American Enterprise Institute

May 21, 2008

[Edited transcript from audio tapes]

8:30 a.m.	Registration and Breakfast
8:45		Panel I: Off-Label Prescribing, Marketing, and Medical Practice
	Presenters:	John E. Calfee, AEI
		Scott Gottlieb, M.D., AEI
	Discussants:	Len Lichtenfeld, American Cancer Society
	Moderator:	Cole Werble, RPM Report
10:30		Panel II: The Legal Environment from Federal Regulation and Enforcement
	Presenters:	John Osborn, University of Oxford
		George Terwilliger, White & Case
		Daniel Troy, Sidley & Austin
	Discussant:	Jeffrey S. Bucholtz, Department of Justice
	Moderator:	Theodore H. Frank, AEI
12:30p.m.	Luncheon
1:15		Panel III: Distortions from State and Private Enforcement
	Presenters:	Brian Anderson, O’Melveny & Myers
		Theodore H. Frank, AEI
		Kyle Sampson, Hunton & Williams
	Discussant:	Michael S. Greve, AEI
	Moderator:	Michael A. Krauss, George Mason University Law School
3:15		Panel IV: Legal Implications for Commercial Speech and Medical Practice
	Presenters:	James Beck, Dechert
		Mark Herrmann, Jones Day
	Discussants:	Margaret Johns, UC Davis School of Law
		Richard Samp, Washington Legal Foundation
	Moderator:	John E. Calfee, AEI
5:00	Adjournment

Proceedings:

Panel I:  Off-Label Prescribing, Marketing, and Medical Practice

John E. Calfee:  Good morning.  I am Jack Calfee.  I’m an economist here at the American Enterprise Institute.  I would like to welcome everyone to this conference on off-label issues.  Credit for organizing it goes mainly to Ted Frank and his crew.

We are going to start off with the session that is basically about off-label prescribing and off-label information and dissemination of that information and controls over that et cetera, et cetera.  And somewhere along the line, probably when Cole Werble takes over for this session, we will actually learn a little bit about what off-label means in this particular context.

But just to preview of what Cole is going to say -- when drugs are approved by the Food and Drug Administration, they are approved with a label, the label says something fairly specific about what the drug will treat, how it is to be used.  But physicians can use a drug pretty much in anyway they want to.  And when they do so in a way that it is not prescribed on the label that is referred to as off-label use, off-label prescribing.

Then there is information about off-label uses, et cetera which in turn becomes an issue and this often gives rise to litigation of one sort or another.  And actually, most of this conference will involve matters related to litigation rather than prescribing per se, although, the prescribing and information is usually somewhere lurking in the background or it is the foundation for litigation.

And so now, without any further ado, I am going to turn this over to Cole Werble who is moderating this first panel and then we will go on from there.  Cole.

Cole Werble:  Thank you.  Good morning.  I’m delighted to be here today; nice to see a good group here to talk about off-label use.  As everyone in the room probably understands deeply, this is one of the defining issues of pharmaceutical policy.  It affects patient access, physician prerogative for professional judgments, drug company growth, payer budgets - private and public payers - and as a budget item, it has become a major of revenue source per a number of government budgets as well as a cost-control mechanism as the settlements have become very easy to extract.  And I am sure a number of attorneys general, justices, and lawyers have become quite heroes for bringing in good sums, at least, heroes to their state legislatures and to the OMB.

It is a good subject to talk about anytime, but I think it is really great that AEI has put together a conference today.  I think this is an especially timely subject because of a change in the role of one of the major players in determining the limits of off-label use and that is FDA.  FDA has been relatively quiet in this area for about 10 years.  I see a number of people in the audience here who helped them be quiet in the area.  And actually, a lot will speak about this later.  But the agency appears ready to burst back on the scene.  I mean we are used to in Washington sunset legislation, this maybe the FDAAA Act legislation last fall maybe sunrise legislation for FDA in this area.

So actually, FDA really sort of started this furor about 15, 18 years ago when Commissioner Kessler took on the oncology community in the use of off-label.  And then it was about 6, 7 years later the FDA was forced out of the area by some private lawsuits, but they did set-up a safe harbor approach, which recently expired at the end of September of 2007.  Currently FDA is getting back in; there is pressure for a new guidance from FDA.  We will hear more about that I am sure.

And what I would like to focus on is there is a provision in the FDAAA to create realms of risk evaluation and litigation strategies, which I think very clearly pushes the agency back into off-label decisions.  Just quickly, I am going to show you two graphics that we have created -- the RPM report to try and explain what the tiers, this are tiers t-i-e-r-s - I am sure that for some other people in the industry, they are t-e-a-r-s - tiers of regulation it will be associated with this realms, they go from a sort of basic labeling-controls not much different apparently up to a very restrictive registry and individual patient monitoring systems that can be required.

Now, the key slide here is that you can then look at that as what it means to off-label use.  So that the top -- in that most restrictive category of pinnacle level of the pyramid, there will be no off-label use and then as you go down, you can see that there are areas where we predict there will really be, again, there will be a system set-up that permits minimal or other lesser restricted levels of off-label use.

And it is very timely because two weeks ago, FDA had advisory committee to look at a drug - an oral fentanyl - products already on the market and the sponsor, Cephalon, was looking for a new indication for a use that was being used in the marketplace.  About 80 percent of the use was off-label.  The drug approved currently is called Fentora.  It is approved currently for non-cancer breakthrough pain in patients that are tolerant to opioids.  And what the company wanted to do is to take it out to breakthrough pain more generally.  And that is where it is being used. 

In a way the company was doing just what the FDA wants, they did clinical trials to show it worked there.  And they went to the agency to get the supplemental “okay.”  But the advisory committee was -- a dreadful situation for them.  They went and instead of the committee looking at the clinical data that they developed, the advisory committee looked at how the product had been controlled currently on the market and said, “You are not doing a very good job of controlling this product.  Eighty percent of the use is off-label and any kind of system you have set-up for a post-market control is not working.”  The company said, “Well, we have a new system that is more restrictive,” and they said, “Why should we believe that?”

The point here is that advisory committees in the future are going to look as much at those post-market control programs as they have in the past at what is the safety and efficacy data.  And that brings FDA back directly into determining off-label.  When the FDAAA bill was being debated, there were some discussions of giving FDA the right to look at marketing plans as part of their approval.  That was taken out of the law, but we think that is going to head right back.  Too bad because there are going to be issues of, “If this is your post-market control plan, how are you are going to show us you are going to behave according to it?”  And commitments made by companies during their discussions about how they are going to control their products, of course, will open them up more to off-label challenges. 

So that is just a little bit of background; I hope that some of the future speakers will get more into how FDA is reasserting its role back in or being forced to reassert a role back in off-label.

I think that the order that we are going to go is Jack first.  Jack is a resident AEI scholar and put together much of this meeting and he has some information about a new study - a survey - that he would like to report on.

 John E. Calfee:  Thank you, Cole.  In 1997, I and LaMar McGinnis, who is an oncologist in Atlanta, Georgia in private practice -- he also has an appointment at Emory Medical School and a strong role in American Cancer Society.  He served as president for a while, is that right?  Yes, afterwards -- not at the time. 

LaMar and I constructed, along with the Roper Starch organization, the successor of the long time famous Roper Polling Firm, a survey of oncologists on information about off-label prescribing, et cetera.  We never published that survey.  We did it quickly and we got it out in public domain quickly because at the time, Congress was debating what became the Food and Drug Administration Modernization Act of 1997.  But the results were summarized in an article in the journal of the National Cancer Institute.

Then earlier this year, LaMar McGinnis and I mounted another survey of oncologists that was almost identical to the one we did in 1997.  And we worked for the same organization, different people; Roper Starch has become Roper Public Affairs.  Roper Public Affairs is now part of a large international market research firm but with the same highly competent folks that we had worked with in 1997.  And that survey was completed just -- we just got the data actually just a week or so ago.  It was conducted during April and in May.

So what I am going to provide here are simply the results of the two surveys and point out some of the things that had happened since then.  And I will see whether this is working -- it is working properly.  I should explain that in both cases, we had a sample of 200 oncologists who were selected randomly from a list of oncology specialists maintained by the American Medical Association.  It is not restricted to AMA members and appears to be the best and most authoritative lists of specialists out there. 

We began by asking them where they get information about drugs, and it is pretty much where you would expect.  We gave them various things to rate in terms of usefulness and the most useful as you might expect in both years were peer review medical journals - of course online access in 1997 was far less important than it was in 2008; the PDR is a Physician’s Desk Reference – that is a big thick volume that is a compilation of all the labels for drugs.

What was that?

 Male Voice:  [Speaks away from the microphone]

 John E. Calfee:  So, I am sorry.  You lost me.  What?

