Questions and Answers - Moderator, Dr. Bernadette Dunham
DR. SENIOR: Looking at -- this question is for Dr. Baker, and it involves the concomitant vaccination slide. Here you have got 8 cases received vaccines. What you are saying there, just as a clarification question, you are saying that of the -- that this represents 8 cases in which there was a reported reaction?
DR. BAKER: Well, what I am saying there is that -- well, vaccination was essentially an exclusion criteria so that dogs were not going to be concurrently vaccinated when they were given ProHeart 6 within 1 month of the administration of ProHeart 6. So what happens is in these cases, 8 animals received, you know, ProHeart 6 and a vaccination within, you know, 1 month of the ProHeart 6 administration. And there was an event that was reported in those cases. And we are not attributing those cases to -- you know, we haven’t made an assessment whether it is due to the vaccine or due to ProHeart 6. We are just stating that these were 8 cases that received concomitant administration of ProHeart 6 and vaccination, and this is what we saw.
DR. SENIOR: Okay --
DR. BAKER: This is exactly what we saw.
DR. SENIOR: But the population of 8 comes from the group that reported a reaction, not --
DR. BAKER: Yes, sir.
DR. SENIOR: Yes.
DR. BAKER: Those are 8 of the 102 reports that we received of clinical events.
DR. HARTOGENSIS: Yes, just to add to that, we don’t have the information available on how many dogs received concomitant vaccinations with -- and ProHeart 6 that did not report an adverse reaction, so --
DR. SENIOR: Right. But you could conclude that because of the RiskMAP that the number of times that there is a concurrent vaccination administration would be relatively low?
DR. BAKER: That would be --- (away from microphone) to say if they followed the exclusion criteria of the RiskMAP, yes.
DR. SENIOR: So there is no way of giving a denominator to this? Okay, all right. Thank you.
DR. BAKER: You are welcome.
DR. DUNHAM: Dr. Poppenga?
DR. POPPENGA: Just a couple of questions. Do you have any data with regard to these adverse drug events compared to the number of times the animals have been exposed to the product? I mean, were most of these in pets that received the product for the first time?
DR. BAKER: That was one of the interesting things we wanted to look at, at what repeated exposures would occur, repeated exposures.
So we are assuming that animals that received it on Day Zero of the program now were at, you know, Day -- 18 months time, you know, whatever --- (away from microphone) -- Day 400, whatever is the time, you know, 18 months 30 days. They would have received, you know, 3 or 4 doses of the product every 6 months, so you can see that effect of repeated exposure. And, you know, as the RiskMAP proceeded, that is how we come to work with an element -- repeated exposure would become more of an element for us to evaluate.
DR. POPPENGA: Some of that data would be available before the RiskMAP, though, correct, to look at?
DR. BAKER: Some of the data before the RiskMAP was started?
DR. POPPENGA: Right. Wouldn’t you have some of that data perhaps?
DR. BAKER: That was only -- before the RiskMAP was started, it was just the typical spontaneous ADE reporting observational data that would say -- that the history of the animal would say I could have received 2 doses prior to that, yes. It could be the clinical history --- (away from microphone).
DR. McCHESNEY: Are you asking from the previous experience up until 2004? Did we --
DR. POPPENGA: Yes, overall. I mean, you would have some data to make some sort of judgment.
DR. BAKER: Right. Every report had a clinical history associated with it, some level of clinical history, so it is just how thorough the report is. And maybe Tatty would
DR. HODGE: Yes.
DR. BAKER: -- you know, how often.
DR. HODGE: Sometimes that is captured, and again, you know, we are looking at previous Fort Dodge captured information.
I think going forward already we can see additional data that we would like to collect to get more information on some of the cases post-relaunch. It was indicated that maybe this was the second dose an animal had received or probably there weren’t any third doses, but it was lacking in some of the cases so it wasn’t available to sort of make a decision about.
But typically for any of the products, if they have previous exposure to it, we would try to capture that, and what was the event, so in other words we are capturing de-challenge and re-challenge information.
So, for instance, with a dose of a flea and tick product, if they have had monthly doses for 13 or 14 months with no problem and then on this particular 15th dose, that is weighed in in terms of trying to figure out, is this product-related or is it something else going on with that animal? That doesn’t mean that it can’t occur. Some events do occur once and then may not occur again with repeated exposure.
So we -- when at all possible, we do try to capture that information.
DR. POPPENGA: Just one more question with regard to the 90 out of the 132 that lacked baseline data. Do you have any idea of what the typical compliance is for getting that baseline data and how that compares with this number?
DR. BAKER: Well, that would just be a -- this is just a sample, of course, of what would happen with all the doses distributed by the -- you know, if you made an inference from that that and said, “You know, this is yet again what poor compliance we would expect on the level of poor compliance that was projected into the rest of the animals that received it.
That is why we want to reemphasize the importance of getting the baseline information. So if you have -- especially if you have repeated exposures, based -- you know, Day -- Zero Day, 6 month-day, 12 months, and we will have some baseline to compare these animals.
DR. HODGE: Yes, and that is another good question. Looking at the subset of the adverse event reports, you can infer to the larger population, but oftentimes it is a unique population as well.
We do know that there were activities that were -- veterinarians were trying to be compliant with the blood work requirement, but they were doing things like taking a baseline sample and freezing it. And then if an adverse event should occur, then they were going to take another sample and unfreeze the original for comparison.
So we do know that all kinds of outside-of-the-box compliance in a way was happening. But as far as real, true compliance with the overall population, we don’t know.
DR. DUNHAM: Additional questions? Dr. Apley?
DR. APLEY: Thank you. Dr. Baker, I had a couple questions related to a slide set. One is you mentioned that ADEs are substantially underreported. I think it is something we probably would assume. But is there a dataset of something to support that or has there actually been a study to give us an idea of how much underreporting or --
DR. BAKER: There is -- I don’t have the -- some of the studies that I could bring with me, but I know that it is either they have some pretty good, some different studies that were done for different publications that talk about that concept. But I don’t have them with me.
DR. APLEY: Okay. The ALT liver enzyme elevation slide, those 5 cases with mild elevation, were those ones that had pre-administration samples? Were they ones that just sampled during that time, had an elevation?
