The importance of expert review to clarify ambiguous situations for (Q)SAR predictions under ICH M7

Advances in Activity/Property Prediction from Chemical Structures

Recent technological advancement in AI modeling of molecular property databases has significantly expanded the opportunities for drug design and development. Quantitative structure-activity relationships (QSARs) are shown to provide more accurate predictions with regards to biological activity as well as toxicological assessment. By using a combination of in-silico models or by combining disparate structure-activity databases, researchers have been able to improve accuracy for a variety of drug discovery and analysis methods, generating viable compounds, which in certain cases, can be synthesized and further studied in vitro to find candidates for potential development. Additionally, the development of compounds of determined toxicology can be discontinued earlier, allowing alternative routes to be evaluated, preventing wasted time and resources. Although the progress that has been made is tremendous, expert review is still necessary for most in-silico generated predictions. Regardless, the scientific community continues to move ever closer to completely automated drug discovery and evaluation.

LC-HRMS and NMR studies for the characterization of degradation impurities of ubrogepant along with the in silico approaches for the prediction of degradation and toxicity

Ubrogepant is the first oral calcitonin gene-related peptide (CGRP) receptor antagonist which is used for the acute treatment of migraine in adults. The present study employs liquid chromatography-high resolution mass spectrometry (LC-HRMS) and nuclear magnetic resonance spectroscopy (NMR) techniques for the identification and characterization of degradation impurities of ubrogepant. The forced degradation study of ubrogepant was performed as per the International Council for Harmonisation (ICH) Q1A and Q1B guidelines. The in silico degradation profile of ubrogepant was predicted by Zeneth. It was observed that ubrogepant was labile to acidic hydrolysis, basic hydrolysis, and oxidative degradation conditions (H2O2), although it was stable in neutral hydrolysis and photolytic (UV light and visible light) conditions. Eight degradation impurities were formed, which were separated on reversed-phase HPLC with a gradient program on an InertSustain C8 column (4.6 × 250 mm, 5 µm) using 10 mM ammonium formate (pH unadjusted) and acetonitrile as the mobile phase. The structures of all the degradation impurities were characterized using the exact masses obtained from the HRMS/MS. Further, NMR studies were conducted on two major degradation impurities (UB-4 and UB-7). A plausible mechanism was proposed to support the structures of all the degradation impurities of UBR. In silico toxicity and mutagenicity assessment were done by DEREK Nexus, SARAH Nexus, and ProTox-II.

In vivo and in vitro metabolite profiling of nirmatrelvir using LC-Q-ToF-MS/MS along with the in silico approaches for prediction of metabolites and their toxicity

Nirmatrelvir (NRV), a 3C-like protease or Mpro inhibitor of SARS-CoV-2, is used for the treatment of COVID-19 in adult and paediatric patients. The present study was accomplished to investigate the comprehensive metabolic fate of NRV using in vitro and in vivo models. The in vitro models used for the study were microsomes (human liver microsomes, rat liver microsomes, mouse liver microsomes) and S9 fractions (human liver S9 fractions and rat liver S9 fractions) with the appropriate cofactors, whereas Sprague-Dawley rats were used as the in vivo models. Nirmatrelvir was administered orally to Sprague-Dawley rats, which was followed by the collection of urine, faeces and blood at pre-determined time intervals. Protein precipitation was used as the sample preparation method for all the samples. The samples were then analysed by liquid chromatography-quadrupole time-of-flight tandem mass spectrometry (LC-Q-ToF-MS/MS) using an Acquity BEH C18 column with 0.1% formic acid and acetonitrile as the mobile phase. Four metabolites were found to be novel, which were formed via amide hydrolysis, oxidation and hydroxylation. Furthermore, an in silico analysis was performed using Meteor Nexus software to predict the probable metabolic changes of NRV. The toxicity and mutagenicity of NRV and its metabolites were also determined using DEREK Nexus and SARAH Nexus.

Prioritization of mycotoxins based on mutagenicity and carcinogenicity evaluation using combined in silico QSAR methods

Mycotoxins and their metabolites are a family of compounds that contains a great diversity of both structure and biological properties. Information on their toxicity is spread within several databases and in scientific literature. Due to the number of molecules and their structure diversity, the cost and time required for hazard evaluation of each compound is unrealistic. In that purpose, new approach methodologies (NAMs) can be applied to evaluate such large set of molecules. Among them, quantitative structure-activity relationship (QSAR) in silico models could be useful to predict the mutagenic and carcinogenic properties of mycotoxins. First, a complete list of 904 mycotoxins and metabolites was built. Then, some known mycotoxins were used to determine the best QSAR tools for mutagenicity and carcinogenicity predictions. The best tool was further applied to the whole list of 904 mycotoxins. At the end, 95 mycotoxins were identified as both mutagen and carcinogen and should be prioritized for further evaluation.