 Male Voice:  [Speaks away from the microphone]

John E. Calfee:  More details --

Male Voice:  [Speaks away from the microphone]

John E. Calfee:  Yes, thank you.  Yes, the V.U. is very useful, fairly useful, only a little useful, et cetera.  For those who do not know, and my mistake -- yes, but it is in there.  And so the PDR is the thing that doctors have kept on the desk for a long time and it contains all of the FDA labels but as we will see, it is one of those things which is tremendously often not used, not nearly as often as it is used to be.

Then there were practice guidelines from professional organizations; those have actually become more prominent in recent years.  My suspicion is that the organizations are becoming somewhat more aggressive in constructing and you might say marketing these things.  And I think the Internet has made a difference also because they are so much more accessible.  Same thing applies - roughly speaking - to government agencies; those guidelines has always been important.  Consultation with colleagues, of course, that is very useful.  Cooperative groups for those who are in the business - these are groups who work together to monitor and even organize clinical trials. 

And then there are the pharmaceutical representatives, the people from the drug companies who go out and visit doctors and talk to them about various things and one of the ongoing issues is the extent to what those representatives can inform physicians about off-label information.  So one thing we asked about is this contact with pharmaceutical reps and in 1997, 18 percent said “very useful,” 48 percent said “fairly useful.”  Those numbers were down quite a bit in 2008.

I will let the physicians and the oncologists in the panel and elsewhere address the question of why those numbers are down.  I suspect they are down because the other information sources have become so much more potent and powerful, but also because there has been an awful lot of discussion in the medical literature and elsewhere about the evils of talking to pharmaceutical representatives.

And then there are the materials provided by the pharmaceutical reps - the same thing, same comments applying.  There is in-house or hospital CME which is Continuing Medical Education which is required by the states.  A lot of the CME is, in fact, sponsored by pharmaceutical firms although the content is to a substantial degree independent of that sponsorship and other kinds of CME.  The package insert is something that goes to the patients.

And having asked them about where they get their information or what they find useful, we asked them about the reliability of the information.  Those numbers pretty much tracked the useful numbers, that is, the more useful they are, the higher the reliability and so all of which makes a lot of sense.  When we get to the pharmaceutical reps, there the reliability numbers dropped a great deal.  Although there are still the top phrases for reliability: the V.C. stands for Very Credible, fairly credible, et cetera.  So here reliability means credibility and credibility for the -- by far, the lowest credibility levels of any of these sources are the pharmaceutical representatives and the materials provided by pharmaceutical companies.

And then we asked them about their actual prescribing behavior but we did not simply say, “How often do you prescribe off-label?”  What we discovered in 1997, during the pre-testing, was that if you simply ask them that, flat out, the numbers we got at the time did not much make sense and LaMar, my co-author and expert on this, informed me that these folks may not have realized just how restrictive these labels tended to be. 

So what we did in 1997 was we had telephone interviewers read brief excerpts from two drugs that were widely-used at the time.  Brief excerpts from the drug labels telling them in those excerpts precisely how those drugs were approved for use, that is, what uses were approved.  And in 1997, those drugs were Taxotere and Platinol, two standard treatments, and essentially, the brief excerpts made clear when those drugs were to be used and how they are to be administered, et cetera according to the label.

And then in 2008, this year, we wanted to use drugs that, in a sense, everything changes very rapidly in this profession, we used essentially newer drugs, one of which was Avastin, a monoclonal antibody, one of modern biotech-based targeted therapeutics that has made such a splash.  And a drug, which I had not heard of, Xeloda, or capecitabine which is a standard - what I guess you might call old-style-chemotherapy that is still nonetheless widely-used.  And again, brief excerpts from the labels for those two drugs were read over the phone although those excerpts were substantially longer than they were in 1997. 

And then we asked them how often they prescribe in accordance with the label.  And the question was:  What percentage of the prescription do you write, do you think are in complete accordance with the PDR or the drug labels?  In 1997, the mean figure was 44 percent, the median was 40 percent, and at this point, no one at AEI or at Roper for Public Affairs can find the original database for this.  And this is the one question or maybe two questions for which we do not have as much breakdown as we would like to have.

In 2008, you can see the breakdown but essentially, most respondents said that they are in complete agreement with the label most of the time - 70 percent or more of the time.  This number surprised us by being quite a bit higher than it was in 1997 - quite a bit higher than the scuttlebutt you hear in the literature and talking to people.  That is one of the things that I will talk a little bit more about later on but we will also be hearing from Len Lichtenfeld, Scott Gottlieb and Cole Werble, again, about what is going on here.

I suspect that for various reasons - including reimbursement restrictions, all the attention that has been paid to off-label prescribing, off-label uses, litigation, criticisms, et cetera - I suspect that physicians are simply more reluctant than they used to be, either to say that they prescribed off-label or even to explicitly recognize that they do. 

But there is another element in all of these which is, as we will see from the next few questions, there are reasons to suspect that a lot of doctors are not intimately familiar with the restrictions on the labels.  In which case, they often are simply guessing as to whether they are on- or off-label.  That I think by itself is a fairly important finding.

And this is a question in which we asked about combination therapy.  A lot of combination therapies are on drug labels but an awful lot are not on drug labels.  And combination therapy has become the norm on oncology as in a few other areas such as HIV-AIDS and as you can see in both years, the vast majority of patients according to our respondents are receiving combination therapy for cancer.  No surprises there; the numbers continue to be very high. 

And then we asked them about how much influence the label or the PDR - and the PDR, again, seems to be a synonym for the FDA-approved labels -- how much influence that has on which drugs to use to treat a particular patient.  And as you can see, doctors said that these labels do not have much influence on what drug to use.  And, again, one of the things that is worth thinking about is, why it is that it has so little influence and yet they think the on-label so often, such a large proportion of the time.  But this I think is one of the fundamental findings.  The label has somewhat more influence in 2008 than it did in 1997 perhaps because reimbursement is focusing more on on-labels uses but still the influence is pretty small.

Same thing when it comes to combination therapy - these numbers are pretty much the same as the numbers we had before.  And then we asked them what you might think of as maybe the most central and the most basic question.  Essentially we said, “Suppose you could only prescribe on-label, how would that affect the patient treatment?”  And in 1997, only 8 percent said it would not affect their ability to provide the best treatment.  In 2008, 28 percent said that it would not, and 70 percent said that it would adversely affect their treatment.  In 1997, 91 percent said that it would adversely affect their treatment.  The 2008 number for Adversely Affect is smaller than it was in 1997; it is still very, very large.

Then we went on the information and here, as Cole mentioned, there has been a continuing debate about the conditions under which pharmaceutical firms can provide the information to physicians.  This was a contingence issue in 1997 when we did the first survey and again we repeated the survey that essentially the same questions this time.  And the first question we asked is, “Do the FDA rules get in the way of providing the best current information to physicians?”  In 1997, a substantial majority - actually 76 percent, 74 percent if you add Strongly Agree and Agree Somewhat, which is what SAE and AS stand for - thought the rules got in the way. 

In 2008, that number was down to 51 percent but, as Cole has mentioned and Scott will talk in considerably more detail, those rules have evolved over the years.  And a law was passed in 1997 that relaxed those rules somewhat and in the meantime the Washington Legal Foundation has made some differences in this environment which also Scott will talk about, not to mention what we will hear later on from the Washington Legal Foundation in later session.  And so there are probably things and, in fact, reasons I think that in fact the rules did get in the way less often than they used to.

Then the next question of whether or not off-label information is inherently misleading and therefore, it should always be prohibited and only 27 percent either Strongly Agreed or Agreed Somewhat in 1997.  And the percentage was much larger in 2008 - 59 percent - and I think that may reflect two things.  It could reflect a lot of things but two things come to mind.  One is, again, the ongoing debate in litigation, et cetera, including in the medical journals that these physicians read, suggesting that there is really something wrong with permitting firms to disseminate off-label information. 

And the other thing is that the Internet has made information including new results - clinical trial results et cetera - that happened not to be on-label or not reflecting on-label.  All that information is much more available than it used to be.  And so I think that to some extent, this simply reflects the fact that the physicians feel themselves less dependent upon the information they get from pharmaceutical reps et cetera than they used to.

And then we asked them about FDA - whether FDA could keep up with off-label information; with new information, revise the labels in order for the labels to stay current with the best medical practice.  And here is the overwhelming majority, roughly 80 percent are so -- again adding the Strongly Agree and Agree Somewhat -- overwhelming majority agreed that the FDA would never be able to keep up with the best, bona fide off-label information; therefore, the labels in some sense will always be out of date.

We did have one question, which in 1997 was suggested by either LaMar or someone else, but it was something that some oncologists had apparently spontaneously expressed their opinions on over the years and that is “Who can decide or who should judge about the value of new information?”  And a very large majority -- but a larger majority in 1997 said the individual oncologist is a better judge of off-label information than the Food and Drug Administration is.

So then we moved on to the FDA explicitly and we asked them essentially about three different areas in which the FDA operates, in which the regulations have some potency and say “VU”.  What is that? 

Female Voice:  Very Useful.

