data that would all be accessible and we can develop new algorithms of data mining and we can develop system data, decision support systems, we could look at field types, we could do things we have never done before, but, again, I think CDISC sees that being able to have that common data there is very, very important.

I think the fact that NIH has started the grant program, the clinical and data sciences grant program. They are telling health care like UNC, Duke, Wake Forest, Maryland, wherever, up and down the east coast or all over the United States. They are saying, and I'm sure many of you are aware of this, but from your bench to your bedside we want best practices. We want standard data collection. We want you to be able to communicate between the CTSA sites around the United States. There is going to be 65 of them.

And to me, establishing that will be a great opportunity for the agency as well if we have made health care and PhRMA link such that you have excellent data extraction of both efficacy and safety data.

And if pursued, the transition to an XML transport format with CDISC content will provide a challenge to stakeholders. This is one of the measured products that we have going in marketing right now. It's not going to be easy. As it says in that lower sentence, we can do it right now with SDTM. We are getting there with that.

The other side of it is probably we will move to XML and I was talking to Bill this morning and I think, I was sort of saying maybe we ought to get started on this and I think everybody in archon is going to take this on. It may take some time and we may take some time make it work. But with time I do think that the industry and standards data organization will probably eventually evolve a better trance format that would meet the needs of your ePlatform and your goals as well.

We hear, for example, like SDTM plus is causing some problems. When we hear that we go tell us what's wrong with it? Tell us. We will work with you. We have a very good SDS team. About 45 people out of the industry that will work with you making it tighter for the submission.

And if it doesn't work well in the Janus, don't just tell it doesn't, tell us how it doesn't and we will work with you. I think there is a lot of interest on the part of the industry to make these things work because we can better understand what we need to do to have a better working relationship with the agency.

Global terminology harmonization with health care terminology. Clear reaction from the FDA will provide stockholders with the confidence to proactively take steps in the adoption of data standards and formats for electronic submissions. I think I've said that enough.

Question 10. What approaches will facilitate the most effective and efficient adoption of the implementation of data standards? An approach that emphasizes accessibility to and usage of standard technology that's easy to use. The willingness of both industry and the regulatory bodies to corroborate and put that technology into place. The standard workflow and processes for analysis of disease population data that is routinely and easily applied to standard data and that results in a decision to support the process.

These SDOs, as I've already said, represented consensus, and as I said here the CDISC, among others, can provide a unique supporting role for these efforts by providing an open, mutual non-profit environment. And the good part about it is when it's all done, you have a consensus.

After we have vetted the whole industry every way we can think of, when we adopt the standard and put it into maintenance mode and say, we think this is what you would want, let's do a pilot, you have a very good vetting process that we are willing to provide all the time.

What key areas require new or expanded electronic submission guidance? The following areas are those that CDISC believes acquire new or expanded guidance. Format and content of the transport architecture. I think, at least from our standpoint, we still would like to talk with you more and see how we can do it better and work with the industry and constituency to make it stronger, tighter, more conforming, yet meet the needs of the industry.

Clear and concise content of the observed, derived and analyzed data as required for the regulatory review process. We finished a big pilot between CDISC and FDA last year as part of the SDTM pilot. It built a standard submission using and demonstrating the interoperability of the standards and FDA provided 12 reviewers that looked at it and we are about to post a final report on it. And I think it will be of really good value and we have moved on to the second pilot. And one of the issues in that was data content and standard domains, required safety and efficiency data.

Number 4. An ePlatform interface. We are excited about the fact that that's probably out there and we want to make sure that the standards are configured correctly to support that. And I think I've already said the difference. So I won't say that again.

12. What lessons learned and best practices should FDA consider as we transition from program specific to enterprise IT solutions using a reusable and modular model? It must have backward compatibility to currently used systems so we don't lose things too quickly at least.

A secondary lesson learned is it's usable and accessible by all parties concerned. Foremost, the information, architecture and the technical architecture must be required by the regulation, and clearly demonstrated to ultimately enhance the efficient compilation of the submission as well as the regulatory review. Further, the enterprise architecture should work effectively on the Internet and should support public and private institutions and interface both open source technology with proper technology.

Like the new registry platform, we have provided our elements to them. We are working with them, because we think it's very important that standard data, standard data collection be in that international registry file.

Last question. What specific concerns, (i.e., security, confidentiality, etc.) exist for a third party entity or entity providing services related to the electronic submission and review and how can they be addressed? A key concern is that the third party must provide compliant and validated software and technology that can meet the requirements of the part 11 and would readily adapt to evolving requirements for the use of electronic data in the real-time assessment of safety and efficacy data.

We think about the two-way interaction and we want to make sure that what we provide will make that work smoothly and it's of value and that it has the information architecture, et cetera, that would really help.

And we have also said this would have to be addressed by a vendor neutral organization which can objectively assess part 11 compliance of the best application and interface of both open source and privately held technology. Again, we are willing to provide our support in that endeavor.

And I have had some interaction like with the open tool box concept. And people want, here at FDA, an open tool box with solutions and tools that can carry out standard analysis and basically what I said to the open tool box group is we just need to make things work together, be compatible, and have good forward thinking technology.

Additionally, the technology and processes of any third party would implement, should take into consideration the global nature of the clinical development and any particular regional country requirements that were appropriate. Again, we try very hard to summon the word to the limits of our resources and we think that should also be kept in mind. Thank you.

MS. BURNETTE: Anybody have any questions? No. Thank you very much. Okay.

