Addressing
Scientific Challenges Associated with Emerging
Regulatory Genetic Toxicology Guidance, Standard
Testing Approaches and Genotoxic Impurities
Invited Speakers and Abstracts
Robert Daniel ("Dan")
Benz, Ph.D.
Food and Drug Administration, CDER
Transforming (Q)SAR Computational Toxicology from a Research Project to an Integral Part of the FDA/CDER Regulatory Process
The Informatics and Computational Safety Analysis Staff (ICSAS) at FDA's Center for Drug Evaluation and Research (CDER) develops quantitative structure-activity relationship (QSAR) prediction models and methodologies, including those useful for predicting the ability of organic chemicals to induce genetic toxicity, carcinogenicity, and other animal and in vitro test effects. ICSAS also has made computational toxicology models to predict the ability of pharmaceuticals to induce adverse human hepatological, renal, cardiological, immunological, and pulmonary clinical effects. Recently, ICSAS has expanded its computational toxicology capabilities to develop approaches for interpreting the combination of QSAR predictions from multiple models and software platforms. This presentation describes ICSAS' efforts to provide validated in silico tools for evaluating the toxicological profiles of chemicals, and introduce, educate, and gain a consensus within and beyond FDA/CDER for how these tools would best fit into the normal process of evaluating the safety of pharmaceuticals. Once a consensus is reached on the best way to use these tools within FDA/CDER, internal MaPPs (SOPs) and external Guidances for industry will be written. When QSAR computational toxicology techniques are accepted as a part of the ordinary procedures for drug safety establishment, they will be performed as a means of reduction, replacement, and refinement for longer, more expensive testing, not as an additional regulatory hurdle. The research and development reported here supports FDA's Critical Path Initiative which encourages the use of computational tools in a regulatory environment to enhance the approval process for FDA-regulated products.
Back to Top ↑
Joel P. Bercu, M.P.H.
Eli Lilly and Company
Mr. Bercu has been with the risk assessment group at Eli Lilly and Company for 8 years. He supports technical guidance for limits of impurities such as genotoxic compounds, residual solvents, reagents, or metals for new drug substances. He also uses risk assessments to support manufacturing issues such as contaminants, cleaning limits, or workplace exposure guidelines. Further, he coordinates structure activity reviews which facilitate genotoxicity testing for potential or measured impurities. In his role, he has been very involved with the risk assessment surrounding genotoxic impurities, including technical support for regulatory questions and guidances, peer-reviewed manuscripts, and presentations at national meetings. Mr. Bercu received his Master's in Public Health specializing in toxicology from the University of Texas in 2001.
Managing Genotoxic Impurities throughout Development of a Pharmaceutical: Control Strategies and Case Studies
Current guidance from the EMEA
CHMP (European Medicines Agency, Committee for
Medicinal Products for Human Use) and draft
guidance from the USFDA (United States Food
and Drug Administration) provide an expectation
of targeted control of genotoxic and/or carcinogenic
impurities. The intent of these guidances is
to ensure drug quality and patient safety. Industry
implementation requires a multidisciplinary
approach between toxicology and chemistry. Control
of impurities is important for active pharmaceutical
ingredients (APIs) used in clinical trials and
for products eventually marketed for use in
patients. Structure-activity analysis using
in silico tools is an effective initial screen
to evaluate synthetic intermediates used to
make the APIs for clinical trials. Greater scrutiny
is given to intermediates that are near the
final steps in the manufacture of an API. For
example, the final intermediates to the API
are automatically tested in an Ames assay because
these intermediates are more likely to be impurities
in the API. When the synthetic route is finalized
for market image material, all the intermediates
are tested in an Ames assay for both product
and workplace safety assessments. A comprehensive
toxicology risk assessment is performed to determine
a safe exposure based on available data such
as structure-activity analysis, laboratory data,
or information in the literature. The limits
derived by toxicology are critical, as they
provide targets for both process chemistry and
analytical method development. The limit for
a genotoxic impurity without carcinogenicity
information is based on the threshold of toxicological
concern (TTC) as a default. However, the TTC
default value is just one consideration, and
other toxicity data should be evaluated, where
available. Control strategies are developed
that are specific to the API, and limits are
dependent on the dose and duration in the clinic
or marketplace. These control strategies will
be demonstrated in descriptions of case studies.
Laura L. Custer, Ph.D.
Bristol-Myers Squibb
Aneugen Detection in the Rat Peripheral Blood Micronucleus Assay
To reduce the numbers of animals used in toxicology
studies and increase productivity, efforts have
focused on integrating genotoxicity endpoints
into rat general toxicology studies. The in vivo rodent micronucleus test has historically
been conducted in bone marrow erythrocytes using
manual microscopic analysis for micronuclei.
A bias against evaluating peripheral blood erythrocytes
exists because the rat spleen removes micronucleated
erythrocytes from circulation, leading to concerns
of reduced test sensitivity. However, accumulating
data suggests that flow cytometric analysis
of large populations of immature rat peripheral
blood erythrocytes can effectively detect genotoxic
compounds. Concerns remain regarding the ability
to detect aneugenic compounds in peripheral
blood and limited data suggesting there may
be strain specific differences in assay sensitivity
remain. The results of experiments designed
to address whether aneugen-induced micronuclei
can be detected in peripheral blood and bone
marrow in Sprague-Dawley and Wistar Han rats
indicate that the rat peripheral blood micronucleus
assay is capable of detecting some aneugens,
but not all aneugens. It is likely that some
aneugens will not be detected in a testing paradigm
consisting of the bacterial reverse mutation
assay and two (2) in vivo mammalian endpoints.
Our results have implications for regulatory
guidances regarding in vivo endpoints in genetic
toxicology assays.
Krista L. Dobo Ph.D.
