New approach methodologies (NAMs): identifying and overcoming hurdles to accelerated adoption.

New approach methodologies (NAMs) can deliver improved chemical safety assessment through the provision of more protective and/or relevant models that have a reduced reliance on animals. Despite the widely acknowledged benefits offered by NAMs, there continue to be barriers that prevent or limit their application for decision-making in chemical safety assessment. These include barriers related to real and perceived scientific, technical, legislative and economic issues, as well as cultural and societal obstacles that may relate to inertia, familiarity, and comfort with established methods, and perceptions around regulatory expectations and acceptance. This article focuses on chemical safety science, exposure, hazard, and risk assessment, and explores the nature of these barriers and how they can be overcome to drive the wider exploitation and acceptance of NAMs. Short-, mid- and longer-term goals are outlined that embrace the opportunities provided by NAMs to deliver improved protection of human health and environmental security as part of a new paradigm that incorporates exposure science and a culture that promotes the use of protective toxicological risk assessments.

Generation of a drug-induced renal injury list to facilitate the development of new approach methodologies for nephrotoxicity.

Drug-induced renal injury (DIRI) causes >1.5 million adverse events annually in the USA alone. Although standard biomarkers exist for DIRI, they lack the sensitivity or specificity to detect nephrotoxicity before the significant loss of renal function. In this study, we describe the creation of DIRIL - a list of drugs associated with DIRI and nephrotoxicity - from two literature datasets with DIRI annotation, confirmed using FDA drug labeling. DIRIL comprises 317 orally administered drugs covering all 14 anatomical, therapeutic and chemical (ATC) classification categories. Of the 317 drugs, 171 were DIRI-positive and 146 were DIRI-negative. DIRIL will be a relevant and invaluable resource for discovery of new approach methods (NAMs) to predict the occurrence and possible severity of DIRI earlier in drug development.

Academic drug discovery: Challenges and opportunities.

There are many different approaches to drug discovery in academia, some of which are based broadly on the industrial model of discovering novel targets and then conducting screening within academic drug discovery centres to identify hit molecules. Here we describe our approach to drug discovery, which makes more efficient use of the capabilities and resources of the different stakeholders. Specifically, we have created a large portfolio of drug projects and conducted small amounts of derisking work to ensure projects are investment ready. In this feature we will describe this model, including its limitations and advantages, since we believe the ideas and concepts will be of interest to other academic institutions and consortia.

Reduced prosocial motivation and effort in adolescents with conduct problems and callous-unemotional traits.

BACKGROUND: Prosocial behaviours - acts that benefit others - are of crucial importance for many species including humans. However, adolescents with conduct problems (CP), unlike their typically developing (TD) peers, demonstrate markedly reduced engagement in prosocial behaviours. This pattern is particularly pronounced in adolescents with CP and high levels of callous-unemotional traits (CP/HCU) who are at increased risk of developing psychopathy in adulthood. While a substantial amount of research has investigated the cognitive-affective mechanisms thought to underlie antisocial behaviour, much less is known about the mechanisms that could explain reduced prosocial behaviours in adolescents with CP. METHODS: Here we examined the willingness to exert effort to benefit oneself (self) and another person (other, prosocial condition) in children with CP/HCU, CP and lower levels of CU traits (CP/LCU) and their TD peers. The task captured both prosocial choices, and actual effort exerted following prosocial choices, in adolescent boys aged 11-16 (27 CP/HCU; 34 CP/LCU; 33 TD). We used computational modelling to reveal the mechanistic processes involved when choosing prosocial acts. RESULTS: We found that both CP/HCU and CP/LCU groups were more averse to initiating effortful prosocial acts than TD adolescents - both at a cognitive and at a behavioural level. Strikingly, even if they chose to initiate a prosocial act, the CP/HCU group exerted less effort following this prosocial choice than other groups. CONCLUSIONS: Our findings indicate that reduced exertion of effort to benefit others may be an important factor that differentiates adolescents with CP/HCU from their peers with CP/LCU. They offer new insights into what might drive low prosocial behaviour in adolescents with CP, including vulnerabilities that may particularly characterise those with high levels of CU traits.

Data science in drug discovery safety: Challenges and opportunities.