DR. BAKER: I would have to look at these individual cases again to see exactly. But I don’t -- we looked at that and --- (away from microphone) --- look at that --- looked at those cases and --
DR. APLEY: To me --
DR. BAKER: -- I am thinking they did have -- my recollection is that they did, but ---
DR. APLEY: Yes, because if they don’t, it could just be a fact that those animals are 2 standard deviate -- more than 2 standard deviations off the mean, and even if there is a change, you almost need a clinical pathology to --
DR. BAKER: Oh, absolutely --- (away from microphone) --- elevated liver enzyme elevation is. You know, it is a clinical medicine discussion about looking at liver function and looking at ultrasounds and doing, you know, a basic, you know, evaluation of what that means clinically. And you are absolutely right.
DR. APLEY: I have got a couple of others I had. Oh, so I go back to this sheet.
So in the discussion of the things that would cause the RiskMAP to be put in place, and this may be generally for the panel, you said it is either -- let me get it right -- either 1 of type of reaction or frequency. So is -- was the impetus for applying a RiskMAP to ProHeart 6 1 of adverse types or like -- not adverse, atypical types of reactions or number of reactions?
DR. BAKER: Well, you look for, you know, like rare events like the -- you are looking to see if you can uncover any more rare events on top of that, so you have a large number of, you know, animals that are being evaluated with that RiskMAP. So we didn’t detect any new signals. So we just strengthened (away from microphone) --- the -- you know, the anaphylaxis/anaphylactoid understanding that we create that level of concern about the ALT elevation.
DR. APLEY: I am actually going back to the original call for a RiskMAP. Was it because of seeing different types of reactions, or the numbers?
DR. HARTOGENSIS: I think I can address that one.
The -- I mean, by and large, our biggest concern -- okay, so by and large, our biggest concern following the recall, and due to the recall, were the reports of anaphylaxis and death, which I think is really well reflected in the RiskMAP. And a lot of dog owners and veterinarians just really were not prepared to deal with the anaphylactic reactions, which is one of the reasons why the RiskMAP really heightened awareness of these anaphylactic reactions and hopefully prevented unnecessary consequences from the anaphylactic reactions.
We were also seeing the liver lesions which cropped up and we included in the label and some of the other hematologic issues, immune-mediated type events.
By and large, most of the adverse events that we were noting really could be attributed to allergic-type reactions, and the RiskMAP was put in place to address those allergic-type reactions and minimize the risks associated with the product and allergic events.
DR. BAKER: Okay. And actually, I guess, the -- you know, the thing from the one slide where I shared the actual signals of concern from 2005, that is what we would use to -- you know, we took a re-look at those signals which Martine just laid them all out, but that one slide lists those actual signals that were discussed at the 2005 meeting.
DR. APLEY: So Dr. Hodge’s statement about -- and your statement, too, Dr. Baker about the perils of taking kind of nebulous numerator and denominators and then comparing between drugs, you know -- there is peril in doing that.
But at the same time, I am curious as to how the anaphylactoid reaction rate, observed rate, probably not the true one, would compare with other drugs. And also, you may not have the answer here, but I think about compare with like anaphylactic reactions reported to the Center for Veterinary Biologics with canine vaccines.
What I am after is not to put them hard on the chart against each other, but some type of context for a risk, because what I am thinking is we are talking about risk, evaluation of risk, and what is acceptable risk versus benefit, so I am trying to put the risk in context with other things we put and develop.
DR. BAKER: But, Mike, that is a great segue into what the FDA is doing for -- it is called the “Sentinel” initiative, where they are actually doing active surveillance where they are looking at databases that have exposed and known exposure of all patients, you know, human patients, and we are looking at it for veterinary medicine as well, to find, you know, databases that have a lot of wellness information in there so we have both exposed animals and unexposed animals, so then we can start to ask those questions and come up with some calculations that have known exposures in the denominator.
But that is called the Sentinel initiative, and there is a lot of discussion on the FDA.gov website about that program, but it is called -- it is an active surveillance system where you are trying to answer those questions. You know, you have a hypothesis and you want to have an answer. Well, you need to go into a database where you have known exposure in that, you know, you have to know all the denominator elements to be able to answer that question, and that is what that program is about.
DR. HODGE: But to be more specific for your question, and it is a great one because really the adverse reactions to biologicals and the known risk of anaphylaxis is extremely great with vaccines, the difference being I believe most practitioners are comfortable with that. If you were to present to them that there are pharmaceutical products that also have allergic reactions associated with them, whether they be anesthetics or other injectable products, they are less likely to be knowledgeable about that.
So -- and then you have to take into account the fact that many of these products are given in -- with concomitant products, so to actually pinpoint which products cause that particular reaction, whether it is hives or urticaria under anesthesia when they receive multiple products, pre-anesthetics, or whether it is -- which particular vaccine in the instance where they are given concomitant vaccines at the same time.
But you are right. In terms of after a while of looking at these, while you are not using a report rate or an incidence rate, you get an idea by comparison, looking at it knowingly, the flaws of such rates, that vaccines are much more likely to cause allergic reactions and then comparatively some of the other pharmaceutical products with greater or lesser reactivity.
DR. APLEY: One last question and I will yield the mike up. This is ignorance on my part. There is a monthly product still available with this active ingredient?
DR. BAKER: Yes.
DR. APLEY: Yes, there is?
DR. HARTOGENSIS: With --- (away from microphone)
DR. APLEY: With Moxidectin?
DR. BAKER: No longer available?
DR. HODGE: No longer available.
DR. APLEY: It is no longer available, the
topical or --
DR. BAKER: (Away from microphone) ---
DR. APLEY: But, I mean, is there another Moxidectin monthly product available, or recently available?
DR. BAKER: No.
(Chorus of “yes”)
DR. APLEY: Okay -- I am sorry. I am in clinical pharmacology, but the daily shift of heartworm prevention of the products I have lost track of, but -- so my question in asking that is, is there a way to look at the monthly administered products with this active ingredient versus the ProHeart 6 method and look at -- do you have a comparison of adverse events between those 2 methods of administration of the product?
DR. HODGE: Yes -- and the truth is we always do that. The Moxidectin oral tablet, I believe, and we are so new to this game, but I believe it is not on the market currently. So there are no other. There are lots of oral, other oral, heartworm preventatives and spot-on heartworm preventatives and we compare, you know, what we can, based on the product class. And we do know that as a class, some of these reactions are typical. But there is no direct comparator to the oral Moxidectin and the injectable Moxidectin.