John E. Calfee:  Very Useful -- very good, and I have forgotten about.  And essentially, they said that when it comes to regulating manufacturing, FDA is very useful with the testing of new drugs, very useful.  Although in both years, the usefulness of FDA regulation for drug testing and by implication approval is less useful than oversight of manufacturing per se, which is interesting.  I suspect the oncologists know an awful lot more about drug testing and approvals than they do about drug manufacturing. 

And then there are the controls over off-label -- dissemination off-label information and very small number of stuff the FDA was Very Useful in that respect but when it comes to Fairly Useful, the number jumped from 20 percent in 1997 to 48 percent on 2008.  Again, I think that reflects a lot of the other things that I have already mentioned including the continuing controversies over FDA regulation of off-label information.

Those are the basic results that we got in these two years.  I mean there are a lot to think about and a lot to talk about.  And I suspect Len and the others will have something to say about them.  Just a couple of thoughts of mine, of my own - one of them is I think -- again, these are things that I have mentioned at least briefly before -- I really think that there are at least two and maybe three forces that are having quite a bit of an effect. 

One is the simple growth of the Internet in Web access, and the extraordinary way in which not only do results of medical journals become easy to get to but also the extraordinary linkage among them.  Seeing a new article, the New England Journal, you go online and you look at it and you get instantly linked to all the other research that are cited.  If the article is six-months-old, nine-months-old, you can instantly see all the other research that had cited the article that you are looking at.  When asked how the American Society for Clinical Oncology when they had their annual meetings at which new clinical trial results are released and that was what?  Just last week?  This week?

 Male Voice:  [Speaks away from the microphone]

John E. Calfee:  And I saw something indicating 50,000?  Five thousand?  That was a misprint what I saw, I hope so.  Five thousand is still a very large -- the results of 5,000 studies basically were being released at the ASCO conference --

 Male Voice:  [Speaks away from the microphone]

John E. Calfee:  Abstract -- 5,000. 

Male Voice:  [Speaks away from the microphone]

John E. Calfee:  What was that?

Male Voice:  [Speaks away from the microphone]

John E. Calfee:  They do not have reservations.

Male Voice:  [Speaks away from the microphone]

John E. Calfee:  Five thousand abstracts and how many oncologists do you think were --

 Male Voice:  [Speaks away from the microphone]

 John E. Calfee:  Are you kidding?

 Male Voice:  [Speaks away from the microphone] No.  A lot of it is international.

 John E. Calfee:  No, do not go to Chicago right now.  But, at least, they have it in May which is about as good as a time which you can have it in Chicago.  So 5,000 abstracts are being released; most, if not all, of which will be accessible from the Internet for those few forlorn oncologists who were stuck with their patients rather than going to Chicago.  And this kind of things happens frequently.  And so the accessibility of new information is far greater than it ever was before.  And I think that does have a pretty substantial impact on a lot of the things that we are talking about or that the physicians were asked about.

And then the other thing is that an off-label information, as Cole mentioned, that has been controversial, at least, since the early 1990’s when David Kessler as FDA Commissioner launched essentially a campaign against that - based partly on his law review article when he got his law degree.  Same time he got his MD and his law review article was on drug marketing and with special attention to things like off-label information. 

Things became fairly controversial then and that was true through the 1990’s leading up to the FDAMA Act, which did have a provision, which to some extent, loosens the constraints on dissemination of off-label information.  These things go on ways and I will think that for the last year or two, things have actually gotten hotter in this particular area, in three respects.  One is that there has been more discussion in the medical literature about, I mean, how should I put it?  New England Journal, Lancet, the British Medical Journal, JAMA, et cetera, when you look at those journals and they are full of new information that is essentially very valuable and is not only on the drug labels. 

But there are also quite a few articles on just how unwise it is for pharmaceutical firms to disseminate any new information that is actually being provided by these medical journals unless that information is actually on the medical labels.  So I think this does have an influence. 

In addition, some of the newer cancer drugs are really pretty darn expensive.  The payers are going to implementing quite a few measures to try to control these costs of newer biotech-based-drugs such as the monoclonal anti-bodies are expensive.  If you go back even two, three, four years ago, these drugs even though they cost $20,000, $30,000 $50,000, $80,000 per year for the patients for which they are being used -- two years ago, that was a pretty small factor in the budgets of Kaiser Permanente, et cetera.  The budget impact is growing very rapidly and it has reached a point where they really are thinking about what they can do. 

I hear a lot of buzz from the people I talked to about the payers who are clamping down on or are trying to constrain the use of the newer cancer drugs.  And the one way they are doing it is by looking somewhat askance at off-label uses of these drugs.  And so a lot of physicians are in situations where they have to go to a quite a bit more trouble when they are prescribing off-label and that may be another reason why they are maybe a little bit reluctant to say they are off-label as often as they are really, really are. 

But again, that is, at this point, speculation and the main thing is that I think there are some things that are very consistent about what the oncologist have thought 11 years ago and today about an awful a lot of issues surrounding off-label usage and information.  And there are a few areas in which their views seem to have changed quite a bit. 

And, Cole, that concludes my remarks.

 Cole Werble:  Well, thank you very much, Jack.  There is a lot of interesting information in there.  I am sure you are all focused on the question “Can” and the results of who can better evaluate off-label information and FDA’s rating went from 12 percent in 1997 to 30 percent in 2008.  And I can only attribute that to Scott’s, our next speaker, tenure at FDA - twice during that period.  I am sure that that is a direct reflection of his contributions.  Our next speaker is Scott Gottlieb.  He is a new resident scholar here -- a permanent scholar, resident alien at AEI and former FDA Deputy Commissioner.

 Scott Gottlieb:  Thanks for the introduction.  I’ll just stay here since I have the computer in front me. 

I want to talk about the public health aspects of off-label prescribing.  I think that we are in a political and legal environment that assumes that there is no public health consequence to restrictions on information, restrictions imposed on the ability of companies to speak to physicians about potential off-label uses of drugs especially in unmet needs.  That is simply not true.  There are public health consequences for these policies.  I do not think it is hyperbole to say that controls on information actually cost lives.  It is a hard thing to quantify which is why if I say it, but hopefully I will show some evidence, at least, anecdotally that I think supports that thesis.

Now those who argue that we ought to maintain a very strict policy in legal environment restricting off-label information and in particular restricting the ability of sponsors to promulgate it, I think we would argue the same.  They would say that the promulgation of off-label information has public health consequences and in some cases, could potentially cost lives.  That is not far-fetched either.  You have unscrupulous drug reps in certain cases.  We certainly know there is anecdotal evidence of that and doctors who are irresponsible who have goaded into prescribing drugs for off-label uses where there is not substantial scientific evidence and patients end up experiencing more of the consequences, more of the side-effects of drugs in situations where there might not be proven benefits so certainly the opposite to what I said is true.

And so I think the question becomes, how do we erect a policy environment that takes into consideration the benefits of information about off-label uses while constraining those circumstances where the dissemination of that information might have harmful effects.  How do we embrace the good aspects of the dissemination of information and control the negative aspects?  And can we do that from policy standpoint. 

I think that doing so is exceptionally challenging and it is challenging for many reasons that are medical and scientific.  But it is also challenging because of misperceptions that we have in the policy environment that we exist in today.  The regime that we erected when it comes to controlling off-label information, I think it is predicated on some false assumptions.  And I want to go through five in all.  I lay out five so you will be able to track my progress during my talk today and you will know when the nap is over.

The first false assumption I think that we have when we think about controls on off-label information is that all information is bad; all off-label information is bad.  If you use the word off-label, if you refer to something as off-label and in the policy environment here in Washington, it is assumed that you are engaging in some kind of criminal activity.  You have like an offshore bank account or something equally nefarious if you are a doctor prescribing off-label, which I think is somewhat reflective in Jack’s survey results. 

And in fact, we are in an environment where I think the legal establishment, the state attorney generals and others who have gotten more and more promiscuous in terms of what they are willing to go after, not creating any clear boundaries to what is bad and what is good information.  We are a long way from some of the initial cases that were prosecuted or brought against the pharmaceutical companies on the basis of off-label promotion. 

And if you bring up the issue with any one in sort of a political environment - I have had this conversation many times - they harken back to Neurontin.  Or some other cases where a company was alleged to have promoted off-label for uses of drugs when there was not really substantial scientific evidence, but we are long way from that.  And I will give you one example of how far we have come from an environment where the types of cases that were brought were brought against companies that were allegedly or knowingly promoting for uses of drugs that really did not have substantial scientific evidence.

I think there is an investigation right now that could potentially be a watershed moment in the off-label environment.  And that is an investigation into alleged promotion of certain uses of Rituxan by Genentech.  As I understand it, there are allegations and these have been publicly reported in their own 10-K that Genentech allegedly sponsored physician education around certain uses of Rituxan back in the 90s.  Rituxan, as many of you know, is a drug for lymphoma.  The uses for which they were alleged to have sponsored physician education were forms of lymphoma that were not treatable - they were universally fatal at that time.  Some of those uses are now approved by the FDA.  And if at that time that Genentech was alleged to have sponsored the physician education, you went to the National Cancer Institute website, a reference to literature to try to find out how you treat a patient with these forms of lymphoma, the answer was Rituxan.

