1
DEPARTMENT OF HEALTH AND HUMAN
SERVICES
FOOD AND DRUG
ADMINISTRATION
CENTER FOR DRUG EVALUATION AND
RESEARCH
ADVISORY COMMITTEE FOR
PHARMACEUTICAL SCIENCE
Advisors and Consultants Staff
Conference Room
2
PARTICIPANTS
Arthur H. Kibbe, Ph.D., Chair
Hilda F. Scharen, M.S., Executive
Secretary
MEMBERS:
Gerald P. Migliaccio, Industry
Representative
Marvin C. Meyer, Ph.D.
Patrick P. DeLuca, Ph.D.
Charles Cooney, Ph.D.
Melvin V. Koch, Ph.D.
Cynthia R.D. Selassie, Ph.D.
Nozer Singpurwalla, Ph.D.
Jurgen Venitz, M.D., Ph.D.
Marc Swadener, Ed.D., Consumer
Representative
SPECIAL GOVERNMENT EMPLOYEES:
Paul H. Fackler, Ph.D.
Gordon Amidon, Ph.D., M.A.
Judy Boehlert, Ph.D.
Leslie Benet
Charles DiLiberti
Laszlo Endrenyi
FDA:
Gary
Buehler, R.Ph.
Ajaz
Hussain, Ph.D.
Helen Winkle
Lawrence Yu, Ph.D.
3
C O N T E N T S
Call to Order, Arthur
Kibbe, Ph.D. 4
Conflict of Interest Statement,
Hilda Scharen, M.S. 4
Bioequivalence of Highly Variable Drugs,
Why Bioequivalence of Highly Variable
Drugs is an
Issue,
Charles DiLiberti, M.S. 11
Highly Variable Drugs: Sources of Variability,
Gordon L. Amidon, Ph.D. 35
Clinical Implications of Highly Variable
Dr. leslie Benet 60
Bioequivalence Methods for Highly
Variable Drugs,
Laszlo Endrenyi, Ph.D. 81
Bioequivalence of Highly Variable Drugs
Case
Studies,
Barbara Davit, Ph.D. 99
FDA
Perspectives, Sam Haidar, Ph.D. 125
Bioequivalence of Highly Variable Drugs
Q&A,
Dale Conner, Pharm.D. 130
Bioinequivalence: Concept and Definition,
Statistical Demonstrations of
Bioinequivalence,
Donald Schuirmann, M.S. 182
Update--Topical Bioequivalence,
Establishing Bioequivalence of Topical
Dermatological
Products, Robert Lionberger,
Ph.D. 225
Future Topics--Nanotechnology, Nakissa
Sadrieh,
Ph.D.
257
Conclusions and Summary Remarks, Ajaz
Hussain,
Ph.D. 270
4
1 P R O C E E D I N G S
2 Call to Order
3
DR. KIBBE: By the clock on the
wall, I
4
think we are at
5
electronics are working so we will be in good
6
shape. We need to start off with
the reading of
7 the
conflict of interest statement.
8 Conflict of Interest
Statement
9 MS. SCHAREN: Good morning.
I am Hilda
10
Scharen. I am the executive
secretary for the
11
Advisory Committee for Pharmaceutical Science and I
13
interest statement for the committee.
14
The following announcement addresses the
15
issue of conflict of interest with respect to this
16
meeting and is made a part of the record to
17
preclude even the appearance of such at this
18
meeting.
19
Based on the agenda, it has
been
20
determined that the topics of today's meetings are
21
issues of broad applicability and there are no
22
products being approved at this meeting.
Unlike
23
issues before a committee in which a particular
24
product is discussed, issues of broader
25
applicability involve many industrial sponsors and
5
1
academic institutions. All
special government
2
employees have been screened for their financial
3
interests as they may apply to the general topics
4 at
hand.
5
To determine if any conflict of interest
6
existed, the agency has reviewed the agenda and all
7
relevant financial interests reported by the
8
meeting participants. The Food
and Drug
9
Administration has granted general matters waivers
10 to
the special government employees participating
11 in
this meeting who require a waiver under Title
12
XVIII,
13
A copy of the waiver statements may be
14
obtained by submitting a written request to the
15
agency's Freedom of Information Office, Room 12A-15
16 of
the
17
Because general topics impact so many
18
entities, it is not prudent to recite all potential
19
conflicts of interest as they may apply to each
20
member and consultant and guest speaker.
FDA
21
acknowledges that there may be potential conflicts
22 of
interest but, because of the general nature of
23 the
discussion before the committee, these
24
potential conflicts are mitigated.
25
With respect to FDA's invited industry
6
1
representative, we would like to disclose that
2
Gerald Migliaccio is participating in this meeting
3 as
an industry representative, acting on behalf of
4
regulated industry. Mr.
Migliaccio is employed by
5 Pfizer. Dr. Paul Fackler is participating in this
6
meeting as an acting industry representative. Dr.
7
Fackler is employed by Teva Pharmaceuticals U.S.A.
8
In the event that the discussions involve
9 any
other products or firms, not already on the
10
agenda, for which FDA participants have a financial
11
interest, the participants' involvement and their
12
exclusion will be noted for the record.
With
13
respect to all other participants, we ask in the
14 interest
of fairness that they address any current
15 or
previous financial involvement with any firm
16
whose product they may wish to comment upon. Thank
17
you.
18
DR. KIBBE: Thank you, Hilda. Just so
19
that our audience knows who all is here, I would
20
like to ask everybody to introduce themselves and
21
give their affiliation. We will
start with Dr. Yu.
22
23
DR. YU:
24
Science, Office of Generic Drugs, Office of
25
Pharmaceutical Science, CDER, FDA.
7
1
DR. BUEHLER: Gary Buehler,
Director,
2
Office of Generic Drugs, Office of Pharmaceutical
3
Science, CDER.
4
DR. HUSSAIN: Ajaz Hussain, Deputy
5
Director, Office of Pharmaceutical Science, CDER.
6
MS. WINKLE: Helen Winkle,
Director,
7
Office of Pharmaceutical Science, CDER.
8
DR. AMIDON: Gordon Amidon,
University of
9 Michigan.
10 DR. VENITZ: Jurgen Venitz, Virginia
11 Commonwealth University.
12 DR. SELASSIE: Cynthia Selassie, Pomona
13 College.
14 DR. BOEHLERT: Judy Boehlert, and I have
15 my own pharmaceutical consulting business.
16
DR. SWADENER: Marc Swadener,
consumer
17
representative, retired from University of
18
19
DR. KIBBE: I am Art Kibbe and I
am
20
Professor of Pharmaceutical Sciences at Wilkes
21
University.
22
DR. MEYER: Marvin Meyer, formerly
23
24
25
DR. SINGPURWALLA: Nozer
Singpurwalla,
8
1
2
DR. KOCH: Mel Koch, the Director
for the
3
Center for Process Analytical Chemistry at the
4
5
DR. COONEY: Charles Cooney,
Professor of
6
Chemical and Biochemical Engineering at MIT.
7
DR. DELUCA: Pat DeLuca,
University of
8
9
MR. MIGLIACCIO: Gerry Migliaccio,
Pfizer.
10
DR. FACKLER: Paul Fackler,
industry
11
representative, Teva Pharmaceuticals.
12
DR. KIBBE: Thank you. We are going to
13
start this morning and Dr. Yu will set us up for
14 our
discussion.
15
Bioequivalence of Highly Variable Drugs
16
DR. YU: Good morning. My slides I guess
17 are
in a different file so I will give my
18
introduction without the slides.
19
Dr. Kibbe, Chair of the FDA Advisory
20
Committee for Pharmaceutical Science, members of
21 the
FDA Advisory Committee for Pharmaceutical
22
Science, distinguished speakers, distinguished
23
guests and distinguished audience, I am
24
Yu. I am Director for Science,
Office of Generic
25
Drugs, Office of Pharmaceutical Science, CDER, FDA.
9
2 and
privilege to introduce to you the first topic
3 of
bioequivalence, bioequivalence of highly
4
variable drugs. The objectives of
this discussion
5 are
to explore and define bioequivalence issues of
6
highly variable drugs, to discuss and to debate
7
potential approaches in resolving them,
8
specifically the pros and cons of the solutions and
9 the
benefits and limitations of these potential
10
approaches.
11
The bioequivalence issues of highly
12
variable drugs have been discussed in many
13
conferences and meetings nationally and
14
internationally. The issue is
obvious because of
15 the
high variability of the drugs or drug products
16
that require a large number of subjects or
17
volunteers in order to pass the confidence interval
18 of
80-125 percent. Despite many, many
discussions,
19
despite many, many publications in scientific
20
literature, to date there is no consensus and no
21
solutions have ever been reached.
In fact, there
22 is
no regulatory definition with respect to the
23
high variability drugs or drug products.
So, there
24 are
various approaches in resolving this in the
25
scientific literature, for example, expansion of
10
1 the
bioequivalence limits; for example, using
2
scaling approaches.
3
We have invited a panel of distinguished
4
speakers this morning to discuss this issue related
5 to
the bioequivalence of highly variable drugs from
6
various perspectives, from practical difficulties
7 of
bioequivalence of highly variable issues, from
8
mechanistic understanding of what causes the high
9
variability of drug or drug products, from
10
understanding of different approaches to resolve
11
understanding of clinical implications why high
12
variability drugs are safer, from case studies and,
13
finally, from regulatory options.
14
At the end of these presentations you will
15 be
asked to discuss or address the following
16
questions. First, what is
actually the definition
17 for
highly variable drugs or drug products?
18
Second, with respect to expansion of
19
bioequivalence limits, what additional information
20
should we gather in order to answer this question?
21 We
also ask you to comment on scaling approaches.
22
With this introduction, I want to turn the
23
podium over to our first speaker, Charlie
24
DiLiberti. Charlie?
25
Why Bioequivalence of Highly Variable Drugs
11
1 is an Issue
2
MR. DILIBERTI: Thank you, Dr.
Yu. Before
3 I
start I need to disclose the potential conflict
4 of
interest in that I am employed by Barr and I am
5
also a shareholder and option holder in the firm.
6
Also, before I get into the actual
7
discussion I would like to say that in the context
8 of
preparing this presentation I had numerous
9
discussions with many of my colleagues in the
10
industry and, based on the feedback that i got from
11
them, it seems to me that the views that I am about
12 to
portray in my presentation are quite widely held
13 in
the industry.
14
[Slide]
15
With that, let's start off with the
16
definition of highly variable drugs.
Oftentimes,
17
highly variable drugs are defined in the context of
18
within-subject variability in terms of a
19
bioequivalence study. I would
like to take it one
20
step further and look at variability within the
21
patient and what does this high level of
22
variability mean to an individual patient taking
23 the
drugs.
24 Commonly, the often used definition
of
25
highly variable drugs is those drugs whose
12
1
intra-subject or, as I characterize it here as
2
intra-patient, coefficient of variation, or CV, is
3
approximately 30 percent or more.
I will use that
4 as
my guideline for the rest of this presentation.
5
[Slide]
6
What are the current criteria?
Just very
7
briefly, for bioequivalence they involve a
8
comparison between test and reference product,
9
involving the natural log transformation of the
10
data. The current criteria are
that the 90 percent
11
confidence intervals around the geometric mean
12
test/reference ratios have to fall entirely within
13 the
range of 80-125 percent.
14
These criteria really apply to all drugs
15
here, in the
16
variability of the drugs. These
criteria do have
17 other
implications. For example, they can be
used
18 by
innovator and, for that matter, generic firms to
19
justify a substantial formulation change so it is
20 not
just in the context of approving a generic.
21
[Slide]
22
This really speaks to the
crux of the
23
issue with highly variable drugs in that it
24
portrays the number of subjects that you would have
25 to
plan on using in a two-way crossover
13
1
bioequivalence study given a particular
2
intra-subject CV. You can see
that for very low CV
3
drugs the number of subjects required is fairly
4
small and quite manageable from a practical
5
standpoint but, as the CV increases, you can see
6
that the number of subjects required can increase
7 to
quite large numbers, possibly in the hundreds.
8
[Slide]
9
Why do we possibly need alternative
10
criteria for highly variable drugs?
Well, first of
11
all, we have an ethical mandate to minimize human
12
experimentation. Second of all,
the prohibitive
13
size of some bioequivalence studies for some highly
14
variable drugs impacts on the availability of a
15 generic
version of that drug, which may mean that
16 in
the absence of a generic many Americans can't
17
afford the reference product so they may go either
18
untreated or they may be subdividing their doses
19
contrary to the prescription.
20
Also, changing criteria will reduce the
21
number of participants in the BE studies and I
22
think it can't be done without compromising the
23
safety and efficacy of the product. Also, there is
24
experience elsewhere in the world with criteria
25
other than 80-125 percent.
14
1
[Slide]
2
This slide shows some of the
3
bioequivalence criteria in other countries and
4
regions in the world. These are
not specific to
5
highly variable drugs and in many cases they don't
6
apply necessarily to all drugs.
That is why I have
7
"most drugs" or "some drugs" listed here. But,
8
certainly, there is experience with certain drugs
9 in
these different regions with confidence
10
intervals that are either wider than 80-125 or, in
11 the
case of
12
confidence interval criterion, just a point
13
estimate criterion.
14
[Slide]
15
What types of drugs are highly variable?
16
Well, the types of drugs really cut across all
17
therapeutic classes and include both new and older
18
products. The potential savings
to American
19
consumers could possibly be in the billions of
20
dollars if generics are approved.
In saying this,
21 I
want to be clear that the bioequivalence issues
22 for
many of these drugs are not the only barriers
23 to
getting a generic. In some cases there
might be
24
patent issues or formulation issues as well, but
25
still the bioequivalence issues do represent some
15
1
sort of a barrier.
2
What are some examples? This is a
very
3
brief list and the list can go on and on but just
4 to
give you some kind of representative examples of
5
drugs that cut across many therapeutic areas, some
6 of
which are on-patent, some off-patent, just to
7
give a flavor.
8
[Slide]
9
Another issue is that as of last year we
10 now
have to meet confidence interval criteria for
11 fed
bioequivalence studies. So now the
variability
12
under the fed state is of concern.
There is
13
generally very little information available on the
14
variability of drugs in the fed state, and we have
15
found that some drugs do show more variability
16
under fed conditions than under fasting conditions,
17
leading to the potential for bioequivalence
18
failures because they may be under-powered. What I
19 am
trying to get across here is that because of the
20
lack of information on many drugs under fed
21
conditions, there may in fact be many more highly
22
variable drugs than we are led to believe.
23
[Slide]
24
Why aren't the current criteria
25
appropriate for some highly variable drugs? Well,
16
1 I
will start off by saying that the current
2
criteria are, I believe, appropriate for drugs with
3 low
to moderate variability because the
4
dose-to-dose variability that a patient would
5
experience is comparable and consistent with the
6
width of the criteria.
7
However, in the case of highly variable
8
drugs this is not true where the dose-to-dose
9
variability experienced by a patient may often be
10
much larger than the width of the criteria. I will
11
illustrate this point later on with some graphs.
12
Highly variable drugs are oftentimes wide
13
therapeutic index drugs. In other
words, they have
14
shallow response curves and wide safety margins. I
15
want to qualify this statement by saying when I say
16
highly variable drugs, highly variable in a patient
17
with respect to the parameter that is variable. If
18 a
patient experiences high variability, that means
19
that the drug is safe and effective despite this
20
wide variability in the patient.
Therefore, I
21
believe that modifying bioequivalence criteria on
22
highly variable drugs to reduce the number of
23
participants in bioequivalence studies could be
24
accomplished while still maintaining safety and
25
efficacy assurance.
17
1
[Slide]
2
Different highly variable drugs may
3
require different approaches. One
size may not fit
4
all. As we can see from the
earlier power graphs
5
that I had plotted, obviously the number of
6
subjects required for a drug with, say, 30 percent
7
coefficient of variation is very different from the
8
number of subjects required for a drug with, say,
9 70
percent intra-subject CV. And, there are
other
10
considerations that we have to take into account.
11
[Slide]
12
Probably one of the more important
13
considerations is whether the drug accumulates in a
14
patient at steady state. Let's
first take the case
15 of
a drug that does not experience significant
16
accumulation to steady state in a patient. These
17 are
typically short half-life drugs, in other
18
words, short half-life with respect to the dosing
19
interval. Here are some
examples. We could
20
possibly consider some sort of modification to the
21
criteria for both AUC and Cmax because an actual
22 patient
would experience significant dose-to-dose
23
variability for both Cmax and AUC because neither
24 is
smoothed out at steady state. Therefore,
the
25
drug could be considered to exhibit a wide
18
1
dose-to-dose variation in blood levels irrespective
2 of
chronic dosing.
3
The same sort of logic could potentially
4
apply to a highly variable drug that is not dosed
5
chronically. One particular
application of the
6
scenario of a relatively short half-life drug that
7
does not undergo accumulation might be the case of
8 a
parent drug with a short half-life and high
9
variability where there is also a metabolite that
10
needs to be measured which has a much longer
11
half-life and low variability. I
could easily
12
envision the case where the confidence interval
13
criteria are somehow modified to accommodate the
14
higher variability of the parent drug but, in the
15
same compound, the current 80-125 criteria could be
16
applied to the metabolite.
17
[Slide]
18
Now let's look at the case of accumulation
19 to
steady state. Typically, this is a case
where a
20 drug is used chronically and with a half-life
long
21
relative to the dosing interval so there is some
22
accumulation going on. Here are a
few examples.
23
In this case, because the accumulation
24
process will tend to reduce the fluctuation in AUC
25 and
Cmax, both at steady state, actually in
19
1
essence, the drug to a patient may not really be
2
highly variable because the variability may be
3 small at steady state. However, the Cmax and AUC I
4
think need to be looked at in a different light.
5 At
steady state the test/reference ratio for two
6
drugs, assuming linear accumulation, will be about
7 the
same as the test/reference ratio that we see in
8 a
single dose study because the accumulation
9
process preserves that test/reference ratio.
10
However, for Cmax, generally speaking, the
11
test/reference ratio that we see at single dose
12 conditions
will be the most extreme and the
13
test/reference ratio observed upon accumulation to
14
steady state will go closer and closer to unity,
15
one. So, that is why I think we
potentially need
16 to
consider these two cases differently in the case
17 of
a drug that accumulates.
18
[Slide]
19
The other possibility with drugs subject
20 to
accumulation is to actually conduct the steady
21
state study but this has all sorts of practical
22 limitations
for some drugs, including toxicity.
23
[Slide]
24
What I have tried to do in this graph is
25 to
get some sense of the magnitude of day-to-day
20
1
fluctuations in a pharmacokinetic parameter--I have
2
plotted this as if it were Cmax but it could
3
equally apply to AUC--in the case of a drug that
4
does not undergo accumulation.
5
What is plotted here, in orange, is
6
simulated data representing the sequential
7
day-to-day Cmax's that might be seen in a given
8
patient taking a single drug over the course of 30
9
days where the drug has a true mean of 100 percent.
10 In
fact, the sample mean here for this set of 30
11
data points is 100 and is the geometric mean, and
12 the
CV of this data set is 10 percent. So,
you can
13 see
that the drug is fairly well controlled within
14 a
fairly narrow range. Just as a yardstick
for
15
variability, I have plotted the bioequivalence
16
limits, the 80 percent limit and the 125 percent
17
limit. I want to make it clear
these limits do not
18
apply to individual day-to-day values, but I am
19
just plotting them here to give some sense of
20
scaling.
21
What I have plotted here, in the green, is
22 a
different formulation, formulation B of the same
23
drug that has a mean here of 125.
So, it is a 25
24
percent higher mean than this. CV
is still 10
25
percent. So, this could be seen
to represent the
21
1
magnitude of change that one would expect upon
2
switching a patient from one formulation to a
3
second formulation with a higher mean.
You can see
4
that there is some degree of overlap between the
5
second formulation and the first but, just
6
eyeballing this, it is not too hard to see that
7
there is visually some discernible shift in the
8
overall levels.
9
[Slide]
10
Let's see what the case looks like for a
11
drug with 30 percent intra-subject CV.
You can see
12
here that there are many more excursions on a
13
single formulation outside the range of 80-125
14
percent. Overall, there is much
more overlap
15
between formulation B and formulation A despite the
16
fact that these two formulations differ by 25
17
percent.
18
[Slide]
19
Let's increase the variability one notch
20
further to 50 percent CV, and we can see even more
21
day-to-day excursions in Cmax for a patient on a
22
given formulation, many of them outside 80-125.
23 You
can see now that the overlap between
24
formulation B and formulation A, again a 25 percent
25
difference here, is almost not discernible at all
22
1 to
the eye.
2
[Slide]
3
Finally, let's turn it up one notch
4
further to 70 percent intra-subject CV.
With a
5
drug that is this variable you end up, while on a
6
single formulation with no switch involved, with a
7
range of Cmax values that could be as far as a
8
5-10-fold range day-to-day. So,
there are wide
9
swings in the Cmax's achieved for a given subject.
10
In light of this, suppose that this is a
11
reference drug that is already approved by the
12
agency and known to be safe and effective, that
13
safety and efficacy is true in spite of the wide
14
variability from day-to-day so, therefore, the drug
15
cannot have a narrow therapeutic index and must
16
necessarily have a relatively wide therapeutic
17
index if it is safe and effective despite such wide
18
variation.
19
Also, you can see that the switch-over
20
product, formulation B, again a 25 percent higher
21
mean, is virtually indistinguishable now from the
22
range of blood levels that you see with formulation
23 A.
24
I think that the criteria, which are still
25
plotted here, 80-125 percent, need to be
23
1
commensurate with the degree of overlap that we are
2
trying to achieve between formulations.
Even
3
though these are the criteria, I would like to
4
point out that in order to pass the criteria the
5
actual observed mean in a bioequivalence study
6
generally has to be in a very narrow range, maybe 5
7 or
10 percent deviant from 100. Outside of
that,
8
your chances of passing a bioequivalence study on a
9
very variable drug are very, very poor.
10
[Slide]
11
There are certain special considerations
12
that we need to take into account in the discussion
13 of
highly variable drugs, one of which is where
14
parallel studies are conducted for long half-life
15
drugs.
16
Oftentimes you can't do a crossover study
17
because the wash-out period would be too long.
18
Powering parallel studies depends on between
19
subject variability rather than within subject
20
variability. Between subject
variability is often
21
large, necessitating large bioequivalence studies
22
just as with highly variable drugs.
However, the
23
high between subject variability does not
24
necessarily imply high within subject variability.
25
Instead, it could be due to inter-individual
24
1
differences in absorption, metabolism, etc. So,
2
these drugs, from a clinical perspective, may not
3
really be highly variable but we are still faced
4
with the powering problems in terms of conducting
5 bio
studies. In these cases, generally
speaking,
6
multiple dose studies are not feasible, and we
7
might consider some sort of alternative criteria
8 for
such studies.
9
[Slide]
10 A second issue that arises and is
directly
11
related to the issue of highly variable drugs is
12 the
issue of pooling data from multiple dosing
13
groups. Because of the large
number of subjects
14
often required for highly variable drugs,
15
oftentimes you have to split up dosing into
16
multiple dosing groups.
17
Currently, the FDA requires a statistical
18
test for the poolability of the data from these
19
multiple dosing groups and the test is a measure of
20 the
significance of the group by treatment
21
interaction terms in the analysis of variance. If
22
this interaction term is statistically significant,
23
then you are not permitted to pool the data from
24 the
multiple dosing groups. The consequence
of
25
this is that each group is then evaluated on its
25
1 own
merit and, because each group is generally
2
considerably smaller than the total pool of
3
subjects, each group will be grossly under-powered
4 to
achieve bioequivalence and, therefore, if you do
5
have a statistically significant interaction term,
6
overall you are likely to have failed the criteria.
7
This procedure results in
discarding and
8
having to repeat about 5 percent of studies based
9 on
random chance alone, even if there is no genuine
10
underlying effect. The concern
here I think is
11
that even if there were some sort of underlying
12
explanation for the statistical significance of the
13
interaction term, for example differences in
14
demographics among the dosing groups, I believe
15
that there is no reason not to use the data from
16 all
the dosing groups because had they been dosed
17
together in a single group it would be perfectly
18
usable and we wouldn't be having this discussion.
19
[Slide]
20
Conclusions--while the current
21
bioequivalence acceptance criteria I believe are
22
appropriate for drugs with ordinary variability, I
23
believe that they may not be appropriate for some
24
highly variable drugs.
25
Current bioequivalence acceptance criteria
26
1
make it difficult or impossible to develop generics
2 in
some cases, which has the public health issue of
3
effectively denying treatment to many patients
4
because of affordability issues.
5
I believe that practical, scientifically
6
sound alternative bioequivalence acceptance
7
criteria could be implemented for highly variable
8
drugs to reduce the bioequivalence study size while
9
still maintaining assurance of safety and efficacy.
10
Different approaches may be needed for
11
different types of drugs depending on accumulation
12
following multiple dosing, and also depending on
13 the
variability of the drug. And, other
related
14
situations, i.e., the issue of parallel studies and
15
multiple dosing groups should also be considered in
16
conjunction with any changes to acceptance criteria
17 for
highly variable drugs. Thank you.
18
DR. KIBBE: Does anybody on the
panel have
19
questions for our presenter to clarify information?
20
Nozer?
21
DR. SINGPURWALLA: Certainly, I
do. I
22
have four questions and five comments.
Do I have
23
time?
24
DR. KIBBE: You have until
everybody
25
leaves to go to the airport!
27
1
DR. SINGPURWALLA: The first
question is a
2
question of clarification. What
is Cmax? when
3
somebody puts C and a max I think of the maximum.
4
MR. DILIBERTI: That represents
the
5
maximum because concentration achieved within a
6
given patient or subject over the course--
7
DR. SINGPURWALLA: So, it is
maximum blood
8
concentration?
9
MR. DILIBERTI: Yes, it is maximum
blood
10
concentration.
11
DR. SINGPURWALLA: Thank you. What is
12
AUC?
13
MR. DILIBERTI: Area under the
curve,
14
which is generally taken to be a measure of the
15
extent of absorption.
16
DR. SINGPURWALLA: The third
question is
17 why
did you take natural logs?
18
MR. DILIBERTI: It is conventional
in the
19
analysis of bioequivalence data to do a log
20
transformation. This is already
established as
21 standard--
22 DR. SINGPURWALLA: Log transformation of
23 the whole data or just the maximum?
24
MR. DILIBERTI: You would log
transform
25
each of the individual Cmax's and then follow that
28
1 by
appropriate analysis of variance. The
same log
2
transformation also applies to the individual AUCs
3
prior to analysis of variance.
4
DR. SINGPURWALLA: Well, I can see
doing a
5 log
transformation of all the data to get
6
approximate normality if the distribution is log
7
normal.
8
MR. DILIBERTI: Yes, that is true.
9
DR. SINGPURWALLA: Just taking the log of
10 the
maximum--I don't know. By geometric
mean, you
11
mean product divided by--what do you exactly mean?
12
MR. DILIBERTI: The geometric mean
is what
13
results from the log transformation.
You do the
14 log
transformation and conduct analysis of
15
variance. From the analysis of
variance you get a
16
least-squares mean on a log transformed variable.
17
When you back-transform that by exponentiating it
18 you
end up with, in essence, a geometric mean.
19
DR. SINGPURWALLA: Okay. Now we will go
20 to
comments. As somebody who is new to all
this
21 and
doesn't know, the thought that first comes to
22 my
mind is that this HVD, highly variable drug,
23
should really be looked at as a bivariate problem.
24 You
have two variables. One variable is the
extent
25 of
absorption and the other variable is the rate of
29
1
absorption. So, I would look at
it as a surface
2
because the following is possible, suppose you have
3 a
drug which has a low variability with respect to
4
absorption but high variability with respect to
5
extent of absorption, how do you classify it? So,
6
what we need is a better measure of classifying a
7
highly variable drug which is a bivariate measure.
8
That is the first comment.
9
You proposed, I think, abolishing the
10
confidence limit notion.
11
MR. DILIBERTI: No, I didn't. I am not
12
here to propose solutions to the problem; I am just
13
here to really identify what the concerns and
14
problems are.
15
DR. SINGPURWALLA: Okay, but do
you have
16 any
sense of what is an alternative?
17
MR. DILIBERTI: Various
alternatives have
18
been proposed, including reference scaling or some
19
fixed point scaling that is different from 80-125--
20
DR. SINGPURWALLA: But you are not
putting
21
those forward?
22
MR. DILIBERTI: I am not really
here to
23
discuss that.
24
DR. SINGPURWALLA: So, your basic
focus is
25
criticizing what is there but without an
30
1
alternative in mind?
2
MR. DILIBERTI: Right, I think
many of the
3
later speakers will address the issue of potential
4
solutions.
5
DR. SINGPURWALLA: Now, in these
charts
6
that you showed, how did you choose the particular
7
patient whose charts you were showing?
8
MR. DILIBERTI: It is simulated
data. It
9 is
log normally distributed random independent
10
variables. It is not patient
data. I am sorry, I
11
thought that that was clear. It
is entirely a
12
computer simulation just to give some sense of the
13
relative magnitude of the variability.
14
DR. SINGPURWALLA: Well, I didn't
get that
15 message.
I thought that was a real patient--
16
MR. DILIBERTI: No, no, no.
17
DR. SINGPURWALLA: --those data
you were
18
showing.
19
MR. DILIBERTI: No.
20
DR. SINGPURWALLA: But you don't
need to
21
show it because if it is simulated we can
22
appreciate it. The last point is
when you talked
23
about pooling the data between two groups, how is a
24
group defined? What constitutes a
group?
25
MR. DILIBERTI: By the day on
which dosing
31
1
occurs. For example, it may be
impractical to dose
2 100
patients or subjects in a clinic all on the
3
same day. So, you may have to
dose half of them
4
today and maybe the other half several weeks from
5
today.
6
DR. SINGPURWALLA: So the groups
are
7
random depending on who shows up.
8
MR. DILIBERTI: Essentially, yes.
9
DR. SINGPURWALLA: Suppose one
were to
10
think about forming these groups based on some
11
other, you know biological or--defining a group in
12 a
certain way, conceivably you could justify
13
pooling. This is completely
random.
14
MR. DILIBERTI: Right, and I
believe that
15 the
way that the groups are conventionally arranged
16 in
a typical bioequivalence study pooling may be
17
justified even if you do have a statistically
18
significant interaction term.
19
DR. SINGPURWALLA: See, what I am afraid
20 of
is that if you did this on some other day and
21 you
had the same policy of pooling at random you
22 may
see a completely different result in the sense
23
that the point you are making may not be made.
24
Well, thank you.
25
MR. DILIBERTI: Thank you.
32
1
DR. KIBBE: Anybody else? Go ahead.
2
DR. SELASSIE: You mentioned that
3 potential savings to patients are in the
billions
4 of
dollars if generics are approved. Can
you tell
5 me
or do you have an idea of what percentage would
6
actually be the lack of savings due to the fact
7
that there are no generics for each of these as
8
opposed to other patent issues?
9
MR. DILIBERTI: That is very
difficult to
10
assess because, for example, in looking at patents
11 you
need to look even beyond the "Orange Book."
12
Some of these formulations have patents that are
13 not
listed in the "Orange Book."
So, to compile
14
data like that would be a Herculean task. However,
15 I
do know from personal experience that the
16
difficulties in meeting bioequivalence criteria do,
17 in
fact, pose a very real barrier to the
18
development of some generics.
19
DR. MEYER: If I could give an
example, if
20
your wife is on premarine you know you insurance
21
co-pays $20.00, because there is no generic
22
currently available because of bioequivalence
23
issues, instead of $5.00.
24
MR. DILIBERTI: Right.
25
DR. MEYER: Since my light is on I
will
33
1 just add that I do agree with you about
pooling
2
data together. A clinical trial,
after all, has a
3
patient come in to a doctor's office; they take a
4
measurement. A week later another
patient comes in
5 and
now you have two groups, and you don't analyze
6
those separately. So, unless
there is really some
7
reason to think that two groups of 50 can't be put
8
together to make one group of 100, I think it is
9
silly not to put them together.
10
DR. KIBBE: Paul?
11
DR. FACKLER: If I could just make
a
12
couple of comments, one addressing the issue of AUC
13 and
Cmax, there are very few drugs where I think
14
Cmax is not highly variable but AUC is.
I would
15 say
that from our experience it is the other way
16
around.
17
DR. SINGPURWALLA: I am sorry, I
missed
18
that. You are saying that the two
are correlated.
19
DR. FACKLER: I am saying that
there are
20
very few examples of drugs that are highly variable
21 on
AUC but not highly variable at Cmax.
Generally
22 it
is the other way around, AUC is not as variable
23 as
Cmax.
24
DR. SINGPURWALLA: So, it makes my
point
25
that you may have a bivariate situation.
34
1
DR. FACKLER: Yes, absolutely.
2
DR. SINGPURWALLA: Thanks.
3
DR. FACKLER: One of the things I
wanted
4 to
ask Charlie was on the simulated data you
5
represented 80 percent and 125 percent.
I am
6
wondering did you happen to calculate the
7
confidence intervals for the simulated data sets to
8
show where the 90 percent confidence intervals
9 would
have resulted? Because I am certain they
are
10 far
beyond 80-125.
11
MR. DILIBERTI: That is
right. No, I did
12 not
go through that calculation.
13
DR. FACKLER: The last point I
wanted to
14
make was that on the graph of the number of
15
subjects needed to get to 80 percent power versus
16 the
variability, it is important to recognize that
17 80
percent power means that one out of five studies
18
under those conditions will fail to show
19 bioequivalence, or only four out of five
will. So,
20
even if a product is tested against itself with,
21 for
instance, 30 percent variability, using the
22
number of subjects in that particular graph one out
23 of
five studies will fail to show that the product
24
against itself is bioequivalent.
25
DR. KIBBE: Shall we move
along? I think,
35
1
Gordon, you are up.
2
Highly Variable Drugs: Sources of Variability
3
DR. AMIDON: I am going to talk
about
4
sources of variability and emphasize mechanisms of
5
absorption and focus on bioequivalence from an
6
absorption point of view. It is
the approach I
7 have
been taking for the past 10 to 15 years.
8
[Slide]
9
If you think about bioequivalence where we
10 are
comparing drug products, then the question of
11
bioequivalence is really a dissolution question.
12
Right, the same drug? So, we
should be looking at
13
mechanism and dissolution and processes that are
14
controlling absorption and develop our tests around
15
that mechanism, what is controlling the process.
16
Of course, plasma levels are the gold
17
standard. Our business is to
ensure that plasma
18
levels match the innovator product used in the
19
clinical testing. That is the
criterion, no
20
question about that; no argument about that. The
21
question is what test.
22
[Slide]
23
So, I want to show some of the factors.
24 We
tend to focus on bioequivalence from a plasma
25
level point of view over here. We
focus on the
36
1
plasma which is the gold standard.
But if
2
absorption is controlled by the dissolution
3
process, dissolution controls the presentation of
4
drug along the gastrointestinal tract and,
5
therefore, controls the rate and extent of
6
absorption. If the rate and
extent of absorption
7 is
the same, then the plasma levels will be the
8
same. So, in the question of
bioequivalence then
9 the
real scientific issue is how do we set a
10
dissolution standard? My position
may be a little
11
extreme because no one seems to want to think about
12
that very much but that is the reality of the
13
science.
14
[Slide]
15
So, I think if you have two drug products
16
that present the same concentration profile along
17 the
gastrointestinal tract, they will have the same
18
rate and extent of absorption and systemic
19
availability. You may want to
think about that,
20 the
same rate and extent of absorption implies the
21
same systemic availability. So,
we need to focus
22 on
product.
