Analysis of Non-Mutagenic Substances in the Context of Drug Impurity Assessment - Few are Potent Toxicants.

ICH Q3A/B guidelines provide qualification thresholds for impurities or degradation products in new drug substances and products. However, the guidelines note that certain impurities / degradation products may warrant further safety evaluation for being unusually potent or toxic. The purpose of this study was to confirm that especially toxic non-mutagenic compounds are rare and to identify classes of compounds that could warrant lower qualification thresholds. A total of 2,815 compounds were evaluated, of which 2,213 were assessed as non-mutagenic. For the purpose of this analysis, compounds were considered potent when the point of departure was ≤ 0.2 mg/kg/day based on the qualification threshold (1 mg/day or 0.02 mg/kg/day for a 50 kg human) in a new drug substance, with an additional 10-fold margin. Only 54 of the entire set (2.4%) would be considered potent based on this conservative potency analysis, confirming that the existing ICH Q3A/B qualification thresholds are appropriate for the majority of impurities. If the Q3A/B threshold, without the additional 10-fold margin is used, 14 compounds (0.6%) are considered "highly potent". Very few non-mutagenic structural classes were identified, including organothiophosphates and derivatives, polychlorinated benzenes and polychlorinated polycyclic aliphatics, that correlate with potential high potency, consistent with prior publications.

Discovery of TRPA1 Antagonist GDC-6599: Derisking Preclinical Toxicity and Aldehyde Oxidase Metabolism with a Potential First-in-Class Therapy for Respiratory Disease.

Transient receptor potential ankyrin 1 (TRPA1) is a nonselective calcium ion channel highly expressed in the primary sensory neurons, functioning as a polymodal sensor for exogenous and endogenous stimuli, and has been implicated in neuropathic pain and respiratory disease. Herein, we describe the optimization of potent, selective, and orally bioavailable TRPA1 small molecule antagonists with strong in vivo target engagement in rodent models. Several lead molecules in preclinical single- and short-term repeat-dose toxicity studies exhibited profound prolongation of coagulation parameters. Based on a thorough investigative toxicology and clinical pathology analysis, anticoagulation effects in vivo are hypothesized to be manifested by a metabolite─generated by aldehyde oxidase (AO)─possessing a similar pharmacophore to known anticoagulants (i.e., coumarins, indandiones). Further optimization to block AO-mediated metabolism yielded compounds that ameliorated coagulation effects in vivo, resulting in the discovery and advancement of clinical candidate GDC-6599, currently in Phase II clinical trials for respiratory indications.

Artificial Intelligence for Drug Discovery: Are We There Yet?

Drug discovery is adapting to novel technologies such as data science, informatics, and artificial intelligence (AI) to accelerate effective treatment development while reducing costs and animal experiments. AI is transforming drug discovery, as indicated by increasing interest from investors, industrial and academic scientists, and legislators. Successful drug discovery requires optimizing properties related to pharmacodynamics, pharmacokinetics, and clinical outcomes. This review discusses the use of AI in the three pillars of drug discovery: diseases, targets, and therapeutic modalities, with a focus on small-molecule drugs. AI technologies, such as generative chemistry, machine learning, and multiproperty optimization, have enabled several compounds to enter clinical trials. The scientific community must carefully vet known information to address the reproducibility crisis. The full potential of AI in drug discovery can only be realized with sufficient ground truth and appropriate human intervention at later pipeline stages.

Development of Duration-Based Non-Mutagenic Thresholds of Toxicological Concern (TTCs) Relevant to Parenteral Extractables and Leachables (E&Ls).

The threshold of toxicological concern (TTC), i.e., the dose of a compound lacking sufficient experimental toxicity data that is unlikely to result in an adverse health effect in humans, is important for evaluating extractables and leachables (E&Ls) as it guides analytical testing and minimizes the use of animal studies. The Extractables and Leachables Safety Information Exchange (ELSIE) consortium, which consists of member companies that span biotechnology, pharmaceutical, and medical device industries, brought together subject matter expert toxicologists to derive TTC values for organic, non-mutagenic E&L substances when administered parenterally. A total of 488 E&L compounds from the ELSIE database were analyzed and parenteral point of departure (PPOD) estimates were derived for 252 compounds. The PPOD estimates were adjusted to extrapolate to subacute, subchronic, and chronic durations of nonclinical exposure and the lower fifth percentiles were calculated. An additional 100-fold adjustment factor to account for nonclinical species and human variability was subsequently applied to derive the parenteral TTC values for E&Ls. The resulting parenteral TTC values are 35, 110, and 180 µg/day for human exposures of >10 years to lifetime, >1-10 years, and ≤1 year, respectively. These parenteral TTCs are expected to be conservative for E&Ls that are considered non-mutagenic per ICH M7(R1) guidelines.