So it was the standard of care at that time, was based on substantial scientific evidence, and yet, the Justice Department is allegedly investigating whether or not Genentech violated some law in allegedly sponsoring physician education around this.  I think that that is a long way from Neurontin; it is a long way from the other cases that get quoted to “support the legal regime that we are in.”  It is an instance where the information, in fact, probably was important to physicians.

Patients did not have alternatives.  The Rituxan is largely non-toxic, although not entirely, certainly, and has gone on to prove a lot of benefit in the forms of lymphoma and eventually made its way on to the label.  So clearly, I think not all information is bad; some of it is exceedingly important.  And then the question becomes, why should the drug companies not going to be given the right to try to promulgate that information?

Well, in the case of Rituxan, it is a complicated drug to prescribe.  It is not always clear how to dose it; how to infuse it; how quickly you can infuse it; and what you need to do concomitant to infusion.  Certainly if a patient had adverse reaction from a doctor who is a prescribing it without support, without medical education, the company would be on the hook for that.  And so to ask the company not to speak to the primary user of the drug about the complicated administration, I think presents certain public health challenges.  And the company, after all, is the only one with deep pockets, in a position to sponsor that kind of medical education.

 The second false assumption, I think, we have is that drug firms cannot be trusted or they have nothing useful to say and I think that is a corollary to the first point.  I think Rituxan certainly demonstrates that the companies, in many cases, do have something useful to promulgate especially when you are dealing with unmet medical needs with its fast moving science where the FDA does not keep up or cannot keep up just by its own statutory limitations.  After all, it takes about two years to get a new indication on a drug label and, that is, if everything goes well just by virtue of the regulations and the way the agency is set up.  So in a fast moving field, like lymphoma, when new drugs do become available, the company potentially does have something very useful to say.

 The other case I will talk about, and I have a slide up relating to it right now, is Herceptin.  And I think this starts to demonstrate why the companies do have something useful to say and why restricting that speech can have consequences.  Back in 2005, some data came out - in fall of 2005 - demonstrating that Herceptin, which is a drug approved for advanced breast cancer, when used early in breast cancer and more frontline uses could reduce breast cancer recurrence by 50 percent.  That is very dramatic.

There are few instances and even the care of cancer patients where the introduction of a single drug can reduce recurrence so dramatically.  I think the Taxol is actually -- I have similar data around them.  But certainly this was very important information.  Well, on this chart you see use of breast cancer in that setting in a so called adjuvant setting - the more frontline setting - you see when the data came out back in 2005, it was released at ASCO, use started to go up of the drug.

And it reached about 50 percent penetrations.  So 50 percent of patients who fit the criteria to be receiving the drug according to the studies that came out were getting it.  FDA delayed the approval of the new indication once, so it took the agency about a year and half to get this on to the label from the date that the data was released.  It was delayed, not because of any real difficulties with the data itself or the veracity of data, but more had to do with difficulties in getting the full compliment of data from the NCI at that time.  So I think it is fair to say that it had to do with a little bit of bureaucratic wrangling rather than the veracity of the data itself.

It was eventually approved in November of 2006 and penetration is now over 80 percent, which is probably where it should be if you talk to most oncologists.  The drug does have certain side effects.  It can cause cardiomyopathy or swelling of the heart so patients who have pre-existing heart conditions do not use it.  Other patients might forego the treatment for other reasons but you figure anywhere between 80 to 85 percent penetration, it is about where it ought to be from a public health standpoint.

Well, clearly, there was under utilization of the drug for a very long time.  You had an initial burst of use probably corresponding to academic physicians or people who were high prescribed user for Herceptin to begin with and felt comfortable, it tailed off and it really did not take up again until after approval, until after Genentech was able to go out and speak to physicians.

I would hypothesize, and there are some data to support this when you look at the prescribing patterns, that the initial burst was in the more academic institutions and the ones who were reluctant to prescribe it were the more community-based physicians, precisely the ones who are often the ones who get information from continue medical education sponsored by drug companies because they are away from the academic centers and they are not necessarily as well into the information flow.

So did that have public health consequences?  Well, if you believe that the restrictions on medical education and speech did in fact contribute to the under-prescribing, certainly, it had public health consequences.  You can quantify how many women who did not get the drug over that period of time - that year and a half period of time - are going to go on to relapse from frontline breast cancer and after all, certain, percentage of people who relapsed are not going to survive the disease.  So again, there are public health consequences to this. 

And getting back to the second false assumption that the companies have nothing useful to disseminate and cannot be trusted, in this case, Genentech has something very useful to disseminate.  The information about how to prescribe Herceptin in this new setting would have been very useful information, but as you can imagine living under the thumb, the specter of a potential investigation into their alleged promotion around Rituxan, I’m sure they were and about to say boo about Herceptin and this new indication.

 So the third false assumption I think the regime that we have now erected is based on, is that the doctors themselves cannot be trusted.  So just like the drug companies cannot be trusted, doctors cannot be trusted.  And this is, I think, somewhat concerning because if you think about it at the core of the investigations that have gone on into particularly Continuing Medical Education. 

We talked about detailing to physicians and information targeted to physicians, even types of restrictions some would like to maintain over the dissemination of journal reprint articles to physicians - all of that has to be at some level predicate on the belief that the doctors cannot be trusted with that information.  And, in fact, you need an attorney from the Justice Department to police what ought to be going to a physician.  That the physicians are too easily lulled into prescribing by sleek marketing information, even bottom line scientific information, journal reprint articles, but for a very earnest attorney who is not necessarily medically trained, the physician could be led to very adverse behavior.  And so, therefore, the doctors should not be trusted.

 The fourth false assumption, I think, the regime we have right now is predicated on is that the attorneys themselves are public health-minded.  And I do not mean this in a derogatory way, I’m sure they are very good people.  I know some of them because they have gone out to work for drug companies, but there is an assumption that the attorneys who work on these cases, the Justice Department and others who are bringing these prosecutions are doing it with a public health orientation, if you will.

And I think that is simply not true.  I think that the criteria that go into selecting the types of cases they bring have to do with how much money can be recouped, the ease of bringing the prosecution, how much evidence there is, whether you have got a smoking gun, someone on a tape or video, some other really good evidence, but not necessarily the medical criteria that is at the heart of the issue.  Whether or not the alleged off-label use, off-label promotion, fell within the realm of medical convention or fell very outside of medical convention.  Is it Neurontin or is it Rituxan?

I would offer that there is not a lot of consideration given to that thought and that is why you see cases like the potential investigation to Rituxan or even the investigation into Evista by Lilly.  It is another drug used for prevention of cancer recurrence where there was an investigation brought against Lilly for allegedly sponsoring continued medical education around that.  I think that case is going to get talked about a little bit later - again another use that is subsequently got approved by the FDA.

 So I would offer that the prosecutors who work on these cases ought to be required, as part of their staff manual, their requirements from their own superiors, to do a public health check, to call FDA, to call NIH and ask the question about whether or not the alleged use falls within the realm of medical convention, whether there is substantial scientific evidence and it is a reasonable prescription for doctors to write or it falls well outside of what we will consider acceptable medical practice.

Now they would argue they do that anyway.  There is no explicit requirement for them to do that and there is certainly no explicit requirement that that ought to be a significant part of the criteria that they use when deciding whether or not to bring a case.  I would offer that it ought to be and I think that would make the cases themselves, the regime and the attorneys who are engaged in this, more public health-minded.

 Finally, the final false assumption I think that our regime is predicated on is that the supplemental new drug application process is efficient.  It is not efficient.  I said at the outset that, by virtue just of how the FDA works and the statute, that it takes upward of two years to get a new indication on a drug label.  That is just by virtue of the fact that there is a 12-month review and it takes six months to get the information into the FDA and there are often delays even before that.

And so you are looking at a best-case scenario, a year, in most cases two years, just by virtue of the way the process itself unfolds.  And you are not going to get a lot of efficiency into that.  It is always going to take the FDA at least six months to review an application for a new indication.  It is always going to take the company at least four or five months to get the application in the format that the FDA wants when it goes about reviewing.

 But the other challenge here is that the supplemental new drug application process itself is counterintuitive and by that I mean that supplemental new drug applications by virtue of the fact that they are supplemental new drug application means the drug is already on the market.  So we already know a great deal about the drug.  It has already been on the market for probably a number of years.  We have a lot of information about its safety profile because it has been used.  We have more information about its effectiveness.