23
[Slide]
24
Some of the processes in the
25
gastrointestinal tract that can lead to the
37
1
variability--and I will just illustrate some of the
2
processes here--would be the gastric emptying,
3
intestinal transit, luminal concentration both of
4 pH
and surfactants, phospholipids, presence or
5
absence of food. When you think
about it, there
6 are
a lot of sources of variability just in the
7
gastrointestinal tract.
8
[Slide]
9
Systemic availability--what should our
10
testing ensure? It is the gold
standard, no
11
question about it. But the
question then is what
12 is
the best test? What is the best test to
ensure
13
plasma levels? And, when plasma
levels are
14
difficult to measure or, in the case of highly
15
variable drugs where it requires a lot of subjects,
16
then I think it really requires us to think what is
17 the
source of that variability and then what type
18 of
test might we set.
19
I would argue that if two highly variable
20 drug
products dissolve the same way in the
21
gastrointestinal tract they will be bioequivalent.
22 It
might require 100 subjects to show that.
I
23
think that is unnecessary. I
think you just do it
24
with a dissolution test and the answer will be far
25
simpler.
38
1
[Slide]
2
So, what are some of the physicochemical
3
factors? Clearly, particle size
and distribution;
4
wetting and solid-liquid contact; and, of course,
5 in
some cases chemical instability such as prodrugs
6 and
esterases and peptidases in the
7
gastrointestinal tract can lead to highly variable
8
absorption and, hence, systemic availability.
9 [Slide]
10
I just put one graph in here showing the
11
dependence here of dissolution time, ranging up to
12 30
hours, and gastrointestinal transit time as a
13
function of particle size. I
can't manipulate this
14 in
this presentation but the dissolution time
15
increases dramatically as the drug solubility
16
decreases. Particle size becomes
a critical factor
17 for
low solubility drugs. Of course,
everyone
18
realizes that but it is not particle size that we
19 put
into the formulation, it is the particle size
20
that comes out of the formulation in the
21
gastrointestinal tract. So, those
process
22
variables are important.
23
[Slide]
24
Some of the factors in the
25
gastrointestinal tract then are gastric emptying,
39
1
intestinal transit, position dependent permeability
2
along the gastrointestinal tract--duodenum,
3
jejunum, ileum and colon and, of course, intestinal
4
mucosal cell metabolism, and in particular CYP3A4
5
which is highly expressed and differentially
6
expressed along the gastrointestinal tract, and
7
potentially PGP expression along the
8
gastrointestinal tract.
9
[Slide]
10
To give you an example of variability in
11
gastric emptying rates, we can just look at the
12
light blue because that is administered with 200
13 ml,
the approximate glass of water that we use.
We
14
used 200 ml here because we did this before we got
15
involved in drug regulatory standards and realized
16
that a glass of water was the
17 the
standard in
18
that out, what is a glass of water in
19
with 200 ml you can see that the variability in
20
gastric emptying. Depending on
when you dose in
21 the
fasting state, it ranges from 5 minutes to
22
about 22 minutes. There is about
a 4-fold
23
variation in gastric emptying rate depending on
24
when you administer to a particular subject. This
25 is
because of the different contractual activities
40
1 in
the fasted state, shown here as phase 1, 2, 3
2 and
4.
3
[Slide]
4
Clearly, intestinal transit--again, this
5 is
a movie but I can't show it with this
6
presentation--transit through the gastrointestinal
7
tract where the drug is released in the duodenum.
8 It
has a very short transit time, maybe 10, 15
9
minutes through the duodenum, jejunum, ileum and
10
colon. The dissolution rate,
particularly of a low
11
permeability drug where the permeability appears to
12 be
the rate-determining step to absorption, the
13
permeability profile along the gastrointestinal
14
tract is very important.
15
[Slide]
16
There are about 10 L of fluid processed in
17 the
gastrointestinal tract per day, actually
18
depending on which book you read, 8 to 10. Of the
19 10
L that are processed, only about 2 L are
20
actually ingested as external.
The other 8 L are
21
ourselves. We are continually
secreting and
22
reabsorbing not only fluids but cells and proteins
23 and
other ions that are secreted into the intestine
24 so
there is a tremendous amount of variability and,
25 of
course, food has a large impact on that as well.
41
1 So,
that is a major factor that can be involved in
2 the
variability and dissolution and absorption in
3 the
gastrointestinal tract.
4
[Slide]
5 I show here just ranitidine, a low
6
permeability drug. This is animal
data. I don't
7
have human data and, in fact, it is very hard to
8 get
human data although there is some data
9
available. The duodenum, jejunum,
ileum--there is
10 a
significant difference in permeability.
So, you
11 can
envision a slowly dissolving ranitidine
12
product--I don't know if there are any, but
13
releasing in the ileum would have very poor
14
absorption. So, dissolution for a
low permeability
15
drug is probably more important because, in
16
general, the permeability in the upper part of the
17
gastrointestinal tract is more important or higher,
18 I
should say.
19
You know, we used to use language like
20
"rapidly but incompletely absorbed." You would see
21
that in the literature after analysis of
22
pharmacokinetic data and I would say how can that
23
be? It doesn't make sense to
me. If it is rapid
24 it
should be well absorbed. Right? Clearly, there
25 has
to be position-dependent permeability and the
42
1
absorption rate must decrease dramatically at some
2
point very quickly after the drug is administered.
3
Presumably, that is the result of drug getting into
4 the
ileum or distal in the small intestine where
5
there is lower absorption.
6
[Slide]
7
PGP--this is some immunoquantitation
8 results on CYP3A4 showing the variation in
the
9
duodenum, ileum and colon, much less in the colon
10 so
that there is less metabolism, particularly if
11
there is a controlled release formulation releasing
12
drug in the colon and, of course, much more in the
13
liver. I don't know, maybe Leslie
is going to say
14
more about the metabolism source of variability,
15
maybe not. You are shaking your
head, no.
16
[Slide]
17
I am going to propose that we classify the
18
drugs, highly variable drugs using BCS.
Here is
19
what I think we would see. We
need to actually
20
look at particular drugs. In
fact, I would like to
21 see
a list of drugs perhaps based on the
22
variability of reference products, whatever we
23
could find today, develop a list of highly variable
24
drugs or that we think might be highly variable,
25 and
then look at their properties and decide what
43
1 are
the likely sources of variability.
2
Anyway, I know there are certain so-called
3
highly variable drugs that are Class I drugs. They
4
have to be low dose, low solubility drugs but they
5 are
soluble enough to dissolve in a glass of water.
6
That is our criteria at the present time. So, if
7
those drug products dissolve rapidly--if they do; I
8
don't know if they do, we should look at that and
9 it
is over; there is no issue. It is all
biologic
10
variability; nothing to do with the product
11
variability. Again, that is a
hypothesis.
12
Probably the majority of the drugs that
13 are
highly variable are in Class II where there is
14 low
solubility, potentially Class IV for some
15
higher molecular weight compounds.
There, the
16
solubility-dissolution metabolism interaction can
17 be
difficult to separate and that is where we would
18
need to look more carefully at the drug products to
19
determine whether it is the solubility and
20
dissolution variability or whether it is a
21
metabolism variability that is leading to the high
22
variability in plasma levels.
23
[Slide]
24
So, I think that the BCS classification
25 can
help focus on the source of the high
44
1
variability. Then, in the case of
rapid
2
dissolution of Class I and Class III drugs a
3 dissolution
standard may be enough. There may not
4 be
too many highly variable drugs because I think
5 the
majority would be the low solubility Class II
6 or
Class IV drugs and there I think metabolism
7
and/or dissolution can be the source of
8
variability. In the case of
metabolism, the
9
metabolism variation would be due to the
10
variability and dissolution and presentation along
11 the
gastrointestinal tract. So, again, it
comes
12
back to a dissolution issues.
13
In fact, I would propose that we look more
14
carefully at the highly variable drugs, the sources
15 of
variability, again asking the critical question
16
what is the best test? What is
the best test? I
17
will go back to the original implementation of BCS
18 in
the case of high solubility, high permeability,
19
rapidly dissolving drugs. Plasma
levels are
20
telling us nothing about the product differences.
21 It
is only telling us about gastric emptying
22
differences at the time of administration of
23
patients or subjects. So, again,
focusing on
24
dissolution and classification I think can help us
25
unravel and simplify some--maybe not all. Maybe
45
1 not
all of the highly variable drugs can be
2
simplified this way but I think some of them can be
3
simplified this way. For those
drugs that are
4
complicated, we just say they are complicated.
5 Take a drug like premarine. You have already
6
mentioned that, Marvin. I think
that premarine is
7 a
complicated drug. That is life; that is
the way
8 it
is. It is too complicated for us to
unravel
9
today because of the way we regulate drugs and
10
approve drugs. So. I am happy to answer any
11
questions by the committee.
12
DR. KIBBE:
Questions, folks? Jurgen?
13 DR. VENITZ:
I agree with you, I am very
14
much in favor of identifying sources of variability
15 and
what you are presenting are obvious sources of
16
variability, and it always bothers me when we talk
17
about highly variable drugs and they are defined
18
phenologically. All we are doing
is a clinical
19
study. We are measuring Cmax and
AUC and we find
20
that they vary a lot, and that is the end of it,
21 and
now let's change criteria to see whether they
22 can
fit bioequivalence. So, I agree with you
on
23
that.
24
What I won't agree with you, at least not
25
fully, is that it is all a dissolution issue. I
46
1
think you are ignoring, in my mind at least, the
2
effects that excipients may have that could be very
3
different between formulations so that may not have
4 an
impact on dissolution but may have an impact on
5 pH,
may have an impact on permeability and may have
6 an
impact on GI metabolism. Now, I don't
know
7
whether that is a significant problem or not but I
8
think it is more than dissolution that you are
9
looking at. It doesn't preclude
what you are
10
recommending, which is basically do dissolution
11
tests and find out if that is an issue and then see
12 how
that matches your in vivo data. That is
just a
13
comment.
14
DR. AMIDON: If we extend the
dissolution
15 to
dissolution of the excipient, that is, the
16
dissolution of the excipient and the drug, then I
17
think we would be okay; I think my statement would
18 be
okay.
19
DR. VENITZ: But if you have
products that
20
have different excipients, that is my point.
21
DR. AMIDON: Yes, okay.
22
DR. VENITZ: As you said, life is
23
complicated. Sometimes it works;
sometimes it
24
doesn't.
25
DR. AMIDON: Right. So, that is the
47
1
function of what is the source of the variability.
2
DR. VENITZ: Yes.
3
DR. KIBBE: Ajaz?
4
DR. HUSSAIN: I worked with Gordon
for
5
many years on developing the BCS guideline, and so
6
forth, and we actually did examine that very
7
question of excipients and their impact not only on
8 the
dissolution process but on permeability and
9
metabolism and it is a serious issue and I think we
10
learn more about transport every day.
Therefore,
11
clearly, I think when Gordon mentioned dissolution,
12 we
have discussed that so many times and we always
13
include that as a source of variability and that
14 has
to be considered.
15
But, Gordon, I wanted to push you in a
16
different direction. One of the
hesitations as we
17
developed the BCS guidance was the reliability of
18 the
in vitro dissolution test. We were not
19
confident that the current test really was good
20
enough to extend it to the slower releasing
21
products. So, that was the reason
we crafted
22
rapidly dissolving and said dissolution is not rate
23
limiting and, therefore, we can rely on the current
24
dissolution test to do that.
25
I think as we move forward here, I think
48
1
what we have done with the PAT initiative is to
2
sort of say, all right, let's really ask the
3
question what are the criteria variables, what are
4 the
root causes of this. So, go back to the
basics
5 as
to particle size, and so forth, and if you
6
really understand those relationships then you have
7 a
better link between your formulation and your
8
excipients; you have your process directed to the
9
clinical relevance. So, that is
the opportunity
10
that technology is offering us to do that without
11
having to do an artificial in vitro test where
12
questions keep continuing and increasing with
13
respect to the relevance of that in vitro test.
14
DR. AMIDON: I certainly obviously
agree,
15
Ajaz. We have talked about these
issues for many
16
years. I did use the word in vivo
dissolution.
17
There is a big step from in vitro to in vivo. I
18
don't think it is magic; it is just complicated and
19 I
think we can figure that out. I think we
can
20
determine for any particular drug what might be a
21
good representative dissolution test, and I might
22
call that a bioequivalence dissolution test rather
23
than a QC, quality control, dissolution test. But
24 you
are absolutely right. The issue is
really in
25
vivo dissolution and how do we capture that in some
49
1 in
vitro methodology. I don't think we have
2
thought about that very hard at all.
I am not sure
3
why. We use the term dissolution
very generically
4
when it should be much more specific.
5
DR. KIBBE: Les wants to comment
and then
6
Nozer. Can you make a comment,
Les, because you
7 are
not part of the committee?
8
DR. BENET: They said as a visitor
I can.
9 I
wanted to comment on BCS and what Jurgen brought
10 up
in terms of the excipients. When we
initiated
11 BCS
I was very strong concerning the potential for
12
excipients on Class I drugs and we have written the
13
rules to make sure that these excipients don't have
14 an
effect. In fact, I now recognize that
with
15
Class I drugs that is not a problem, that the
16
excipients won't be a problem in terms of affecting
17 at
least the transporters. But they will be
a
18
problem with Class III drugs.
19
So, so far I have been very opposed to
20
moving the Class III drugs because I can make a
21
Class III formulation that will pass dissolution,
22 any
dissolution, and fail. The reason is
that
23
Class III drugs need uptake transporters to get
24 absorbed
and, therefore, I can block an uptake
25
transporter in the gut with a substance that has no
50
1
dissolution criteria. So, I still
think we are a
2
little early in translating this dissolution
3
criteria beyond Class I, but I think we were
4
correct in Class I and the extra safeguards we put
5 in
actually really turn out not to be necessary.
6
DR. SINGPURWALLA: I like this
concept of
7
looking at the causes of variability.
I see this
8 as
a first step towards going to a Bayesian
9
alternative for the existing methodology that was
10
criticized by the first speaker.
But I do have a
11
question perhaps both for you and also for the
12
first speaker. Has anybody looked
at the
13
reliability of the testing instrument itself?
14
Because if the testing instrument itself shows a
15
large variability--if the instrument itself shows a
16
large variability then you don't know whether the
17
variability is coming from the instrument or from
18 the
particular drug or the combination of the
19
instrument, the drug and the patient.
20
DR. KIBBE: Anybody? Who wants to handle
21 that?
22
DR. VENITZ: I think by instrument
what
23 you
mean is the human being used in those studies.
24 Are
you talking about dissolution or are you
25
talking about in vivo?
51
1
DR. SINGPURWALLA: Both.
2
DR. VENITZ: Well, then let's talk
about
3 in
vivo and I will leave it up to you to talk about
4
dissolution. What you are looking
at is the Cmax's
5 and
the areas under the curves. They do not
only
6
depend upon absorption and dissolution; they depend
7 on
everything that happens after the drug gets in
8 the
body, which is something we are not interested
9
in. If that contributes
significantly to the
10
variability, then you are looking at primarily
11
variability and disposition which determines why we
12
have a highly variable drug, not because there is
13
variability in absorption. So,
your instrument
14
would be a very noisy instrument I think, to use
15
your lingo.
16
DR. SINGPURWALLA: Right. You have an
17
instrument by which you measure these things, like
18 a
thermometer. If your thermometer is
bad--
19
DR. VENITZ: I am saying that for
some
20
drugs it could well be that you have a very noisy
21
instrument and the noise is not related to what you
22 are
trying to measure.
23
DR. SINGPURWALLA: Exactly.
24
DR. KIBBE: Let me just take the
25
prerogative of the chair for half a second and then
52
1 I
will let you speak. It is very difficult
for us
2 to
understand the real noise level of the
3
instrument. The instrument is the
bioequivalency
4
test itself and the agency gets submissions with
5
bioequivalency tests that are passed.
The question
6 is
how many were done that failed before the one
7
that passed, and what was done to make that work?
8
I think if you go back and we got a bunch
9 of
data together, which we can't but it would be
10
interesting to look at, we would find that the
11
instrument is very crude and the reason we live
12
with it is that it is close to the clinical
13
therapeutic outcomes that we really want to measure
14 in
terms of steps away from that outcome.
What
15
Gordon is recommending is that we even eliminate
16 the
human from our decision-making process, which
17
brings us further away from the ultimate goal which
18 is
to know that it therapeutically equivalent, and
19 we
have to be sure that our predictor is going to
20
hold true. Those are the problems
I think that we
21 all
have been struggling with for 25 years.
22
DR. HUSSAIN: Now I have three
comments.
23
With respect to the instrument variability, I think
24 it
is a very important question. In the
case of
25
bioequivalence testing we try to minimize that and
53
1 try
to make it more precise and more accurate by
2
doing a crossover study. We test
the two products
3 in
the same patient in a crossover fashion.
So,
4
that is our attempt to minimize that.
The other
5
attempt that we had to minimize is to get a group
6 of
more similar individuals but we wanted to move
7
away from that in the general population because
8 the
crossover is a way to minimize that. I
also
9
pointed out with respect to variability the
10
dissolution test. I think as we
think about that,
11 we
need to address that.
12
But the point I think, going back to the
13 key
question, is what are the important questions
14
here? Dr. Kibbe's comment was, in
a sense,
15
bioequivalence. For therapeutic
equivalence our
16
approach is very simple. First
you need to be
17
pharmaceutically equivalent and then, if there is a
18
need, you do a bio study. For
example, for
19
pharmaceutical equivalence for solutions you don't
20
need a bio study. So
pharmaceutical equivalence,
21
bioequivalence and then therapeutic
22
equivalence--those come together to define that. I
23
could sort of generalize what Gordon has said, in a
24
sense if we understand our formulations, if we
25
understand our processes, if we understand the
54
1
mechanisms, pharmaceutical equivalence essentially
2 is
defining therapeutic equivalence.
3
DR. AMIDON: To come back to your
question
4
about the dissolution apparatus, there is a range
5 of
dissolution apparatus in the USP that are used
6 internationally,
and you can study many of the
7
variables that change in vivo by pH and surfactants
8 in
those apparatus. The apparatus
themselves have
9
been proven perhaps historically to be very
10
reliable, although you could argue maybe today that
11 we
could design a better apparatus but that is very
12
complicated because these things are used in many
13
companies internationally with defined procedures
14
that are approved by the regulatory agencies and
15
making change in an apparatus is a very complex
16
process.
17
But, yes, we can study the various
18
variables in vivo and I think that a dissolution
19
test that included changes in pH and surfactant to
20
reflect what is happening in vivo is something we
21
should do. We don't do that; we
just do fixed pH
22 and
follow the dissolution as a function of time.
23 So,
I don't think we use our apparatus very
24
insightfully actually.
25
DR. KIBBE: I would argue that the
way we
55
1 use
dissolution is reliable but insensitive, and we
2
need to do a lot more to be able to make that
3
conversion. Anybody else?
4
DR. MEYER: Gordon, I listened to the PAT
5
stuff all day yesterday and what I got out of it is
6
that it is applicable to this so the idea of why do
7 we
have variability--right now we are proposing to
8
potentially change our release specifications
9
because our product is too variable and that is not
10
acceptable in the manufacturing arena.
You go back
11 and
figure out why it is too variable. I
wonder
12 how
much data is really available on if I gave
13
myself a rapidly absorbed drug once for the next
14
three weeks, what would my profiles look like? I
15
don't know that there is a lot of data that shows
16
reproducibility in a subject, unless it was the old
17
multiple dose studies where the drug was
18
essentially eliminated in 24 hours.
19
So, I think we need some more information.
20 I
don't know, maybe the agency does this, but when
21 the
innovator firms do special populations and they
22
find the elderly are different than the young, do
23
they have to then go further and explain is that
24
gastrointestinal pH, is it transit, is it
25
metabolism, what is the reason for it.
Because I
56
1
think then we can get some background information
2 on
source of variability.
3
Just to bounce off an idea which is
4
undoubtedly ludicrous, do we need in a sense to
5
prescreen some subjects so we have a calibrated man
6 or,
if you will, a USP man or woman that is then
7
allowed into the study so if they have less
8
variability they get into our study?
Could we do
9
that? One thing that really
troubles me is the
10
current policy, and I understand why it is and I
11
think I support it, of having different mechanisms
12 of
release tested against each other in a
13
bioequivalence study, an oral study versus a
14
particular dosage form.
Intestinal transit can
15
have a profound difference on those two so if you
16
have a uniform man, that uniform man may show them
17 to
be equal but if you throw in a vegetarian, that
18
vegetarian might show the oral tablet is excreted
19 in
four hours and the other person may take much
20
longer. So, just some support
really for the idea
21 of
knowing where the problems are; can we reduce
22
variability somehow; are subjects legitimately--is
23
that a viable approach?
24
DR. AMIDON: I don't know, I am
not sure I
25
would want to take on preselecting subjects because
57
1
what criteria are you going to use?
Normal in what
2
sense?
3
DR. MEYER: I am thinking more in
terms
4 of,
say, rapid metabolism or poor metabolism.
We
5 do
that now somewhat routinely.
6
DR. AMIDON: Right.
7
DR. MEYER: So, we might give a
8
panel--CROs now, they use the same subjects over
9 and
over again anyway. Let's characterize
them
10
first before they are allowed into subsequent
11
studies.
12
DR. KIBBE: Paul, go ahead.
13
DR. FACKLER: If I can just
comment on
14
that, we used to do bioequivalence studies in males
15
only and restricted their ages from 18 to 45, I
16
believe. The agency has recently
requested that BE
17
studies be done in a larger group of people, more
18
representative of the American population so we now
19
include females and we include the elderly, and it
20
just makes the variability problem that much worse.
21 I
mean, I agree completely that ideally if we would
22 get
15 people all exactly the same way, all exactly
23
with the same physical habits, generally with the
24
same diet, it would make BE studies easier to pass
25
because we have reduced the variability in the
58
1
subjects. But the agency has been
going, at least
2
recently, in the opposite direction, making these
3
products in particular less likely to pass against
4
themselves again.
5
DR. KIBBE: It is my impression,
and I am
6
sure the FDA people will correct me, that they are
7
trying to get two answers using one study, and that
8 is,
are the two formulations behaving the same,
9
should be their behavior independent of the
10
subjects studied, and are there variabilities
11 between
product-subject interactions that might be
12
significant in special populations.
I think it is
13
really hard to do that in one study, and that is
14 one
of the problems you are running into.
What I
15
think Gordon is suggesting is if we understood the
16
variables we might not have to use that blunt a
17
tool to estimate what will happen in the average
18
patient.
19
I would love to see us be able to do that.
20
There was a wonderful report done--Les will
21
remember because he is almost as old as I am--by
22 the
agency that looked at dissolution and tried to
23
correlate it with bioequivalency data that they had
24
almost twenty years ago and there was absolutely no
25 way
that dissolution predicted any of the results
59
1
that they got on those studies.
So, it is more
2
complicated than it first appears.
3
DR. AMIDON: I got involved in
this
4
process about that time, and my position is you
5
just did the wrong test.
Okay? That is the
6
problem. So, it is a matter of
refining the
7
dissolution test to make it more relevant to the
8
variables that we need to control to ensure
9
bioequivalence. We haven't done
enough of that.
10
DR. KIBBE: Ajaz, you have a
comment?
11
DR. HUSSAIN: The key aspect I
think is
12
that we need to keep the focus on asking the right
13
questions and if a bioequivalence study is only
14
for, you know, males 18 to 45, is that the right
15
question from the public health aspect because the
16
product is going to be used in all populations?
17 So,
you really have to go and look at the
18
fundamentals of what is a bioequivalence study. If
19 it
is just confidence interval criteria, then that
20 is
one aspect.
21
DR. SINGPURWALLA: Why not have a
separate
22 set
of drugs for different categories of people?
23
Like, you know, you have cholesterol drugs 20 mg,
24 10
mg and you specify your milligrams based on the
25
population.
60
1
DR. HUSSAIN: That is a major
aspect of
2
dose finding and then labeling that goes into the
3 new
drug development process itself. The
4
bioequivalence essentially has been a quality
5
assurance approach to making sure that a
6
pharmaceutically equal product has an in vivo rate
7 and
extent of absorption similar to the innovator.
8
That is one of the main reasons for doing the bio
9
study, to make sure that your assumptions and your
10 in
vitro methods are more reliable or at least
11
conform from that perspective.
12
DR. KIBBE: Thank you. Unless someone
13
else has a comment we will let you off the hook for
14 a
few minutes, and go to Dr. Benet who will
15
enlighten us.
16
Clinical Implications of Highly Variable Drugs
17
DR. BENET: I am older!
18
[Laughter]
19
Thank you. It is a pleasure to be
here.
20 I
think the last two times I have appeared before
21
this committee I stayed in my office but it is nice
22 to
be here in person, and I thank you for the
23
opportunity.
24
25
We have been discussing at an
61
1
international level, I was reminded as I heard
2 this,
for 15 years--we held our first sort of
3
consensus conference in 1989 to try to develop
4
standards for bioequivalence and we are still at
5 it.
6
[Slide]
7
This was said by the first speaker but
8 this
is a slide that is now maybe 12 years old, or
9 at
least parts of it. The current U.S.
Procrustean
10
bioequivalence guidelines: the manufacturer of the
11
test product must show using two one-sided tests
12
that a 90 percent confidence interval for the ratio
13 of
the mean response--usually the area under the
14
curve and Cmax--of its product to that of the
15
reference product is within the limits of 0.8 and
16
1.25 using log transformed data.
It is
17
Procrustean, and those of you who don't remember
18
your mythology, the Procrustes himself was a robber
19
that took people when they came through his gate
20 and
put them on his bed, the Procrustean bed.
If
21
they were too long he cut off their feet. If they
22
were too short he stretched them out until they fit
23 the
bed. And, that is exactly what we have,
24
Procrustean guidelines that say all drugs must fit
25 the
same criteria no matter what the issues are.
62
1
Now, BCS, biopharmaceutical classification
2
system, is non-Procrustean. It is
an advance and
3 the
obvious answer, Arthur, to why a study failed
4 in
looking at dissolution is that we didn't
5
understand the flawed classifications.
So, the
6
only time dissolution is going to have any
7
relevance to bioequivalence or bioavailability is
8 for
Class I and Class III drugs. Since we
looked
9 at
all drugs about 20 years ago, we were obviously
10
going to fail. So, we are making
some advances.
11 But
I strongly believe and have suggested over a
12
number of years that there need to be other
13
non-Procrustean advances and that is what I will
14
talk about today.
15
[Slide]
16
What are we trying to solve? What
are the
17
bioequivalence issues and what concerns patients
18 and
clinicians so that they have confidence in the
19
generic drugs that are approved by the regulatory
20
agencies so that they feel there are no questions
21
related to their therapeutic efficacy?
22
It doesn't help to tell them--and that is
23 a
true fact, it doesn't help to tell them that
24
there has never been a drug that passed the U.S.
25 FDA
bioequivalence issues that ever caused any
63
1
therapeutic problems in a prospective study. That
2
doesn't help them because they always say, well, it
3 is
the next drug and they have a lot of emphasis
4 out
there from people who would like them to
5
question the bioequivalence criteria.
So, this is
6
always in my mind, that one of the major issues
7 that we face is not necessarily scientific but
it
8 is
creating an environment where the American
9
public has confidence in the regulations that we
10 use
and the drugs that we say can go on the market.
11
But what we have done and what our
12
concerns are now with therapeutic index drugs, NTI,
13 we
need to have practitioners have assurance that
14
transferring a patient from one drug product to
15
another yields comparable safety and efficacy, and
16 a
few years ago we termed that switchability and we
17
developed or tried to develop a number of
18
statistical criteria to approach that.
The issues
19 we
are facing today are for a wide therapeutic
20
index, highly variable drugs which do not have to
21
study an excessive number of patients to prove that
22 two
equivalent products meet the preset one size
23
fits all statistical criteria.
So, these are the
24
issues I want to address and ask the committee to
25
take cognizance of.
64
1
[Slide]
2
Now, it was not obvious a few years ago
3 but
it is very obvious today that if you have a
4
narrow therapeutic index drug it is very easy to
5
pass the bioequivalence criteria, and that is
6
because narrow therapeutic index drugs, by
7
definition, must have small intra-subject
8
variability. If this were not
true for narrow
9
therapeutic index drugs, patients would routinely
10
experience cycles of toxicity and lack of efficacy,
11 and
therapeutic monitoring would be useless.
So,
12 in
fact, it is not an issue. Narrow
therapeutic
13
drugs we take care of and we do very well from a
14 scientific issue. We might not have the
15
confidence, and I will come back and address that.
16
[Slide]
17
Let's look at some narrow therapeutic
18
index drugs. They have high
inter-subject
19
variability and they have low intra-subject
20
variability. That is why we don't
have to worry;
21
when we get the patient to the right place, they
22
stay there. The question was are
they all Class I,
23
Class II. Theophylline is a Class
I drug. So,
24
there are drugs on this list that are Class I drugs
25
although most of them are Class II drugs.
65
1
Getting back to the reliability of the
2
instrument, I would just like to make a comment.
3
Look at the warfarin sodium intra-subject
4
variability. The clinical measure
that the
5
clinician uses to judge the status of the patient
6 in
terms of his blood thinning capability, the INH
7 measurement,
is significantly more variable. So,
8 in
fact, what the clinician does in testing if the
9
drug is working is more variable than the patient
10 is
going to experience from dose to dose in terms
11 of
the criteria for this particular drug.
So,
12
these are interesting questions.
13
[Slide]
14
Now, we tried to address this
15
switchability issue over a long period of time with
16 the
concept called individual bioequivalence, and I
17 chaired
the expert panel for about three years and
18
tried to address this issue. The
ideas about
19
individual bioequivalence were that we were going
20 to
get these promises, we would address the correct
21
question, switchability in a patient.
We would
22
consider the potential for subject by formulation
23
interaction. There would be
incentive for less
24
variable test products. Scaling
would be based on
25
variability of the reference product both for
66
1
highly variable drugs and for certain
2
agency-defined narrow therapeutic range drugs.
3
And, we would encourage the use of subjects more
4
representative of the general population.
5
In fact, none of that worked and we gave
6 up
on it. So, did it address the correct
question?
7
Well, the question was, was there even a question
8 and
was there any necessity for this at all, and
9
there is no evidence that the present regulations
10 are
inadequate and that we need to be more rigorous
11 in
our definition related to switchability.
12
[Slide]
13
Consider that the subject by formulation
14
interaction turned out to be an unintelligible
15
parameter from both the agency and the exterior
16
scientific community.
17
Incentive for less variable test products,
18
yes, but that could be solved by average
19
bioequivalence scaling and that is what at least I
20 am
here to talk about today.
21
Scaling based on variability of the
22
reference product both for highly variable drugs
23 and
for certain agency-defined narrow therapeutic
24
index drugs, again average bioequivalence with
25
scaling could solve this issue.
67
1
Encourage the use of subjects more
2
representative of the general population, that was
3 a
good hope but it completely failed in terms of
4 how
people designed their study. So, it
didn't
5
work.
6
[Slide]
7
I recognized in Lawrence's introduction
8
that the FDA doesn't have a definition for highly
9
variable drugs. This is the
consensus definition
10
that came out of a number of international
11
workshops, highly variable drugs should be those
12
when the intra-subject variability is equal or
13
greater than 30 percent. The idea
is that for wide
14
therapeutic index highly variable drugs we should
15 not
have to study an excessive number of patients
16 to
prove that two equivalent products meet this
17
preset one size fits all statistical criteria.
18
This is because, by definition, again
19
highly variable approved drugs must have a wide
20
therapeutic index, otherwise there would have been
21
significant safety issues and lack of efficacy
22
during Phase III testing. In
fact, highly variable
23
drugs fall out; don't get to the market.
They fall
24 out
in Phase II because the company can't prove
25
that they work and they can't prove that they are
68
1
safe. So, we don't have highly
variable narrow
2
therapeutic index drugs. We only
have drugs that,
3
with this tremendous variability that we
4
potentially saw in the first speaker's slide, don't
5
have any problems. And, those
individual patients
6
having very high levels one time, low levels the
7
next time, high areas under the curve one time, low
8
areas under the curve the next time get through.
9 In
fact, for those highly variable drugs we don't
10
need to worry about the genetic differences in
11
their enzymes. It has already
been shown that,
12
yes, there are tremendous differences.
Somebody is
13
going to have very high levels because they lack
14 the
enzyme; somebody is going to have very low
15
levels but still they are safe and effective
16
because they are wide therapeutic index drugs.
17
[Slide]
18
But it makes it very difficult, as was
19
also pointed out by the first speaker, to get them
20 to be
bioequivalent and here is my champion or what
21 I
think is the champion from the data that I have
22
seen, and this is progesterone which I believe is
23 the
poster drug for highly variable variability.
A
24
repeat measures study of the innovator's product
25 was
carried out in 12 healthy post-menopausal
69
1
females and it yielded intra-subject variability in
2 an
AUC of 61 percent for the coefficient of
3
variation and intra-subject coefficient of
4
variation for Cmax of 98 percent.
5
If you did the calculations, it came out
6
that you needed 300 women just to meet the
7
statistical criteria and, in fact, this was not a
8
study that a generic company, or at least the
9
company interested in this, could afford to carry
10 out
because, for sure, we know that the way we
11
design the studies there is a chance, even if you
12 had
the right numbers, that one out of ten or one
13 out
of five studies would fail just on statistical
14
chance and you have carried out a study with 300
15
people in it to prove that this highly variable
16
drug is bioequivalent. This is
the issue that we
17 are
asking you to talk about today, and can we
18
solve this problem so that we don't have highly
19
variable, very safe, wide therapeutic index drugs
20 for
which we can't prove bioequivalence because of
21 the
inherent variability of the innovator product.
22
[Slide]
23
I appeared before this committee three and
24 a
half years ago to give the recommendations of the
25 FDA
expert panel on individual bioequivalence, and
70
1
these are some of the recommendations.
One that I
2
didn't put on here is that all generic drug studies
3
must be submitted to the agency, and I am very
4
pleased that that has happened and congratulations
5 to
the agency.
6
Our recommendations at that time were that
7
sponsors may see bioequivalence approval using
8
either average bioequivalence or individual
9
bioequivalence, and we recommended that the subject
10 by
formulation parameter be deleted since no one
11
knew what to do with it and we couldn't justify it
12
statistically.
13
We asked that scaling for average
14
bioequivalence be considered, that the agency and
15 the
statistical group go into this and it be
16
something to be followed up and presented to this
17
advisory committee at some time in the future.
18
We recommended at that time that if an IBE
19
study, individual bioequivalence study, was carried
20 out
and the test product fails you could not then
21
reanalyze with average bioequivalence because in
22
those days we said you had to pick one or the
23
other.
24
Here is something that we recommended that
25 I want to bring up again today because this
has to
71
1
do with confidence. We recommended the point
2
estimate criteria be added, and we added this not
3 on
any scientific basis that we are going to rule
4 out
products, we said that these criteria are
5
always met today and what we have is a conception
6 or
a view outside that it would be possible to have
7
products that differ by 25 percent, and that we
8
would be well served if we would say let's put a
9
point estimate criterion in addition to our
10
criteria--AUCs of at least plus/minus 15 for point
11
estimate criteria and Cmax plus/minus 20 percent no
12
matter what you do, and if you have narrow
13
therapeutic index drugs make it even smaller, make
14 the
point estimate plus/minus 10 percent for AUC
15 and
plus/minus 15 percent for Cmax.