Toward Quantitative Models in Safety Assessment: A Case Study to Show Impact of Dose-Response Inference on hERG Inhibition Models.

Due to challenges with historical data and the diversity of assay formats, in silico models for safety-related endpoints are often based on discretized data instead of the data on a natural continuous scale. Models for discretized endpoints have limitations in usage and interpretation that can impact compound design. Here, we present a consistent data inference approach, exemplified on two data sets of Ether-à-go-go-Related Gene (hERG) K+ inhibition data, for dose-response and screening experiments that are generally applicable for in vitro assays. hERG inhibition has been associated with severe cardiac effects and is one of the more prominent safety targets assessed in drug development, using a wide array of in vitro and in silico screening methods. In this study, the IC50 for hERG inhibition is estimated from diverse historical proprietary data. The IC50 derived from a two-point proprietary screening data set demonstrated high correlation (R = 0.98, MAE = 0.08) with IC50s derived from six-point dose-response curves. Similar IC50 estimation accuracy was obtained on a public thallium flux assay data set (R = 0.90, MAE = 0.2). The IC50 data were used to develop a robust quantitative model. The model's MAE (0.47) and R2 (0.46) were on par with literature statistics and approached assay reproducibility. Using a continuous model has high value for pharmaceutical projects, as it enables rank ordering of compounds and evaluation of compounds against project-specific inhibition thresholds. This data inference approach can be widely applicable to assays with quantitative readouts and has the potential to impact experimental design and improve model performance, interpretation, and acceptance across many standard safety endpoints.

A cross-industry collaboration to assess if acute oral toxicity (Q)SAR models are fit-for-purpose for GHS classification and labelling.

This study assesses whether currently available acute oral toxicity (AOT) in silico models, provided by the widely employed Leadscope software, are fit-for-purpose for categorization and labelling of chemicals. As part of this study, a large data set of proprietary and marketed compounds from multiple companies (pharmaceutical, plant protection products, and other chemical industries) was assembled to assess the models' performance. The absolute percentage of correct or more conservative predictions, based on a comparison of experimental and predicted GHS categories, was approximately 95%, after excluding a small percentage of inconclusive (indeterminate or out of domain) predictions. Since the frequency distribution across the experimental categories is skewed towards low toxicity chemicals, a balanced assessment was also performed. Across all compounds which could be assigned to a well-defined experimental category, the average percentage of correct or more conservative predictions was around 80%. These results indicate the potential for reliable and broad application of these models across different industrial sectors. This manuscript describes the evaluation of these models, highlights the importance of an expert review, and provides guidance on the use of AOT models to fulfill testing requirements, GHS classification/labelling, and transportation needs.

Management of pharmaceutical ICH M7 (Q)SAR predictions - The impact of model updates.

Pharmaceutical applicants conduct (Q)SAR assessments on identified and theoretical impurities to predict their mutagenic potential. Two complementary models-one rule-based and one statistical-based-are used, followed by expert review. (Q)SAR models are continuously updated to improve predictions, with new versions typically released on a yearly basis. Numerous releases of (Q)SAR models will occur during the typical 6-7 years of drug development until new drug registration. Therefore, it is important to understand the impact of model updates on impurity mutagenicity predictions over time. Compounds representative of pharmaceutical impurities were analyzed with three rule- and three statistical-based models covering a 4-8 year period, with the individual time frame being dependent on when the individual models were initially made available. The largest changes in the combined outcome of two complementary models were from positive or equivocal to negative and from negative to equivocal. Importantly, the cumulative change of negative to positive predictions was small in all models (<5%) and was further reduced when complementary models were combined in a consensus fashion. We conclude that model updates of the type evaluated in this manuscript would not necessarily require re-running a (Q)SAR prediction unless there is a specific need. However, original (Q)SAR predictions should be evaluated when finalizing the commercial route of synthesis for marketing authorization.