DR. DUNHAM: Just for one moment, I am going to ask Dr. Ann Stohlman to come to the microphone and she will just give you an update on the products that are available.
DR. STOHLMAN: (Away from microphone) -- there is a topical ---
DR. DUNHAM: Thank you, Dr. Stohlman. Dr. Apley, further questions? You are okay? Okay -- thank you.
DR. McCHESNEY: Ann, is that an FDA-regulated product or an EPA-regulated?
DR. STOHLMAN: Yes. It is a combination of --- (away from microphone)---
DR. McCHESNEY: Thank you.
DR. DUNHAM: Dr. Griffin?
DR. GRIFFIN: In the process of collecting the data for the 102, was other data collected, such as breeds of dogs, or any more detailed information about the dogs, including what types of vaccines in the 8 cases?
DR. BAKER: Yes -- yes. We have --- (away from microphone) breeds were represented in what we call the general, you know, representation of the breeds in the population that --- reported on the one side with a snapshot. Slide number --
DR. GRIFFIN: I saw that. Just kind of -- it seemed like glossed over. I mean, I --
DR. BAKER: No, we -- well, I thought we pointed out that it was nothing other than we expect of in just the general occurrence of those breeds in the population and then the gender, the gender comment.
DR. GRIFFIN: And no information about the
DR. BAKER: Yes, yes -- the vaccines, there was different -- you know, injectable vaccines and intranasal, bordetella. You know, it is all in the case reports, but -- yes.
We don’t have the specific lot number, though, of the vaccine like the label that you would get as a practitioner and stick one in your medical record. That is not on the ADE report.
DR. GRIFFIN: And that would include any adjuvants that were used with those vaccines?
DR. BAKER: Only if you had the label that typically practitioners stick on the medical record when they vaccinate, but that is not a part of the ADE report, so we could, you know, hopefully -- you could get that from the medical records if you did a call-back and tried to see if you could get that information.
DR. DUNHAM: Yes, Dr. Murtaugh?
DR. MURTAUGH: I would appreciate some additional information on -- related to the scientific value of the data that was presented. The -- since there is a high level of subjectivity in the reporting, are there differences in veterinary reports versus public reports of adverse events for the different indications that you showed, such as death versus anaphylactic responses or liver involvement? And then -- well, okay, that is the first question.
DR. BAKER: Well, you know, I would think maybe if you had a mild case of pruritis and of vomiting and your veterinarian or the owner didn’t -- just dismissed it and didn’t bring it to the attention of the vet or however didn’t -- you know, as opposed to something more dramatic that, you know, clinically dramatic, that caught everyone’s attention. But -- so maybe some of the more subtle anaphylactoid cases maybe because they didn’t rate. But hopefully the RiskMAP raised that level of awareness to look for these things because it was -- you know, it was in the educational materials, the communication materials, to look for these events and let your veterinarian know, so we really emphasized to have that enhanced reporting ---
DR. MURTAUGH: What is the period of time after administration that adverse events are recorded?
DR. BAKER: It is any time. It is once the drug -- you know, once the drug is administered ---
DR. MURTAUGH: So veterinarians would not be likely to report adverse events that occurred after the dog left the clinic?
DR. BAKER: Unless the owner, you know, called back in and said -- you know, called the receptionist and said, “My dog just did this or that or whatever,” and then the vet can say, “Well, bring it back in and we will take a look at it” and that sort of thing.
DR. MURTAUGH: So then, what is the time frame of adverse event reporting after treatment in your case summary for all of the events?
DR. BAKER: It varies. I mean, we have the exposure, the data ---
DR. MURTAUGH: I mean, to me it makes a difference if we take the death events, if that was reported a year after the treatment --
DR. BAKER: Right.
DR. MURTAUGH: -- it is less significant than if it occurred within three days.
DR. BAKER: Right, absolutely. That is part of the causality assessment you could do on that. When you drill down into the individual case and look at that case which they are included for you to look at, those cases are there in your briefing package that you can look at the actual report and see what we looked at.
DR. HODGE: I could explain a little more --- (away from microphone) collected. So --
DR. DUNHAM: Could you bring the microphone back over, please? Thank you very much.
DR. HODGE: That -- the 800 number is available for people to call at any time. Generally, adverse events of the allergic type are reported quite quickly after they occur because they are calling at that moment or requesting sort of information on treatment and so forth.
But we record not only when it occurred but also when it was reported, and based on that data, we develop a case description for each type of event.
So the digestive events may occur within the first 3 to 4 days; the allergic events, as John Baker said, generally occurring within the first 24 hours, more likely within that first hour. That doesn’t mean they were reported to us in -- within that 20-minute period. They may have been reported -- sometimes we get reports 2 years after the fact, or a year after the fact.
But all of that goes into the database and is trended, along with age, weight, breed, when it occurred, when the first administration was, how long the event lasted, when did it resolve, what was the resolution, what other concomitant products were given? So all of that is collected and trended when that information is available, so -- but reporting practices do vary --
DR. MURTAUGH: Thank you.
DR. HODGE: -- and you will tend to see more reporting of serious events.
We have -- there is something about seizures. They are serious, difficult for the owners to watch. They are almost always reported. So you will see a clinical sign of seizure reported for almost every product in a database. So there is a preferential reporting, depending upon the reaction. But all of that kind of information is taken into account when you trend this kind of information.
DR. MURTAUGH: So that applies to the capturing of events since Pfizer took control in December of 2009?
DR. HODGE: Yes.
DR. MURTAUGH: And is it fair to say that that is a substantial change in the data capture from the period of time from which the data that are being presented today represent?
DR. HODGE: Okay, so we looked at the Fort Dodge data from the beginning of the RiskMAP until today and the information was complete. There are additional fields that
-- actually based on questions and so forth, we may enhance and want to ask. I found the information to be fairly complete, realizing that Fort Dodge as a result of the previous situation did enhance their pharmacovigilance processes quite a bit.
DR. MURTAUGH: Thank you. Then I had one last question about the data analysis. It looked to be primarily a numerical presentation of results. And was the recommendation relating to the liver enzymology based on just the numerical differences that were observed or was there a statistical analysis that that recommendation was, or question raised based on?
DR. BAKER: So you are back to that slide where you try to understand exactly -- the liver enzyme elevation slide?