16
[Slide]
17
So, what I am suggesting here today and
18
what I am recommending to the committee to do is
19 ask
the agency to develop methodology, and we are
20
going to hear some, to allow approval based on
21
weighting of average bioequivalence analytical for
22
highly variable drugs so that we can bring some
23
drugs to the market that can't be studied because
24 of
the progesterone example. Also, that the
point
25
estimate criteria be added to the criteria because,
72
1 in
fact, all products will pass these criteria at
2 the
present time and we won't be harmed, or we will
3
increase the confidence of those that say, you
4
know, you could have two products that differ by 50
5
percent because look at what the FDA criteria say.
6
Now, the FDA criteria, as they used to be
7
written two years ago, were easily misinterpreted
8 but
that also changed two years ago and now the
9
criteria are written in a way that no clinician can
10
understand them in the first place so they won't be
11
misinterpreted.
12
[Laughter]
13
They still say exactly the same thing but
14
they can't be misinterpreted to say you could have
15 two
products that differ by 50 percent. So,
these
16 are
my recommendations. Thank you for
listening to
17 me.
18
DR. KIBBE: Questions for Dr.
Benet?
19
DR. SINGPURWALLA: I have a
comment not
20
just to you but to everyone else.
This example of
21
highly variable drugs shows, to me, how the drug
22
industry is buried under the tombstone of
23
frequentist methods. Such methods
ignore clinical
24 and
biopharmaceutical knowledge, and it is bogged
25
down by its own weight.
73
1
DR. BENET: I disagree.
2
DR. SINGPURWALLA: Why?
3
DR. BENET: I think you are coming
to this
4
fresh and that is good, but what we are interested
5 in
is safety and efficacy, and in all cases
6
measures of safety and efficacy are more variable
7
than any pharmacokinetic measure.
What we are
8
really interested in, what the agency is interested
9 in
is safety and efficacy.
10
DR. SINGPURWALLA: Who said that
Bayesian
11
methods do not incorporate high variability? It is
12
these confidence intervals and these confidence
13
limits, and the comment you make is a failure to
14
understand Bayesian methods.
15
DR. BENET: I understand Bayesian
methods.
16
DR. SINGPURWALLA: No, you don't;
you
17
wouldn't say this.
18
DR. BENET: Well, I welcome the
19
committee's spending the time discussing this with
20 you
and if you adjourn I get to go home.
21
[Laughter]
22
DR. MEYER: I think I agree with
23
everything you have said and it embarrasses me no
24 end
to say that!
25 [Laughter]
74
1
Is there still going to be a perceived
2
problem when you have, let's say, a Cmax point
3
estimate of plus/minus 15 percent?
Isn't that
4 going to solicit illustrations of, well,
look, my
5
Cmax was 115 units and their Cmax was 85 and the
6
high and low can be switched in the marketplace?
7
DR. BENET: I think we are never
going to
8 get
around that. There are always going to
be
9
people who will take the present situation and use
10 it
to their marketing advantage. So, I
don't think
11 we
can get around that. You know, we have
the same
12
issues today. I am not sure that
everyone on the
13
committee is aware that in terms of BCS Class I,
14
where you don't have to do a clinical study--I
15
don't know of a generic company that has used that
16 for
exactly the reason you are bringing up, Marvin.
17
They would be afraid that someone will go out there
18 and
say this product has never been tested in
19
humans; it was approved on the basis of a
20
dissolution. You have confidence
in this product
21 so
that people that use BCS Class I at the present
22 time
are the innovators who use it when they have a
23
SUPAC change or something like that.
So, I think
24 we
are always going to face that, and I think what
25 we
need to do is just try to do the best job that
75
1 we
can in making it happen.
2
DR. KIBBE: Let me just ask about
an
3
application of one of your recommendations to your
4 own
example. If you use methodology that is
5
developed as a weighted average, how would that
6
play out with progesterone? In
other words, what
7
kind of numbers would we start to work with?
8
DR. BENET: I mean, I do agree
with
9
weighting to the variability of the innovator
10 product.
In other words, that would be the term in
11 the
denominator that you would weight. But
there
12 are
different statistical issues that have to be
13
addressed that I can't do so we need the expert
14
statisticians to tell us how to approach that. But
15
that is what I want. I would want
a weighting on
16 the
variability of the innovator product in terms
17 of
the coefficient of variation for Cmax as one
18
criterion and for AUC as another criterion.
19
DR. KIBBE: I have always found
20
intellectually attractive the concept of three ways
21
where we could look at variability and then compare
22 it
to the generic. Is that going to help us
get to
23 the
numbers that we need to make these kinds of
24
decisions?
25
DR. BENET: Well, there is going
to have
76
1 to
be some measure of intra-subject variability.
2 We
need to know that, and I have requested the
3
agency for many years to make this a requirement
4 for
new drugs, that a measure of intra-subject
5
variability in humans or even in patients be
6
included in the approval process and be included in
7 the
package insert. So, we do have to have
that
8
measure some place.
9
I am very encouraged, even though the
10
agency does not require that, that we are starting
11 to
see with many new products, when you look at
12
their package insert, measures of intra-subject
13
variability included because it is important
14
criteria and value that clinicians want to know.
15
What is the inherent pharmacokinetic variability so
16
that then I can say is the pharmacodynamic
17
variability more than this inherent pharmacokinetic
18
variability. If they don't know
the inherent
19
pharmacokinetic variability, then they have a tough
20
time making any decision about whether the change
21 in
efficacy is related to pharmacokinetics or to
22
real variability. So, somebody
has to do this,
23
Arthur, and I think that has to come out of what
24 you
recommend.
25
DR. MEYER: Les, you put a little
bit less
77
1
weight on Cmax than you do on AUC; there is a less
2
stringent requirement. Is that
because Cmax is
3
more variable because we don't measure it very
4
precisely, or is it because Cmax is less important
5 than
AUC? And, I would quarrel that we don't
have
6
enough data for the latter conclusion.
7
DR. BENET: Well, in some cases we
do but,
8 as
was initially discussed, it is confounded.
As
9 we
all know, Cmax is a very confounded measure and
10 the
agency and many academics have spent years and
11
years in trying to develop a new measure. None of
12
them turned out to be any better.
So, it is very
13
confounded and, as was stated, is always more
14
highly variable than AUC. I know
of no case.
15
DR. MEYER: But it is the only
measure we
16
have that has any component of rate in it.
17
DR. BENET: That is correct, but
it is
18
more variable.
19
DR. VENITZ: Les, I agree with
your
20
additional recommendation to put constraints on the
21
point estimates. You mentioned
one of the reasons
22
being that the public needs to be reassured that,
23
indeed, no matter whether it is unintelligible
24 regulation or not, we do have generics that
are
25
bioequivalent.
78
1
What I am personally not certain about is
2
whether I agree with the reference scaling--and,
3
again, we are going to have some more presentations
4 on
that--because you are now, in my mind,
5
aggregating variance and mean differences, and I am
6 not
sure whether one can offset the other.
In
7
other words, if you have a large mean difference,
8 can
that be offset by differences in variance?
9
When we had the discussion last time with IBE,
10
surprisingly there were drugs out there in the
11
database that the FDA provided us with that passed
12 IBE
but wouldn't have passed ABE, which I think was
13
counter-intuitive for most of us, at least on the
14
committee, in terms that we expected IBE to be much
15
more conservative than ABE and it didn't turn out
16
that way. So, I still personally
withhold judgment
17 on
the reference scaling but I am very much in
18
favor of putting in additional constraints.
19
DR. BENET: Let me just answer
that. I
20
think having the additional constraints solves part
21 of
the problem.
22
DR. VENITZ: Yes, that was the
reason why
23 I
think the committee at that time went along with
24
that because we were worried about the IBE not
25
being conservative enough. Right
now you are
79
1
basically breaking drugs down into two categories,
2
NTIs and non-NTIs, in terms of the criteria that
3 you
are going to use or that you are proposing to
4 be
used for BE assessment.
5
DR. BENET: Yes.
6
DR. VENITZ: Can you think of
additional
7
criteria along the lines that we heard Gordon talk
8
about, that if we understand where the variability
9
comes from we might use different criteria? In
10
other words, is NTI the only thing that we have in
11
some decision tree that decides which way we are
12
going to go?
13
DR. BENET: As I said, the NTI
statement
14
there has nothing to do with science because it is
15
easy to prove bioequivalence of NTI drugs. It just
16 has
to do with confidence. So, that is why I
made
17 it
lower, because it is easy to pass.
18
I definitely believe that as we progress
19 we
are going to have different criteria, and I
20
think BCS has a real potential for it.
I have a
21 big
list, my BCS list, and I looked to see what
22
drugs were there and that is why I made sure that
23
theophylline was a Class I drug.
I think as we
24
progress--and I presented to the agency last
25
November my newest concepts in terms of using BCS
80
1 or
some sort of variant of BCS to actually predict
2
drug disposition, and I think we are going to
3
progress a lot in the next few years.
4
DR. KIBBE: Nozer?
5
DR. SINGPURWALLA: Well, just a
general
6
comment. I was pleased to hear
you acknowledge
7
that newcomers can identify things like
8
confounding, but I also think that newcomers can
9
look at an old problem and come up with new methods
10 of
addressing that. Therefore, I urge you
to pay
11
more attention to alternate methods and not get
12
committed to an old, archaic notion of confidence
13
intervals. These have been
criticized in the
14
literature. And, what we see here
is repeated use
15 of
confidence limits, and the difficulty that
16
confidence limits poses both to the FDA and also to
17 the
drug industry in getting their drugs approved.
18 So,
I am going to urge you to start paying more
19
attention to alternatives and don't dismiss it.
20
DR. BENET: I don't dismiss it,
and my
21
colleague, Dr. Scheiner, has spent a lot of time
22
informing the committee and the agency of these
23
approaches and the Bayesian approach, and I think
24 we
are all well aware of it and do recognize it.
25 It
is important to have fresh eyes and fresh views
81
1 of
these kinds of issues, but it is also important
2 to
recognize that the agency's criteria are safety
3 and
efficacy, and when we have criteria that have
4
never failed it is tough to say that we move beyond
5
that criteria to untested criteria in terms of this
6
particular issue. So, that is why
the agency must
7 be
very careful in the changes that they make.
8
DR. KIBBE: Thank you, Les. We have one
9
more speaker before the break.
Dr. Endrenyi,
10
welcome.
11
Bioequivalence Methods for Highly Variable Drugs
12
DR. ENDRENYI: Thank you.
13
[Slide]
14
This presentation was put together with
15
Laszlo Tothfalusi and I would like to acknowledge
16
that.
17
[Slide]
18
I would like to raise a number of
19
questions which I believe that this committee will
20
have to make recommendations about eventually that,
21
certainly, the agency ought to consider.
I would
22
like to go through the first part fairly quickly
23
because much of that has already been considered.
24 So,
we have the usual criterion of comparing two
25
formulations and the confidence limits for the
82
1
ratio of geometric means should be between 0.8 and
2
1.25. This has already been
stated.
3
[Slide]
4
It has also been stated that for highly
5
variable drugs this presents a problem because with
6
large variations it is very easy to hit that 0.8 to
7
1.25 and, therefore, many subjects may be needed in
8
order to satisfy that.
9
[Slide]
10
For the purpose of this presentation but
11 not
necessarily as the final word at all, the
12
coefficient of variation has been considered
13
exceeding 30 percent for highly variable drugs.
14
[Slide]
15
This slide would simply ask is there an
16
issue and this has already been asked and the
17
answer was probably yes. In this
case, two
18
formulations of isoptin are considered in the same
19
subject repeatedly, and two different occasions
20
different relationships between the two
21
formulations were obtained. So,
it looks as though
22 the
drug is not really bioequivalent with itself
23 and
that is a concern, but this has already been
24
demonstrated by Dr. DiLiberti.
25
[Slide]
83
1
This is perhaps more recent. This
was
2
obtained from Diane Potvin, from MDS, who
3
demonstrated that, indeed, things look reasonable
4 as
long as the intra-individual CV is up to about
5 70
percent but beyond that it is very difficult to
6
satisfy the criteria. There are
many, many studies
7
submitted that failed.
8
[Slide]
9 Then she went on, very kindly, to
look at
10
details of these highly variable drugs.
From this,
11 one
could conclude that there is a relationship
12
between the coefficient of variation and failure
13
rate, higher failure rate with higher coefficient
14 of
variation. Mind you, these are all
submitted
15
studies so this analysis is still biased because
16 the
company submitted them in the hope that they
17
would pass, so these are not all studies at all.
18 The second conclusion is that, indeed,
19
AUCs fail less frequently than Cmax's but they
20
still fail with a high frequency.
So, the
21
variation of AUCs should not be dismissed.
22
[Slide]
23
Study condition--perhaps I would omit this
24
almost entirely because it is considering single
25
dosing versus steady state. In
the U.S. this is a
84
1
non-issue because U.S. goes by single
2
administration even though it has been demonstrated
3 and
we know that frequently in steady state we get
4
lower variation--not frequently but not always.
5
[Slide]
6
This is a study showing that and in the
7
U.S. I think this is largely at the moment
8
irrelevant.
9
[Slide]
10
Study designs, which one to choose?
A 2 X
11 2
traditional or replicate design? It need
not be
12 a
4-period replicate design; it could be 3.
13
[Slide]
14
Now, the advantage of replicate designs
15
includes that one gets clear estimates of
16
within-subject variations.
Particularly the
17
concern would be to get a clear estimate of
18
within-subject variation for the reference product.
19 I
would note that this design is favored by K.K.
20
Midha who has worked long years and is certainly
21 one
of the foremost experts on the bioequivalence
22 of
highly variable drugs and drug products.
So,
23 his
voice ought to be respected.
24
Secondly, on the other hand, my concern is
25
that one can have a pooled criterion which could
85
1
have better properties, pooled criterion related to
2 the
test and reference products together.
3
There are issues that these replicate
4
design studies can be evaluated by various
5
procedures, and a question is whether these
6 procedures
would give the same results and,
7
therefore, would agencies be able to check how
8
those results would be calculated and were
9
calculated.
10
Another question arises, namely, is a test
11
comparing the variations of test and reference
12
products useful; is it needed?
Or, perhaps is an
13
estimate of these variations simply sufficient or
14 is
that needed?
15
[Slide]
16
Turning to the 2 X 2 crossovers, they are
17
simple; simple to execute, simple to evaluate. An
18
advantage is that there are many studies on file
19 and
they could be evaluated retrospectively.
20
Another comment is that the ratio of
21
within-subject variabilities could be estimated.
22
There are procedures that would permit this even
23
from 2 X 2 crossover studies. For
example, the
24
procedure suggested here by Guilbaud and Gould is
25 to
have for each subject the sum of the test and
86
1
reference response, AUC or Cmax in this case, and
2
then the difference of the two; plot them against
3
each other, have a linear regression and evaluate
4 the
slope, and then apply the slope in that fashion
5
which gives the ratio of the estimated variances.
6 So,
it would be possible to evaluate this ratio
7
from 2 X 2 crossovers. However,
features of this
8
procedure have not been studied and they ought to
9 be
evaluated.
10
[Slide]
11
Now, various possible methods of
12
evaluation, the usual procedure is unscaled average
13
bioequivalence with a criterion of 0.8 to 1.25 for
14 the
ratio of geometric means, the GMR. It is
also
15
possible to apply unscaled average bioequivalence
16
with expanded bioequivalence limits.
One way of
17
doing it is to present these bioequivalence limits.
18 It
has been shown that some jurisdictions do this.
19 For
example, the ratio of GMR could be between 0.75
20 and
1.33 or 0.7 to 1.43. This is one
possibility
21
which is practiced in some areas, or to expand the
22
bioequivalence limits flexibly depending on the
23
estimated variation. I shall talk
more about these
24
procedures.
25
Another approach is the scaled average
87
1
bioequivalence and, again, I shall refer to this
2 and
shall talk about this, and I also should
3
mention scaled individual bioequivalence for
4
comparisons only.
5
[Slide]
6
To talk about unscaled average
7
bioequivalence--these scissors are supposed to be
8
less than or equal signs so instead of scissors, it
9 is
less than or equal--the unscaled average, as we
10
have seen--this is a bit more formalized but, as
11 you
see here, the ratio of geometric means should
12 be
between, say, 0.8 or 1.25 or 0.75 and 1.33.
13
This is the same statement as saying that the
14
logarithmic bioequivalence limits should be plus
15 and
minus and in between is the difference of the
16
logarithmic means, and that is a useful way to look
17 at
it. Now, the procedure is simple but as
the 0.8
18 and
1.25 limits were arbitrary so would be any
19
other criteria.
20
But another concern is that whatever way
21 it
would be decided, if this is the way to go, then
22
0.75 to 1.33 is a partial solution because it may
23
help drugs with, say, 30, 40 percent intra-subject,
24
intra-individual variation but not those which have
25
higher variations and 50, 60 percent would still be
88
1 the
cut off.
2
[Slide]
3
Another approach would be to expand the
4
limits in proportion to the estimated variation.
5
This has been suggested by Boddy and coworkers.
6 So,
here there is a proportionality factor, and the
7
other factor is the estimated standard deviation,
8
intra-subject variation. This
procedure has the
9
advantage that the usual testing procedure can be
10
applied with some proviso. The
statistical power
11 is
independent of the variation and the statistical
12
power is higher, much higher than the unscaled
13
average bioequivalence with the usual criterion so
14 we
need fewer subjects.
15
On the other hand, the criterion is that
16 bioequivalence
limits, as shown there, are really
17
random variables because they include the estimated
18
standard deviation, estimated intra-subject
19
variation. So, the limit itself
is a variable.
20
Therefore, the two one-sided test procedure is not
21
quite correct, however, it is becoming
22
approximately correct with large samples.
23
[Slide]
24
Scaled average bioequivalence is very
25
similar to the previous one except that the S from
89
1 the
bioequivalence limits, here, came over to the
2
measure that we apply. So, it is
formally very
3
similar and we have developed and have recommended
4
procedures for setting the bioequivalence limits.
5
Again, the advantages are that the
6
statistical power is independent of the variation
7 and
with the same sample size is much higher than
8 the
unscaled average bioequivalence. I am
going to
9
demonstrate this. There is a
sensible
10
interpretation. The first
interpretation is very
11
similar to that applied with individual
12
bioequivalence, namely, the expected change to
13
switching is being compared with the expected
14
difference between replicate administrations and
15 one
can make sense of that.
16
A second interpretation is that the
17
standardized effect size is being applied which is
18 a
clinical interpretation. There are procedures
to
19
evaluate confidence limits. If it
is a 2 X 2
20
crossover, then non-central t-test can be applied,
21 or
there is a procedure recommended by Hyslop and
22 her
coworkers which is somewhat more involved but
23
still reasonable I think.
24
[Slide]
25
This is a demonstration comparing the
90
1
procedures and effectiveness of various approaches.
2
They include the scaled individual bioequivalence,
3
scaled average bioequivalence and unscaled average
4
bioequivalence. You see the
probability of
5
acceptance. These are results of
simulations. It
6
amounts to the probability of acceptance at various
7 distances
between the two means. The first thing
8 you
can see is that for individual bioequivalence
9 the
range is very wide. Ranges are much
narrower
10
with scaled average bioequivalence.
So, this wide
11
range raised the concern of Dr. Benet.
The second
12
observation is that scaled average bioequivalence
13 is,
indeed, much more powerful than unscaled
14
average bioequivalence. So, we
again need fewer
15
people.
16
[Slide]
17
What is the limiting variation for highly
18
variable drugs? This is obviously
a subject of
19
regulatory decision, as are the others.
The
20
procedure could be that we apply unscaled average
21
bioequivalence if the variation is less than the
22
cut-off measure and use some kind of a different
23
procedure appropriate for highly variable drugs if
24 the
variation is higher.
25
Perhaps I should go down here.
This is
91
1 the
same kind of mixed model that was suggested for
2
individual bioequivalence but, just as Dr. Benet
3
suggested, it is not reasonable that a sponsor
4
should play both ways. The
sponsor should declare
5 the
intention of using one procedure or the other
6 in
the protocol.
7
I wouldn't necessarily dismiss these other
8
possibilities. For example, K.
Midha recommends 25
9
percent. The outcome of those
probabilities that
10 you
have seen on the previous slide depend on how
11 you
set these limiting variations.
Obviously, 30
12
percent is stricter than 25 percent.
In all cases
13 you
and the agency will ask what is the practically
14
reasonable criterion that one can live with, the
15
agency can live with and the industry can live
16
with, and the public can live with.
So, don't
17
necessarily set everything on the 30 percent; do
18
consider what the effect of, say, 25 percent would
19 be.
20 [Slide]
21
Now, this method of the secondary
22
criterion has arisen in connection with the
23
features of individual bioequivalence.
So, we talk
24
about two approaches, that of individual
25
bioequivalence and today we are talking about
92
1
highly variable drugs. There are
two very
2
different concerns.
3
First of all, we have already seen that
4 for
highly variable drugs the potential variation
5 is
smaller than with individual bioequivalence.
In
6 the
case of individual bioequivalence the
7
deviations arose because the regulatory criterion
8 was
changed. A much more liberal regulatory
9 criterion
was introduced whereas in the case of
10
highly variable drugs it is a natural change of the
11
variability between the two means.
You know this
12
very well. With the usual kind of
drug the
13
variation between the means just fluctuates
14
slightly. Most of the differences
are probably
15
between the two means and are within the range of
16 10
percent. But with highly variable drugs
those
17
means also fluctuate much more.
So, to impose a
18
constraint of 10-15 percent on this natural
19
variation means that the natural fluctuation is
20
altered so the sources of the concern are very
21
different. Whereas in the case of
individual
22
bioequivalence you have to deal with the criterion,
23 here
you have to deal with the natural variation.
24
[Slide]
25
So, I would like to raise some caution.
93
1 In
addition, the imposition of the secondary
2 criterion has serious consequences. I present this
3
from my life earlier when I dealt with individual
4
bioequivalence because we had the results then; I
5
don't have many results for average bioequivalence.
6
But, again, here you have the results for
7
individual bioequivalence. This
is the probability
8
curve for the constrained criterion alone and this
9 is
then the application of the combined criterion.
10
The combined criterion is expected and
11
does always run below the two separate criteria.
12 But
when the GMR criterion is highly constricting,
13 as
in this case, then the combined criterion is
14
really a GMR criterion essentially and has nothing
15 to
do, or very little to do with the bioequivalence
16
criterion. So, if you were to
consider the
17
secondary criterion, then this slide suggests to do
18 it
with great caution and after serious
19
consideration.
20
[Slide]
21
Here are the questions again which I have
22
raised for the committee's consideration and for
23 the
agency's consideration. They certainly
suggest
24
that many of these issues require further
25
consideration and further investigation.
94
1
Originally I wanted to end with this loose and
2
compliant mode, however, I looked at the questions
3
being raised and, since after this I may have to
4
shut up, I would like to call attention to question
5
2(b) in which the application of scaling is
6
combined with the application of this secondary
7
criterion. I would like to call
your attention to
8 the
fact that these are two separate questions.
9
Both of them ought to be studied further but, to my
10
mind, the restriction criterion is much more
11
controversial and requires thorough exploration for
12 its
need as well as for its application. So,
I
13
would recommend a separation of those questions.
14
Also, I have a question about reference
15
scaling. I would certainly like
to be an advocate
16 for
scaling, but whether the scaling ought to be
17
reference scaling I would like again to be a
18
subject for study. Thank you.
19
DR. KIBBE: Thank you. Questions?
20
Jurgen?
21
DR. VENITZ: I have a question
about your
22
first simulation slide where you compare the IBE to
23 the
ABE and scaled ABE. My question basically
is
24
that you are assuming for the purposes of
25
simulation that the COVs for both test and
95
1
reference are the same, 40 percent.
Is that
2
correct?
3 DR. ENDRENYI: Yes.
4
DR. VENITZ: What would happen if
you had
5
differences in COVs between test and reference? In
6
other words, let's assume that the test product has
7
much less intra-individual variability than the
8
reference, how would that affect your curves?
9
DR. ENDRENYI: It does affect the
curves,
10 but
mainly the curve of the individual
11
bioequivalence. It affects little
the average
12
bioequivalence curve.
13 DR. VENITZ: What about the scaled average
14 BE?
15
DR. ENDRENYI: The same. But that is an
16
artifact in a way because here we consider the
17
scaling by reference product so we didn't
18
have--these were 4-period studies.
Your question
19 is
relevant if you consider the 2-period studies.
20
DR. VENITZ: Right, right.
21
DR. ENDRENYI: Which we haven't
done, but
22
that is an interesting question.
It would be worth
23
investigating.
24
DR. VENITZ: So, the answer that
you are
25
using then is the reference variation.
96
1
DR. ENDRENYI: That is right.
2
DR. VENITZ: So, you are assuming
that you
3
know but you wouldn't necessary do a 2 X 2--
4
DR. ENDRENYI: No, the estimated
5
reference.
6
DR. VENITZ: So, you could get
that from a
7 2 X
2 design?
8
DR. ENDRENYI: Well, it is a
different
9
interpretation. Yes, we could but
it has to be
10
validated whether it works or not.
We haven't done
11
that.
12
DR. KIBBE: Anybody else? Ajaz, do I see
13 you
leaning forward? No? Go ahead.
14
DR. SINGPURWALLA: I just have a
technical
15
comment. Somewhere in your slides
you had a
16
restricted maximum likelihood.
Right?
17
DR. ENDRENYI: Yes, as a possible
18
procedure.
19
DR. SINGPURWALLA: As a possible
20
procedure?
21
DR. ENDRENYI: Yes.
22
DR. SINGPURWALLA: Well, this is a
23
technical comment, the maximum likelihood is
24
advocated because of its asymptotic properties in
25 the
sense that it converges to the center.
You
97
1
know, you get the central limit theorem.
When you
2
restrict your maximum there is no assurance that
3 you
converge, the central limit theorem.
4
Therefore, the value of that process cannot be
5
really evaluated. I don't know
what impact all
6
that has on the proposals you have made but I just
7
want to caution you.
8
DR. ENDRENYI: You are absolutely
right,
9 but
the point I think was that in the case of
10
replicate design probably the procedure of
11
evaluation would have to be defined very clearly
12 and
very strictly, otherwise one can go in all
13
different directions and that will be another task
14 if
the agency goes that way.
15
DR. KIBBE: Go ahead.
16
DR. BENET: Just a quick follow-up
on
17
Laszlo's comment, I think it would be worthwhile if
18 the
agency went back and looked at the content
19
uniformity criteria and published two sets of data.
20 I
think it would be worthwhile to go back and look
21 at
the bioequivalence data and look and see how
22
often it falls within certain criteria.
You have a
23 big
database and it would be nice to see what those
24
numbers were, and I think that would be useful for
25 the
committee on the secondary criteria.
98
1
DR. YU: Actually, you will see
that in
2 the
last talk. Sam is going to talk about
data.
3
DR. KIBBE: It is always good to
have data
4
when we are having a discussion.
No one else?
5
Marv?
6
DR. MEYER: This is probably a
7
statistically ignorant question but under the
8
scaled condition, however you want to scale it, is
9 it
possible to have a product with a scale
10
confidence limit that was, say, 60-90?
If so, then
11
let's say the ratio would be somewhere around 75
12
percent and that wouldn't be acceptable.
So,
13
without the point estimate constraint you have a
14
potential for allowing 60-90 approved and 120-140
15 to
be approved.
16
DR. ENDRENYI: No--
17
DR. MEYER: Two different studies?
18
DR. ENDRENYI: In two different
19
studies--within each study it should be one and I
20
wouldn't envision between study variation and I
21
don't--I doubt it very much.
22
DR. MEYER: Even if the test product only
23
released 70 percent of its dose and the innovator
24
released 100 percent of its dose the true ratio
25
would be 0.7 and you wouldn't know that; you would
99
1 be
looking for 1.0. It is not possible?
2
DR. ENDRENYI: No, I think if it
is
3
inter-study variation, then with the low variation
4
drugs each of them would be between 0.8 and 1.25
5 but
the two in comparison with each other could be
6
quite different. That is equally
possible but it
7 is
not likely. If it is the same reference
8
product, then it is not possible.
9
DR. KIBBE: I see no other
questions.
10
Thank you very much. We will take
our break now.
11 We
will be back at 10:52.
12
[Brief recess]
13
DR. KIBBE: We have open public
hearing at
14 one
o'clock but there are no presentations to be
15
made at that time so what we will be able to do is
16
modify our schedule to try to get everything done
17 and
get back on schedule. I know there is a
lot of
18
interest in what we are talking about so we might
19
allow our speakers a little extra time and some
20
questions and answers to go a little further. I
21 see
our next speaker is at the podium, ready to go,
22
Barbara Davit.
23 Bioequivalence of Highly
Variable
24 Drugs Case Studies
25 DR. DAVIT: I am pleased to be able to
100
1
respond to one of the questions that Les raised, in
2
that we do have a survey of some of the data that
3 has
been submitted to the Division of
4
Bioequivalence.
5
[Slide]
6
When Dale and I were talking about putting
7
this presentation together for the advisory
8
committee, one of the things we thought we would
9
consider is looking at what has been submitted to
10 the
Division of Bioequivalence and to answer the
11
question of whether highly variability is a
12
significant issue in these bioequivalence studies
13 in
ANDA submissions.
14
By looking at these data and focusing on
15
some case studies, we thought also we could maybe
16
answer the questions in a limited number of cases
17 of
what is contributing to the variability or what
18 are
some of the sources of this variability.
19
[Slide]
20
So, what we were trying to do is see if
21
there is a significant problem with highly variable
22
drugs, and I would like to mention, first of all,
23
that this obviously represents a biased sample
24
because we receive predominantly studies that have
25
passed the 90 percent confidence interval criteria.
101
1 So
we obviously don't see the big picture like
2
people from industry would be seeing.
We don't see
3
what percentage that is of the total number of
4
drugs in a company's pipeline for example.
5
But of the submissions we saw, which are
6
passing studies, what percentage were for highly
7
variable drugs? Did these studies
involve
8
enrolling a large number of subjects because that
9 has
been one of the issues that has been raised
10
today, the large number of subjects that might be
11
necessary to show bioequivalence for these generic
12
products of highly variable drugs?
Also, how
13
narrow and wide are these 90 percent confidence
14
intervals? That goes along with
how many subjects
15 are
necessary for a passing study.
16
[Slide]
17
We collected data from all the
18
bioequivalence studies that were submitted to the
19
Division of Bioequivalence in 2003.
We used the
20
root mean square error as an estimate of
21
intra-subject variability. I
realize this is just
22 a
rough estimate and it is not a pure estimate of
23 the
intra-subject variability but, unfortunately,
24
most of the studies that we had to look at were
25
two-way crossover studies so the best estimate that
102
1 we
could get of the intra-subject variability was
2 the
root mean square error.
3
We defined a highly variable drug as one
4
with a root mean square error which is greater than
5
0.3, representing 30 percent intra-subject
6
variability. The data that I am
going to present
7 is
only solid oral dosage forms, and I would like
8 to
point out that all the studies that I am going
9 to
be presenting passed our 90 percent confidence
10
interval criteria, but that is because for the most
11
part we don't receive submissions of studies where
12 the
product did not pass bioequivalence criteria.
13
[Slide]
14
First of all from 2003, this was a total
15 of
212 in vivo bioequivalence studies. Of
these
16
212, looking at only those studies in which the
17
root mean square error of AUC or Cmax was greater
18
than 0.3, in 15.5 percent of these studies, AUC or
19
Cmax, was greater than 0.3. In
other words, in
20
about 15 percent of our studies the drug would
21
qualify as having highly variable characteristics.
22
Most of this was due to Cmax and this has been
23
discussed today. So, in about 13
percent of the
24
total only Cmax was highly variable.
There were no
25
studies in which only AUC was highly variable. But
103
1
there were 5 studies in which both AUC and Cmax
2
were highly variable, and this was 2.5 percent of
3 the
total.
4
[Slide]
5
This goes along with the previous slide
6 and
it just shows the number of studies in which we
7 saw
a root mean square error of a particular value
8 for
Cmax. There is an error in this
particular
9
slide in your handout but this is the correct
10
slide. Really, obviously, for
most of the Cmax
11
values the root mean square error is below 0.3. I
12 have
a line here representing 0.3. I think I
said
13
earlier that 15 percent of all the studies, 15.5
14
percent of all the studies that came in had a root
15
mean square error for Cmax of greater than 0.3.
16
[Slide]
17
This is for AUC. Of course, the AUC is a
18 lot
less variable than Cmax. Really, for the
most
19
part the root mean square errors were hovering
20
around 0.1 to 0.15, so quite a bit less variability
21 in
AUC than Cmax.
22
[Slide]
23
One of the questions that we wanted to ask
24 was
what is contributing to this variability.
25
Since for a lot of products we look at
104
1
bioequivalence studies in fasted subjects as well
2 as
fed subjects, we wanted to see what impact was
3
having on variability. I
mentioned 33 studies.
4
This represented a total of 24 of the ANDAs that
5
were submitted and reviewed in 2003.
Of these,
6
both AUC or Cmax were highly variable in both the
7 fed
and fasted studies. In 8 of these the
8
pharmacokinetic parameters were highly variable in
9
only the fed study, and for 7 the PK parameters
10
were highly variable in only the fasted study. But
11
this is a little bit skewed too because we have
12
submissions, for whatever reason, which contain
13
only a fed study and submissions that contain only
14 a
fasted study--not a lot but it does happen.
15 [Slide]
16
This shows some of our data. I
think
17
these are all the Cmax values from the 212 studies
18 I
was talking about in which Cmax was variable in
19
only the fed study and not the fasting study. So,
20 this
would suggest, of course, that we are seeing
21
variability because of food effects.
I am not
22
giving the names of the drugs but I have
23
illustrated them by class.
24
There is a variety of reasons I think for
25 the
variability. Some of these are
prodrugs. We
105
1
have a number of angiotensin converting enzyme
2
inhibitors and most of these are prodrugs.
3
Generally the parent is present at low
4
concentrations so this could contribute to the
5
variability. A number of these
drugs also are
6
highly metabolized and this would contribute to the
7
variability. But, in this case,
obviously there
8 was
a food effect. The variability was
observed in
9 the
fed state, not in the fasting state. In
these
10
studies too the number of subjects ranged from
11
about 27 to 51 I guess, so all over the place in
12
terms of numbers of subjects.
13
[Slide]
14
It is pretty unusual to only see a highly
15
variable Cmax in the fasting study and not the fed
16
study, and this occurred in only two cases last
17
year. These were both angiotensin
converting
18
enzyme inhibitors, both prodrugs.
For one of them
19 the
bioequivalence was based on measuring the
20
parent. For the other one the
company could not
21
measure the parent despite I guess a number of
22
attempts. This is actually true
for pretty much
23
everyone who has worked with this particular drug.
24 So,
the bioequivalence here is only based on the
25
metabolite. But that is quite
rare. In the vast
106
1
majority of submissions that we have the
2
bioequivalence is based on the parent.
3
[Slide]
4
This table shows the Cmax data where Cmax
5 was
highly variably in both fed and fasted studies.
6 So,
for this drug product obviously there will be
7
highly variable regardless of whether it is the fed
8
study or the fasted BE study.
This was six drug
9
products, various drug classes, various reasons for
10
variability; some prodrugs, some highly metabolized
11
drugs; some drugs that undergo extensive first-pass
12
metabolism. The number of
subjects varied from I
13
guess 18 to 57.
14
[Slide]
15
Finally, this table is for two-way
16
crossover studies and shows the data for which both
17 AUC
and Cmax were highly variable, and this was for
18
four drug products. For the one
that I have shown
19 in
yellow, for this particular product both AUC and
20
Cmax met the highly variable criteria in both the
21 fed
and the fasting state. For the other
drugs
22
there was high variability in the fed but not
23
necessarily the fasted, or Cmax and not necessarily
24
AUC. So, this was four drugs that
fell in this
25
class. The number of subjects that
the companies
107
1
used varied from 26 to 62.
