Use of in silico systems and expert knowledge for structure-based assessment of potentially mutagenic impurities.

Genotoxicity hazard identification is part of the impurity qualification process for drug substances and products, the first step of which being the prediction of their potential DNA reactivity using in silico (quantitative) structure-activity relationship (Q)SAR models/systems. This white paper provides information relevant to the development of the draft harmonized tripartite guideline ICH M7 on potentially DNA-reactive/mutagenic impurities in pharmaceuticals and their application in practice. It explains relevant (Q)SAR methodologies as well as the added value of expert knowledge. Moreover, the predictive value of the different methodologies analyzed in two surveys conveyed in the US and European pharmaceutical industry is compared: most pharmaceutical companies used a rule-based expert system as their primary methodology, yielding negative predictivity values of ⩾78% in all participating companies. A further increase (>90%) was often achieved by an additional expert review and/or a second QSAR methodology. Also in the latter case, an expert review was mandatory, especially when conflicting results were obtained. Based on the available data, we concluded that a rule-based expert system complemented by either expert knowledge or a second (Q)SAR model is appropriate. A maximal transparency of the assessment process (e.g. methods, results, arguments of weight-of-evidence approach) achieved by e.g. data sharing initiatives and the use of standards for reporting will enable regulators to fully understand the results of the analysis. Overall, the procedures presented here for structure-based assessment are considered appropriate for regulatory submissions in the scope of ICH M7.

In vivo flow cytometric Pig-a and micronucleus assays: highly sensitive discrimination of the carcinogen/noncarcinogen pair benzo(a)pyrene and pyrene using acute and repeated-dose designs.

Combining multiple genetic toxicology endpoints into a single in vivo study, and/or integrating one or more genotoxicity assays into general toxicology studies, is attractive because it reduces animal use and enables comprehensive comparative analysis using toxicity, metabolism, and pharmacokinetic information from the same animal. This laboratory has developed flow cytometric scoring techniques for monitoring two blood-based genotoxicity endpoints-micronucleated reticulocyte frequency and gene mutation at the Pig-a locus-thereby making combination and integration studies practical. The ability to effectively monitor these endpoints in short-term and repeated dosing schedules was investigated with the carcinogen/noncarcinogen pair benzo(a)pyrene (BP) and pyrene (Pyr). Male Sprague-Dawley rats were treated via oral gavage for 3 or 28 consecutive days with several dose levels of Pyr, including maximum tolerated doses. BP exposure was administered by the same route but at one dose level, 250 or 125 mg/kg/day for 3-day and 28-day studies, respectively. Serial blood samples were collected up to Day 45, and were analyzed for Pig-a mutation with a dual labeling method (SYTO 13 in combination with anti-CD59-PE) that facilitated mutant cell frequency measurements in both total erythrocytes and the reticulocyte subpopulation. A mutant cell enrichment step based on immunomagnetic column separation was used to increase the statistical power of the assay. BP induced robust mutant reticulocyte responses by Day 15, and elevated frequencies persisted until study termination. Mutant erythrocyte responses lagged mutant reticulocyte responses, with peak incidences observed on Day 30 of the 3-day study (43-fold increase) and on Day 42 of the 28-day study (171-fold increase). No mutagenic effects were apparent for Pyr. Blood samples collected on Day 4, and Day 29 for the 28-day study, were evaluated for micronucleated reticulocyte frequency. Significant increases in micronucleus frequencies were observed with BP, whereas Pyr had no effect. These results demonstrate that Pig-a and micronucleus endpoints discriminate between these structurally related carcinogenic and noncarcinogenic agents. Furthermore, the high sensitivity demonstrated with the enrichment protocol indicates that the Pig-a endpoint is suitable for both repeated-dose and acute studies, allowing integration of mutagenic and clastogenic endpoints into on-going toxicology studies, and use as a short-term assay that provides efficient screening and mechanistic information in vivo.

Tandem optimization of target activity and elimination of mutagenic potential in a potent series of N-aryl bicyclic hydantoin-based selective androgen receptor modulators.

Pharmacokinetic studies in cynomolgus monkeys with a novel prototype selective androgen receptor modulator revealed trace amounts of an aniline fragment released through hydrolytic metabolism. This aniline fragment was determined to be mutagenic in an Ames assay. Subsequent concurrent optimization for target activity and avoidance of mutagenicity led to the identification of a pharmacologically superior clinical candidate without mutagenic potential.

Probabilistic neural network multiple classifier system for predicting the genotoxicity of quinolone and quinoline derivatives.

