Carcinogenicity prediction for Regulatory Use Natalja Fjodorova Marjana Novič, Marjan Vračko, Marjan Tušar National institute of Chemistry, Ljubljana,

Slides:



Advertisements
Similar presentations
May Beneficiary 1 : Istituto di Ricerche Farmacologiche Mario Negri Beneficiary 2 : Istituto Superiore di Sanità Beneficiary 3 : Federazione.
Advertisements

ENTITIES FOR A UN SYSTEM EVALUATION FRAMEWORK 17th MEETING OF SENIOR FELLOWSHIP OFFICERS OF THE UNITED NATIONS SYSTEM AND HOST COUNTRY AGENCIES BY DAVIDE.
EPAA Annual conference November Regulatory acceptance of alternative approaches for pharmaceuticals Jean-Marc Vidal Safety & Efficacy of Human Medicines.
On Comparing Classifiers: Pitfalls to Avoid and a Recommended Approach Author: Steven L. Salzberg Presented by: Zheng Liu.
Perspectives from EPA’s Endocrine Disruptor Screening Program
Analysis of High-Throughput Screening Data C371 Fall 2004.
Design of Experiments Lecture I
Enhancing Data Quality of Distributive Trade Statistics Workshop for African countries on the Implementation of International Recommendations for Distributive.
Carcinogen Classification Criteria Patricia Richter Ph.D., DABT Tobacco Products Scientific Advisory Committee June 8, 2010.
1 Pharmacology/Toxicology information to submit an IND for an anticancer drug.
1 Development & Evaluation of Ecotoxicity Predictive Tools EPA Development Team Regional Stakeholder Meetings January 11-22, 2010.
PROBABILISTIC ASSESSMENT OF THE QSAR APPLICATION DOMAIN Nina Jeliazkova 1, Joanna Jaworska 2 (1) IPP, Bulgarian Academy of Sciences, Sofia, Bulgaria (2)
Software Quality Ranking: Bringing Order to Software Modules in Testing Fei Xing Michael R. Lyu Ping Guo.
Chemical Category Formation: Toxicology and REACH Dr Steven Enoch Liverpool John Moores University 14 th May 2009.
Decision Making: An Introduction 1. 2 Decision Making Decision Making is a process of choosing among two or more alternative courses of action for the.
NSF/ANSI STANDARD 61 FRAMEWORK FOR RISK ASSESSMENTS For use by Toxicology Sub-committee only Please do not copy or distribute.
Key issues in biopesticide regulation Pesticides Safety Directorate 19 June 2007.
Supervised classification performance (prediction) assessment Dr. Huiru Zheng Dr. Franscisco Azuaje School of Computing and Mathematics Faculty of Engineering.
What Do Toxicologists Do?
QSAR Modelling of Carcinogenicity for Regulatory Use in Europe Natalja Fjodorova, Marjana Novič, Marjan Vračko, Marjan Tušar, National institute of Chemistry,
Value of in vitro assays in your REACH dossier Frédérique van Acker 18 November 2014.
An Overview of Risk Assessment Bernard D. Goldstein, MD University of Pittsburgh Graduate School of Public Health.
FAO/WHO CODEX TRAINING PACKAGE
June 16-19, USEPA Cancer Guidelines: Mode of Carcinogenic Action 1 ICABR – Impacts of the Bioeconomy on Agricultural Sustainability, the Environment.
Attention Deficit Hyperactivity Disorder (ADHD) Student Classification Using Genetic Algorithm and Artificial Neural Network S. Yenaeng 1, S. Saelee 2.
Environmental Risk Analysis
Reporting & Ethical Standards EPSY 5245 Michael C. Rodriguez.
Health and Safety Executive UK Approach to Risk Assessment of Genotoxic Carcinogens in the Occupational Setting Dr Susy Brescia Chemicals Regulation Directorate.
Review of methods to assess a QSAR Applicability Domain Joanna Jaworska Procter & Gamble European Technical Center Brussels, Belgium and Nina Nikolova.
VTT-STUK assessment method for safety evaluation of safety-critical computer based systems - application in BE-SECBS project.
Evaluation of software engineering. Software engineering research : Research in SE aims to achieve two main goals: 1) To increase the knowledge about.
Project design & Planning The Logical Framework Approach An Over View Icelandic International Development Agency (ICEIDA) Iceland United Nations University.