Early de-risking of drug targets and chemistry is essential to provide drug projects with the best chance of success. Target safety assessments (TSAs) use target biology, gene and protein expression data, genetic information from humans and animals, and competitor compound intelligence to understand the potential safety risks associated with modulating a drug target. However, there is a vast amount of information, updated daily that must be considered for each TSA. We have developed a data science-based approach that allows acquisition of relevant evidence for an optimal TSA. This is built on expert-led conventional and artificial intelligence-based mining of literature and other bioinformatics databases. Potential safety risks are identified according to an evidence framework, adjusted to the degree of target novelty. Expert knowledge is necessary to interpret the evidence and to take account of the nuances of drug safety, the modality, and the intended patient population for each TSA within each project. Overall, TSAs take full advantage of the most recent developments in data science and can be used within drug projects to identify and mitigate risks, helping with informed decision-making and resource management. These approaches should be used in the earliest stages of a drug project to guide decisions such as target selection, discovery chemistry options, in vitro assay choice, and end points for investigative in vivo studies.

PLM-ARG: antibiotic resistance gene identification using a pretrained protein language model.

MOTIVATION: Antibiotic resistance presents a formidable global challenge to public health and the environment. While considerable endeavors have been dedicated to identify antibiotic resistance genes (ARGs) for assessing the threat of antibiotic resistance, recent extensive investigations using metagenomic and metatranscriptomic approaches have unveiled a noteworthy concern. A significant fraction of proteins defies annotation through conventional sequence similarity-based methods, an issue that extends to ARGs, potentially leading to their under-recognition due to dissimilarities at the sequence level. RESULTS: Herein, we proposed an Artificial Intelligence-powered ARG identification framework using a pretrained large protein language model, enabling ARG identification and resistance category classification simultaneously. The proposed PLM-ARG was developed based on the most comprehensive ARG and related resistance category information (>28K ARGs and associated 29 resistance categories), yielding Matthew's correlation coefficients (MCCs) of 0.983 ± 0.001 by using a 5-fold cross-validation strategy. Furthermore, the PLM-ARG model was verified using an independent validation set and achieved an MCC of 0.838, outperforming other publicly available ARG prediction tools with an improvement range of 51.8%-107.9%. Moreover, the utility of the proposed PLM-ARG model was demonstrated by annotating resistance in the UniProt database and evaluating the impact of ARGs on the Earth's environmental microbiota. AVAILABILITY AND IMPLEMENTATION: PLM-ARG is available for academic purposes at https://github.com/Junwu302/PLM-ARG, and a user-friendly webserver (http://www.unimd.org/PLM-ARG) is also provided.

A generative adversarial network model alternative to animal studies for clinical pathology assessment.

Animal studies are unavoidable in evaluating chemical and drug safety. Generative Adversarial Networks (GANs) can generate synthetic animal data by learning from the legacy animal study results, thus may serve as an alternative approach to assess untested chemicals. AnimalGAN, a GAN method to simulate 38 rat clinical pathology measures, was developed with significant robustness even for the drugs that vary significantly from these used during training, both in terms of chemical structure, drug class, and the year of FDA approval. AnimalGAN showed comparable results in hepatotoxicity assessment as using the real animal data and outperformed 12 conventional quantitative structure-activity relationship approaches. Using AnimalGAN, a virtual experiment of 100,000 rats ranked hepatotoxicity of three structurally similar drugs in a similar trend that has been observed in human population. AnimalGAN represented a significant step with artificial intelligence towards the global effort in replacement, reduction, and refinement (3Rs) of animal use.

'It wasn't the strategies on their own': Exploring caregivers' experiences of accessing services in the development of interventions for autistic people with intellectual disability.

LAY ABSTRACT: Many autistic individuals with intellectual disability experience anxiety, and for those who use few or no words, anxiety may present as behaviour that challenges, such as self-injury and avoiding anxiety-provoking situations. Families report difficulty accessing support from services for autistic individuals experiencing anxiety. Moreover, once receiving support, effective interventions for autistic people with intellectual disability are limited. We completed individual and group discussions with 16 caregivers of autistic people with intellectual disability, to (a) explore their experiences of accessing services for anxiety and/or behaviour that challenges for their child; and (b) understand what matters to caregivers when developing interventions that have been designed for them and the autistic individual with intellectual disability that they support. Caregivers reported that services, in their experience, did not deliver the support that they expected, and that they often needed to 'fight' for support. Caregivers considered services and families working together, the inclusion of peer support, and families being offered interventions that are flexible to individual circumstances to be important. These considerations are valuable for clinicians and researchers developing interventions and aiming to improve outcomes for autistic people with intellectual disability and their families.