Skin sensitization in silico protocol.

The assessment of skin sensitization has evolved over the past few years to include in vitro assessments of key events along the adverse outcome pathway and opportunistically capitalize on the strengths of in silico methods to support a weight of evidence assessment without conducting a test in animals. While in silico methods vary greatly in their purpose and format; there is a need to standardize the underlying principles on which such models are developed and to make transparent the implications for the uncertainty in the overall assessment. In this contribution, the relationship between skin sensitization relevant effects, mechanisms, and endpoints are built into a hazard assessment framework. Based on the relevance of the mechanisms and effects as well as the strengths and limitations of the experimental systems used to identify them, rules and principles are defined for deriving skin sensitization in silico assessments. Further, the assignments of reliability and confidence scores that reflect the overall strength of the assessment are discussed. This skin sensitization protocol supports the implementation and acceptance of in silico approaches for the prediction of skin sensitization.

Genetic toxicology in silico protocol.

In silico toxicology (IST) approaches to rapidly assess chemical hazard, and usage of such methods is increasing in all applications but especially for regulatory submissions, such as for assessing chemicals under REACH as well as the ICH M7 guideline for drug impurities. There are a number of obstacles to performing an IST assessment, including uncertainty in how such an assessment and associated expert review should be performed or what is fit for purpose, as well as a lack of confidence that the results will be accepted by colleagues, collaborators and regulatory authorities. To address this, a project to develop a series of IST protocols for different hazard endpoints has been initiated and this paper describes the genetic toxicity in silico (GIST) protocol. The protocol outlines a hazard assessment framework including key effects/mechanisms and their relationships to endpoints such as gene mutation and clastogenicity. IST models and data are reviewed that support the assessment of these effects/mechanisms along with defined approaches for combining the information and evaluating the confidence in the assessment. This protocol has been developed through a consortium of toxicologists, computational scientists, and regulatory scientists across several industries to support the implementation and acceptance of in silico approaches.

Extending (Q)SARs to incorporate proprietary knowledge for regulatory purposes: is aromatic N-oxide a structural alert for predicting DNA-reactive mutagenicity?

(Quantitative) structure-activity relationship or (Q)SAR predictions of DNA-reactive mutagenicity are important to support both the design of new chemicals and the assessment of impurities, degradants, metabolites, extractables and leachables, as well as existing chemicals. Aromatic N-oxides represent a class of compounds that are often considered alerting for mutagenicity yet the scientific rationale of this structural alert is not clear and has been questioned. Because aromatic N-oxide-containing compounds may be encountered as impurities, degradants and metabolites, it is important to accurately predict mutagenicity of this chemical class. This article analysed a series of publicly available aromatic N-oxide data in search of supporting information. The article also used a previously developed structure-activity relationship (SAR) fingerprint methodology where a series of aromatic N-oxide substructures was generated and matched against public and proprietary databases, including pharmaceutical data. An assessment of the number of mutagenic and non-mutagenic compounds matching each substructure across all sources was used to understand whether the general class or any specific subclasses appear to lead to mutagenicity. This analysis resulted in a downgrade of the general aromatic N-oxide alert. However, it was determined there were enough public and proprietary data to assign the quindioxin and related chemicals as well as benzo[c][1,2,5]oxadiazole 1-oxide subclasses as alerts. The overall results of this analysis were incorporated into Leadscope's expert-rule-based model to enhance its predictive accuracy.

Principles and procedures for handling out-of-domain and indeterminate results as part of ICH M7 recommended (Q)SAR analyses.

The International Council for Harmonization (ICH) M7 guideline describes a hazard assessment process for impurities that have the potential to be present in a drug substance or drug product. In the absence of adequate experimental bacterial mutagenicity data, (Q)SAR analysis may be used as a test to predict impurities' DNA reactive (mutagenic) potential. However, in certain situations, (Q)SAR software is unable to generate a positive or negative prediction either because of conflicting information or because the impurity is outside the applicability domain of the model. Such results present challenges in generating an overall mutagenicity prediction and highlight the importance of performing a thorough expert review. The following paper reviews pharmaceutical and regulatory experiences handling such situations. The paper also presents an analysis of proprietary data to help understand the likelihood of misclassifying a mutagenic impurity as non-mutagenic based on different combinations of (Q)SAR results. This information may be taken into consideration when supporting the (Q)SAR results with an expert review, especially when out-of-domain results are generated during a (Q)SAR evaluation.

