The
following is an internal report, which has not been reviewed. A verbatim transcript will be available in approximately
two weeks, sent to the Division and posted on the FDA website at
http://www.fda.gov.dockets/ecomments.
Slides shown at the meeting will be available at the same website.
All external requests for the meeting transcripts should be submitted to the CDER Freedom of Information office.
_________________________________________________________________________________________
The
Clinical Pharmacology Subcommittee of the Advisory Committee for Pharmaceutical
Science, Food and Drug Administration, Center for Drug Evaluation and Research
met on
Clinical Pharmacology Subcommittee of the Advisory Committee for
Pharmaceutical Science Members (voting):
Jürgen Venitz, M.D., Ph.D., David D'Argenio, Ph.D., Marie Davidian, Ph.D., Hartmut Derendorf, Ph.D., David Flockhart, M.D. Ph.D., Marc Swadener, Ed.D., William J. Jusko, Ph.D., Gregory L. Kearns, Pharm.D., Ph.D., Ph.D., Howard L. McLeod, Pharm.D., Mary V. Relling, Pharm.D., Wolfgang Sadee, Dr.rer.nat., Lewis B. Sheiner, M.D.
Advisory Committee for Pharmaceutical Consultants (voting):
Acting Industry Representative (non-voting):
Efraim Shek, Ph.D.
Guest Speakers:
Peter Bonate, Ph.D., Richard Hockett, MD, Pertti Neuvonen M.D.
FDA Guest Speakers:
Hae-Young Ahn, Ph.D., Albert Chen, Ph.D., Joga Gobburu, Ph.D., Peter Hinderling, M.D., Ph.D., Shiew-Mei Huang, Ph.D., Leslie Kenna, Ph.D., Peter Lee, Ph.D., Lawrence Lesko, Ph.D., Stella Machado, Ph.D., Ameeta Parekh, Ph.D., William Rodriguez, M.D.
FDA Participants:
Open Public Hearing Speakers:
No speakers were signed-up to orally present at the Open Public Hearing. Pfizer submitted a written document, for Committee members’ comment, entitled: “Quantitative analysis using exposure-response for End-of-Phase 2A (EOP2A) meeting and use of clinical trial simulation for PK-QT study design.
These summary minutes for the November 17 and 18, 2003 of the Advisory Committee for Pharmaceutical Science of the Food and Drug Administration were approved on ___12/12/03_____________.
I certify that I attended the
________//S//____________________ ________//S//____________________
Hilda Scharen, M.S. Jürgen
Venitz, M.D., Ph.D.
Executive Secretary Chair
The
Subcommittee discussed the following: 1) quantitative analysis using
exposure-response: proposal for End-of-Phase2A (EOP2A) meeting and use of
clinical trial simulation for PK-QT study design; and 2) pediatric decision
tree: examples for applying the pediatric decision tree.
Jürgen Venitz, M.D., Ph.D.
(Committee Chair), called the meeting to order at
Day 1:
Introduction
Proposal for End-of-Phase-2A (EOP2A)
meetings
Issues proposed to be discussed at EOP2A
and their impact Peter
Lee, Ph.D., FDA
Case Studies Ameeta
Parekh, Ph.D., FDA
Hae-Young
Ahn, Ph.D., FDA
Joga
Gobburu, Ph.D., FDA
Break
Committee discussion
Lunch
PK-PD (QT) study design: points-to-consider Peter
Lee, Ph.D., FDA
Use of clinical trial simulation (CTS) for
PK-PD QT studies Peter
Bonate, Ph.D., Ilex Oncology
Case Studies Leslie
Kenna, Ph.D., FDA
Committee discussion
Introduction
Case Studies Peter
Hinderling, M.D., FDA
Albert
Chen, Ph.D., FDA
Methods for determining similarity of
exposure-response Stella
Machado, Ph.D., FDA
between pediatric and adult populations
Break
Research experience in the use of pediatric
decision tree Gregory
Children's
Regulatory experience in using the
pediatric decision tree Bill
Rodriguez, M.D., FDA
Committee Discussion
Concluding Remarks Jürgen
Venitz, M.D., Ph.D.