Pfizer Global Research and Development
Krista is the Senior Director of the Genetic Toxicology Department at Pfizer Global Research and Development in Groton, CT. She holds a B.S in Biology from the Indiana University of Pennsylvania, and received her Ph.D. in Environmental Toxicology from the University of California, Riverside. During her graduate years she applied various in vitro genotoxicity assays to evaluate mechanisms of clastogenicity and mutagenicity. She was a post-doctoral fellow for 1 year in the Genetic Toxicology Department at Bristol-Myers Squibb. She then joined the Genetic Toxicology group at Pfizer in 1997. Currently, in addition to her role leading the Genetic Toxicology group, Krista also provides subject matter expertise to drug development project teams regarding genotoxic impurity related issues.
Application of the Staged TTC to a High Dose Drug Intended for Short Term Use
Starting materials and intermediates used in
the synthesis of pharmaceuticals are by necessity
reactive in nature and may be present as impurities
of active pharmaceutical ingredient (API). In
some cases starting materials and intermediates
may be known or suspect mutagens and/or carcinogens.
Limiting human exposure (and therefore risk)
to genotoxic impurities during drug development
requires analytical and safety diligence beyond
that prescribed in the ICH Q3A guidance. To
address this, more specific guidance on the
limits of genotoxic impurities has been developed
both in Europe and the United States. There
are many shared principles regarding acceptable
limits of genotoxic impurities within the respective
guidance documents. For example, both indicate
the use of the threshold of toxicological concern
(TTC), 1.5 mcg/day, as an allowable limit for
lifetime daily exposure for marketed product,
as well as higher allowable limits (staged TTCs)
for shorter duration treatment during investigational
phases. Another shared principle is that the
limits are general recommendations and certain
circumstance may allow for some flexibility.
Some factors to take under consideration would
include duration of treatment, proposed indication,
as well as the feasibility of controlling impurity
levels. This presentation is intended to highlight
a case study in which a number of such factors
may justify long term application of the staged
TTC limit (i.e. beyond the investigational stage).
Back
to Top ↑
Paurene Duramad, Ph.D., M.P.H.
Genentech, Inc.
Paurene Duramad, PhD, MPH is a Toxicologist at Genentech, Inc. where she provides safety assessment support in the preclinical and clinical development of biologics and small molecule inhibitors for multiple therapeutic indications including rheumatology, asthma, oncology, and cardiovascular disease. Paurene received her MPH and PhD from UC Berkeley and completed a post-doctoral fellowship in Pathology at Harvard Medical School/Brigham and Women's Hospital.
Flow Cytometry, High-throughput Drug Screening, and Biomarkers of Immunotoxicity
Flow cytometry has emerged as a powerful tool
for quantitative, single-cell analysis of both
surface markers and intracellular antigens.
In addition, flow cytometers can now be used
to measure intracellular signaling cascades
and phosphorylation events and are employed
extensively in high-throughput drug screening.
Quantitative flow cytometric analysis is being
applied in many areas of biology, from the study
of immunology in animal models or human patients
to high-content drug screening of pharmacologically
active compounds. Case studies of the development
and application of immunological biomarkers
will be presented.
Back
to Top ↑
Kathleen L. Gabrielson, D.V.M.,
Ph.D., D.A.C.V.P.
Johns Hopkins University School of Medicine
Dr.
Kathleen Gabrielson received her DVM from North
Carolina State University Veterinary College
of Medicine in 1989 and entered a postdoctoral
fellowship in veterinary pathology at Johns
Hopkins University School of Medicine. After
completing a PhD program in toxicology at the
Johns Hopkins University School of Public Health,
she worked as a diagnostic pathologist and became
board certified by the American College of Veterinary
Pathologists.
Currently, she is an Assistant Professor of
Pathology at Johns Hopkins University School
of Medicine in the Department of Molecular and
Comparative Pathology and has a joint appointment
in the Department of Environmental Health Services
in the Johns Hopkins Bloomberg School of Public
Health and a faculty member for the Center for
Alternatives to Animal Testing. Dr. Gabrielson
is pursuing research focused on the signal transduction
of cardiovascular toxicities in vitro, in cardiomyocyte
culture and in vivo, using rodent models. She
is studying the role of erbB2 and Hsp90 in the
heart, two proteins that are up-regulated during
doxorubicin chemotherapy and during metal toxicities.
Her background of imaging stems from her research
where she routinely uses ultrasound, optical
and SPECT imaging. Her work is currently funded
by NIH and FAMRI. Her clinical activities at
JHU include Veterinary Anatomical Pathologist
and teaching pathology to DVMs in JHU training
program.
Dr. Gabrielson is on the editorial board of
both Toxicological Pathology and Cardiovascular
Toxicology since 2008. She has a membership
in the American Heart Association/ American,
American College of Veterinary Pathologists,
Society of Toxicology, Cell Stress Society International,
and the Society of Toxicological Pathologists.
She is on the national committees of the American
College of Veterinary Pathologists, subspecialty
Toxicological Pathology, officer, since 2005;
Society of Toxicology, Comparative Veterinary
Sub-Specialty group, Vice President elect (2008)
Vice President-2009, President-2010), councilor,
2006-08; and Continuing Education committee
for the Society of Toxicology from 2009-2012.
She is director of the Johns Hopkins University
Cardiovascular Research Phenotyping Program
Development Committee, and the BRB Imaging Facility
Board or Directors and Johns Hopkins NIEHS Center
for Urban Environmental Health. She directs
and founded the course Toxicological Pathology,
and regularly teaches in Comparative Pathobiology
and Genetically Engineered Mice, Models of Disease,
Laboratory Animal Medicine in Pathology, Pathology
of Cancer, Phenotyping Genetically Engineered
Mice and Directs the course Basic Mechanisms:
Pathobiology program.