Quinolone and quinoline are known to be liver carcinogens in rodents, and a number of their derivatives have been shown to exhibit mutagenicity in the Ames test, using Salmonella typhimurium strain TA 100 in the presence of S9. Both the carcinogenicity and the mutagenicity of quinolone and quinoline derivatives, as determined by SAS, can be attributed to their genotoxicity potential. This potential, which is measured by genotoxicity tests, is a good indication of carcinogenicity and mutagenicity because compounds that are positive in these tests have the potential to be human carcinogens and/or mutagens. In this study, a collection of quinolone and quinoline derivatives' carcinogenicity is determined by qualitatively predicting their genotoxicity potential with predictive PNN (probabilistic neural network) classification models. In addition, a multiple classifier system is also developed to improve the predictability of genotoxicity. Superior results are seen with the multiple classifier system over the individual PNN classification models. With the multiple classifier system, 89.4% of the quinolone derivatives were predicted correctly, and higher predictability is seen with the quinoline derivatives at 92.2% correct. The multiple classifier system not only is able to accurately predict the genotoxicity but also provides an insight about the main determinants of genotoxicity of the quinolone and quinoline derivatives. Thus, the PNN multiple classifier system generated in this study is a beneficial contributor toward predictive toxicology in the design of less carcinogenic bioactive compounds.

Predicting the genotoxicity of polycyclic aromatic compounds from molecular structure with different classifiers.

Classification models were developed to provide accurate prediction of genotoxicity of 277 polycyclic aromatic compounds (PACs) directly from their molecular structures. Numerical descriptors encoding the topological, geometric, electronic, and polar surface area properties of the compounds were calculated to represent the structural information. Each compound's genotoxicity was represented with IMAX (maximal SOS induction factor) values measured by the SOS Chromotest in the presence and absence of S9 rat liver homogenate. The compounds' class identity was determined by a cutoff IMAX value of 1.25-compounds with IMAX > 1.25 in either test were classified as genotoxic, and the ones with IMAX < or = 1.25 were nongenotoxic. Several binary classification models were generated to predict genotoxicity: k-nearest neighbor (k-NN), linear discriminant analysis, and probabilistic neural network. The study showed k-NN to provide the highest predictive ability among the three classifiers with a training set classification rate of 93.5%. A consensus model was also developed that incorporated the three classifiers and correctly predicted 81.2% of the 277 compounds. It also provided a higher prediction rate on the genotoxic class than any other single model.

Predicting the genotoxicity of thiophene derivatives from molecular structure.

We report several binary classification models that directly link the genetic toxicity of a series of 140 thiophene derivatives with information derived from the compounds' molecular structure. Genetic toxicity was measured using an SOS Chromotest. IMAX (maximal SOS induction factor) values were recorded for each of the 140 compounds both in the presence and in the absence of S9 rat liver homogenate. Compounds were classified as genotoxic if IMAX >or= 1.5 in either test or nongenotoxic if IMAX < 1.5 for both tests. The molecular structures were represented by numerical descriptors that encoded the topological, geometric, electronic, and polar surface area properties of the thiophene derivatives. The classification models used were linear discriminant analysis (LDA), k-nearest neighbor classification (k-NN), and the probabilistic neural network (PNN). These were used in conjunction with either a genetic algorithm or a generalized simulated annealing to find optimal subsets of descriptors for each classifier. The quality of the resulting models was determined by the number of misclassified compounds, with preference given to models that produced fewer false negative classifications. Model sizes ranged from seven descriptors for LDA to three descriptors for k-NN and PNN. Very good classification results were obtained with all three classifiers. Classification rates for the LDA, k-NN, and PNN models were 80, 85, and 85%, respectively, for the prediction set compounds. Additionally, a consensus model was generated that incorporated all three of the basic model types. This consensus model correctly predicted the genotoxicity of 95% of the prediction set compounds.

Predicting the genotoxicity of secondary and aromatic amines using data subsetting to generate a model ensemble.

Binary quantitative structure-activity relationship (QSAR) models are developed to classify a data set of 334 aromatic and secondary amine compounds as genotoxic or nongenotoxic based on information calculated solely from chemical structure. Genotoxic endpoints for each compound were determined using the SOS Chromotest in both the presence and absence of an S9 rat liver homogenate. Compounds were considered genotoxic if assay results indicated a positive genotoxicity hit for either the S9 inactivated or S9 activated assay. Each compound in the data set was encoded through the calculation of numerical descriptors that describe various aspects of chemical structure (e.g. topological, geometric, electronic, polar surface area). Furthermore, five additional descriptors that focused on the secondary and aromatic nitrogen atoms in each molecule were calculated specifically for this study. Descriptor subsets were examined using a genetic algorithm search engine interfaced with a k-Nearest Neighbor fitness evaluator to find the most information-rich subsets, which ultimately served as the final predictive models. Models were chosen for their ability to minimize the total number of misclassifications, with special attention given to those models that possessed fewer occurrences of positive toxicity hits being misclassified as nontoxic (false negatives). In addition, a subsetting procedure was used to form an ensemble of models using different combinations of compounds in the training and prediction sets. This was done to ensure that consistent results could be obtained regardless of training set composition. The procedure also allowed for each compound to be externally validated three times by different training set data with the resultant predictions being used in a "majority rules" voting scheme to produce a consensus prediction for each member of the data set. The individual models produced an average training set classification rate of 71.6% and an average prediction set classification rate of 67.7%. However, the model ensemble was able to correctly classify the genotoxicity of 72.2% of all prediction set compounds.