And so in that kind of environment, you ought to think that the FDA’s tolerance for risk is higher, that it ought to be willing to go quicker on the supplemental new drug application because it knows more.  But the exact opposite is true and it is counterintuitive, I know, but it is true.  The agency often sets a higher bar on the supplemental new drug applications for a whole variety of reasons often relating to the fact that they did not get all their questions answered the first time and so now this is their second shot on goal and they are going to try to answer all of the ancillary questions that they could not answer in the original application in this supplemental application.

So they are going to ask the company to study other things that are ancillary to the new indication to try to get at that information.  So it is often the case that you see the supplemental applications dragged on or wind on longer than the original NDA and often encompass much larger clinical studies than the original NDA.

 I just want to show two cases that we can contemplate: one relating to Avastin and breast cancer.  As you know Avastin was recently approved for use in breast cancer in February 22nd, 2008.  The initial approval for Avastin in colon cancer was four years ago, in 2004, and over that time, you had a lot of data come out, some very significant studies including the study in December of 2007 that was really the basis for this supplemental approval, so you can see the time it took from the publication of the initial results that the approval is based on and then the subsequent approval. 

And so the question becomes, when did this drug become effective in breast cancer?  Did it become effective in breast cancer at the time at which FDA approved it?  Clearly not.  I do not think any of us would argue that.

At what point was there sufficient information for this to be a reliable, reasonable decision for physicians to make?  At what point was this information credible enough that the sponsor in this case either had a right or responsibility to promulgate it?  And, again, we are talking about a drug that is complicated to prescribe in an indication for which the doctors who would be prescribing it might not be familiar with it.  After all, a lot of oncologists specialize in certain kinds of solid tumors, so a doctor treating breast cancer might not be treating colon cancer patients and might not know how to use Avastin.  And this is not something that you just take the pill and call me in the morning.  This is a difficult infusion process where you want the physician to be trained and you want the staff to be trained, frankly, on how to prescribe the medication.

 So the second one is Eloxatin - this is a drug used for the treatment of colon cancer, a little bit more of a torturous process through the agency.  It was first approved in France in 1996.  Before we start to think that the French are more enlightened than us, it is made by a French company so they often cut their own sponsors same breaks.  But it was approved in France in 1996 for more advanced metastatic colon cancer.  It was subsequently approved in more frontline colon cancer in 1998 or 1999 and there was a wealth of data produced along the way demonstrating that the drug was more effective than we even initially thought.

And, now, it is used in frontline colon cancer and believed to be very effective in colon cancer.  It was not approved by FDA until 2005 in the initial indication of more advanced cancer and it was not approved in a broader range of tumors and more frontline settings until 2006.  So once again, when did the drug become effective?  When was it reasonable for information to be communicated?  When was it reasonable for physicians to make a prescribing decision to use this drug in more frontline settings?  Was it 1996?  Was it 2003 when the results of a very large study that became the basis for approval came out?  Was it 2005 when the FDA initially approved it or 2006 when they subsequently approved it for more widespread use?

 So in conclusion, I think the question we should contemplate when we are looking at this issue is:  Do we need to embrace a decerebrate policy?  And by that, I mean a brainless policy.  Do we need to have a legal regime and a policy that is a one-size-fits-all hammer in the name of trying to enforce a regime or create a regime that prevents the bad behavior; the behavior that we might agree could have negative public health consequences.

Or is there a way to create more nuances to our policy approach to enable some of the information dissemination that I think reasonable people could agree has public health benefits while restricting the kind of information that reasonable people could agree does not offer a lot of useful value in the medical practice setting.  I think the case of Herceptin, if you look at it closely, in frontline breast cancer shows that the information restrictions do have consequences, that there are physicians who are not going to be comfortable or even fully aware of new uses of drugs especially in unmet medical needs, especially in complicated cases, complicated drugs, drugs that are difficult to prescribe.

And so can we make these distinctions?  At the very least, I think, it is probably appropriate to ask the question of whether or not the people who are engaged in this activity, the state attorney generals, the lawyers and the Justice Department, should be paying more heed to these public health considerations and should be looking at the literature and using that as a criteria for judging what cases they go forward with.  I know they say they do that.  I know they are very well-intentioned people, they certainly have good interests at heart, but it is not an explicit criterion.  And that is why you see things like the investigation into Rituxan, which I think if it goes forward is going to dramatically change the environment for medical education.

 So in summation, I think there is a way to nuance this.  I think there is a way to bring more rational discussion to these policies.  I do not think that we are doing that now.  Thanks.

 Cole Werble :  Thank you, Scott.  I think we just realized again that as well as being a knowledgeable scholar and regulator, Scott is a very clever and brilliant wordsmith.  I hope you recognized that his first reference to state attorneys general was to call them promiscuous.  I think he is right.  They are using broader -- but I think he very carefully chose that word and as you heard later, he does not have, particularly, high regard for their role in this area.

Okay, our third speaker is Len Lichtenfeld.  He is a deputy chief medical officer at the American Cancer Society and he is going to comment on some of the two first speeches and especially the survey done by Jack.

 Len Lichtenfeld:  Scott, since you are controlling the computer -- yeah, he has to bring my slides set up, he has to do that.  While he is doing that, let me just assure you that I did not coordinate this conversation with Scott before I presented today, so you are going to hear some of the same information but perhaps in a little bit different approach.  Nor did I -- the title you are going to see come up, it is somewhere up there, it is -- that one right there, let’s bring the slides up

 Male Voice:  [Speaks away from the microphone]

 Len Lichtenfeld:  Well, I do not know if it would be clinically correct.  You know something there is a -- we always respectfully disagree.  I told Scott that I had never met him before but I have known him from CNBC and I know he likes to have arguments and discussions so I feel grateful for that.

But the reality is there is a lot we do not know.  And I’m going to disvalue right from the beginning by saying that if you think we have the answers to how doctors make decisions on what they do, the answer is we do not.  I, for example, have been talking, in this whole issue, about how doctors prescribe and why they prescribe, we do not have the information, we do not have the quality monitors.  That is really one of the failings of our - and I will say - non-system of medical care.  I use that advisedly.  I’m very pro-medicine.  I’m an advocate for medicine.  It is a part of my life, my professional life, but we do not have the information that we need to understand how things are done and why they are done.

And let me just expand this for a moment.  I am deputy chief medical officer for the American Cancer Society.  I do have my own opinions much of which are going to be my own opinions today but I also have a different outlook.  I have served for a number of years on a committee that sets the Medicare physician fee schedule.  I have been very much involved in medical economics and the impact that some of these decisions and how they play out over time not just for an off-label issue but how they then in turn impact on public policy and where some of the issues are.

So with that, let me just go right here to the first slide -- I set up -- and the truth is somewhere in between in terms of off-label use and that is from a completely regulated environment to a non-regulated environment.  And I firmly do believe that there is a middle ground and I hope by the end of this discussion, I can complete it.  I will make a suggestion that may resonate with some of you in the room.

And I think this is really the bottom line of this whole discussion.  Cancer care in the United States is off-label medicine.  And that is the nice way of saying what I was really going to be putting up there.  I told Jack when I walked in, since I’m a guest in his house, that slide was going to say PDR?  Hah!  As in exclamation point, that is, if you go to a doctor and you have cancer, and say, “Doctor, the treatment you are going to give me today, are you using the PDR to guide my treatment?”  Well, that maybe the wrong way of phrasing it.  There are lots of ways you can phrase a statement.  But if you ask that question to suggest that that is what the doctor is using to tell the doctor what medicines to give you, that day in his or her office, I would walk out of that office and go somewhere else. 

The doctors do not and should not be relying on PDR information to decide what the regimens are they are going to be using to treat you for cancer with one exception.  And I was thinking about this, Scott, as you were giving your talk, and that is when a drug first comes on to the market for approval, when you have a single agent like an Avastin, that has been demonstrated to be effective in colorectal cancer and although there was a huge lag time and people hopefully are out there using it but it cannot come to market and be sold until it gets that FDA approval - that is the moment when that PDR probably is effective.  Beyond that, you can forget about it because the drug will be used in a variety of different ways and the information in the PDR really will not be very useful in current medical knowledge.

 Well, I’m going to talk a little bit.  The gray hair thing that I have here gives me a 35-year background on what has happened between then and now.  Going back I have been training in the NCI back in 1972.  Then I practiced in Baltimore in 1977 and back then there was no such thing as oncology in the community.  There were, there were some people out there who giving some drugs and making people throw up in a bucket and I mean that sincerely, I do not mean to be impolite.  That is where I walked into when I went into private practice.

We did change all that when we tried to develop a true oncology practice.  Today, fortunately, that is much more the norm than the exception.  We did not have a whole lot of drugs.  Almost everything is intravenous.  We did not have much successful treatment of Hodgkin’s disease, for example, some childhood leukemia; there certainly was not much of that.  There was not anything in the way of the Internet.  And the outpatient chemotherapy was something the insurers literally, literally would not pay for.