” Particulates „ Characterisation of Exhaust Particulate Emissions from Road Vehicles Key Action KA2:Sustainable Mobility and Intermodality Task 2.2:Infrastructures.
Exploratory IND Studies
Adding Safety Pharm Endopoints To General Tox Studies - II Michael J Engwall, DVM, PhD Principal Scientist Safety and Exploratory Pharmacology Toxicology.
Mike Comber Consulting TIMES-SS Assessment of skin sensitisation hazard Presented on behalf of the TIMES-SS consortia.
Research & Science Advancing Risk Assessment Presentation March Association of Chemical Industry of the Czech Republic Monique Marrec Fairley.
Use of Machine Learning in Chemoinformatics Irene Kouskoumvekaki Associate Professor December 12th, 2012 Biological Sequence Analysis course.
Institute for Advanced Studies in Basic Sciences – Zanjan Kohonen Artificial Neural Networks in Analytical Chemistry Mahdi Vasighi.
Exploiting Context Analysis for Combining Multiple Entity Resolution Systems -Ramu Bandaru Zhaoqi Chen Dmitri V.kalashnikov Sharad Mehrotra.
Prof. Dr. Wolfgang Dekant Department of Toxicology University of Würzburg Germany Risk, Hazard, and Innovation.
Paola Gramatica, Elena Bonfanti, Manuela Pavan and Federica Consolaro QSAR Research Unit, Department of Structural and Functional Biology, University of.
Mike Comber TIMES-SS Application of Reactivity Principles in Screening for Skin Sensitisers Presented on behalf of the TIMES-SS consortia & International.
QSAR Study of HIV Protease Inhibitors Using Neural Network and Genetic Algorithm Akmal Aulia, 1 Sunil Kumar, 2 Rajni Garg, * 3 A. Srinivas Reddy, 4 1 Computational.
McKim Conference on Predictive Toxicology
Consider Incorporating Respiratory Safety Pharmacology Measurements into Your Next Repeat Dose Toxicology Study September 14, 2012 Jeff Tepper, PhD, DABT.
Selection of Molecular Descriptor Subsets for Property Prediction Inga Paster a, Neima Brauner b and Mordechai Shacham a, a Department of Chemical Engineering,
O PTIMAL NANO - DESCRIPTORS AS TRANSLATORS OF ECLECTIC DATA INTO PREDICTION OF THE CELL MEMBRANE DAMAGE BY MEANS OF NANO METAL - OXIDES A LLA P. T OROPOVA.
Barcelona April, 2008 Overview of the QSAR Application Toolbox Gilman Veith International QSAR Foundation Duluth, Minnesota.
MUTAGENICITY OF AROMATIC AMINES: MODELLING, PREDICTION AND CLASSIFICATION BY MOLECULAR DESCRIPTORS M.Pavan and P.Gramatica QSAR Research Unit, Dept. of.
Science Symposium, 26 May 2014, New Delhi, India Dr Gerald Renner Director Technical Regulatory Affairs Cosmetics Europe EU scenario on alternatives in.
THE PRAGMATICS OF CORRELATION OR HOW MODELS RESHAPE THE GOVERNMENT OF TECHNICAL OBJECTS BRICE LAURENT & FRANÇOIS THOREAU CENTRE DE SOCIOLOGIE DE L’INNOVATION,
Use of Machine Learning in Chemoinformatics
1 Biopharmaceutics Dr Mohammad Issa Saleh. 2 Biopharmaceutics Biopharmaceutics is the science that examines this interrelationship of the physicochemical.
Draft EU Risk Assessment and Mapping Guidelines for Disaster Management 8th Meeting of Working Group F on Floods October 2010 Commission ECHO.C4.
Summation of Toxicity Data in Vitic Andrew Thresher
Abstract A step-wise or ‘tiered’ approach has been used as a rational procedure to conduct environmental risk assessments in many disciplines. The Technical.
Toxicity vs CHEMICAL space
General Concepts in QSAR for Using the QSAR Application Toolbox
JMP Discovery Summit 2016 Janet Alvarado
QSAR Application Toolbox: Step 12: Building a QSAR model
General Concepts in QSAR for Using the QSAR Application Toolbox
ADME/Tox PredictionTox Prediction. The characterization of Absorption, Distribution, Metabolism, and Excretion (also known as ADME) and Toxicity are essential.
NBA Draft Prediction BIT 5534 May 2nd 2018
Biopharmaceutics Dr Mohammad Issa Saleh.
network of simple neuron-like computing elements
IDEA International Dialogue for the Evaluation of Allergens
The Category Approach for Predicting Mutagenicity and Carcinogenicity
EFSA’s Chemical Hazards Database
Introduction to Risk Assessment
Presentation transcript:

Carcinogenicity prediction for Regulatory Use Natalja Fjodorova Marjana Novič, Marjan Vračko, Marjan Tušar National institute of Chemistry, Ljubljana, Slovenia

Kemijske Dnevi September 2008 UNIVERZA MARIBOR

Overview 1. EU project CAESAR aimed for development of QSAR models for prediction of toxicological properties of substances, used for regulatory purposes. 2. The principles of validations of QSARs which will be used for chemical regulation. 3. Carcinogenicity models using Counter Propagation Artificial Network

It is estimated that over industrial chemicals used in Europe require additional safety testing to meet requirements of new chemical regulation REACH. If conducted on animals this testing would require the use of an extra million animal experiments. Quantitative Structure Activity Relationships (QSAR) is one major prospect between alternative testing methods to be used in a regulatory context.

aimed to develop (Q)SARs as non-animal alternative tools for the assessment of chemical toxicity under the REACH. FR6- CAESAR European Project Computer Assisted Evaluation of Industrial chemical Substances According to Regulations Coordinator- Emilio Benfenati- Istituto di Ricerche Farmacologiche “Mario Negri”

The general aim of CAESAR is 1. To produce QSAR models for toxicity prediction of chemical substances, to be used for regulatory purposes under REACH in a transparent manner by applying new and unique modelling and validation methods.

2. Reduce animal testing and its associated costs, in accordance with Council Directive 86/609/EEC and Cosmetics Directive (Council Directive 2003/15/EC)

CAESAR is solving several problems: Ethical- save animal lifes; Economical- cost reduction on testing; Political- REACH implementation- new chemical legislation

CAESAR aimed to develop new (Q)SAR models for 5 end-points: Bioaccumulation (BCF), Skin sensitisation Mutagenicity Carcinogenicity Teratogenicity

The characterization of the QSAR models follows the general scheme of 5 OECD principles: 1.A defined endpoint 2.An unambiguous algorithm 3.A defined domain of applicability 4.Appropriate measures of goodness- of-fit, robustness and predictivity 5.A mechanistic interpretation, if possible.

Principle1- A defined endpoint Endpoint is the property or biological activity determined in experimental protocol, (OECDTest Guideline). Carcinogenicity is a defined endpoint addressed by an officially recognized test method (Method B.32 Carcinogenicity test – Annex V to Directive 67/548/EEC).

Principle2- An unambiguous algorithm Algorithm is the form of relationship between chemical structure and property or biological activity being modelled. Examples: 1. Statistically (regression) based QSARs 2. Neural network model, which includes both learning process and prediction process.

Transparency in the (Q)SAR algorithm can be provided by means of the following information: a) Definition of the mathematical form of a QSAR model, or of the decision rule (e.g. in the case of a SAR) b) Definitions of all descriptors in the algorithm, and a description of their derivation c) Details of the training set used to develop the algorithm.

Principle3- A Defined Domain of Applicability The definition of the Applicability Domain (AD) is based on the assumption that a model is capable of making reliable predictions only within the structural, physicochemical and response space that is known from its training set. List of basic structures (for example, aniline, fluorene..) The range of chemical descriptors values.

The assessment of model performance is sometimes called statistical validation. Principle4- Appropriate measures goodness-of-fit, robustness (internal performance) and predictivity (external performance)

Principle5- A mechanistic interpretation, if possible Mechanistic interpretation of (Q)SAR provides a ground for interaction and dialogue between model developer, and toxicologists and regulators, and permits the integration of the (Q)SAR results into wider regulatory framework, where different types of evidence and data concur or compliment each other as a basis for making decisions and taking actions. Example: enhancing/inhibition the metabolic activation of substances may be discussed.