An integrated approach for early in vitro seizure prediction utilizing hiPSC neurons and human ion channel assays.

Seizure liability remains a significant cause of attrition throughout drug development. Advances in stem cell biology coupled with an increased understanding of the role of ion channels in seizure offer an opportunity for a new paradigm in screening. We assessed the activity of 15 pro-seizurogenic compounds (7 CNS active therapies, 4 GABA receptor antagonists, and 4 other reported seizurogenic compounds) using automated electrophysiology against a panel of 14 ion channels (Nav1.1, Nav1.2, Nav1.6, Kv7.2/7.3, Kv7.3/7.5, Kv1.1, Kv4.2, KCa4.1, Kv2.1, Kv3.1, KCa1.1, GABA α1ß2γ2, nicotinic α4ß2, NMDA 1/2A). These were selected based on linkage to seizure in genetic/pharmacological studies. Fourteen compounds demonstrated at least one "hit" against the seizure panel and 11 compounds inhibited 2 or more ion channels. Next, we assessed the impact of the 15 compounds on electrical signaling using human-induced pluripotent stem cell neurons in microelectrode array (MEA). The CNS active therapies (amoxapine, bupropion, chlorpromazine, clozapine, diphenhydramine, paroxetine, quetiapine) all caused characteristic changes to electrical activity in key parameters indicative of seizure such as network burst frequency and duration. The GABA antagonist picrotoxin increased all parameters, but the antibiotics amoxicillin and enoxacin only showed minimal changes. Acetaminophen, included as a negative control, caused no changes in any of the parameters assessed. Overall, pro-seizurogenic compounds showed a distinct fingerprint in the ion channel/MEA panel. These studies highlight the potential utility of an integrated in vitro approach for early seizure prediction to provide mechanistic information and to support optimal drug design in early development, saving time and resources.

DeepAmes: A deep learning-powered Ames test predictive model with potential for regulatory application.

The Ames assay is required by the regulatory agencies worldwide to assess the mutagenic potential risk of consumer products. As well as this in vitro assay, in silico approaches have been widely used to predict Ames test results as outlined in the International Council for Harmonization (ICH) guidelines. Building on this in silico approach, here we describe DeepAmes, a high performance and robust model developed with a novel deep learning (DL) approach for potential utility in regulatory science. DeepAmes was developed with a large and consistent Ames dataset (>10,000 compounds) and was compared with other five standard Machine Learning (ML) methods. Using a test set of 1,543 compounds, DeepAmes was the best performer in predicting the outcome of Ames assay. In addition, DeepAmes yielded the best and most stable performance up to when compounds were >30% outside of the applicability domain (AD). Regarding the potential for regulatory application, a revised version of DeepAmes with a much-improved sensitivity of 0.87 from 0.47. In conclusion, DeepAmes provides a DL-powered Ames test predictive model for predicting the results of Ames tests; with its defined AD and clear context of use, DeepAmes has potential for utility in regulatory application.

TransOrGAN: An Artificial Intelligence Mapping of Rat Transcriptomic Profiles between Organs, Ages, and Sexes.

Animal studies are required for the evaluation of candidate drugs to ensure patient and volunteer safety. Toxicogenomics is often applied in these studies to gain understanding of the underlying mechanisms of toxicity, which is usually focused on critical organs such as the liver or kidney in young male rats. There is a strong ethical reason to reduce, refine and replace animal use (the 3Rs), where the mapping of data between organs, sexes and ages could reduce the cost and time of drug development. Herein, we proposed a generative adversarial network (GAN)-based framework entitled TransOrGAN that allowed the molecular mapping of gene expression profiles in different rodent organ systems and across sex and age groups. We carried out a proof-of-concept study based on rat RNA-seq data from 288 samples in 9 different organs of both sexes and 4 developmental stages. First, we demonstrated that TransOrGAN could infer transcriptomic profiles between any 2 of the 9 organs studied, yielding an average cosine similarity of 0.984 between synthetic transcriptomic profiles and their corresponding real profiles. Second, we found that TransOrGAN could infer transcriptomic profiles observed in females from males, with an average cosine similarity of 0.984. Third, we found that TransOrGAN could infer transcriptomic profiles in juvenile, adult, and aged animals from adolescent animals with an average cosine similarity of 0.981, 0.983, and 0.989, respectively. Altogether, TransOrGAN is an innovative approach to infer transcriptomic profiles between ages, sexes, and organ systems, offering the opportunity to reduce animal usage and to provide an integrated assessment of toxicity in the whole organism irrespective of sex or age.