2
[Slide]
3
In trying to explore some of the sources
4 for
this variability, we wanted to compare the
5
intra-subject variability for the test versus the
6
reference product. We don't see
very many
7
replicate design studies anymore.
In this
8
particular class of drugs we only had two
9
submissions last year so these are the data from
10 the
two submissions.
11
These data are a good sign because what
12
they show is that the variability, based on the
13
root mean square error, was comparable for the test
14 and
the reference product for both of these drug
15
products. That is obviously what
we are looking
16 for
because we want to see people achieve a generic
17
product that is the same as the reference product.
18 So,
in this case I would say the variability was
19
comparable, test versus reference.
20
One study used 33 subjects. The
other,
21
this would obviously fall into a category where it
22
necessitated a lot of subjects because this was not
23
only 77 subjects, it was also a replicate design so
24 it
meant that each of these 77 subjects received
25 the
drug product four times, on four occasions.
108
1 So,
this was quite an extensive study.
2
[Slide]
3 Another question we wanted to ask
was are
4
there ever cases in which the pharmacokinetic
5
variability is a function of the drug product as
6
opposed to the drug substance. We
found two
7
instances last year, two different drug products
8 and
I will call them drug C and drug D. This
was
9 the
same RLD for both studies for drug C and the
10
same RLD for both studies with drug D.
Drug C was
11 an
extended release tablet. Drug D was an
12
immediate release tablet.
13
We will look at drug C first. In
one
14
study, conducted by one applicant, using I guess 33
15
subjects in the fasted and 35 subjects in the fed,
16
this product would not qualify as a highly variable
17 drug. Notice root mean square errors of 0.18,
18
0.11, 0.21, 0.24. However, for
the same reference
19
product, in other words it is the same product,
20
different formulation, another company, 0.31, 0.38,
21
0.25, 0.34.
22
This could be due to a number of reasons.
23 I
looked at the data and, obviously, the extended
24
release dosage forms are more complex than the
25
immediate release dosage forms and the two
109
1
formulations were quite different.
So, there could
2
have been, you know, differences in variability due
3 to
the formulation. Also, the
bioequivalence
4
studies were done at different sites.
I looked at
5 the
assays. They were both LCMS assays. I didn't
6 get
the specifics of the extraction methods but I
7
noticed that the two studies had different limits
8 of
quantitation and there were different doses in
9 the
two studies. I am not sure how much of a
10
factor this was. This was an
extended release
11
product for which I believe there were three
12
different strengths. One company
submitted a study
13 on
the highest strength and I think used two times
14 15
mg, which was 30. The other company did
studies
15 on
5 mg and used 4 times 5 mg, which was 20.
So,
16
different doses in the two studies.
So, there are
17 all
these factors that could be contributing to the
18
variability. At least, those are
the factors I
19
could think of.
20
Drug D--this was an interesting issue.
21
Once again, in the hands of one sponsor, one
22
applicant, we saw root mean square errors of 0.16,
23
0.25, 0.13 and 0.2; the other applicant, 0.38,
24
0.55, 0.22 and 0.24. This was an
immediate release
25
product and I noticed that the formulations of
110
1
these two were qualitatively identical;
2
quantitatively there were some differences.
3
These were done at two different sites and
4 in
this particular application the bioanaytical
5
methods were done at a CRO that we have had some
6
issues with in the past. They
seemed to be having
7
problems with some of their data.
So, it could
8
have been a contributing factor here.
9
I would like to stress that of all the
10
applications that we saw last year, these were the
11
only four in which we saw that there was a
12
difference which was possibly due to drug
13
formulation or possibly due to where the studies
14
were done that was contributing to the high
15
variability.
16
[Slide]
17
Then we thought we would look at how many
18
study subjects are usually enrolled in these
19
studies. Once again, I emphasize
that this is
20
really a biased sample because we only see the
21
studies that have passed. We
don't know how many
22
tries this represents. We don't
know how many
23
studies were done where the company just couldn't
24 get
the study to pass the confidence interval
25
criteria so these are just the passed studies.
111
1
I was expecting to see a much bigger
2
increase in the number of subjects as we went above
3 0.3
and we really didn't. This could
probably be
4 in
part because, as you know, the root mean square
5
error is not really a true estimate of variability;
6 it
is just a rough estimate. But in
general, I
7
guess of all the studies that came in last year,
8
that came in and were reviewed, there were only 14
9
that enrolled more than 50 subjects, and for those
10
that met our high variability criteria there were
11
only 5 that enrolled more than 50 subjects. I
12
think there are about 14 with root mean square
13
errors greater than 0.3 that enrolled more than 40
14
subjects. But in some cases we
are seeing high
15
numbers of subjects but this particular graph shows
16
that it is possible for companies to do a study
17
with under 40 subjects with a drug that is
18
considered highly variable and still pass
19
confidence interval criteria.
20
[Slide]
21
Then I wanted to see what would happen if
22 we
plotted the width of the confidence interval
23
versus the number of subjects, and this was done
24 for
Cmax. These are the 33 bioequivalent
studies
25 in
which the root mean square error of the Cmax was
112
1
greater than 0.3. Really not a
big surprise. As
2 the
number of subjects increased the width of the
3
confidence interval became narrower, suggesting, as
4 has
been mentioned this morning, that with a higher
5
number of subjects it is much easier to meet the
6
confidence interval criteria because the confidence
7
interval of the product becomes narrower.
8
[Slide]
9
These are the data for AUC. We
have data
10
from a combination of fed and fasted studies. I
11
would like to point out the two at the top. These
12 are
fed bioequivalence studies and they don't meet
13 our
present confidence interval criteria, but these
14
studies were submitted before the new food guidance
15 was
put into effect, which was in January.
If a
16
study was submitted before January of 2003, we were
17
evaluating the study based on our old criteria for
18 fed
bioequivalence studies, meaning that only the
19
point estimate had to fall within the limits of 0.8
20 to
1.25. That is why, if you look at the
21
confidence intervals, if these studies had been
22
done later these would not have met our criteria
23 but
they met our criteria at the time.
24
Once again, you can see a trend where, as
25 the
number of subjects increases, the confidence
113
1
interval narrows. I suspect that
these two
2
products, with more study subjects, probably would
3
have been able to squeeze into the 0.8 to 1.25
4
confidence interval.
5 [Slide]
6
In conclusion, I would just like to sum up
7
that these are observations from the data that we
8
looked at from 2003, and 15.5 percent of all the
9
bioequivalence studies that were submitted and
10 reviewed
last year were for drugs that met the
11
highly variable criteria. Cmax
was more variable
12
than AUC. In general, higher
pharmacokinetic
13
variability occurred in the fed bioequivalence
14
studies. The two replicate design
studies that we
15
were able to look at showed comparable
16
pharmacokinetic variability for the generic and the
17 RLD
product.
18
[Slide]
19
In two cases for two drug products the
20
variability was associated with the formulation or
21
other factors in conducting the bioequivalence
22
studies. In general, the width of
the 90 percent
23
confidence interval narrowed as the number of
24
subjects increased. Of the 212
passing
25
bioequivalence studies, only 14 enrolled more than
114
1 50
subjects. Of the 33 passing
bioequivalence
2
studies of highly variable drugs, only 5 enrolled
3
more than 50 subjects.
4
[Slide]
5
I would like to acknowledge the members of
6 our
working group at the FDA. This is a
group of
7
individuals who have been discussing the highly
8
variable drug issues and what types of
9
presentations to put together for the advisory
10
committee meeting today. I would
like to give a
11
special thanks to Devvrat Patel, one of our
12
reviewers in the Division of Bioequivalence, who
13
collected all the data that I showed you today. I
14 would
also like to thank all of our reviewers for
15
their hard work in putting the reviews together
16
from which Dev was able to collect these data.
17
Thank you for your attention.
18
DR. KIBBE: Questions, folks? Go ahead,
19
Jurgen.
20
DR. VENITZ: Just one
clarification. This
21 was
an interesting presentation, Barbara but just
22 one
clarification, the root mean square error that
23 you
calculated, is that the pooled intra-individual
24 variability
across test and reference?
25
DR. DAVIT: Yes, it is.
115
1
DR. VENITZ: Then if you go back
to your
2
slide number 14, this is where you look at the
3
effect the drug product may have and you compare
4 the
extended release and the immediate release.
In
5
case number one, I guess manufacturer number one,
6 it
looks like it is a low variability drug and for
7
manufacturer number two is a high variability drug.
8
DR. DAVIT: Right.
9
DR. VENITZ: Could that indicate
that the
10
test product for manufacturer two actually has a
11
higher variability and the reference drug still has
12 the
same, whatever variability, it has?
13
DR. DAVIT: Oh, absolutely. I mean, yes,
14
there is no way to tell.
15
DR. VENITZ: So, this might then
16
contradict one of the statements that you made
17
later on because you are saying that test and
18
reference in the replicate design studies--
19
DR. DAVIT: For those two
products.
20
DR. VENITZ: Right, for those two
products
21 it
could well be that the test product has higher
22 variability than the reference product.
23
DR. DAVIT: For this product, yes,
it is a
24
possibility.
25
DR. VENITZ: But all the replicate
design
116
1 studies that you looked at--
2
DR. DAVIT: Which was only two.
3
DR. VENITZ: Right, you found for
those
4 two
at least that test and reference had the same
5
intra-individual variability.
6
DR. DAVIT: Yes.
7
DR. VENITZ: How does that compare
to the
8
overall experience, going back beyond your survey?
9 Do
you have any idea? Because I know they
talked
10
about this in 2001 the last time we met.
11
DR. DAVIT: You know, that is a
really
12
good question and we didn't have the time for this
13
presentation. We were only able
to collect data
14
from last year. We do have a lot
of replicate
15
design data from 2000, 2001 and I guess some from
16
2002 and I think we would like to expand this study
17 and
go back a number of years because we would have
18
more replicate design studies to compare test and
19
reference variability. Yes, this
is all we have,
20
unfortunately.
21
DR. BENET: First of all, what you
22
presented is very interesting but it wasn't what I
23
asked for. So, let me make clear
what I think the
24
committee could use. There is an
issue about the
25
point estimate. In 1999 Commissioner
Heaney
117
1
published in JAMA an article where she looked at
2 all
the drugs approved in '97, showed the content
3
uniformity and the Cmax and AUC with the means and
4 the
standard deviations. What I am asking
you to
5 do
is to go back and give the committee information
6 on
the point estimates. Where are the point
7
estimates on all those studies?
How much
8
variability? Are you going to do
that?
9
DR. YU: That is actually going to
be
10
presented by the next speaker.
11
DR. BENET: You set me up. Barbara said
12 she
was answering my question! Many of you
saw the
13 MDS
abstract at the AAPS in November of 2002.
If
14 you
didn't, I have two slides here that I talk
15
about all the time. MDS looked at
800 fasting
16
studies in terms of approval or non-approval. Of
17
course, you can have a highly variable drug that 12
18 people
pass because sometimes statistics work.
19
I think the most interesting piece of data
20
from that is that they looked at the number of
21
subjects enrolled and how many studies failed.
22
When they looked at 49-60 subjects enrolled in a
23
study, 68 percent of the studies failed.
When they
24
looked at greater than 60 subjects 12 percent of
25 the
studies failed.
118
1
Now, why is that? It has nothing
to do
2
with statistics. It has nothing
to do with going
3
back and saying how are you going to run the study.
4 It
has to do with generic companies CEOs, and I
5
have seen it many times. The
scientists say to the
6
company "we have run the preliminary study. We ran
7
six. We need 96 people to make
sure that we meet
8 the
confidence intervals," and the president says,
9
"96 people? Do you know how
much that costs? I am
10
feeling lucky, run 24." And,
that is exactly what
11
happens. If the 24 they get
it. If the 24 doesn't
12
pass, they either give up or they run another
13
study. So, you can't conclude
anything from the
14
data that you are seeing here in terms of
15 variability and the ability to pass. I want to
16
warn you on that. I think it is
really important
17 to
realize that until you start to see all the
18
data, which you will now, you really can't make
19
comments about whether highly variable drugs can
20
pass or whether you could have a progesterone study
21
that passed based on 50 subjects.
You could
22
easily; you just have to be lucky and lots of
23
people are lucky.
24
DR. DAVIT: Oh, I agree. I thought the
25
exact same thing when I looked at all these studies
119
1
with the number of subjects and number 24 and 25
2
came up again and again and I wondered if it was
3 something
like that.
4
DR. KIBBE: I just have a question
about
5 the
data. You had 212 studies you analyzed
but
6
that wasn't for 212 different compounds--
7
DR. DAVIT: Right.
8
DR. KIBBE: --there were multiple
9
submissions for the same compound.
10
DR. DAVIT: Right.
11
DR. KIBBE: So, the question that
I come
12
back to is on that early slide where you showed
13
five studies had AUC and Cmax problems, which
14
represented 2.5 percent. How many
drugs was that?
15
DR. DAVIT: Oh, that was five
different
16
drugs.
17
DR. KIBBE: Five different drugs,
not just
18
five studies by different companies?
19
DR. DAVIT: Right, it was five
different
20
drugs.
21
DR. KIBBE: That isn't the same
though for
22 the
33 with AUC or Cmax?
23
DR. DAVIT: Correct.
24
DR. KIBBE: There would be cases
where you
25 had
studies where there were multiple studies
120
1
showing the same drug having variability in each
2 one
of the studies?
3
DR. DAVIT: Right. I actually had a slide
4 like
that at one point and I took it out. But
the
5
answer to your question is yes.
6
DR. DELUCA: I noticed that your
data is
7
just for the approved drugs.
8
DR. DAVIT: Yes.
9
DR. DELUCA: But I had a
question. Maybe
10
Les--with the data he just mentioned because he has
11
data there of approved and non-approved, you have
12 212
here, is there a feel in relation to how many
13
drugs were not approved that did not meet the
14 specs? Maybe the industry or the data that Les has
15
might be able to give an estimate of what that
16
might be.
17
DR. BENET: Well, this is
something that
18
Laszlo said. I mean, the MDS data
obviously--you
19
know, what they showed was that if you had CVs less
20
than 30 percent, only like one-quarter of the
21
studies failed. If you had CVs
greater than 30
22
percent, 62 percent failed. And,
Laszlo was giving
23 you
the data for greater than 35 percent and all of
24
them failed. But, again, these
could have easily
25
been under-powered. I think most
of these are
121
1
under-powered in terms of the studies that MDS ran
2 but
it is 800 studies. I mean, they got data
from
3 800
studies.
4
DR. SINGPURWALLA: I want to
respond to
5 Dr.
Benet again before he goes away.
6
[Laughter]
7
Now, you raised this dichotomy of the
8 surprise that the test passed and then it
failed,
9 or
something like that. I am not sure
exactly how
10 you
said it. But there is a procedure in
11
statistics called sequential analysis which I am
12
sure you are aware of. The government,
not the FDA
13 but
the Department of Defense uses this procedure
14 for
acceptance sampling of products, whatever
15
product they are interested in.
The whole idea
16
behind that is you test one item at a time and you
17
make a decision either to accept or to reject. If
18 you
cannot make a decision to accept or to reject,
19 you
take another sample. You keep taking a
sample
20
until you make a decision, let's say, to accept.
21 The
government then buys tons and tons of
22
transistors or whatever it is based on this nice,
23
little sequential test, codified an put out as
24
military standard 414, version C, which is how I
25
last remember it.
122
1
Now, if you use that particular procedure
2 and
let's say the procedure says accept and, for
3
fun, you don't accept and go on testing more, guess
4
what happens. The procedure leads
to rejection.
5 So,
an early acceptance could be a bad thing had no
6
tested more. It seems that the
same phenomenon is
7
happening here. The culprit there
again is this
8
concept of type 1 and type 2 errors that come into
9
play. These procedures have been
discussed and
10
shown to be incoherent. I suspect
similar things
11 are
happening here. Thank you.
12
DR. KIBBE: Paul?
13
DR. FACKLER: I have a couple of
comments.
14
Ordinarily I agree with Les but I think he might
15
have over-simplified the generic industry.
16
Admittedly, a study with 96 subjects costs a lot of
17
money and there is a statistical probability that
18
with a highly variable drug you will pass with 10
19 or
12 subjects. Decisions are made based on
a lot
20 of
factors. Part of it is the probability
of
21
passing. Part of it is the
economics of what a
22
product might bring back to a generic company. I
23
will leave it at that.
24
As far as the analysis that the FDA has
25
done, the generic industry has been asked to submit
123
1
failed studies and I believe it is almost part of
2 the
Federal Register now that those are required.
3 But
those are failed studies on products that are
4
submitted to FDA. There are a
number of products
5
that the generic industry works on that never come
6 to
FDA because the BE studies haven't been able to
7 be passed.
So, MDS have a larger data bank of
8
studies than FDA but, of course, it is confidential
9
information and MDS can't really share all of the
10
details about that with FDA.
11
A couple of other comments, the one slide
12
that showed the two products that had differing
13
root mean squares, you suggested it might be
14
formulation differences that caused the difference
15 in
the variabilities. I am not sure that
you can
16
draw that conclusion. You did
qualify it by saying
17
that there could be other reasons for those errors.
18 It
could be as simple as different populations of
19
patients or subjects in these cases.
We have seen
20
examples where doing a highly variable product by
21 one
CRO can give a dramatically different
22
variability than another CRO just because of the
23
variability of the subject population that the CROs
24 are
able to gather. A CRO in the inner city
is
25
going to have a dramatically different patient
124
1
population than a CRO in the country in the
2
northern part of a very isolated corner of the
3
United States.
4
Then, I wanted to make the same comment
5
about the slide that showed only five studies with
6
more than 50 subjects. The
studies with 50, 60,
7 70,
80 and 90 subjects often fail and FDA never
8
becomes aware of those. Those
projects are often
9
dropped after two or three failures because there
10
doesn't seem to be a way to meet the 0.8 to 1.25
11
confidence intervals.
12
I would suggest, if the resources are
13
there, the FDA go back and look at all those
14
replicate design studies that were submitted two
15 and
three years ago when we were looking at IBE as
16 a
possibility and scaled bioequivalence. I
think
17 you
will find that the variability between test
18
product and test product is really not different
19
than the variability you see between the reference
20
product in those replicate design studies.
21
DR. KIBBE: Anybody else? Ajaz?
22
DR. HUSSAIN: I think I just want
to put
23
some issues back, important issues back on the
24
table. I think one of the reasons
we wanted Gordon
25
Amidon to come and speak here I think was to focus
125
1 on
what the root causes of variability are.
2
Because often we have these discussions, and so
3
forth, and we get so bogged down in the numbers and
4 the
statistics that we forget what the real
5
questions are that we were really asking. So, I
6
just want to remind us.
7
DR. KIBBE: Anybody else? No?
Thank you.
8 Now
Dr. Sam Haidar.
9 FDA Perspectives
10
DR. HAIDAR: Good morning,
everyone.
11
[Slide]
12
For my talk I will present regulatory
13
perspectives on the issue of bioequivalence of
14
highly variable drugs.
15
[Slide]
16
We are interested in this issue because it
17 has
several potential benefits, including reduction
18 in
regulatory burden and easier market access for
19
drugs which are safe and effective but also highly
20
variable.
21
[Slide]
22
Initially I would like to present a quick
23
overview of the regulatory requirements if
24 different
agencies including the FDA. For example,
25
Health Canada, CPMP in Europe and the FDA
126
1
equivalent in Japan.
2
[Slide]
3
The FDA criteria for bioequivalence has
4
been more precisely defined earlier so I will just
5
repeat that we have 80-125 percent limits on the 90
6
percent confidence interval for both AUC and Cmax.
7
These criteria are applied to drugs of low and high
8
variability.
9
[Slide]
10
In contrast, Health Canada has the same
11
criterion on AUC, the 80-125, however, no
12
confidence interval criteria for Cmax.
They just
13
have a constraint on the point estimate test to
14
fall between 80 and 125. In June
of last year,
15
these criteria were judged flexible enough to
16
handle highly variable drugs by an expert advisory
17
committee meeting.
18
[Slide]
19
In Europe they have the same limits on the
20
confidence interval for AUC and Cmax, however, they
21 do
make an exception in certain cases with regard
22 to
Cmax where wider limits are acceptable and they
23
cite the 75-133 as an example.
24
[Slide]
25
In Japan also they have the 80-125 percent
127
1
limits for AUC and Cmax, however, in cases of
2
failure they do allow for add-on studies.
3
[Slide]
4
From this, we conclude that major
5
regulatory agencies do have some flexibility in
6
their regulations to handle special cases of the
7
highly variable drugs. To
evaluate the performance
8 of
the FDA criteria a survey was taken of ANDA
9
submissions between 1996 and 2001.
I will present
10 the
point estimate distribution for Cmax and AUC.
11
[Slide]
12
This is the point estimate distribution
13 for
AUC and we have the percent of total studies
14
submitted. We can see that there
is a clustering
15
around the ratio of 1.0 and with a closer look we
16 saw
that for 95 percent of the studies--the in vivo
17
bioequivalence studies, 95 percent were within
18
plus/minus 10 percent.
19
[Slide]
20
For Cmax, which is a more variable
21
parameter, it is expressed with a wider
22
distribution. However, we also
see a clustering
23
around the ratio of 1.0. In the
case of Cmax, 85
24
percent of the studies were within plus/minus 10
25
percent.
128
1
[Slide]
2
From this data set we created a subset
3
that included highly variable drugs and highly
4
variable drug products. We see a
somewhat similar
5
distribution, also clustering around a ratio of 1.0
6 for
the AUC.
7
[Slide]
8
The same is true for Cmax but also to a
9 lesser extent as expressed by the greater
10
distribution.
11
[Slide]
12
From this we conclude that although the
13 FDA
criteria allow for a mean difference of
14
plus/minus 20 percent, the vast majority of the
15
submissions were within plus/minus 10 percent.
16
This was also observed for highly variable drugs.
17
[Slide]
18
In dealing with the special case of highly
19
variable drugs there are several options, including
20 a
scaling approach based on intra-subject
21
variability in Cmax and AUC, or direct expansion of
22 the
regulatory limits.
23
[Slide]
24
For scaling approaches, they would result
25 in
a reduction in sample size. The limits
are not
129
1
fixed but they are defined as a function of the
2
variability. There may also be a
need for a point
3
estimate constraint.
4
[Slide]
5
In contrast, direct expansion of the
6
limits, which may be applied only to Cmax or Cmax
7 and
AUC, the limits are fixed, for example 70 to
8
143, for drugs which are considered highly variable
9 or
are classified as highly variable. There
may
10
also be a need for a point estimate constraint in
11
this approach as well.
12
A major concern with this method for drugs
13
which are borderline around the 30 percent cut-off
14 is
how do we classify those drugs, and who does it?
15
Because, obviously, there are major commercial
16
advantages with a drug being classified as highly
17
variable under those circumstances.
18
[Slide]
19
A study conducted by Walter Hauck, which
20 was
supported by the FDA when the food effect
21
guidance was under development--they wanted to look
22 at
the impact of expanded limits around Cmax on
23
study design since fed studies in general tend to
24 be
more highly variable. The interval test
25
evaluated was 70-143 around Cmax.
The outcome was
130
1 60
percent reduction in sample size on average.
A
2
concern was expressed in that study that Cmax
3
ratios of up to 128 percent still passed using this
4
limit.
5
[Slide]
6
Finally, if a decision is made to modify
7 the
regulations to accommodate highly variable
8
drugs, we feel like either approach would result in
9 a
significant reduction in sample size although an
10
additional regulatory criterion might be needed
11
constraining the point estimate.
However, based on
12 our
previous experience, it is very likely that a
13
clustering around a ratio of 1.0 would still be
14
observed although, in theory, it could fluctuate to
15 a
greater extent.
16
[Slide]
17
Now Dr. Dale Conner would chair the
18
question and answer session.
19
Bioequivalence of Highly Variable Drugs Q&A
20
DR. CONNER: Good morning. I was asked to
21
simply not have a presentation but come up and
22
field questions. You know,
anything on this topic
23 is
fair game I think, although you can try and get
24
some other ones in if you like.
However, I decided
25 to
start it off to get the ball rolling by making a
131
1 few
remarks. First off, I can truly say when
I sit
2
through a lot of advisory committee topics and
3
discussions I am not extremely stimulated by them.
4 I
hate to admit that but sometimes some of the
5
topics to me, personally, are not very exciting.
6
This one however I found extremely exciting from
7
beginning to end, and perhaps that is just because
8 it
is bioequivalence; it is what I do all the time
9 and
it is a problem that has been discussed for a
10
long time and, due to the experts and this
11
committee, we are finally starting to make some
12
progress towards doing something about it.
13
So, I would like to thank both the
14
committee and all the excellent speakers who really
15
gave us quite a lot to think about and discuss.
16
Because there were so many issues, I sat there with
17 my
little list of points that I was going to make
18
which, hopefully, wouldn't have taken very long but
19 it
started to expand at a very alarming rate with
20
each of the speakers and the excellent points they
21
were making. I consider it kind
of a scaling
22
effect that my points were expanding with the
23
variability and quality of the speakers.
So, I
24
will try and keep it to a minimum and perhaps be a
25
little Procrustean in cutting off both ends.
132
1
These comments, these points I am making
2 are
my own take on it so one shouldn't necessarily
3
interpret this as FDA policy or even FDA thinking,
4 but
I tend to deal with these types of questions on
5 a
day-to-day basis in a practical sense and it
6
really seems to me, and what should have come out
7 of
this if you get down to the real issue, what we
8 are
looking at here for the most part is an
9
economic issue. In other words,
from the drug
10
company's point of view it is really economics.
11
These studies cost too much.
12
When I want to develop this drug--if I am
13 a
generic company and I want to develop this drug
14
and, as has been stated before, my statistician
15
comes back and say you have to do 120 subjects, I
16 am
sure that the bean counters at the firm are very
17
alarmed and saying, "my God, I'm used to paying for
18 a
24-subject trial and you've just told me it's
19
five times as expensive."
For a small company that
20
could mean three or four other products that I
21
don't have the funds to develop.
So, they have to
22
make a choice. Is this product
worth all that
23
money to spend or should I do four or five other
24
ones and forget about this?
25
It doesn't necessarily mean those products
133
1
aren't going to be developed by someone and be on
2 the
market but it will decrease the players.
Only
3
those with deep pockets will be able to develop it.
4 So,
of course, in the marketplace you will have
5
very much lower competition, which is not good for
6 the
consumer.
7
So, there is a variety of economic
8
considerations that this brings into play and if we
9
consistent somehow, using scientifically valid,
10
good regulatory methods, alleviate some of that,
11
that would be good. The FDA has a
motivation as
12
well. As has been mentioned, we
have a mandate to
13
eliminate or decrease unnecessary or excessive
14
human testing so we have a motivation as well. Our
15
motivation isn't strictly economic but we don't
16
want to expose normal subjects or patients in these
17
trials anymore than we have to because although
18
most of these trials are very low risk, they are
19 not
no risk. So, it is up to us to develop
20
scientifically valid ways to determine
21
bioequivalence with confidence, yet efficiently
22
with the least number of subjects we can get away
23
with. That is our motivation.
24
So, you could simply say from the firm's
25
point of view because of this criterion, because of
134
1
this inflexible criterion I am having to do an
2
unreasonable or excessive number of subjects. Now,
3
what is that? I mean, how do you
define
4
"unreasonable" or "excessive?" I am sure that some
5
people say that anything above 24 is unreasonable
6 and
if I asked everybody in this room what is an
7
unreasonable number of subjects, what is the
8
maximum number of subject you think you should have
9 to
do in any bioequivalence study for any product,
10 I
would probably get as many answers as there are
11
people in this room.
12
So, one of the ways you could start is
13
simply empirically saying, okay, I am going to set
14 a
number of samples that I don't want to go above,
15 no
matter what. A very simple way would be
to work
16
your way backward from that number.
Say, 60 was
17 the
highest you ever wanted to do, work your way
18
back saying, well, this is the variability I am
19
looking at. This is the allowable
true mean
20
difference. This is the power I
want. Work your
21 way
back and through simulation you get a set of
22
criteria that would achieve that goal, static
23
criteria. You could do something
like that.
24
Static criteria, I think as the last
25
couple of talks have outlined, in this case has
135
1
some problems because you have boundary conditions,
2
things where a drug product in one CRO's hands is
3
highly variable. You know, you go
to another CRO
4 and
it is not highly variable. So, who gets
the
5
benefit of your highly variable technique? When
6 you
look at creating a method to deal with this you
7
don't want to reward highly variable.
You don't
8
want sponsors, whether they intend to or not, to
9
force themselves into the highly variable state
10
just to get the benefit of whatever techniques you
11 are
dealing with. So, you want to adequately
deal
12
with the problem without, whether unintentionally,
13
encouraging bad or highly variable formulations.
14
What was mentioned by Les in the
15
individual bioequivalence, that was an attempt to
16
promise the fact that system would encourage firms
17 to
make lower variability products that still
18
matched and fit within the accepted criteria,
19
therapeutic criteria that were established in the
20
NDA. Even if we are not able to
do that, we
21
certainly don't want to do the opposite.
We
22
certainly don't want to unintentionally encourage
23
people to make their products or do their studies
24 in
a more variable manner just to get a benefit and
25 an
easier pass. So, just keep that in mind
when
136
1 you
are considering anything. You don't want
to be
2
counterproductive. You know, help
people in one
3 way
and then be counterproductive in another way
4
and, therefore, decrease the quality of the generic
5 and
maybe even the innovator products that we are
6
putting out. So, that is
something that always has
7 to
be kept in mind.
8
The topic that also worries me--I again
9
said in these scientific discussions--I will make
10
another admission, a portion of my mind is always
11 on
the scientific discussions and a portion of my
12
mind is, you know, on the practical sense of how
13 the
heck am I going to implement this.
Because the
14
scientists in industry, the firms and the review
15
staff at the FDA are the ones that are going to
16
have to live with this, are going to have to find a
17 way
to implement these techniques, to make them
18
work, and a lot of times the little details that we
19
don't talk about in rooms like this are the things
20
that kill you, that make this an almost unworkable
21
system.
22
For example, we can all agree and discuss
23
that we like 30 percent as the cut-off but, again,
24 now
do you determine that 30 percent? Is it
25
determined before you do any studies, from pilot
137
1
studies? Is it determined from
the literature? Is
2 it
determined from the NDA? What happens
when the
3
entire literature and available information says
4
that something is 28 percent and somebody does a
5
study and it is 34 on that one product?
Every
6
other product that is done, similar product, is
7
still 29, what do you do in that case?
Or the
8
opposite? You know, every other
study has been 32,
9 33,
34. It is considered a highly variable
drug.
10
Somehow you do one study, have one formulation and
11 it
is 28. What do you do then? So, that really is
12 a
very practical thing. These things that
are on
13 the
borderline could have the benefit or the remedy
14
applied to them or not applied to them depending
15 how
their data comes out. It is probably an
16
advantage for a proper scaling method rather than
17
simply increasing the study confidence intervals.
18
With that said, I will field the questions
19 if
there are any.
20
DR. KIBBE: Shall we start? Les wants to
21 ask
a question.
22
DR. CONNER: Les gets very antsy
unless he
23
talks about every ten minutes.
24
DR. KIBBE: If you do all the
paperwork,
25
Les, you can sit at the table.
138
1
DR. BENET: That is exactly why I
am not
2
sitting at the table.
3
[Laughter]
4
Dale, going back to Sam's data and just
5
following up exactly what you said, you have some
6
products that passed where the point estimate on
7
Cmax was 1.2 and they were supposed to be in the
8
highly variable group. Have you
gone back and
9
looked at that data? Was it a
huge number of
10
subjects or was there no variability on that study?
11
DR. CONNER: Usually with that
type there
12 are
only like one or two instances. I mean,
we
13
have done all sorts of periods and done that data,
14 and
I actually like that way of presenting it
15 rather
than the Heaney article--
16
DR. BENET: I like that way too.
17
DR. CONNER: --and subsequent
article
18
which just gave point estimates.
I always expect,
19 you
know, that I am going to see that are out at
20 1.8
or 1.9 or, you know, kind of close to the edge
21 but
not quite there, and I always react with horror
22
when I see that particular data point.
When we
23
really go in--I am not really sure; I would
24
actually have to direct it to Sam to put that
25
together because it doesn't make sense to me that
139
1
something could have a point estimate that far out
2 and
be highly variable unless they used a lot of
3
subjects--I mean a lot. So, I
will direct it to
4 him
but, on its face, it doesn't seem to make sense
5
because I have seen that type of data presented in
6
other ways and when I looked into it, it was a low
7
variability product. It was
something that just
8
squeaked by, had low variability and they used
9
sufficient subjects so even the alarmingly close
10
point estimate was still okay by our criteria.
11
DR. KIBBE: We are recording the
activity
12 so
you have to talk into a microphone.
13
DR. HAIDAR: I will have to go
back and
14
look at that study but based on what I have seen
15
there were maybe one or two studies that were above
16
1.20, and the reasons could be large number of
17
subjects or just purely by chance.
18
DR. BENET: I agree they passed
but I
19
think it would be instructive to go back and look
20 at
those boundary conditions and see what are the
21
characteristics of those studies.
I am sort of
22
thinking back to the generic drug scandal, you
23
know, where we saw some unbelievably low standard
24
deviations that nobody else ever saw at any other
25
time and I just think we ought to look at that data
140
1
carefully.
2
DR. CONNER: Sometimes you can get
low
3
standard deviations when you study the same drug
4
against the same drug.
5
DR. FACKLER: Can I address that
point?
6
DR. KIBBE: Please, go ahead.
7
DR. FACKLER: Confidence intervals
weren't
8
required for fed studies prior to 2002.
9
DR. DAVIT: I was just going to
say the
10
same thing.
11
DR. FACKLER: A lot of the fed
studies
12
from '96 to 2002 only needed a point estimate to
13
pass so 1.20 was perfectly within FDA's
14
acceptability criteria.
15
DR. BENET: I am aware of that too
but you
16 didn't separate them out, Sam? Those were both fed
17 and
fasted conditions?
18
DR. DAVIT: It is everything.
19
DR. BENET: I think we need to
separate
20
them out.
21
DR. HAIDAR: They were not separated.
22
This was our initial look.
23
DR. DAVIT: I would like to say
too that
24
probably we didn't start seeing consistently fed
25
studies that passed confidence interval criteria
141
1
until about six months ago. So,
before that all
2 the
fed studies were point estimate criteria.
3
DR. CONNER: So, we are doing an
unusual
4
analysis. We are taking that data
and calculating
5 confidence
intervals but it was never designed or
6
powered to do that. So, in a way
we are being
7
unfair to the data although, I mean, it is still
8
useful to look at it but, you know, to expect it to
9
pass confidence intervals when that was never the
10
intent and the statisticians that designed them
11
never powered it that way.
12
DR. BENET: Right, I can
understand that.
13 If
it was really true, then my recommendation would
14 be
impossible so that is why I want to see data
15
that looks at that. I think you
do too, or the
16
committee should too.
17
DR. CONNER: It is also important
to
18
remember that point estimates--you know, people
19
like to look at them because they are easy and they
20
seem to be the mean but you have to really look at
21
them very carefully because the say the statistics
22
work that isn't the true mean of the product. That
23 is
simply an estimate of the center of the data of
24
your small sample of the universe.
So, although it
25 is
interesting to look at them and they can be a
142
1
good indicator, you have to be very careful when
2 you
look at point estimates because it is not the
3
true mean.
4
DR. KIBBE: Ajaz?