Electrochemically Enhanced Delivery of Pemetrexed from Electroactive Hydrogels

Electroactive hydrogels based on derivatives of polyethyleneglycol (PEG), chitosan and polypyrrole were prepared via a combination of photopolymerization and oxidative chemical polymerization, and optionally doped with anions (e.g., lignin, drugs, etc.). The products were analyzed with a variety of techniques, including: FT-IR, UV-Vis, ¹H NMR (solution state), ¹³C NMR (solid state), XRD, TGA, SEM, swelling ratios and rheology. The conductive gels swell ca. 8 times less than the non-conductive gels due to the presence of the interpenetrating network (IPN) of polypyrrole and lignin. A rheological study showed that the non-conductive gels are soft (G' 0.35 kPa, G″ 0.02 kPa) with properties analogous to brain tissue, whereas the conductive gels are significantly stronger (G' 30 kPa, G″ 19 kPa) analogous to breast tissue due to the presence of the IPN of polypyrrole and lignin. The potential of these biomaterials to be used for biomedical applications was validated in vitro by cell culture studies (assessing adhesion and proliferation of fibroblasts) and drug delivery studies (electrochemically loading the FDA-approved chemotherapeutic pemetrexed and measuring passive and stimulated release); indeed, the application of electrical stimulus enhanced the release of PEM from gels by ca. 10-15% relative to the passive release control experiment for each application of electrical stimulation over a short period analogous to the duration of stimulation applied for electrochemotherapy. It is foreseeable that such materials could be integrated in electrochemotherapeutic medical devices, e.g., electrode arrays or plates currently used in the clinic.

Combining DELs and machine learning for toxicology prediction

DNA-encoded libraries (DELs) allow starting chemical matter to be identified in drug discovery. The volume of experimental data generated also makes DELs an attractive resource for machine learning (ML). ML allows modeling complex relationships between compounds and numerical endpoints, such as the binding to a target measured by DELs. DELs could also empower other areas of drug discovery. Here, we propose that DELs and ML could be combined to model binding to off-targets, enabling better predictive toxicology. With enough data, ML models can make accurate predictions across a vast chemical space, and they can be reused and expanded across projects. Although there are limitations, more general toxicology models could be applied earlier during drug discovery, illuminating safety liabilities at a lower cost.

Instructive electroactive electrospun silk fibroin-based biomaterials for peripheral nerve tissue engineering

Aligned sub-micron fibres are an outstanding surface for orienting and promoting neurite outgrowth; therefore, attractive features to include in peripheral nerve tissue scaffolds. A new generation of peripheral nerve tissue scaffolds is under development incorporating electroactive materials and electrical regimes as instructive cues in order to facilitate fully functional regeneration. Herein, electroactive fibres composed of silk fibroin (SF) and poly(3,4-ethylenedioxythiophene):polystyrene sulfonate (PEDOT:PSS) were developed as a novel peripheral nerve tissue scaffold. Mats of SF with sub-micron fibre diameters of 190 ± 50 nm were fabricated by double layer electrospinning with thicknesses of ∼100 μm (∼70-80 μm random fibres and ∼20-30 μm aligned fibres). Electrospun SF mats were modified with interpenetrating polymer networks (IPN) of PEDOT:PSS in various ratios of PSS/EDOT (α) and the polymerisation was assessed by hard X-ray photoelectron spectroscopy (HAXPES). The mechanical properties of electrospun SF and IPNs mats were characterised in the wet state tensile and the electrical properties were examined by cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS). The cytotoxicity and biocompatibility of the optimal IPNs (α = 2.3 and 3.3) mats were ascertained via the growth and neurite extension of mouse neuroblastoma x rat glioma hybrid cells (NG108-15) for 7 days. The longest neurite outgrowth of 300 μm was observed in the parallel direction of fibre alignment on laminin-coated electrospun SF and IPN (α = 2.3) mats which is the material with the lowest electron transfer resistance (R_et, ca. 330 Ω). These electrically conductive composites with aligned sub-micron fibres exhibit promise for axon guidance and also have the potential to be combined with electrical stimulation treatment as a further step for the effective regeneration of nerves.