Overview- Introduction -Applicability/Feasibility of Rodent Imaging
In the last decade, in vivo imaging methods
have become established tools in basic science
research. This lecture is designed to provide
a general overview of imaging modalities utilized
in basic science with applications to drug discovery,
pathology and toxicological research.
This review will focus on five in vivo imaging
modalities including: MR, PET, SPECT, optical
and ultrasound. Each imaging modality will be
presented with specific examples relevant to
drug discovery and toxicological research.
Several applications of MRI in oncology and
morphological phenotyping during neurotoxicological
studies will be included. Applicable to toxicological
research, PET/CT and SPEC/CT imaging of radioactively
labeled compounds will be reviewed. These studies
provide comprehensive evaluation that enables
the toxicologist to study structure, function,
in tandem with distribution and metabolism of
radioactively labeled drugs. Optical imaging
will be reviewed in the context of imaging gene
expression in transgenic mice engineered with
promoters from metabolism enzymes (e.g., CYP1A2
or CYP3A4) driving luciferase expression. Ultrasound
imaging will be overviewed using applications
of toxicity and drug discovery in cardiovascular,
reproductive and cancer therapy studies. Following
completion of this lecture, attendees should
have basic understanding of the various imaging
modalities and applications to drug discovery
and toxicological research.
Sheila M. Galloway, Ph.D.
Merck Research Laboratories
Dr Galloway received her B.Sc., from the University of Edinburgh, Scotland, followed by a PhD in cytogenetics, at the Medical Research Council, in Edinburgh where she worked on chromosome changes with aging in the human population, and was one of the first to assay sister chromatid exchanges (SCEs), particularly in chromosomal instability syndromes such as ataxia telangiectasia. During a postdoctoral fellowship with Dr Sheldon Wolff at UC San Francisco, she studied mechanisms of SCEs and chromosome aberrations induced by chemicals. From 1979-1983 she was head of the Cytogenetics group at Litton Bionetics, Kensington, MD, responsible for screening of chemicals for chromosome damage for National Toxicology Program and for commercial clients; and carried out human cytogenetic monitoring of people occupationally exposed to ethylene oxide. Since 1983 Dr Galloway has been at Merck Research Laboratories where she is Executive Director, Head of Genetic and Toxicology and Molecular Carcinogenesis. Responsibilities include assessing drug development candidates for potential genotoxicity, both regulatory testing and working with research groups to guide selection of candidate compounds, and supporting drug synthesis with identification and control of potentially genotoxic impurities. She is a former president of the Environmental Mutagen Society, (EMS) and recipient of the Alexander Hollaender Award from EMS. Current collaborations include work on International Working Groups on Genetic Toxicology (IWGT), development of the Comet assay and assessing its suitability for integration into toxicology assays, and work on updating international genotoxicity test guidelines.
Towards Developing an Integrated Approach to Safety Assessment of Potentially Genotoxic Impurities, Degradates and Metabolites of Pharmaceuticals
In pharmaceutical development,
assessment of potential genotoxicity based on
analysis of chemical structure and/or on genotoxicity
testing is done for multiple reasons. Drugs
are tested to support clinical trials in people;
synthetic pathways are reviewed to enable prospective
control of potentially genotoxic impurities,
by assessing starting materials and process
intermediates that might carry over into the
drug. Often, the materials used in production
also require testing for genotoxicity for occupational
health purposes or to meet transport regulations.
It is not unusual that synthetic intermediates,
especially in late steps in the synthetic pathway,
or degradates, are also metabolites. Regulatory
guidelines address genotoxic impurities, but
since the only available guidance on safety
testing of metabolites is quite general, companies
are encountering differing regulatory requests
for testing metabolites for genotoxicity. Issues
include whether structural alerts are part of
the trigger for testing, until recently a lack
of consistent definition of a "disproportionate"
human metabolite, and the concern that if a
human metabolite is synthesized and tested in
a biological assay, it is further metabolized,
confounding interpretation of the relevance
of the results to people. There is mixed regulatory
feedback on whether a potentially genotoxic
impurity that is a metabolite has to be controlled
in the drug to meet the TTC, the threshold of
toxicological concern (or acceptable daily intake)
defined in EMEA guidance on genotoxic impurities.
At a recent DIA meeting there was discussion
among regulatory and industry scientists on
the need to develop an integrated approach to
safety assessment of impurities, degradates
and metabolites with potential genotoxicity.
It was pointed out that the TTC adopted for
potentially genotoxic impurities is extremely
conservative, being based on the more potent
rodent carcinogens and linear extrapolation
of the rodent tumorigenic dose to a level expected
to cause 1 in 100,000 excess cancers. A different
approach is more appropriate for metabolites.
SAR (structure-activity relationships for mutagenicity)
is routinely used to prioritize potential impurities
for genotoxicity testing and control in drugs,
but a more selective approach is needed for
metabolites, and some of the difficulties of
SAR interpretation were highlighted. Philosophies
expressed ranged from the concept that a genotoxic
metabolite that is a low percentage of a drug
(e.g. 1% of a drug given at 1 gram/day) is no
different from a genotoxic drug given at 10
mg/day and therefore the metabolite should be
tested just as thoroughly as a drug, to the
concept that we have accepted for decades that
potentially genotoxic metabolites are adequately
tested in preclinical testing, initially through
use of S-9 in genotoxicity assays and ultimately
in carcinogenicity studies. If metabolites are
considered qualified in S-9, there should be
no need to carry out SAR on those metabolites,
even if they are intermediates and potential
impurities. It would be helpful to define the
information needed at different stages of clinical
development, and to reconcile the needs for
patient safety evaluation with the requirement
to test an intermediate, even if it is a metabolite,
at high, "regulatory" concentrations
to support occupational health in manufacturing,
and transport of materials.
