1. CADASTER has joined the EUROECOTOX network
2. Data collection: Final compounds for testing were selected
3. Models development: Local models for CADASTER chemicals
4. Integration of QSARs within chemical decision making
5. The dissemination of information: Why and How?
6. CADASTER Workshop on the use of QSAR models in REACH at the CMTPI conference
7. SETAC Europe 2011
8. Meet CADASTER

Within the FP7 Environment Programme, the coordination action EUROECOTOX was recently funded by the European Commission. EUROECOTOX is a European Network established to promote the integration of European activities on the replacement and reduction of animal experiments in ecotoxicology. EUROECOTOX aims to foster the exchange of knowledge, data, ideas and available European expertise to accelerate the R&D, (pre)validation and regulatory acceptance of alternative ecotoxicity tests and methods.
Read more

Animal welfare is nowadays the subject of significant societal ethical concerns and there is a strong European support to Refine, Reduce and Replace (3Rs) animals linked to regulation compliance. The creation of the EUROECOTOX network was originally inspired by the 3Rs ambitions and motivated  due to the implementation of EU regulations on chemicals and their transformation products (REACH) that no doubt will require an intensification of eco-toxicological safety testing, requiring the submission of appropriate data for risk assessment for the registration of chemicals, pesticides, biocides and pharmaceuticals. This could result in an increase in the use of laboratory animals, and in order to avoid it and pursue the 3Rs goals, you must first understand the status of the validation and regulatory approval process of alternative ecotoxicological test methods in Europe and the existing conditions and stages of the alternative test methods and strategies.

The EUROECOTOX network is open to European R&D groups from universities, public research institutions, animal welfare groups and other non-profit organizations, specialized ecotoxicological contract laboratories, industry and other stakeholders developing alternative test methods. CADASTER is one of the stakeholders that has joined the network. This was done not only to join a platform for exchange of information and approached that are of use for CADASTER but also to allow for more efficient dissemination of project results within a broader, relevant, platform.

Further information on the the EUROECOTOX Network may be found at the following website:
https://sites.google.com/a/euroecotox.eu/networkeuroecotox/HOME

The following link may be used to join the network:
https://sites.google.com/a/euroecotox.eu/networkeuroecotox/HOME/about-euroecotox/how-to-join-the-network

Final compounds for testing were selected according to structural representativity, toxicity and physical chemical properties, REACH relevance and analytical possibilities for polybrominated diphenylethers (PBDE), perfluoroalkylated substances (PFA), substituted musks/fragrances and triazoles/benzotriazoles ((B)TAZ).
Read more

In the beginning of 2009 the work on the Cadaster project began with the collection of existing experimental data. Data search on all endpoints of relevance was performed for the environ­mental risk and hazard assessment of the groups of chemicals included in the case studies. The survey of the existing QSAR/QSPR models was performed for the four classes of chemicals selected in this project (PBDEs, PFAs, fragrances and (B)TAZs). An overview was prepared of the non-testing options given under REACH to either replace experimental testing, or to strengthen confidence in experimental results (QSARs, read-across, category approaches, and exposure based waiving). The extended database on existing experimental data and (Q)SAR models was created on the four selected group of chemicals.

In WP2, also new experimental data on the most essential endpoints of assessment are generated. With regard to this task, the final compounds for testing were selected according to structural representativity, toxicity and physical chemical properties, REACH relevance and analytical possibilities for polybrominated diphenylethers (PBDE), perfluoroalkylated substances (PFA), substituted musks/fragrances and triazoles/benzotriazoles ((B)TAZ).

The activity on prioritization to optimize the experimental testing was performed in WP3 during the first two years of the project. Similarity analysis and multivariate ranking methods were applied for experimental design and the identification of priority chemicals in the four chemical classes of interest. The prioritization was performed:

1. on the basis of toxicological and chemical-physical profiles defined for the studied classes of chemicals. The basic idea is that the priority chemicals are those more hazardous sensu latu, thus those that have any demonstrated kind of toxicity. Owing to the general paucity of experimental data, in particular for SIDS endpoints, for these chemicals, the available experimental data on toxicity end points (also mammalian toxicity and endocrine disruption activity) were used for the characterization of toxicological profile. In addition to the available experimental data also data predicted by the new “ad hoc” QSAR models developed by UI (WP3, task 3.5) were used.
2. on the basis of chemical structure, in order to select representative chemicals in the structural space of each studied chemical class.

