Anfrage der FDA zu Zoloft im Jahre 1996

Liebe Leser,

als Dr. Mosholder von der FDA Anfang 2004 die Ergebnisse seiner  Untersuchung zum Suizidrisikos der SSRI-Antidepressiva auf einer Sitzung des Advisory Committees präsentieren sollte, passte dies einigen seiner Vorgesetzten nicht. Denn Dr. Mosholder hatte ein ca. doppelt so hohes Suizidrisiko bei Kindern und Jugendlichen festgestellt und dringend empfohlen, die Fachinformationen und Produktbeilagen entsprechend zu ändern und widersprach damit der bisherigen Meinung der Fachgesellschaften. So wurde die Tagesordnung der Sitzung geändert und Dr. Mosholders von der Liste gestrichen. Als die Presse hiervon erfuhr und berichtete, wollte dann auch der US-Kongress wissen, was bei der FDA vor sich ging und berief eine Untersuchung der Vorfälle ein.

Hierbei erfuhr der Kongress, dass es schon im Jahre 1996 eine Anfrage der FDA zum Suizidrisiko von Zoloft an Pfizer gegeben hatte:

“… Throughout our investigation, we have learned that as far back
as 1996, 8 years ago, a medical review at FDA, Dr. James Knudsen,
raised the question of an increase in suicidality in pediatric
clinical trials of a drug called Zoloft. There was also an analysis in
1997 of Luvox, another anti-depressant, where the review, the
same Dr. Mosholder, noted that there was an increase in hostility
in children versus adults. The issue is noted in the Luvox labeling
as a result.
The fact that children taking anti-depressants were experiencing
psychiatric adverse events at greater rates than adults was known
at the agency as far back as 1996 and 1997…” (s. Seite 11 des Berichts)

Hierzu wurde auch Dr. Mosholder befragt:

“…Mr. WALDEN. Were you aware that apparently Dr. Knudsen also
had some warnings that go back to 1996?
Mr. MOSHOLDER. I recall the——
Mr. WALDEN. For Zoloft.
Mr. MOSHOLDER. [continuing] something about that at the time.
Then, of course, in preparation for this hearing I’ve been reminded
of that. Yes…” (siehe Seite 67 des Berichts)

Später wurde dann auch Dr. Knudsen befragt (s. Seite 71 des Berichts)

“…

[Brief recess.]
Mr. WALDEN. I am going to call this hearing back to order and
ask our next panel of witnesses to come up: Dr. Robert Temple,
Food and Drug Administration; Dr. Paul Seligman, Food and Drug
Administration; Dr. Thomas Laughren, Food and Drug Administration;
Dr. Tarek Hammad, Food and Drug Administration; and Dr.
James Knudsen, Food and Drug Administration. Please come up to
the witness table, if you would.
You are aware the committee is holding an investigative hearing
and, when doing so, has the practice of taking testimony under
oath. Do you have any objection to testifying under oath? Do any
of you have an objection?
Let the record show they all indicated they have no objection.
The Chair then advises you that, under the rules of the House
and the rules of the committee, you are entitled to be advised by
counsel. Do you desire to be advised by counsel during your testimony
today? Mr. Knudsen? Could you turn on your mike, sir, and
then we will need you to identify your counsel. Dr. Knudsen.
Mr. KNUDSEN. My name is Dr. James Knudsen.
Mr. WALDEN. Okay. Yes, you actually have to get kind of close
to that. Sorry. If you could identify for the record, Dr. Knudsen,
your counsel, please.
Mr. KNUDSEN. My counsel?
Mr. WALDEN. Oh, I thought you said you wanted to be represented
by counsel.
Mr. KNUDSEN. No, I did not.
Mr. WALDEN. Okay, fine. No? Okay. Dr. Temple? Dr. Laughren?
Mr. LAUGHREN. No, sir.
Mr. WALDEN. Dr. Seligman?
Mr. SELIGMAN. No.
Mr. WALDEN. Okay. So let the record show, none of them is being
represented by counsel.
In that case then, would you please rise and raise your right
hand, and we will take your testimony under oath. Let the record
show, they all indicated yes.
[Witnesses sworn.]
Mr. WALDEN. Thank you. You may be seated.