Collaborative Study on 15 Compounds in the in vivo Liver Comet Assay Integrated into 2- and 4-Week Studies
A study was designed to help address several
questions on the use of the in vivo Comet assay
in pharmaceutical development: Is this DNA damage
assay of appropriate sensitivity to detect in vivo genotoxins when tested at the maximum tolerated
dose for 2 to 4 weeks? Is the liver Comet assay
an effective part of a test battery, to complement
the micronucleus test in blood or bone marrow?
Is it prone to give "false positive"
results with compounds that are non-genotoxic
but induce significant cell proliferation or
hypertropy in liver?
Experimental conditions were defined in great
detail, in coordination with the ongoing JaCVAM
trials of an acute Comet assay. All the compounds
were given by oral gavage. Samples for Comet
analysis were taken 3 hr after the final dose.
Practical aspects of integration into a repeat
dose study were also assessed, such as effects
of interim bleeds (as if for toxicokinetics,
hematology and/or serum biochemistry); sacrifice
time (3 or 24 hr after the final treatment);
and effects of freezing cell samples for later
analyses. Each of the 13 participating laboratories
in Europe (6), Japan (3) and the USA (4) provided
Comet and histopathological analyses on liver.
Some also contributed Comet data from other
tissues, or measured micronuclei in bone marrow
and/or blood (manually or by flow cytometry).
For 13 of the 15 compounds, acute Comet data
existed in the literature, or was generated
in this study following 3 administrations of
test compound. The 15 compounds represented:
Liver-specific genotoxins, including mutagenic
(Ames test positive) rodent carcinogens that
were negative in vivo in cytogenetics assays;
2 compounds that induce micronuclei in an acute
protocol but were reported negative in a 4-week
micronucleus assay; 3 compounds considered non-genotoxic
rodent carcinogens; 3 compounds used in the
JaCVAM acute Comet study, and 3 further pharmaceutical-like
compounds. Laboratories evaluated their own
data, and an evaluation committee discussed
data with individual investigators to ensure
a clear justification for the "calls".
Overall the study confirmed the appropriate
sensitivity of the assay: Most of the known
in vivo liver genotoxins were detected in the
2/4 week liver Comet assay. The study also affirmed
the value of the combination of the in vivo
micronucleus and liver Comet assays for detection
of genotoxic carcinogens. The non-genotoxic
carcinogens were negative.
Overall, either an acute or a repeat-administration
protocol for the two-test battery would detect
the genotoxins. In two cases where an acute
protocol gave positive results but the repeat
dose Comet assay was negative, somewhat higher
doses were used acutely, but previous data also
indicated a reduction in exposure with time,
a circumstance that would preclude use of the
repeat-administration protocol under the recommendations
in the proposed revised ICH S2 guideline. In
three cases the repeat-dose protocol was more
sensitive (a lower dose was detectable). The
questions raised related to appropriate criteria
for evaluating results, not only statistical
methods but whether, or how, to use toxicity
data to include or exclude Comet results (and
which data, e.g., hedgehogs, apoptosis, necrosis,
hyperplasia, inflammation, serum biochemistry?).
Published data and the information from the
current study do not indicate that toxicity
is a confounding factor for the Comet assay
in vivo, although it has to be carefully considered
in vitro.
Jim Harvey, Ph.D.
GlaxoSmithKline
Genotoxic Impurities That Are Also Metabolites: What's Next?
The standard in vitro and in vivo genotoxicity
tests outlined in the international ICH S2R
guideline were designed to assess the potential
genotoxicity of a parent drug substance and
its potential metabolites. Regulatory guidance
also exists to ensure the potential genotoxicity
of drug related metabolites in humans is adequately
assessed in these genotoxicity studies, with
additional recommendations for the assessment
of human specific or disproportionate metabolites.
The adequacy of metabolite assessment in the
existing in vitro and in vivo genotoxicity studies
is a subject of much debate (e.g. what level
of metabolite formation in vitro is enough?)
although typically these uncertainties are eventually
addressed upon the conduct of standard lifetime
carcinogenicity studies of the parent drug in
rodents. The recent genotoxic impurities guidelines
on the limits of genotoxic impurities in drug
substance has led to an increased focus on genotoxicity
testing of synthetic intermediates, impurities
and degradants related to parent drug substance.
This impurities guidance has led to further
complications in the generic approaches for
assessing the potential genotoxicity of metabolites
of parent drug substance. Specifically issues
can arise when potentially genotoxic isolated
intermediates and/or impurities are also known
to be part of metabolic pathway of the parent
drug substance in rodents and humans. Is an
assessment of an isolated synthesis intermediate
in the Ames equivalent to an assessment of this
same intermediate when it is formed as a metabolite
of parent under Ames test conditions? What methods
could be used to determine whether the level
of formation of a mutagenic metabolite in humans
is considered acceptable or unacceptable? While
the EMEA guidance recognised this issue "Genotoxic
impurities that are also significant metabolites
may be assessed based on the acceptability of
the metabolites" the definition of "acceptability"
needs further discussion and clarification.
Michael L. Homiski
Pfizer Global Research and Development
Different Approaches to Automation of the in vitro Micronucleus Assay Using Image Analysis
The in vitro micronucleus assay is well established
as an early screen to evaluate the genotoxic
potential of drug candidates by some pharmaceutical
sponsors and is currently under consideration
by regulatory agencies as a suitable alternative
to the in vitro chromosome aberration (CA) assay
and mouse lymphoma assay. At Pfizer we have
over twelve years of experience using microscopic
evaluation of the cytokinesis-block in vitro
micronucleus assay (CBMN) to screen drug candidates
and have recently reported 86.6% concordance
with the CA assay. In addition we have over
four years of experience using an automated
in vitro micronucleus platform, developed in
house to measure micronuclei induction in mononucleated
cells. This prescreen assay is used at very
early stages of drug development and is highly
sensitive (>99% of prescreen positive compounds
are CBMN positive) but due to the inherent limitations
of the platform, specificity is lower with a
potential of up to 30% false negatives. Since
commercially available automated CBMN platforms
were insensitive for detecting weak pharmaceutical
positives, we developed a second system capable
of examining micronuclei in binucleated cells.
