A knowledge-based approach to the in silico assessment of toxicity - - PowerPoint PPT Presentation

A knowledge-based approach to the in silico assessment of toxicity

Carol Marchant carol.marchant@lhasalimited.org

About Lhasa Limited

• Not-for-profit company and educational

charity

• Interests in knowledge and data sharing

in chemistry and the life sciences

Booth 70

Meteor An expert system for the assessment

• f xenobiotic metabolism

Some Lhasa Limited Products and Projects

Derek for Windows & Derek Nexus Expert systems for the assessment

• f toxicity

Vitic & Vitic Nexus Structure-searchable toxicity databases Zeneth An expert system for the assessment

• f chemical degradation pathways

Scope of Presentation

• In silico assessment of toxicity from chemical

structure

• Single component, organic chemicals of low to

medium molecular weight

• Human health-related

Outline

• Overview of expert systems in the in silico

assessment of toxicity

• Approaches to overcoming data availability

issues

Statistical model Expert system Molecular model Derek HazardExpert OncoLogic

I n Silico Toxicity Assessment

System which makes use of rules compiled by human experts and stored in a knowledge base

Derek Knowledge Base Structure

Typical rules describe: Presence of structural alerts Physicochemical property dependencies Known toxicity data Endpoint extrapolations Species-specific effects

Rules Comments References Structural alerts Comments References Validation comments Examples

Evidence might include consideration of: chemistry, mechanism of action, metabolism, pharmacology, physicochemical properties, toxicology etc

• Each rule that is applied to a query chemical provides a

qualitative indication of the likelihood of toxicity for a particular toxicological endpoint

• A typical rule might take the form:

If [Mutagenicity alert] is [certain] then [Mutagenicity] is [plausible]

• Arguments for and against toxicity from such rules are weighed

up against each other to arrive at an overall assessment

Derek Reasoning Process

Judson et al, Journal of Chemical Information and Computer Sciences 43 1364-1370 2003

Advantages of an Expert System Approach

• Human experts are able to work with diverse data types,

also allowing for inconsistent or incomplete data

Use can be made of all available information

• Assessments are meaningful and transparent

Suitable for both prediction and interpretation

• Knowledge can be expanded and updated without being

completely rebuilt

Readily adapted to incorporate proprietary knowledge in-house

Vitic editor

Sharing of Data and Knowledge

Vitic Derek Derek editor

Generate knowledge Enter knowledge Enter data Share knowledge Share data

Major Endpoint Categories in Derek

• Carcinogenicity
• Chromosome damage
• Genotoxicity
• Hepatotoxicity
• HERG channel inhibition
• Irritation
• Mutagenicity
• Ocular toxicity
• Reproductive toxicity
• Respiratory sensitisation
• Skin sensitisation
• Thyroid toxicity

I mpact of High Throughput Screening at Hoffmann-La Roche

Müller et al, Computational Toxicology (ed. S Ekins) pp 545-579 2007 10 20 30 40 50 60 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Year Percentage Positive Ames GLP Positive Ames screen

Introduction of in silico testing

I mpurity Assessment

Starting materials Intermediates APIs Starting materials Intermediates APIs Categorisation & management Categorisation & management

75% of chemicals can be assessed

• n the basis of structure alone

+ +

Dobo et al, Regulatory Toxicology and Pharmacology 44 282-293 2006

Process through Derek SciFinder, TOXNET & database search Synthetic route review

European Medicines Agency (2006) Guideline on the Limits of Genotoxic Impurities

http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2009/09/WC500002903.pdf

European Medicines Agency (2009) Q & A on the CHMP Guideline on the Limits of Genotoxic Impurities

http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2009/09/WC500002907.pdf

EMEA Guidance on I mpurities

US Food and Drug Administration (2008) DRAFT Guidance for Industry, Genotoxic and Carcinogenic Impurities in Drug Substances and Products: Recommended Approaches

http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm079235.pdf

DRAFT FDA Guidance on I mpurities

Continuous I mprovement

Application

Consensus modelling Integrated testing strategies

Presentation

OECD Principles for (Q)SAR validation

Performance

Availability of data Structure of knowledge

Data Availability

• Mutagenicity is a relatively easy endpoint to

derive knowledge for, at least in part because of the extent of available Ames test data

• For other endpoints, knowledge development

can become more problematic as complexity increases and data availability becomes an issue

Data may not exist Data may be in an inaccessible format Data may be proprietary

Strategies for I mproving Data Availability

• Data may not exist

Custom testing Precursory endpoints

• Data may not be in an inaccessible format
• Data may be proprietary

Chromosome Damage of Aromatic Thioamides

• Ethionamide induces in

vitro chromosomal aberrations in the absence of S9 mix

• Thioacetamide does not

N NH2 S

⇒Can we create a structural alert which associates all

aromatic thioamides with the induction of in vitro chromosomal aberrations?