…

Mr. WALDEN. Thank you. Well, we appreciate all of you here
today to share with us this information as we continue to look at
what happened in this area and maybe what needed to happen,
and where we are today and where we will be when the FDA
makes it decision relative to the Advisory Committee’s recommendations.
Dr. Knudsen, could you turn in our big binder there to Tab 71
and 72? While you are looking at that, these are the two versions
of a letter under your signature sent to Pfizer Pharmaceuticals on
March 19, 1996, Tab 71 and 72.
Tab 71 has a FAX cover page filled out in someone’s handwriting
to Martha Brumfield of Pfizer from James Knudsen. The top of
that page indicates it was sent at 10:18 and shows FDA
Neuropharm on it as well.
Does this appear to be your handwriting on the FAX cover sheet,
sir?
Mr. KNUDSEN. It does appear to be.
Mr. WALDEN. It does. Okay. The letter attached to this FAX has
lots of typographical errors in it as well as different fonts being
used for various words. If you would turn to Tab 72, it appears to
be the same letter in substance as Tab 71. However, the typos are
removed, and the font is consistent. The letters alone have a different
FAX time Sent stamp on them, and show them coming from
a different section of FDA. Yet does the signature on both these letters
appear to be yours?
Mr. KNUDSEN. Tough question, isn’t it? They appear to be, but
then again—yes, they appear to be. Back in 1996 when I was—my
penmanship may have been a bit better than now. It varies somewhat.
But I will answer the question as it appears to be. I have
to equivocate a week bit, just because of the duration of time and
the instability of my penmanship.
Mr. WALDEN. All of our penmanship tends to suffer with age, sir.
Mr. KNUDSEN. Thank you so much.
Mr. WALDEN. Was it your practice to send a draft letter to a
pharmaceutical company requesting information, then resend a
cleaned-up version later on, though? Would you have sent it as a
draft and then send a different version later?
Mr. KNUDSEN. No, I don’t—I mean, once again I have to preface
a statement by, regrettably, this was done in 1996. So it is some-
what precarious for me to forage around in the limited gray matter
that is available to answer that concretely.
Mr. WALDEN. Is it a practice you recall doing throughout your career?
Do you usually send a draft and then another?
Mr. KNUDSEN. I do not—no, I do not usually send a draft and
another. That’s correct.
Mr. WALDEN. I mean, this wouldn’t be a normal practice, I
wouldn’t think.
Mr. KNUDSEN. No. That’s correct.
Mr. WALDEN. Okay. I don’t know. I mean, I’m not the best speller
in the world, but——
Mr. KNUDSEN. Well, quite frankly, I chatted with this—I mean,
last week I talked with the subcommittee staffers, and I was rather
appalled at what—with the typographical mistakes. I am rather
fastidious most of the time. There are periods whereby I could deviate
from that, but I mean, this is—this being Tab 71 is a mess.
Draft or otherwise, I wouldn’t be sending it to Martha, best I can
recall anyway.
Mr. WALDEN. I understand that. Do you have any explanation for
the fact that two versions of this letter exist?
Mr. KNUDSEN. No, but I suspect others do. I am unable to come
up with an explanation.
Mr. WALDEN. Were you able to find this letter in the files at
FDA?
Mr. KNUDSEN. I checked—no, to answer your question. I did
check the document room. My own files are in—not trying to generate
excuses, but they are in boxes which I invite you to my office
and it is extremely difficult to even find a box. But they are all
there. We are getting ready to relocate. So maybe with another 40
days and 40 nights I could find it.
Mr. WALDEN. Well, should there have been a copy of this letter
in the NDA files?
Mr. KNUDSEN. I would—yes, and I would have kept a copy myself
in my Certraline file. I keep everything.
Mr. WALDEN. Your files are in boxes?
Mr. KNUDSEN. I as unable to locate it in the document room——
Mr. WALDEN. Right.
Mr. KNUDSEN. [continuing] when I was there. I checked in a cursory
way in my office, just trying to find the Certraline file that
I have. In fact, I did find the Certraline file, parts of it, but I could
not locate this particular document.
Mr. WALDEN. Where did you obtain a copy of your March 19,
1996, letter, and which version did you see?
Mr. KNUDSEN. I obtained two copies, one from the Division.
Mr. WALDEN. The Division?
Mr. KNUDSEN. HFD, the Division I am in, the day before I left
to go to Maine. I took it with me, in addition to other things, other
documents, and then the subcommittee members sent via Federal
Express another document. I mean the same one.
Mr. WALDEN. Another copy of that same document?
Mr. KNUDSEN. Yes, sir. Yes, sir.
Mr. WALDEN. And where did the agency get the version they sent
to you?
Mr. KNUDSEN. I did not inquire.
Mr. WALDEN. Dr. Temple, do you know?