This automated platform is state of the art
with a QC interface that is fast, reliable and
used to confirm the classification of micronucleated
cells and we fully anticipate it to out perform
the manual microscopic CBMN assay. Based on
our experience, the advantages and disadvantages
of these in vitro MN screens will be compared
and contrasted within the context of supporting
the early development of drug candidates.
Timothy E. Johnson, Ph.D.
Merck Research Laboratories
Assessing the Potential Risk of a Positive in vivo Micronucleus Result
In the genotoxicity testing of pharmaceuticals, very few compounds in the in vivo micronucleus assay are positive. However, when a positive in vivo result is obtained a risk assessment must be done. Micronuclei can arise through whole chromosome loss (potential for aneuploidy) from disruption of the mitotic apparatus or cell division. Unlike a direct DNA damaging effect where there is no generally accepted safe dose, there is scientific consensus that there is a threshold level for aneugenic compounds. One approach for determining if an in vivo micronucleus positive molecule is inducing chromosome loss is to rule out chromosome breakage in vivo by a metaphase aberration assay in rodent bone marrow. Alternatively, staining for centromeres/kinetochores can be done on micronuclei in vitro or in vivo to identify chromosome loss. However, in cases where a threshold argument is made, it has not been clearly defined what additional data are needed for regulatory submissions and how large a safety (exposure) margin is necessary. Positive in vivo micronucleus results can also result from a disturbance of bone marrow hematopoiesis which only occurs at high doses and therefore would not be seen at clinically relevant exposures. The goal of this session is to share industry experience through case studies of pharmaceutical compounds that induce the potential for aneuploidy or cause micronuclei through nongenotoxic mechanisms and to obtain regulatory guidance in this area.
Back to Top ↑
Peter Kasper, Ph.D.
Federal Institute for Drugs and Medical Devices
(BfArM)
Dr. Kasper has served as the vice president of the German Environmental Mutagen Society and national Councillor of the European Environmental Mutagen Society. He is member of the Steering Committee of the International Workshops on Genotoxicity Testing (IWGT) and of the Genotoxicity and Carcinogenicity Expert Team at the European Centre for the Validation of Alternative Methods (ECVAM) in Ispra, Italy.
Genotoxic Impurities and Risk: Managing the Unknown
Traces of genotoxic impurities in pharmaceuticals
are not so uncommon and their occurrence is
in most cases inherently linked to the manufacturing
process. An EMEA guideline and a recent FDA
draft guidance both suggest consideration of
the daily intake of 1.5 µg of a genotoxic
impurity as a threshold of toxicological concern
(TTC), and to treat this as a risk level that
is thought to pose negligible safety concerns,
i.e. excess cancer risk of <10-5 over a lifetime.
The TTC concept is based on an accumulation
of worst-case assumptions and the resulting
value can be considered as an over conservative
and therefore safe limit. In the absence of
carcinogenicity data it is generally impossible
to determine whether an Ames-positive impurity
might be found to array toward the stronger
or weaker end of the TTC carcinogenic potency
spectrum (or to be carcinogenic at all). Therefore,
control of an Ames-positive impurity to TTC
level will in many cases likely exceed the actual
risk by orders of magnitude. However, the use
of this default limit approach is considered
inevitable when essential cancer risk assessment
data are missing.
Deviation from the standard TTC approach might
be acceptable when more extensive and appropriate
data for a proper assessment of potential risks
are available, e.g., data from carcinogenicity
studies with the impurity or mechanistic data
providing evidence for a threshold (example
EMS). Other conditions under which deviations
from the standard TTC limit might be accepted
or necessary include short-term treatment in
clinical trials, treatment of life-threatening
condition, short life expectancy of patients,
presence of multiple genotoxic impurities or
genotoxic impurities that are also metabolites.
This presentation will provide an overview with
case examples from the more than two years experience
since the guideline on genotoxic impurities
came into force in the European Union. Overall,
the TTC concept has basically been proven as
a suitable and pragmatic approach for regulation
of genotoxic impurities in new drug substances.
Nagu Keshava, Ph.D.
US-EPA
Introduction to Genotoxic Mode of Action and Its Use in Cancer Risk Assessment
Genetic toxicology information has been used in several scenarios. For instance, genotoxicity of a chemical is utilized in industry for screening of pharmaceutical drugs; in academic institutions, the main interest is to understand the mechanism of action of a chemical/physical agent; and in regulatory agencies, genotoxicity information is mostly used in hazard identification of a chemical during analysis of risk to an exposure. It is also used to understand the dose response of a chemical. Genotoxicity information has been traditionally obtained by a battery of assays that predict genotoxic potential of a chemical. However, there are certain challenges during analysis and interpretation of genotoxicity data due to specificity and sensitivity of the assays performed. Recent advances in science and sensitive technology such as toxicogenomics have assisted in understanding and differentiating genotoxicity/mutagenicity data from other types of data. In this symposium, the presentations will highlight the importance of toxicogenomics technology in understanding and differentiating chemicals that have genotoxic mode of action. Furthermore, the presentations will discuss the concerns and considerations of sound science of using in vitro genotoxicity testing to predict carcinogenicity. In addition, the symposium will also have discussion on how this data is being used in cancer risk assessment.
Disclaimer: The views expressed are those of the authors and do not necessarily reflect the views or policies of the U.S. EPA.
Prof. David Kirkland, B.Sc.,
CBiol, Ph.D.