DR. HARTOGENSIS: I just wanted to add one thing to your previous question. The Client Information Sheet actually details and explains how long the product will remain in the body and that -- the purpose of that Q&A there is to call attention to the fact that the product is -- does remain in the body for 6 months and may potentially spur a client or a dog owner to report during the course of the treatment, so --
DR. MURTAUGH: I apologize for continuing here, but that raises another question, because I think in the literature we received there was an indication that the ProHeart 6 will provide protection for up to 2 years after a single administration, which suggests that in fact that it will -- it may be maintained at lower levels but still present in the body beyond 6 months.
DR. HODGE: Okay. You saw my eyebrows go up. It would never be our recommendation to say that it was effective out to 2 years and there are negligible amounts --
DR. MURTAUGH: I agree.
DR. HODGE: -- in the body at 6 months. So I would like to see what you are seeing.
DR. MURTAUGH: I will look for it. Thank you.
DR. HODGE: Thanks.
DR. BAKER: Do we answer those --
DR. DUNHAM: John?
DR. BAKER: The liver? Still I could discuss the liver question? The liver ---
DR. MURTAUGH: Yes.
DR. BAKER: Okay. And, Mike, I think there was -- I looked back at the liver cases real quick. I think there was 1 case of elevated liver enzyme that had a baseline that was elevated. That was before the -- that we had baseline information on 1 of those 7. And, you know, you are looking at the conclusion: “No apparent association with ProHeart 6 administration.” That was just a -- no statistical, specific statistical, analysis was done to make that statement.
DR. MURTAUGH: Pardon me?
DR. BAKER: No specific statistical calculation was done to make that statement.
DR. MURTAUGH: Thank you.
DR. DUNHAM: Dr. Apley?
DR. APLEY: 1 follow-up, if I could, on the -- back on the seizures. Could someone tell me, if I were to look at 300,000 dogs over 18 months, how many spontaneous new cases of epilepsy I would run into? Is there data on that out there?
DR. BAKER: That gets back to the active surveillance question, you know, the Sentinel mission. That is why the FDA is really making it an initiative, to get -- to develop those databases where you can query where you have known exposure, wellness cases, and all that sort of thing to kind of figure out, you know, that kind of background incidence information.
DR. HODGE: Yes.
DR. BAKER: But that is really -- we are really working hard on that.
DR. APLEY: I think I am going to be on PubMed over lunch trying to see if I can see anything because it ---
DR. HODGE: What -- I will just add to that. One thing that is really missing in companion animal medicine is baseline natural history of disease, if you will. Some of the -- as John said, some of the databases available, most notably through Banfield, make an attempt to give us some kind of prevalence incidence of some of these diseases. And that is why it is so difficult to calculate what we call “attributable risk,” which is you have a baseline of disease in the population and what is actually attributable above that to a given product? But since we really are lacking in that baseline natural history of disease in many cases --
Immune-mediated hemolytic anemia is another one. Vaccines tend to get blamed for that, as well as other products, and we are just beginning to collect data on what is the baseline occurrence of idiopathic, if you will, immune-mediated hemolytic anemia.
DR. APLEY: Maybe one way to get a rough feel for it is to look if there is kind of a baseline noise level of seizures across multiple drugs that you get ADEs for. I know is there always -- does there tend to be always a smattering of seizures across all these drugs? And if so, that might give us some idea of baseline. Obviously, when you start looking at huge population data that is observational rather than a prospective study, you are into all kinds of confounders.
Back to the ALT, I always like to ask the question of positive and negative predictive value of those tests in a very low-incidence population, which becomes -- it is like heartworm testing in cats. All of a sudden you have a test that is basically -- that you are unable to interpret because of the potential predictive value implications in a low-incidence population. So that is why I keep asking back to the ALT and trying to get a baseline for: Are these even above normal occurrence in that many dogs?
DR. HARTOGENSIS: I just wanted to add to -- on the comment and question on seizures. The seizure signal is actually nothing really new. We saw that signal, I believe, in the pre-approval studies as well as before the 2004 recall, and we consistently did have some reports during the RiskMAP, so that particular signal seems to be fairly consistent.
DR. DUNHAM: Yes, Dr. Altier?
DR. ALTIER: I have a question about some of the warnings that are currently on this product. This is just a clarification. The warning about the concomitant use of vaccinations with the product, that was instituted because of the need to reduce confounding factors in investigation; there is no reason to believe that the use of the vaccine with the product would have exacerbated adverse reactions. Is that correct?
DR. HARTOGENSIS: That is correct. However, it would be very, very difficult, using the data that we get, to really tease that out unless we had specific studies that are designed to tease out that information. But that -- yes, that is correct.
DR. ALTIER: And similarly with the other warnings, one was for blood work. There was no blood work abnormality that would have precluded a veterinarian from using this product. It was simply a means of gathering data before the product was administered?
DR. HARTOGENSIS: Well, the warning that -- against using the product in dogs with pre-existing disease, that sort of goes together with the blood work because we really want the veterinarians to be treating healthy patients and not to be treating -- you know, because there is obviously these liver signals that are on the label.
So, for example, a dog with elevated liver enzymes for baseline may not -- you know, depending on that -- you know, the veterinarian, the patient, et cetera, all those parameters may not be appropriate for that particular patient. So, yes, that --
DR. ALTIER: But -- so there is a biological reason for that. But there was no specific recommendations given to veterinarians about what abnormalities might prevent them from treating a dog with this?
DR. HARTOGENSIS: No, not specifically.
DR. ALTIER: And the age, the maximum age recommendation, that was also just to reduce confounding factors in older dogs?
DR. HARTOGENSIS: Correct.
DR. ALTIER: Okay. Thank you.
DR. BAKER: And what happens with that, with the age factor, is that when animals stay into the -- once they are entered into the RiskMAP, at 7 years of age, well, then they are going to be -- you know, or 6 years of age, they are going to hit the age limit, so we had to adjust the RiskMAP to allow for animals that were receiving the product to “age out of it” to stay in the RiskMAP. So that was a modification that was made.
DR. WOLF: I just want to follow up on Dr. Altier’s comment because I think you are driving at the right point, that you are talking about -- so there are 8 concurrent -- or concomitant -- vaccinations, so thinking of -- and these are extremely rare events. I am just thinking of what we were even talking about on 300,000 cases. So, if anything, you might expect that your -- if there is any kind of noise with that, some of these are actually caused by a vaccination and not by PH 6, that it may be a little bit lower.