Well, the landscape has shifted.  We have a lot of drugs, combinations, targeted therapies.  Drugs are moving to oral settings.  The Internet is certainly pervasive in our lives even though some doctors are still reluctantly adapting to it but they are getting there slowly but surely.  And we have had explosion of studies and reports.  I want to emphasize a little back and forth that Jack and I had during his presentation.  Thursday night at 9:00 p.m. last week was the bingo moment when I came to ASCO releasing its abstracts.

Traditionally, what ASCO used to do was through a book, send it out to us, we would get the book - everything was embargoed - and then they would time the release of the study until the presentation at the annual meeting.  Well, 5000 abstracts is a whole lot of abstracts so the people actually do comb through those abstracts - I do not know if I’m sorry to say this or not - but the people who pay more attention to those abstracts than anybody else are a) the doctors or b) the financial analysts.  If you selected b, you would be right.

The financial analysts spent the night, so the next morning, we were sort of joking about CNBC, but where you found out where the hard information was, was by following where the business took you.  Five thousand abstracts -- let me also disvalue of any thought that doctors sit around all day reading their journals.  In the old days, we would have piles of journals.  We say, “I’m going to read that tomorrow.  I’m going to read that tomorrow.  I’m going to read that tomorrow.”  When the pile got so big, you throw them in the trashcan.

We do not have that anymore.  What we now have are downloads onto our computers which pile up on another file, that pile up, pile up and pile up and we still do not read them.  So unfortunately, the amount of information, to be able to process that information, the ability to make use of that information is indeed very limited.

Let me give you a concrete example where we were and where we are.  Drug up their VP-16, that was the code name for drug that is now known as etoposide.  In fact, I could not even find anything on etoposide in the PDR to show you how influential the PDR is.  Once a drug is generic or once a drug does have a market, it does not necessarily have to be in the PDR or somebody is going to pay to have it there.  So you cannot even go to the PDR, which I tried last night to find out, I could not even find the information on etoposide or VP-16 or Vepesid, whatever name you want to try.

Having said that, my first run in with off-label indications was back in 1977 when I started my practice.  There is a fellow back at the Cancer Institute 1973 or ’74 - I saw what we call the Lazarus syndrome.  That does not happen very often.  The Lazarus syndrome was a man lying on his bed, with small cell cancer.  Friday afternoon we gave him the first dose of VP16 at the Baltimore Cancer Research Center and we thought he was going to die over the weekend.  We came back on Monday that man was sitting up and eating a meal.

It was truly one of the moments you have very rarely.  Gleevec is another example that we could use back in 2002 but that was a breakthrough moment.  It quickly became the standard of care.  And however, when I went into practice, I had the drug investigation, they shipped it to me.  I gave it -- Medicare, they do not want to pay for outpatient chemotherapy anyway, but they would not pay me because I used the drug that was off-label.

And that thing became sort of my advocacy career back then 1977, writing letters, calling and everything else to congressmen, senators, whoever would listen to get through to somebody and say, this is inappropriate.  And eventually, they agreed but this was the standard of care.  There was no other drug which will help the patient, really help the patients with small cell lung cancer.  So that was the beginning where I started.

But let’s talk about where we are and I think we have to bear in mind that not everything is rosy out there.  And this is an example.  These are blog shots.  I do a blog.  I do not have to worry about peer review.  I just put out what I think.  It is my opinion.  Nobody tells me.  They do not have to review it, you know, the editorial board.  I just put it out there but I guess I have been pretty careful because nobody has told me to be quiet yet.

Anyway this is a blog shot from the recent ODAC hearings, Oncology Drug Advisory Committee hearings.  They were held here in Washington, down in Gaithersburg; it was back in February - I cannot read the date.  But the bottom line is here is a drug that was approved in 1990 something or rather.  It was approved for avoiding transfusion in patients getting active treatment with chemotherapy.  It developed a huge off-label use and came what we called the cancer-related anemia - those patients who are not being treated who have anemia related to their cancer, it became - I do not have the data - a substantial portion.

And also, for better or for worse, if you are seating in the audience from the company, I apologize but this was the fact -- well, the drug company was creating financial incentives for doctors to give the drug.  It was documented, the articles were there and in fact, what is really sad was that the ODAC hearing, they had to be asked five times before they said the word “yes” to the question, “Are you still providing rebates to oncologists to provide those drugs?”

And I will tell you that there are many oncologists out there who were furious about this but what we ended up with was a huge market in ESA’s Medicare drug budget by about $15 billion in Part B; $1.9 billion of that a year before this started - $1.9 billion was related to the use of these drugs.

That is a big amount of money, folks, and it impacts what doctors get paid, all right.  At one point, it was a rapid ascent in the use of that drug.  And then it turned out there were some clinical trials out there that people were not talking about and guess what: it looked like more patients were dying than were being helped from this drug in that indication.  And then the ball started rolling.

The Cancer Letters started in.  The FDA had some comments about the [indiscernible].  They came to the FDA and that was started a year ago and that has led all the way up to the fact that now these drugs are under increased scrutiny and so are the practices of the companies.  And direct to consumer advertising also - you cannot forget it - it played a big role.  And a lot of oncologists were upset about that.  They felt that the financial incentives were inappropriate.  They felt the fact they did not know the risk information is inappropriate.

One of the questions I will leave you with that I do not have an answer to in the off-label -- I sat in the ODAC hearing and the question came up about off-label use.  Dr. Patterson -- “We cannot discuss that.  That is off the table.  We cannot discuss that as part of our discussions, the panel asks the question.”  We cannot discuss it, still cannot discuss it.

And I think the question has to be asked at what point should the FDA be able to answer the question, “are we doing more harm than good?”  Because there are situations in cancer medicine - bone marrow transplant and breast cancer are excellent examples - where the advocacy gets ahead of the facts and just because it is out there and people do it does not make it right.  And risks and harms to patients must be kept in mind.

So I agree with the executive summary that I had from the presentation, oncologists feel the PDR has a little or no influence, I certainly agree.  The PDR has declined drastically - that is in the data sets - I agree.  And online access to medical libraries is more credible than 11 years ago but is far from perfect.  Just because an article appears in the New England Journal it does not mean that it shows up in the doctors’ inbox the next day.  Unless they do subscribe to the New England Journal, they cannot get to the article.  So there is a big issue on how we exchange our knowledge.

Medical knowledge has exceeded the capabilities of any one physician to keep up to date.  And I do not care if it is in oncology; I do not care if it is in cardiology, it is impossible for doctors to do all the work they have to do.  Internet access has increased and we already talked about the ASCO effect, it is very real.  However, information is not readily accessible and it is not well organized.  You cannot go to one place and say, “Give me the latest information on this particular topic.”

It is pertinent to me today as I’m standing here before you because I have written a blog last week and the week before talking about -- I will talk more about the National Comprehensive Cancer Network.  Well, they do collate some of that information, do give you expert opinion.  It is in the public domain, and I will share with you, with the unfortunate illness of Senator Kennedy, I have spent a lot of time collating data over the past 24 hours because of interviews and so forth and some writing I did - the reality is that probably is the one source I do go to.  Our data on the cancer.org site is primarily geared towards patients.  And NCI does have an excellent website where they do have health professional information, but it is not necessarily as up to date as it should be. 

Now let’s take my view of the Herceptin situation.  I wish I had Scott’s slides superimposed.  I will tell you something, Scott.  I think actually the uptake was fantastic.  I think it was great because the normal uptake -- you are going to fall over when you hear this normal uptake of technology and new procedures in medicine is.  I should ask, raise your hands - five, 10, 15 or 17 years?  It is 17 years.  That curve normally is 17 years so we have compressed the curve tremendously.

And I do not know the reasons why the curve took so long.  This is the headline we had on our website, the date; you can see the same date.  It goes back -- that is April 27th, that is when we report out of ASCO, but the bottom line is that I was in the room when that data was presented.  And let’s remember an abstract is an abstract.  It is not a published peer-reviewed article.  It is limited peer review.  It is not all the information you need to make a medical decision.  I have written that up so I will show you that comment so just, I’m not saying it here today - it is not all the information needed -- I was in the room.  There were two rooms, ladies and gentlemen, there were 10,000 doctors.

Were you there, Madam?  No?  Okay.  Let me tell you, let’s set the scene - 10,000 doctors, two rooms, 5000 each piping it in and when those -- there were three abstracts so when they got done, when each one got done, I have never seen this happen in medical meeting, there was a standing ovation and literally it was called a rock star moment.

There were cover stories in news journals.  There were newspaper articles.  It was all over the television.  Our oncologists knew about it.  That is one of the very few times, I will say, in my medical career where I’m willing to say, and I’m a very conservative person, that a new standard of care was established at that moment.  And Scott was absolutely correct.  Fifty percent reduction in recurrence for women with a very aggressive form of breast cancer and decreased death and improved survival - that was a defining moment and one of the few absolute defining moments of medical oncology.