Our next presenters are from Octagon Research. We have Nancy Smerkanich and David Evans.

MS. SMERKANICH: Just a little bit of background on myself and Dave. I was here in December for those who don't know me. I'm in charge of regulatory affairs at Octagon. I spent the better part of 24 years on compiling investigations of new drug applications and obviously have a stake in everything that's said today.

And I'll let David introduce himself.

MR. EVANS: I'm Dave Evans. I'm the CIO at Octagon. I have been involved with agency interaction at least for 23 years. We put in the first electronic submission in '85 and the agency has done a number of them since then. My oversight at Octagon deals with the implementation of the standard and consulting with organizations.

MS. SMERKANICH: And for purposes of full disclosure we always want to tell everybody that we are both a service and a software provider. So we are not going to talk about any of that today because in addition to our own views on the questions that were presented in the docket, we also want to give credit to the invisible presenters are who aren't here which are the people who responded to a survey that we sent out with selected questions from the docket and we will get to that later. We did that via e-mail blast and I think what's significant about that is that we were attempting to reach people who many of which did not even know about this meeting.

There is a huge section of PhRMA that is small companies. They may not even necessarily have regulatory expertise in-house and we certainly wanted to gain their concerns and their interests and when we do present the results of the survey, they are verbatim, non-edited. So we warned the FDA not to take anything personally. We will be including that in this presentation as well.

The agenda for today is basically we took the docket questions and kind of reordered them and grouped them according to topic, electronic submissions, data standards and we took a very literal interpretation of data standards specifically around CDISC & SDTM which today we will be addressing as well as took a stab at talking about IT plans. As I mentioned, we will distinguish which questions we answered internally versus those we put out in the survey.

The first question was docket question number 1 about the quality of the electronic submissions and you are probably going to think that myself and a bunch of the other presenters got together in a room and all made up our slides together. I think that's good for you because there is a clear take away message there. Our initial concern and I understand something that's being worked on this year is to make sure that the eCTD validation criteria are made public. That's very important. That will help us give you the technical pieces that you need to process.

In addition to that, since the guidance for electronic submissions have been around since '99, unfortunately we have never been able to document a lot about navigation and there is a lot of anecdotal information about what's a requirement versus what's nice to have. That's very hard as a regulatory person to point to. So any time that you can document -- and I'll going through some of the formats that we think would be a good way to document, those kind of navigational requirements, I think that would certainly help the industry. Because believe it or not people are still struggling with just basic things like what is PDF navigation, how many hyperlinks, are we giving you what you want. So that's certainly important.

And I think Allen Jones mentioned this morning how incredibly powerful feedback from the reviewers can be. It's great that it's done in the format of a DIE meeting, but for those folks who can't get out to meetings and get that kind of feedback from the reviewers, any kind of venue that you can use and again we have some ideas for the venues not from us necessarily, but from the stakeholders. Ways that people don't have to leave their desk and can still access this kind of information will be very important.

For the second question in terms of the quantity, so you really only have two choices. You can either mandate or give incentives, and it seems like legislatively incentives are going to be difficult thing. Reducing user fees after they just went up quite a bit is one way of advising people, but certainly if you can look at reducing review times.

People who have to sell electronic submissions to their senior management don't have a lot to hold on to. There is a small business waiver out there and that's great. There are a lot of companies that do not fall into that under 250 employees and yet this is the financial burden of doing electronic submissions substantial for them.

We think if you mandate, we think that mandating according to a schedule where there is certainly a group of applications that may be grandfathered is certainly one way to go. The way the physicians' labeling rule was ruled out I think was pretty effective. That's one model that we would like to see you look at.

The next docket question is prioritizing quality over quantity. I have never prioritize quantity first. Getting a ton of poorly done electronic submissions is going to be worse than getting them in paper. I would always choose quality. The issue around making quality submissions both on our side and your receiving side is going to be process etiquette.

Once you have a process for doing quality submissions, you can repeat it over and over again. I think that's going to be essential. As everybody has said today, for companies who are struggling with playing on a global playing field, you need to align everything that you are doing here with some of the global efforts that are going on elsewhere.

We are jumping ahead to question 11 because these are the ones that we attested to internally answer and I keep a running list on my white board of what I would like a guidance written on and then somebody wiped out my white board, but I still came back with the following list for expanded guidance. I mentioned the validation specifications. I think we've heard today there is a lot of people asking for that.

Specifically around eCTD there is still a lot to interpret out there. There is like 45 documents that you have to read and interpret. Things around granularity, life cycle management are really critical. There are specific documents that can be troublesome, annual reports. How do we do safety updates? How do we merge safety information? The quality overall summary for the chemistry piece. There are a lot of different ways that these are being done and we don't know which ones you like, you don't like, you find useful, you don't fine useful.

So if you could, again, document that either in the form of guidance, best practice, Q & A, anything that you can get into somebody's hands, I think that will go a long way to both addressing the quality and the quantity issues.

And then the other thing that we struggle with a lot is guidance documents have been around since '99 around the data piece of its submission. But it still is a huge man hour effort to get the data sets into compliant format. With the withdrawal of the '99 guidance we found that the data set throughout is pretty inadequate. Jim might have mentioned it this morning, program files and what do you want programs for? There is a lot of questions out there.

So we are looking to encompass that information and bring those guidances forward, something that you can point to and say this is why you need to do it this way.