Mr. TEMPLE. I could be wrong about this. My understanding is
that Dr. Knudsen got a copy of the letter from the committee.
Maybe I’m wrong about that.
Mr. KNUDSEN. Yes, I just said that.
Mr. TEMPLE. And that we never were able to find it in our files
and got it from Pfizer.
Mr. WALDEN. There you go. So you had to go to Pfizer to get it?
Mr. TEMPLE. Yes.
Mr. WALDEN. That’s what you provided to the committee. Right?
Mr. TEMPLE. I’m not sure, but we could not—what I am sure of
is that we were unable to find a record of this letter anywhere in
our files. That, I am sure of. I am not sure about the rest.
I should say that it is unusual. Letters don’t ordinarily go out
under a medical officer’s signature. They would ordinarily go out
under Dr. Katz’s signature or Dr. Lieber’s or whoever was in
charge at the time, and a copy would be in the New Drug Application,
in the file. So this was unusual.
Mr. WALDEN. All right. Dr. Knudsen, was it your practice as a
medical review officer in 1996 to directly correspond with a pharmaceutical
company on a matter you were reviewing, and then request
information or did you need to apprise any of the supervisors
of your request for additional information from the pharmaceutical
company?
Mr. KNUDSEN. It was not my practice to do so.
Mr. WALDEN. So you would have—was it your practice to tell
your colleagues or supervisors that you were seeking such information
from a pharmaceutical company?
Mr. KNUDSEN. Correct, 86 to 95 percent of the time. There is always
a slight opportunity for me to—I mean, once again, I mean,
I answered the question as best I could that it is not my practice
to do so. In fact, I received my copy from the Division via—of
course, I guess the Division received it from Pfizer. I wasn’t aware
of that. I had no need to question that anyway. I just wanted to
take some materials with me to Maine.
Mr. WALDEN. Isn’t it a requirement of FDA regulations these
types of correspondent documents be kept on file by the agency?
Mr. KNUDSEN. Yes.
Mr. WALDEN. All right. And yet in this case, that doesn’t appear
to be what happened. Right?
Mr. KNUDSEN. That is correct.
Mr. WALDEN. All right. In these letters, you state ‘‘We note that
there appears to be an increased frequency of reports of suicidality
in pediatric adolescent patients exposed to Certraline compared to
either placebo or Certraline treated adult OCD patients. If this is,
in fact, the case, what would be a plausible explanation?’’ That is
what is in the letter that you signed or you think you signed and
sent to Pfizer.
You asked for summary tables from Pfizer to compare data from
adult and pediatric patients in their data base. Is it fair to say that
you wrote this letter to Pfizer because you noticed an increase in
suicide related behavior in the pediatric OCD trials relative to the
rates in the adult trials, and that that was of concern to you? Is
that why you wrote this letter to Pfizer?
Mr. KNUDSEN. Yes.
Mr. WALDEN. All right. And was it of enough concern that you
wanted answers from the company?
Mr. KNUDSEN. Correct.
Mr. WALDEN. And approximately 10 days after you sent this letter
to Pfizer, you complete a safety update to Zoloft. We have put
selected pages of your safety review at Tab 81, 81, if you want to
refer to that, sir.
In your safety update you note on page 15 that, ‘‘In the small pediatric
adolescent pool population of OCD patients, the incidence of
suicidality in the Certraline treated patients was fivefold greater
than the adult OCD Certraline treated patients.’’
You go on to note that 4 of 6 Certraline pediatric patients had
comorbid depression and, ‘‘Depression is an important risk factor
for suicide.’’ You then cite an article published in the Journal of
American Academy of Child and Adolescent Psychiatry that indicated—
that also noted the same phenomenon with kids being treated
with Prozac.
What did you do other than note these concerns in the safety update?
Where did you take it from here?
Mr. KNUDSEN. I was trying to see whether or not that was instrumental
in my sending the letter to Pfizer, just to garner some
additional information. This was March 28, 1996. The letter to
Pfizer was October, was it?
Mr. WALDEN. I think the letter to Pfizer, you will see, is dated
March 19.
Mr. KNUDSEN. March 19, before.
Mr. WALDEN. So like 9 days later——
Mr. KNUDSEN. Well, in fact, in reviewing the NDA, this was a
final document that was signed off, the one that—the document in
Tab 81. So prior to finalizing this document, Tab 81, I found this
information to be—at the time anyway, certainly of concern to me
to make some further inquiries to Pfizer, and realizing, of course,
when I finalize this document, I believed that Pfizer had not responded yet to this.
Mr. WALDEN. That would be correct, based on the timeline I have
seen. But 9 days before you wrote to Pfizer asking for this additional
