DR. SCOTT: Well, I think the committee is relieved to hear that this is not for a vote but only for your discussion and thoughts. So what I want to introduce is just a proposed FDA operational definition of a possible new variant CJD case for the purpose of deciding whether there should be a quarantine or withdrawal of blood or blood products from such a possible case when information is missing that would lead to a firm diagnosis of new variant CJD in a blood donor.

This is just to summarize what has happened previously. I think most people here are familiar with it. That is, in August 1995, and then revised slightly in December 1996, the FDA issued a memorandum recommending deferral of all donors with CJD risk factors from donating that included family history in one or more family members, or if they were pituitary growth hormone recipients or had received dura mater.

And it was also recommended to withdraw all products, including plasma derivatives, if a donor developed CJD, had a positive -- strong positive family history with two or more family members with CJD, was a pituitary growth hormone recipient, or a dura mater recipient.

This was all revised and the revision was announced in late August 1998 by Dr. Satcher. And this revision was based on epidemiologic evidence. It was extensively reviewed, which you've already heard about, or at least has been very much alluded to, would show that there was no evidence so far of any transmission of CJD by blood products.

And this was supported by lab-based scientific evidence which showed at least a diminution of titer of the CJD or TSE agents in processing of plasma.

So you've already been through this today. Obviously, our concerns about new variant CJD is that there is a lack of experimental data showing whether or not blood can transmit this particular infection, and also we don't know much about partitioning during manufacturing of the new variant agent. In fact, we don't really know anything yet.

In addition, we do know, as Dr. Prusiner has pointed out several times, that the new variant agent is biologically different from the classical CJD agent, so we can't necessarily extrapolate all of the information that we have on classical CJD to new variant.

For example, he talked about the differences in the protein and its behavior, and we also know that there is enhanced expression of the new variant agent in lymphoid tissues compared with CJD. And we don't know much about its virulence or infectivity compared with the classical CJD.

And, of course, we haven't had time to get or enough patients or subjects or transfused people to get the kind of epidemiologic data that we have which tells us that transmission of classical CJD by blood or blood products at worst is rare and may not occur.

So, currently, the diagnosis of new variant CJD is based upon neuropathology, and these are the three most characteristic features -- numerous widespread kuru type amyloid plaques, which obviously can occur in a few other kinds of CJD but are quite common in new variant CJD; spongiform change, which is predominant in certain areas of the brain; and a high density prion protein accumulation, especially the cerebrum and the cerebellum by immunohistochemistry, and tonsillar biopsy may ultimately play a role in this diagnosis as well as analysis of prion glycoforms.

You can't see the top of this, but actually it's in your handout. And what I have there is CDC suspected new variant CJD case definition for use when pathology is not available. In other words, there isn't always going to be a neuropathological specimen to examine, or it might not be big enough, I guess.

And so we do need clinical criteria to try to tell if we have a possible new variant CJD case, and the CDC has developed such criteria and this is mostly based on the findings that are described by the CJD surveillance unit in the United Kingdom.

And I want to point out that this kind of list is going to be subject to change as clinical and diagnostic methods and experience evolve. However, the current CDC definition -- the suspected new variant CJD case would include all nine of the following -- current age, and, of course, we're talking about in donors for our purposes, but the CDC is also using this kind of definition for their own surveillance.

Current age, if alive, or age at death, less than 55. Since the typical age of a new variant patient is about late 20s, and the typical age of a classical CJD patient is about 65, this is one criteria that is useful. And new variant patients tend to have persistent painful sensory symptoms early in presentation and/or psychiatric symptoms.

I can go into this further if people want to know about it. But there were a couple of articles published in the Lancet from the CJD surveillance unit in September 1997, which goes into this in great detail.

In addition, the patient must have dementia and a delayed development of neurologic symptoms, particularly movement disorders, about a four-month delay. And, again, this is somewhat different from classical CJD in its course. They may have a normal or abnormal EEG, but not the diagnostic EEG, which is a pseudo periodic sharp wave that's often seen in classical CJD.

The duration of illness should be greater than six months. Again, this is in marked distinction to most cases of classical CJD which average four to four and a half months of duration. Whereas, the new variant case typically is around 14 months duration, although there is a spread.