And getting at your question number 1 here is: Can you make a modification to the RiskMAP to have the language be referring to a precaution in knowing that? Basically, if it just says “use with precaution,” you are basically telling clinicians to use it.
They are not -- I don’t know what “precaution” actually means other than maybe that they will have some additional follow-up, but they are still going to probably give the vaccination with -- they are going to give the PH6 with the vaccination. I am assuming that is probably correct? That would be the behavior change you would see?
DR. BAKER: Also, things like, just common sense things like when the animal receives the vaccine in the ProHeart 6 had to wait in the waiting room for an extended period of time just to watch it.
DR. WOLF: Stay around, yes ---
DR. BAKER: Stay around. So that is -- you know, typically with a small animal, when I vaccinate it, I just want to have it stay around, for the toy breeds, just to watch it.
DR. WOLF: But they would probably proceed to go forward with it.
I guess the question I was going to ask is: Even if you are looking at these rates of adverse events, you know, which are extraordinarily low, and again I am coming from general medicine and not from -- this is not my area of expertise, but I know that from med safety in humans that we have a whole lot of medications, including over-the-counter products, available and approved by the FDA that have much higher rates of, you know, nasty stuff like acute liver failure with acetaminophen.
But if you -- would you change -- would you do anything different with the product if you did associate with this current level of rates of rare, but serious, adverse events? Would you do anything different with the product beyond just wanting to potentially get more data by not having it be able to be administered with vaccination?
DR. HODGE: Go ahead.
DR. BAKER: Well, you know, the RiskMAP provides that coverage if we introduce vaccinations as a concurrent practice element with ProHeart 6 administration. Well, you know, with the RiskMAP with that enhanced reporting, if that -- if there was a problem start to develop, we would hopefully pick it up quickly since the -- you know, the reporting, those reports, and we have a very fluid communication with the sponsor that they let us know what is going on. So that -- you know, the advantage of the RiskMAP is provide that oversight.
DR. WOLF: I am just trying to get at this behavior which Dr. Altier pointed out. The only reason that you would want to not have that language in here is that you would think that in the future you might get better data that would better isolate the association between these adverse events and ProHeart 6?
DR. HARTOGENSIS: Well, not only that but, as you heard before, we heard from a lot of veterinarians and dog owners that they really were not able to -- in these very, very heartworm prevalent areas, they were not adequately able to protect their animals from heartworm disease without the use of the product.
So it is really -- it is also an availability issue, is that some practitioners and the clients to have a difficulty doing the ProHeart 6 and vaccinations because logistically it poses a difficult situation, especially if the dog has -- if the client has a difficulty getting the dog in the car and all of those logistical issues. So that is -- you know, it is also an availability --
DR. WOLF: But that adds to the benefit, so like the benefits making the change?
DR. HODGE: Yes.
DR. WOLF: So I guess I am trying to figure out -- I am looking at this, and again I am really driving at this question because even with the limitations of the data, which clearly there are limitations, why wouldn’t you want to go with this if the benefits outweigh the risks because it has a protection, a protective effect, for heartworm and the data is not really giving you a signal otherwise, other than that there are some very rare occurrences that you knew about already with these signals? And the only benefit that you would get by not going that route is that you would still have access to potentially get better data, but you don’t know that you are going to get any better data.
DR. HODGE: Right.
DR. WOLF: So --
DR. DUNHAM: Thank you, Dr. Wolf. Additional questions? Dr. McKean?
DR. McKEAN: Perhaps I missed this, but ADEs can be reported by either consumers or by veterinarians. Do we have a -- do we know what percentage are represented in this slide set, one to the other?
DR. HODGE: I think the majority of these were reported by the veterinarians. I think in general for most products, I think it might be 30 percent pet owners and 70 percent veterinarians, but I think it varies quite a bit what the product might be, whether it is a flea and tick or an injectable product. But that is roughly right.
I did want to mention the vaccine issue again and just that -- to remember that in international markets there is no restriction on vaccines and they have experienced very low rates of adverse events. That doesn’t mean that if we remove that restriction, things won’t change, but we do -- we would have to make even a greater effort to capture exactly what the vaccines were and the time periods and the recovery and outcome.
DR. DUNHAM: Dr. McKean?
DR. McKEAN: I want to have a follow-up, then. It appears from what you presented, Dr. Baker, that this listing that you gave was an enumeration of reactions that showed up after the use of the product at some extended period of time. The seizures I have got ranging from 4 to 60 days.
What -- when you enumerate this data, what do you do to attribute that to one product versus other products that might be in the history? How do you do this so that when you get this list, we have an idea that it is related to the product that it is assigned to the list?
DR. BAKER: That is an excellent question. It is part of the causality assessment you would do on that individual case, looking at, you know, how many medications were given, looking at the kinetic behavior of each compound that was given, the known -- you know, the known pharmacologic profile, toxicological profile, of each, you know, product that was given.
So if you have all those different elements identifying, you know, the drug -- you know, this drug lasts for -- half-life is 14 days versus 1 that has got a, you know, 1-day half-life, and, you know, you make those scientific, you know -- you bring that science into it and look at re-challenge, de-challenge, and the clinical history of the patient, pre-existing diseases. That is the whole -- you know, all that information is taken together as part of the causality assessment that is done. Does that help?
DR. McKEAN: Well, that does help, but then I want to be clear. The materials that you presented in the slides, those have all gone through that process and you have some level of confidence that there is a causation rather than just an enumeration related to the ProHeart 6 ---
DR. BAKER: Everything that was in the calculations was based on a possible, or probable, causality. And then we even were even more transparent in the actual cases that -- for every event that was seen in the 2005 VMAC, we were totally transparent and laid it all out there so you could see all the cases irregardless of causality in the individual slides. But the calculations were based on cases that were possibly, or probably, based on our causality assessment algorithm that we use. That is what was used to determine those actual numbers.
But even -- you know, but we gave you all the deaths here, all 8 cases, but, you know, they all didn’t -- only 2 of the 8 deaths came out as a possible -- you know, the causality is possibly drug-related. But we put all 8 out there just as complete transparency so you could look at the case reports yourself.
DR. McKEAN: I want to continue on specifically with the seizures. How do you equate the use of the product with seizures that occur 28 to 60 days after that product use and take it back to the actual use of the product?