5
DR. HUSSAIN: I think Dale
mentioned
6
something which I think is important and I want to
7
sort of repeat that because I think the whole
8
aspect of bioequivalence is to confirm that two
9
pharmaceutically equal products would behave as we
10
would expect them to behave. And,
I think we keep
11
missing that discussion and I think this discussion
12
also will not get to that but I want to keep
13
pounding on that. If there are
differences in the
14
variability of the product in terms of rate and
15
extent of absorption, that is the concern. That is
16 a
regulatory decision that has to be evaluated,
17
whether a high level of variability compared to a
18
lower level of variability in the innovator product
19 is
acceptable or not.
20
But the key aspect here is differences in
21 the
two products of the same drug. The drug
is the
22
same here. The formulation is
different. That is
23 the
focus of our entire discussion and, again, we
24 get
into the discussion on numbers and so forth but
25 we
never ask the question--since generally drug
143
1
approval is evaluation of the chemistry
2
manufacturing controls and then there is the
3
bioequivalence study which is one study.
If we
4
remember the clay feet of the bioequivalence
5 argument that Prof. Levy has always argued,
the
6
connection never gets discussed and somehow we have
7 to
rethink that process.
8
DR. KIBBE: Nozer?
9
DR. SINGPURWALLA: First I would
like to
10
comment on vocabulary. I prefer
that you use the
11
word within-subject versus between subject instead
12 of
this inter- and intra-, whatever it is.
13
DR. CONNER: I agree. I always get mixed
14 up
by that too.
15
DR. SINGPURWALLA: I think that is
a minor
16
comment. But the significant
comment is in the
17
handout questions on your last slide.
18
DR. CONNER: Those aren't really
my
19
questions. Those are the
questions for the
20
committee.
21
DR. SINGPURWALLA: Right, but are
we ready
22 to
talk about these?
23
DR. KIBBE: We are ready if you
are.
24
DR. SINGPURWALLA: Right, I
am. Now, this
25
whole morning's presentation, which I agree with
144
1 you
was not boring but very interesting, makes one
2
point clear, that this problem of bioequivalence
3 and
highly variable products calls for an
4
application of risk-based decision-making. The FDA
5
should serve as a benevolent decision-maker and
6
formulate the problem as one of decision-making
7
under uncertainty, keeping in mind the interests of
8 the
population, of the subjects; keeping in mind
9 the
interests of the drug companies or the
10
pharmaceutical companies and balancing and trading
11 off
those risks.
12
You can retain the technology of scaling.
13 You
can retain the investigation of causes of
14 variability.
There is nothing in the framework
15
that denies those things. But
what is really
16
needed is a change of mind set and a shift in the
17
paradigm. You have to get away
from the notion of
18
confidence intervals which have, I am told, just
19
been introduced two or three years ago, and move on
20
into a paradigm of decision-making under certainty,
21
bringing in utilities, bringing in those kinds of
22
considerations into this problem, otherwise you are
23 just spinning your head against the
wheel. That is
24 my
comment.
25
DR. KIBBE: Anybody? Marvin?
145
1
DR. MEYER: If we go with the
static
2
change, 70-143 for example, that smacks a bit of
3
being arbitrary which is a problem to defend and,
4
without a point estimate, allows, according to
5
Walter Hauck, 128 percent to pass.
That can be
6
taken care of by a point estimate such as Les
7
suggested. So, the arbitrariness
of that bothers
8 me.
9
But if we go to what I think is more
10
scientific-based, based on the variability of the
11
reference, albeit necessary to do a replicate at
12
least on the reference, then you have a situation
13
where you have confidence limits varying by study,
14 by
sponsor, by whatever else and then the
15
marketplace becomes chaotic because I am sure you
16
will have people arguing, well, our confidence
17
limits are narrower than their confidence limits
18 and
we have, therefore, a better product. Of
19
course, then you will send out a letter and say you
20
can't say that. So, I don't
know--I guess I would
21
favor the scale because it has some elements of
22
being tied to real data, and then somehow figure
23
out--I think Les said don't worry about what people
24
think in some sense. So, if the
confidence limit
25
ranged within two sponsors, maybe it isn't going to
146
1 be
a big issue once people understand what you did.
2
Those are just some comments really.
3
DR. CONNER: Over the years I have
been in
4
many meetings, internal and external, where we have
5
discussed widening or tightening the confidence
6
interval depending on the topic and the drug under
7
discussion. The tendency that
always disturbs me,
8 and
still disturbs me to this day, is people say,
9 oh
well, it is more variable so we should widen the
10
confidence intervals; let's do 70, let's do
11
plus/minus 30. You query where
did you get this
12
number. Well, it is wider. Well, how do you
13
support that? What makes you
think that is wide
14
enough to deal with the problem?
Maybe you have
15
gone too far. Or tightening the
confidence
16
interval limits, static limits are the same. I
17
mean, what makes you think that is tight enough to
18
deal with the perceived problem?
People tend to
19
just jump to the next--you know, they say it is
20
going to be wider and they go to the next five or
21
ten. But we rarely ever have
anyone come in and
22
support that with data. Maybe it
is just because
23 the
data is hard to come by but it disturbs me to
24
this day that most of these discussions are not
25
supported by any kind of scientific support that
147
1
this change is truly going to be able to
perceive
2 the
problem.
3
You know there are decisions and there are
4
problems with scaling methods, especially if you do
5
mixed scaling where you have a transition point
6
where it goes from a constant or static limit to a
7
scaled limit, which we saw proposed in individual
8
bioequivalence. There are some
boundary problems
9
around that transition point.
Again, you know,
10
which side do I fit? Where do I
get a better deal?
11
That type of situation. But I
don't really think
12 it
is as big a problem. You know several
different
13
sponsors might have slightly different limits
14
because those limits are determined by their own
15
data, their study, their data.
If, say, one CRO is
16 a
little more sloppy--I don't mean necessarily
17
negatively, and their variability of doing their
18
study is a little bit higher, that scaling would
19
account for that because you would have that across
20 the
board for both reference and test. I
mean,
21
scaling does have some properties that if it is
22
properly done it is probably a little more elegant
23 way
to deal with this problem. Still, you
have to
24 do
it properly and you have to think it through
25
very carefully. You can't just
jump into a method
148
1
without careful study.
2
DR. KIBBE: Jurgen?
3 DR. VENITZ: I am trying to get us to
4
start working on question number one, and it has to
5 do
with the comment that I made earlier.
Gordon
6
talked about mechanisms. We heard
Ajaz talking
7
about the need to understand where the variability
8
comes from, and that really is something that I
9
personally am missing. And, I
won't even get into
10 my
pet peeve about what is the clinical relevance
11 of
all of this.
12
But if I can identify, and I think I am in
13
agreement with Nozer that we have to use risk-based
14
assessment. Well, risk to me
means I have to
15
understand where are the key variability sources
16
that I am impacting on. What if
the variability is
17
primarily driven by systemic metabolism, then the
18
area under the curve and Cmax do not reflect
19
primarily product performance.
They reflect
20
something else which presumably is not affected by
21
changing products. So, is there
any way that you
22 can
incorporate that in some kind of algorithm,
23
some kind of decision tree where you decide what
24
rules you are going to use depending on what you
25
know about the drug? Maybe I am
not as strongly
149
1
statistically Bayesian as you are, but I do believe
2
that the current system disregards anything that we
3
know about the product. It just
says compare
4
product A to product B and roll the dice. It
5
ignores everything that we know about the
6
pharmaceutic characteristics of the drug substance
7 and
what we might know about a specific product in
8
question, whether it is extended or immediate
9
release classification.
10
So, I would like for the FDA to think
11
about how you could come up with an algorithm, a
12
decision tree where you would incorporate that in
13 the
early stages and then, by the time that you get
14 to
the end of your tree, there are different rules
15 but
those rules are then based on what you know
16
about the drug, not about something
17
arbitrary--well, in order for me to avoid a large
18
number; in order for me to pass some arbitrary
19
criteria I have to do this. To
me, it is the tail
20
wagging the dog as opposed to trying to use the
21
understanding that we have and a lot of those
22
products that you are looking at have been out for
23 a
long time so we know a lot about them but we
24
ignore that when it comes to the bioequivalence
25
assessment.
150
1
As far as scaling is concerned, the way I
2
understand it right now I am still wary about the
3
scaling and I really haven't formed an opinion yet.
4 In
order to do the average bioequivalence scaling,
5
right now what you need and probably the most
6
important problem I guess is within-subject
7
variability in the reference product.
How would
8 you
get that? You couldn't get it from a 2 X
2
9
study design. So, whose
responsibility then is it
10 to
provide that information? Because it
presumably
11
requires either a replicate design study or a
12
specific study just to identify the within-subject
13
variability in the innovator product.
Whose
14
responsibility is that? Is the
FDA going to pay
15 for
all those studies?
16
DR. CONNER: Usually it is the
sponsor's.
17
DR. VENITZ: Okay, so the generic
company
18 has
to do at least two studies or a replicate
19
design study.
20
DR. CONNER: With that approach,
if that
21 is
the type of scaling you designed requiring
22
replicate designs as we tried to do in the past,
23
probably a replicate design would be in order.
24
But, you know, there are a variety of things in the
25
literature and other proposals where that may or
151
1 may
not be necessary. But if you did pick
that
2
type of approach, yes, the sponsors would end up
3
probably doing some type of replicate design.
4
DR. KIBBE: Lawrence?
5
DR. YU: I want to make a
comment. I
6
guess a lot of speakers, especially FDA speakers
7
from the Office of Generic Drugs, paid a lot of
8
attention this morning to the generic application.
9
Yes, it is absolutely necessary that a part of the
10
requirement for generic approval for the market.
11 But
I want to remind you that we are developing an
12 FDA
policy to equally apply for innovator
13
manufacturers. What I
specifically mean is that I
14
think we have data to show that innovators, during
15 the
drug development process, during the approval
16
process or postmarketing, will make significant
17
changes, for example in excipients, formulation and
18
manufacturing facilities, and so on and so forth.
19 They will be required to conduct a
bioequivalence
20
study to make sure they are equal.
Therefore, for
21
highly variable drugs it is also equally applied
22 for
the innovator, not just simply the generic
23
companies. I want to make sure
that is
24
understandable.
25
Secondly, in terms of if we go forward, we
152
1 are
seeking your advice on which approach we should
2
take so that we can spend time on the right track
3 and
then come back to you with recommendations on
4
what approach we should take. If
the committee
5
advises us to move forward with the reference
6
scaling how do we determine within-subject
7
variability? That is an excellent
question.
8
Certainly it would be very difficult to get a
9
two-way crossover study. We would
have to go to
10 the
three-way crossover study at least a
11
replicative design from the reference list product
12 to get the number. Thank you.
13
DR. KIBBE: Marv?
14
DR. MEYER: To kind of follow-up
on
15
Jurgen's comment about what we know about the drug,
16 I
think there are a couple of simple yardsticks.
17 If
you can give a patient an intravenous and then
18
transfer them to IR, or if you can give them IR and
19
transfer them to CR, or back and forth, there is
20
probably no issue with Cmax there.
If the product
21 is
only available in one strength, 200 mg, and I
22
take it and small people take it and old people
23
take it once a day or twice a day, there is
24
probably no real issue with AUC or Cmax so you
25
could have somewhat less stringent requirements for
153
1
those kinds of drugs.
2
One other comment, add-in designs--it has
3
shown up here and there but we haven't really
4
addressed it and, to me, that seemed to be one
5
approach, provided that there are some constraints
6 on
that and you don't just keep on adding three
7
subjects until you get it right.
Add-ons have some
8
capability of eliminating excessive use of
9
subjects.
10
DR. CONNER: By add-on, I think
you mean
11
sequential.
12
DR. MEYER: Yes.
13
DR. CONNER: In other words, you
do the
14
first group, you look at the results, you make a
15
decision whether to go on or not.
16
DR. MEYER: Right, the point
estimate
17
looks good--
18
DR. CONNER: What we refer to as
add-on
19 is,
you know, you plan to do 24 subjects and you
20 get
a whole lot of dropouts. You haven't
looked at
21 the
data; you have no decision based on the results
22 but
you realize you are going to come up short so
23 you
get some alternates, recruit some more and put
24
them in. That is what we consider
an add-on. So,
25
there is no real decision based on results.
154
1
Whereas a sequential design is a plan ahead of time
2 to
do a certain number and generally, as has been
3
mentioned before, the true sequential design where
4 you
do one sample at a time is really not very
5
practical in these types of studies.
It would take
6 you
years maybe to do the right trial.
7
So, what we are talking about is a partial
8
sequential where you do groups.
If you were going
9 to
plan an overall 36, you were going to do 12 at a
10
time or 18 at a time, look at the results, make a
11
decision--you know, a correct statistical penalty
12 for
that look and that decision and then go on.
We
13
don't currently accept that but we are working on
14
it. In several venues, PQRI and
otherwise, we have
15
some working groups looking at that very carefully
16 and
the proper statistics to do on that. So,
we
17
hope to have some results on that pretty soon.
18
DR. KIBBE: Ajaz?
19
DR. HUSSAIN: No, I think I just
wanted to
20 say
a couple of things after Jurgen's comment.
21
From what he discussed, I think there are a
22
probably a few questions which are not on the
23
screen. So, are we asking the
right question also
24 is
the topic and I totally agree with him in a
25
sense because we continue to use the black box
155
1
approach. We don't know anything
about it so we
2
have to pass through this goalpost and the goalpost
3
often tends to be arbitrary to start with.
4
Then also, I think we essentially move
5
towards a check box exercise because that is easy
6 to
implement, and so forth. Clearly, I
think you
7
have to balance the ease of the process of doing
8
something and the scientific rigor and so forth.
9 So,
I think clearly as we move forward we will be
10
looking at what are the right questions also and
11
what are the right opportunities.
12
Two things that I think will open this up
13
further and new opportunities will come is the
14
prior knowledge. For example,
currently if you
15
look at an ANDA submission or even an NDA
16
submission you don't have much information to make
17
decisions with respect to formulation, process and
18 so
forth, what are the critical variables.
In
19
ICH-Q8 we have essentially moved forward with
20
pharmaceutical development as a basis for making
21
more scientific, mechanistic based decisions. So,
22 I
think we are trying to bring that know-how into
23 the
agency to do that.
24
Also, I look at submissions of all
25
studies, all bio studies done as an opportunity to
156
1 use
all that knowledge to make more rational
2
decisions and set more appropriate specifications,
3 and
so forth. So, clearly, I think there is
4
opportunity that is opening up and what you see in
5
front of you are questions of trying to make
6
decisions in the current mode and the future might
7 be
quite different.
8
DR. KIBBE: Let me just throw out
some
9
thoughts from listening to everyone.
We have been
10
trying to take a complicated situation and make an
11
easy rule, a simple rule. I think
Jurgen hit one
12 of
the points dead-on, and that is, I think we
13
really need a decision tree that looks at the
14
characteristics of the product we are dealing with
15 and
the therapeutic ranges that it is effective in.
16 We
have lots of data on a lot of these products in
17
terms of their therapeutic concentrations in the
18
body and how wide that can be and still get
19
reasonably safe therapeutic effects.
20
To make a rule that only responds to the
21
fact that the product is variable and doesn't have
22 a
basis for why we are allowing that variability or
23 why
we shouldn't allow that variability just
24
doesn't sound good to me. The
thought of going
25
outside the box with some solutions to some of
157
1
these problems, instead of going straight to
2
another bio study and redesigning a bio study--and
3
Gordon said, you know, what is wrong with designing
4
better dissolution testing? One
thing no one ever
5
said is, well, what is wrong with a different
6
animal model? You know, I have
had quite a bit of
7
success with the pig. The pig is
a good animal
8
model for human absorption in the GI tract and that
9 is
really what we are caring about--and the
10
controls, the negative controls are always tasty.
11
[Laughter]
12
The question here is too complicated for a
13
simple answer and whether we have enough data to
14 get
really a quality answer today is problematic.
15 I
am intrigued by scaling but only when the
16
supportive data makes sense that we should scale to
17
allow something to happen.
18
I love three-way and four-way studies
19
because they get at what we have been trying to get
20 at
for years, which is how variable is the product
21 and
how variable are the people we test it on.
22
I worked for a couple of years in a
23
contract research lab. We did ten
bio studies a
24
month and I will tell you that I don't think the
25
agency gets to see 40 percent of what we do, and I
158
1
think a lot of the companies, when they find that
2
they can't successfully formulate, they kill it and
3
none of those studies show up.
You might get some
4
useful qualitative information from the contract
5
research labs by just simply asking them to tell
6 you
how many studies they do that never make it to
7 the
agency so you have a handle on the denominator,
8 if
you will.
9
I think you have a real bear by the tail
10
here and I wish I had as much confidence in
11
Bayesian that would answer every question as my
12
colleague does, but I think really we have to
13
apply--we have to be willing for the agency not to
14
have a rule that everybody can look at in one
15
sentence and say I meet that rule or I don't meet
16
that rule. We have to be willing
to say good
17 science supports my product, good science
doesn't
18
support my product, and the agency can make a
19
decision based on a whole set of criteria.
20
The last thing is that there is lots of
21
information I would love to learn about the
22
process, supporting Gordon's idea of really
23
understanding the variables and understanding what
24 is
going on so I can make better decisions and I am
25
stuck with the single question of who is going to
159
1 pay
for that, and I don't see everybody rushing to
2 the
forefront to throw millions of dollars for
3
understanding it when what they really want to do
4 is
get the product on the market.
5
DR. MEYER: A really quick
follow-up to
6
Art, maybe there is a way for the innovator firms
7 to
have a little expansion of exclusivity if they
8
seek answers to some of the questions you would
9
really like to know about mechanisms.
It wouldn't
10
cost you a cent. It would cost
the American public
11 a
little bit but the return might be good.
12
DR. SINGPURWALLA: Well, I was
very
13
pleased to hear Art talk about using decision trees
14 but
was a little concerned when he said he doesn't
15
have that much faith in Bayesian methods. Well,
16 the
two are isomorphic, my friend.
17
DR. KIBBE: Pat?
18
DR. DELUCA: Just a comment, it
seems to
19 me
that the innovator wants a drug approved so it
20
should be incumbent upon the innovator to seek
21
answers to why there is that high variability. I
22
don't know if we need to give any more exclusivity,
23 I
think it ought to be incumbent upon them to do
24
this and to see whether that high variability is a
25
formulation or a physical property, as Gordon had
160
1
outlined.
2
DR. MEYER: Pat, when I was
talking about
3
exclusivity I was talking about something already
4
approved, much like the pediatric carrot--if you do
5
pediatric studies you get an extra six months; if
6 you
do mechanism studies you get another three
7
months, or whatever. And, if you
are making a
8
million a day that is a pretty good incentive.
9
DR. KIBBE: I don't know where the
10
incentive is for the company. If
I am an innovator
11
with an approved product on the market which has a
12 lot
of variability but is still approved and it is
13
clinically effective, and it has been sold and now
14 I
am producing X billion dollars worth of product
15
every year, do I really want to carefully define
16
that product so someone else can copy it? Or, do I
17
want to keep claiming that the trace elements in it
18
that come from the natural source are so important
19 if
they have to be assays so that I don't have to
20
have the problem? I mean, I think
Marvin is right,
21 you
have to have an incentive for them to get that
22
data for you.
23
DR. HUSSAIN: I just want to point
out
24
what Dr. Benet and Jurgen also pointed out, that I
25
think as you go through an approval decision to
161
1
approve a new drug product, the basis of it is
2
safety and efficacy and the risk/benefit decision
3
that is made. Often when you go
through that
4
process what results is that it is a safe and
5
efficacious product.
6
Now, PK variability, yes, we can measure
7
it. We know it is there and it
may not have any
8
bearing on that and that is what Dr. Benet started
9
discussing, and so forth. So,
keep in mind--Jurgen
10
keeps raising his hand again--what is the clinical
11
relevance. If we can measure it
and it is highly
12
variable, if it is not relevant we should probably
13 not
be measuring it.
14
DR. KIBBE: Paul, go ahead.
15
DR. FACKLER: I was just going to
make a
16
comment along the same lines. We
are aware of some
17
NDAs that have been approved without the BE studies
18
having to meet the traditional confidence intervals
19 on
both Cmax and AUC--
20
DR. HUSSAIN: Yes, it is done all
the
21
time. It is a clinical decision;
it is not a PK
22
decision.
23
DR. FACKLER: Right, where the
Division
24 can
say, you know, the Cmax isn't that relevant to
25
this therapeutic endpoint and, of course, then the
162
1
generic industry still has to meet Cmax even when
2
reference versus reference can't possibly pass it.
3 So,
I think it is a real problem for the generic
4
industry. Lawrence is right,
these rules apply to
5 new
drugs but the divisions have the authority I
6
suppose to waive a particular data requirement.
7
You know, as far as granting
extra
8
exclusivity to find the variability in a new drug,
9 I
don't think the generic industry is opposed to
10
doing replicate design studies, doing four-way
11
studies, to define the variability of the reference
12
product and I am guessing that the Division of
13
Bioequivalence would be interested to know the
14
variability in the test product they are
15
considering. So, I am sure there
are lots of ways
16 of
getting the information one needs to make a
17
decision about whether a product is highly
18
variable. I just wanted to point
out that there
19 are
products approved for which there is no way for
20 a
generic product to gain approval without
21
reference scaling, wider goalposts, whatever the
22
committee decides to recommend.
There needs to be
23 a
process for a certain fraction of products that
24 are
on the market today in the U.S.
25
DR. KIBBE: And I would hope that
the FDA
163
1
staffers will go away and give us a decision tree
2
with some understanding of the therapeutic outcomes
3 and
the risk/benefit of that product, how narrow
4 the
therapeutic range has to be, how variable it
5 is,
and then the goalposts can move based on a
6
decision tree and not have us reestablish another
7 set
that are just ticked. I think we have
lived
8
quite well with 80-125 but it was still picked.
9
Someone came up and put that down.
10
The other point I just wanted to emphasize
11 is
what Les said about how this plays out in the
12
public and among healthcare professionals, and his
13
concept of adding the point determination with what
14
would appear to be to them a narrower range or
15
allowable range might be something also to look
16
into.
17
DR. YU: I guess we will have to
look into
18
long-term solutions, the short-term solutions,
19
long-term objectives and short-term objectives. I
20
think mechanisms understanding of the causes of
21
variability and having some kind of decision tree
22
which you imagine is a great idea.
I think that is
23
long-term we ought to be looking for.
We ought to
24 be
looking at moving in this direction. We
also
25
have to balance the short-term objectives. If we
164
1
have not seen what the decision tree will look
2
like, how to implement them right now basically the
3
policy for the short-term is 80-125 percent
4
confidence interval, and you have already heard
5
that some drugs may be difficult to meet, maybe
6
never to be put on the market.
So, I guess this is
7 a
question put to the committee we will have to
8
discuss. In other words, we are
waiting to also
9
develop the great idea of long-term objective
10
decision trees and then put basically, given this
11
short-term period for the next decade, you will not
12
have those products. So, that is
a decision for
13
which we are seeking your advice.
Thank you.
14
DR. CONNER: One comment, I mean,
you
15
mentioned that we should come forward with a number
16 of
sets of data, including the therapeutic range of
17 the
product. Now, having been involved in at
least
18 one
working group where we were looking at NTI
19
drugs and saying, well, can we have a definition,
20 you
know, that can always be supported for a given
21
drug or new drug to say what its therapeutic window
22 or
therapeutic range is, we spent I think about
23
four or five years and realized we couldn't do it
24
because the data, even in an NDA, does not really
25
tell the true therapeutic range.
I mean, they have
165
1
some indicators that if I go up above a certain
2
point, you know, I don't get anymore efficacy and I
3
start to get something with disturbing side effects
4
but, you know, they usually don't have a full
5
characterization of therapeutic range.
Plus, the
6
definition of what indicators I look for and, you
7 know,
do I take a ratio and do I look at this
8
versus that, you know, for any given drug it is
9
just not there. Even if you had
infinite amount of
10
data, it is very hard to decide what I should look
11 at
and what I should use. So, it is easy to
say I
12
want to know about the therapeutic range but the
13
data to determine it so that everyone will agree on
14 it,
and determine it with certainty just isn't
15
there. We spent a long time
really trying to do
16
that and finally gave up, unfortunately.
We are
17
still interested in the topic but, you know, we
18
realize that it is a lot harder to do than most
19
people realize.
20
DR. KIBBE: Marvin?
21
DR. MEYER: Since it is in the
interest of
22
everyone who is trying to sell a highly variable
23
drug, it would seem to me that a number of
24
companies would give a release to MDS, if asked, to
25
just have a disguised set of data--you don't even
166
1
have to say class of drugs.
2
I kind of favor number two, reference
3
scaling with a test of a reference stipulation, but
4 I
would be interested in seeing the 66 percent of
5
studies that failed and what would you have to do
6 to
your limits in order to get them to pass and
7
work with real data. Right now it
is somewhat
8
hypothetical.
9
DR. CONNER: Well, that could be
done and
10 it
would provide more evidence and information
11
about the immediate problem but I don't think that
12
partial knowledge of those would get at the root
13
cause, which is what some members have said they
14
want to see. I mean, you really have
to know what
15 the
nature of those drug substances is and how they
16 are
formulated. You have to know a lot of
factors
17 and
relate that to what you saw in trying to get at
18 the
problem and its root causes.
19
DR. MEYER: Yes, that is a
laudable goal
20 but
I thought we wanted an answer while we are
21
still alive.
22
[Laughter]
23
DR. CONNER: I expect you will be
around
24 for
quite a while.
25
DR. KIBBE: Gary?
167
1
DR. BUEHLER: We do want an answer
while
2 you
are still alive, Marvin. This is a big
issue
3 for
us. I am one that always says that we
have to
4
bring difficult issues to the advisory committee
5
and, you know, we bring sort of soft balls to you.
6 I
really made the point that this is a very
7
difficult scientific issue. It is
an economic
8
issue. As was brought up today,
clearly it is an
9
economic issue but the Office of Generic Drugs is
10 an
economically driven office that makes scientific
11
decisions and makes these economic decisions in a
12
scientific way. We have products
out there that we
13
can't get generics of, and that was made evident by
14 Dr.
Fackler, from Teva, and he knows that probably
15
better than I do. But generic
drugs are a big
16
political issue. They are a big,
passionate issue
17
with the American public today.
People see
18
generics as an answer to the high cost of
19
prescription drugs today.
20
So, what we are bringing to you today--I
21 am
not saying that a decision tree is not a good
22
idea; I think it is a great idea, but I agree with
23 my
colleagues from the Office of Generic Drugs that
24 a
decision tree can be an awfully long process to
25 put
together and to gather the data from all the
168
1 various
drugs, and I am not sure I have the staff
2 to
be able to do that within the next millennium or
3
so. So, what we would like from
you today is some
4
direction as to which way to go.
If you would be
5
able to provide that to us somehow, we would
6
appreciate that.
7
DR. KIBBE: I think we have heard
from Dr.
8
Meyer that he prefers scaling.
How many of us
9
think that that is an option for situations that
10
seem to be highly variable and need an evaluation
11
outside of the current rules?
12
DR. SINGPURWALLA: I am sorry, I
think one
13 has
to put things sometimes rather bluntly.
I feel
14
that those questions that you have asked are the
15
wrong questions, or there should be additional
16
questions, namely, what are the alternatives? The
17
decision tree, as Jurgen puts it, is a very good
18
alternative.
19
The second point is the scaling.
The
20
scaling has been talked, and talked, and talked
21
about. There is a simple reason
why you do the
22
scaling. The scaling is a
transformation which
23
tries to bring the variability down.
If the
24
scaling does not bring the variability down no
25
statistician will do it. Its main
purpose is
169
1
two-fold, to make everything look approximately
2
normal and, in the process, bring the variability
3
down. So, the scaling is a
technical exercise
4
which nobody should question or criticize whenever
5 it
is appropriate and it is not a debatable issue.
6
The issue that is really debatable is are
7
those the right kind of questions and do we want to
8
pursue that line of questioning.
What I would like
9 to
do, if Mr. Chairman would allow me sometime
10
later in the afternoon, is to ask everyone around
11 the
table what do they mean by confidence limits;
12
what is its interpretation; and how is it
13
understood. And, I will make a
bet that fifty
14
percent will get the answer wrong, at least. Thank
15
you.
16
DR. KIBBE: It is against federal
17
regulation to gamble in Washington, D.C.
18
[Laughter]
19
There will be no betting going on!
20
Jurgen, what do you think?
21
DR. VENITZ: I think it is time
for lunch.
22
DR. KIBBE: No, no, no one is
going to
23
lunch until we answer his question.
What do we do
24 in
the short term?
25
DR. VENITZ: Well, in the short
term I
170
1
don't think there is anything wrong with reference
2
scaling the way I understand it.
I had some
3
question about how you are going to get the
4
estimate for your within-subject reference
5
variability but if that is part of a replicate
6
design study or separate study, I think I can live
7
with that. I still think, as I
said before with
8
Les, you should have additional constraints on the
9
point estimates, and it might just be for public
10
consumption so everybody on the outside that is the
11
recipient of whatever we come up with today feels
12
comfortable, yes, the rules are not being bent to
13
make bad products look good or, you know, highly
14
variable drugs look good. But
other than that, I
15 can
live with this as a band aid. I do think
you
16
should start working on the long-term strategy,
17
which comes back to the decision aspects.
18
DR. KIBBE: Gordon, what do you
think?
19
DR. AMIDON: I would agree with
the
20
scaling. Again, the question of
how you get the
21
reference scaling, I think the last point on the
22
reference scaling is a good starting point to look
23 at
in trying to make that in a concrete decision
24
rule. I still think that the
mechanism, you know,
25
what is going on with highly variable drugs, where
171
1 the
problems are, that is the real long-term
2
solution, understanding what the problem is and the
3 FDA
should in some way put some resources into
4
that, and I think Marvin Meyer's suggestion is an
5
excellent one. You know, just
provide some
6
incentive for industry to fund research into what
7 is
happening with these drugs.
8
DR. KIBBE: Judy?
9
DR. BOEHLERT: I have thought
about this a
10 lot
and I also would agree with using the scaling
11
factors but I think you are still going to be in a
12
position of having to make decisions and having
13
some kind of decision tree, even if it is not
14
formal because Dr. Davit presented data this
15
morning that showed that the same product with two
16
different laboratories had different values. So,
17 you
are going to be in that situation where one
18
manufacturer uses scaling and the other one doesn't
19 so
you are going to have to make some decisions
20
around those issues. It is not so
straightforward,
21
particularly when you get around that 30 percent
22
number.
23
DR. BUEHLER: Making decisions is
"an
24
understood" for me so I can accept that. But I
25
echo Jurgen. We have to make sure
that whatever we
172
1
decide will provide a good scientific method so
2 that
the generic products that go on the market as
3 a
result of this are unequivocally bioequivalent
4
and, of course, safe and effective.
5
DR. KIBBE: My colleague who
hardly ever
6
speaks?
7
DR. SELASSIE: I agree with what
Jurgen
8
said because I think that there needs to be a
9
mechanistic basis as to what type of scaling
10
factors you use, and it seems to me that that is
11
really important and we need to understand the
12
physicochemical parameters that are involved in
13
dissolution and it seems like that is missing and
14 is
arbitrary in trying to set some scaling factor,
15 and
we are not taking those types of phenomena into
16
consideration. So, I think a
decision tree in the
17
long-term would be a good idea but I guess in the
18
interim you can use something like reference
19
scaling.
20
DR. KIBBE: Marc?
21
DR. SWADENER: I think it is a
little bit
22
naive to think that all of us here, at every stage
23
along the line, don't use a decision tree of some
24
sort. Formalizing it to the stage
that people are
25
talking about here is a little diplomatic but we
173
1 all
have a decision tree that we use.
Whether I
2
came here to this meeting or not, I used one.
3
Whether it was formalized or not is a different
4
question.
5
I do know enough about statistics to know
6 you
can't believe them all the time. You
have to
7 be
very, very careful about them. So, I
think I
8
would encourage looking at a decision tree not as
9 the
short term as it will take time, and do the
10
best you can. I do agree with
Jurgen, you really
11
have to look at what are the real fundamental
12
questions you are dealing with too.
With my
13
representation on this committee, the public just
14
needs to know that what they are getting is safe
15 and
will do what it says it will do. Now,
that is
16 a
very simple approach but they don't know all this
17
stuff and they are relying on you to do the best
18 you
can. I don't see that you are not doing
that.
19
DR. KIBBE: Dr. Koch?
20
DR. KOCH: Yes, I would agree with
the
21
summary that you came up with and certainly stay
22
with the reference scaling short term but something
23 has
to be put together to address the decision tree
24
approach.
25 DR. COONEY: I think that the change in
174
1
limits is not acceptable because it is very
2
arbitrary and if there are options I think the use
3 of
reference scaling makes fundamental sense.
4
Furthermore, a decision to do that is a decision
5
down the path of a decision tree so it is a logical
6
step to take and I want to underscore the
7
importance of continuing to gather the data and
8 establish
the criteria around which these
9
individual decisions are made and you will be in a
10
better position to do this goring forward.
11
DR. KIBBE: Pat?
12
DR. DELUCA: I am certainly an
advocate of
13
getting those drugs that are safe and effective to
14 the
market. Certainly, the public would
benefit
15
from those. I guess I favor in
the short term--I
16
think in the long term something more substantial
17 has
to be done with regards to decision making, but
18
reference scaling I think is very important here.
19
Again, I don't like the arbitrary nature of
20
widening the limits but if that is something that
21 can
be approached in the short term, then I would
22 be
for it.
23
I still think that we need to encourage
24 the
innovator to finance for the high variability
25
that exists. Whether it is
offering incentives in
175
1
some way, so be it but certainly an incentive if
2
they can reduce the variability.
It seems to me
3
they gain something when the generic has to go into
4 the
reference scaling, they have improved the
5
product so I think that is also an incentive to do
6 it.
7
DR. KIBBE: Our industry
representatives
8
have a comment one way or the other?
I would just
9
like to wrap up and go to lunch but with one
10
comment. I think if we go ahead
and make a change
11 in
the way we approve highly variable drugs, then I
12
think we ought to consider seriously also Les'
13
other point which is to come up with something that
14 is
going to reassure the public that the changes we
15 are
making are not getting drugs that can vary by
16 50
percent on the marketplace but, rather, that
17
they really are tighter than that so they
18
understand it better. So, I would
end with that.
19
With that said and no one else waving frantically
20 to
get my attention, we will break for lunch and we
21
will be back at 1:40.
22
[Whereupon, at 12:40 p.m. the proceedings
23
were recessed for lunch, to resume at 1:40 p.m.]
176
1
A F T E R N O O N P R O C E E D I
N G S
2
[Because the Chairman reconvened the
3
proceedings before 1:40 p.m., part of the text is missing.
4
There were no speakers who wished to speak
5
during the open public hearing but there was a public
6
submission from Zeb Horowitz, M.D.]
7
Bioinequivalence: Concept and Definition
8
[Slide]
9
DR. YU: ...bioavailability is
rate and
10
extent of absorption and it is the site of drug
11
action. So, normally you give a
drug to a healthy
12
volunteer or patient and measure the plasma
13
concentration against time, as shown in this
14
figure. Then you get a Cmax here;
you get the AUC,
15 AUC
is area under the curve. Cmax is a
surrogate
16 for
the rate of drug absorption; AUC is basically
17 for
the extent of absorption so this is defined as
18
bioavailability.
19
[Slide]
20
Bioequivalence basically is defined as the
21
absence of a significant difference in the rate and
22
extent to which the active ingredient or active
23
moiety in the pharmaceutical equivalents or
177
1
pharmaceutical alternatives become available at the
2
site of drug action when administered at the same
3
molar dose under similar conditions in the
4
appropriately designed study. So,
this basically
5 is
the Federal Register Notice definition.