In addition, routine investigations will not suggest an alternate diagnosis. And this is a criteria, really, for the U.S. There should be history of possible exposure to BSE; that is, consumption of local beef products as resident or traveler to a BSE-affected country.

And there is only two more. No history of iatrogenic exposures that are related to development of classical CJD, and, finally, of course, such a patient, if they had a prion protein gene mutation, it was associated with familiar CJD. That would not fall under -- that would not be a patient that we would worry about new variant CJD in.

Certainly, other criteria may be added, as I mentioned, in particular the CJD surveillance unit is expected to publish something about MRI studies, looking in great detail at certain areas of the brain which might be very useful in making the clinical diagnosis without neuropathology of new variant CJD.

Well, if we used all of those nine criteria to consider whether or not we should quarantine or withdraw a blood product in a case of -- a suspected case of new variant CJD, we might run into a problem.

And one of the possible problems is that two of these criteria are time-based, so one is the time course of disease greater than six months and the other is that a period of four months should have elapsed before development of neurologic symptoms but after the initial symptoms.

And it's conceivable that a true new variant case could come to our attention where this time has not elapsed. And, secondly, travel history and symptom history might not be available or they might not be very accurate.

So from the FDA point of view, what we have been considering is whether or not to lower our threshold for considering withdraw and quarantine of a product, where we don't even have all of the information needed for the CDC criteria for suspected new variant CJD.

So we have proposed the following that -- and, again, I'm sorry, the heading is missing. But that for such a case to be considered even as a possible, or I should say potential, new variant case, it will be a donor who had a physician's clinical or pathological diagnosis of either CJD or new variant CJD.

And the donor would be young, less than 55 years of age. And, of course, such a donor would not have risk factors for classical CJD. And that's what we would call a possible new variant CJD case. And I should point out that although we would include all three of these criteria, from the point of view of reporting to the CDC, we would want to ask plasma establishments and blood banks to also report donors who were young but had risk factors for classical CJD that came down with disease.

And the proposed actions for possible new variant cases with this low threshold of consideration by FDA for disposition of blood and plasma products -- the actions that we would propose would be an immediate investigation and review by CDC and FDA of all of the available case information, and followed by an expeditious decision by the FDA on a case-by-case basis as to whether blood products from such a patient should be withdrawn as a precaution.

So just in summary, obviously, this is already built in, that any definite new variant CJD case would result in quarantine and withdrawal of all products. In addition, we're proposing that suspected cases meeting all nine of the CDC criteria would also be quarantined and withdrawn.

And that criteria for possible CJD, the young age, the diagnosis of any kind of CJD, would trigger a rapid investigation followed by an expeditious decision about a precautionary withdrawal and quarantine of material.

So that's what I have, and I open it, then, to discussion or comments.

DR. ROOS: Thanks, Dr. Scott. So we're not asked to take a vote, but just to discuss these issues. Yes?

DR. NELSON: I'm concerned a little bit about the explanation for the age criteria, and I can see that this is very useful because the one thing you do know, when somebody gets sick, you can estimate what their age is. And so that's an easy -- you know, an easy early marker for a possible case that's not classical.

And I assume that probably the reason for the classical CJD patients being much older is that the incubation period is so long that they probably had an exposure much longer. But as this epidemic -- or as the -- if it's exposure to the BSE agent from the epidemic, it seems like over time this age criteria will probably change, and that the under 55 may no longer be a useful criteria 10 years from now or 40 years from now.

And I just wonder if Larry or anybody could comment on that.

DR. SCHONBERGER: We definitely agree, and it underscores the evolving nature of these diagnoses. All I can say is the age is an excellent and easy criteria for us to use now. All cases, as you know, in the world of new variant CJD have been under age 55. In fact, I think the oldest was -- I think the median age is like 29 or so, 28 at onset and 29 at death. So that's why that particular criteria came into existence.

However, obviously, if the epidemic should change and we should start seeing older cases, then, obviously, we would have to change.

There is some semantic problems. We actually investigate every case under 55. So, in a sense, all cases under 55 in the United States could be regarded as under investigation or possible. We have not used the word "probable," in part because that's the word they use in the United Kingdom, and they count those cases as amongst the cases of new variant CJD that we count.