Use of Lhasa Limited Products for the In Silico Prediction of Drug Toxicity

Lhasa Limited have had a role in the in silico prediction of drug and other chemical toxicity for over 30 years. This role has always been multifaceted, both as a provider of predictive software such as Derek Nexus, and as an honest broker for the sharing of proprietary chemical and toxicity data. A changing regulatory environment and the drive for the Replacement, Reduction and Refinement (the 3Rs) of animal testing have led both to increased acceptance of in silico predictions and a desire for the sharing of data to reduce duplicate testing. The combination of these factors has led to Lhasa Limited providing a suite of products and coordinating numerous data-sharing consortia that do indeed facilitate a significant reduction in the testing burden that companies would otherwise be laboring under. Many of these products and consortia can be organized into workflows for specific regulatory use cases, and it is these that will be used to frame the narrative in this chapter.

Bacterial mutagenicity test data: collection by the task force of the Japan pharmaceutical manufacturers association

Background

Ames test is used worldwide for detecting the bacterial mutagenicity of chemicals. In silico analyses of bacterial mutagenicity have recently gained acceptance by regulatory agencies; however, current in silico models for prediction remain to be improved. The Japan Pharmaceutical Manufacturers Association (JPMA) organized a task force in 2017 in which eight Japanese pharmaceutical companies had participated. The purpose of this task force was to disclose a piece of pharmaceutical companies’ proprietary Ames test data.

Results

Ames test data for 99 chemicals of various chemical classes were collected for disclosure in this study. These chemicals are related to the manufacturing process of pharmaceutical drugs, including reagents, synthetic intermediates, and drug substances. The structure-activity (mutagenicity) relationships are discussed in relation to structural alerts for each chemical class. In addition, in silico analyses of these chemicals were conducted using a knowledge-based model of Derek Nexus (Derek) and a statistics-based model (GT1_BMUT module) of CASE Ultra. To calculate the effectiveness of these models, 89 chemicals for Derek and 54 chemicals for CASE Ultra were selected; major exclusions were the salt form of four chemicals that were tested both in the salt and free forms for both models, and 35 chemicals called “known” positives or negatives for CASE Ultra. For Derek, the sensitivity, specificity, and accuracy were 65% (15/23), 71% (47/66), and 70% (62/89), respectively. The sensitivity, specificity, and accuracy were 50% (6/12), 60% (25/42), and 57% (31/54) for CASE Ultra, respectively. The ratio of overall disagreement between the CASE Ultra “known” positives/negatives and the actual test results was 11% (4/35). In this study, 19 out of 28 mutagens (68%) were detected with TA100 and/or TA98, and 9 out of 28 mutagens (32%) were detected with either TA1535, TA1537, WP2uvrA, or their combination.

Conclusion

The Ames test data presented here will help avoid duplicated Ames testing in some cases, support duplicate testing in other cases, improve in silico models, and enhance our understanding of the mechanisms of mutagenesis.

Supplementary Information

The online version contains supplementary material available at 10.1186/s41021-021-00206-1.

Assessing the impact of expert knowledge on ICH M7 (Q)SAR predictions. Is expert review still needed?

The ICH M7 (R1) guideline recommends the use of complementary (Q)SAR models to assess the mutagenic potential of drug impurities as a state-of-the-art, high-throughput alternative to empirical testing. Additionally, it includes a provision for the application of expert knowledge to increase prediction confidence and resolve conflicting calls. Expert knowledge, which considers structural analogs and mechanisms of activity, has been valuable when models return an indeterminate (equivocal) result or no prediction (out-of-domain). A retrospective analysis of 1002 impurities evaluated in drug regulatory applications between April 2017 and March 2019 assessed the impact of expert review on (Q)SAR predictions. Expert knowledge overturned the default predictions for 26% of the impurities and resolved 91% of equivocal predictions and 75% of out-of-domain calls. Of the 261 overturned default predictions, 15% were upgraded to equivocal or positive and 79% were downgraded to equivocal or negative. Chemical classes with the most overturns were primary aromatic amines (46%), aldehydes (45%), Michael-reactive acceptors (37%), and non-primary alkyl halides (33%). Additionally, low confidence predictions were the most often overturned. Collectively, the results suggest that expert knowledge continues to play an important role in an ICH M7 (Q)SAR prediction workflow and triaging predictions based on chemical class and probability can improve (Q)SAR review efficiency.