It did in fact change the standard of care.  So if the doctors did not know about it or a patient - I do not expect patients to know everything - you have to be under a rock.  And I do not know the answer to the question, why it took so long.  But I will tell you there is a normal reticence, for good or for bad, in the medical profession for adopting new information.  Some are early adopters, some are okay, some are late adopters - all the reasons that go into it - maybe it is the test, maybe it is the cost, maybe it is the insurers, maybe it is a whole lot of things, but I will tell you that information was blasted.

Nobody, no drug representative, maybe the drug rep could have reinforced it but go back and look at all the newspaper articles, Google and you will see the word rock star moment again and again.  So I’m not sure why it did not happen.  And then the New England Journal article Scott referred to - there, it is October 23rd, 2005 - confirmed it and just reinforced it and once again got substantial amount of press.  But then we had this situation and I think that this is very telling, particularly telling to where we are sitting here today.

And this is a blog I did; this is dated as May 4th of 2006.  And it was generated by an article appeared in the Wall Street Journal.  And it talked about the difference in getting Herceptin in the United States versus what was going on in the European Union at that time.  And those of you who are maybe not as familiar with the drug process known as the FDA, European Union has an FDA equivalent but then passes or does not pass.  And then once it passes the European Union, then it goes to the individual countries to make a decision.

If you do not think it was economically driven, what was going on in Europe at that time -- if you are a woman in England, you have a darn hard time getting Herceptin for breast cancer.  In the United States, if you could afford it, if your insurer probably is going to pay for it, for reasons I can get into later, insurers were probably going to pay for it, but if you did not have insurance and a whole lot of folks do not have insurance, you still had the option of getting that drug.  In Europe, you did not have the option because NICE, the English equivalent of the people who make the recommendations, had not said it was okay.  It was too expensive, it was too much money.

Actually I went back and reread, I was pretty surprised at what I have written.  There is a lot to be understood about how drugs get disseminated and access to drugs.  By the way, the Europeans approved it before we did but we were able to use it long before they were.  And I said the same thing that Scott said.  I said it was almost criminal not to be able to have access to this drug because it really and absolutely demonstrated -- once a breast cancer recurs, unfortunately it is fatal.  It does not mean people die right away.  They can go for many, many years; ultimately, it turns out for most situations fatal.

You want to prevent that recurrence from happening.  Not to offer the drug will be a criminal event.  Here is the article that came from the New York Times.  You can see, the British clinic is allowed to deny medicine solely because the lady lived in a particular - equivalent of zip code -- it was called postal code lottery.  And the courts upheld that they did not have to do it and then the court said, “Well, she does have the right to do it.”  It was only financially driven.

And by the way, just so that everybody is on the same page, the same time that New England Journal article was coming out, the Lancet came out, a very respected international medical journal, scientific journal and said “Herceptin should not be approved for use in adjuvant therapy of breast cancer.”  And I wrote in my blog, I said to the editor - I do not know the editor, I never asked them - “I could not imagine why they came to the conclusion in editorial that Herceptin should not be approved other than the fact that it was going to cost the budget a whole lot of money.”

And then, of course, we eventually, as Scott mentioned, got around.  This is the FDA website shot of approving this drug and it was in November 2006.  But well, by that time, it was hopefully in wide use, but, as Scott has pointed out, it could be better. 

Well, okay, here, what about Avastin?  I think Avastin represents another side of this coin.  This is a screen shot of a story from San Francisco Chronicle and the date on there is April 16th of 2005 -- 2005 I believe, yes, I can see it better now.  And this is -- there was an announcement -- ASCO is coming up saying years the Herceptin announcement came -- ASCO was coming up -- were you at the FDA then, Scott?  Remember 2005?  Were you?  Okay.  So here you have NCI calling a press conference to announce we have breakthroughs with the treatment of lung cancer and breast cancer --

 Male Voice:  [Speaks away from the microphone]

 Len Lichtenfield:  -- okay, all right.  Well, I do not hold you responsible.  It was an interesting moment.  Here it is.  They were at the press conference saying Avastin works in breast cancer and lung cancer.  And although the buzz was not as great as it was for the Herceptin study, they presented that day -- we got a late breaking abstract at the ASCO that year that presented the data - lung cancer and breast cancer.

Well, guess what happened on the way to the store?  The lung cancer approval went through but the breast cancer approval did not.  Scott has already appropriately pointed that out.  This is a blog in December of 2007.  And the reason that blog is there is because that was the day the FDA ODAC committee, the Oncology Drug Advisory Committee, said we are going to vote not to approve the use of Avastin for the treatment of women with breast cancer.

This is now two and a half years that the company had to go back and get additional data, additional information that the drug worked.  And why were they not going to approve it?  They were not going to approve it because although Avastin increased the progression-free survival, that is where the disease is the day you started the treatment to when that disease starts to show progression, was twice as long as the control arm of the study.

So you had an additional, I think it was about six months and that is a median.  And those of you who understand statistics, 50 percent above, 50 percent below, so some did not have that advantage but there are women out there who had a big advantage beyond that period of time - double, double, progression-free survival.  It did not demonstrate improvement of survival.

Now what I said on that blog was, well, you know, let’s step back for a minute because the commissioner used to be with the NCI.  When I said the NCI, he was the one who had the press conference.  But I understand the commissioner does not get involved in these decisions.  But the other part was in another drug called Tykerb.  The argument was it did not improve survival, but Tykerb, in fact, is another drug, lapatinib - used to treat breast cancer in women.  That drug has similar profile for response and yet the FDA approved it without demonstrating improved survival.  So I was sitting here saying, “Wait a minute.  You did it for this one.  You are not going to do it for that one?”  So stay tuned.  Eventually, it came up and they approved it.  And that is what this slide is about. 

Here is the FDA’s dilemma.  That was a conditional approval based on whether or not the drug demonstrates improved survival in a later study; the post-marketing studies that Scott has referred to.  There have been drugs, like Iressa, which is used in lung cancer, which helped a small number of patients.  Ten percent of the patients who received it had excellent responses.  It was pulled off the market because, ultimately, it did not show improvement in survival, plus some other issues about side effects and post-marketing studies.  So it is not all clear. 

To me, the overriding question is if the FDA had not approved Avastin and if you have doctors out there who are seeing women - and they are out there, I promise you - who are seeing dramatic responses to Avastin in terms of pain relief, improved function.  And if you believe that six months or doubling the time of progressive survival when your quality of life will be improved - by the way, they neglected to measure quality of life in the study, which was the key problem - if you have that, is not that a value? 

And if the FDA had said, “No, we are not going to approve it,” I think that would have been a very interesting test of what insurers and others would do, because up to this point, you could go on off-label.  Once you got past that point, maybe you could continue off-label, but the FDA position would have been a little bit more difficult. 

Well, here we are.  Where are we today?  Well, it used to be, “I’m a doctor.  I will do what I want to do.  Leave me alone.”  No more.  The reality is that individual physician - and I’m sorry to tell you this, folks - when you talk to oncologists in practice, and they are in a group and if they have not taken the time to standardize their therapies, there is a possibility, if not a probability - I have not done the study, so I do not know - there is a possibility that different doctors in the same group will use the same drugs in different doses and different changes in the dosage. 

It is not standardized to the practice level.  Remember I said when I started we do not have the information.  We do not have the quality parameters we need to know and we do not know.  So we cannot make those assumptions. 

Individual physician lack of standardization - there is no nationwide reference for you to form chemotherapy regimens.  There is no place I can go, press the button and say, “For colorectal cancer, I want to treat it.  How do I treat it?”  No generally, shall we say, arbited, something reviewed standard for that.  It is very hard to find.  What is then is now. 

And costs are becoming an increasing concern.  You cannot have drugs that cost - not the cost of administering, the cost of the drugs - you cannot have them costing $200,000 a year.  Two hundred thousand dollars a year for the drug and not have somebody on a combination, and have somebody paying attention to that.  That, my friends - maybe it is not a lot of money for the folks in this room - that is a lot of money.  That is a lot of money.  And Medicare, I promise you, I look at those numbers. By the way, the ESA numbers are down.  That was when all the constrictions are going on.  They started to tail down, but the other numbers are heading on the way back up on the chart, targeted therapy.  So it is sort of like the squishy balloon.  You hit it in one place, comes back somewhere else. 

Just to wrap up here.  National Comprehensive Cancer Network, if you do not know about it, find out about it, particularly those of you who are involved in regulatory and congressional issues.  This is an organization of 22 leading cancer centers around the country, and they do what I call living guidelines.  These are guidelines that are revised every year.  They are updated, and they are updated as needed, as evidence develops. 

It is significantly important.  They represent published studies, they represent strong evidence, they represent consensus evidence, and in my mind and those of others, they represent the standard of care in oncology.  In fact, CMS - for those of you do not know - had a demonstration project in 2005 and 2006.  They asked the oncologists, “Do you follow ASCO or NCCN guidelines in the care of this patient?”  And if you checked yes, you got a couple extra dollars.  The issue why it happened is another story.  You got a couple extra bucks for that patient visit.  They continued that through 2006, and then they dropped it in 2007.  