After prioritization procedure the final selection of chemicals for experimental testing was carried out taking into account the REACH relevance of chemicals and the analytical possibilities (method availability and availability of chemicals on the market). Selected chemicals are listed in Table 1 -Table 4.

Table 1. PBDE congeners selected for experimental testing

No	Selected PBDEs	No. Br
1	BDE-2	1
2	BDE-66	4
3	BDE-77	4
4	BDE-119	5
5	BDE-126	5
6	BDE-190	7
7	BDE-197	8
8	BDE-198	8
9	BDE-203	8
10	BDE-204	8
11	BDE-207	9

Table 2. List of fragrances selected for the experimental testing

	Mol ID	Name	Functional Class	CAS Nr.
1	FRA-001	Acethyl cedrene	Terpenes	032388-55-9
2	FRA-006	Benzyl cinnamate	Cynnamic acid der.	000103-41-3
3	FRA-022	Hexyl salycilate	Salycilic acid der.	006259-76-3
4	FRA-023	Hexylcinnamaldeyde	Cynnamic acid der.	000101-86-0
5	FRA-024	HHCB (Galaxolide)	Musks	001222-05-5
6	FRA-041	Methyl dihydrojasmonate	Terpenes	024851-98-7
7	FRA-049	Quinidine	Ref.Comp. (Drug)	000056-54-2
8	FRA-054	α-amylcinnamyl alcohol	Cynnamic acid der.	000101-85-9
9	FRA-056	Musk ambrette	Musks	000083-66-9
10	FRA-065	Cyclopentadecanolide	Musks	000106-02-5
11	FRA-069	Benzyl Benzoate	other compounds	000120-51-4

Table 3. List of PFCs selected for the experimental testing

No	Selected PFCs	CAS Nr.
1	PFBA	000375-22-4
2	1-Pentanol, 2,2,3,3,4,4,5,5-octafluoro-	000355-80-6
3	PFOA	000335-67-1
4	APFO	003825-26-1
5	PFNA	000375-95-1
6	PFDA	000335-76-2
7	PFUnA	002058-94-8
8	PFDoA	000307-55-1
9	PFOS	001763-23-1
10	Butanoic acid, heptafluoro-, ethyl ester	000356-27-4
11	Methacrylic acid, 2,2,3,3,4,4,4-heptafluorobutyl ester	013695-31-3
12	1-Butanesulfonamide, N-ethyl-1,1,2,2,3,3,4,4,4-nonafluoro-N-(2-hydroxyethyl)-	034449-89-3
13	2-Propenoic acid, 2-methyl-, 2- ethyl (nonafluorobutyl)sulfonyl amino ethyl ester)	067939-33-7

Table 4. Priority list of (benzo)triazoles on the basis of their toxicological profile and structural similarity