MR. EVANS: So going back to question number 4, what data standards are needed to implement these improvements, we clearly feel that it is SDTM. That's the standard that's been worked on recently and clearly from '99, an organization that has come together industry-wide to find out what is the best way to have the nomenclature content and structure standards, the delivery of information to the agency.

Clearly to mandate this, having 20 years of experience delivering data to the agency, really the standard way is the way to go so that there is interoperability among the agency and the industry, and the industry itself, which I'll get to in a little bit.

One challenge is the legacy data conversion. Taking data from a non-standard and move forward will always be a challenge. I will talk about that. It is really though a transitional period. It is not a long term problem.

Next slide, please. The docket question number 7. What lead time is needed for stakeholders to respond to and be in alignment with the initiatives? We have taken some very clear examples of work that we have done with clients. Our feeling on late stage SDTM conversion, that is that point of getting data from whatever format it is in into then a submission ready format, that it takes anywhere from up to 8 weeks or maybe beyond for that very first study.

After that, converting studies, assuming that they are alike, we have a three-week period of time. After that you have to look at how do you build the tools in order to actually do that. There is not really any commercial tools available to do this, so we have to build them customly. If we want to integrate SDTM into the collection process what we're really looking at is how do you change the way you do data standards now within an organization if you are at all doing data standards.

If you are going to take an organization that has it, we feel that anywhere from 6 to 9 months is the optimal or optimistic view of that. Clearly it will take much longer if you have not adopted any data standards.

What I put before you is two charts that are out there. The first one, data flow is typically how we do clinical research today. That's typically how data flows through the pipes. What we are looking at in the bottom right-hand sider there is the legacy data conversion going into the agency.

We then look at the other chart which is essentially where standards are going to have to go, you'll see the impact, the complexity necessary to move those standards back through that life cycle all the way back. So there is not necessarily a simple process to go through this. The impact of getting the data to the agency also has impact further and further upstream and I'll refer to those charts.

MS. SMERKANICH: We didn't make a slide of this on purpose.

MR. EVANS: In order to do this on a slide it would be difficult in order to get the resolution. So I provided this to the team.

Next slide. So our feeling is what data standards areas provide the great challenges. I feel where we are now, the short term transient time is this legacy data conversion. Why? Because there is a failure of small and mid-size companies to adopt standards that have not had them in place. As you do one, two, three or more trials using different CROs or different ways of collecting the data it then becomes much more complex to bring that all back together.

Failure of CROs and clinical research organizations to adopt data standards so they have their own way of delivering data to organizations. Especially for the small and mid-size firms that are using CROs more and more. We had a lack of standardized CRFs. What about that collection vehicle that we're using at the site that collects the information? It changes study to study, company to company and even within departments within companies.

Ed talked about lack of control terminology. This is a theme. It's not just about the structure and the nomenclature, but it's also about the content of getting the terminology correct.

SDTM is relatively new. It's within the last three, four years. So the knowledge base within the industry is still relatively small. We talked about SDTM, everybody knows the acronym, but really what is the granular details associated with SDTM?

We've had the privilege to train about 1400 professionals in the last two and a half years over about 80 venues worldwide, but only that small number considering the total population of people that actually touched data during the clinical trial. So we see this as expanding especially if mandated. To a certain extent the vertical structure of SDTM, the delivery mechanism of getting that data to the agency does not bode well for it being used in operational context, as you'll see. Structures would have to be changed.

So therefore we remove the content and the nomenclature layer from the structure layer and that would be better.

Next question. What approaches facilitate the most effective and efficient adoption? Again as Nancy said, we feel there are really two choices here. Mandating its use and then also maybe some type of incentive for compliance with a data standard.

Next slide.

MS. SMERKANICH: To move on to how we did the survey, we called out the questions 5, 7, 8, 9 and 11, and we sent that out in the e-mail blast with a field for folks to provide additional comments and sent it out really very widely to everyone on our e-mail contact list.

As far as the next slide it shows -- we didn't get a ton of responses. The last time I was here I was able to show you 77 responses which I think was a more representative number. We did this on a relatively short order based on the timing of this meeting and the timing of the receiving the questions.

But we did get -- it was interesting who we didn't hear from. We heard about back from tier 1, big PhRMA, revenues of more than $1 billion and up and we heard back from twice as many people actually from tier 3, which is small PhRMA. And I find their comments probably the most valid to this presentation only because I think the people from the PhRMA group really did a great job of presenting their issues. So that's who we heard from.

Who we didn't hear from is any CROs which I think is unfortunate. As Dave mentioned there are a lot of companies that were extremely heavily on the clinical research organizations and if they are not all on the same page as the rest of us, companies who rely on them are going to be lagging behind as well. We also didn't hear from the academic environment. We know that a tremendous number of investigators INDs are out there probably contributing to that huge number that you have.

So we were a little disappointed that we didn't hear from those folks and you'll notice no tier 2 companies. Those tier 2 companies, what I figured out is that those are the people who are really kind of getting squeezed. They don't have the resources of a tier 1 PhRMA, they don't have the flexibility and maybe the workload even of a tier 3. And they are kind of caught in the middle. My guess is they probably either didn't even read the e-mail blast or didn't have time to answer the questions. Unfortunately, we don't have that kind of information to present to you.

If you go on to the next slide I know it's busy, but we took the questions that were in the docket and then we kind of supplemented them with some additional queries that we had specifically around the data standards piece because we were really trying to get at some, things about who does this kind of work and where do you see the benefits. These are the questions that we actually asked and we will run through those and give you the answers now.