DR. BAKER: That is an excellent question. I think, you know, looking at the half-life of the drug and trying to make some decisions based on how long the drug is in the system and at what levels.
You know, the peak of the drug, the C-max and T-max, those sort of elements help you give, you know, some relationship between, you know, exposure and the event that you see. But we put them -- you know, put them all out there so you would see those onsets for yourself, so you actually saw what we saw, and just for transparency and completeness, that you can see that.
Now, if you are asking my opinion, that -- you know, I wasn’t going to give you my opinion, but that is -- they were all given a possible causality based on our assessment, those 4 cases.
DR. WOLF: Could I ask a quick question of Dr. Hartogensis?
DR. HARTOGENSIS: Yes, Dr. Wolf?
DR. WOLF: In the presentation, you mentioned about updates to the Client Information Sheet. That makes sense, but -- and I noticed that in an earlier presentation one of the other barriers to use was that they didn’t like -- either it said there was a note that along with having, you know, the multiple trips back to the vet, that they didn’t like to sign the form off or have to be written confirmation that they received the education about the use of the medication, the precaution. Would that still be a requirement in the current -- in what is being suggested in revisions of the RiskMAP, that they require the family member would have to sign off?
DR. HARTOGENSIS: That is our recommendation. However, we leave that open to the panel to make -- provide feedback on that as well. But CVM feels that that is an important part of the whole circle of the RiskMAP in getting that feedback and communication.
DR. DUNHAM: Yes, Dr. Mathew?
DR. MATHEW: Just a general question about how CVM looks at risk. In one of the earlier presentations, it was mentioned that the World Health Organization levels of .9 I guess per 10,000 is considered rare. Do you apply the WHO classifications to how you look at risk in animals as well?
And I ask that question because I am -- I don’t know, but I would assume that the reporting of adverse effects is going to be a higher percentage in human populations than in pets, but that may or may not be the case.
DR. BAKER: Maybe Tatty might want to -- it was in her slides, so she could ---
DR. HODGE: Yes. Again, those rates, we do use them and what we do in pharmacovigilance, we use them just as sort of indicators, realizing the discrepancy. When rates like that are created on the human side, they certainly have greater, A, baseline data for natural history of disease. They have insurance and medical records to look at, they have pharmacy records. So they have a lot more data that goes into that kind of calculation on the human side.
I would also admit that they have greater reporting. However, we are sort of comparing to the RiskMAP where we have enhanced reporting of all events. And I do believe that is true, because looking at the cases we have received, we had many veterinarians within this small group of 130 cases multiple cases reported by one veterinarian, and that was not uncommon in this group, but it is uncommon in general to have people reporting like that.
So that is from my perspective. I think that with the enhanced reporting, we have a little bit more assurance that we are picking up most of the cases that are occurring. I do believe, by comparison, it is a rare event, particularly when you are looking at the international market as well. But those rates are not commonly used across human and animal.
DR. DUNHAM: Additional questions for the panel? Yes, Dr. Poppenga?
DR. POPPENGA: I just have one related to the use of the product in Europe and Australia and Canada. Just in terms of -- so none of these enhanced requirements are required in those countries at all?
DR. HODGE: That is right.
DR. POPPENGA: --- education, veterinary certification, that type of thing?
DR. HODGE: Yes, that is correct. They have been marketed in those countries for now 9 years, and continuously without restriction.
DR. POPPENGA: And the incidence rate there or the ADEs reported there are actually lower than in this country. Do you think that is related to the enhanced awareness in the U.S. of the concern with the product?
DR. HODGE: Yes, it is multi-factorial. U.S. tends to report a lot. We are just good reporters of things, observers. And -- but that is typical. But still, with consistency, even not looking at the rate, looking at the types of events, they are so similar that we almost have an identical picture whether or not the rate is exactly the same, allowing for differences in reporting.
Just to give you an example, in Italy, for instance, it used to be that if a veterinarian reported an adverse event, they were immediately open to inspection by the authorities. So what does that do? That automatically tells them not to report any adverse events.
It is increasing in all countries, but Canada is an equal, almost an equal, reporter as the U.S. and Japan as well; they tend to report almost everything that happens -- it is part of their culture.
So it does vary from country to country, but we have to look at not only the overall picture and the rate but in addition it is really critical that you look at each individual report and that is what is really most important.
DR. McCHESNEY: I would like to ask either Martine or John to talk a little bit about the causality system --- (away from microphone) --- because I think it goes to some of the questions I have heard, of what is the importance of -- is 4 days really more important than 28 or 4 months out? And what -- I think what hasn’t been said, and what isn’t in the packet, and so folks on the panel and also in the room know, is that this goes to a causality scoring system where there is each -- there are 6 or 7 factors and --
DR. BAKER: Right.
DR. McCHESNEY: -- I think it is probably worth putting that out so everyone knows that something that occurs at 4 days may not receive the same weight or score as something that occurs at 60 days. So either John or Martine might ---
DR. BAKER: I brought a copy of the algorithm. It is called a “Modified Framer Algorithm” that we use, and there are 6 elements.
The first one is “Previous Experience with the Drug.” And there are scores associated with that of +1, -1 or 0. And I think the algorithm, you actually can get to it on our website if you go to our -- I think it is -- you can find it there.
Then there is -- “Alternative Etiologic Candidate” is the second choice. And there are different choices
given. You can score +2, a 0 or -1. Then there are, again, descriptions on how you make that number choice for how you score that particular alternative etiologic candidate.
Then there is a “Timing of the Event” score which gets into the half-life and the timing based on like C-max to max-effects -- you know, peak exposure effects.
And then there is “Evidence of Overdose” and it is another element of this assessment.
And there is a “Dechallenge” and a “Rechallenge” aspect of that assessment.
So all 6 of those things are done to look at those 6 elements in the causality assessment. And then you come up with an overall score that gives you a choice of whether it is “possibly,” “probably” or “definitely” drug-related or remotely drug-related. And so it is all based on a numeric score that you could have it available to you.
DR. McCHESNEY: And what is related in this, what is being reported here, John, is “probable?”
DR. BAKER: At least “possibly.” But we went even further than that for transparency. We reported -- you know, all the events that were discussed in 2005, we wanted to have complete for that -- you know, as much, you know, disclosure to show you everything that we saw just to let you see the events.