information, why didn’t you include in this update the fact
that you were awaiting additional information from the company to
explain the fivefold increase? Would that have been a prudent
thing to do?
Mr. KNUDSEN. Yes, it would have been.
Mr. WALDEN. Well, my time has expired. I will now recognize the
ranking member of the subcommittee at this time, the gentlelady
from Colorado.
Ms. DEGETTE. Thank you very much. Dr. Temple and Dr.
Laughren, I am wondering if you can tell me, knowing what you
know today, do you believe that Dr. Mosholder’s initial conclusions
about the increased risk of suicidality exists in pediatric populations
taking anti-depressant medication to treat MDD? Dr. Temple?
Mr. TEMPLE. The reanalysis that Columbia did, did not change
the overall direction of the results. So——
Ms. DEGETTE. So your answer would be yes?
Mr. TEMPLE. Would be yes. Dr. Hammad’s analysis and Dr.
Mosholder’s are slightly different analyses, but in fact the relative
proportions of suicidality are similar to what Dr. Mosholder found.
Ms. DEGETTE. What about you, Dr. Laughren?
Mr. LAUGHREN. Yes, I agree. The relative risk for both analyses
is roughly twofold. So it is essentially the same. There are some
differences across drugs. The signal gets a little stronger for some
drugs, a little weaker for others, but overall I agree that it is
roughly the same result.
Ms. DEGETTE. There was about 8 months between his findings
and when, I think, the FDA took action. I guess my question to
both of you: Do you wish that the agency would have taken him
more seriously and allowed him to present the findings so that we
could have warned parents and physicians about the increased
suicidality rates instead of waiting these 8 months?
Mr. TEMPLE. Let me say a few things. Our concern, as I said before,
was that the action we take be based on the best possible
data. Let me describe the kind of data we had here.
The usual way we expected to evaluate increased suicidal risk is
by looking at the scales that patients in trials are given that ask
them how suicidal they are. Dr. Laughren in his comments on Dr.
Knudsen’s review points out that we are going to have more data
on this question.
Those analyses revealed nothing in any of these trials. There was
no increased suicidality by that measure. What we got was something
unexpected, namely the adverse reaction reports, when interpreted,
when translated, revealed an excess of these suicidal behaviors.
What we had very little experience with was what those
things mean.
We thought, as we looked at them, that somebody—that people
expert in interpreting these behaviors needed to look at them. Dr.
Mosholder specifically in his review says he did not try to reevaluate
each of these cases, because he was no longer blinded. That
conclusion——
Ms. DEGETTE. But Dr. Mosholder also said that he only looked
at the most—I’m not a researcher, but he only looked at the most
serious cases and, in fact, Dr. Temple, you yourself in your opening
statement said that the comment you had made about the face
slapping you now regretted that, because he didn’t take those
things into account.
Mr. TEMPLE. Let me explain. He had—in response to the concern
that these cases might not be a true bill, might not be what they
seemed to be, he offered several approaches. One was to only look
at the serious cases. That is clear, and you can see in his review,
if you look at the cases that were included and not included, that
many of the trivial cases were excluded by the decision to look only
at the serious cases. That is perfectly true.
There were, however, additional cases where you didn’t know
what they meant, and he was in no position to reevaluate them.
Let me just——
Ms. DEGETTE. I apologize, but they only give me 10 minutes. So
if you can make your answer concise, I would appreciate that.
Mr. TEMPLE. Okay. I wanted to explain one other point about it.
Ms. DEGETTE. Very briefly.
Mr. TEMPLE. He also said that, if there is noise in the system,
if it is inaccurate, that would tend to hide a finding rather than
to create one, and that is true.
What is also true, however, is if there was a bias toward interpreting
certain things that the drugs do, like agitating people or
making them hostile, as suicidality, that could give you the wrong
picture. It could cause you to think there were suicidal events
when, in fact, they were not.
That is why we thought we needed an independent look at these
cases in——
Ms. DEGETTE. Okay. But at the time that Dr. Mosholder came
up with his findings, there was already the British study that had
come out earlier that year.
Mr. TEMPLE. No, the British were using the same data we were.
Ms. DEGETTE. Right, but they had concluded this increased risk
of suicidality.
Mr. TEMPLE. But we don’t know that they——
Ms. DEGETTE. But I mean there were two.
Mr. TEMPLE. Let me make it clear. There was nothing wrong