ID	Selected (B)TAZ	CAS
1	Diclobutrazol	075736-33-3
2	1H-1,2,4-Triazole-1-ethanol, beta-(4-chlorophenoxy)-alpha-(1,1-dimethyl)-,(alphaR,betaR)-rel-	082200-72-4
3	Uniconazole-P [ISO]	083657-17-4
4	Diniconazole [ISO]	083657-24-3
5	Triadimenol A	089482-17-7
6		098519-02-9
7		098519-41-6
8		098532-70-8
9		098532-73-1
10	Flupoxam [ISO:BSI]	119126-15-7
11	Sulfentrazone [ISO]	122836-35-5
12	Ipconazole [ISO]	125225-28-7
13	Ssf 109	129586-32-9
14	ipfencarbazone	212201-70-2
15	(2-hydroxy-3,3-dimethyl-2-[1,2,4]triazol-1-ylmethyl-cyclopentyl)-(4-chlorophenyl)-methanone (Ref: CL 382389)
16	2-(2,4-dichlorophenyl)-3-(1H-1,2,4-triazol-1-yl)propan-1-ol (Ref: M14360-alcohol)
17	2-(1-chlorocyclopropyl)-1-(2-chlorophenyl)-3-[5-(methylthio)-1H-1,2,4-triazol-1-yl]propan-2-ol
18	2-chloro-3-{2-chloro-5-[4-(difluoromethyl)-3-methyl-5-oxo-4,5-dihydro-1H-1,2,4-triazol-1-yl]-4-fluorophenyl}propanoic acid
19	(E)-2-(4-chlorobenzylidene)-5,5-dimethyl-1-((1H)-1,2,4-triazol-1-ylmethyl)-cyclopentan-1,3-diol (Ref: RPA 404766)
20	trans-5-(4-chlorophenyl)-dihydro-3-phenyl-3- (1H-1,2,4-triazole-1-ylmethyl)-2-3H-furanone (Ref: RH-9130)
21	4-(4-chlorophenyl)-2-(methyl-1H-1,2,4-triazole)-4-oxo-2-phenylbutanenitrile (Ref: RH-6467)
22	cis-5-(4-chlorophenyl)-dihydro-3-phenyl-3-(1H-1,2,4-triazole-1-ylmethyl)-2-3H-furanone (Ref: RH-9129)
23	(2RS)-1-(4-chlorophenyl)-4,4-dimethyl-2-(1H-1,2,4-triazol-1-yl)pentan-3-one (Ref: CGA 149907)

In WP3 several local models for the selected classes of CADASTER chemicals have been already developed, presented to various international meetings and published in ISI scientific journals. Some other models for various not-SIDS toxicity end-points have been also developed and their predictions were used to create priority lists of compounds to be suggested for experimental tests performed by CADASTER Partners
Read more

In particular, models for SIDS physico-chemical properties have been developed for PBDEs (MP, VP, S, Kow, Koa), PFCs (MP, BP, S, VP, CMC) and (B)TAZs (MP, VP, S, Kow), and applied to compounds of the same chemical class present in the ECHA list, verifying the structural applicability domain. The applicability of the CADASTER local models to the specific class of compounds is generally high and the quality of the predictions are always better than those obtained by general models, as those in EPISuite. For MP and BP of PFCs consensus predictions, based on different models developed by the WP3 partners, have been also proposed.

At present, all the WP3 partners are modeling data of fish toxicity, specifically on Oncorhynchus mykiss for (B)TAZs and Pimephales promelas (heterogeneous dataset): the predictions of the different approaches (MLR, PLS, ANN), with external validation performed during the modeling step, will be verified on two common external prediction sets of (B)TAZs and predicted data by consensus will be shared among partners and proposed as better predictions for (B)TAZs present in ECHA list and belonging to models’ AD.

Some other models for various not-SIDS toxicity end-points (oral and inhalation toxicity in rat and mouse measured for PFCs, endocrine disruptive potency of PBDEs, and citotoxicity of fragrances) have been also developed and their predictions were used to create priority lists of compounds to be suggested for experimental tests performed by CADASTER Partners.

A goal of REACH is to regulate the use of chemical compounds under a proper balance between societal, economic and environmental objectives. Regulating a chemical make us ask if the chemical compound is dangerous and how dangerous it is. Risk assessment offers a framework to digest current background knowledge on the query compound into information that allows a rational decision to be made.
Read more

However, our knowledge is never complete, and instead of waiting for this to happen, we have to consider information from varying levels of strength in relation to the best possible knowledge. QSARs offer an alternative mean to provide information compared to having experimentally derived data on the query compound. QSARs are therefore useful for decision making because it opens up for balancing between the cost of regulation, time and avoiding animal testing, and the potential impact of compounds. The implementation of QSARs into decision making on chemical compounds within REACH is the focus of the fourth work package of the CADASTER project.