Questions
to the Committee:
Topic #1: EOP2A Meetings
1. Please comment on the goals of the proposed
EOP2A meeting and the impact that such meetings could have on optimizing dose
selection strategies and efficiency in clinical pharmacology drug
development. What major obstacles would
be expected to stand in the way of achieving the goals of the EOP2A meeting and
how can they be avoided?
The Committee agreed that an EOP2A meeting could be helpful because
brings to attention to exposure-response at EOP2 (sponsor and agency) and may
prevent suboptimal dose finding, which is often the reason for non-approval. The Committee acknowledged that there is a
fair amount of uncertainty in the drug development process and the reasons for
failure of clinical trials are also unknown. The members recognized that
discussing perspectives on dose response and risk benefits earlier on in this
voluntary meeting could be beneficial; however, additional “red-tape” was to be
avoided.
It was discussed that this is a pilot program for EOP2A meetings over a
period of two to three years that will allow for improvements if needed; the
primary outcome of the pilot program is likely to be customer satisfaction
since other development outcomes are subject to other uncontrollable factors.
The Committee concluded that flexible expectations have to be set and a
collaborative meeting with congruence of FDA and the sponsor is crucial to a
successful outcome. The Committee agreed the approach should be quantitative
and mechanistic, where biomarkers and utility functions along with quantitative
integration of in-vitro and preclinical PK/PD data should be discussed, while
considering the end point goal of improving the approval and labeling process.
2. Based on the examples of quantitative
analysis of exposure-response data to assess benefit/risk presented to the
committee, are these the approaches that are best used to optimize dose
selection strategies? What
considerations should be given to the prerequisite studies and data, methods of
analysis, assumptions and certainty of results at this point in time of drug
development, in order to maximize the value of an EOP2A meeting?
The Committee proposed that the concept of a utility function may be
necessary for optimal dose finding. The members defined this meeting would
serve to identify the problem issues and how they could impact the development
process. The Committee emphasized that retrospective data could identify issues
that need to be studied prospectively.
The Committee agreed that pre-clinical and in vitro information need to
be part of the quantitative analysis. The Committee concluded that the goal is
to give guidance to Industry and define what will be at stakes for the
different issues. It was also agreed upon that other items should include the
potential payoff and impact on the guidance process.
3. What benchmark measurements and metrics for
measuring the future impact of the EOP2A meeting should FDA consider?
The Committee suggested that customer
satisfaction may be the only way to measure the outcome of a change in the
development process. The Committee agreed that dose change or dose reduction in
post approval may be a useful metric that can be measured over time.
The members felt that the benchmarks were
defined in the FDA strategic planning steps as reducing the: time, cost, and
uncertainty of developing new drugs. It was agreed that the goal is defined and
the outcome can be measured.
The Committee identified specific scenarios
where EOP2 meeting would be the most helpful.
The members emphasized that for newer drugs, where less prior
information is available, such a meeting may be less
advantageous. The members concluded that such a meeting would be most
beneficial for drugs where a fair amount of knowledge is available ie. preferably drugs used to treat symptomatic versus chronic
conditions, as the payoff may be earlier regardless of the pharmacology of the
drug. The Committee concluded that the EOP2 meeting could also be quite helpful
in bringing forward necessary studies for special populations the sponsor will
need to address.
Topic #2: PK-PD (QT) Study Design
1. What additional study design points would
the committee recommend for consideration in the analysis of PK-QT data?
The Committee agreed that the use of
standard “corrections” for heart rate (Bazet’s or Fridericia’s formula) do not
remove all influence of heart rate and that bivariate analysis of QT
(unoicffected) and HR (or its reciprocal RR) might better separate drug effects
from placebo and HR effects
2. Please comment on the case studies
presented to the committee and the pros and cons of using clinical trial
simulation (CTS) approaches to evaluate PK-PD (QT) study design. Are there other methods of analyzing PK-QT
data that FDA should consider?