Genetic Toxicology Consultant
David will shortly take over as President of the European Environmental Mutagen Society (EEMS), is Past President of the UK Environmental Mutagen Society (UKEMS), has edited 3 books on mutagenicity test guidelines (on behalf of UKEMS), has published >90 scientific papers, is on the editorial board and is Special Issues editor for Mutation Research, has been a UK expert to OECD, is a member of the UK Government Advisory Committee on Mutagenicity (UK COM) and has organised four major international workshops on harmonisation of genotoxicity test procedures (IWGT). In recognition of these efforts, David was awarded Fellowship of the UKEMS and was recently made an Honorary Professor of the University of Wales Swansea.
David is now recognised globally as an expert in genetic toxicology, and is frequently consulted by companies large and small for his expert interpretation of the genotoxic and carcinogenic risk of new and existing substances.
The OECD In Vitro MN Guideline: Cytotoxicity Issues
In the 10 years since the first draft of OECD
guideline 487 for the in vitro micronucleus
test (MNvit) there have been many issues to
resolve. A retrospective validation of the method
from previously conducted trials (Japanese,
GUM, SFTG) had to be performed. There were some
scientists who only wanted the assay to be performed
in the presence of cytochalasin B, so that a
population of cells that had divided (binucleate
cells) could be identified for scoring. However,
tests in the absence of cytochalasin B were
accepted, but this raised a new (recent) issue
namely what should be the recommended methods
for measuring cytotoxicity. Many EU scientists
(including UK as sponsor of the guideline) wanted
to recommend measures of proliferation such
as relative population doubling (RPD) or relative
increase in cell count (RICC). It was felt that
these were more comparable to measuring replicative
index (RI) or cytokinesis block proliferation
index (CBPI) in the presence of cytochalasin
B. It was also felt that such measures would
be more likely to avoid "misleading"
positive results that could arise by underestimation
of true cytotoxicity by such simple measures
as relative cell count (RCC). A collaborative
trial involving 10 EU and 2 US labs has recently
been completed. Five different cell types have
been used. For the EU labs, 11 known genotoxins
with different modes of action and different
profiles of in vitro clastogenic or aneugenic
activity were agreed. The US labs agreed to
test a further 3 chemicals for which the existing
data gave less confidence that they would reproducibly
induce MN. All participating labs performed
treatments in the absence of cytochalasin B
and compared RCC, RICC and RPD for selection
of target range toxicity (50 + 5% reduction
in the measure), and those labs with experience
of treatments in the presence of cytochalasin
also provided data with RI as the toxicity measure.
The results show that, although some chemicals
produced quite weak responses in some cells,
irrespective of the measure of cytotoxicity,
there were no differences between the various
measures of cytotoxicity in their ability to
identify positive responses with all 14 chemicals.
Thus RPD and RICC should be acceptable as measures
of cytotoxicity for the in vitro micronucleus
test in the absence of cytochalasin B.
Hans-Joerg Martus, Ph.D.
Novartis Institutes for BioMedical Research
Follow-up of in vivo Positive Genotoxicity Results: Novartis Experience
According to ICHS2 guidelines, genotoxicity
testing of drug candidates comprises two in vitro tests (bacterial and mammalian) is complemented
by at least one in vivo test, which is a rodent
micronucleus test, if the in vitro data indicate
absence of a genotoxic potential. However, in
a significant number of cases at least one in vitro test, mostly the mammalian test with a
cytogenetic endpoint, shows positive results
which necessitates a more elaborate follow-up
testing, in general by additional in vivo studies.
At Novartis, we have established to routinely
combine the acute rat bone marrow micronucleus
test with the comet assay in organs considered
relevant, which often is liver, blood or gastrointestinal
tissue. Following this approach it has been
possible to qualify the majority of in vitro
positive results as not of relevance for the
in vivo situation, thus allowing the drug candidate
to proceed into clinical development. Infrequently,
though, positive in vivo micronucleus results
are obtained, and four case studies are discussed.
In the first example, an anti-tumor compound
was found to be negative in the Ames and human
lymphocyte chromosome aberration (HL CA) test.
In the rat bone marrow micronucleus test, a
dose-dependent induction of micronucleated polychromatic
erythrocytes was seen, together with a decrease
in reticulocyte count, indicating bone marrow
toxicity. Data obtained from repeated-dose studies
up to 4 weeks indicated bone marrow toxicity
and abnormal erythropoietic activity, potentially
reactive to inflammatory processes. A subsequent
rat bone marrow chromosome aberration test was
negative, so that it was concluded that the
positive micronucleus test was a consequence
of a disturbed bone marrow physiology without
relevance for a human genotoxic risk. In the
second example, a platelet fibrinogen receptor
inhibitor which inhibits platelet aggregation,
foreseen for oral treatment of thromboses, was
found essentially negative in the Ames test
and HL and V79 CA, as well as mouse lymphoma
tk tests. in vivo, a dose-dependent statistically
significant increase in bone marrow micronuclei
was found in mice when treated by the intravenous
route. In a follow-up study exploring different
treatment and sampling times, the result was
not reproduced, despite bone marrow toxicity
and hematological disturbances indicating anaemic
processes. An oral bone marrow study was negative
as was a rat liver micronucleus test. In summary,
the original MN induction was assigned to bone
marrow disturbances and not considered indicative
of a genotoxic potential. In the third example,
an oral drug that binds iron, negative in the
Ames and HL CA tests, a dose-dependent statistically
significant increase in micronucleus frequencies
was seen at severely myelotoxic doses in the
rat bone marrow. A time-course study confirmed
the result, indicating a delayed response maximum.
A rat liver micronucleus test was negative,
so that the overall conclusion was, again, that
a disturbed bone marrow physiology was considered
to be responsible for the elevated micronucleus
frequencies rather than a genotoxic effect.