DR. HARTOGENSIS: And just for clarification, the Modified Framer Algorithm is used for each -- to assess each clinical sign associated with that report. So when there are multiple clinical signs, they may get different scores based on timing and when they occurred because things, different things, happened during the course of the report, so it depends on -- each clinical finding gets its score.
DR. DUNHAM: Thank you, Dr. Baker. Additional questions from the panel? Yes, sir?
DR. ALTIER: This is 1 more for Dr. Hodge. You had a slide in which you showed the international adverse event reporting rate, and those rates took a rather precipitous drop between 2002 and 2003, and that coincided with changes in the manufacturing. Am I to conclude from that that there was something in the product prior to that time that was causing these reactions which has since been removed?
DR. HODGE: Yes, I will probably pass this to Steve, but it was the removal of residual solvents from the product.
DR. SUTHERLAND: So there was an observation that lots that were manufactured that had low, no detectable residual solvents were correlated with less allergic mediated adverse events.
So the company made the decision to only manufacture product with Moxidectin active ingredient that had no detectable residual solvents. And that then was correlated in the future. Some immuno-toxicity testing was done to see whether or not residual solvents actually do have allergic potential and the studies indicated that those solvents did, so that further validated that manufacturing change.
DR. ALTIER: Can you tell me then what the similar data would have been for the United States at that time? So you would have had data from the time it was introduced until 2004. Was there a similar sort of drop in the adverse reaction rate?
DR. HODGE: There was. It didn’t drop that precipitously, but -- and I don’t have all of the collection of that data, but there was a decrease in the occurring in the adverse event report rate.
DR. DUNHAM: Martine?
DR. HARTOGENSIS: Just to add to that, the product was recalled in September of 2004 and the change was made in 2002 and the product has a 2-year shelf life, so it appeared that there was a decline in those later months but it was difficult to statistically show whether there really in fact was a decline or not based on the shelf life of the product. We did do a lot analysis and it wasn’t completely clear, statistically.
DR. DUNHAM: Thank you. Additional questions from the panel? Yes, sir, Dr. Senior?
DR. SENIOR: The Pfizer Animal Health is -- has been following this RiskMAP for a brief period. The predominant duration of this RiskMAP was under the auspices of FDAH. And I just wondered if there was -- can we presume that the level of surveillance that you outlined in your presentation matched that that was the level of surveillance that FDAH was performing?
DR. HODGE: Yes. All I can say is I reviewed every single report out of FDAH and discussed with the colleagues there their procedures and watched their procedures as they continued to handle the RiskMAP during that period of October to middle of December.
So I can only say, you know, that I -- first off, I had to satisfy myself with the data, and I did that, and then I had to satisfy myself with the process in place. And so I am fairly certain that they followed the RiskMAP properly.
DR. SENIOR: Okay. The data that was presented, it was numerical in nature and there was no attempt to apply any statistical analysis to comparison of pre-2004 withdrawal and the incidence recording, if you can call it that, the FDA recording after the reintroduction under the RiskMAP.
Do you have any data that can tell us the relative risks or any change in risk, you know, the simple consequences, understanding fully that we are under a great degree of surveillance and reporting would be more common? I just wondered if you had run those numbers.
DR. BAKER: Well, we did have a statistician. We -- you know, we had an epidemiologist and statisticians look at the information. So I don’t know if it is appropriate for us to -- you know, if they would want to talk about it, because they are present, the statistician, but, you know, because I couldn’t address that because I am not a statistician. But we did have that -- you know, that, and here she comes.
DR. DUNHAM: Dr. Mary Bartholomew will come and answer this question. Thank you, Mary.
DR. BAKER: Thanks, Mary.
DR. BARTHOLOMEW: May I speak from this microphone? Okay. I am Mary Bartholomew. I have been at the Center for about 20 years now.
In terms of doing statistical analysis, this is what I told the group. We have an estimate of doses distributed for the period for 2004 of 18,000,000. Now, how much rounding do you think went into that number? Anywhere between -- you know, from 17,500,000 up to 18,499--something. And so I believe that we have a closer estimate of number of doses for the RiskMAP period.
I did not feel comfortable, given the amount of noise that was in the 18,000,000 estimate, to say flat-out “I am going to do -- perform a statistical test.”
So we were stuck with then looking at the proportion of events of a certain type out of the total that were reported. And the most serious one to us, of course, was death. And when we compare the rate, or the proportion, of death out of all events that were reported pre-RiskMAP to RiskMAP, there was a substantial reduction. About a fourth as many of the cases went on to die. And that is kind of where I have left it.
If you look at the histogram which is based on the noise in that 18,000,000 figure, those are rates per 10,000 based on that. What amazes me is with that amount -- level of noise, the similarity between then and now. And so even with all that noise, and you can figure out if you have a numerator and you divide by 17,500,000 versus 18,400,000, it is not going to make much difference in the ratio.
So although I did not feel like attaching a significance level to something divided by something around 18,000,000, I did feel that this represents that we are not seeing much difference in the reporting rates now and then and that we did see a substantial reduction in deaths, which was one of the key components of the RiskMAP, was to mitigate that risk of primarily, and others as well, but that was the biggest achievement, I think, of the RiskMAP overall.
DR. SENIOR: Okay, thank you.
DR. DUNHAM: Thank you, Dr. Bartholomew. Dr. Senior?
DR. SENIOR: Just to stay with it for a bit, you mentioned that in -- regularly, it is 70/30 ratio of veterinarians reporting versus clients reporting, and that would apply to all product. What are the numbers for this product during the RiskMAP?
DR. HODGE: I did speak to that, and I feel that it was primarily veterinarians reporting. But there were -- some pet owners had direct calls and reported an event. But it was primarily veterinarians. But I would really have to go back to the data to pull out a real number for that, but primarily it was the veterinarians reporting.
DR. SENIOR: Okay. When you look at the -- whether it is WHO-considered rare, I am having trouble with these, this terminology, because I am trying to differentiate between WHO-considered rare and acceptable. And, of course, if -- and this would also be determined by what kind of reaction you are dealing with if an “A” was always “death,” then maybe “rare” isn’t very good. If an A involves many other things which clients would more or less handle easily and not have too much problem and the pet just goes on, then you have no -- you know, it would be very, very different.