The 40 cases in the UK, I think, includes one, is it? One probable? That was a case in a teenager whose brain tissue was unavailable for study. And they indicate that it's too early in the epidemic. Their experience is too small for them to be absolutely sure about that, but they're willing to -- at this point to call it a case.

And I've been told that with these new MRI criteria, and so on, that maybe we'll be able to call cases without necessarily having the tissue, depending on what they find the specificity and sensitivity of those to be. So all cases essentially under 55 right now are under investigation.

Plus, we have established amongst pathologists the concept that any case that has the pathology of new variant CJD, regardless of age, or even regardless of whether they've diagnosed it as CJD, should be reported. And those two would count as new variant even though they are not under 55.

DR. ROOS: Just a quick question, Larry. What is your timeframe of reporting, or what is the goal here? Obviously, with respect to these new guidelines, you want to identify these cases fairly quickly and make some disposition as far as blood products.

DR. SCHONBERGER: Precisely because we are looking at all cases under 55, I was encouraging FDA to encourage the blood establishments -- or the first to identify these cases at least, and that has been the history -- to report to us any case of CJD under 55.

Once we get that report, it may be very easy for us and very quickly making it -- to very quickly make a determination that we're dealing with, say, a dura mater case or a human growth hormone case. But then, another part of FDA will probably become interested in that.

So we think it's worth the blood establishments reporting all of their cases in donors. There just are not that many CJD cases that are going to occur among donors that the blood establishment is going to be able to identify that quickly. But if they do, we want it reported right away.

DR. ROOS: Just a quick question. So, I mean, how about if this patient donates to some large blood pool or has donated whole blood? It doesn't go back to the blood establishment. It goes to a neurologist, gets diagnosed, etcetera. What's the timeframe then?

DR. SCHONBERGER: Well, frequently, our experience with the withdrawals -- and I'll use the Utah case as an example as that came out -- we handled that very, very rapidly. But even handling it very, very rapidly, you'll find that huge, huge numbers of recipients were exposed to this donor's blood products.

So the withdrawal program is relatively inefficient, compared to what we just did, which was to get deferral criteria. And I think that's why it was important to try to be preemptive in a sense and have the deferral criteria up front.

The withdrawal procedure, even when you do it very quickly as in the Utah case, I would not encourage people to depend on that for considerable safety. What we will do is we will modify and ameliorate the situation. But it certainly won't eliminate even the majority of the risk.

DR. ROOS: I just think it might be good to publicize these new policies widely to the neurological community, so that they alert you, Larry, or the FDA quickly. The Utah case, in fact, was kind of a very aberrant case. It could be that there are other cases that get less sophisticated care. And if you really want to identify things in a timely manner, you obviously have to publicize the program and new policies to the neurological community.

DR. SCHONBERGER: Well, let me clarify that the primary group doing the surveillance on this are blood establishments. And if this group wants to recommend that blood establishments, you know, provide blood donors with cards or something that would, you know, speed up any type of reporting, that's possible.

The surveillance that CDC is conducting is not designed for that type of rapid turnaround or rapid identification in reporting. That's another weakness of the system and relying on this withdrawal system for tremendous protection of the population.

DR. ROOS: Peter?

DR. LURIE: My question/concern is whether or not requiring all nine of these criteria is too restrictive a set of criterion. I guess the data question that I have is: of the 30-odd new variant CJD cases in Britain, how many of them have met all nine of these criteria?

DR. SCOTT: Well, could I also respond to that question?

DR. LURIE: Yes, please do.

DR. SCOTT; I don't know the answer to how many have had all nine of those criteria, but most. However, the CJD surveillance unit has somewhat altered their criteria with time such that the current organization is similar to this but not the same. And most critically, they have gotten rid of the age criteria and added an MRI criteria. But this is not yet published material, and it's very recent. We just got that information on May 31st.

And I think the other thing to mention is that we weren't considering only using all nine criteria. But, really, that's the purpose of the third way, if I can say it, which is to have a very low threshold for identifying even potential cases and then to make a rapid decision on a case-by-case basis.

But what we're anticipating is probably what you're thinking, that not all of those criteria are going to be met, just due to a lack of information, time hasn't passed, we don't have material to analyze. And so I think what we're anticipating is that we would be -- we would err on the side of caution unless investigation showed us that it was most unlikely that this was a new variant case.