Understanding the effects from using QSARs instead of experimental data in chemical safety assessment will help to achieve a balanced treatment of uncertainty in chemical regulation. Information derived from model predictions, instead of experimentally tested data, contains uncertainty due to the fact that models cannot fully represent reality. However, experimental data are also associated with different kinds of errors and naturally occurring non-reducible uncertainty (variability). At best we would like to eliminate knowledge-based uncertainty, but we have to accept some extent of uncertainty and learn how to deal with it when making decisions.

The reliability of background information for a regulatory decision depends on the predictive abilities of used QSARs and the extent of uncertainty associated to the predicted information. It is relevant to known the extent of this predictive uncertainty and what role it has on regulatory decisions. To this end, we have identified approaches that characterize predictive uncertainty from QSARs as variability in experimental data, empirical estimates from independent data samples or by explicit modeling of uncertainty. These methods are currently evaluated from theoretical, empirical and practical perspectives. By working on the integration of QSAR models into a probabilistic risk assessment, i.e. a risk assessment where uncertainty is characterized by probabilistic methods, we are from a practical decision problem addressing general methodological questions related to decision making under uncertainty.

The requirement of strong background information for chemical regulation, i.e. experimental testing, should be related to the potential consequences of a chemical compound. Therefore one aim is to explore the possibilities for economic valuation of substitution of chemicals from within chemical classes. As a starting point we are using estimated costs associated with impacts of polybrominated diphenyl ethers (PBDEs) on a population of peregrine falcons. So far, very few studies have assessed the external costs of chemical impacts on ecosystem properties or functions. We are working on a case study assessing the costs of adverse effects of PBDEs on peregrine falcons (Falco peregrinus). We use a population model in which effects of PBDEs on peregrine falcon population size are mediated through reduced reproduction success. We parameterized the model for the peregrine falcon population of California from 1992 through 2020. Cost estimates are based on the number of captive-bred individuals needed to compensate for the impacts of PBDEs. Three exposure scenarios were designed to account for uncertainties in future environmental concentrations of PBDEs. Uncertainties in ecological and toxicological model parameters were accounted for by a probabilistic approach. Results will be ready soon.

The implementation of QSARs into decision making is also addressed from a practical point of view. We have evaluated ECETOC’s Targeted risk assessment tool for its ability to identify chemicals of concern, at varying levels of information. We found that it would be difficult to use this tool as a part of the project for on the fly risk assessments since there are great shortcomings in the terms of the usability of the tool. This means that there is a need of tools for chemical regulation that aid smaller companies without the skills or expertise. This is an ongoing work by other projects.

The CADASTER project has produced a number of data, models, theoretical approaches to be used for the risk assessment within the REACH. The dissemination of this information to the end users, in particular to SMEs, risk assessors and national chemicals authorities, is the important part of the CADASTER project strategy to promote application of QSAR approaches and the use of non-animal testing strategies. CADASTER uses different ways to provide the widest possible dissemination of its results.
Read more

The CADASTER web site http://www.cadaster.eu has been created and supported from the begging of the project. It provides dissemination of the recent activities within the project, such as information about published articles, upcoming meetings, conferences, contains oral and poster reports and lists articles published in peer-reviewed literature. Up to know CADASTER has produced 13 article in peer-reviewed journals, 26 posters and 16 oral presentations at 20 conferences. A detailled list can be seen here

The data and models produced during the project are available through QSPR-THESAURUS web site. This web site is based on Open Chemical Modelling Environment OCHEM, which has been developed by the HelmholtzZentrum Munich. Recently the description of the platform has been published elsewhere (Sushko et al, 2011). The OCHEM platform was provided as a background information to the CADASTER and it was further customized to the project requirements within the framework of the project. The OCHEM itself comprises more than 100,000 lines of Java, C++, and shell script code and it was developed by a team of 5 people over three years. The majority of data collected within CADASTER project were hidden and password protected. Starting end of December 2010 these data were released to public. These data are freely available at the http://www.qspr-thesaurus.eu and can be searched by molecular name, Chemical Abstract Service (CAS) number, substructure search, property, condition, article as well as by published models. The available data include those collected from published literature as well as measured during the project in laboratories of project.
The browser with selection of published CADASTER models at the QSPR-THESAURUS web site.