To the first point, again, very consistent message with everything else that you are hearing about how to involve the stakeholders. Everyone really likes the idea of pilot submissions, especially tier 1 companies. The FDA Website is the go to place. Again, not only for tier 1s, but tier 3s as well. Participation of the FDA in meetings as well as industry is going to be huge as well as involving the industry groups. And we have already seen that's happening just by your presence today.

As far as the same question and the responses from the tier 3, I think they are crying out to say please don't forget about us. We want to be involved in testing as well. How you get to these folks? I'm not sure. I honestly don't know. Because they are looking to use communication vehicles that clearly indicate that they can't go anywhere. They want newsletters, online courses, electronic questionnaires, information available on the Website. Because, again, my feeling is, and this is just my gut, telling me that these people are so busy doing their work that they don't have time to read the Federal Register, never mind respond to the docket. So they need to be able to get information in some other way.

They like the idea of having a hotline with an FDA expert. I'm not sure how that would ever be implemented, but I thought it was an interesting idea. A lot of repetition of the comment that -- and I think it was mentioned earlier today as well, the HL7 SPL team model, the working group seemed to really work very well. There was a concern about making sure that the FDA had dedicated team members.

The use of focus groups that involved PhRMA and both tech and devices and making sure that cross-functional people were exposed to these standards. It's not just a regulatory thing or it's not just an IT thing. I think it's really important. And again, they are looking for the theme of FDA communicating with industry, gaining their perspective. This is a perfect example of that. Unfortunately, I don't represent all of PhRMA, but wanted to bring these tier 3 concerns to you and we got a lot of responses.

The next question in terms of lead time, again, looking at everything from as little as three months, which that's an extremely fast response time, to big PhRMA saying, yeah, this could take us a couple of years to the tier 3 companies obviously having a lot more trepidation and saying if I'm looking at redesigning, I could be looking at four to seven years.

Some other comments, companies not being ready for eCTD as the electronic submission format in January, looking to grandfather how legacy data and legacy studies are handled and I'm reminding everyone that all PhRMA are not billion dollar companies. And interestingly, because I think that fair number of people said I'm not going to do this until you make me do it. Somebody still thinks that changes should be voluntary. Well, changes have been voluntary since 1999 and I think we are done, we need to be done with that.

MR. EVANS: The next slide is how should FDA coordinate with stakeholders on the adoption and implementation of data standards. We see some of that coming back from both tier 1 and tier 3. Industry group meetings, articles and journals, notification to participate in meetings like this through guidances, clear guidances. Tier 3, simplify the standards. Some companies think they are too complex, that there should be a single point to come to the FDA for answers to questions they have. Other ones think that the current communication is working fine. Others want personal communication with the reviewer.

Next slide. What data standards areas provide the greatest challenges. We see in tier 1 it's the implementation and development of the vocabularies and taxonomies associated with this that are not defined within the SDTM model. So it's those custom domains and therapy domains that have to be worked out.

Tier 3, some say that ODM, which is a transfer vehicle, it's on your maps, it's essentially the way to go between different stages from EDC to data management systems. Some say it's impossible. We had examples of that where it is not impossible to do, that you have end-to-end solutions going all the way from trial design all the way through the submission using that format.

Others feel SDTM is cumbersome, but doable. Yes. When we are talking about clinical data, we are not talking about ten variables, we are talking about thousands of variables spread across numerous domains to correct the science of a clinical research. So it is going to be cumbersome, but it has to be put into some type of standard.

They feel some changes to the structure and standards should apply to work already completed. This interesting thing, adverse events and efficacy criteria are not defined in SDTM. I think what they are meaning is the derived fields, as far as the ADAM standard goes, how you do derived fields or summary variables, that's not clear in SDTM but it is clear as you look in the ADAM standard.

And more experienced FDA personnel with technical and regulatory experience. We've had the privilege of being able to train FDA reviewers. So the FDA is taking a leadership role on becoming more knowledgeable. Next slide.

MS. SMERKANICH: So with regard to expanded guidance, again, this is coming from the survey responders is they are looking for a final rule. They are looking for additional help with biologics and areas that may not be addressed within the specifications around that.

Additional detail on how study report data is provided, again, relating directly to the way some of the biologics and the therapies are developed and some of the additional components that they have. And as I mentioned previously, the SAS data submissions and SAS programs' submissions looking for more guidance on that area.

With regard to the tier 3 companies, they feel that there needs to be additional guidance around therapeutic areas that are not expanded fully, including oncology. I'm not sure I even understand what the piece about the FDA or EMEA guidance that would eventually need CDISC models. As I said, these are results of a survey.

More training for medical officers, reviewing data. Help unrolling the DMAs. I think someone mentioned earlier this morning, EDC and how that fits into the picture. Electronic processing and looking at additional guidance about AERS reporting. It's great that we have the AERS and the gateway for the eCTD, but making sure that those systems can interface with each other and again looking for some additional information about how to file electronic IND formats.

We are kind of shoe-horning a lot of the IND pieces into that and as we go through this and gain more experience with it there is some space there for some additional guidance.

MR. EVANS: We are going to break this up into two slides. It's talking about how do we do end-to-end? How do we go from beginning to the end of submission? A lot of companies, at least at this point in time that have to do this end of pipe or late stage conversion in order to get it to the agency. And the answers that came back is what are the factors determining the chosen approach, whether to do the whole process or late stage conversion. The tier 1s are saying we need to see what's happening from the agency standpoint.