DR. LURIE: I'm still left -- I'm afraid after that answer, it -- which may be the best you can give. I'm still left with uncertainty. I mean, it seems to me that that is a basic question. And if independent of data that are unavailable for the reasons that you point out there are people who do not have myoclonus, or whatever, and they don't have the right time course of disease, etcetera, we might -- and they may be too restrictive.

I think, at a minimum, it would be interesting to find out the answer to that question, and that might inform us better.

DR. SCOTT: Right. I can also tell you that in terms of the course of the neurologic progression, they reported I think it was 14 or 17 patients, and three of them would not have met, for example, that criteria because they got their movement disorders before four months had elapsed.

So you're absolutely right. Likewise, it was the psychiatric. So we would not be using the nine criteria per se in a potential case, as including or excluding the possibility of withdrawal.

DR. ROOS: Yes. I guess I kind of agree with Peter that I might have felt more comfortable if all of the cases satisfied the criteria of suspected cases, plus others that then turned out not to have new variant.

In other words, you want to throw somewhat of a larger net to take care of a lot of the comers, especially when you only have 40 cases that have presently been identified.

DR. SCOTT: That's right.

DR. ROOS: Yes?

DR. BELAY: I just wanted to say that all of the new variant CJD patients in the United Kingdom meet all of this criteria. In fact, in addition, a certain proportion of classic CJD patients could also meet this criteria, all nine criteria. So by no means this criteria is just specific to new variant CJD.

The only criteria that we added was item number 7, which is a history of possible exposure. Again, even in new variants we get patients that would -- that would still be present, because most of them resided in the UK.

DR. ROOS: Yes, Will?

DR. HUESTON: Three thoughts. One -- if, in fact, a case meets the three -- the three criteria for definite CJD diagnosis, you don't need to go through the rest.

DR. SCOTT: That's correct, yes.

DR. HUESTON: Right. So some of the cases were identified because they met these criteria. They were defined without going through all of the rest of the history.

Point number 2, in terms of the nine -- and I just mentioned to Larry -- for all practical purposes, I think number 7 ought to be simply revised to say, "Resident or traveler to a BSE-affected country." The bottom line -- you do not know what you've eaten.

(Laughter.)

DR. HUESTON: You don't know to what you've been exposed. So it's -- the second thing is it draws -- I think it gives a false sense of security and directs, potentially, attention to the wrong products, because the average person thinks of beef as primal cuts of beef. And that's, at this point, the least likely of the sources of exposure, given meat products.

The third comment is that I personally am very concerned about the proposed -- this criteria of possible new variant CJD by FDA. And I have two major reasons for that. The first is that I see the potential for conflict arising between FDA and CDC, where FDA is stepping forward or making a pronouncement of possible new variant CJD, and at the same time CDC says, "We're still investigating; you know, it's premature."

And I think that puts the FDA in a very awkward position, and I think an inappropriate -- Larry is telling me that they are investigating 25 --

DR. SCHONBERGER: There's about 25 cases under 55 a year.

DR. HUESTON: So my fear -- here is my fear based on my experience. Item number 2 says, "Donor has physician's clinical or pathologic diagnosis of CJD."

DR. SCHONBERGER: They're not all donors, by the way. Very few of them are donors. Okay?

DR. HUESTON: Okay. Fair enough. But once you get a terminology like this established, my concern is that it's going to spread further, that people are going to say, "Well, the FDA would have called this a possible case."

Number 2 says, "Has a physician's clinical or pathologic diagnosis," it doesn't say anything about the physician. And no offense to my distinguished colleagues, but there are a number of physicians that are simply not in the position to make a clinical diagnosis or a pathologic diagnosis of Creutzfeldt Jakob. That has not precluded some of these same physicians from making a proclamation.

Third, I think that the public health and the risk communication implications of this are potentially massive. And having been on the firing -- you know, on the other end of trying to deal with these, you know, the press grabbing hold of a case and blowing it totally out of proportion and creating a great deal of concern, I don't see why you need another term.

I think you coordinate with the CDC, you coordinate your investigation when it comes back from a blood collection center that you have a donor less than 55 years of age, where you have some suspicion of Creutzfeldt Jakob Disease. You go through the same CDC workup, and you base -- on a case-by-case basis, you base your decision on that coordination with CDC.