The Committee felt that the main challenge
in this approach is determining the maximum and differentiating from random
fluctuations. The Committee agreed that clinical trial simulation is very
difficult and there are limitations to what the QT interval should be, as there
are many variants and genes, there is a need for positive control. The members
discussed that prospective genotyping and preselecting of patients at risk for
TdP may be appropriate. The Committee agreed FDA is on the right track, as FDA
can do these simulations based on available in-house real-life data; models for
drug effects should not be limited to only concentration. The Committee
recommended reviewing and modeling heart rate along with QTc and considering
any drug effects on heart rate. The members agreed that a more realistic and
forward thinking approach needs to used, including available in- vitro and preclinical
information.
3. What critical design elements influence the
outcome of a PK-QT study that has as its goal to identify a meaningful change
in QT?
The Committee proposed
that in drug interaction studies, it is crucial to understand the potential effect
both drugs have on QT. The Committee agreed that generalizations can only be
made if drugs are tested independently; as the implications of a pharmcodynamic
interaction may be far greater than a pharmacokinetic interaction. The Committee discussed a meaningful change
in terms of a risk benefit analysis for drugs with real benefits, in extreme
situations. The Committee concluded that clinical significance varies a lot
depending on the benefits of the drug and the parameters of risk versus benefit
and suggested further exploration of the definition of “meaningful QTc effect”.
Topic #3: Pediatric Bridging: Pediatric Decision Tree
1. Please provide feedback on the pros and
cons of the current pediatric decision tree and the changes that have been
proposed in light of the examples that have been presented?
The Committee agreed with the proposed
changes to the current pediatric decision tree. The Committee proposed that an
approach of exposure-response mechanism be used to guide the development of drugs, that are well understood (e.g., have a
well-characterized mechanism of action and/or known concentration-effect data
in adults) or that have a wide therapeutic index.
The Committee discussed that pharmacodynamic
endpoints to be evaluated in the context of a clinical trial must be directly
linked to drug effect through its mechanism of action and also, must be
appropriate (i.e., technically feasible without adding more than minimal risk,
scientifically valid), for assessment in
infants and children. . The members felt that, in the course of
pediatric drug development, testing of certain drugs is quite difficult because
some methods/techniques recommended for use to assess drug effect/efficacy cannot be
validated with clinical data in pediatric subjects. . In addition, the
committee discussed that in the context of the trials, one is sometimes forced
to examine endpoints
that may be unrelated to the effect of the drug. The Committee agreed that
pediatric investigation should be driven by the known clinical pharmacology of
the drug (i.e., mechanism of action, concentration-effect determined in adults)
using the E-R based decision tree.
2. Please comment on the relevant adult data
and information, as well as quantitative methods of analysis that determine the
similarity between E-R in adults and pediatric patients.
The Committee discussed that there is a
tremendous amount of interpretation that needs to go on between the Office of
Clinical Pharmacology and the Review divisions so there can be agreement
earlier on in the process, to ensure that the studies that are truly necessary
in children are completed. The Committee argued that disease presentation and
progression in adults and children can
be very different. They also noted that
much off-label (i.e., contrary to the approved adult indication) drug use in
pediatric patients is not driven by specific disease based indications but
rather, by the expected pharmacologic effect (i.e., mechanism of action) for a
given drug. , However, the Committee
felt that if the decision tree is appropriately revised and correctly
implemented, it could facilitate pediatric drug development by improving its
efficiency, maximizing the amount of information produced from clinical trials
while reducing the overall risk and providing valid scientific and clinical
information that would support pediatric drug labeling as intended by the Best
Pharmaceuticals for Children Act.
Finally, the Committee suggested that in certain instances, the decision
tree could be utilized to incorporate preclinical information (e.g., primate
data, in vitro reaction phenotyping using pediatric livers) to facilitate use
of an E-R approach.
3. How do we know that by adjusting dose and
exposure we achieve efficacy and safety in all populations? Under what circumstances
do they predict deviations will occur?
The Committee argued that adjusting dose and
exposure does not necessarily optimize clinical outcomes. The Committee felt
that extrapolation is predicated by reasonable assumptions derived from a
scientific and clinical perspective and successful methods. The Committee
concluded that the pediatric initiative has made some great advances and is a
work in progress. The Committee argued that when adjusting dose based on
exposure of parent drug, one has to be careful in the case of highly
metabolized drugs, as it is unknown how any active metabolite exposure may
change as a result.