But can you -- you must have numbers. I mean, anyone who has been in practice for any period of time has seen a vaccine reaction of an anaphylactic nature. If you have vaccinated any animals at all over a period of a career, you have seen these. Surely you have numbers on -- and this goes to anyone on the panel -- sure, you have numbers regarding reported vaccine reactions of an anaphylactic nature, and if you do, I would like to hear what they are so that we can get some kind of comparison with the situation under this RiskMAP.
DR. BAKER: Well, one thing I should point out, that we do not regulate biologics, vaccines. USDA regulates those for animals, so that is why we wouldn’t have that, you know, information just in terms of just the reporting rates for those because that is a USDA-regulated product. So that is first of all.
Unlike, you know, the FDA other centers, Center for Biologics, Center for Devices, they regulate human biologics, so that is all under one FDA -- you know, one under the FDA. So I just wanted to point that out, that we don’t have that information from the fact that we don’t regulate the products.
DR. HODGE: We do have numbers, of course, but it is really what are -- what is the clinical relevance of these numbers? We are called upon primarily for marketing and to provide to clinicians in the field to provide numbers, and so I will give you some.
There are some vaccines that we have reported rates of 7 to 10 per 10,000 doses administered. Sometimes, depending upon a vaccine or a vaccine product, it may be as high as 14 to 15 per 10,000. But that is all within a particular safety profile for a particular product.
Now, again it is all reporting dependent and how you are calculating your doses administered, and actually it is kind of a hot button and it is pertinent to understanding pharmacovigilance. But on the biological side, you at least have one dose per animal when you are talking about companion animals. When you are talking about pharmaceuticals, you have a wide range of doses, from Chihuahuas to Great Danes, so that dose, or animals treated, is very difficult to compare.
So on the biological side, those rates are a little more relevant. So I could say to you that, on the average for most vaccines, we are looking at rates 1 or below 1 per 10,000 doses administered. And again, that is reported rates, and since it is an expected event with biological products and it is not a requirement that we report those, so there is underreporting and so forth, but at least the denominator is quite a bit more accurate because it is 1 dose per animal.
So that is typically what -- but, if you were going into a practice database to look at that same information, you may get something like 20 to 30 per 10,000 doses actually administered because that is the accurate and that is the most accurate representation of that data because it is actually recorded as it is occurring.
So it varies all over the place, and that is why my stance in pharmacovigilance is the rates are important and it is important for you as a group to understand how widely they vary, but it is more important, really, the individual case analysis. And that -- and when we have drilled down to that level, that is really the gold standard for comparison.
DR. BAKER: I just wanted to take a moment to just reemphasize the value of this guidance for industry that I -- if you have a chance to look through this at some point, they are good pharmacovigilance practices and pharmaco-epidemiologic assessment. It talks about all those elements in great detail and points out a lot of things that Tatty pointed out.
It is just a very well put together document for explaining exactly what we just talked about for the last, you know, 3 minutes here, but it gives you some really good understanding of how CDER, you know, and CBER, who have got tremendous history looking at, you know, this reporting rate information, you know, the limitations, what you can and cannot say about reporting rate information, and how you would look at, you know, pharmacovigilance. It is a great, great document to look at. So it is on the slide that you could take a look at.
DR. WOLF: Can I ask a quick question? Again, not being in this field, what is the rate of infection of heartworm among unvaccinated or, you know, on pets that are not given any preventive measure?
DR. HODGE: It really depends on what area of the country you are talking about. But when I worked at the Katrina, 100 percent, without question. If they are not on preventative and if they are not 100 percent compliant with the product, they are going to have heartworm disease. Just as an aside, I saw heartworm disease which most young people in different parts of the country have never seen before, but animals actually dying in front of your eyes from heartworm. So that was an eye-opener that -- you know, it is still a very, very big issue, depending upon where you are in the country.
DR. DUNHAM: Dr. Apley?
DR. APLEY: If I am doing my math right, and we would look at number needed to treat and number needed to harm, if you get down there where you were after Katrina, down around New Orleans, and you have 100 percent infection rate without treatment, and out of your 321,000, how many reports of lack of efficacy did you have?
DR. HODGE: We had 30 reports of lack of efficacy, but only 1 was 1 of any possible sort of true lack of efficacy.
DR. APLEY: Even at 30, by the time you round it out, I would propose the number needed to treat in that environment would be 1 to see a beneficial effect where you prevent the disease.
If you have a number needed to treat of 1 and a number needed to harm, if you take all 102 reported adverse events over 321,290 doses, granted, not dogs but doses, but each one of those represents the application of the drug to the dog, if you apply all 100 to -- and say those are caused by the drug, which we aren’t saying, but if we say all 100 were, that would give you a number needed to harm of 3,000, if I do my numbers right.
So that is another way of looking at it, a number needed to treat of 1. Or even if it was a place with a 50 percent infection rate and you assumed almost complete prevention with the drug, you would have a number needed to treat of 2 with the attributable risk reduction of 50 percent there. But you are looking at number needed to treat of 1 or 2 animals versus the number needed to harm of 3,000. And when we actually take those down to ones who are defensively due to the drug, you may have the number needed to harm closer to 30,000.
So I would invite everyone to kind of mull that over in their head if that is another way to look at it, and I think that is becoming increasingly more accepted on the human side when we have the data to drive that, as comparing your number needed to treat to number needed to harm in weighing that matter.
DR. DUNHAM: Thank you, Dr. Apley. Do we have additional questions from the panel?
DR. DUNHAM: Well, we do value the Q&A. It is here to help you. And we do value your advice that will come back to us later this afternoon.
Seeing no further questions from either, I am going to turn this back over to Aleta Sindelar and she will give us some instructions for lunch and coming back this afternoon. Thank you.
MS. SINDELAR: Thank you, Dr. Dunham. Just a couple of announcements.
I would like to recognize the two individuals from FDA’s Office of the Commissioner, press officers. Those are Ira Allen and Siobhan Delancey. Thank you. And from CVM we have a press officer here as well, Laura Alvey. Thank you.
Important, but not affecting all of us, is our parking issue. (Parking logistics)
Finally, it is just about 12:00 noon. We will break for lunch. It will be 1 hour. We will see you back here I think 1:00, and we will begin with the Open Public Hearing portion of the meeting.
Thank you very much.
(Luncheon recess from 11:56 a.m. to 1:00 p.m.)