DR. SCOTT: Right. So we would leave those products on the market if the patient hadn't had six months of disease, for example. You see, there has --

DR. HUESTON: I'm suggesting that you do it on a case-by-case basis --

DR. SCOTT: Right.

DR. HUESTON: -- in association with CDC. And you may decide to take action prior to meeting all of those criteria.

DR. SCOTT: Right.

DR. HUESTON: I'm concerned about putting forth yet one more term that I believe will be misinterpreted. It will create more misinformation than it will help clarify the situation.

DR. ROOS: Just so I understand, Will, the term is this possible new variant. So maybe it could just be stated that cases were under investigation at that point, rather than label it potential or possible. And I must say, I kind of thought FDA and CDC were working together on these cases. That was kind of my assumption. Okay. So -- Dr. McCullough?

DR. McCULLOUGH: I have the same concerns from the standpoint of the blood banking system. It isn't clear to me exactly when the process of the market withdrawal begins. But if it starts earlier than the resolution of the case by -- based on the nine criteria, what we have under the proposed criteria is someone that some physician says has CJD and is under 55 years of age.

And if something close to that triggers the market withdrawal, potentially involving very large amounts of plasma derivatives, and all of that sort of thing, I have a lot of concerns about that. I think those actions need to be much -- to be initiated much farther along in the investigation of the case. So I have the same concerns about these very minimal criteria.

DR. SCOTT: Well, if I could interject -- I think what I intended to convey was that those small, three criteria would trigger an investigation that the FDA would be involved in, but not necessarily a withdrawal.

DR. McCULLOUGH: I'm reassured if you can assure me the FDA wouldn't, from time to time, decide to start things sooner, which could happen, I think.

DR. ROOS: Yes?

DR. EWENSTEIN: I think we should also remember that these patients, whatever their subsequent diagnosis, may be the recipients of products that the FDA regulates, and not just the source of products. And so I think it's important to have a low sensitivity for the -- I mean, we talk about hemophiliacs never having been diagnosed with CJD.

Well, you need a low sensitivity to make sure that you're not missing that sort of thing. There are, obviously, other groups that are certainly in a high risk in terms of receiving biologic products.

DR. ROOS: I had a question. I didn't see any real criteria used related to the abnormal glycoform of new variant. And it was my understanding that all new variant cases had a specific electrophoretic mobility after the proteinase treatment. And why isn't that one of the definite criteria here?

In other words, if you did a brain biopsy that was normal, let's say, or looked pretty normal, or had, you know, just minimal changes, and you saw this distinctive glycoform, would that be adequate by British standards, or should it be adequate by our standards?

Larry, do you want to --

DR. SCHONBERGER: I don't know of any of the cases that don't have the definite diagnosis criteria -- that don't have that and have the glycoform alone. I've had it the other way around, for example, even with the Utah case. We did it based on a biopsy, and there was insufficient material, as I recall, to get the glycoform --

DR. ROOS: No. I had heard that it was -- it was -- it did not look like a BSE new variant.

DR. SCHONBERGER: No, I'm --

DR. ROOS: On the basis of --

DR. SCHONBERGER: No, I understand that. What I'm saying is we had an inadequate specimen for the glycoform. We were able to get the Type I protein fragment at 21 KV, which sort of ruled out the new variant. But we were not able to get the glycoform pattern, certainly right away. I don't know if he ultimately got it. I don't think he even ultimately got that.

Do you remember that, Ermias?

DR. BELAY: I'm a little concerned about adding this glycoform ratio as a case definition for two reasons. The first one is there is no standardized kind of methods that are being used by different groups. That the group in the United Kingdom -- namely, Collinge group -- would use a different criteria compared with other groups within the United States.

So that part of the, you know, method -- the immunoblotting or the Western Blot method -- has not been characterized or has been -- has not been standardized. And the second concern I have is there are other diseases potentially that could have the same kind of glycoform ratio. And Dr. Pedro probably can correct me on this. FFI, I think, has been reported to have a similar kind of glycoform ratio also.

DR. PICCARDO: Yes. Let me back up for a second. First, I agree with what you've said. If the standardization of prp res, Western Blotting, is -- it is still under discussion.