In the last example, a spindle fiber poison
intended for tumor treatment was shown positive
in the CA and micronucleus test in vitro and
bone marrow MNT in vivo but negative in the
Ames and comet assay in rat jejunum and liver
cells. In this case the genotoxic activity was
attributed to the pharmacological activity which,
however, is considered relevant for humans,
with the according risk assessment and mitigation
consequences. In summary, in vivo positive results
are rare in drug development at Novartis. Should
they occur, risk characterization needs to involve
the analysis of alternative relevant endpoints
and tissues, and the creation of a mechanistic
hypothesis why an observed effect can be considered
irrelevant for humans which, for the bone marrow
micronucleus test, in many cases is a disturbed
hematopoiesis as the cause for an elevated micronucleus
frequency.
Anu Mudipalli, Ph.D.
US-EPA
In, 2005, Anu Mudipalli joined as a Biologist in US EPA's National Center for Environmental assessment and continues to work in the areas of developing scientific assessments for key chemical contaminants.
Dr. Anu Mudipalli has published in such peer reviewed journals as Mutation research, Molecular Carcinogenesis, International journal of Oncology, Leukemia, Cancer research, in vitro Cell developmental Biology, Toxicology in vitro and Indian Journal of Medical Research and has presented in professional societies such as American Association of Cancer Research and Society of Toxicology.
Her work in EPA's Environmental carcinogenesis Division has resulted in the EPA's Scientific and technological achievement award for publication of her work on Arsenic in 2006 and a bronze medal from the agency for her work in developing the Scientific document on lead.
In Vitro Genotoxicity Tests: Concerns and Considerations for Sound Science
Genetic toxicity testing is routinely performed
to identify potential genotoxicity associated
with exposure to diverse environmental chemicals
and therapeutic agents. During the past three
decades the major focus of genetic toxicity
testing has been hazard identification using
short term tests. The experience gained over
the past 30 years has consistently demonstrated
that no single genetic toxicity testing method
is capable of conclusively detecting all types
of genetic effects. The characterization of
genotoxic potential of a chemical based on in vitro genotoxic test methods is confounded by
the inherent issues associated with the sensitivity,
reliability and predictability of these methods.
For example, when a battery of two or three
in vitro genotoxicity tests was performed, at
least 80% of the 177 non-carcinogenic compounds
tested gave a false positive result in at least
one test. The specificity of the bacterial mutagenicity
tests was 73.9% while the corresponding value
for mammalian cell mutagenicity and clastogenicity
tests was below 45% (Witte et.al, 2005 and Kirkland
et.al, 2007). In another analysis of 600 pharmaceutical
compounds submitted to the German Health Authority
between 1995-2005, 25-35% gave positive results
in one or more of the mammalian cell assays
and yet few were carcinogenic (Kirkland et.al;2005,
2007). Recognizing these limitations in the
current methods to properly detect genetic effects,
guidelines from different regulatory agencies
recommend a battery of in vitro and in vivo
genetic toxicity test methods, although the
guidelines vary from different regulatory agencies.
Pressures for alternative animal testing such
as in the European cosmetic directive will lead
to in vitro studies not being confirmed with
in vivo studies leaving regulatory agencies
to make decisions based on in vitro results.
Although, the in vitro genotoxicity tests have
been established for decades, and protocols
consistently refined, the poor specificity of
the current cell-based assays suggests that
there are intrinsic problems in the protocols.
This presentation will provide an overview on
the concerns and considerations of in vitro
genetic toxicity tests that provide preliminary
and important information for hazard identification
and mode of action analysis of chemicals.
Sashi Nadanaciva, D.Phil.
Pfizer Global Research and Development
Strategies to Reduce NCE Attrition Due to Mitochondrial Toxicity - Novel Screening Methods
Toxicity is a major reason for compound attrition
in preclinical drug development and post-market
drug withdrawals. Recent retrospective analysis
of drugs that were withdrawn from the market
due to safety issues or have Black Box warnings
by the FDA demonstrates that mitochondrial dysfunction
is an important contributor of organ toxicity.
These drugs belong to a wide spectrum of classes
and impair mitochondria in several ways. Mitochondrial
toxicity testing of new chemical entities is,
thus, essential for diminishing compound attrition.
This talk will provide a general overview of
mitochondria, describe the diverse mechanisms
by which drugs can impair mitochondrial function,
and discuss the high throughput screening assays
being used for detecting mitochondrial dysfunction.
These novel assays include measurement of (a)
mitochondrial oxygen consumption by water-soluble
oxygen sensing fluorescent probes, (b) cellular
oxygen consumption and glycolytic rates by solid
state fluorescent probes, (c) activities of
oxidative phosphorylation enzymes by immunocapture
methods and (d) mtDNA-encoded protein levels
by high content imaging.
M. Vijay Reddy, Ph.D.
Merck Research Laboratories
Synthetic Intermediate That is Also a Metabolite, Impurity and Degradate: What Are the Appropriate Risk Evaluation Options?
Case studies will be presented as to how the
genotoxicity risk of a structurally-alerting
synthetic intermediate that is also a metabolite
(detected in microsomes, hepatocytes and/or
animals), an impurity (due to penultimate intermediate),
and a degradate (e.g., chemical hydrolysis in
stomach acidic conditions) could be evaluated.
Traditionally, safety assessment of pharmaceuticals
has assumed that metabolites are tested in assays
of the parent, with the rat S9 metabolic activation
system. With more recent emphasis on early identification
of potentially genotoxic metabolites (and impurities)
it may be appropriate to actually demonstrate
the presence of certain metabolites under the
incubation conditions of the Ames assay; the
question also arises whether it is necessary
to control the same metabolite as an impurity
at the staged TTC (threshold of toxicological
concern) level during early clinical trials.