The QSPR-THESAURUS contains a number of additional features that were elaborated specifically for the CADASTER project. This includes a possibility to upload models developed within the CADASTER project, similarity search, calculation of applicability domains, support of description of models in QRMF format, experimental design and selection compounds for experimental testing using methodology being developed within the project. All these features allows external users to better understand and to exemplify all necessary steps performed by experimentalists, chemoinformatians and computational chemists from the stage of planning of experimental measurements, data collection, curation, filtering to model development, analysis, validation and interpretation. These steps are required to develop good QSAR models and to ensure their application for the risk assessment.

The first CADASTER workshop will be organized in Maribor, Slovenia, as a part of the CMPTI conference and will take place from September 1st to 2nd 2011. The workshop aims to assist the risk assessors and national chemicals autho­rities, particular in Eastern European countries, with the use of the QSAR tools for the environmental risk assessments in REACH.
Read more

The SETAC Europe 2011 meeting was organized in Milan, 15-19 May 2011. The theme of this year’s conference was “Ecosystem Protection in a Sustainable World: a Challenge for Science and Regulation.” Meeting sessions spanned a multitude of topics to discuss and exchange latest information leading to sustainable environmental quality and ecosystem integrity under the stress of economic development.
The session “Risk Assessment of chemicals within REACH integrating alternative methods and non testing strategies “ was chaired by Dr. Ester Papa and Dr. Michael Neumann.
Read more

The aim of this session was to provide a forum to exchange experiences gained by regulators and academics within REACH and REACH-related FP7 EU projects such as CADASTER, OSIRIS and OPENTOX. This session contributed among others to the development, validation and application of QSAR/QSPR models focussed on the use and reliability of Read Across, Intelligent Testing Strategies or Weight of Evidence approaches.


All the platform presentations, mainly presented by the academic branch of SETAC, were focussed on the practical approach and the future application of their result into the REACH-system. Two platforms were presented by CADASTER partners: an overview of the project CADASTER (Dr. W. Peijnenburg - coordinator) and updates in the field of characterisation of QSAR models uncertainty (Dr. T. Oberg). Other platforms presented actual improvements or updates in the field of QSARs applied to Threshold of Toxicological Concern, transformation products assessment, level of stakeholders’ acceptance of Integrated Testing Strategies, and the OECD QSAR Toolbox. The platform session was well attended by over 100 audiences.

At the same time 24 posters of this session presented a wide range of topics and attracted high interests throughout the day. Seven posters were presented by CADASTER partners.
Overall this session reflected experiences in chemical regulation, proposed new and emerging approaches for decision making in the future. It also provided a forum for participants from around the world to stimulate the exchange between eco-toxicologists, substance property experts, (Q)SAR model developers, and non-(Q)SAR specialists requiring access to models and data.

CADASTER presentations at SETAC

Brandmaier, S.; Abdelaziz, A.; Sahlin, U.; Öberg, T.; Tetko, I. V. Stepwise D-Optimal design based on latent variables, SETAC Europe annual meeting, Milan, Italy May 15-19, 2011, presentation by Stefan Brandmaier, download poster

Brandmaier, S.; Srivastava, V.; Tetko, I. V.; Körner, R.; Teetz, W.; Novotarskyi, S.; Pandey, A.; Sushko, I. MOPAC@home - a structural database for small chemical compounds, SETAC Europe annual meeting, Milan, Italy May 15-19, 2011, presentation by Stefan Brandmaier, download poster

Tomas Öberg, T; Iqbal, M. S. Uncertainty in fugacity-based multimedia modeling: probabilistic and non-probabilistic methods, SETAC Europe annual meeting, Milan, Italy May 15-19, 2011, download poster