Science-led regulatory strategies in nonclinical development of new medicines.

The development of a pharmaceutical is a stepwise process involving an evaluation of both animal and human efficacy and safety information. Regulations around drug development exist to protect people and the environment from harm and should create a level playing field for business, allowing well-run companies to thrive. However, adherence to good science should guide decisions rather than rigorously following guidelines, and there is almost always more than one way to get to the ultimate goal.

Artificial intelligence and real-world data for drug and food safety - A regulatory science perspective.

In 2013, the Global Coalition for Regulatory Science Research (GCRSR) was established with members from over ten countries (www.gcrsr.net). One of the main objectives of GCRSR is to facilitate communication among global regulators on the rise of new technologies with regulatory applications through the annual conference Global Summit on Regulatory Science (GSRS). The 11th annual GSRS conference (GSRS21) focused on "Regulatory Sciences for Food/Drug Safety with Real-World Data (RWD) and Artificial Intelligence (AI)." The conference discussed current advancements in both AI and RWD approaches with a specific emphasis on how they impact regulatory sciences and how regulatory agencies across the globe are pursuing the adaptation and oversight of these technologies. There were presentations from Brazil, Canada, India, Italy, Japan, Germany, Switzerland, Singapore, the United Kingdom, and the United States. These presentations highlighted how various agencies are moving forward with these technologies by either improving the agencies' operation and/or preparing regulatory mechanisms to approve the products containing these innovations. To increase the content and discussion, the GSRS21 hosted two debate sessions on the question of "Is Regulatory Science Ready for AI?" and a workshop to showcase the analytical data tools that global regulatory agencies have been using and/or plan to apply to regulatory science. Several key topics were highlighted and discussed during the conference, such as the capabilities of AI and RWD to assist regulatory science policies for drug and food safety, the readiness of AI and data science to provide solutions for regulatory science. Discussions highlighted the need for a constant effort to evaluate emerging technologies for fit-for-purpose regulatory applications. The annual GSRS conferences offer a unique platform to facilitate discussion and collaboration across regulatory agencies, modernizing regulatory approaches, and harmonizing efforts.

US regulations to curb alleged cancer causes are ineffectual and compromised by scientific, constitutional and ethical violations.

The 1958 Delaney amendment to the Federal Food Drug and Cosmetics Act prohibited food additives causing cancer in animals by appropriate tests. Regulators responded by adopting chronic lifetime cancer tests in rodents, soon challenged as inappropriate, for they led to very inconsistent results depending on the subjective choice of animals, test design and conduct, and interpretive assumptions. Presently, decades of discussions and trials have come to conclude it is impossible to translate chronic animal data into verifiable prospects of cancer hazards and risks in humans. Such conclusion poses an existential crisis for official agencies in the US and abroad, which for some 65 years have used animal tests to justify massive regulations of alleged human cancer hazards, with aggregated costs of $trillions and without provable evidence of public health advantages. This article addresses suitable remedies for the US and potentially worldwide, by critically exploring the practices of regulatory agencies vis-á-vis essential criteria for validating scientific evidence. According to this analysis, regulations of alleged cancer hazards and risks have been and continue to be structured around arbitrary default assumptions at odds with basic scientific and legal tests of reliable evidence. Such practices raise a manifold ethical predicament for being incompatible with basic premises of the US Constitution, and with the ensuing public expectations of testable truth and transparency from government agencies. Potential remedies in the US include amendments to the US Administrative Procedures Act, preferably requiring agencies to justify regulations compliant with the Daubert opinion of the Daubert ruling of the US Supreme Court, which codifies the criteria defining reliable scientific evidence. International reverberations are bound to follow what remedial actions may be taken in the US, the origin of current world regulatory procedures to control alleged cancer causing agents.

Parenting boys with conduct problems and callous-unemotional traits: parent and child perspectives.