We see that it's efficient and ease of use, it's a good business case process for it. But we have to revamp our entire process. And for large PhRMA, that's an enormous undertaking.

Next slide. Tier 3. Some of them -- so tier 3, what we are looking at is saying whether there are new studies that can actually meet into the standard. There is no true commercial systems that are available for these standards. Again, end of pipe. They are saying too late. They need to build it earlier. So tier 3 companies are seeing it doing it more and more upstream. And the CRO capabilities.

Next slide. So what we looked at in slide number 6 is which approach, what are the factors most important. The company's commitment to change I think is number one, is the executives actually behind the change. Do they buy into the need for the standards? For a tier 3 company it's the commercial availability of a certified system. If we have a standard, how can we ensure that it is a standard and it is certified?

And the last point, reconfiguring the information to the appropriate level. They feel that that is an important for them to get it together so that they can do the safety analysis.

Next slide. Again, how long will it take before the benefits are realized. Tier 1s are saying because they had the work force in order to do it, that we can do anywhere from four to eight hours a week over three months or eight weeks for SAS programmers, wherein the tier 3, they don't have the state level. They don't have the capital in order to spend to do this. Typically they are going to outsource it and it really depends on the system. We find it really depends on the individual not the system.

Next slide. And so in this case how long will benefits be realized. Some of them need understand the benefits before they can even judge on that. Really the consistency across studies is a big issue. They feel it's really truly a concern to have consistent information as it goes across the studies. They want shorter review by FDA and can bring the product to market earlier, obviously for commercial gain.

Tier 3, benefits will only be realized when the standard is not just a set of criteria, but a true mandated practice. Running clinical trials will be much smoother. You now agree with that. You are now going to have eCTD systems that are going to capture and process and move the data through. And again faster review time.

Additional comments. Nancy, you want to speak to this?

MS. SMERKANICH: Yeah. Tier 1 had none. Tier 3 companies, again, it's just looking for communication, I believe. Looking for communication and guidance and with a sensitivity to ICH and things that are done globally as well as looking at resources and being sensitive to the lack of resources that a lot of companies have.

Next slide. You know, I think that again, this is pretty much what everyone has been saying today. Communication is going to be important. A mandate is going to be important. A cross functional approach is going to be important. Again, those tier 3 companies want you to understand what they are pain points are.

MR. EVANS: In the final slides we just want to talk about IT plans. Most of this has been talked about by the people that have spoken prior to us. It's about maintaining enterprise focus, standard-based involvement, good IT practices that are in place.

Next slide. Leverage industry experience, business requirements, functional requirements, good practices, find a partner that has done this before.

Next slide. The whole idea of confidentiality, security, authentication, again good IT business practices are a necessity to even consider this for the agency.

Finally, the overall summary of this is that standards and formats are available today, they are in use today, they are in use in industry today. We feel they need to be mandated by the agency and it needs to be implemented in a feasible and cost sensitive time frame for this to actually get done.

Finally, training and communication directly from the agency would be critical to the success of the adoption of these standards industry-wide. Thank you.

MS. BURNETTE: Any questions?

MS. SMERKANICH: We are the only ones who used up almost our entire time. Sorry.

MS. BURNETTE: Thank you very much. And our final scheduled speaker is Mark Dente from GE Healthcare. We do have some time allotted at the end for general comments. If there are going to be people who want to have a brief comment, for the sake of moving it along, I would ask that -- when Mark gets closer to the end, if you would kind of come over to the side and we can start, motion for you to come up when you are done. Mark.

MR. DENTE: Thank you. First of all, I think that most of the comments, in fact, much of the comments were scheduled, and also it's at the end of the day and Friday afternoon. The folks that are here should be commended.

I'm Mark Dente, vice president for GE Healthcare. We are part of a larger health care entity, about a $17 billion company. We work with the FDA on many events and issues. We understand the importance of safety, the need for safety and patient transparency of process and how that all takes on.

We also work sparingly. We have the luxury of some of the other larger companies to see what is going on at international levels. In the UK our medical diagnostics global vigilance function uses a database application to track adverse events. We have been doing this in systems that use processes, we can provide monitoring, provide reporting and we can do this in a timely manner.

What's interesting and what is important about some of the global initiatives is that I was going on and we feel that we want to make sure that that voice is also heard. That clearly half of all trials are not done in this country anymore. There is a lot of work done internally and we have to make sure we have leveraged all that research and make sure that is provided in a standardized format.

Ed talked about it from CDISC, the electronic records. We can go through some of the topics I'm going to talk about today, but I have been doing bioformat text for about 17 years. It's a bias of mine, understanding the value of electronic health records. How do we allow for clinical decision support and education for patients and physicians? How do we as clinicians deal with that growth of knowledge. It's by allowing technology to aid us.

A dumb computer is never going to practice medicine. What it can do is say wait a minute, can I automate that adverse event reporting process and knowledge transfer? I'm going to go through some of those issues as we go forward. Can we go to the next slide? Again, this just to kind of let you know where we are coming from.

Information technology is a cornerstone of where we think medicine is going. We actually have what we call our informed, predict, diagnose, treat, monitor program, where it states that we are in all these different businesses. It's really information technology that allows us to internally do this. Not to say we have all the answers, because we are promoting CDISC and all of the other interoperability standards.