Sahlin, U; Aldenberg, T.; Öberg, T. Towards guidance on how to characterize predictive uncertainty in QSAR regression models, SETAC Europe annual meeting, Milan, Italy May 15-19, 2011, download presentation

Sahlin, U; Bhhatarai, B; Papa, E.; Öberg, T.; Gramatica, P. A QSAR-based compound prioritization for lab-testing for chemical safety assessment, SETAC Europe annual meeting, Milan, Italy May 15-19, 2011, download poster

Papa, E.; Cassani, S.; D’Onofrio, E.; Kovarich, S.; Roy, P. P.; Rahmberg, M.; Nilsson, S.; Gramatica, P. QSAR prediction of aquatic and mammalian toxicity of triazoles and benzo-triazoles, SETAC Europe annual meeting, Milan, Italy May 15-19, 2011, presentation by Ester Papa, download poster

Kovarich, S.; Bhhatarai, B,; Papa, E.; Rahmberg, M.; Nilsson, S.; Liu, T.; Öberg, T.; Jeliazkova, N.; Kochev, O.; Teetz, W.; Brandmaier, S.; Tetko, I. V.; Gramatica, P. QSAR and QSPR models for emerging pollutants: WP3 activities within the FP7 European Project CADASTER, SETAC Europe annual meeting, Milan, Italy May 15-19, 2011, presentation by Simona Kovarich, download poster

Chirico, N.; Papa, E.; Gramatica, P. On the agreement of external validation parameters for linear regression QSAR models, SETAC Europe annual meeting, Milan, Italy May 15-19, 2011, presentation by Paola Gramatica, download poster

Bhhatarai, B,; Gramatica, P. Physico-chemical property prediction of emerging pollutants: PFC and (B)TAZ for environmental distribution, SETAC Europe annual meeting, Milan, Italy May 15-19, 2011, download poster

Roy, P. P.; Kovarich, S.; Papa, E.; Gramatica, P. Exploring the QSARs for OH Tropospheric Degradation of VOCs using freely available online descriptors, SETAC Europe annual meeting, Milan, Italy May 15-19, 2011, presentation by Partha Pratim Roy, download poster

Peijnenburg, W.; Durjava, M.; Gramatica, P.; Öberg, T.; Rahmberg, M.; Tetko, I.; Jeliaskova, N.; Huijbregts, M.; Comber, M. Exemplification of the integration of tools within REACH: the CADASTER project, SETAC Europe annual meeting, Milan, Italy May 15-19, 2011, presentation by Willie Peijnenburg, download presentation

Peijnenburg, W.; Ding, G. Toxicity of poly- and perfluorinated compounds to lettuce (Lactuca sativa) and green algae (Pseudokirchneriella subcapitata), SETAC Europe annual meeting, Milan, Italy May 15-19, 2011, poster by Willie Peijnenburg, download poster

• OpenTox InterAction Meeting

  9 - 12 August 2011 Munich, Germany

• CADASTER Workshop

  1 – 2 September 2011 Maribor, Slovenia

• 6th International Symposium on Computational Methods in Toxicology and Pharmacology Integrating Internet Resources (CMTPI 2011)

  3 – 7 September 2011 Maribor, Slovenia

• German Conference on Bioinformatics 2011

  7 - 9 September 2011 Munich, Germany

• EuCheMS International Conference on Chemistry and the Environment (ICCE 2011) – Workshop on Per- and Polyfluorinated Chemical

  11 September 2011 Zurich, Switzerland

• Conferentia Chemometrica (CC 2011)

  18-21 September 2011 Sümeg, Hungary

• 16th International Symposium on Environmental Pollution and its Impact on Life in the Mediterranean Region, organized by MESAEP (Mediterranean Scientific Association of Environmental Protection)

  24 - 27 September 2011 Ioannina, Greece

• 7th German Conference on Chemoinformatics

  6 – 8 November 2011 Goslar, Germany

• SETAC North America 32nd Annual Meeting

  13 – 17 November 2011 Boston, Massachusetts, USA