with Dr. Mosholder’s analysis, the ratios he designed, any of those
things. That is not——
Ms. DEGETTE. Well, right. In fact, it has now turned out he was
completely right.
Mr. TEMPLE. No, that is not at issue. What was at issue was
what the cases were, whether they really showed suicidality, and
to answer that question you either have to look at them closely or
decide that they could not have been biased.
Ms. DEGETTE. Well, let me ask you this. In the spring or summer
of 2003, Wyeth came to the FDA, and they wanted on their own—
we heard this in the last hearing—to strengthen warnings on
Efexir, and the FDA asked them not to do that. Is that right?
Mr. TEMPLE. Not quite. They were allowed to do that, and they
did it until we created a new stronger warning or—you can call it
strong or not—a different warning in march of 2004. That warning
was in the warning section. It prominently said you really need to
watch patients, and we thought that was a more trenchant warning.
That was in response to the Advisory Committee.
Ms. DEGETTE. Okay. Now do you think that the FDA is going to
adopt this most recent recommendation about the black box warnings?
Mr. TEMPLE. Our public statement said that we were going to do
all the things they said. We want to think about the conversation
they had about the black box. It is true it was 15 to 8, but there
were a lot of people that said a lot of things.
You know, I don’t want to——
Ms. DEGETTE. Does that mean no?
Mr. TEMPLE. No, it absolutely doesn’t mean no. It means we
haven’t finished our decision yet. We want——
Ms. DEGETTE. Well, what is the FDA’s goal with respect to labeling
of these anti-depressants for off-label use for pediatrics? What
is the goal at this point, knowing the information you know about
increased risk of suicidality?
Mr. TEMPLE. Well, we are unquestionably going to explain that
the drugs themselves appear to be—are associated with or cause an
increased risk of suicidality. That is a given. The only question is
what form it will take.
The discussion the Advisory Committee had was——
Ms. DEGETTE. What kinds of forms do you have that you can
take with it?
Mr. TEMPLE. Oh, you could put a warning—I mean, the alternative,
you could put a warning in dark print, something like that,
or you can put it in a box. Those are probably the two choices.
Ms. DEGETTE. So the choice would be to put it on the bottle. No?
Mr. TEMPLE. No, no.
Ms. DEGETTE. To put it on the box?
Mr. TEMPLE. Well, a box warning is the very first thing you read
in the label.
Ms. DEGETTE. Right. Open it up.
Mr. TEMPLE. A warning comes a little bit later. Those are prominent,
too, and we sometimes do one and sometimes do the other.
The particular——
Ms. DEGETTE. If there is a black box, that has to be in the advertising,
too. Right? So if Zoloft has an ad, it has to have a warning,
may cause suicidality in pediatric use, or something like that.
Mr. TEMPLE. Yes. The contents of the black box would have to
appear, but——
Ms. DEGETTE. It seems to me you would want to do that.
Mr. TEMPLE. Wait, wait, wait. The content of the warning would
have to be there, too.
Ms. DEGETTE. Well, sure. I understand, but that’s the effect of
a black box versus some of these other warnings. Right?
Mr. TEMPLE. No. The requirement for advertising is you have to
balance the information. If there was a prominent dark print box,
that would have to be there, too. I’m not trying to discourage a
black box. I am just trying to reflect the fact that people who spoke
to us were concerned that people who were at risk of killing themselves
would not be treated if we scared people too much.
I’m not saying I agree with that. We put the idea of the black
box before the committee. You know, we are not shrinking from it,
but they said multiple things.
Ms. DEGETTE. Well, I would imagine you would share my concern.
My concern is that off-label prescription of these nonapproved
drugs for pediatrics with, at best, no effect on these depressed kids
and, at worst, increased risk of suicidality will continue unabated.
I would assume that is the FDA’s role to decide that. Right?
Mr. TEMPLE. One of the problems with off-label use and not having
enough data is that you don’t know what the answer is. The
Advisory Committee—many, many people said we know how the
studies came out; they are not impressive; they weren’t able to
show effectiveness. But they clearly were concerned that maybe as
a second line drug these drugs probably should be available and
probably worked in people.
That is not the same as knowing, because we know the studies
largely failed.
Ms. DEGETTE. I think we can probably all agree that it would
help to have more clinical trials in this area, would it not?
Mr. TEMPLE. Yes, but they—Again, I am talking for them. I am
not telling you what we decided to do. They were very concerned
VerDate