We learned early on that we had to standardize nomenclature, we had to standardize terms. There were technical people in the room that can talk about underlying things of what has to happen. I will tell you as a clinician one of the largest challenges out there is when you think about data submissions. I call hypertension, hypertension. You call it HTM.

How does a computer know what that means? Those are the issues that we are trying to solve internally and we think that if we can get guidance and bring that together and also recognize that there are other people out there, other parts of the government, other agencies that are also working on standards, there are multiple programs out there. We are asking for and looking for coordination across not only what the FDA is doing, but also across government agencies.

We are at a threshold that allows us this opportunity, why? Because we still have relatively low adoption of the game changer here and that's electronic health records. So while we are talking about data submission here, the value and the ability to expedite the process of trials more safely with greater patient safety, with the ability to have more transparency in the process really has almost a by-product of what other parts our health care industry are doing. Right now about 40 percent of all health centers have already implemented electronic health records. There are national standards for certification. The last presenter talked about that.

What makes someone take away a barrier to purchase some of these solutions? They are expensive. One of the barriers is whether you buy from a GE or agency or a small company, it's certified. Those standards are already out there. Let's leverage those standards as we go forward.

Finally, I want to say that GE has a vested interest in this just from our own employees. We have about 340,000 employees about a million -- We are looking for the most effective care. We are looking for the ability to bring new therapeutics to market faster and most important safely and we want to figure out how to leverage those and train our own clinicians that treat our families and loved ones as well.

Next slide, please. So what's the role of the electronic health records? It's a comprehensive longitudinal patient record. Other speakers spoke about data solutions. Here I'm going take us up to a 30,000-foot level and really think about transforming what we are doing today to a new process. Let's not come up with an electronic bugaboo. We have opportunity now, whether we want to use our very big strong components of that system that works for us internally. But those type processes and the root cause analysis and think about process redesign and that's very important, it can be used in thinking about work tools. This is really all about workflow.

What we talked about is standards for -- what we really talked about is how do we stop -- how do we stem the flow that 60 percent of all principal investigators only do one clinical trial? What's the cost to society, the buried ebb to their practice? Now it encumbers them so much that they choose not to do a second. So it's about using clinical decision support, it's about utilizing evidence based medicine and model electronically.

But the underlying all of those technologies requires standardization of vocabularies, standardization of terminology, harmonization of those different standards not only within the FDA but also what the NIH is doing, what biosurveillance is doing.

We at GE are being asked literally in the last eight days, it seems things run in groups. In the last eight days I have had no less than 70 leads from four different agencies of government trying to understand how we are dealing with these issues as we develop applications. I'll tell you it's expensive. It's a burden to our development staff. Every time there is another standard out there we have to be able to develop mapping.

There are times when the standards out there aren't evolved enough so it's nice to say let's use SNOMED, but they are not sophisticated enough. So in fact, we actually need greater standards and greater granularity in terminology of vocabularies. Other standards entities out there are doing great work. Let's leverage those.

As we think about this, let's think about the increased adoption of the EHRs. This is one of the State of the Union Address a number of years ago. Electronic health record for every patient in ten years or so. One of the four criteria was, the benefit was that we are going to aggregate data for research. And we will be able to have those standards for that interoperability. So whether it's coming from our system, coming from another system to our system, wouldn't it be great to have a clinical trial.

Now I can actually have real-time data monitoring, not near real-time, real-time data monitoring. We have a prototype. We have about 7.8 million lives in the database that many of the -- they are using post-market standards, using it for inclusion, exclusion to develop. The only reason we can do this and there is another great organization -- the UK can do this well.

Seven years ago before a lot of the other standards were adopted we said there was enough of a demand -- enough of a demand to understand the information and to treat our patients. Really we weren't thinking about PhRMA. We were thinking about --

We actually standardized data across multiple instances of the install regardless of business concerns. It allows us now to have this wonderful resource. Our next challenge is saying how do we use CDISC standards and become a mode of data for a virtual trial?

Those are the opportunities that are out there and there are people outside of the PhRMA world more in the IT health care delivery side and we are doing this already. And there are a whole bunch of teams of individuals that have worked on these standards as well. That's leverage that work. I'm not quite sure how that occurs given my last ten days.

CMS has standards they want us to work on. There will be standards for NIH, there is all these different standards. What we really need is maybe the Office of the National Coordinator that takes on this challenge. So we are not here to provide the solutions, but I am here to say that we want to participate in the process and are willing to do that.

I'm going to end on this one slide on one last thing. I want to ensure that we do not compromise workflow. We are absolutely looking at it from how do we think more broadly about adverse event reporting, more broadly about surveillance monitoring, especially with electronic health care records, you are really impacting clinical workloads. So the other thing we can do is make any rule or data requirement so restrictive that we impinge on that 8 to 12 minutes that the patient gets in front of their physician.

So we are always balancing. Saying can you get this other piece of data. Hey, you have this really interesting technology. Sure we can. The reality is that every time we put in an additional data requirement we know we will lose some percentage of physicians that will say it's too restrictive, I just want to get paid.

So we are always balancing what we can think about as value for data and where we need to be judicious in how we think about what data we truly need versus what data we would like to have and I would like those type issues to be thought of as you work through your processes as well.

Next slide, please. When we think about data aggregation, clearly the primary use from an EHC perspective is to take care of that patient. They have a concern and they want to be taken care of. That's what we are here to do.