If the metabolite is not formed in the Ames
assay incubations or if it is human specific
(detected in human hepatocytes only), it may
then warrant adapting the metabolism system
(e.g., human S-9) or testing of the metabolite
itself in the Ames assay. Options could be to
test the metabolite at the maximum amount of
250 µg/plate instead of 5000 µg/plate
used in the standard protocol, or to test in
an in vivo genotoxicity assay selected based
on whether systemic exposure is expected. It
is not practical to identify and carry out structural
alert analysis on all metabolites in early development.
However, some obvious hydrolysis products/metabolites
could be recognized which are not alerted when
embedded in the parent molecule, but would be
alerted if released. Examples of certain aromatic
amines will be discussed, which also illustrate
the strengths and weaknesses of in silico SAR
systems depending on the extent of available
"training set" molecules. Carcinogenicity
data for compounds with structural similarities
to a metabolite could be used to estimate acceptable
exposure levels. As the drug further progresses
into the later phases of clinical trials, it
could be necessary to further address the safety
of the metabolite using a transgenic mouse carcinogenicity
study of the parent drug or even the metabolite,
with the demonstration of metabolite in circulation
and/or tissues.
Maik Schuler, Ph.D.
Pfizer Global Research and Development
Strategies for Better Understanding In Vivo Micronucleus Positives With Kinase Inhibitors
The human 'kinome' space consists of 518 kinases
that could potentially present important pharmaceutical
targets for a wide variety of diseases. Most
kinases inhibitors target the much conserved
ATP binding site and thereby increasing the
likelihood of off-target inhibition of undesired
kinases like those involved in mitotic division.
It is not surprising that most kinase inhibitors
produce positive responses related to the induction
of aneuploidy in the in vitro mammalian genetic
toxicology battery of tests, especially the
in vitro micronucleus assay. The aim of this
presentation is to provide insight into the
aneugenic mechanism of various kinase inhibitors
and to offer up experimental strategies for
the identification of kinase targets responsible
for the aneugenic response in vitro. Using molecular
genetics techniques like siRNA and Analog-Sensitive
Kinase Allele (ASKA) technology, we will demonstrate
that inhibition of Glycogen synthase kinase
-3 (GSK-3), a kinase involved in mitotic division,
might not be responsible for the induction of
aneuploidy. In addition, the use of in vivo
genetic toxicology studies to assess the risk
and relevance to human safety will be discussed.
Kevin S. Sweder, Ph.D.
Bristol-Myers Squibb Company
At Bristol-Myers Squibb, Dr. Sweder pursues investigative research in genetic toxicology. Research interests focus on understanding how organisms maintain genomic integrity and the consequences of DNA damage, especially the genotoxic and non-genotoxic mechanisms of clastogenesis. Dr. Sweder strives to incorporate new technologies and model systems to address genetic toxicological issues that impact regulatory submissions of drug candidates. An important challenge is to develop new assays to improve the ability to predict mutagenicity and carcinogenicity, and to help discern which assays should be added/removed from the regulatory guidelines to assure patient safety.
Kinase Inhibitors and In Vitro Genetic Toxicology Assays
The conjoining of molecular biology and pharmaceutics has opened a plethora of therapeutic targets for drug development. Protein kinases involved in signal transduction pathways provide particularly robust targets. However, given the high degree of homology/similarity among kinases, undesirable off-target effects may negatively impact drug development. We have investigated drug candidates that inhibit a protein kinase in the standard genetic toxicology assays. The first candidate induced micronuclei in an in vitro Chinese hamster ovary (CHO) cell assay, despite being negative in a bacterial reverse mutation assay, and an in vitro CHO chromosome aberration assay. A rat bone marrow micronucleus assay integrated into a 2-week oral toxicokinetic/tolerability study revealed a sub-threshold elevation in micronuclei. A second, stand-alone oral rat bone marrow micronucleus assay tested to higher doses over 3 days revealed a clear elevation in micronuclei formation. Subsequent antibody-based kinetochore staining of cultured CHO cells indicated that most micronuclei contained kinetochores. Our data are consistent with an aneugenic threshold-based mechanism of micronucleus induction that may be related to off-target drug effects present at very high concentrations above the intended therapeutic concentration.
Back to Top ↑
Michael D. Waters, Ph.D.
Integrated Laboratory Systems (ILS), Inc.
Understanding and Differentiating the Genotoxic Mode of Action Using Toxicogenomics Data
Scientific knowledge of the human health significance
of chemical exposures has grown rapidly; however,
experimental methods for evaluating the potential
carcinogenicity of environmental agents have
remained essentially unchanged over the past
20 years. Newer global methods (transcriptomics,
proteomics, and metabolomics) used in in vivo
and in vitro toxicogenomics studies, with appropriate
statistical classification methods, have progressed
such that it may be possible to use them to
predict carcinogenicity. A number of research
groups have identified cancer-relevant gene
sets that appear to discriminate carcinogens
vs. non-carcinogens. In general, the strategy
employed involves selecting a training set of
known carcinogenic and non-carcinogenic compounds
to which to expose mice or rats in vivo or cultured
cells in vitro for periods of from 1 to 90 days.
Statistical classification techniques are used
to identify expressed genes that discriminate
the possible outcome categories. The training
set genes are then used in classifying untested
chemicals. Published in vitro and in vivo studies
on a variety of microarray platforms have been
reviewed and will be presented to illustrate
the utility of the approach. The multi-factorial
nature of observed changes provide evidence
that a combination of pathway-associated gene
expression profiles that provide insights into
underlying mechanisms will be advantageous in
the prediction of chemical carcinogenesis. Together,
these studies and others demonstrate that toxicogenomics
is becoming robust and able to differentiate
genotoxic as well as various nongenotoxic modes
of action.