Computational Toxicology and Drug Discovery.

The use of computational toxicology methods within drug discovery began in the early 2000s with applications such as predicting bacterial mutagenicity and hERG inhibition. The field has been continuously expanding ever since and the tasks at hand have become more complex. These approaches are now strategically integrated into the risk assessment process, as a complement to in vitro and in vivo methods. Today, computational toxicology can be used in every phase of drug discovery and development, from profiling large libraries early on, to predicting off-target effects in the mid-discovery phase, to assessing potential mutagenic impurities in development and degradants as part of life-cycle management. This chapter provides an overview of the field and describes the application of computational toxicology throughout the entire discovery and development process.

Principles and procedures for implementation of ICH M7 recommended (Q)SAR analyses.

The ICH M7 guideline describes a consistent approach to identify, categorize, and control DNA reactive, mutagenic, impurities in pharmaceutical products to limit the potential carcinogenic risk related to such impurities. This paper outlines a series of principles and procedures to consider when generating (Q)SAR assessments aligned with the ICH M7 guideline to be included in a regulatory submission. In the absence of adequate experimental data, the results from two complementary (Q)SAR methodologies may be combined to support an initial hazard classification. This may be followed by an assessment of additional information that serves as the basis for an expert review to support or refute the predictions. This paper elucidates scenarios where additional expert knowledge may be beneficial, what such an expert review may contain, and how the results and accompanying considerations may be documented. Furthermore, the use of these principles and procedures to yield a consistent and robust (Q)SAR-based argument to support impurity qualification for regulatory purposes is described in this manuscript.

Synchronous down-modulation of miR-17 family members is an early causative event in the retinal angiogenic switch.

Six members of the microRNA-17 (miR-17) family were mapped to three different chromosomes, although they share the same seed sequence and are predicted to target common genes, among which are those encoding hypoxia-inducible factor-1α (HIF1A) and VEGFA. Here, we evaluated the in vivo expression profile of the miR-17 family in the murine retinopathy of prematurity (ROP) model, whereby Vegfa expression is highly enhanced at the early stage of retinal neovascularization, and we found simultaneous reduction of all miR-17 family members at this stage. Using gene reporter assays, we observed binding of these miRs to specific sites in the 3' UTRs of Hif1a and Vegfa. Furthermore, overexpression of these miRs decreased HIF1A and VEGFA expression in vitro. Our data indicate that this miR-17 family elicits a regulatory synergistic down-regulation of Hif1a and Vegfa expression in this biological model. We propose the existence of a coordinated regulatory network, in which diverse miRs are synchronously regulated to target the Hif1a transcription factor, which in turn, potentiates and reinforces the regulatory effects of the miRs on Vegfa to trigger and sustain a significant physiological response.

Assessment of machine learning reliability methods for quantifying the applicability domain of QSAR regression models.

The vastness of chemical space and the relatively small coverage by experimental data recording molecular properties require us to identify subspaces, or domains, for which we can confidently apply QSAR models. The prediction of QSAR models in these domains is reliable, and potential subsequent investigations of such compounds would find that the predictions closely match the experimental values. Standard approaches in QSAR assume that predictions are more reliable for compounds that are "similar" to those in subspaces with denser experimental data. Here, we report on a study of an alternative set of techniques recently proposed in the machine learning community. These methods quantify prediction confidence through estimation of the prediction error at the point of interest. Our study includes 20 public QSAR data sets with continuous response and assesses the quality of 10 reliability scoring methods by observing their correlation with prediction error. We show that these new alternative approaches can outperform standard reliability scores that rely only on similarity to compounds in the training set. The results also indicate that the quality of reliability scoring methods is sensitive to data set characteristics and to the regression method used in QSAR. We demonstrate that at the cost of increased computational complexity these dependencies can be leveraged by integration of scores from various reliability estimation approaches. The reliability estimation techniques described in this paper have been implemented in an open source add-on package ( https://bitbucket.org/biolab/orange-reliability ) to the Orange data mining suite.

Localized heuristic inverse quantitative structure activity relationship with bulk descriptors using numerical gradients.