National Institute of Chemistry in Ljubljana (NIC-LJU) is responsible for development of models for predicton of carcinogenicity

DATA ON CARCINOGENICITY 1.Studies of carcinogenicity in humans 2.Carcinogenicity studies in animals 3.Other relevant data additional evidence related to the possible carcinogenicity Genetic Toxicology Structure-Activity Comparisons Pharmacokinetics and Metabolism Pathology

Cancer Risk Assessment IARC International Agency for Research of Cancer IARC For animals Group Classification Explanation Classification Group AHuman Carcinogen sufficient human evidence for causal association between exposure and cancer Group B1Probable Humanlimited evidence in human Group B2Probable Human inadequate evidence in humans and sufficient evidence in animalsclear evidence Group C Possible Human Carcinogenlimited evidence in animalssome evidence Group D Not Classifiable as Human Carcinogenicityinadequate evidence in animalsequivocal Group E No Evidence of Carcinogenicity in Human at least two adequate animal tests or both negative epidemiology and animal studiesno evidence

Predictive Toxicology Approaches 1. Quantitative models (QSARs) Continuous data prediction on the basis of experimental evidence of rodent carcinogenic potential (TD50 tumorgenic dose) 2. Categorical models based on YES/NO data. (P-positive; NP-not positive)

Dataset: 805 chemicals were filtered from 1481compounds taken from Distributed Structure-Searchable Toxicity (DSSTox) Public Database Network as.html as.html which was derived from the Lois Gold Carcinogenic Database (CPDBAS) The chemicals involved in the study belong to different chemical classes, (noncongeneric substances)

Descriptors: MDL descriptors were calculated in program MDL QSAR. 2. Descriptors dataset was reduced to 27 MDL descriptors, using Kohonen map and Principle Component Analisis.

Counter Propagation Artificial Neural Network Step1: mapping of molecule Xs (vector representing structure) into the Kohonen layer Step2: correction of weights in both, the Kohonen and the Output layer Step3: prediction of the four- dementional target (toxicity) Ts

Investigation of quantitative models shows us low results RESPONCE- TD50mmol 1. Correlation coefficient in the external validation is lower then 0.5

Continuouse data models (Quantitative models) Models Reduction of descriptors method, model TRAININGTEST R_trainRMSER_testRMSE CP ANN_model 250MDLdescriptors CP ANN_model 86MDLdescriptors Kohonen map CP ANN_model 27MDLdescriptors PCA SVM_model (Thomas Ferrary) 86MDLdescriptors

Investigation of categorical models shows us satisfactory results YES/NO principe RESPONCE: P-positive-active NP-not positive-inactive

Characteristics used for validation of categorical model true positive(TP), true negative (TN) Accuracy(AC), AC=(TN+TP)/(TN+TP+FN+FP) TPrate=Sensitivity(SE)=TP/(TP+FN) TNrate=Specificity(SP)=TN/(TN+FP)

Categorical model for dataset 805 chemicals (Training=644 and Test=161), using 27 MDL descriptors TrainingTest ACC, % SE, % SP, % ACC, % SE, % SP, % Model _ Model _

Confusion matrix TR(644)/TE(161) classes (Positive- Negative) Class Positive (predict.) Negative (predict.) Number TR(TE) 644(161) Positive (experim.) 329(65) 3(24)332(89) Negative (experim.) 47(27)265(45)312(72) FP FNTP TN

How we find optimal model, using threshold Threshold=0.45 Accuracy=0.68 SE=0.73 SP=0.63

Changing of threshold allows us to get models with different statistical performances. TrSESPACC

ROC(Receiver operating characteristic) curve Training set Test set The area under the curve is and in the training and test sets, respectively.

How requrements of REACH reflect development of models To focus model to high sensitivity in prediction of carcinogenicity From regulatory perspective, the higher sensitivity in predicting carcinogens is more desirable than high specificity Sensitivity- percentage of correct predictions of carcinogens Specificity- percentage of correct predictions of non-carcinogens

Conclusion 1.We have bult the carcinogenicity models in accordance with 5 OECD principles principle of validation 2. We have got satisfactory results for categorical models with accuracy 68% which is good for carcinogenicity as it meet the level of uncertanty of test data. 3. The goal of our future investigation will be dedicated to research of relationship between results of carcinogenicity tests and presence of Genotoxic, non Genotoxic alerts using TOX TREE program.

Acknowledgements The financial support of the European Union through CAESAR project (SSPI ) as well as of the Slovenian Ministry of Higher Education, Science and Technology (grant P1-017) is gratefully acknowledged.

THANK YOU