The other issue though is the secondary aggregation of that data for research is the value. That's the payback of why clinicians gave up their lovely paper and don't get to dictate and sometimes come running very excited and others come completely screaming. Well, it's because of secondary use of that clinical data. We can go to CMS and do paper performance reporting and help them understand what percentage of our patients have an A1C greater than 10 or what part of our patients are doing a great job of A1C less than 7. That same data can be used by this industry to say wait a minute, if I have an inclusion/exclusion criteria, what is it, the likelihood of capturing a number of patients for the protocol if I change the A1C value from 7 to 8? What is the effect there?

By the way, that takes all of 30 seconds to do. I can do from a potential of 700 patients to a potential of 850 patients. I don't know what the number is. It's the secondary use of that clinical number. Only thirteen can be done though through data standards and controlled medical vocabularies. As you think about some of the other transactions, processes of data transmissions, let's not lose site of the actual need for terminology consistency across the industry.

If we think about identification of patients for clinical trials, we touched on this already briefly, but if you really think about the use of the clinical decision support rules engine, does it really care from a rules engine perspective, gee, again do we -- how many of our patients are discharged post MI that are put on an ace inhibitor?

By the way, other than a very small subset of the population, anyone that you know that's had an MI that's been discharged, the 0 plus percent should also be an ace inhibitor. Dr. Glenn published a report a few years ago in the New England Journal of Medicine that showed that had occurred about 54 percent of the time -- under 50 percent of the time that occurred. How come we can't track that? Isn't that information available as we think about patient outcome and safety?

Again, if you use the same technology you can also say let's think about how we are going to go to look for PhRMA surveillance and think about alternatives to the med watch form.

It's been years. I haven't touched a patient in years. A computer tech physician. Back many, many years ago it would have to be so egregious that I would stop what you are doing to find the form to go, if you even had it, okay, fill it out, and deal with it. The reality is that we all know the data comes in fragmented at best, self-reported and under-reported.

In the UK we have systems over there that automate their yellow sheet which is analogous to our med watch form. The same clinical support rules engine can help. Doctor, you -- they can say if this was an adverse effect, what are the lab values that you find valuable to have with that report.

So think about standards, but also think about the implementation of new technologies that will change the game of how drug delivery and drug development is done. Because it's coming, it's the freight train, and the good news is that it's actually happening.

Maybe unlike a decade ago where I had a lot more scars on my back. The good news is that it is happening.

Next slide, please. Interoperability, standards harmonization, momentum. The reality is that there is several initiatives that are underway in the U.S. Department of Health and Human Services that are looking at this.

There are road maps out there that we need to start to think about. The standard could be leveraged to support surveillance and automate -- the private sector has engaged in processes, CDISC, in collaboration with IAG, and --

The same system, if you can leverage the standards can be -- we are not saying that, we are not saying that an EHR should want to take on the burden also. We also want to consider the burdens of CFR part 11. What would be considered burdensome? Something as simple as we have to have unbelievable amount of up time. If your medical record is now electronic.

If Microsoft puts out a patch to go through a validation process, that delay really incumbers our ability to take on some of the more, and leverage some of the added technologies that EHR will be able to bring. If you don't want to do that -- there are multiple binders sitting over there. It is not difficult at all. In fact, it's done already that an ERH could open up through a window, have standards of how the past medical data and single sign on is. So now the clinician doesn't have to it all to even populate out the appropriate the ECD form. Close that, submit that and it handles all that.

Again, you streamlined the workflow. You helped the office staff, you helped the physician staff and what we are trying to do is to see how we can make this process more efficient.

Next slide, please. My format is coming up differently on my computer, so I apologize for that.

Personalized medicine. The three revolutions that we see in medicine, the first was Doctor, wash your hands. Like, a hundred years ago. The next revolution. Hey, we have this thing called an antibiotic in the late 40s and 50s. Now what we have is the genome. The FDA put out the release eight weeks ago. Outstanding. What it did was it was a wake-up call. So why we all understand the value, we have -- the part of the challenge of this organization is to educate the hundreds of thousands of physicians. Frankly, like myself. I'm not a scientist.

So how do we provide education, how do we provide that knowledge transfer back to the ones who are actually doing the front line work? One is through standards. The same data standards that would allow us to submit for adverse events will also allow us to fire a rules engine.

Again, patient safety, transparency of care is what we are all looking at. To be very honest, we have to recognize that there is a tension between privacy and an individual patient's need for that privacy. We believe that while we absolutely value and understand the need for patient privacy, the ability to aggregate clinical data for clinical research is paramount to further development of health care solutions and therapy in the country.

And we want to make sure that while we are all thinking about the specific standards, on a national level we also have our voice heard throughout FDA and other regulatory agencies that the importance of contributing anonymized data and aggregating anonymized data of clinical research.

Next slide, please. Here is the one that I got to read because there is a bunch of things that I just don't really remember all the different names of. One, the EHR road map provides a model for the FDA to benchmark.

Harmonization of new standards does exist and there are working groups already there. How do we also look at CMS? How do we look at biosurveillance? How do we go across the government and think about standards and the importance of those standards. We want to establish milestones for profits. We think what gets measured gets done and if you don't measure it, it won't get done. But make those milestones realistic.

It's one thing for an organization like mine to have a bit more resources. Let's be sensitive to the fact that we want to be as inclusive as possible going forward. In fact, the whole purpose of interoperability standards is to allow a wired choice for our customers, for our PhRMAs for all the folks that actually put the money into the system to develop the new therapies and bring them to market.