But NCCN said -- we do not have a lot of things in the medical profession that have what I call living guidelines.  I have been in this business a long time.  This is a place you can go.  And you can go, too, by the way, if you want to wade through the pros and the algorithms to see what the recommended treatments are for a variety of cancers.  It is very helpful advice and a helpful place.  But you will see here they have a Drug and Biologist Compendium.  And what they do on this Drug and Biologist Compendium is they will say what the FDA-approved use is and they will say what the expert consensus is about the appropriate use of particular drugs.  That gets passed, and that includes off-label utilization.  And I’m pleased to say - because we have supported this - CMS has agreed to adopt this compendium as a standard to measure effectiveness of drugs, and so do the insurers and the drug companies.  The insurers will look at this and CMS will look at it.  So here is a standard in the product domain and privately supported, it is generally accepted. 

Oncologists even in my small hometown, I work in Atlanta and live in a little town south of West Georgia called Thomasville, Georgia.  It is a small town.  Even the doctors in Thomasville, Georgia - I’m proud to say - when I had a conversation one evening in a dinner said, “We will not start treating the patient before we look at these guidelines.”  And hopefully, NCCM will come up with that chemotherapy regimen so they have something that is standardized. 

Let me just point out how things happen.  This is an abstract that was released last week.  It talks about Avastin - you can see - Avastin in the use of brain tumors, malignant gliomas and glioblastomas.  It talks about the results it had in recurrent disease.  But unfortunately, I do not want to get too much into the weeds on this because of the current press issues.  It is a very unfortunate situation that has been brought to the fore in the past 24 hours. 

That is an ASCO abstract.  It is the early information, but the latest information available.  You cannot see it, but I will tell you.  It says, “FDA use and NCCN-recommended use”.  You will see that Avastin is on that list.  I assure you if we did not have the opportunity to use these drugs off-label, we would be in a very serious situation when it comes to very lethal cancers. 

So with that, I thank you very much.  Obviously, it is time for discussion.  But it is a critical question, and there has got to balance.  We have got to figure out how to do it, how to get it right.  I think there are some solutions.  And I will look forward to the discussion.  Thank you.

 Cole Werble:  Len, thank you.  It was very helpful and encouraging hearing your words about the Comprehensive Care Network has new living compendia, because that compendia issue has been troublesome from back when it was started in the early ‘90s.  It was sort of thrown in there as a way to get payment for some drugs, and the compendia that existed back then were not particularly good.  So that was very interesting. 

Now, we have time for questions and answers.  I hope there are a lot from out in the audience.  To get things started, though, I would like to follow up on some comments that Len and Scott made.  And Jack, I hope you will chime in also.  But both Len and Scott provide some interesting perceptions on how FDA keeps up on new information in approved drugs, both good and bad.  I mean, how does FDA catch up with labeling?  Scott says they do not.  And how do they catch up when there is a safety issue?  Are they rapid in catching up with that too? 

So the FDA law, as part of the REMS, has a new look back process for products that are approved, that the post-market controls that are approved with them will be looked at, at some set date, and the schedule is set up, individually, for each drug that is approved.  I hope you all recognize that there had been a number of these first REMS approvals.  There have been about a half dozen.  And at least four of them have these look back schedules, so the drugs will be looked at. 

I believe, on almost all of the first ones, they picked 18 months, three years, and seven years.  This relates to the discussion of whether you actually have standardized follow-up reviews of drugs that are already approved.  Do either of you see that?  Or any of the three, do you see that as an opportunity to keep up with new approvals and also to catch safety issues?

 Scott Gottlieb:  I will start.  Yeah, I do not think that the fact that FDA does not keep up with the supplemental knowledge about a drug in a way that is reflective of clinical practice, I do not think this is criticism of FDA.  I think if you had anyone from the oncology division up here and said, “Does the label on these newer drugs define the scope of clinical practice, or what ought to be clinical practice?”  I think they would say no.  I think in these presentations you have many of the statements that the agency has made with respect to off-label prescribing.  That is especially true in unmet medical needs and things like cancer, where the oncologists even inside the agency recognize the value of enabling off-label prescribing. 

So they set a different standard than what the practice environment ought to be setting for itself through the kinds of groups that we are talking about here.  As far as FDA/AAA, I have read RPM, so I know that the journal sees it as creating a very different environment for enabling risk benefit balancing, and maybe giving the agency more confidence at the time of approval to embrace more uncertainty.  I think that that is absolutely untrue. 

I do not think the agency’s posture with respect to how much uncertainty they are willing to embrace at the time of approval is going to change with FDA/AAA.  The agency is unwilling to make any kind of concession that more rigorous tools, post-market, to require data to be collected or even more resources to have data streaming into the agency, post-markets, that you will recognize both side effects and new uses more quickly gives them any ability to titrate the binary decision to approve a drug. 

Now, the medical device side of FDA’s house will make that concession.  If you go to the center director at the medical device side of FDA’s house and you say, “If you had better tools to track and monitor devices, post-market, would that give you more confidence to embrace a little bit more uncertainty on the approval decision?”  They would say, “Absolutely.  If we were able to get information and quickly spot adverse events before they become catastrophic, we would have more confidence to maybe approve some products with a little bit more uncertainty than we do right now.”  If you ask the same question to folks in CDER, whether they are in a leadership position or a mid-level position, I have been through this exercise.  You ask that same question.  They will say, “No.  Our requirements for approval would not change.”

 Cole Werble:  Scott, did not I read some articles in Health Affairs by people like Mark McClellan and Scott Gottlieb, who said that things were going to change in the FDA, in exactly the way that they have not changed?

 Scott Gottlieb:  I do know which article I wrote, but I think that we ought to be, in this coming --

 Cole Werble:   Could change, not would change.

 Scott Gottlieb:  I think that we should be willing to embrace more uncertainty if we have better tools, post-market, so that in a case where, right now, if a drug adversely affects the liver, FDA does not find out about it until 10 patients show up requiring liver transplants two weeks after taking a drug.  That is something that the system currently recognizes.  What the system ought to recognize is 1,000 patients who, a week after they get a new drug on a routine blood monitoring test, have an incidental elevation of liver function tests. 

If we were able to move to that kind of system that spotted these problems while they were still simmering before they became catastrophic, would the agency have more confidence to make different decisions?  That is the question.  I think the answer is no.  I think it ought to embrace more uncertainty in that kind of environment, that kind of system.  But it does not. 

I think my comments with respect to the risk maps, in particular, in the article I wrote in Health Affairs basically said that they were going to use these tools to attenuate off-label prescribing.  Because in many situations, FDA might feel confident prescribing and approving a drug in a certain narrow setting where they have judged the benefits that outweighed known risks or unknown risks.  But they are very concerned about how the drug might be used in a wider environment, where there is less certainty about the benefits and some certainty about the risks.  So I think with risk maps, which try to regulate how drugs get prescribed by physicians, you can start to attenuate that behavior.  And in fact, you have seen risk management tools used in that regard. 

And Cole, you talked about fizzy fentanyl.  What is the name of it, the Cephalon -- Fentora.  It is an orally disintegrating fentanyl that comes like Alka-Seltzers in the mouth so that people call it fizzy fentanyl.  Anyway, that is a classic example, I think, where you see risk management tools applied to try to attenuate off-label prescribing.

 Len Lichtenfeld:  Well, only in the sense that my comments will be that this is a bifurcated issue.  There are certain drugs, which go into widespread use.  And the risk tolerance -- and I’m not an FDA person.  I’m just talking, this is me.  If you put a drug -- and the Vioxx situation really put a scare in a whole lot of people.  And the ability of risk assessment - and many of you in this audience probably know much more about that than I do - but it really set a very low threshold for drugs that are on widespread use.  Cancer medicine has traditionally been in a situation where we understand there are serious risks associated with the drugs.  The profile changes substantially, so their willingness to take risk in putting something like that on the market is greater. 

The other reasons that the ODAC committee turned down the approval for breast cancer was there was a higher risk of mortality in the Avastin-treated group.  You heard about Herceptin.  Well, there is a 4 percent risk of congestive heart failure in women who get Herceptin.  We had a 30-year-old who gets congestive heart failure.  That is a serious issue.  But nonetheless it comes out. 

I will say that I have been a defender of the FDA in situations where they had taken the chance.  And Iressa, I think, is a good example.  Although I do not necessarily agree with what is the reason in pulling it off the market, but the reality was they said, “We are going to go out there and put it out there.”  And I think that we could argue about whether it was good to take it off or not.  There were some side effects seen in Japan that were not otherwise anticipated.  They did not show up until the drug got to widespread use.  But let’s understand the risk tolerance in cancer medicine is a different ballgame than risk tolerance in the more typical medications.

 Cole Werble:   Well, I see that we have about five to 10 minutes left here.  Okay?  Good.  Thank you.  We just got a dispensation to run long here.  But I would like to turn this open to the audience.  I see a question over here.