NH2 S

Lhasa Limited collaboration with the National Institute of Health Sciences

+ +

• w+

w+

Aberrations

• S9 + S9

Custom Testing of Aromatic Thioamides

NH2 S NH2 S Cl Cl N H S

Thiobenzamide Chlorthiamid N-Phenyl thiobenzamide

• Structural alert scope based on available test data
• Postulated mechanism suggested to involve formation
• f electrophilic S-oxide metabolites

I mplementation of Structural Alert for Chromosome Damage of Aromatic Thioamides

R1 N S R2 H R1 = C (aromatic) R2 = H, C# (sp3), C (aromatic) C# cannot be attached to further heteroatoms Ortho substituted aromatic rings are excluded

Precursory Endpoints

• If insufficient data exist to develop knowledge for a

particular endpoint, it may be possible to develop knowledge for a precursory effect

• Mitochondrial dysfunction, for example, may be a precursor

to hepatotoxic or other adverse events

N N H NH N H2 NH N H N H NH N H2 NH N H N H NH N H2 NH

Mitochondrial data Structural alert describing mitochondrial dysfunction of bisguanides

Lhasa Limited collaboration with Pfizer; Fisk et al. Poster 206-005

N H2 N H N C NH NH # * N# - no heteroatoms allowed C* - cannot be double, triple or aromatically bonded

Strategies for I mproving Data Availability

• Data may not exist
• Data may not be in an inaccessible format

Semi-automated data extraction

• Data may be proprietary

Schema Vocabularies Queries Optical character recognition Text mining Check data Transfer data Read document I dentify facts Enter data Check data

Extracting Data from Legacy Reports

Manual extraction Semi-automated extraction

Lhasa Limited collaboration with Linguamatics and Pfizer

Semi-automated Extraction of Repeat Dose Study Reports

• Toxicity data can be captured using a semi-

automated extraction process with consistency and standardisation

• Extracted data has high accuracy and acceptable

recall

• Time taken is similar to manual extraction but

can be expected to improve over time

Strategies for I mproving Data Availability

• Data may not exist
• Data may not be in an inaccessible format
• Data may be proprietary

Derivation of non-proprietary knowledge Data sharing

Structural Alerts from Proprietary Repeat Dose Data

Proprietary data set (731 compounds) Clustering 34 structural alerts for hepatotoxicity

Lhasa Limited collaboration with the National Institute of Health Sciences

3-Furoic acid and derivatives structural alert

O R4 O R3 R1 R2 * * R1-R3 = H, C R4 = O , OH, OC Bonds marked * can be single or aromatic

• 32 non-proprietary

structural alerts for hepatotoxicity Expert analysis Expert review

Proprietary Data Sharing Model

Organisation 1 Organisation 2 Organisation 3 Organisation 4 Organisation 5 Organisation 6 Organisations 1-6

Database Data Data

Vitic I ntermediates Project

• Aims to provide data for:

Refinement and validation of in silico models Assessment of impurities for genotoxicity

• Involves sharing of proprietary Ames test data,

particularly for synthetic intermediates

• Consists of a consortium of 7 pharmaceutical companies
• Shared database of more than 500 chemicals to date

Application of Vitic I ntermediates Data

• Data has been used to:

Assess the performance of

existing structural alerts

Suggest new structural

alerts

20 40 60 80 100 120 2 7 12 15 16 19 27 28 33 39 49 69 303 304 305 315 318 326 329 344 349 351 352 353 354 471 645 Alert number Positive predictivity %

O O R1 O H R3 R2 R1-R3 = H, C (any)

4-Oxy-3(2H)-furanone structural alert

I MI Project

• Innovative Medicines Initiative (IMI) is a Joint

Technology Initiative between the European Federation

• f Pharmaceutical Industries and Associations (EFPIA) +

European Community within the European Union’s 7th Framework

• Aims to:

Establish a high-quality database of in vitro and in vivo data from

legacy pre-clinical studies

Build and validate new, integrated prediction models for target

• rgan toxicity
• Consists of a consortium of 13 EFPIA and 12 public

partners

Conclusions

• The use of in silico methods in toxicology is

commonplace and regulatory acceptance is increasing

• Expert systems offer one approach to the in

silico assessment of toxicity

• All methods are dependent on data for their
• ngoing improvement and we therefore need to

continue to be resourceful in terms of maximising data availability

Acknowledgements

• Linguamatics

Paul Milligan David Milward

• National Institute of Health Sciences

Makoto Hayashi (now at BSRC) Akihiko Hirose Masamitsu Honma Eichii Kamata

• Pfizer

Nigel Greene Russ Naven

• Lhasa Limited

Katharine Briggs Alex Caley Lilia Fisk Diana Suarez David Wilkinson Richard Williams