Next slide, please. In summary, I think that when we think about transformation across all the phases, what we are talking about is a safety issue, it is about education, it is about understanding the leverage transformational technologies that are already being adopted. Again, 40 percent of the health centers have online or in their pipe electronic health records. We need help in trying to think about virtual clinical trials.

In fact, the downside of this is that -- maybe this would be an incentive to help adopt EHR in our country. Now a physician that is in a more rural community, they will want to participate in the trial or a marketing surveillance study and it might be that way. Using this technology we have driven down a cost of delivery and the cost the process of the drug discovery process.

If we can help recreate patients, whether it's through our system or multiple systems and we can change four to six months after recruitment time of the trial, we have to find patients. If we can find them in a month and a half versus seven months, isn't there the potential for -- society all wins. I will say that --

Next slide, please. We are about to be submitting follow-up comments within the 15-day period. It will be more extensive than what we are doing today. I want to thank you all for allowing us to present our thoughts and processes and if there are any questions I would be happy to take them now.

MS. BURNETTE: Any questions? Thank you. That does conclude the scheduled presentations we have. Do we have anybody that would like to make a -- I believe we have one, for a few minutes, make a few comments; correct?

AUDIENCE PARTICIPANT: My name is Mel Brevan, and I'm president of Public Health Resources. I'm glad our last speaker covered many of the areas he covered because it goes into some of the points I think are very important for a lot of perspectives. Also, I have to let you know I retired from the FDA in 2001, and I'm delighted to see these activities increasing in scope.

It's important that the FDA is growing systems across centers and, in particular, as many of the issues that were discussed focus on drugs and biologics. Clearly, a commitment was made to go beyond that and involve the other centers, too.

Basically, as the IOM report talked about, looking at products across the life cycle of the product is increasingly important. For many years we focused on electronic submission systems when the key issue is post-market surveillance. For many years the drug side and the device side of reporting were incompatible. You couldn't bring the two together and deal with the results across the board and, hopefully, this will be an issue dealt with, too, in the future as we get more compatible systems, and again dealing with that across the life cycle.

I think basically the bottom line is that the, as many people feel about it, and I have had discussions over the years, and one thought I have in mind is how we can have some of these issues come up at annual meetings and deal with some of these issues. And FDA has been very active. Management meetings and we had a fascinating one on imaging issues. But really coming across the broader perspectives of bringing people together at the annual meeting would be very useful. The other hat I wear is I try to chair one of the tracks for the annual meeting.

Basically, I think certainly the FDA has a very good track record of working with a variety of organizations across a variety of lines. I think we need to also look at the lines across traditional drug boundaries. As an example, just looking at medical imaging. Medical imaging devices are largely regulated by CRH. They are used in drug development and to evaluate the effectiveness and safety of drugs. Clearly, there is a really interest to bring those two segments together. It's very important for the drug side of the house to know what went into the decisions for the device side of the house and to make sure those things work together effectively.

Clearly, the agency has been involved with national and international standards organizations for many years. I'll give you a little story. Back in the early '80's I was, I grew up in the public service hospital system. Unfortunately, when I did my second residency I learned about what the FDA was doing in radiology and it reactivated my career.

Another thing was basically in terms of the international activities. The track record is a good one. Back in the late '80's the head of the European Commission made sure that the head of the American Standards Institute was contacted because the EC was supporting standards activities at that time, and basically wanted to coordinate with the U.S., have the U.S. working with it.

We wrote the letter and asked basically that we work together internationally in those areas. We couldn't do it all. Clearly a lot of progress has been made. So I look forward to this happening with the things we are talking about now.

Clearly, Janet Woodcock, we talked about all the activities that were taking place. Clearly HDSA was talked about also. Clearly for a long time, and increasingly, requiring the results of its research be made available to the public. One thing that's important is for the FDA and other agencies to work together to see how they can share data, and it was also brought up by one of the earlier speakers, so that FDA data could be made available to them and other data could be used by FDA in a positive way.

Basically, the bottom line is that I would like to see the FDA continue its collaboration with other organizations. It would be wonderful if we could take advantage of the DOD healthcare system for data and availability to test systems and FDA could expand to other venues. The bottom line is again we needed to coordinate it more effectively, again as was implied by the previous speaker, at least I feel, with the office of the national coordinator, and bring together the various kinds of activities so it does become a joint public/private sector collaborative effort even better than it is now.

MS. BURNETTE: Thank you very much. Do you have any questions? Do we have any other general comments? Okay.

If not, I do really want to thank you all for coming. It really is helpful to us as we are trying to write the plan to hear your input and even if we have heard some of it before, it's good to hear it a number of times. We do appreciate you coming out here. I also want to thank the panel for sitting here and listening and taking good notes. And with that, remember you have 15 days if you want to submit any comments. The docket page is on the agenda here. So thank you.

(Whereupon, at 2:32 p.m. the meeting adjourned.)

UNITED STATES OF AMERICA )
ss:
DISTRICT OF COLUMBIA )
I, ROBERT M. JAKUPCIAK, an RPR and Notary Public within and for the District of Columbia do hereby certify:
That the within is a true and accurate transcript of the meeting of the FDA on October 19th, 2007.
I further certify that I am not related to any of these parties to this action by blood or marriage and that I am in no way interested in the outcome of this matter.
IN WITNESS WHEREOF, I have hereunto set my hand this 31ST day of October, 2007.

________________________

My Commission Expires:
December 14, 2008