State-of-the-art quantitative structure-activity relationship (QSAR) models are often based on nonlinear machine learning algorithms, which are difficult to interpret. From a pharmaceutical perspective, QSARs are used to enhance the chemical design process. Ultimately, they should not only provide a prediction but also contribute to a mechanistic understanding and guide modifications to the chemical structure, promoting compounds with desirable biological activity profiles. Global ranking of descriptor importance and inverse QSAR have been used for these purposes. This paper introduces localized heuristic inverse QSAR, which provides an assessment of the relative ability of the descriptors to influence the biological response in an area localized around the predicted compound. The method is based on numerical gradients with parameters optimized using data sets sampled from analytical functions. The heuristic character of the method reduces the computational requirements and makes it applicable not only to fragment based methods but also to QSARs based on bulk descriptors. The application of the method is illustrated on congeneric QSAR data sets, and it is shown that the predicted influential descriptors can be used to guide structural modifications that affect the biological response in the desired direction. The method is implemented into the AZOrange Open Source QSAR package. The current implementation of localized heuristic inverse QSAR is a step toward a generally applicable method for elucidating the structure activity relationship specifically for a congeneric region of chemical space when using QSARs based on bulk properties. Consequently, this method could contribute to accelerating the chemical design process in pharmaceutical projects, as well as provide information that could enhance the mechanistic understanding for individual scaffolds.

In silico methods combined with expert knowledge rule out mutagenic potential of pharmaceutical impurities: an industry survey.

With the increasing emphasis on identification and low level control of potentially genotoxic impurities (GTIs), there has been increased use of structure-based assessments including application of computerized models. To date many publications have focused on the ability of computational models, either individually or in combination, to accurately predict the mutagenic effects of a chemical in the Ames assay. Typically, these investigations take large numbers of compounds and use in silico tools to predict their activity with no human interpretation being made. However, this does not reflect how these assessments are conducted in practice across the pharmaceutical industry. Current guidelines indicate that a structural assessment is sufficient to conclude that an impurity is non-mutagenic. To assess how confident we can be in identifying non-mutagenic structures, eight companies were surveyed for their success rate. The Negative Predictive Value (NPV) of the in silico approaches was 94%. When human interpretation of in silico model predictions was conducted, the NPV increased substantially to 99%. The survey illustrates the importance of expert interpretation of in silico predictions. The survey also suggests the use of multiple computational models is not a significant factor in the success of these approaches with respect to NPV.

Explanation for main features of structure-genotoxicity relationships of aromatic amines by theoretical studies of their activation pathways in CYP1A2.

Aromatic and heteroaromatic amines (ArNH(2)) represent a class of potential mutagens that after being metabolically activated covalently modify DNA. Activation of ArNH(2) in many cases starts with N-hydroxylation by P450 enzymes, primarily CYP1A2. Poor understanding of structure-mutagenicity relationships of ArNH(2) limits their use in drug discovery programs. Key factors that facilitate activation of ArNH(2) are revealed by exploring their reaction intermediates in CYP1A2 using DFT calculations. On the basis of these calculations and extensive analysis of structure-mutagenicity data, we suggest that mutagenic metabolites are generated by ferric peroxo intermediate, (CYP1A2)Fe(III)-OO(-), in a three-step heterolytic mechanism. First, the distal oxygen of the oxidant abstracts proton from H-bonded ArNH(2). The subsequent proximal protonation of the resulting (CYP1A2)Fe(III)-OOH weakens both the O-O and the O-H bonds of the oxidant. Heterolytic cleavage of the O-O bond leads to N-hydroxylation of ArNH(-) via S(N)2 mechanism, whereas cleavage of the O-H bond results in release of hydroperoxy radical. Thus, our proposed reaction offers a mechanistic explanation for previous observations that metabolism of aromatic amines could cause oxidative stress. The primary drivers for mutagenic potency of ArNH(2) are (i) binding affinity of ArNH(2) in the productive binding mode within the CYP1A2 substrate cavity, (ii) resonance stabilization of the anionic forms of ArNH(2), and (iii) exothermicity of proton-assisted heterolytic cleavage of N-O bonds of hydroxylamines and their bioconjugates. This leads to a strategy for designing mutagenicity free ArNH(2): Structural alterations in ArNH(2), which disrupt geometric compatibility with CYP1A2, hinder proton abstraction, or strongly destabilize the nitrenium ion, in this order of priority, prevent genotoxicity.