So the UK -- Collinge group -- has one classification, up to seven different forms of normal prp while in the UK. In the U.S., basically, there is a Type I and Type II that have been recognized. So that is under intense discussion as we speak right now. So I would not base the diagnosis on that. That's for sure. And even at the pathologic level -- let me see, I had to walk out for a second because I had to get a taxi, but -- so I have to ask you a question. You were talking about that Utah case, and you were talking about the biopsy, right?

So I think at this point in time for the pathologist to make the diagnosis we'll need the full autopsy. I mean, with a small piece of tissue, with a lot of spongiform changes, with plaques, even in that biopsy, even with florid plaques, I would not feel comfortable in making the diagnosis, because you can have rare forms of sporadic CJD in which you have a lot of spongiform changes.

And if you have a minimal amount of amyloid of plaque there, it will be florid, because it will be surrounded by vacuoles. So I think in order to make the diagnosis of new variant from a pathologic point of view, you need the full autopsy.

DR. SCHONBERGER: Generally, I agree with you. We were able in this instance, however, to show that it was not a Type II protein, but, rather, a Type I, which was -- which gave us hard data that was inconsistent with the new variant as reported in the UK. But generally, obviously, most pathologists are going to want the entire brain to deal with.

DR. PICCARDO: I'm not arguing against. All I'm saying is I think we have to be extremely careful. And the only way to be sure about all of this would be the full autopsy. And then work the -- the ratios, glycoforms, etcetera, etcetera -- I mean, we need more time for that.

DR. ROOS: Larry, the definition of suspected and definite -- this corresponds to the CDC classification at the moment or --

DR. SCHONBERGER: Yes. In fact, they had asked us to come up with this definition, and that's where that comes from.

DR. HUESTON: It's compatible with the Brits, too.

DR. SCHONBERGER: And it is definitely compatible with the UK, although I'm in fairly regular touch with Rob Will, and he tells me that they are changing their criteria and that's why I was emphasizing that people have to regard these criteria as something in progress. It's a model being made.

DR. ROOS: Good point. Any other questions? Peter?

DR. LURIE: Just to be clear, if any one of these nine criteria is not present for reasons of the examination not being done, like an EEG, or not enough time having elapsed, it will count as if it is, in fact, present, right?

DR. SCHONBERGER: Yes, that's right. We would not count the absence of information as being negative. So that's why if a person is alive at five months, that doesn't -- he hasn't really lived greater than six months, that doesn't rule that case out.

DR. ROOS: But it sounds like the action that might be taken by the FDA in a particular case is done on a case-by-case basis. In other words, we are leaving a certain amount of discretion up to them in their investigations, which I think at this point is probably appropriate, rather than putting every little detail --

DR. SCHONBERGER: I'm sure if Jay saw that we had five months, and that was the only difference, we'd be withdrawing that blood.

DR. ROOS: Yes?

DR. PICCARDO: I think we have to be very careful and very flexible with all of this. Setting the criteria now I think is good, as a working thing. But I think we have to be extremely careful, because in the unfortunate event in which heterozygotes nv will start developing the disease, they might have a completely different phenotype.

So this is just a work -- in my opinion, this is a working hypothesis, and we've set this criteria and we will have to modify that accordingly. I think that's the way to go.

DR. ROOS: It sounds like we are all in agreement about this being a good template to follow, and that maybe we shouldn't introduce a new term probable or possible Creutzfeldt Jakob, and that the FDA should look carefully and on a timely basis at these cases.

I would suggest that you do publicize these actions to the neurological community because I think they're the ones that probably are going to have these cases come to them, rather than blood banks specifically.

Yes?

DR. ROHWER: Ray, I just wanted to draw attention again to number 7. It seems to me like while that's very helpful in implicating a case, it shouldn't be an absolute criteria for putting it in this category because it eliminates the possibility of discovering cases which may arise de novo from other causes in our midst -- for example, this Utah case.

DR. ROOS: I agree. If there are no further cases, I guess I'm going to call this session to an end and thank the committee members and other discussants.

Tomorrow morning is?

DR. FREAS: Tomorrow morning we will reconvene at 8:30 in the morning. I ask the committee members not to leave anything on their desks. The hotel may clear off the table tonight, and we do not want you to lose any of your papers. Thank you. See you tomorrow morning at 8:30.

(Whereupon, at 5:43 p.m., the proceedings in the foregoing matter went off the record.