So,
6
bioequivalence basically means the absence of a
7
significant difference in the rate and extent of
8
drug absorption.
9
[Slide]
10
This morning we discussed bioequivalence
11 and
we said a 90 percent confidence interval for
12 the
extent or AUC for the rate as a surrogate or
13
Cmax 80-125 percent. Now, passing
or meeting the
14
bioequivalence standards allows marketing access
15
basically as one of the standards for approval of
16 the
applications. Of course, you have to meet
17
very many other requirements, especially with
18
respect to the chemistry, manufacturing controls
19
with respect to quality of the products.
So,
20
demonstration of bioequivalence makes the generic
21 to
be approved and the innovator basically
22
demonstrates that the marketed formulation is
23
equivalent to the clinical formulation.
24
[Slide]
25
The question is why do we define the
178
1
bioinequivalence concept? What
are you talking
2
about here? Why do you define
this? It is because
3 FDA
receives studies that attempt to reverse a
4
previous finding of bioequivalence.
In other
5
words, you approve a product to put on the market
6
when some manufacturer conducts a study to show or
7
fails to show the bioequivalence.
Also, in the
8
public literature there are claims of
9 bioinequivalence. In reality, it is simply a
10
failure to demonstrate bioequivalence.
So, there
11 is
a concept you need to clarify, what is called
12
bioequivalence and what is called bioinequivalence.
13
What is the difference when you fail to demonstrate
14
bioequivalence and bioinequivalence?
15
[Slide]
16
There are many reasons, as we discussed
17
this morning, for high variability--under-powered
18
study designs, study samples, many, many reasons
19
that can make a study fail. Of
course, the easiest
20
way, as we discussed this morning, is to use a
21
small number of subjects. So, it
is easy to fail
22 to
show bioequivalence by a small number of
23
subjects and, certainly, there will be other
24
considerations like study design, study sample,
25
data analysis. There are many,
many other reasons
179
1 and
these are just several of them.
2 [Slide]
3
What should bioequivalence mean if we
4
define a definition for bioinequivalence? As we
5
said, bioequivalence leads to market access.
6
Basically a study that demonstrates bioequivalence
7 is
clear and convincing evidence of equivalence.
8
Bioinequivalence may lead to market exclusion. Of
9
course, we have to consider many, many other
10
factors too as we discussed this morning--safety,
11
efficacy, pharmacokinetics, pharmacodynamic
12
relationship and so on. But a
bioequivalence study
13
demonstrated by equivalence is clear and convincing
14
evidence of potential problem for the specific
15
product.
16
[Slide]
17
So what do we specifically mean here?
I
18
want to spend a little time on this slide. When
19 you
conduct a study, if a study is properly
20
designed, the 90 percent confidence interval is
21
between 80 to 125 percent. Now,
if this study is
22
under-powered and if this study has a small number
23 of
subjects, there is a greater possibility that it
24
fails to demonstrate bioequivalence.
What this
25
specifically means is simply that the manufacturer
180
1 or
sponsor does not use enough subjects for
2
example, of course, among many other reasons, to
3
conduct a study. If the study is
powered enough,
4
there is a greater possibility that you can narrow
5 the
confidence interval and make this a passing,
6
successful study. Coming back to
so-called
7
inequivalence is to make sure that the test product
8 has
a difference more than 20 percent or, for
9
example is underneath the 80 percent or above 125
10
percent. Of course, there is also
the failure to
11
demonstrate a bioequivalence study because simply
12 the
top limit above 80 or, on the other side of the
13
lower limit it may be below 125.
14
I think it is in the best interest of the
15
public and us, for clarification, that we want to
16
define the bioequivalence, bioinequivalence,
17
failure to demonstrate bioequivalence and failure
18 to
demonstrate bioinequivalence. This is a
concept
19 that
we have to be very clear about because in many
20
cases in the published literature or studies
21
submitted to the Food and Drug Administration are
22
simply that. For example, the top
limit is above
23 125
percent or the lower limit is below 80 percent
24 if
you use enough power and increase the subjects
25 of
the studies the study will become a successful,
181
1
passing study instead of failure to demonstrate
2
bioequivalence.
3
Yet, because of confusion because there is
4 no
clear definition with respect to bioequivalence,
5 in
the end any study, whether the lower limit is
6
below 80 and upper limit is above 125, the sponsor
7 or
other parties will have bioinequivalence.
The
8
reality is simply to fail to demonstrate
9
bioequivalence. In other words,
the true
10
difference is acceptable, however, the study is not
11
properly designed because it is under-powered, or
12
many, many other reasons where the confidence
13
interval does not meet the regulatory criteria,
14
which is 80-125 percent. At the
end, the claim is
15
basically bioinequivalence and in reality, as I
16
said, is a failure to demonstrate bioequivalence.
17 So,
I want to make it clear, I want to clarify the
18
concept.
19
[Slide]
20
So, the objective at FDA is to develop
21
bioinequivalence criteria that are scientifically
22 sound, statistically valid and fair to all
parties
23
and, hopefully, easy to use.
24
With this introduction, I want to turn the
25
podium to Don and I am sure a lot of people know
182
1
him. He is the developer of the
original 80-125
2
percent criteria for FDA standards.
He will be
3
speaking about how to statistically establish
4
bioinequivalence.
5
DR. MEYER: A real quick question,
I don't
6
quite catch the fail to demonstrate
7
bioinequivalence for the one where the right-hand
8
tail is barely across 80 but the point estimate is
9
well to the left of 80. It seems
to me that still
10 is
a bioinequivalent product.
11
DR. YU: This one?
12
DR. MEYER: Yes, with the point
estimate
13
falling well to the left. It
seems to me changing
14 the
N won't help that one. It will just make
the
15
confidence limits fall, totally bioinequivalent.
16
DR. YU: Yes, most likely if this
study is
17
powered--to increase, for example, the power of
18
this study this product is bioinequivalent. This
19
time it is because the confidence interval above 80
20 statistically
speaking, as Don can clarify, failed
21 to
demonstrate whether it is truly bioinequivalent
22 or
not. I think that Don is the better
person to
23
answer the question.
24
Statistical Demonstrations of Bioinequivalence
25
MR. SCHUIRMANN: One clarification
to what
183
1
Lawrence said, I did not have anything to do with
2
choosing 80-125 as the limits for bioequivalence.
3
[Laughter]
4
[Slide]
5
This presentation is joint work with
6
colleagues in the Quantitative Methods and Research
7
staff of the Office of Biostatistics and also in
8 the
Office of Generic Drugs, and the bulk of the
9
presentation was put together by my colleague, Dr.
10
Qian Li, who originally was scheduled to give this
11
presentation but she just recently had a baby so
12 she
is having a little deserved maternity leave.
13
[Slide]
14
We hope to go over the definition of
15
bioinequivalence, comments on claiming
16
bioinequivalence if you fail to show
17
bioequivalence, proposing a criterion to use for
18 one
PK endpoint--PK is pharmacokinetic, and talk
19
briefly about strategies when you are looking at
20
three pharmacokinetic endpoints.
21
[Slide]
22
The usual definition of the bioequivalence
23
interval on the ratio of the population geometric
24
mean of the test product over the population
25
geometric mean of the reference product is that it
184
1
should fall within the limits of 80 percent to 125
2
percent. That is what is called
the bioequivalence
3
interval. It is never correct to
refer to that as
4 a
confidence interval. So, it is obvious
to define
5 the
bioinequivalence region as just the complement.
6 If
you are not in the bioequivalence interval, then
7 you
are in the bioinequivalence region which
8
consists of the two disjoint regions.
9
[Slide]
10
So, the question that I first want to look
11 at
is, is it appropriate to claim bioinequivalence
12 if
a study fails to show bioequivalence?
Two
13
products may, in fact, be bioequivalent but they
14 may
not be shown to be bioequivalent by the study.
15 The
primary reason for that is inadequate power.
16
There could possibly be other reasons.
17
[Slide]
18
In doing our standard testing for
19
bioequivalence, it is an application of statistical
20
hypotheses testing where we have a null hypothesis
21
that says either the ratio of geometric means is
22 too
low, below 80 percent, or else it is too high,
23
above 125 percent, and we test that against the
24
alternative hypothesis that the ratio of geometric
25
means is within the interval. The
way that we have
185
1
typically tested this statistical hypothesis is by
2
doing two one-sided statistical tests, and each of
3
those tests is carried out at the alpha equals 0.05
4
level of significance.
5 Now, it turns out that doing these two
6
one-sided tests--it is an example of what is called
7
intersection union test--is algebraically
8
equivalent to calculating a two-sided 90 percent
9
confidence interval and seeing whether it falls
10
within the equivalence interval.
So, that is why
11 you
hear a lot of talk about confidence intervals
12
today even though we are not using the confidence
13
interval as a confidence interval; we are using the
14
endpoints of the confidence interval as test
15
statistics. What we are doing
here is statistical
16
hypothesis testing. As I said,
the type 1
17
error--we have to reject both one-sided null
18
hypotheses, both H-nought 1 and H-nought 2. If we
19
reject both one-sided null hypotheses, then we
20
conclude that this alternative is true, that is,
21
that we have average bioequivalence.
22
[Slide]
23
So, we need to reject the hypothesis of
24
inequivalence with high confidence and the
25
rejection region is selected to make the chance of
186
1
doing that incorrectly to be small, and that is the
2
level of significance which, as I said, is alpha
3
equals 0.05.
4
[Slide]
5
So, what is the error associated with
6
claiming inequivalence if you don't claim
7
equivalence? Well, if you are
looking for a
8
procedure for testing to see if you have
9
inequivalence, then we need to control the error
10
wrongfully rejecting equivalence to be small. If
11 you
are going to base it on the equivalence test,
12
that means you want the equivalence test to have
13
large power. However, the power
for the
14
bioequivalence test, as you will in a moment, may
15 not
be large overall values of the geometric mean
16
ratio in the equivalence region.
17
The testing for bioequivalence focuses on
18
controlling the type 1 error and then other aspects
19 of
the test, such as high power if the alternative
20 is
true, are gotten, if they can be. So, we
may
21 not
have adequate power to claim bioequivalence
22
even when bioequivalence is true.
23 [Slide]
24
Here is an example. If we had a
product
25 and
we are going to design a two-period, two
187
1
sequence bioequivalence trial, and we assume we
2
have within-subject variance of 0.04, that is to
3 say
within-subject standard deviation of 0.2, and
4 we
are willing to assume that the ratio of
5
geometric means deviates from 1 by no more than 5
6
percentage points and, if that is true, we want to
7 be
at least 85 percent sure that we will reach a
8
conclusion of equivalence, you can then crank the
9
numbers and you come up with the sample size of 22
10
subjects.
11
Well, if you have 85 percent power, that
12
means you are 85 percent sure of concluding that
13 the
products are equivalent. That means you
could
14
have as much as a 15 percent chance of not
15
concluding that they are equivalent.
So, even with
16
this design study you could have products that are
17
equivalent but you have as much as a 15 percent
18
chance, or even more, of failing to conclude that
19
they are equivalent. In fact, if
the variance,
20
unbeknown to you, is higher than you thought or if
21 the
geometric mean ratio deviates from 1 by more
22
than 5 percentage points, the power will be lower
23 so
the chance of not concluding bioequivalence will
24 be
higher. So, it should be apparent that
that is
25 not
a basis for concluding inequivalence.
188
1
[Slide]
2
The rejection region for the
3
bioequivalence test--what do I mean by rejection?
4 I
mean rejecting the hypothesis of inequivalence
5 and
concluding equivalence--is determined by the
6
variability associated with the point estimate of
7 the
geometric mean ratio, which is illustrated here
8 on
the log scale. The higher that standard
9 deviation
of the estimator is, the further away
10
from the actual limits--delta-2 here is 1.25;
11
delta-1 is 0.8--the narrower that rejection region
12
will be and the lower the power will be.
So, it
13
isn't enough for the point estimate to be within
14 the
equivalence interval. It has to be
comfortably
15
away from the edges of the equivalence interval in
16
order to conclude equivalence.
17
[Slide]
18
This is an example of power curves for the
19
test for equivalence. The blue
lines correspond to
20 a
standard deviation on the log scale of 0.5.
The
21 red
lines correspond to a standard deviation of 0.3
22 and
the green lines correspond to a standard
23
deviation of 0.1. The solid lines
are for a study
24
with 60 subjects. The
corresponding dashed lines
25 are
for a study with 30 subjects.
189
1
Let's take this example, a 60-subject
2
study but the standard deviation is 0.5, even if
3 you
are exactly equivalent, you are identical,
4
there is a very good chance that you will not
5
conclude equivalence so that is no basis for
6
concluding inequivalence.
7
[Slide]
8
So, instead of trying to use the
9
equivalence test as a means for establishing
10
inequivalence, we need to develop a testing
11
procedure aimed specifically at inequivalence.
12
Here we have done that by reversing the hypotheses.
13 Now
the null hypothesis, in statistical jargon, is
14
that the geometric mean ratio is within the
15
interval. The alternative is that
it is either
16
below 80 percent or else it is above 125 percent.
17 So,
once again try to test this hypothesis by doing
18 two
one-sided tests, each at a level of 0.05, and
19 in
the case of equivalence we had to reject both of
20 the
one-sided hypotheses but in this case we have
21 to
reject one or the other. So, we will
reject
22
bioequivalence and conclude bioinequivalence if one
23 of
these two one-sided hypotheses is rejected.
It
24
says here under certain conditions and, in fact,
25
under most conditions the overall level of that
190
1
procedure will not be appreciably different from
2
0.05, however, we can find mathematically cases
3
where it could be higher. It
could be as high as
4
0.1.
5
[Slide]
6
Before I showed you the region for
7
concluding equivalence and that is these lines,
8
here. Now the region for
concluding inequivalence
9 is
given by this line and this line. You
have to
10
fall higher than this with the point estimate or
11 you
have to fall lower than that in order to
12
conclude inequivalence. So, once
more you need to
13 be
comfortably away from the actual boundary before
14 you
reach your conclusion.
15
[Slide]
16
This is what the power of that test looks
17
like. The color scheme and the
solid and dashed
18
line scheme is the same before.
These vertical
19
lines are the 0.8 to 1.25 lines and so if you are
20 in
the interval, 0.8 to 1.25, the probability never
21
gets higher noticeably than 0.05.
But if you are
22
outside of the interval, then you have a greater
23
chance of concluding bioinequivalence.
I might add
24
that it is symmetric in the reciprocal of the
25
ratio. In other words, here is a
ratio of 0.5 and
191
1 it
has a certain high probability of concluding
2
inequivalence. To have the same
probability over
3 on
the other side, it would have to be equal to 2,
4
which is the reciprocal of 0.5.
5
[Slide]
6
We are going to try to control that error
7 to
0.05. It is a function of what in this
slide is
8
designated sigma-T, which is the standard deviation
9 of
the estimator that is used as the basis for the
10
test statistic. As that sigma-T
gets larger you
11
could possibly have more than a 5 percent chance of
12
wrongfully concluding inequivalence.
13 [Slide]
14
Well, how big would the variance have to
15
be? Dr. Li ran some
calculations. In this
16
example, here, N equals 10. A
bioequivalence study
17
with only 10 subjects I don't think would even be
18
allowed. These studies tend to be
considerably
19
larger than 10 subjects. But it
just illustrates
20 the
fact that even for that tiny sample size the
21
standard deviation, which is on the log scale, has
22 to
be quite large before the chance of making the
23
wrong decision, that is to say concluding
24
inequivalence when, in fact, the products are
25
equivalent gets unacceptably high.
If you have a
192
1
more reasonable number of subjects it has to be
2
astronomically high before you start running into
3
that problem.
4
There could be cases perhaps with a
5
parallel design of a highly variable drug where we
6
might have to do some adjustment to the
7
significance level, and there do exist methods in
8 the
literature to do that.
9
[Slide]
10
So, that is the corresponding test to the
11
bioequivalence test for one parameter, but
12
generally we assess bioequivalence studies with
13
respect to three endpoints--I said parameter,
14
didn't I? Pharmacokineticists
love to use the word
15
"parameter" to describe AUC and Cmax; statisticians
16
don't. Typically, we require for an
equivalent
17
study you have to show equivalence for area under
18 the
curve to the last sampling time; area under the
19
curve extrapolated to infinity; and maximum
20
observed concentration. What are
we going to do
21
about bioinequivalence? In
concept the products
22 are
bioinequivalent if they are bioinequivalent
23
with respect to just one of these three.
24
[Slide]
25
So, what statistical criteria shall we
193
1
use? We are looking at a number
of strategies.
2
Strategy one is to say, well, if you conclude
3
bioequivalence with respect to just one of the
4
three pharmacokinetic endpoints, then you will
5 reach a conclusion of bioinequivalence. The things
6 in
favor of that is that it is quite intuitive.
7 The
arguments against it are that you now have
8
three chances--if you have a case where the
9
products are close to being inequivalent but they
10
aren't inequivalent, then you have three chances to
11
make a mistake and you may inflate the overall type
12 1
error rate.
13
[Slide]
14
So, if you are worried about that, here is
15
another strategy which says, well, you have to show
16
that it is inequivalent with respect to all three
17 of
the PK endpoints. Then you can tightly
control
18 the
type 1 error rate. Type 1 error in this
case
19
means concluding inequivalence when, in fact, the
20
products are equivalent. But the
argument against
21
this strategy is that it is not going to have
22
reasonable power against alternatives of interest.
23
[Slide]
24
Another possibility would be to say, well,
25 you
need to prespecify which endpoint you are going
194
1 to
look at. In the slide here AUC was used
as an
2
example. Possibly Cmax would be
another choice.
3
This will control the type 1 error but if the
4
endpoint you chose is not the endpoint for which
5 the
products are inequivalent, then you are not
6
going to have a reasonable chance of reaching a
7
proper conclusion.
8
[Slide]
9
Other strategies require that you show
10
equivalence for all three but you adjust the alpha
11
levels so the overall level is maintained but you
12
have more power for each individual test. A method
13 to
do this which doesn't require the levels to be
14 the
same for all three endpoints is currently under
15
development in QMR.
16
Other possibilities--one that occurred to
17 me
is you might say, well, before you can conclude
18 that
the products are inequivalent with respect to
19 AUC
you have to show inequivalence for both AUC to
20 the
last time point and also for AUC to infinity
21 but
we will look at Cmax separately. But
there
22
could be regulation complications in all of these
23
proposals.
24
[Slide]
25
So, the main focus of this presentation
195
1 was
on power to make the right decision and what we
2 are
calling error, which is making the wrong
3
decision and controlling the probability of making
4 a
wrong decision. There could be other
statistical
5
issues as well. Thank you.
6
DR. KIBBE: We will open it up for
7 questions
from the panel. Marvin, go ahead.
8
DR. MEYER: Don, a practical
example of
9
what would happen with strategy one if the type 1
10
error were inflated and the three PK endpoints are
11 now
highly correlated--
12 MR. SCHUIRMANN: I apologize, Dr. Meyer, I
13
didn't bring those numbers with me--
14
DR. MEYER: Just conceptually.
15
MR. SCHUIRMANN: Suppose you had a
product
16 for
which the ratio of the population of the
17
geometric means was something like 124 percent for
18 all
three parameters. Then, the chance that
you
19
will conclude inequivalence for at least one of
20
them could be something like 15 percent, in that
21
neighborhood, depending on the sample size;
22
depending on how tightly correlated the AUC is with
23 the
Cmax. I can't give you a very quick
answer.
24 In
some of the simulations that Dr. Li did about 15
25
percent was the highest I saw.
So, it depends on
196
1
whether you are interested in controlling that
2
overall level or whether you are merely interested
3 the
level in each individual endpoint.
4
DR. BENET: Since this would be a
test to
5
take a drug approved off the market, have you
6
considered that maybe we need more than one study?
7
Have you talked about that? Have
you thought about
8
that in your thinking about it?
9
MR. SCHUIRMANN: I can't speak for
the
10
Center. I have not thought about
that much.
11
DR. BENET: Well, I was reacting
to
12
Barbara's data where you looked at the two
13
different studies with two people running it with
14 significantly
different variance in the two
15
different studies, and that could be an issue here.
16 You
know, I think it is a statistical issue but it
17 is
also a policy issue in terms of, you know, is
18 one
study going to be adequate? No matter
which of
19
these terrible suggestions you pick, is one study
20
going to be adequate?
21
MR. SCHUIRMANN: On the one hand,
22
requiring two studies would bring it more in line
23
with what we require for Phase III clinical trials
24
where we want a reproducible result so you have to
25
show us more than once. On the
other hand, we
197
1
approve generic drugs with only one bioequivalence
2
study. So, what would be the
basis for requiring
3 two
studies for the opposite claim?
4
DR. VENITZ: But you already have
two
5
studies. Don't you have a prior
study that led to
6 its
approval as a bioequivalent generic and now you
7
have a study to disprove it. So,
my question is
8
somewhat related to what Les is asking, how do you
9
incorporate the prior information that you have
10
from the fact that your drug got approved based on
11 a
bioequivalence study? Because you now
have one
12
study done God knows how long ago--
13
MR. SCHUIRMANN: Yes.
14
DR. VENITZ: --but it passed
15
bioequivalence. Now you have done
a study, no
16
matter what method you use, that shows
17
bioinequivalence. Are you going
to pool the
18
studies? Are you going to use
Bayesian to
19
incorporate your prior information or are you
20
completely ignoring the fact that in order to get
21
approval it must have passed a bioequivalence
22
study? And this is not a question
to you but to
23
everybody.
24
DR. BUEHLER: Usually when we get
a
25
challenge study now we will inform the generic
198
1
sponsor of the generic application that their
2
bioequivalence has been challenged, and that they
3 can
come back to us with additional data, usually
4
another study, which would refute the study that
5
came in. We usually review the
study extensively,
6 the
challenge study extensively to make sure that
7 the
study was conducted properly. We review
it as
8 far
as it was powered correctly, etc. Then
we give
9 the
generic firm that was challenged the
10
opportunity to come back to us with a study or else
11
face being downgraded in the "Orange Book."
12
DR. VENITZ: But if they don't
come back
13 do
you ignore the fact that they must have done a
14
study in the first place that demonstrated
15
bioequivalence?
16
DR. BUEHLER: No, we don't ignore
that
17
fact. That is why we leave them
on the market
18
while they get the additional data to us. I mean,
19
they did submit a study to us that showed their
20
product to be bioequivalent. Now,
whatever
21
happened along the way, you know, whatever water
22
flowed under the bridge between then and the time
23
when we have had the challenge study, you know,
24
sometimes it is a long time.
Sometimes
25
formulations change or reference listed drugs
199
1
change so we give them the opportunity to come back
2 to
us with another study to show that they are
3
still bioequivalent.
4
DR. KIBBE: Go ahead, Marc.
5
DR. SWADENER: Is my intuitive
notion that
6
these strategies one, two and three could result in
7
failure to agree on the hypothesis that it turns
8 out
that is not inequivalent doesn't necessarily
9 say
that it is equivalent? Aren't there
parts
10
where it is really not equivalent?
11
MR. SCHUIRMANN: We are talking
about
12
studies here and there is such a thing as an
13
inconclusive study, a study that does not establish
14
that two products are equivalent and the study also
15
does not establish that the two products are not
16
equivalent.
17
DR. SWADENER: Exactly.
18
MR. SCHUIRMANN: That isn't to be
confused
19
with the actual reality unknown to any human being
20
whether they are or aren't equivalent.
With these
21
strategies, depending upon how stringent you make
22 it,
you could very well have data that, as a
23
clinical, you look at and it worries you--"gosh,
24
these products sure differed a lot in this
25
study"--but it is not conclusive that they are
200
1
inequivalent. So, yes, you could
have that
2
situation very easily.
3
DR. KIBBE: Marvin?
4
DR. MEYER: Gary, did I understand
you to
5 say
that if in the initial study the generic
6
product came in, let's say, at 80-125, just hit the
7
upper and lower limit, and then the challenge study
8
came in at 79-125 the generic would have to redo
9
their study?
10
DR. BUEHLER: No. That is part of the
11
reason for this exercise, that is, we do face
12
situations like that where we will get very
13
marginal challenge studies submitted and where a
14
reasonable person could say, gee, you if threw
15
another six patients or subjects into that study
16 and
you probably would have been 80 or 81.
So,
17
what we are looking for here is to try to set up
18
some guidelines as to what will be acceptable as a
19
challenge to the bioequivalence.
20
DR. KIBBE: Let me ask Jurgen's
question,
21
which is when the challenge comes in is there any
22
thought to how clinically significant it is what
23 the
challenge study shows? I mean, is it
24
clinically significant relative to the use of the
25
drug itself?
201
1
DR. BUEHLER: The challenge study
is
2
reviewed and we make an assessment as to whether
3 the
challenge study, as I said, was conducted
4
properly and powered properly. If
the condition is
5 that we believe that more subjects, you know,
would
6
have thrown it over the line we normally make the
7
generic do another study to prove their
8
bioequivalence. Now, that is a
value judgment with
9
respect to what to do. Again, that
is one of the
10
reasons we are here right now. We
would like to
11
have a little more certainty in making this
12
decision as to when a generic has to repeat their
13
study.
14
DR. KIBBE: Gordon, go ahead.
15
DR. AMIDON: Yes, I think we are, again,
16
treating the BE test just as a simple empirical
17
test, yes or no. I think there
have to be other
18
underlying reasons for why there is now a large
19
difference in the performance of the dosage form in
20
vivo, things such as dissolution.
I think one
21
should look at other data and have other facts or
22
information supplied by a company saying that it
23 has
attempted the bioinequivalence study that they
24
come up with something that suggests that it is
25
bioinequivalent. There should be
other facts that
202
1
support that conclusion, in particular dissolution
2
methodology. So, you should look
for more
3
information.
4
DR. YU: That is correct. Of course, when
5 we
receive such challenge studies we have to make
6
sure that the study is properly designed and
7
conducted and the conclusion is valid.
Secondly,
8 we
look at the quality of the sample used to
9
conduct the studies. From the
cGMP perspective,
10
from the quality perspective we look at the
11
dissolution of the stability and potency, and so
12 on,
all the quality standard samples.
Certainly,
13 we
also look at the process. As I said, in
14
bioinequivalence we want evidence to show
15
inequivalence and we certainly look at many, many
16
other factors. In other words, we
want to say that
17 the
decision we are making is a systematic decision
18
instead of being based on one parameter.
19
DR. KIBBE: Nozer?
20
DR. SINGPURWALLA: Yes, C,
subscript t,
21 and
C subscript infinity--
22
MR. SCHUIRMANN: You mean AUC
subscript--
23
DR. SINGPURWALLA: Yes, is that
the time
24
index?
25
MR. SCHUIRMANN: It is not an
estimate of
203
1 the
time.
2
DR. SINGPURWALLA: It is some
index?
3
MR. SCHUIRMANN: When you do these
studies
4 you
give the products to subjects and then you
5
start taking blood samples from them at specified
6
sampling times, at however many hours, and one of
7
those has to be the last one.
Maybe it is 24
8
hours. It would depend on the
drug product. So,
9 you
can calculate the area under the blood level
10
time curve up to that last blood sampling time for
11
that subject. You have the data
for each sampling
12
time and the trapezoidal rule is used to calculate
13 the
area. So, that is AUC sub t.
14
Now, there is a way of taking the last
15
several blood concentrations when you are in what
16 is
called the terminal elimination phase, and
17
estimate the elimination rate, and to use that
18
estimated elimination rate to extrapolate that
19
calculated area to theoretical infinite time. That
20 is
the AUC infinity.
21
DR. SINGPURWALLA: I got the
message that
22 AUC
infinity is when t goes to infinity.
23
MR. SCHUIRMANN: Yes.
24
DR. SINGPURWALLA: I am not sure
that this
25
question is germane, but is there a danger or a
204
1
pleasure, depending on which side of the fence you
2
are, that you may make a certain decision for a
3
certain time t and your decision would be reversed
4
about your hypothesis had t been something else?
5
MR. SCHUIRMANN: That is really
not a
6
question that I am qualified to address.
I am sure
7
there could be aspects of the profile, the blood
8
concentration over time profile where the action is
9 in
a certain time interval, and if that happened to
10 be
the last time you sampled--
11
DR. SINGPURWALLA: In other words,
how
12
sensitive is your hypothesis?
13
MR. SCHUIRMANN: I would yield to
the
14 pharmacokineticists
in the room for that question.
15
DR. KIBBE: Les?
16
DR BENET: There is definitely
that
17
possibility. There was a famous
brochure that the
18
Upjohn Company--so that is how long this is--put
19 out comparing two different drugs and they
showed
20
equivalence making that error of picking an early
21
time point so that they actually had very
22
different--if they had gone to infinity they had
23
very different times. So, that is
very critical
24 and
usually what the agency will do or what anyone
25
will do, you want to know that the area under the
205
1
curve up to t is a very high percentage of your
2
total area under the curve infinity or you would
3 not
qualify this as a reasonable study to make a
4
judgment on.
5
DR. SINGPURWALLA: I have a
follow-up, a
6
word of caution, are you familiar with the filer
7
problem?
8
MR. SCHUIRMANN: Yes
9
DR. SINGPURWALLA: Do you think
you would
10 be
a victim of that particular problem here?
11
MR. SCHUIRMANN: There are in
12
bioequivalence assessments but usually not with
13
pharmacokinetic bioequivalence assessments. We
14
sometimes are not doing the analysis on the log
15
transformed endpoints but, instead, there are other
16
types of bioequivalence studies where we are
17
analyzing the untransformed endpoints and we do,
18
indeed, do two one-sided tests based on linear
19
inequalities like mu-T minus 1.25 times mu-R and
20 you
will reject those two one-sided hypotheses if,
21 and
only if the 90 percent filer's confidence
22
interval falls within the interval.
So, we use
23
that method. Which aspect of the
problem are you
24
referring to?
25
DR. SINGPURWALLA: Well, the
filer's
206
1
problem is the following, that when you have two
2
normal distributions with unknown means and when
3 you
take the ratio of their means, then it is
4
possible to get confidence limits which are from
5
minus infinity to plus infinity but with the
6
coverage probability less than 1.
7
MR. SCHUIRMANN: I am aware of
that. If
8
your data is such that that would happen, then you
9
would not reject the two one-sided tests and you
10
would not reach a conclusion of equivalence.
11
DR. SINGPURWALLA: But then we
would have
12
addressed the comment my colleague made that you
13
will have an inconclusive answer.
14
DR. SWADENER: My question really
related
15 to
rejecting the case that it is non-equivalent.
16
That doesn't mean that it is equivalent.
Right?
17
Because there are some outliers; there are places
18
between the two.
19
MR. SCHUIRMANN: There are
experimental
20
outcomes that are inconclusive.
If you reject
21
equivalence, then you conclude inequivalence. If
22 you
reject inequivalence, then you conclude
23
equivalence. But there are data
sets for which you
24
would not reject either.
25
DR. SWADENER: But I thought you
said the
207
1
rationale for trying to define inequivalence,
2
rejecting equivalence doesn't mean inequivalence.
3
MR. SCHUIRMANN: No, I said
failing to
4
conclude equivalence doesn't necessarily mean
5
inequivalence. Perhaps that
sounds like word games
6 to
you but I assure you it isn't.
7
DR. SINGPURWALLA: I think you are
facing
8 a
statistician.
9
DR. SWADENER: No question about
it,
10
right.
11
[Laughter]
12
DR. KIBBE: And a frequentist
statistician
13 at
that.
14
DR. SINGPURWALLA: It pains my
heart!
15
MR. SCHUIRMANN: The take-home message
of
16 my
presentation was it is not reasonable to
17
conclude bioinequivalence if you do a
18
bioequivalence test and don't conclude
19
bioequivalence. You have to aim a
test
20
specifically at seeing whether you can show
21 bioinequivalence.
22
DR. KIBBE: Thank you, Don. Ajaz wants to
23 say
something and I guess, Lawrence, you want to
24 get
back to the questions?
25
DR. YU: Yes.
208
1
DR. KIBBE: Good.
2
DR. HUSSAIN: Well, I think this
3
discussion has been focused primarily on the bio
4
topic but the principles, concepts and issues go
5
beyond that and how does this relate to that? Does
6
somebody have any thoughts on that?
7
DR. SINGPURWALLA: Actually, I do.
8
DR. KIBBE: I knew you would!
9
DR. SINGPURWALLA: Again, a
problem like
10
this is a problem which should be cast in the
11
framework of decision making or, in other words, it
12
should be cast in the framework of a Bayesian
13
setup, and that is the way you address this kind of
14 a
problem where you may have three decisions, three
15
actions--equivalence, inequivalence or
16
inconclusive. That could be a
decision and that
17
provision could be made. Of
course, it could also
18 be
made in the frequentist framework. But I
think
19
this is another example of decision making and it
20
should be cast in the same framework.
21
DR. KIBBE: I think the problem we
are
22
facing here is the difference that we have in a
23
court of law between preponderance of evidence and
24
beyond a reasonable doubt, and we accept drugs as
25
equivalent when we have the preponderance of
209
1
evidence. Do we now ask for
something beyond a
2
reasonable doubt to reject what we have already
3 accepted, and I think that is
interesting. Paul?
4
DR. FACKLER: I just wanted to ask
a
5
question, recognizing that there are thousands of
6
generic products on the market and I understand
7
that there have been challenges to those, do you
8
have any idea how many of those have turned out
9
post-approval to be inequivalent to the innovator?
10
DR. KIBBE: He wants a success
rate for
11
challenges.
12
DR. BUEHLER: All right. Well, I have to
13
think. I know we have had at
least one that I can
14
remember where we had a challenge and when we
15
threatened to downgrade they removed the product
16
from the market. I know that
because that was when
17 I
was in the Office of Generic Drugs. I am
not
18
sure how many more there have been but I do know
19
that there was at least one.
20
DR. YU: I think we just had one
right
21
now. In fact, the study is
under-powered so it has
22
come back--
23
DR. BUEHLER: But that wasn't
removed from
24 the
market.
25
DR. YU: It was not removed.
210
1
DR. FACKLER: Could I ask a
question then,
2 how
important an issue is this?
3
DR. BUEHLER: I think the
importance of it
4
depends on the amount of work it generates to the
5
Office of Generic Drugs with each specific
6
challenge that we get because I have the
7
understanding that the challenge studies are sort
8 of
bioequivalence studies, sort of masquerading as
9
bioequivalence studies but they are really
10
channeled to show bioinequivalence or showing
11
failed bioequivalence. Therefore,
we look at them
12
really with a fine-tooth comb and, as Lawrence
13
said, we look at all aspects of the drug product
14
that was used in the challenge study.
We go out
15 and
actually make site visits to inspect the CRO
16
that conducted the challenge study to make sure
17
that the study was conducted properly.
So, it
18
really involves a significant amount of resource
19
allocation when we get one of these challenge
20
studies because we take them very seriously. If
21
someone challenges the bioequivalence of a product
22
that is currently on the market we, in the Office
23 of
Generic Drugs, take that challenge very
24
seriously and we do put a lot of resources into
25
making sure that it is either valid or invalid.
211
1
Because of that, we would like to have a little bit
2
better framework under which to sort of, like,
3
unleash these dogs. You know, if
we don't have to
4
turn the dogs out we really want to but right now
5 we
are.
6
DR. KIBBE: Do you have a comment?
7
DR. DAVIT: Yes, I would like to
add to
8
what Gary was just saying. I was
directly involved
9 in a challenge several years ago and it was a
10
tremendous amount of work to sort out what was
11
going on. I was a team leader at
the time. I
12
pretty much had my entire team working on it. We
13 had
project managers working on it. We got
the
14
clinical division involved; we had the
15
statisticians involved. We looked
at the
16
dissolution. We looked at the
RLD. We made many
17
visits to the clinical division to discuss what was
18
going on. We sent an inspection
out to the site
19
where the challenge study was conducted.