Parenting children with conduct problems (CP) is challenging, yet very little research has examined parenting using both quantitative and qualitative methods, from the perspective of the child and their parent/caregiver, and separately for those with high vs. low levels of callous-unemotional traits (HCU vs. LCU). One hundred and forty-six boys aged 11-16 [Typically developing (TD) n = 31; CP/HCU n = 35; CP/LCU n = 35] and their parents/caregivers completed the Alabama Parenting Questionnaire and provided a written qualitative statement describing their respective experiences of parenting/being parented. Parents/caregivers of CP/HCU boys reported more difficulty with child monitoring and supervision than parents of TD boys. This was echoed in qualitative reports of parents of CP/HCU boys reporting concerns regarding their child's safety. Parents/caregivers of both groups of CP boys reported more inconsistent discipline than parents of TD boys. Parental qualitative descriptions of challenging behavior in CP/HCU boys, and difficulties with setting boundaries and motivating CP/LCU boys, provided further insight to the potential triggers for inconsistent discipline. Qualitative reports from boys with CP indicated that they understood the parenting challenges their parents/caregivers faced. These findings replicate and extend previous work on the associations between parenting and CP. Children with CP/HCU and CP/LCU show some commonalities and differences in their parenting experiences and CP children and their parents/caregivers do not necessarily share all the same perceptions or concerns. CP interventions often involve parent/family engagement and this research highlights the continued importance of examining both parent and child perspectives.

Insulin Titration Guidelines for Patients With Type 1 Diabetes: It Is About Time!

PURPOSE: A proposal that an Insulin Advisory Committee develop insulin titration guidelines 100 years after its discovery. FINDINGS: Glucose control metrics remain poor despite significant advances in diabetes technology. SUMMARY: A century after the introduction of insulin, health care providers and patients with type 1 diabetes have worldwide access to a variety of insulin delivery devices (IDDs), glucose monitors, bolus calculators (BCs), continuous glucose monitors (CGMs), and automated insulin delivery (AID) systems. However, these advances have not enabled most patients to achieve today's clear A1c and time-in-range goals. Much of this failure arises from the lack of clear insulin titration guidelines for determining appropriate insulin doses. The lack of dosing clarity results in local physicians, clinics, and individual patients managing insulin titrations as they see fit, creating significant inefficiencies for reaching recommended glycemic goals. This review (1) details the widespread problems generated by nonphysiological dose settings in today's BCs, insulin pumps, and AID systems; (2) presents a method to develop and implement optimized total daily doses of insulin to correct the most common problem of hyperglycemia; (3) discusses using large device databases to provide clear insulin titration guidelines that optimize BC settings from an optimized total daily dose (TDD) of insulin for patients with T1D; and (4) recommends the formation of an Insulin Advisory Committee to clarify the steps to take toward universal insulin titration guidelines, optimized BC settings, and a systematic logic for their use in insulin delivery devices.

Corrigendum: DeepCarc: Deep learning-powered carcinogenicity prediction using model-level representation.

[This corrects the article DOI: 10.3389/frai.2021.757780.].

Adaptability of AI for safety evaluation in regulatory science: A case study of drug-induced liver injury.

Artificial intelligence (AI) has played a crucial role in advancing biomedical sciences but has yet to have the impact it merits in regulatory science. As the field advances, in silico and in vitro approaches have been evaluated as alternatives to animal studies, in a drive to identify and mitigate safety concerns earlier in the drug development process. Although many AI tools are available, their acceptance in regulatory decision-making for drug efficacy and safety evaluation is still a challenge. It is a common perception that an AI model improves with more data, but does reality reflect this perception in drug safety assessments? Importantly, a model aiming at regulatory application needs to take a broad range of model characteristics into consideration. Among them is adaptability, defined as the adaptive behavior of a model as it is retrained on unseen data. This is an important model characteristic which should be considered in regulatory applications. In this study, we set up a comprehensive study to assess adaptability in AI by mimicking the real-world scenario of the annual addition of new drugs to the market, using a model we previously developed known as DeepDILI for predicting drug-induced liver injury (DILI) with a novel Deep Learning method. We found that the target test set plays a major role in assessing the adaptive behavior of our model. Our findings also indicated that adding more drugs to the training set does not significantly affect the predictive performance of our adaptive model. We concluded that the proposed adaptability assessment framework has utility in the evaluation of the performance of a model over time.