The meeting was adjourned at
approximately
On
Jürgen Venitz, M.D., Ph.D.
(Committee Chair), called the meeting to order at
Day 2:
Introduction
Introduction Shiew-Mei
Huang, Ph.D., FDA
Evaluation of CYP2B6-based interactions David
Flockhart, M.D., Ph.D., FDA
Evaluation of CYP2C8-based interactions Pertti
Neuvonen, M.D.,
Committee Discussion
Introduction
Academic perspectives David
Flockhart, M.D., Ph.D., FDA
Industry perspectives Richard
Hockett, M.D., Eli Lilly
"Practitioner perspectives" Mary
V. Relling, Pharm.D., St.
Jude Children's
Committee Discussion
Committee Discussion and Concluding Remarks
Questions to the Committee:
Topic
#3: Drug Interactions
Please
discuss the implications of drug interactions involving CYP2B6 and CYP2C8, and
what recommendations that FDA should provide to sponsors with regard to in vitro and in vivo drug-drug interaction studies?
The
Committee recommended for CYP2B6 the following substrates in-vitro: efavirenz,
bupropion, and in some cases S-mephenytoin. The Committee suggested the only
selective in vitro inhibitor to be thioTEPA.
However, the Committee felt that that were not any specific inducers,
nor any specific in-vivo probes.
In addition,
the Committee recommended as a CYP2C8 substrate (in-vivo and in-vitro)
repaglinide, because it is the most sensitive marker. The Committee suggested
that gemfibrozil be used in-vivo or in-vitro as a non-selective inhibitor,
because it is most potent; trimethoprime is less potent but more selective.
Further studies are needed to find optimal probe substrates and inhibitors,
particularly for in vivo evaluations.
As
for other in-vitro DDIs, the in-vitro concentrations need to be considered in
the determination of relative contribution of specific metabolic pathways.
The
Committee advised to consider not only two-way but multiple drug interactions
given the increasing polypharmacy in clinical practice. In addition to the mean exposure, the
variability in the in-vivo DDI should be considered in evaluating the clinical
significance. Assessment o population
variability will depend on corresponding population-based clinical
studies. Furthermore, the Committee
recommended that large medication use databases be mined for drug interactions
these databases are becoming more available and reliable, and can be used to
assess clinical significance and estimates of prevalence for drug interactions
Topic
#4: Pharmacogenetics
Are
the approaches presented to study the influence of pharmacogenetics on
exposure-response sufficient and appropriate?
Are there other criteria or approaches that FDA should consider
recommending to sponsors?
The
Committee defined that it is necessary to find a way to formulate individualized
dosages for patients, though there may be different dosage requirement for
different patients. A utility function is necessary to address the issue of
clinical significance (safety, efficacy) of PG information. Population-based studies may be necessary to
determine the prevalence of clinical significant genetic polymorphisms in
practice as part of risk assessment.
The
Committee recognized that in order to use PG optimally, there needs to be a
high level of mechanistic, quantitative understanding of the contribution of PG
to PK and PD. The Committee suggested that product labels should include all
information that is known with respect to the different factors, because the
complexity can help clarify the information that otherwise could be misleading.
The
Committee felt that only the known genetic polymorphisms be included in the
label, and others can be added as they are better understood. Also, the
Committee discussed that is difficult to distill down incomplete pieces of PG
information and use it to educate practitioners how to adjust dosages. However, there was disagreement as to how
detailed PG information should be provided in the drug product label in order
to effectively and efficiently translate PG knowledge into clinical practice.
The Committee suggested that PG testing be recommended/required in areas of the most clinical relevance
where the stakes are high for inadvertent over-/underdosing, and the PG
mechanisms are well understood quantitatively, e.g., genotyping differences can
be measured in terms of exposure response, and the clinical relevance is high.
Finally, there was lively discussion on how
PG information may be similar to or different from other clinical covariates
used in dosage adjustment. It was
recognized the PG information adds to the multidimensional nature of covariate
effects on drug exposure and/or response.
The meeting was adjourned at
approximately