We had
20
meetings with the generic company.
So, it was
21
very, very involved. And, the
outcome of that
22
particular situation was positive but it took many,
23
many man hours of work from many different people
24 to
sort things out.
25
DR. YU: In other words, the
effort we put
212
1 in
to clarify some of the concept is well worth it.
2
DR. KIBBE: Do you think we should
3
approach it like they do with a challenge flag in a
4
football game? That if they
uphold the challenge
5
they still keep their time outs?
And, if the
6
challenge hasn't been upheld they lose their time
7
outs? So, if a company wants to
challenge they
8
have to put a bond up to pay for the expense of FDA
9
adjudicating the challenge?
10
DR. BUEHLER: That would be okay!
11
DR. YU: Basically, in many cases
if a
12
study comes back it fails to demonstrate
13
bioequivalence instead of bioinequivalence study.
14 As
Don has very clearly pointed out, if you test
15 for
bioequivalence you simply fail to show
16
bioinequivalence. So with a
guidance, if you do
17
want to show that it is bioinequivalence, here you
18
are, this is how to conduct a study so there is no
19
confusion or ambiguity. It is a
very clear
20
definition, clear evidence for agency to take
21
action so we can spend all the time to approve
22
generic applications. We received
over 500 this
23
year.
24
DR. KIBBE: Marc?
25
DR. SWADENER: Have you thought
about
213
1
whether if, in fact, you clearly defined
2
inequivalence it is going to increase your
3
challenges? Will it, in fact,
make your life
4
easier?
5
DR. YU: I think my life would be
a lot
6
easier. There is no doubt about
it; I am very
7
confident.
8
DR. SWADENER: It may double the
number of
9
challenges, or triple.
10
DR. YU: That is certainly a
hypothetical
11
question and I am very confident.
12
DR. KIBBE: Jurgen?
13
DR. VENITZ: I am trying to get
back to
14 the
questions that you want us to answer, Lawrence.
15 I
would say you have demonstrated to me that it is
16
different whether you prove equivalence or you
17
prove inequivalence. In other
words, they are two
18
different objectives, meaning they require two
19
different studies. So, failing to
show
20
bioequivalence is not the same as demonstrating
21
bioinequivalence, which I think is what your first
22
question is all about.
23
DR. YU: Thank you very much--
24
DR. VENITZ: Well, that is my
personal
25
answer; I can't speak for the committee.
The
214
1
second one, as far as the challenge study is
2
concerned, in order to demonstrate bioinequivalence
3
which, as I said, is not the same as failing to
4
show bioequivalence, you have to have an adequate
5 and
well-controlled study to do that, which
6
includes all the characteristics that you are
7
familiar with. From my
perspective, in addition to
8
that you have to have preexisting information
9
suggesting that the drug is bioequivalent because
10
that is what is being challenged in the first
11
place. So, in my mind, the burden
of proof is upon
12 the
challenger t have an adequate and
13
well-controlled study demonstrating beyond any
14
reasonable doubt, to use Dr. Kibbe's terminology,
15
that they are truly bioinequivalent.
So, among the
16
strategies that you are proposing I would use the
17
most conservative one, which I think is number two,
18
meaning that all three endpoints have to
19 demonstrate
bioinequivalence. Only underlying
20
those circumstances would you move to the next step
21
which would be removing, I guess, the generic from
22 the
market.
23
DR. YU: And some others too.
24
DR. VENITZ: I am sorry?
25
DR. YU: Assuming the quality--
215
1
DR. VENITZ: Right, just speaking
about
2 the
testing procedures. I am sure there are
other
3 things
that you look at.
4
DR. YU: Yes.
5
DR. VENITZ: So, I would say
number one is
6 the
difference between showing bioequivalence and
7
showing bioinequivalence. Number
two, a study to
8
demonstrate bioinequivalence has to be adequately
9
well-controlled, or the equivalent thereof. Number
10
three, the burden of proof is on the challenge
11
sponsor to demonstrate that, and I would suggest
12
strategy two as the most conservative one.
13 DR. YU: Thank you.
14
DR. KIBBE: Anybody else? Marvin?
15
DR. MEYER: I agree with part one,
that
16
this is needed. I think, just
from a conceptual
17
point of view, if approval means everything has to
18 be
between 80 and 125, then for inequivalence
19
everything needs to be less than 80 percent or
20
above, as you have drawn it on your little diagram.
21
I don't know that it is fair to require
22 all
three to fail. I think any one should be
23
enough because, after all, it is not fair to expect
24 AUC
to always fail along with Cmax.
Sometimes AUC
25 is
fairly stable and Cmax isn't. So, I
would say
216
1
number one rather than all three.
2
DR. VENITZ: Can I just give you
the
3
reason why I disagree with you on that?
4
DR. KIBBE: Please, go ahead.
5
DR. VENITZ: Because you already
have a
6
study that demonstrated bioequivalence in the first
7
place. Otherwise, I would be in
agreement with
8
you. But it is not like the study
stands on its
9
own. You are basically trying to
meta-analyze two
10
studies.
11
DR. MEYER: But I would argue that you are
12
just setting it up for the inequivalence people to
13
fail by requiring all three.
14
DR. VENITZ: But I think there is
another
15
study demonstrating that there is bioequivalence.
16
DR. KIBBE: And we are waiting for
Les to
17
clarify everything for us--
18
[Laughter]
19
--but I think the first point is true,
20
that we need to have the study design to show
21
bioinequivalence, not just that you do a
22
bioequivalency study and if it fails that doesn't
23
work. That is clear. But the argument over
24
whether you want all three items or not, I think we
25
need to fall back on what is the clinical relevance
217
1 of
the thing failing the Cmax component of the
2
bioinequivalency study to a drug that has a large
3
therapeutic index. I think you
probably need to
4 put
more emphasis in terms of area under the curve
5 if
you are going to pick one instead of three.
So,
6 I
would be inclined to go with my colleague Jurgen
7 and
say let me see all three out of whack and then
8 I
am ready to get the generic company to do
9
additional studies to balance out what we are
10
doing. Les?
11
DR. BENET: I would like to make a
12
comment--
13
DR. KIBBE: Good.
14
DR. BENET: --and it is something
I have
15
worried about for a long time, and that is the
16
stability of the innovator's product from study to
17
study. It sort of gets to Ajaz'
question. I have
18
always been concerned about the innovator or
19
generic, at the end of the shelf life, is the
20
product equivalent to the product when it was first
21
approved? So, I think in this
criteria there needs
22 to
be something that is an evaluation of the data
23 of
the innovator product, that it is, in fact,
24
representative of what the agency knows.
Because I
25
know that there are situations where you could have
218
1 end
of the shelf life drugs that would fail versus
2
when they are first manufactured.
So, I could see
3 how
this could easily be manipulated, if I was a
4
manipulative person which I am not, right--
5
[Laughter]
6
--to make a failed study. I don't
think
7 it
would be that difficult with some drugs.
So, I
8
think there needs to be an additional criteria,
9
again no matter which of these three you pick, that
10 the
agency has confidence that the innovator data
11 is,
in fact, representative in this study.
Maybe
12
that is already true, Gary. I
don't know.
13
DR. BUEHLER: As part of the
review of the
14
study I believe we do look at that parameter.
15
DR. KIBBE: Anybody else?
16
DR. COONEY: Just one point to
come back
17 to,
in trying to resolve the distinction between
18
one, two or three PK parameters to make the
19
decision on, the issue of clinical relevance that
20
several of you have spoken to strikes me as the
21
most important part of that part of the question.
22 So,
my question is it doesn't matter what decision
23 is
made, whether it is one, two or three of these
24
parameters, how do you factor into the analysis
25
that you are doing that you have chosen the
219
1
parameters that, in fact, are clinically relevant
2 for
each individual case?
3
DR. KIBBE: Do you want to give an
answer?
4
DR. YU: Instead of giving an
answer, I
5
guess we have to make some kind of recommendation
6
that, indeed, when we look at those challenge
7
studies the clinical division is heavily involved.
8 We
are working as a team in resolving some of the
9
challenge studies, instead of pharmacologists or
10
chemists acting alone.
11
DR. COONEY: Then the question
becomes how
12 do
you factor that working into the recommendation
13
that is being made so that it isn't just an
14
arbitrary one, two or three or the parameters but
15
that a judgment call is clearly defined in the
16
decision process?
17
DR. YU: That is, indeed, a
challenge. We
18
will certainly look at case by case but we do want
19
some kind of clarification so that people know what
20 is
going on and what to do.
21
DR. KIBBE: Marvin?
22
DR. MEYER: Two comments. One, I know a
23
body in the street is not a good measure but if the
24
generic product has been out there and has sold
25
five million units, it is probably not that bad if
220
1
your adverse reaction reports aren't alarming.
2
Secondly, I think the approach of having
3 the inequivalence confidence limit be
totally to
4 the
left of the right of 80 or 125 is a fairly
5
rigorous kind of assessment because your point
6
estimate then has to be well to the left or right
7 of
the upper limit, in other words, quite a ways
8
away. So, I think one is probably
all you need,
9
Cmax or AUC.
10
DR. KIBBE: And, if you are going
to go
11
with one I would go with AUC.
Gordon?
12
DR. AMIDON: I can readily see how
a
13
contention of bioinequivalence could generate an
14
awful lot of work for the agency, and it could be
15
done almost frivolously.
Therefore, I would be in
16
favor of requiring that it be all three parameters
17 to
be bioinequivalent, plus other supporting data
18
like dissolution data to support that there is
19
something really to go after here and that would
20
merit the action and activity, investigation by the
21
agency. Yes, I am all in favor of
having a bond
22 posted.
If you don't pass, then you lose your
23
money. It is not gambling, is it?
24
[Laughter]
25
DR. KIBBE: That is not
legal. But this
221
1 is,
so it couldn't be gambling.
2
DR. BENET: I support that, Art.
3
DR. KIBBE: Les is going to
comment. Go
4
ahead.
5
DR. BENET: Thank you. I want to support
6
Marvin's position because this is, as is the
7
difficulty of the correction now--I mean we have
8
very good criteria for approving bioequivalence.
9 The
way you have defined bioinequivalence is very
10
difficult criteria that has to be outside the
11
boundary and the confidence interval has to be
12
outside the boundary. For sure,
that is going to
13 be
so hard to do, and if there is one, then it is
14
real and I think that if one of those three
15
parameters is outside I would go for the one. I
16 think
Marvin's argument is a very good argument.
17
DR. MEYER: You agreed with me
before.
18
DR. KIBBE: I want somebody to
make note
19 of
the historical events that Marvin and Les have
20
been agreeing everywhere.
21
[Laughter]
22
If you and I are going to have to back
23
off, then I suggest you look seriously at the area
24
under the curve, more seriously than Cmax. I
25
think, if anything that might actually meet this
222
1
criteria where the other two wouldn't, it would be
2 the
Cmax. It is the most open to pushing one
way
3 or
the other.
4
DR. AMIDON: I think I am still a
little
5
confused, Les and Marvin. You
want to do one
6
parameter. You want to do a test
and if any one
7
parameter falls--what is the correct statistical
8
language?--doesn't show bioequivalence or shows
9
bioinequivalence as opposed to all three must
10
showit--it depends on how you word it, all three
11
must show bioinequivalence, that would be tougher,
12
right? That is what I am saying
and it is what you
13 are
saying. You are saying, Les and Marvin,
that
14 is too
tough. I am not sure. It makes the agency
15
look like they are trying to sweep everything
16
possible under the rug by having such criteria that
17 it
will almost never happen.
18
DR. KIBBE: But the bioequivalence
19 criteria
is that way. It requires, you know, both
20
Cmax and AUC to be--
21
DR. MEYER: But if one fails, it
fails;
22 not
all three. I mean, if Cmax fails it
doesn't
23
matter what the AUC was, you failed.
24
DR. KIBBE: We can go around and
around on
25
this. One of the nice things
about an advisory
223
1
committee is that we give advice and the agency can
2
just ignore us if they want, and they can look at
3
everybody's advice and when the committee is split
4
they can take the input of each member of the
5
committee and weigh one against the other and do a
6
Bayesian analysis of it and pick the right
7
decision. All I am saying is that
if you are going
8 to
accept that the study has shown inequivalence
9
because it has shown inequivalence in one of the
10
three parameters, then I would be careful to make
11
sure it was the area under the curve parameter and
12 not
a Cmax. I would have less confidence in
that
13
personally and I am sure that is biased.
14
DR. SINGPURWALLA: Mr. Chairman,
the point
15 you
raise has to have one thing in mind. Are
these
16
three criteria interdependent? If
they are, it
17
makes a big difference. If they
are not, it makes
18
another difference. I suspect
they are
19
interdependent and that is what you should keep in
20
mind. So, rejecting one is as
good as rejecting
21 all
if they are interdependent. If they are
not,
22
then the kind of things you mentioned do become
23
serious, or the kind of things that Marvin
24
mentioned do become serious. I am
asking the
25
question are they interdependent in your judgment.
224
1
DR. VENITZ: I think they are
2
interdependent and I think the differences between
3 the
two strategies are marginal. In other
words,
4 if
you reject AUC infinity you are likely to reject
5
AUC-t as well. There is a little
less
6
interdependence between the Cmax and the area
7
estimates. So, you are really
splitting the
8
difference that is very small.
9
DR. SINGPURWALLA: Did I agree with you?
10
DR. KIBBE: I don't know. I need a
11
decision tree to find out whether we agree or not.
12 Has
anybody got anything else? Lawrence, do
you
13
need anything else from us or have we given you
14
enough information to help you go forward?
15
DR. YU: I think so.
16
DR. KIBBE: Then I propose that we
take
17 our
break. We have two more topics to cover
after
18
break. We are breaking right on schedule. We will
19 be
back to do topical bioequivalence at a few
20
minutes before 3:00.
21
[Brier recess]
22
DR. KIBBE: We have a cadre of
taxis
23
waiting at 4:30. We want to be
finished. We want
24 to
have time for topical bioequivalence, such a
25
wonderful topic and Lawrence again is going to
225
1
start off, only he has no slides.
2 Update--Topical Bioequivalence
3
DR. YU: The October, 2003
advisory
4
committee meetings report and, in fact, manuscript
5
have reviews and systematic reviews of the
6
challenges in developing pharmaceutical or
7
bioequivalence criteria for topical products. I
8
think we sent it to you one month ago and this is
9 the
work that was developed in collaboration with
10 Dr.
Jonathan Wilkin. It also further
developed the
11 Q3
concept.
12
So, today we want to share with you and
13
seek your feedback. For example,
are we on the
14
right track? We will publish this
manuscript very
15
soon to initiate a dialogue and then bring back to
16 you
the formal proposal. We will have Dr.
17
Lionberger give you an overview of this paper.
18
Rob?
19
Establishing Bioequivalence of Topical
20 Dermatological Products
21
DR. LIONBERGER: Today I am going
to give
22 you
an update on our current efforts to develop
23
methods to demonstrate bioequivalence of topical
24
dermatological products.
25
[Slide]
226
1
The current state of topical
2
bioequivalence is that for almost all products, for
3
almost all locally acting dermatological products
4
clinical trials are necessary to demonstrate
5
bioequivalence. So, I am just
going to give you
6
some quick examples of the kind of clinical trials
7
that are actually needed for this demonstration.
8
These are just recent submissions to the Office of
9
Generic Drugs.
10
[Slide]
11
As you can see here, the number of
12
subjects used in these comparisons--these are all
13 for
topical antifungals, there were three-arm test
14
references placebo studies in patients.
They used
15
700, 400 and 400 subjects. Here
is just the
16
percent cure rate for the test and the reference
17
product. The reference product is
the RLD. Then,
18 the
90 percent confidence interval on the
19
difference between the test and reference cure
20
rate. The goal for this is to be
within minus 20
21 to
plus 20.
22
So, you can see that even with these large
23
numbers of subjects these studies still came close
24 to
failure. So, if you retrospectively
looked at
25 the
power of these studies, you would find that
227
1
these studies probably had at least a 50 percent
2
chance to fail even with that large number of
3
subjects.
4
[Slide]
5
So, there are consequences to having this
6
cost to demonstrating bioequivalence.
It is a
7
barrier to product improvement and also the access
8 of
generic products to the market.
Innovator
9
products need to use bioequivalence studies after a
10
formulation change. These
clinical endpoints have
11 high
variability and so, if you think of what the
12
purpose of bioequivalence is, it is to demonstrate
13
formulation similarity and these are just clinical
14
endpoint and there are just not good methods to do
15
that. Also, these lead to possibly
unnecessary
16
human testing in these studies that have hundreds
17 of
patients to say unapproved products.
18
[Slide]
19
So, based on this, some of the goals that
20 we
have are to identify when clinical studies are
21 not
necessary to demonstrate bioequivalence of
22
topical products and to provide some alternative
23
methods that will still assure product quality.
24
[Slide]
25
In this talk I am going to outline and
228
1
give you an update on a strategy to reach these
2
goals. Our bioequivalence
strategy starts with a
3
mechanistic understanding of the topical drug
4
absorption process. Then we will
identify the key
5
parameters that affect bioavailability.
You heard
6 a
similar approach in Prof. Amidon's talk this
7
morning where he talked about the mechanistic basis
8 for
oral absorption and how that led to a
9
biopharmaceutical classification system, and the
10
possibility for bio waivers based on an
11
understanding of the mechanistic processes
12
involved.
13
So, once these key parameters are
14
identified, then we can choose the in vitro and in
15
vivo tests that best measure and detect differences
16 in
these key parameters. As part of the
selection,
17 we
are going to look at classification of
18
formulation similarity. If two
formulations have
19 exactly
the same components, exactly the same
20
compositions we might focus a different set of
21
tests than if they had different excipients and
22
vastly different formulations.
23
This talk is just giving you the first
24 step
to presenting a decision tree that will allow
25 us
to decide when we might not need clinical
229
1
studies to demonstrate bioequivalence.
This
2
decision tree will be specific for different sites
3 of
action. So first we will look today
primarily
4 at
products that are targeting the very top layer
5 of
the skin, the stratum corneum. Finally,
I will
6
talk about some of the external research projects
7 that we have under way to support
development of
8
this decision tree.
9
[Slide]
10
So, the first thing I am going to talk
11
about is just an overview of the topical drug
12
absorption process. Here I have a
schematic of the
13
skin showing different layers. If
you think about
14
what happens when you apply a topical product,
15
first the vehicle is applied to the skin and then
16 the
drug must dissolve in the vehicle, if it is not
17 already
dissolved, and fused to the surface of the
18
skin.
19
So, the top layer of the skin is the
20
stratum corneum and this is a very dense layer,
21
about ten microns thick, and it is the primary
22
barrier to keep things outside of the body. There
23 are
two paths across the stratum corneum, either
24 the
drug can partition from the vehicle which is
25
placed on the surface of the skin into the stratum
230
1
corneum and diffused through the stratum corneum,
2 or
there is the possibility that drugs applied to
3 the
surface of the skin can travel through the hair
4
follicles and bypass the stratum corneum.
5
If we look at sort of the various areas
6
available for transport by these two mechanisms and
7 we
assume that there is no bias in the drug
8
choosing one path over the other, the flux through
9 the
stratum corneum will be about 30 times more
10
than the transport through the hair follicles if
11
there is no bias between the two pathways, if the
12
drug is equally likely to go into one path or the
13
other.
14
Once the drug gets across the stratum
15
corneum, then the tissue behind that is much less
16
dense. The drugs can fuse much
faster; this is
17
much less of a barrier to the drug finally reaching
18 the
systemic circulation.
19
[Slide]
20
So, as we think about this process we have
21 to
remember that we are looking at bioequivalence
22 and
the goal of bioequivalence is to detect
23
differences in the formulations.
It is not really
24
about how complicated this absorption process is
25 and
how well we can understand that. It is
really
231
1 how
well we can detect differences in the
2
formulations that have already been demonstrated to
3
contain drugs that work in clinical trials.
4
[Slide]
5
Again, as Lawrence has said and we have
6
heard many times today, bioequivalence is defined
7 as
no significant difference in the rate and extent
8 of
absorption at the site of action. So, if
we are
9
looking at products where the site of action is
10
this top layer of the skin, the two sort of rates
11
that can possibly important for determining this
12
are, first the rate at which the drug might leave
13 the
formulation and, second, the rate at which the
14
drug might cross this barrier of the stratum
15
corneum. So, if we understand
those two rates,
16
then we can understand what rate is actually
17
controlling the rate at which the drug actually
18
reaches the site of action. That
is the thing that
19 we
are after in bioequivalence, to demonstrate that
20 the
two formulations will perform the same.
21
[Slide]
22
Usually, in almost all cases, the stratum
23
corneum is the limiting resistance and we
24
characterize this limiting resistance by
25
permeability. The permeability
just includes
232
1
contributes from the diffusion of the drug through
2 the
stratum corneum, the thickness of this layer
3 and
the partition between the vehicle and the
4
stratum corneum. So, we can write
an expression.
5 The
J is the total flux. That is the sort of
rate
6 at
which drug is reaching the body and that is what
7 we
are interested in when we are making our
8
comparison of bioequivalence.
This is related to
9 the
permeability times the area that is available
10
times the concentration of the drug that is present
11 in
the vehicle. We can sort of do a little
bit of
12
manipulation with this partition coefficient here,
13
where S is just the solubility of the drug either
14 in
the membrane stratum corneum or the vehicle.
15
So, this is sort of split up
into
16
contributions that are just properties of the skin
17 and
just properties of the formulation. From
this,
18 you
can see it is the thermodynamic activity, the
19
ratio of the concentration to the solubility in the
20
vehicle that is the driving force for what the flux
21
is. So, if the membranes were the
same between two
22
products and presumably if they were applied to the
23
same person it is the same skin and you would think
24 that
these two products would be the same, and it
25 is
just essentially this activity in the
233
1
formulation that would determine how fast the drug
2
arrives at the site of action.
3
But the most sort of important
4
complication here and the thing that we are sort of
5
worried about when we are looking at what methods
6 are
best to develop, bioequivalence methods, is
7
that properties of the formulation can alter the
8
barrier properties of the skin.
So, if by applying
9 the
formulation, either the formulation itself or
10 the
excipients in it, if they can alter the
11
properties of the skin they will change this flux
12 independent of what is happening in the
13
formulation. There is a whole
technology and
14
design in topical formulations to, say, improve
15
bioavailability where there are lots of adjuvants
16
that are known to reduce the barrier and increase
17 the
flux. This is not just hypothetical
18
possibility but a known situation that can happen.
19
[Slide]
20
Once we recognize tat this is sort of the
21 key
mechanism. Then we can sort of identify
what
22 are
possible causes of bioequivalence for products
23
that have the same drug content.
So at different
24
stages in the absorption process we can identify
25
things that possibly can go on.
234
1
First of all, at the application stage if
2 the
two products spread differently on the
3
skin--say, the viscosity or the rheology is
4
different, you could have different outcomes in
5
terms of how they contact the skin, the amount of
6
area each product has if they are applied
7
similarly. If we look into the
formulation we can
8
imagine a case where, well, what if a drug doesn't
9
leave the formulation at all?
Say, the drug is
10
present in the formulation as suspended particles
11 and
these particles just don't dissolve, the drug
12
never leaves the formulation so, even though you
13
have the same amount of drug in the formulation but
14 it
doesn't get out of the formulation, the two
15
products might not be equivalent.
16
Again, the thermodynamic activity in the
17
vehicle might be different. In
one case the drug
18
might be dissolved into a cream and partitioned
19
between the oil and water phases and you have one
20
concentration of drug, one free concentration of
21
drug in the vehicle. If you had a
suspension where
22 the
particles were dissolving the dissolution rate
23
might control what the free drug concentration is.
24
And, this could happen if you had the same overall
25
drug content.
235
1
Finally, when you reach stratum corneum,
2
again as I said, formulations might have different
3
effects on the stratum corneum or you might have
4 one
formulation preferring the follicular pathway.
5
This is particularly known to happen when you have
6
particles of certain sizes that might bias toward
7
this particular transport pathway.
So, that is
8
primarily a concern when you have the drug present
9 in
the formulation as a suspension.
10
So, if you we think about the mechanism
11 and
possible reasons why products might not be
12
equivalent, that leads us to think about how can,
13 or
is it possible that in vivo or in vitro tests of
14 the
formulation can measure these differences
15
adequately enough to replace clinical trials.
16
[Slide]
17
So, I just quickly want to point out two
18
sort of most important in vitro tests that are
19
relevant to these types of products.
The first is
20
diffusion cell. Just a quick
description of what
21
that it is, in a diffusion cell it measures the
22
rate at which the drug leaves the formulation and
23
crosses an artificial in vitro membrane into
24
receptor fluids. So, in most
implementations of
25
diffusion cells the membrane in the diffusion cell
236
1 is
very permeable to the drug so the membrane is
2 not
the limiting resistance. In this case,
this
3
really measures how fast the drug is actually
4
released from the formulation or diffused from the
5
formulation, and also the rate of release and
6
diffusion is also proportional to the fraction of
7 the
free drug. So, it gives you a sense of
whether
8 or
not the drug is actually bound to the
9 formulation
or is free to transport into the skin.
10
Because of this fact that these devices
11 are
usually used with highly permeable membranes
12
they are not very predictive of bioavailability in
13
vivo because in vivo bioavailability is usually
14
controlled by the resistance due to the stratum
15
corneum itself. But these tests
have been shown to
16 be
very sensitive to formulation differences.
17
There is also an important safety role for
18 this
test. If you imagine applying a topical
19
product to damaged skin where the barrier function
20 of
the stratum corneum has been breached for some
21
reason, perhaps by disease, then the drug release
22 to
the patient is going to be determined by how
23
fast it is released in the formulation, which is
24
exactly what is measured in this type of test.
25
The other key in vitro test is a measure
237
1 of
the rheology or how the formulation flows.
This
2
would determine how vehicle spreads on the skin.
3
This type of characterization is also important to
4
classifying the proper dosage form for the
5
formulation. At the last advisory
committee
6
meeting you heard about a decision tree to classify
7
different topical semi-solid dosage forms, and part
8 of
that decision tree involved evaluating rheology
9 or
how easy it was to make a formulation flow.
So,
10 that is part of the testing that is already
11
involved in these products.
12
[Slide]
13
If you have a drug present in a suspension
14
form you have additional tests that might be very
15
relevant to apply. It might be
direct measurements
16 of
particle size in the formulation or measurements
17 of
the dissolution rate in the vehicle as well.
18
[Slide]
19
There are also in vivo tests that can be
20
used to characterize topical formulations. The two
21
most important ones in this case are a skin
22
stripping method where you apply the formulation to
23 the
skin, after a certain amount of time remove it,
24
then remove the layers of skin and assay them for
25 the
actual drug content in the skin layers, or
238
1
microdialysis techniques where you insert a
2
capillary under the skin and you measure the
3
concentration that passes through the skin into the
4
lower layers of the dermis.
5
There have been experimental reports in
6 the
literature on how they are used. But in
this
7
context, please remember that the important role of
8 in
vivo tests is to quantify the effect of the
9
formulation on the skin itself.
If we didn't
10
believe that there is any possibility that the
11
formulation would change the barrier properties of
12 the
skin we would be much more confident that just
13
assays of the in vitro performance would be
14
sufficient to determine whether or not two products
15
were bioequivalent. But since we
have reason to
16
believe that formulations can affect the skin
17
properties, then we would like to at least have our
18
battery of tests in some way to measure this
19
effect. So, the role of these in
vivo tests is
20
sort of very specific.
21
They tell you a lot more information than
22
this. They tell you about the amount
of
23
experience, concentration, presence of different
24
aspects of the skin as well. We
are specifically
25
here looking formulation effects since we are
239
1
looking to determine bioequivalence.
2
[Slide]
3
Now that we have sort of identified the
4
whole list of tests, the question is how do you
5
decide which tests should be relevant to which
6
types of products. So, again,
here we are going to
7 be
talking specifically about using formulation
8
similarity as part of that classification. So,
9
here we define Q1 similarity as products that have
10 the
same components. Q2 similar products
have the
11
same components but also present at exactly the
12
same amounts. So, Q3 means we
have the same
13
component and the same amount, but they also have
14 the
same arrangement of matter or microstructure of
15 the
material so that they are sort of identical not
16
just in composition but also in the arrangement of
17 the
material. So, based on classification of
the
18
formulation difference between test and reference,
19 we
want to choose the appropriate in vivo or in
20 vitro test.
21
So, in all the following discussions,
22
since we are talking about bioequivalence we are
23
really talking in the beginning, before we even
24
talk about bioequivalence, about products that are
25
pharmaceutically equivalent and that means they
240
1
have the same active ingredient in the same dosage
2
form so we are comparing a cream versus a cream,
3 not
a cream versus an ointment or versus a
4
solution, at the same strength, the same dosage
5
form of the active ingredient and also targeting
6 the
stratum corneum. So, again, all those
things
7 are
sort of prerequisites to determining if the
8
products are bioequivalent.
9
[Slide]
10
So, if we start at the sort of highest
11
degree of similarity, if we know the products are
12 Q3
similar and have the same composition, the same
13
structure, you might regard them as identical and,
14 by
definition, bioequivalent.
15
One example in sort of a regulatory scheme
16
where this comes up is for topical solutions. If
17 it
is a solution it is in thermodynamic
18
equilibrium. If you know that it
is Q1 and Q2, has
19 the
same composition, then because it is in
20
thermodynamic equilibrium you know it has the same
21
arrangement of matter as well.
So, we often give
22 bio
waivers for products that are true solutions.
23
Unfortunately, for formulations
that are
24
more complex than simple solutions it is harder to
25
directly tell that they are exactly identical in
241
1
their formulation, and possibly manufacturing
2
differences might result in products that have the
3
same composition having different arrangements of
4
matter. A simple example of that
might be a case
5
where you have the same composition but in one
6 formulation your particle size is different
from
7 the
other one. So, that is something that is
a
8
non-equilibrium state and usually comes from
9
differences in the manufacturing process of the raw
10
materials. So, those are sort of
the origins of
11
cases where products might have the same
12
composition but have differences in their Q3
13
identity.
14
[Slide]
15
Now if we step down a little bit and look
16 at
products where we just know that they are Q1 and
17 Q2
identical, we want to sort of identify what kind
18 of
differences they could possibly have.
So, here
19 it
is sort of thinking if you deliberately took
20
products with the same composition and you tried to
21
manufacture them in a way where you actually get
22
differences in product formulation, what kind of
23
things would you have to do?
24
So, one of those is that rheology might be
25
different. The flow maybe might
be different. If
242
1 you
take a cream or some sort of emulsion and you
2
changed the particle size of the droplets you might
3
actually change dramatically how the material
4
flows. It might adhere to the
skin differently and
5 you
would end up with different performance even
6
though the products have exactly the same
7
composition. By having some
non-equilibrium
8
formulation in manufacturing, you might be able to
9
change the solubility of the free drug by
10
increasing the sort of surface area of, say, an oil
11
phase. You know that in these
products you have
12 the
same excipients. Presumably they should
have
13
mostly the same effect on changing the barrier
14
products of the skin. But you
might have a case
15
where in one formulation the excipients might be
16
released at a different rate and if you have
17
suspensions, as mentioned before in the particle
18 size
example.
19
If we think about these things, these are
20 all
sort of manufacturing differences and the
21
question we want to ask is are the in vitro tests
22
that we have able to detect these types of
23
manufacturing differences? So,
again, the rheology
24 we
can measure directly. In vitro release
is a
25
very sensitive measure of are things diffusing
243
1
through the formulation at the same rate; will
2
there be any differences in how sort of excipients
3 or
drug reach the skin itself from the formulation.
4
Those two can be directly measured.
5
So, the question that sort of hinges on
6
this is for products where you know that they are
7
pharmaceutically equivalent, you know they have
8
exactly the same composition, in this case are in
9
vitro tests sufficient to ensure bioequivalence?
10
Again, all of these differences, all these possible
11
differences are really due to manufacturing
12
processes. As I said before, in
vitro tests are
13
probably the most sensitive and best evaluation
14
methods for detecting manufacturing differences
15
rather than relying on clinical trials, which are
16
very insensitive to those types of differences.
17
[Slide]
18
If we sort of step down the level one more
19
time and we look at products that are just Q1
20
identical, they just have the same components but
21
maybe in different amounts, in this case we might
22 be
more concerned that the different amounts of,
23
say, excipients in the formulation might have
24
different effects on the skin barrier.
They might
25
change the solubility of the drug in the
244
1
formulation. So, in these cases
we might be more
2
likely to say that in this category you might want
3 to
do some sort of in vitro test to ensure that the
4
change in the formulation does not have a
5
significant effect on the barrier properties.
6
[Slide]
7
Finally, if you go down to products that
8 are
Q1 different, which means they might have a
9
different excipient between test and reference
10
products, again similar discussion to the previous
11
tests for the in vitro tests, but here it seems
12
that you would always want to do some sort of in
13
vitro test to make sure that the new excipients are
14 not
having a different effect on the skin barrier
15
process.
16
[Slide]
17
Just summarizing sort of a little bit of
18 our
current thinking, we go to the beginning of the
19
process of developing this type of decision tree
20 and
looking at classifications based on formulation
21
similarity and the level of in vitro and in vivo
22
testing that you might want to do in those
23
different categories.
24
[Slide]
25
So, as we were sort of developing this, we
245
1
sort of identified key problems that we wanted to
2
look at. So, we have sort of two
ongoing external
3 research
projects, one which with Colorado School
4 of
Mines where we are looking at the in vivo skin
5
stripping method, specifically looking to reduce
6
variability and also accuracy of the method to
7
measure both the diffusion coefficient and the
8
partition into the formulation, so measuring
9
effects of the formulation on the stratum corneum
10 and
its partition in it. The key aspect
there is
11 as
you are doing the skin stripping, measuring the
12
thickness of skin removed via transepidermal water
13
loss.
14
We also have another project going on.
15 So,
we have emphasized sort of in vitro
16
characterization and its ability to detect
17
manufacturing differences. We
have a project with
18 the
University of Kentucky where they are
19
manufacturing different formulations that are Q1
20 and
Q2 identical, so exactly the same composition
21 but
using different manufacturing processes,
22
primarily for cream formulations so oil and water
23
emulsions, and then looking at these known
24
differences and seeing how much difference can we
25
manufacture looking at the ability of the
246
1 rheological
and in vitro release tests to detect
2
these manufacturing differences.
3
With that, I would like to thank you for
4
your attention and answer any questions that you
5
might have.
6
DR. KIBBE: Anybody have any
questions?
7
DR. FACKLER: I have one.
8
DR. KIBBE: Good.
9
DR. FACKLER: Looking at the
decision tree
10 and
then at the examples that you gave at the very
11
beginning, the three examples, to me, showed
12
products that were similarly efficacious and I am
13
wondering if in your decision tree you are
14
suggesting that--I don't know if those products are
15 Q1
and Q2 or Q3--but being that they are similarly
16
efficacious, is it important whether or not the in
17
vitro tests for those products pass?
18
DR. LIONBERGER: Well, I think we
are
19
trying to provide an alternative framework so the
20
idea is that, certainly, you can have products that
21
will give similar efficacy and they won't match at
22 all
the in vitro tests. It is certainly
possible
23 to
come up with products that have different
24
viscosities, different in vitro release rates,
25
especially since that is not a limiting step, and
247
1
still be bioequivalent in a clinical study. So, we
2 are
trying to provide sort of an alternative
3
pathway. It is not that sort of
this decision tree
4
will determine bioequivalence; it is sort of an
5
alternative pathway to doing a clinical study. So,
6 it
is basically up to the sponsor to decide do we
7
want to try to characterize our product very well
8 in
vitro or just do some sort of clinical study,
9 and
they have to balance the costs to those two
10
different pathways.
11
DR. FACKLER: The only reason I
ask is
12
thinking back on the nasal products, there is a
13
requirement for bioequivalence that they pass both
14 the
in vitro studies and the clinical study.
So, I
15 am
wondering if that is the same direction FDA is
16
going in for the topical products.
17
DR. HUSSAIN: I think right now
this is
18
simply our current thinking of moving away from ten
19
years on DPT and so forth, and starting fresh.
20
Again, going to a mechanistic basis, here is
21
another highly variable situation and I think the
22
mechanistic basis decision tree up front as an
23
approach to providing all possible alternatives is
24 the
direction. But at the same time, I think
we
25
need to keep in mind that in many of these cases
248
1 some
of these attributes are critical variables and
2
they will need to be controlled during
3
manufacturing lot-to-lot anyway.
4
DR. KIBBE: Judy?
5
DR. BOEHLERT: Have you also
considered in
6
these studies looking at how creams or ointments,
7 or
whatever, age? Because there can be
differences
8
that develop that are formulation dependent or not.
9 For
example, it comes out a solution; you could get
10
crystal growth if it is not in solution to begin
11
with. So, what seems to be
equivalent to start off
12
with may not be as the product ages.
13
DR. LIONBERGER: That would be
part of
14
sort of the chemistry manufacturing controls to
15
ensure the stability of the product over its shelf
16
life. Is that what you are
talking about?
17
DR. BOEHLERT: Exactly, that is
what I am
18
talking about. Over the shelf
life of a lot of
19
creams you will get crystal growth and the efficacy
20 of
that cream will change because the crystals
21
start to grow and they don't have the same
22
transport property that they did.
23
DR. LIONBERGER: You would want to
have in
24
vitro tests for stability to evaluate those
25 differences,
if they occurred.
249
1
DR. KIBBE: Gordon?
2
DR. AMIDON: Yes, Bob, I would
like to
3
commend you. I think you have
really brought a
4 good
focus to how to apply and rationally go about
5 in
vitro testing for topicals, which are more
6
complicated than oral, as you have described. That
7 is
why I have stayed away from it. The
dilution
8
that you get in the stomach is an enormous
9
advantage to regulating oral products, but I think
10 the
enumeration of the factors you are really very
11
much on track with, simplifying or quantitating the
12
differences.
13
I like the idea of starting out by looking
14 at
formulations that have qualitative similar
15
components because they are maybe going to have
16
similar effects on the permeability; similar
17
effects on the thermodynamic activity; similar
18
evaporation rates of spreading rates--start with
19
something that is manageable and then go off into
20
different excipients where it is more complicated
21 and
determine how you might characterize that.
I
22
think it is a very difficult process and you are
23 not
going to be able to simplify everything but you
24 can
simplify some things and at least characterize
25
where we feel confident about the in vitro test and
250
1
where in vivo testing is needed.
So, I think it is
2
really an excellent start.
3
DR. KIBBE: Ajaz?
4
DR. HUSSAIN: I think this is more
focused
5 on
understanding the mechanisms first and then
6
deciding what is critical and what is not critical,
7 and
how it relates to performance. I totally
agree
8
with you, here is a much more complex system from a
9
physical-chemical perspective compared to the
10
tablets and how that happens, and here is a highly
11
variable drug situation also. So,
this example
12
relates totally to the previous disease that we had
13 on
highly variable drug products.
14
DR. KIBBE: Anybody? No?
Good.
15
DR. SELASSIE: In terms of your Q1
16
differences, have you looked at the role of
17
hydrophilicity, especially in terms of the
18
different excipients and what effect they have on
19
follicular transport versus stratum corneum?
20
DR. LIONBERGER: Yes. Certainly the
21
partition between sort of the effect of the
22
formulation on different transport paths would be
23
determined by the partition between the two phases.
24 So,
I don't think that just sort of changing the
25
excipients will have a big effect on partition
251
1
between the two things since they are both
2
partitioning from the same vehicle phase into
3
either the stratum corneum of the sebaceous fluid.
4 So,
it is not going to be sort of different unless
5 you
have some sort of mechanism by which it can be
6
biased toward the follicle, things like
7
micro-motions or things like that.
8
DR. KOCH: I just had a question
and it is
9
related but not necessarily. We
heard that using
10
this topical evaluation is perhaps more complicated
11
than the dissolution one would have in the stomach.
12 But
what about another form, a suppository?
Are
13 there methods in place--and obviously it is
not
14
exactly what you would call a topical, but are
15
there similar equivalence studies in place that
16
either can be drawn from, particularly as you go
17
into some of the European dosage forms, to validate
18 or
add to this particular study?
19
DR. HUSSAIN: I think the key is
that we
20
often struggle when the site of action is local.
21
Now, rectal suppositories often are for systemic
22
absorption. If they are for
systemic absorption,
23
then our current system handles it fairly nicely.
24 But
if they are for local effects, and anything
25
that we have to deal with for localized effects, we
252
1
have challenges, inhalation, topical and so forth,
2
where the site of action is the tissue adjacent to
3
where the delivery is. So, those
are sort of the
4
common challenges we face.
5
DR. KIBBE: Pat, do you have
something?
6
DR. DELUCA: I just wonder if this
is
7
going to extend to the transdermal delivery
8
devices, the patches, and all?
9
DR. LIONBERGER: This is mainly
for
10
products that are locally acting so if you can
11
measure concentration in the blood and sort of
12
reduce the standard pharmacokinetic measurements to
13 do
bioequivalence.
14
DR. KIBBE: We are clearly talking
about
15
drugs that act in the stratum corneum.
But the
16
direction that drugs move from the applied product
17 is
into the stratum corneum and then out.
So, now
18
that begs the question where they go after that,
19 and
can we measure it there as a surrogate for it
20
being in the stratum corneum. I
will argue that
21 our
ability to measure trace amount of things has
22
gotten better. I remember the
reason we actually
23
even started doing pharmacokinetics is because the
24
Bratton Marshall was invented and we actually could
25
measure sulfa drugs and therapeutic concentrations
253
1 for
the first time. So, has anyone thought
about
2 the
possibility of looking for trace amounts just
3 to
show that it has crossed and penetrated into the
4
capillaries?
5
DR. LIONBERGER: Sometimes there
is
6
concern that the site of action really is the
7
stratum corneum. You don't know
how much is
8
accumulating there versus other parts of the skin.
9
DR. HUSSAIN: I think the
discussions have
10
always been in terms of two aspects, safety and
11
efficacy aspects. Now, if the
site of action is
12 the
stratum corneum or the dermis or the follicles,
13 and
so forth, clearly that is important from an
14
efficacy perspective. But where
it goes next also
15 is
important from a safety perspective and often we
16
will have some coverage of that, and so forth.
17
But I think the challenge we
have had for
18 the
last ten years is that the localized delivery
19 to
site of action is the focal point for discussion
20 and
looking at systemic circulation because, after
21
topical application, you could look at urinary
22
excretion or even blood levels but that is
23
generally considered from a safety perspective, not
24 to
demonstrate bioequivalence because it has
25
crossed over and it is not the site of action.
254
1
DR. KIBBE: But you recognize that
2
bioequivalence has always been aimed at evaluating
3 the
dosage form.
4
DR. HUSSAIN: Yes.
5
DR. KIBBE: So, once it gets into
an
6
individual stratum corneum, no matter how long it
7
takes to get out, that is a direct measure of how
8
well it got out of the dosage form, and if you can
9
find it and quantitate it, it is a measure of what
10
happened before. So, I think as
we get better with
11 LC,
MSMS and we can find them it might even be
12
better for some of these companies rather than
13
doing 728 patients to look at percent cure rate.
14 If
you can find it with trace amounts with a lag
15
time of an hour and a half, and look at it for
16
three or four hours, wouldn't that be acceptable?
17
DR. HUSSAIN: It has not been
acceptable
18 for
the last ten years. That has been the
debate
19
because, if you recall the debates that we have had
20 it
was the localized concentration that the
21
clinicians wanted. I could
actually argue that
22
measuring systemic circulation can actually
23
indirectly give you that assessment, but we haven't
24
been able to convince the rest of the world on that
25
yet, especially the dermatology community. So, I
255
1
think that is a challenge. But
also I was hoping
2
that we can also bring a lot of imaging
3
technologies to bear on this.
4
DR. KOCH: That is exactly the
next point
5 I
was going to make because a lot of the imaging
6
technologies, as they are now being applied for
7
physical measurements--I have seen different things
8
showing up that have to do with--well, just the
9
thing I mentioned yesterday about studying
10
coatings. Using the same
technology we are now
11
able to get below some of those levels down to 100
12
microns or increasing all the time, and the
13
sensitivity is improving. So, at
least from an in
14
vitro method, I think a series of imaging
15
technologies should be able to begin showing some
16
value there.
17
DR. HUSSAIN: We just started a process to
18
look at terahertz microscopy, a spatial aspect of
19
looking at chemical distribution within membranes,
20 and
so forth. The technology is evolving
rather
21
quickly so we may see some solutions out there.
22
DR. KIBBE: Anybody else?
23
DR MEYER: Silence is interpreted
as
24
negative. I think you have
incorporated almost
25
everything we talked about here that we would like
256
1 to
do for many things. We would like to
know more
2
about the mechanism. We would
like simpler, or
3
dissolution tests that meant something, or in vitro
4
tests that mean something. We
would like a
5
decision tree. It seems like
everything everybody
6
mentioned about some of the other problems you are
7
incorporating. You are testing in
vitro, trying to
8
look at manufacturing and effects on topical
9
variability. So, I think you are
covering a lot of
10
basis and doing a good job.
11
DR. KIBBE: Yes, I think you are
right.
12 The
one thing that you need to keep in the back of
13
your mind is that the source of the excipient is
14
going to have a dramatic effect sometimes on their
15
viability and their physical and chemical nature.
16 So,
the company ought to have good characterization
17 for
all their excipients coming in when you get the
18
chemistry data for the Q1 and Q2 evaluations
19
because they don't really characterize the
20
excipients coming in. It is hard
for them to be
21
assured that they have gotten a good, consistent
22
product.
23
DR. COONEY: If I can just add one
more
24 point,
when I think back on studies that I have
25
done where I have made mistakes, the most common
257
1
mistake is not to have looked, really looked at
2
what I am doing. So, using
imaging and microscopy
3 to
visualize what is there should not be
4
overlooked.
5
DR. YU: Since the imaging
technique has
6
been mentioned a couple of times, I just want to
7
update you. In fact, as we are
speaking right now
8 the
studies being conducted, hopefully, will have
9
some results very soon on topical imaging at the
10
University of Kentucky. Thank
you.
11
DR. KIBBE: Is the agency happy
with the
12
discussion? Okay? Well, we have our last
13
presentation and then Dr. Hussain will summarize.
14
DR. HUSSAIN: As Nakissa is coming
over to
15
talk, all the topics that we have discussed are
16
interconnected, and one of the issues that Nakissa
17
wants to bring to your attention is the issue of
18
nanotechnology-based drug delivery systems.
19
Currently, there are a number of issues--confusion
20 to
a large degree with respect to nomenclature,
21
definition and so forth. So, as she
talks about
22
that, I think you will see what we are trying to do
23 to
address some of these.
24 Future Topics--Nanotechnology
25
DR. SADRIEH: Good afternoon.
258
1
[Slide]
2
The last presentation at this advisory
3
committee meeting will be on nanotechnology. This
4 is
an awareness topic so this is going to be a very
5
short presentation.
6 [Slide]
7
Why the interest? Nanotechnology
is a
8
rapidly growing area of science.
You just have to
9
look at the number of publications with the word
10
nanotechnology in the title. With
regards to CDER
11
interests, it is anticipated to lead to the
12
development of novel and sophisticated applications
13 in
drug delivery systems. The private
sector,
14
academic centers and federal agencies are all
15
developing substantial programs in nanotechnology,
16 and
there are significant research dollars being
17
invested in this area.
Approximately 3.7 billion
18
dollars have been invested by the U.S. government
19
projected for the next four years.
So, this is a
20
major area of research.
21
[Slide]
22
This talk will focus on the regulatory
23
considerations of nanotechnology, and specifically
24 as
they apply to CDER products. We have
identified
25
four areas that we would like to talk about. The
259
1
first one is nomenclature, and quality, safety,
2
facility/environmental issues. I
will just go over
3
each one of these things right now briefly.
4
[Slide]
5
For nomenclature the National Science
6
Foundation has a definition for nanotechnology
7
presently, which is anything with a dimension less
8
than 100 nanometers is considered nanotechnology.
9
However, for CDER purposes we need to first define
10
what are some of the nomenclature criteria, and
11
then having defined these criteria we will need to
12
develop a definition that will be appropriate for
13
CDER, and then identify the potential
14
nanotechnology applications to CDER.
15
[Slide]
16
Regarding quality, for products that are
17
going to be called nanotechnology we need to
18
consider these five elements here.
The first one
19 is
characterization of the nanomaterials;
20
description of the critical attributes; assurance
21 of
stability; manufacturing and controls; and then
22
drug release and bioequivalence testing issues.
23
These all have to be identified and described.
24 [Slide]
25
For safety, pharmacology and toxicology
260
1
studies have normally addressed the safety issues.
2
Currently, we believe that the studies that we
3
require for any drug the pharmacology and
4
toxicology are adequate for nanotechnology products
5
also. However, since this is a
new area and there
6
might be some unique areas of concern, we might
7
need to think about possibly new testing models and
8
whether they be in vitro or in vivo.
So, these
9
issues will have to be discussed and this is purely
10
going to be based on scientific issues.
11
For the environmental aspect the things
12
that we have to consider are facility design and
13 the
potential impact of nanotechnology products in
14 the
environment, whether they be from an industrial
15
setting or other.
16
[Slide]
17
The last few slides just identify some of
18 the
challenges that we anticipate having to address
19
regarding nanotechnology. At CDER
we have decided
20 to
meet this challenge by crating a
21
multidisciplinary working group.
This working
22
group will identify the regulatory challenges
23
related to the timely scientific assessment of drug
24 and
drug-device combination products. We
have to
25
consider the drug-device combination products in
261
1
this area because in nanotechnology this might be a
2
very important consideration.
Also, this working
3
group will propose solutions to overcome these
4
challenges.
5
Presently, the members for this group are
6
from the Office of Pharmaceutical Science, Office
7 of
New Drugs, Office of New Drug Chemistry, Office
8 of
Generic Drugs, Over-the-Counter Drugs and Office
9 of
Clinical Pharmacology and Biopharmaceutics.
The
10
co-chairs of this group are in the Office of
11
Pharmaceutical Science. There is
also one member
12
from the Office of the Commissioner because in the
13
Office of the Commissioner there is an interest
14
group for nanotechnology and we would like to
15
maintain a connection between the CDER working
16
group and that Office of the Commissioner interest
17
group so we have that member there to maintain
18
that.
19
[Slide]
20
The goals and objectives basically of this
21
working group are to provide a definition and to
22
craft the terminology; to develop a position paper,
23 a
White Paper, possibly in the future; to identify
24
areas of concern and propose suggestions towards
25 the
development of regulatory guidance documents;
262
1 to
identify training and research needs; and to be
2
involved in the coordination of the above-stated
3
activities and also collaboration for potential
4
research activities in the future.
5
So, having said that, that is the end of
6 the
presentation. I said it was a
"nanotalk."
7
Thank you.
8
DR. KIBBE: It was a
"nanotalk." I like
9 that. Are there any questions or comments? Go
10
ahead.
11
DR. SINGPURWALLA: I am very
curious. I
12
have seen nanotechnology operate at Sandia Labs.
13
What I saw was miniature gears and miniature
14
machines that they were making.
So, as far as
15
manufacturing is concerned or building things is
16
concerned, I saw the relevance of nanotechnology.
17 Can
you tell us how nanotechnology is relevant to
18 the
kind of things that you do?
19
DR. SADRIEH: Are you talking
about
20
devices?
21
DR. SINGPURWALLA: I saw little
gears
22
being made.
23
DR. SADRIEH: That sounds more
like a
24
device. We are going to focus
mostly on drugs.
25 So,
you know, there might be drug-device
263
1
combinations with the gears that you are talking
2
about, but we specifically are focusing on drug
3
issues for CDER.
4
DR. SINGPURWALLA: Right. So, I just need
5 to
get a sense of what you have in mind.
6
DR. SADRIEH: For example, we have
7
nanoparticulate drugs or, you know, platforms.
8
Sometimes somebody designs a platform and it has
9
several different components in it and there might
10 be
an imaging component and a treatment component,
11 a
targeting component, and all of this might be
12
within a size that actually would be within the
13
nano range. So, that is more the
direction that we
14 are
going in.
15
DR. SINGPURWALLA: What advantage
do you
16 see
in it?
17
DR. HUSSAIN: Before I answer that
18
question direction, I think one of the challenges
19 we
face is that we often get calls from higher-ups
20
from everywhere, saying, how many nanotechnology
21
products do you have, and so forth, and it is very
22
difficult to answer that because there are a lot of
23
products which have been in nanoscales for years,
24 and
every solid material that goes into solution
25
goes into a nanoscale. So, from
one aspect, every
264
1
product we have is nanotechnology so the definition
2 out
there is not really applicable. So, we
want to
3
avoid the confusion of what is nanotechnology.
4
The type of products that we have where
5
nanotechnology is being utilized is to reduce
6
particle size to increase bioavailability, and so
7
forth. That is one but that is
simply
8
micromization to a nanoscale, right?
But other
9
than that, I think you are looking at design of
10
drug delivery systems. These
could be nanosomes.
11
These could be other ones which are more target
12
oriented where you want to distribute the drug
13
differently, and so forth. So,
these are mostly
14
drug formulation or drug delivery devices in the
15
nanorange. Then, as Nakissa said,
you will have
16
combinations where, you know, you have a drug
17
delivery device which is a device, a machine with
18
drug loaded on to that. So, there
are many
19
possible combinations. So.
20
DR. KOCH: I was going to add
there
21
because I think this committee or working group
22
that you are talking about needs to just take a
23
step back to put on the list those things which may
24 be
obvious present products that may go all the way
25
from aerosols through a number of micromization
265
1
products but I think that also then takes you into
2
excipients and things that are related.
Then,
3
there are the proactive ones where you would
4
actually be involved with, say, nanotubes, etc.,
5 for
sustained release and things like that.
So, it
6
seems like you first need to begin putting
7
everything on paper that exists and plan as to
8
proceeding or encouraging.
9
DR. SADRIEH: But that is what we
are
10
doing. We are presently preparing
a database of
11
what we have already in-house, what we have already
12
approved.
13
DR. KOCH: So, we will hear that
when you
14 get
to the micron presentation.
15
[Laughter]
16
DR. SADRIEH: Sure.
17
DR. KIBBE: Gordon?
18
DR. AMIDON: What I can see in the
19
research area are things like polymerized mice
20
cells. You know there are new
technology methods
21
being developed and I can see where there are going
22 to
be questions what are the things we should be
23
concerned about--oral, topical ophthalmic, rapid
24
dissolving, and I don't know the answer.
I think
25 it
is being proactive to look at that and, yet we
266
1
have systems to go through the nanoparticle size
2
range today and we are seeing new technologies to
3 do
that and direct use in delivery systems.
4
Do you have any products or any product
5
areas that you are initially looking into? Let's
6 say
nanoparticle polymerized oral delivery system
7 or
something like that, to kind of focus on what
8 issues do we have to address if we are
presented
9
with one of these as an NDA application, or
10
probably earlier during the process of developing a
11
delivery system? Because likely
it would be a new
12
material so then you have the drug master file
13
issues, but maybe not. If it is
not, then I think,
14
yes--if it is a material that is used in humans but
15
processed differently then you have to ask the
16
question what standards are we going to set for
17 that.
18
DR. HUSSAIN: Let me give you an
example.
19 The
challenges are in a sense same material that we
20
have always used now nano-sized, and what issues do
21
they raise? One of the things we
had to look at
22
was, for example, titanium dioxide and zinc oxide
23 in
sunscreen preparations. You bring them
down to
24
nanoscale, you have translucence in sunscreen
25
preparations.
267
1
Traditionally these are USP
materials and
2 USP
does not have physical attributes as
3
specifications so they are USP.
Whether nano or
4
micro it doesn't matter, they are USP.
That raises
5 the
same set of issues in terms of do we have the
6
characterization methods? Are
these stable? Are
7
there photocatalytic issues, and so forth? Also, I
8
think we are sort of working with the NCTR, the
9
National Center for Toxicology Research that has
10
started a program on looking at skin penetration
11 and
photocatalytic activity leading to some
12
toxicity issues. So, we have a
small program
13
looking at all those things.
14
But from a general perspective, what we
15
have seen happening is physics become more
16
important now from a stability perspective.
17
Generally, if anything, we will focus--because we
18
don't do physics well today with current products,
19 we
have to do physics much better in nanotechnology
20 products.
That is an area of gap that we want to
21
fill from a characterization perspective.
22
Also, you will see a lot of issues in the
23
press. There are a lot of
concerns being raised,
24 and
so forth, so we just want to make sure we are
25
rational, science-based with our approaches and
268
1
proactive in our approaches because, otherwise,
2
this area will get stifled and we don't want to do
3
that.
4
DR. SADRIEH: We currently think
that we
5 are
addressing the issues pretty well with our
6
existing system. We just want to
make sure. This
7
working group is going to consider all the issues
8 and
just make sure that we really are; is there
9
anything that we might not have thought about
10
because, as Ajaz said, we get asked a lot of
11
questions. So, I think it is
primarily to just
12
make sure.
13
DR. AMIDON: You are right, it may
bring
14 new
technologies for quality control and stuff, and
15
things that we aren't familiar with within the
16
typical pharmaceutical manufacturing formulation
17
area. Yes, I think this is a good
step to be
18
proactive and think about what we may be faced
19
with. In fact, you will be; it is
a matter of
20
when.
21
DR. COONEY: I would also like to
add my
22
compliments to taking a very proactive view towards
23
this area. I would also suggest
that you look at
24 it
as a continuum of the activities you have in
25
place now because it is a continuum from interest
269
1 in
topical application of drugs. It is a continuum
2
from some of the things that have been looked at in
3 the
drug delivery area. So, it doesn't stand
out
4 by
itself but it connects back to so many
5
activities that are in place.
6
One of the things that I find very
7
positive about this is that by anticipation and by
8
taking this proactive approach you will be able to
9 put
in place the assets, the people, the mind set
10 to
be prepared when things come to you and you are
11 not
going to be trying to catch up. You will
be
12
right on line if not even ahead of the game.
13
DR. SADRIEH: Right.
14
DR. KOCH: If I could add
something to
15
that, I think this would be a good opportunity for
16 the
MOUS or NSF where NSF is looking to take a role
17 in
nano, but to build on what you have already
18
established if you got involved with
19
characterization of tools that would help take the
20
continuum down. I still feel that
there is
21 probably in your particle size distributions
or
22
registrations an area, as we have talked about,
23
that is called below 400 mesh.
That could be a
24
very critical area to what is actually happening in
25
some of the dissolutions and other things. So, it
270
1 is
a continuum again, but just to move into
2
perfecting the characterization tools that will
3
allow you to move forward.
4
DR. KIBBE: Let me just add a
couple of
5
science fiction items. The rate
of technology
6
change is exponential and has been exponential for
7
known recorded history. Right now
we are at a rate
8
which is astronomical. There are
some people who
9
have written, really knowledgeable people in terms
10 of
science who have written about singularity in
11 the
year 2014 and you can't predict what is
12
possible after that because of the rate, and all
13
that. And, it is really good to
see the agency,
14
even if it is gradually getting its feet wet in an
15
area that is potentially spectacular in terms of
16
therapeutics which combine what might be called
17
nano devices with drugs or that kind of thing--so,
18 I
think some of the issues that you will deal with,
19 and
this working group might be the busiest working
20
group in the agency in about five years.
So, it is
21
really good to see that. Anybody
else have any
22
comments? If not, we get to let
Ajaz have the
23
final word; it is kind of the rule around here.
24 Conclusions and Summary
Remarks
25
DR. HUSSAIN: Well, I think I have
271
1
actually thoroughly enjoyed the discussion and I
2
think, especially today, the morning discussion was
3
very useful.
4
But let me go back to day one and try to
5
summarize some of the talks and at least some of my
6
conclusions which I think I was able to reach, and
7 I
want to share that internally as we start
8
tomorrow and we get back to our work.
9
On day one we started with the process
10
analytical technology update. We
provided you a
11
brief summary that covered history, evolution,
12
current status and next steps. I
think the
13
committee was generally satisfied with the progress
14 of
this initiative and essentially agreed with the
15
direction in which it is going.
16
I think the suggestions we received from
17 you
for this topic were that we need to consider
18
more objective metrics, especially for a training
19
program, to see how effective they are.
Look
20
towards international harmonization is another
21
message that we heard and we are pursuing that and
22
will continue to do that. Also, I
think Dr. DeLuca
23
pointed out the need to encourage publications and
24
research in this area, and I think this links to
25
nanotechnology. Everything is
connected in some
272
1 way
or form and we will try to do that as much as
2
possible.
3
The afternoon discussion was PAT provider
4
tech. As I sort of summarize the
talks here, the
5
Office of Biotechnology Products is a new office in
6 the
Office of Pharmaceutical Science. They
were
7 not
part of the initial team building and the
8
training and certification program that we had for
9 our
CDER staff members. Since the guidance
is a
10
framework guidance, the framework is applicable to
11 any
manufacturing. The reason the Office of
12
Biotechnology Products was not included within the
13
scope of the guidance was because they were not
14
part of the training.
15
So, the afternoon discussion was to give
16 our
Office of Biotechnology Products and CBER
17
colleagues an opportunity to discuss with you
18
challenges of the complexity they are facing in
19
their area, and how PAT might be applicable to
20
biotechnology products. I think
we discussed a
21
number of emergent technologies and then potential
22
applications, not only by the members here but also
23 in
open session.
24
I think the question focus primarily in
25
particular was on how should the training program
273
1 be
structured as we go to the next training
2
program. The general discussion
and what we heard
3
from the committee was that training needs to be
4
emphasizing more critical thinking problem solving.
5 We
did not really get a sense that it has to be a
6
technology focus, and so forth, because we cannot
7 do
that. If we focus on general principles,
if we
8
focus on the concepts and approach that technology
9
will evolve and we can always gather that
10
information rather quickly.
11
Based on that sort of discussion--I had a
12
chance to talk to Helen also, I think we have an
13
opportunity to think a bit differently than we
14
were. What I am proposing now is
that as we move
15
forward, since we already have a mature PAT process
16
within the Office of Pharmaceutical Science and
17
since we never excluded biotechnology products from
18 the
PAT guidance because our Office of New Drug
19
Chemistry probably has more biotechnology products
20
than the Office of Biotechnology Products right
21
now, so I don't see a need to exclude our Office of
22
Biotechnology Products from the scope of the
23
guidance that we finalize.
24
The key issue there is that of training
25 and
certification. Because of the
infrastructure
274
1
already in place with our OPS PAT team and others,
2
through consultation, and so forth, we can actually
3 build
a bridge to that and get the second training
4
program started but not have to exclude our Office
5 of
Biotechnology Products from the guidance.
So,
6
that is the thought process that sort of evolved,
7 and
I think Helen an I thought this might be a
8
better approach as we finalize to include them.
9 So,
the guidance will only exclude CBER products
10
because CBER was not on board from that
11
perspective. So, that is how we
think we will
12
proceed with that. So, I think
the discussion was
13
very useful to make that sort of a decision and I
14
hope you agree with that. If you
don't, obviously
15 you
will tell us before we leave.
16
I think discussions today were very
17
valuable and I am really pleased with how we sort
18 of
came up with a decision with respect to highly
19
variable drug products, at least a sharpened
20
decision. But I do want to sort
of emphasize a
21
couple of things. In a sense the
discussion was on
22
highly variable drug products because
23
bioequivalence deals with formulation of products;
24 it
doesn't deal with the drug. If you
inject a
25
drug, a very simple solution of drug into a human
275
1
being and you see a lot of variability in the PK
2
parameters, that is a highly variable drug with
3
respect to the disposition characteristics--you
4
know, metabolism, excretion, elimination and so
5
forth. Now, if you give the same
solution orally,
6
then you add on the variability, the physiologic
7
variability of gastric emptying, and so forth. So,
8
that is a highly variable drug by itself. For the
9
sake of assumptions, it is a simple solution; the
10
formulation is not an effect.
11
But then you put that drug in a solid
12
dosage form, or any other dosage form, or a topical
13
dosage form and you have a set of variabilities
14
there. If the variability is the
same as what we
15 had
after intravenous administration the
16
formulation really did not add or subtract from
17
that variability. So, it is a
highly variable drug
18 and
the product did not alter that variability.
19
But you can also have scenarios where the
20
product that you design can increase or actually
21
reduce that variability. For
example, I think we
22
have seen more recently some drugs, especially
23
Class II drugs, which have significant food effect
24
when you administer them in a conventional dosage
25
form. If you can design a
formulation, for example
276
1 a
nanoparticle formulation or a cyclodextrin-based
2
formulation you can actually eliminate the food
3
effect so you have reduced the variability. So,
4
here is a formulation design strategy that can
5
actually reduce the variability.
6
So, I think the highly variable discussion
7
really is a focus of discussion of highly variable
8
drug products. The variability is
no different
9
from the variability of the innovator.
That is not
10 an
issue. When the variability is higher
then that
11
becomes a decision issue, whether it is acceptable
12 or
not.
13
But for the last ten years or so that we
14
have discussed that, all the discussion has focused
15 on
the statistical criteria and actually trying to
16 clear the check box exercise. The simple answer, I
17
think it is simply an arbitrary number.
I hate
18
those check box exercises and it is easy, we can
19
make a decision. So, I am not
comfortable with
20
sort of arbitrary numbers defining that.
So, I
21
think that is the gap that will remain.
22
But the scaling approach, if we address
23 the
arbitrariness of that and make it more
24
comparative scaling to a reference variability is a
25 way
forward, and I think that was the general
277
1
conclusion of this committee and I think you gave
2 us
the signal to move forward. I will ask
Lawrence
3
next month to have that ready for you.
4
[Laughter]
5
I think that was a very useful discussion
6 and
I think we will move forward very quickly to
7
sort of hone in on that. At the
same time, I think
8 the
decision tree approach is built in there.
It
9 is
a logical decision tree that will evolve and I
10
think we will move there and I can be assured how
11 it
can be done with the topical discussion that
12
followed, and that is a highly, highly variable
13
scenario right there.
14
So, I think the discussion was very useful
15 and
helped us move forward in terms of being more
16
confident about the direction we want to move
17
forward in. With Lawrence and his
team I am very
18
confident. Probably, if
necessary, we can bring
19 our
proposal to you in October. That might
be an
20
option. I don't want to put
pressure on Lawrence
21 but
I think we can do it.
22
The topic of bioinequivalence I think is
23 to
address currently I think a procedural nightmare
24
that comes from the aspect that our Office of
25
Generic Drugs has to deal with. I
think we want a
278
1
solution to use our resources more effectively and
2 so
forth. So, Lawrence and the group
presented
3
this proposal and I think generally we came to the
4
general understanding that it might be very useful
5 to
move forward.
6
But I do want to sort of remind ourselves
7 of
a couple of things. This is an important
8
concept. It is not a trivial
concept because we
9
have to really think beyond the application that we
10
discussed today and how it applies to the entire
11
regulatory scenario. For example,
out of
12
specification results and how do we deal with those
13 is
a major issue, and how does this relate to that
14
discussion I think is a very serious discussion
15
that probably needs to be considered more
16
carefully. One can think about
misuse of this in
17
some ways. If a product is out of
specification
18 and
the company does a bioequivalence study and
19
fails to establish bioequivalence, and they come
20
back and say there is no clinical relevance so why
21 do
you want to recall the product? So, all
those
22
implications are there, which we did not discuss
23
today. So, it is not a simple
matter and how it
24
relates to the big picture needs to be looked at
25
very carefully.
279
1
The other positive aspect of this is that
2 I
hope it will force people to ask the question
3
why, why is it bioinequivalence?
That gets to a
4
road to a mechanism understanding, and I think
5
without that the numbers game and the check box
6
exercise will continue. As Helen
pointed out in
7 her
opening remarks, we really don't like check box
8
exercises--at least Helen and I don't like them,
9 and
we want to move away from that and be more
10
science based. But the challenge
is when you go
11
towards that without proper training, without a
12
proper quality system for our review staff and
13 review
processes, it has the potential of creating
14
more questions and so forth. So,
we want to manage
15
that very, very carefully.
16
Now, the other two topics that we
17
discussed, I think topical bioequivalence again is
18 a
10-15 year old saga. We have debated and
19
discussed this, and so forth, and the only solution
20
that we could find was to step away from all that
21 we
have done for 15 years and to start fresh.
22
Lawrence and Dr. Lionberger really took the step
23
backwards and said let's rethink this and work with
24 Dr.
Wilkin to rethink the mechanism perspective.
25
Again, the misgiving, if I have any, is in
280
1 the
sense as a professor of pharmaceutics we knew
2
this 15 years ago. There is
nothing new in that.
3 But
it is unfortunate that at FDA we have to now go
4
back to the basics that we have been teaching. So,
5
that is a bit of a frustration but I think we have
6
taken a positive step, in my opinion, in that
7
direction and with the support of our clinicians I
8
think we will move forward very quickly.
9
Now, nanotechnology--I think it is simply
10 a
starting point for discussion and we actually
11
have a number of products which companies want to
12
discuss with us, and PAT is actually very well
13
connected to nanotechnology. If
you read the
14
guidance, there is a sentence in that and many of
15 the
things that we are looking at--particle size
16
reduction, for example, particle size analysis, you
17
cannot just take a sample and send it to the lab
18 and
do this. Most of the particle size
reductions
19 are
based on on-line assessment of particle size.
20
So, every discussion topic was
21
interconnected and I was thinking that, in a sense,
22 I
was going to apologize for quality by design of
23 our
advisory committee agenda because I think the
24
topics were a bit lighter on day one; we had more
25
time left, and a bit heavy on day two.
But the
281
1
sequence that we had in mind was if you look at the
2
discussion of PAT and biotech, and if you look at
3 the
discussion of highly variable drug products,
4
there was a quality control check right there from
5 our
speakers that we had invited. Everything
was
6
connected. The sequence was there
but I think the
7
material should have been more in depth on day one.
8 So,
we will work on quality by design for our
9
agenda more. With that, Helen, do
you want to say
10
something?
11
MS. WINKLE: I just want to say
that I
12
agree with Ajaz. I thought the
conversation, both
13
yesterday and today, was excellent.
I think that
14
yesterday there was total agreement on the
15
direction we are going with PAT.
I think that the
16
committee has been very supportive for what we have
17
been doing in PAT and I think we have moved ahead,
18 and
I think it is going to be really a very good
19
undertaking for industry, FDA and the public, and I
20
appreciate the committee's support of that
21
initiative.
22
Today's discussion was especially valuable
23 to
us. I think there are a lot of things in
the
24
area of bioequivalence as well as inequivalence
25
that we are still learning and still need to make
282
1
changes. It is constantly
evolving and I think
2
today's conversation will help move us forward in
3 the
direction we need to go to in making some of
4 the
really necessary changes that can reduce the
5
regulatory burden and really get the products out
6 on
the market quicker. So, I appreciate the
7
conversation on that as well.
8
DR. KIBBE: Do I get to say we are
9
adjourned? Good. We are adjourned.
10
[Whereupon, at 4:10 p.m., the proceedings
11
were adjourned.]
12 - - -